Trajectories of In-Person and Remote Care Utilization: A Sequence Analysis of Primary Care Modalities Pre- and Post-COVID-19

Trajectories of In-Person and Remote Care Utilization: A Sequence Analysis of Primary Care Modalities Pre- and Post-COVID-19 | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Trajectories of In-Person and Remote Care Utilization: A Sequence Analysis of Primary Care Modalities Pre- and Post-COVID-19 Taylor Rapson, Kathryn E Kemper-McIsaac, Namuun Clifford, Elizabeth B Sherwin, and 4 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8176924/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 9 You are reading this latest preprint version Abstract The COVID-19 pandemic propelled the growth of virtual care, but few studies have examined how it reshaped care utilization patterns longitudinally across diverse populations. We used multichannel sequence analysis to classify 10,671 diabetic adults at two San Francisco health systems by utilization patterns from April 2019-March 2023. Patients who transitioned to digital care were Black or White, were Medicare beneficiaries, or had higher comorbidity burdens. Individuals who increased their combined in-person and virtual healthcare utilization were Hispanic with Spanish language preference and had greater disease burden. Those who decreased care utilization overall were Asian with Chinese language preference and had low patient portal engagement. While technology-enabled care models have the potential to enhance access, certain populations may be at risk of reduced access to care. Health systems should consider targeted interventions to ensure equitable access to telehealth for populations at risk of digital exclusion with the growth of remote care. Health sciences/Diseases Health sciences/Health care Health sciences/Medical research Sequence analysis health disparities telehealth between-visit care healthcare utilization Figures Figure 1 Figure 2 Introduction The COVID-19 pandemic spurred the growth of telehealth (scheduled video and/or telephone encounters) and between-visit care, such as unscheduled telephone visits and secure messaging in the patient portal, as alternative modalities for receiving healthcare 1 – 3 . Use of digital health and personal devices to engage in remote care has been linked to improved quality of care and clinical outcomes, particularly among individuals with chronic disease 4 – 9 . Evidence suggests that using the patient portal to send secure messages, request refills, and schedule appointments is linked to patient empowerment and engagement in their care, as well as improved patient-provider communication 6 – 12 . Similarly, use of the patient portal and telehealth has been shown to have a positive effect on self-management behaviors and clinical outcomes across chronic conditions, such as improved symptom control and medication adherence, which are critical drivers of health 4 , 6 – 8 , 13 – 16 . Despite the well-documented increase in the use of telehealth visits and between-visit care as additional access points for engagement with primary care clinical teams in recent years, the longitudinal healthcare utilization patterns comprising multiple modalities of care and their differences across sociodemographic groups are understudied. Prior research has documented unequal uptake of digital healthcare across racial and ethnic groups, language preference, age, and gender, as well as differences in use for those facing barriers to accessing broadband internet, devices, or data (e.g., due to cost) 7 , 10 , 14 , 15 , 17 – 26 . In particular, older adults, especially those with non-English language preferences or who identify as a racial or ethnic minority, experience significant disparities in digital health adoption due to disparities in device access, data access, or digital literacy 8 , 18 , 22 , 27 . Together, these barriers create a “digital divide” that may worsen inequities in healthcare settings that increasingly rely on digital health tools, especially in systems serving low-income, diverse populations 8 , 14 , 21 , 28 , 29 . As digital health becomes further embedded in healthcare delivery, these barriers to the uptake and use of digital health tools may ultimately shape healthcare utilization and outcomes for marginalized populations 30 . These concerns are particularly salient for patients with Type 2 Diabetes Mellitus (T2DM), one of the most common chronic conditions and one that disproportionately affects marginalized populations. Patients with uncontrolled diabetes and complications are disproportionately represented in minoritized racial and ethnic groups and lower socioeconomic neighborhoods, highlighting the need for equitable care modalities 14 , 31 , 32 . Because effective diabetes management requires ongoing monitoring, communication, and care coordination, patients with T2DM may benefit substantially from expanded modalities for accessing care. Remote tools can help address barriers to in-person care, support continuity between visits, and encourage better disease self-management. Yet, given persistent disparities in both digital health adoption and diabetes control, it remains uncertain whether the expansion of telehealth and patient portal use during the pandemic has narrowed or widened existing inequities. Understanding healthcare utilization patterns is, therefore, critical—not only to evaluate how patients with T2DM and other chronic conditions engage with multiple modalities of care, but also to identify where gaps in care delivery may exacerbate disparities. However, most prior studies rely on repeated cross-sectional data, examining a single care modality and classifying individuals as either “high” or “low” utilizers of care 33 – 36 . Such designs capture snapshots rather than trajectories, limiting the ability to understand the multifaceted nature of healthcare utilization and how it evolves over time, particularly in an era of expanded digital health use 2 – 4 . In contrast, a longitudinal, sequence-based approach can reveal how individuals transition between in-person, telehealth, and between-visit modalities, illuminating dynamic patterns that reflect real-world care experiences. Identifying patterns of underutilization across modalities can further pinpoint unmet needs and opportunities to improve equity. For example, a recent study highlighted that when telehealth, between-visit care, and in-person care were available to patients, there were fewer differences in utilization by race and ethnicity and preferred language, suggesting that multiple options for care utilization can reduce disparities 30 . Still, research remains limited on how these alternative access points to care can alleviate barriers to care and interact over time to supplement traditional care-seeking patterns 8 , 33 . Sequence analysis is a data-driven approach that compares and groups longitudinal trajectories of events, making it well-suited for evaluating healthcare utilization patterns, as it preserves the timing, order, and duration of event 37 – 39 . While studies often summarize utilization as aggregate counts over fixed windows, multichannel sequence analysis can reveal when and in what order patients use in-person, telehealth, and between-visit care, identifying recognizable patterns (for example, “transitioned to digital” or “reduced care engagement” 39–41 . This moves beyond simple categorizations of “high” or “low” utilizers and can surface nuanced variation in utilization patterns. The ability to cluster patients based on the timing, order, and duration of their healthcare encounters can provide a more granular understanding of patient heterogeneity and reveal hidden patterns and subgroups 42 , 43 . This study advances the existing research by employing multichannel sequence analysis and cluster analysis to create clusters of patients with similar temporal sequences of in-person, between-visit, and telehealth interactions across two health systems. We examine sociodemographic differences between clusters of healthcare utilization to assess whether specific patient characteristics are disproportionately represented in certain clusters. By comparing clusters across two health systems, we aim to 1) identify heterogeneities in multi-modality healthcare utilization over time, 2) assess whether clusters of care utilization are patterned by socio-demographic factors, which could suggest potential inequities in access and utilization, and 3) assess whether care utilization patterns are consistent across health systems or vary by site, potentially reflecting differences in patient populations, resources, and care delivery models. Given the potential role of remote care engagement to either mitigate or perpetuate existing disparities, the evidence from these comparisons can shape policies that promote equitable integration of digital health across diverse healthcare settings. Methods Study Setting and Data Source We conducted a 4-year retrospective longitudinal cohort study of adults with Type 2 Diabetes Mellitus (T2DM) receiving care at two health systems in San Francisco: the University of California, San Francisco (UCSF) and the San Francisco Health Network (SFHN). UCSF is an academic tertiary care center with four primary care practices. SFHN is an integrated, urban safety-net health system with 14 primary care practices. The study period was April 1, 2019 – March 31, 2023 at UCSF and October 1, 2019 – March 31, 2023 at SFHN. The difference in start date was due to the adoption of a new EHR software in August 2019 at SFHN. We used electronic health records (EHR) from the two health systems, capturing sociodemographic data, clinical characteristics, and encounter data. Since the study procedures involved only a review of data collected for routine clinical care, informed consent was waived due to minimal risk to patients in the cohort. All study procedures were approved by the UCSF Institutional Review Board (IRB #20-31253) in accordance with the Declaration of Helsinki. We report our procedures and results in accordance with the STROBE guidelines. Study Sample We identified a cohort of adult patients \(\:\ge\:\) 18 years old with T2DM who were assigned a primary care provider as of April 1, 2019, at UCSF and SFHN. The creation of the cohort is described in detail in a previous study 30 (Methodological Appendix). In brief, patients with T2DM were identified based on the following criteria before the study start date: an active DM health maintenance modifier; having a T2DM-related ICD-9/ICD-10 code on their active problem list; the most recent HbA1c test result ≥6.5%, and/or having an active diabetes medication on their medication list. To ensure active engagement in care, patients were only included if they had at least one encounter with their diabetes (primary care or endocrinology) team between April 1, 2019 – March 31, 2021, and at least one healthcare interaction (including patient portal messages, phone calls) in the last 12 months of the study period (April 1, 2022 – March 31, 2023). Exposure variables We included a comprehensive set of patient-level sociodemographic characteristics as of 4/1/2019 that were extracted from the EHR. We included age (as a continuous variable and categorized as 18–34, 35–49, 50–64, 65–74, or 75+), race and ethnicity (categorized as Hispanic/Latino(a); non-Hispanic Asian; non-Hispanic Black; non-Hispanic White/Caucasian; or other non-Hispanic races, which included American Indian or Alaskan Native, Native Hawaiian or Other Pacific Islander), preferred language (English, Chinese, Spanish, or Other), sex (male/female), health insurance as of April 2019 (Commercial/Private, Medicare, Medicaid, Healthy Workers [an SF-specific program that provides health insurance coverage to people working in In-Home Supportive Services, foster care, or other publicly funded roles who do not qualify for traditional Medicaid or employer-sponsored coverage], or Uninsured), and neighborhood socioeconomic status (nSES) based on geocoded addresses, and categorized as quintiles from low to high 44 . Further details about insurance status and nSES are included in the Methodological Appendix. Clinical characteristics included the Charlson comorbidity index (CCI, modified to exclude diabetes) from 03/01/2019-02/28/2020, baseline A1c as of 04/01/2019, and baseline blood pressure control as of 04/01/2019. We dichotomized CCI as 0–2 or 3 + to represent the severity of disease comorbidities 45 . Baseline A1c control was categorized as: controlled A1c (A1c \(\:\le\:\) 8), uncontrolled A1c (A1c \(\:>\) 8), or missing. Baseline blood pressure was categorized as: \(\:\le\:\) 120/80 (no hypertension), >120/80 and \(\:\le\:\) 140/90 (controlled hypertension), or >140/90 mmHg (uncontrolled hypertension). Patient portal enrollment was included as a proxy measure of digital access. Patient portal enrollment status was extracted from the EHR as of 4/1/2020 and documented as active or inactive (yes/no). Construction of Care Trajectories and Cluster Analysis We conducted sequence and cluster analysis to group patients into homogeneous clusters of healthcare utilization over time. This approach enabled us to measure the similarity of each patient’s sequence for each encounter modality across quarters and cluster similar sequences together 37 . For example, a patient whose sequence includes frequent in-person visits early in the study period followed by a shift to telehealth visits may be grouped in a cluster representing patients who transitioned from traditional to digital care modalities. To account for multiple modalities (in-person visits, telehealth visits, and between-visit interactions), we employed multichannel (also called multidimensional) sequence analysis, which allows creating three distinct channels, one for each encounter type 46 . Analyses were conducted in R Version 4.2.1 using the “TraMineR,” “cluster,” and “UniversalCVI” packages 47 – 50 . Step 1: Building individual sequences across three channels We summarized care utilization across three distinct types of patient encounters with the healthcare team (comprised of physicians, advanced practice providers, nurses, social workers, pharmacists, dietitians, medical assistants, or embedded behavioral health clinicians): In-person visits : scheduled in-person visits with a member of the care team. Telehealth visits : scheduled video or telephone visits with a member of the care team. Between-visit interactions : care that occurred between visits, documented in the EHR as unscheduled telephone calls or patient portal messages. For each patient, quarterly counts of encounters for each modality were grouped into ordered categories based on utilization levels. These categories were designed to capture meaningful differences in utilization while maintaining sufficient sample sizes within each category (details in Methodological Appendix). In both health systems, Q2 2020 (April–June 2020) was excluded from the analysis due to the unique encounter patterns during the COVID-19 Shelter-in-Place period (further detailed in Appendix B). For each modality, quarterly counts were binned into the following ordered categories: In-person visits : UCSF: 0, 1, 2, 3+; SFHN: 0, 1, 2, 3–5, 6+ Telehealth visits : UCSF: 0, 1, 2+; SFHN: 0, 1, 2 Between-visit interactions : UCSF: 0,1,2,3,4–6,7+; SFHN: 0, 1, 2,3–4, 5+ Step 2: Sequence analysis to quantify differences between trajectories After building the individual sequences, we then used multichannel sequence analysis to compare patients’ trajectories over time, computing differences between sequences for each modality using the dynamic Hamming distances (DHD) method 51 (further detailed in Methodological Appendix). This approach allowed us to assess whole-person care patterns over time and identify clusters of patients with similar utilization trajectories. Step 3: Cluster analysis to group similar trajectories We grouped patients into healthcare utilization clusters using agglomerative hierarchical clustering with Ward’s linkage. Cluster quality was evaluated using indicators such as the Average Silhouette Width (ASW), the Hubert’s C index (HC), and the Calinski–Harabasz (CH) index to identify cluster solutions with high homogeneity within clusters and distinction between clusters (see Appendix C for more details). Since several cluster solutions may have similar quality, we also used expert judgement to make considerations on parsimony and salience and choose the final number of clusters 52 . To ease our comparison of clusters within and between health systems, we conducted post-hoc analyses to categorize clusters based on similar overall trends in utilization. We grouped clusters into categories based on their trajectories after the shelter-in-place mandate was lifted, relative to their pre-pandemic behavior. We further characterized clusters by pre-pandemic in-person utilization levels. Statistical analyses We described patient characteristics overall and by healthcare utilization trajectory. In addition to the exposure variables listed above, we included an interaction term between race and ethnicity and preferred language to capture the intersectional and integrated nature of these two variables. Categorical variables were compared using chi-squared tests. Analyses were conducted separately for each health system, with cross-system comparisons of care trajectories and pre-pandemic utilization presented descriptively. Sensitivity analyses Sensitivity analyses were conducted to assess the robustness of the clustering approach, ensuring the stability of cluster solutions under different assumptions and methodologies. We repeated the sequence and cluster analysis using an alternative sequence dissimilarity measure and compared the resulting cluster solutions and interpretations. Results First, we separately outlined the visit trajectories within each healthcare system, given their unique patient populations and different uptake of telehealth and between-visit care during and post-pandemic. UCSF Patient Characteristics After excluding patients without a visit in the last 12 months of the study period, we identified a cohort of 3,860 patients with T2DM at UCSF (Table 1 ). Over half (53%) of patients were aged 65+. Most patients were English-speaking (82%), and the cohort included high proportions of Asian (40%, including 12% with non-English preference) and White (24%) patients. Patients were relatively evenly distributed across nSES quintiles, with 17% residing in the lowest and 18% in the highest quintile. Most patients in the cohort had insurance coverage through Commercial/private (44%) and Medicare plans (43%). Almost a quarter (23%) of patients had high comorbidity scores (CCI 3+), and nearly a third had uncontrolled A1c at baseline. Digital engagement was high, with 78% of patients having an active MyChart account at baseline. Description of Care Utilization Clusters We identified seven distinct care utilization clusters as the final number of clusters within the UCSF diabetes cohort, representing diverse patterns of engagement with in-person, telehealth, and between-visit care modalities over time. Figure 1 displays the distribution of each care modality at each time point by cluster. Index plots showing all the individual trajectories in each cluster are displayed in Appendix E.1. High Utilizers who added Telehealth (n = 177): These patients had very high in-person and between-visit care before the pandemic. About a year after the shelter-in-place mandate, they returned to very high in-person care and sustained high levels of between-visit care alongside moderate telehealth uptake. High utilizers who added Telehealth and Between-Visit care (n = 333): Patients in this cluster had high in-person and moderate between-visit care before the pandemic. Following the shelter-in-place mandate, they returned to prior in-person care levels, sustained moderate increases in telehealth, and increased between-visit care. High utilizers who transitioned to Telehealth (n = 212): These patients had high in-person, very high between-visit care, and some telehealth use before the pandemic. Post-shelter-in-place, they shifted to high telehealth and between-visit care without returning to prior in-person rates. Moderate utilizers who transitioned to Telehealth (n = 584): These patients had moderate in-person and between-visit care before the pandemic. Post-shelter-in-place, they shifted to moderate telehealth and increased between-visit care without returning to prior in-person rates. Moderate utilizers who lost care (n = 173): These patients had moderate in-person and between-visit care before the pandemic but did not recover either modality after the shelf-in-place period, and telehealth uptake was minimal. Low utilizers who added Telehealth (n = 769): These patients had low in-person and between-visit care before the pandemic. Post-shelter-in-place, they returned nearly to prior in-person rates, while adding telehealth and slightly increasing between-visit care. Low utilizers who lost care (n = 1,612): These patients had low utilization across all modalities before and after the shelter-in-place mandate. They did not return to pre-pandemic in-person rates, demonstrated minimal telehealth uptake, and had no between-visit care. Sociodemographic Characteristics by Post-Pandemic Care Trajectories Table 2 details the seven clusters classified by pre-pandemic utilization and post-pandemic changes in in-person, telehealth, and between-visit care trajectories over time. For comparisons of sociodemographic and clinical characteristics, clusters were grouped into three post-pandemic trajectories: added digital care (telehealth and/or between-visit care), transitioned to digital care (telehealth and/or between-visit care), and lost care. Among the 3,860 patients, 33% added digital care, 21% transitioned to digital care, and 46% lost care. Sociodemographic composition varied significantly across trajectories (Table 3 , full descriptive tables for each cluster are presented in Appendix D.1). Patients who added digital care were more likely to be female (p = .02) and Black/African American (p < 0.001), or Hispanic with non-English language preference (p < 0.001) compared to patients who lost care or transitioned to digital care. These patients were also slightly more likely to be covered by Medicaid (p = .031), have high comorbidity scores (CCI 3+) (p < 0.001), and uncontrolled A1c at baseline (p < 0.001). Patients who transitioned to digital care were more likely to be White, Black/African American, or Hispanic with an English-language preference (p < 0.001), and be covered by Medicaid or Medicare (p = .031) compared to patients in other care trajectories. Individuals who transitioned to digital care also had higher comorbidity scores (CCI 3+) (p < 0.001), were less likely to have missing A1c or blood pressure values (p < 0.001 and p = 0.001, respectively) and were more likely to have an active patient portal at baseline (p < 0.001). Patients who lost care were more likely to be male (p = 0.02), Asian - particularly with a non-English language preference (p < 0.001), and commercially insured (p = .031) compared to patients in other care trajectories. Patients who lost care were also more likely to have lower comorbidity scores (CCI 0–2) (p < 0.001), missing baseline A1c or blood pressure results (p < 0.001 and p = 0.001, respectively), and low patient portal enrollment (p < 0.001). SFHN Patient Characteristics After excluding patients without a visit in the last 12 months of the study period, we identified a cohort of 6,811 patients with diabetes at SFHN (Table 1 ). The SFHN diabetes cohort was younger than the UCSF cohort, with two-thirds of patients under the age of 65. Most patients at this health system were non-English speakers (57%), including 28% identifying as Hispanic and Spanish preference and 26% identifying as Asian and Chinese preference. Most patients were insured through Medicaid (40%), Medicare (31%), or Healthy Workers (15%), and nearly two-thirds (64%) resided in neighborhoods within the lowest two nSES quintiles. While 85% of patients had lower comorbidity scores (CCI < 3), 39% had uncontrolled A1c at baseline. Digital engagement was notably low, with less than 10% of patients having an active MyChart account at baseline. Description of Care Utilization Clusters We identified seven care utilization clusters as the final number of clusters within the SFHN cohort, described below. Figure 2 displays the distribution of each care modality at each time point by cluster. Index plots showing all the individual trajectories are displayed in Appendix E.2. High utilizers who added Telehealth and Between-Visit care (n = 319): These patients had high in-person and between-visit care prior to the pandemic. Within one year of the shelter-in-place mandate, they returned to pre-pandemic in-person care levels while adding more between-visit care and moderate telehealth uptake. High utilizers who added Between-Visit care (n = 470): These patients had high in-person and moderate between-visit care before the pandemic. A year after the shelter-in-place mandate was lifted, they returned to prior in-person rates, while increasing between-visit care. Telehealth uptake was minimal and not sustained over time. High utilizers who transitioned to Telehealth (n = 285): These patients had high in-person and between-visit care with some telehealth use before the pandemic. Post-shelter-in-place, they did not return to pre-pandemic in-person care levels but maintained high between-visit care and had high adoption of telehealth. Moderate utilizers who transitioned to Telehealth and Between-Visit care (n = 364): These patients had moderate in-person and high between-visit engagement prior to the pandemic. Post-shelter-in-place, they did not return to pre-pandemic in-person levels but had high telehealth uptake and increased between-visit care. Moderate utilizers with minimal change (n = 707): These patients had moderate in-person and low between-visit care prior to the pandemic. After the shelter-in-place mandate lifted, their in-person care levels nearly returned to pre-pandemic levels, with slight telehealth uptake and between-visit engagement that was not sustained over time. Moderate utilizers who lost care (n = 973): These patients had moderate in-person and between-visit care utilization prior to the pandemic. After the shelter-in-place mandate, they did not recover in-person care levels, which continued to decline over time, with any telehealth and between-visit care are also trending downward. Low utilizers who lost care (n = 3,693): These patients had low in-person and between-visit utilization before the pandemic. Post-shelter-in-place, they did not return to in-person levels and maintained minimal between-visit care, with any telehealth uptake was not sustained. Sociodemographic Characteristics by Post-Pandemic Care Trajectories We classified the seven SFHN clusters by pre-pandemic utilization and post-pandemic changes in in-person, telehealth, and between-visit care (Table 4 ) and grouped them into four trajectories: added digital care (telehealth and/or between-visit care), transitioned to digital care (telehealth and/or between-visit care), lost care, and sustained minimal change (Table 5 , full descriptive tables for each cluster are presented in Appendix D.2). Patients who added digital care were more likely to be \(\:\ge\:\) 50 years old (p = .037), Hispanic with a non-English language preference (p < 0.001), and insured through Medicaid or Medicare (p < 0.001) compared to other care trajectories. Clinically, these patients were also more likely to have high comorbidity scores (CCI 3+) (p < 0.001) and uncontrolled A1c levels at baseline (p < 0.001). Patients who transitioned to digital care were more likely to be 75+ years old (p = .037), Black/African American or White (p < 0.001), and much less likely to have a non-English language preference (p < 0.001) compared to other care trajectories. These patients were most likely to be covered by Medicare, had high comorbidity scores (CCI 3+), have controlled baseline A1c levels, and have an active patient portal at baseline (p < 0.001 for all characteristics). Patients who lost care were more likely to be Asian -particularly with a non-English language preference (p < 0.001) and be enrolled in Healthy Workers (p < 0.001) compared to other care trajectories. Clinically, these patients were more likely to have lower comorbidity scores (CCI 0–2), missing baseline A1c and blood pressure results, and low patient portal enrollment (p < 0.001 for all characteristics). SFHN patients who had minimal change in care patterns were more likely to be 35–49 years old (p = .037), Hispanic/Latino with a non-English language preference (p < 0.001), and uninsured (p < 0.001) compared to other care trajectories. Clinically, they were also more likely to have lower comorbidity scores (CCI 0–2), slightly more likely to have uncontrolled A1c at baseline, and a very low likelihood of being enrolled in the patient portal (p < 0.001 for all characteristics). Key Patterns in Care Trajectories Across Health Systems Care Trajectories by Post-Pandemic Changes in Utilization Next, we descriptively compared these trajectories across healthcare systems to identify important patterns and/or differences. UCSF saw higher proportions of patients who added or transitioned to digital care, with 33% of patients adding digital care and 21% transitioning to digital care (Table 3 ). In contrast, SFHN had lower proportions in these categories: 12% added digital care and 10% transitioned to digital care (Table 5 ). At UCSF, those who added digital care (Clusters UCSF 1, UCSF 2, UCSF 6) demonstrated sustained moderate to high telehealth uptake and varying rates of between-visit interactions, but with an increase in use overall (Fig. 1 ). At SFHN, however, those who added digital care (Clusters SFHN 1, SFHN 2) showed low to moderate uptake of telehealth (that was not sustained) and higher volumes of between-visit care (Fig. 2 ). At both UCSF and SFHN, those who transitioned to digital care (Clusters UCSF 3, UCSF 4, SFHN 3, SFHN 4) showed moderate to high uptake of telehealth. Patients who transitioned to digital care had heterogeneous uptake of between-visit care, as some clusters increased their interactions (Clusters UCSF 4, SFHN 4) while others maintained moderate to high uptake (with no sustained increase in volume; Clusters UCSF 3, SFHN 3). SFHN exhibited higher proportions of patients who either maintained prior care patterns with minimal changes or experienced care disengagement: 10% maintained in-person visit rates with minimal to no uptake of digital care, and 69% experienced care disruptions and lost care (compared to 46% of patients who lost care at UCSF). Patients who lost care (Clusters UCSF 5, UCSF 7, SFHN 6, SFHN 7) demonstrated low adoption of telehealth and between-visit care, often characterized by consistent non-usage or decreasing trends in use. In-person visit rates also did not recover to pre-pandemic levels among the clusters that lost care. Care Trajectories by Pre-Pandemic Utilization To compare care trajectories across both health systems, we focused on how care patterns changed for participants who were high, moderate, or low utilizers in the pre-pandemic period (see Appendix F for a detailed description). In brief, high utilizers mostly returned to their pre-pandemic in-person rates of care and incorporated both telehealth and/or between-visit care with sustained use of between-visit care modalities. Moderate utilizers had more heterogeneity, with some transitioning to digital care, some losing care, and some showing minimal changes. Low utilizers were the largest cluster in both health systems; some low utilizers lost care, while some added telehealth care. Sensitivity Analysis Sensitivity analyses using Optimal Matching (OM) instead of Dynamic Hamming Distance (DHD) produced broadly similar trajectory types. While OM yielded some variation in the distribution of patients, the overall care patterns and substantive interpretations remained consistent with the DHD results (Appendix G). Discussion In a large, diverse sample of patients with T2DM across two large San Francisco health systems, we characterized clusters of multi-modal healthcare utilization patterns across three types of visits (in-person, telehealth visits, and between-visit encounters) from April 2019 to March 2023 using sequence and cluster analysis. Across both UCSF and SFHN, three distinct patterns emerged: adding digital care, where in-person visit volumes returned to baseline while telehealth and between-visit communications remained elevated; transitioning to digital care, characterized by sustained reductions in in-person visits alongside increased telehealth and/or between-visit use; and loss of care, with neither in-person nor digital services returning to pre-pandemic levels. At SFHN, a fourth pattern—minimal change—was observed, reflecting stable use of in-person visits with persistently low use of telehealth and between-visit care. Within these patterns, we noticed heterogeneity not only in the degree of telehealth uptake and sustainability of between-visit care, but also in the timing of these shifts—differences that were captured through our use of sequence analysis—highlighting nuanced variation both across and within health systems. We then examined whether sociodemographic and clinical factors were associated with distinct healthcare utilization trajectories. Our findings provide critical insights into the factors shaping transitions to digital care, sustained in-person use, and care loss. Post-Pandemic Shifts in Care Trajectories A key finding was the identification of characteristics among the clusters that adopted digital care. Several clusters reflected a transition from primarily in-person care to primarily digital engagement after the shelter–in–place mandate, which was more common among Black or White patients with an English-language preference, those covered by Medicare, and those with higher comorbidity burdens. These patterns suggest that digital modalities may be particularly attractive or accessible for managing complex health needs or for those facing barriers to consistent in-person care, such as transportation, work schedules, or mobility issues, which are more pronounced among marginalized groups 53 , 54 . These results align with the existing literature, which indicates the value of hybrid models of care in increasing accessibility and supporting chronic disease management 4 , 6 , 7 , 9 , 29 , 53 – 55 . These findings are particularly insightful because they both corroborate prior literature on digital shifts in care delivery and extend it to a broader population, including both academic and safety-net settings. By analyzing two health systems serving demographically distinct populations, we demonstrate that digital care adoption is not limited to well-resourced populations but also emerges as a critical tool for sustaining care engagement in more diverse, underserved settings. Notably, UCSF clusters experienced a higher uptake of telehealth compared to those at SFHN, which is surprising since SFHN primarily relied on generally more accessible between-visit interactions or phone visits, whereas patients at UCSF had higher proportions of video visits 30 . Heterogeneity in Sociodemographic Characteristics Race and ethnicity, preferred language, and health insurance were associated with each care trajectory, yet their distributions differed across systems. This heterogeneity implies that demographic factors do not determine care utilization but can vary depending on the context, such as the health system’s capacity or digital health outreach approach. Notably, race and ethnicity and preferred language were similar across both health systems, with patients who added digital care being more likely to be Spanish-speaking Hispanic, particularly at SFHN; patients who transitioned to digital care were more likely to be White or Black, and those who lost care were more likely to be Asian with a non-English language preference. Differences by race and ethnicity and language indicate that cultural and linguistic factors may shape how patients engage with digital care, reflecting persistent barriers as well as opportunities to enhance access. At the same time, system-level elements, such as language-concordant communication, digital literacy support, and targeted outreach, likely play a critical role in determining which patients are able to adopt and sustain digital care use, underscoring the need for equity-oriented implementation strategies 23 , 56 , 57 . The fact that individuals who added digital care were more likely to be Spanish-speaking suggests that both health systems were successful in offering and supporting the adoption of digital care options for Spanish speakers and can serve as a model for engaging diverse populations who may face barriers to digital engagement. Moreover, there was notable heterogeneity in health insurance coverage across the two health systems, highlighting how structural context may shape care engagement. These differences reflect the broader roles of each system: UCSF as a tertiary referral center with a commercially insured base, and SFHN as a safety-net system serving populations with more limited financial and digital access. Insurance status corresponded closely to care trajectories. At UCSF, patients who lost care were more likely to be commercially insured, a group that may experience weaker continuity with a single health system due to broader care alternatives, potentially reflecting data loss rather than true change in care patterns. At SFHN, uninsured patients were overrepresented among those who lost care or exhibited minimal changes in utilization. For uninsured patients, financial barriers and limited access options likely constrain sustained engagement, while for commercially insured patients, greater flexibility in provider choice may diffuse continuity across systems. Insurance type functions as a structural determinant of healthcare engagement, influencing not only access but also the stability and modality of care over time. Health systems serving distinct payer mixes, therefore, face different equity challenges: safety-net systems must continue to address affordability and digital access, while commercially oriented systems may need to strengthen coordination and continuity within fragmented care networks. Addressing these gaps through targeted interventions—such as insurance navigation support, proactive outreach, and integrated digital care supports—may improve continuity and mitigate disparities across diverse insurance contexts. Additionally, examining how policy-level factors, such as insurance or network restrictions, influence care engagement could inform system-wide strategies to prevent care loss across diverse populations. Care Trajectories Based on Pre-Pandemic Utilization When examining care trajectories through the lens of patients’ pre-pandemic utilization levels, clear trends emerged across both health systems. High utilizers prior to the pandemic were the most likely to adopt or transition to digital care, demonstrating moderate to high telehealth uptake and between-visit care engagement. This suggests that the most engaged patients were more likely to incorporate digital modalities into their care. Notably, there was one UCSF cluster of low utilizers that also added care, demonstrating some heterogeneity. No clusters among moderate utilizers showed consistent trends of adding care. Those who transitioned to digital care tended to be moderate to high utilizers prior to the pandemic, adopting telehealth and between-visit care as a partial substitute for lost in-person visits. However, for both the trajectories that added digital care or transitioned to digital care, sustained between-visit growth was not universal, suggesting that engagement in care between visits may ebb and flow depending on patients’ needs and interests. Future studies should expand upon this work to evaluate if and how between-visit care supports care access. Patients who lost care constituted the largest segments at both systems, mainly comprised of low or moderate pre-pandemic utilizers with limited telehealth uptake. This pattern underscores how individuals with historically lower engagement may be particularly vulnerable to care disruption and highlights the limited capacity of digital modalities alone to re-engage these patients. Notably, SFHN had a larger group of low utilizers who added care, suggesting that targeted outreach or leveraging more accessible modalities of engagement (such as greater emphasis on telephone calls versus video visits) may mitigate disengagement. Future Directions These results underscore the complex interplay between digital health adoption, socioeconomic factors, and healthcare disparities in urban contexts. By examining how patients navigate and combine multiple care modalities, we can better understand how these options may create complexity, overwhelm, or mismatch with patient needs, ultimately leading to reduced engagement. In this study, those who lost care or sustained minimal changes in their care were more likely to speak Chinese or Spanish. Several recent studies demonstrated that Chinese and Spanish speakers were disproportionately more likely to struggle with specific tasks required to use digital health tools effectively, reinforcing that linguistic and cultural barriers influence how patients access and interact with different modalities and aspects of care 18 , 59 , 65 . However, a recent study at SFHN found that language-concordant digital skills training particularly benefits Spanish speakers and patients with lower digital engagement, highlighting a possible strategy to address barriers to and reduce disparities in engagement among those at risk of care disengagement 66 . Designing tailored interventions and strategies for moderate and low utilizers at risk for care disengagement, including language-concordant support, digital literacy programs, and outreach efforts, has the potential to address barriers such as perceived usefulness and digital skills. The differences in cluster patterns at UCSF and SFHN also indicate that digital engagement is not one-size-fits-all, and the importance of tailoring digital adoption strategies to the structural and patient-level realities of each organization, including digital preferences and needs 30 , 53 , 54 . Clusters at SFHN demonstrated moderate to low telehealth engagement, compared to UCSF, which had multiple clusters with high telehealth adoption; but SFHN patients were more likely to have sustained or increased their between-visit engagement. This highlights the continued importance of telephone calls (as opposed to patient portal messages) as an approach to engage historically excluded populations 30 . Lastly, future studies should consider integrating clinical outcomes. Linking utilization care trajectories to longitudinal health outcomes (such as glycemic control, blood pressure, or hospitalizations), will further clarify whether digital care adoption translates into measurable health benefits, and for whom. Limitations This study has some limitations. First, as this is an observational study using EHR data, we cannot rule out residual confounding, particularly related to unmeasured factors such as patient motivation, access to devices and internet, and lifestyle factors or social norms that may influence attitude towards diabetes and remote engagement in care. While we included a range of clinical and sociodemographic covariates, selection into different care modalities is non-random and likely reflects underlying health engagement, care-seeking behavior, or attitudes toward different care modalities. Inclusion in the study was restricted to those who had at least one interaction with the healthcare system both pre- and post-pandemic, and results are therefore only representative of an empaneled, care-seeking population. Lastly, results are likely not generalizable to rural settings due to differences in digital access and infrastructure. Conclusion In two urban health system cohorts of adults with Type 2 diabetes analyzed over multiple years, we found that care utilization trajectories differed by sociodemographic and clinical characteristics. Specifically, in the post-pandemic era, we saw a substantial proportion of the patient population at both health systems embrace digital care (both telehealth and between-visit care), suggesting that increased digital accessibility supplements and allows patients to engage in care in ways that better suit their needs. However, a significant proportion of patients either lost access to care and experienced care disruptions or did not adopt digital care, suggesting a need for future research and tailored strategies to ensure equity in digital engagement and to close gaps in care continuity. As healthcare systems continue to expand hybrid models of care, it is critical to ensure equitable access and effectiveness across diverse populations. Future work should build upon this research and explore ways to mitigate disparities in the use of all care modalities, particularly for diverse patients and healthcare systems. Declarations Author Contributions: Conceptualization: Elaine C Khoong Data curation: Kathryn E Kemper-McIsaac, Elizabeth B Sherwin, Anna Rubinsky Formal analysis: Taylor Rapson, Kathryn E Kemper-McIsaac, Elizabeth B Sherwin Funding acquisition: Elaine C Khoong Methodology: Elizabeth B Sherwin, Lucia Pacca, Anna Rubinsky Project administration: Kathryn E Kemper-McIsaac Supervision: Elaine C Khoong Visualization: Kathryn E Kemper-McIsaac, Elizabeth B Sherwin Writing – original draft: Taylor Rapson, Kathryn E Kemper-McIsaac Writing – review & editing: All authors Funding: The following authors received this funding to support part of this work. No funders played any role in the study procedures. - Residual class settlement funds in the matter of April Krueger v. Wyeth, Inc., Case No. 03-cv-2496 (US District Court, SD of Calif. (Kemper-McIsaac; Rubinsky; Khoong) - National Institute of Nursing Research, T32NR019035 (Clifford) - National Institute of Diabetes and Digestive and Kidney Diseases, P30 DK092924 (Lyles; Khoong) - National Heart, Lung, and Blood Institute, K23HL157750 (Khoong) - National Center for Advancing Translational Sciences, KL2TR001870 (Khoong) Declarations: No authors have any competing interests as defined by Nature Portfolio, or other interests that might be perceived to influence the results and/or discussion reported in this paper. Acknowledgements Authors of this paper for sharing their code: Eisenberg-Guyot J, Peckham T, Andrea SB, Oddo V, Seixas N, Hajat A. Life-course trajectories of employment quality and health in the U.S.: A multichannel sequence analysis. Soc Sci Med. 2020 Nov;264:113327. doi: 10.1016/j.socscimed.2020.113327. References Rodriguez JA, Saadi A, Schwamm LH, Bates DW, Samal L. Disparities In Telehealth Use Among California Patients With Limited English Proficiency. Health Aff Proj Hope . 2021;40(3):487–495. doi: 10.1377/hlthaff.2020.00823 Ramsetty A, Adams C. Impact of the digital divide in the age of COVID-19. J Am Med Inform Assoc . 2020;27(7):1147–1148. doi: 10.1093/jamia/ocaa078 Robinson J, Borgo L, Fennell K, Funahashi TT. The Covid-19 Pandemic Accelerates the Transition to Virtual Care. Published online 2020. Reed M, Huang J, Graetz I, et al. Outpatient Electronic Health Records and the Clinical Care and Outcomes of Patients With Diabetes Mellitus. Ann Intern Med . 2012;157(7):482–489. doi: 10.7326/0003-4819-157-7-201210020-00004 Herrin J, Da Graca B, Nicewander D, et al. The Effectiveness of Implementing an Electronic Health Record on Diabetes Care and Outcomes. Health Serv Res . 2012;47(4):1522–1540. doi: 10.1111/j.1475-6773.2011.01370.x Harris LT, Koepsell TD, Haneuse SJ, Martin DP, Ralston JD. Glycemic Control Associated With Secure Patient-Provider Messaging Within a Shared Electronic Medical Record. Diabetes Care . 2013;36(9):2726–2733. doi: 10.2337/dc12-2003 Lyles C, Schillinger D, Sarkar U. Connecting the Dots: Health Information Technology Expansion and Health Disparities. PLOS Med . 2015;12(7):e1001852. doi: 10.1371/journal.pmed.1001852 Kruse CS, Argueta DA, Lopez L, Nair A. Patient and Provider Attitudes Toward the Use of Patient Portals for the Management of Chronic Disease: A Systematic Review. J Med Internet Res . 2015;17(2):e40. doi: 10.2196/jmir.3703 Goldzweig CL, Orshansky G, Paige NM, et al. Electronic Patient Portals: Evidence on Health Outcomes, Satisfaction, Efficiency, and Attitudes: A Systematic Review. Ann Intern Med . 2013;159(10):677–687. doi: 10.7326/0003-4819-159-10-201311190-00006 Rodriguez JA, Casillas A, Cook BL, Marlin RP. The Language of Equity in Digital Health: Prioritizing the Needs of Limited English Proficient Communities in the Patient Portal 2.0. J Health Care Poor Underserved . 2021;32:211–219. Ralston JD, Rutter CM, Carrell D, Hecht J, Rubanowice D, Simon GE. Patient Use of Secure Electronic Messaging Within a Shared Medical Record: A Cross-sectional Study. J Gen Intern Med . 2009;24(3):349–355. doi: 10.1007/s11606-008-0899-z Lyles CR, Nelson EC, Frampton S, Dykes PC, Cemballi AG, Sarkar U. Using Electronic Health Record Portals to Improve Patient Engagement: Research Priorities and Best Practices. Ann Intern Med . 2020;172(11_Supplement):S123-S129. doi: 10.7326/M19-0876 Antonio MG, Petrovskaya O, Lau F. Is research on patient portals attuned to health equity? A scoping review. J Am Med Inform Assoc . 2019;26(8–9):871–883. doi: 10.1093/jamia/ocz054 Sarkar U, Karter AJ, Liu JY, et al. Social disparities in internet patient portal use in diabetes: evidence that the digital divide extends beyond access. J Am Med Inform Assoc . 2011;18(3):318–321. doi: 10.1136/jamia.2010.006015 Deshpande N, Arora VM, Vollbrecht H, Meltzer DO, Press V. eHealth Literacy and Patient Portal Use and Attitudes: Cross-sectional Observational Study. JMIR Hum Factors . 2023;10:e40105. doi: 10.2196/40105 Ancker JS, Barrón Y, Rockoff ML, et al. Use of an Electronic Patient Portal Among Disadvantaged Populations. J Gen Intern Med . 2011;26(10):1117–1123. doi: 10.1007/s11606-011-1749-y Ancker JS, Hafeez B, Kaushal R. Socioeconomic Disparities in Adoption of Personal Health Records Over Time. Am J Manag Care . 2016;22(8):539–540. Khoong EC, Rivadeneira NA, Hiatt RA, Sarkar U. The Use of Technology for Communicating With Clinicians or Seeking Health Information in a Multilingual Urban Cohort: Cross-Sectional Survey. J Med Internet Res . 2020;22(4):e16951. doi: 10.2196/16951 Tieu L, Schillinger D, Sarkar U, et al. Online patient websites for electronic health record access among vulnerable populations: portals to nowhere? J Am Med Inform Assoc . 2017;24(e1):e47-e54. doi: 10.1093/jamia/ocw098 Gibbons MC. A Historical Overview of Health Disparities and the Potential of eHealth Solutions. J Med Internet Res . 2005;7(5):e50. doi: 10.2196/jmir.7.5.e50 Eberly LA, Kallan MJ, Julien HM, et al. Patient Characteristics Associated With Telemedicine Access for Primary and Specialty Ambulatory Care During the COVID-19 Pandemic. JAMA Netw Open . 2020;3(12):e2031640. doi: 10.1001/jamanetworkopen.2020.31640 Ryskina KL, Shultz K, Zhou Y, Lautenbach G, Brown RT. Older adults’ access to primary care: Gender, racial, and ethnic disparities in telemedicine. J Am Geriatr Soc . 2021;69(10):2732–2740. doi: 10.1111/jgs.17354 Sarkar U, Karter AJ, Liu JY, et al. The Literacy Divide: Health Literacy and the Use of an Internet-Based Patient Portal in an Integrated Health System—Results from the Diabetes Study of Northern California (DISTANCE). J Health Commun . 2010;15(sup2):183–196. doi: 10.1080/10810730.2010.499988 Davies AR, Honeyman M, Gann B. Addressing the Digital Inverse Care Law in the Time of COVID-19: Potential for Digital Technology to Exacerbate or Mitigate Health Inequalities. J Med Internet Res . 2021;23(4):e21726. doi: 10.2196/21726 Lyles CR, Harris LT, Jordan L, et al. Patient Race/Ethnicity and Shared Medical Record Use Among Diabetes Patients. Med Care . 2012;50(5):434. doi: 10.1097/MLR.0b013e318249d81b Anthony DL, Campos-Castillo C, Lim PS. Who Isn’t Using Patient Portals And Why? Evidence And Implications From A National Sample Of US Adults. Health Aff (Millwood) . 2018;37(12):1948–1954. doi: 10.1377/hlthaff.2018.05117 Lyles CR, Nguyen OK, Khoong EC, Aguilera A, Sarkar U. Multilevel Determinants of Digital Health Equity: A Literature Synthesis to Advance the Field. Annu Rev Public Health . 2023;44:383–405. doi: 10.1146/annurev-publhealth-071521-023913 Alkureishi MA, Choo ZY, Rahman A, et al. Digitally Disconnected: Qualitative Study of Patient Perspectives on the Digital Divide and Potential Solutions. JMIR Hum Factors . 2021;8(4):e33364. doi: 10.2196/33364 Nouri S, Khoong EC, Lyles CR, Karliner L. Addressing Equity in Telemedicine for Chronic Disease Management During the Covid-19 Pandemic. Catal Non-Issue Content . 2020;1(3). doi: 10.1056/CAT.20.0123 Nouri S, Lyles CR, Sherwin EB, et al. Visit and Between-Visit Interaction Frequency Before and After COVID-19 Telehealth Implementation. JAMA Netw Open . 2023;6(9):e2333944. doi: 10.1001/jamanetworkopen.2023.33944 Li P, Zhang P, Guan D, et al. Changes in racial and ethnic disparities in glucose-lowering drug utilization and glycated haemoglobin A1c in US adults with diabetes: 2005‐2018. Diabetes Obes Metab . 2023;25(2):516–525. doi: 10.1111/dom.14894 Sheon AR, Bolen SD, Callahan B, Shick S, Perzynski AT. Addressing Disparities in Diabetes Management Through Novel Approaches to Encourage Technology Adoption and Use. JMIR Diabetes . 2017;2(2):e16. doi: 10.2196/diabetes.6751 Lyles CR, Grothaus L, Reid RJ, Sarkar U, Ralston JD. Communication about diabetes risk factors during between-visit encounters. Am J Manag Care . 2012;18(12):807–815. Jorge A. Rodriguez MD, Joseph R. Betancourt MD, Thomas D. Sequist MD, Ishani Ganguli MD. Differences in the Use of Telephone and Video Telemedicine Visits During the COVID-19 Pandemic. 2021;27. Accessed August 12, 2025. https://www.ajmc.com/view/differences-in-the-use-of-telephone-and-video-telemedicine-visits-during-the-covid-19-pandemic Nath B, Williams B, Jeffery MM, et al. Trends in Electronic Health Record Inbox Messaging During the COVID-19 Pandemic in an Ambulatory Practice Network in New England. JAMA Netw Open . 2021;4(10):e2131490. doi: 10.1001/jamanetworkopen.2021.31490 Patel SY, Mehrotra A, Huskamp HA, Uscher-Pines L, Ganguli I, Barnett ML. Trends in Outpatient Care Delivery and Telemedicine During the COVID-19 Pandemic in the US. JAMA Intern Med . 2021;181(3):388–391. doi: 10.1001/jamainternmed.2020.5928 Abbott A, Tsay A. Sequence Analysis and Optimal Matching Methods in Sociology. doi: 10.1177/0049124100029001001 Emery K, Berchtold A. Comparison of two approaches in multichannel sequence analysis using the Swiss Household Panel. Longitud Life Course Stud . 2023;14(4):592–623. doi: 10.1332/175795921X16698302233894 Eisenberg-Guyot J, Peckham T, Andrea SB, Oddo V, Seixas N, Hajat A. Life-course trajectories of employment quality and health in the U.S.: A multichannel sequence analysis. Soc Sci Med . 2020;264:113327. doi: 10.1016/j.socscimed.2020.113327 Savaré L, Ieva F, Corrao G, Lora A. Mining and evaluation of patients’ diagnostic therapeutic paths through state sequences analysis. arXiv . Preprint posted online September 9, 2022. doi: 10.48550/arXiv.2209.04384 Kempa-Liehr AW, Lin CYC, Britten R, et al. Healthcare pathway discovery and probabilistic machine learning. Int J Med Inf . 2020;137:104087. doi: 10.1016/j.ijmedinf.2020.104087 Mathew S, Peat G, Parry E, Sokhal BS, Yu D. Applying sequence analysis to uncover ‘real-world’ clinical pathways from routinely collected data: a systematic review. J Clin Epidemiol . 2024;166:111226. doi: 10.1016/j.jclinepi.2023.111226 Cezard G, Sullivan F, Keenan K. Understanding multimorbidity trajectories in Scotland using sequence analysis. Sci Rep . 2022;12(1):16485. doi: 10.1038/s41598-022-20546-4 Gomez SL, Glaser SL, McClure LA, et al. The California Neighborhoods Data System: a new resource for examining the impact of neighborhood characteristics on cancer incidence and outcomes in populations. Cancer Causes Control CCC . 2011;22(4):631–647. doi: 10.1007/s10552-011-9736-5 Quan H, Sundararajan V, Halfon P, et al. Coding algorithms for defining comorbidities in ICD-9-CM and ICD-10 administrative data. Med Care . 2005;43(11):1130–1139. doi: 10.1097/01.mlr.0000182534.19832.83 Agborsangaya CB, Lau D, Lahtinen M, Cooke T, Johnson JA. Health-related quality of life and healthcare utilization in multimorbidity: results of a cross-sectional survey. Qual Life Res . 2013;22(4):791–799. doi: 10.1007/s11136-012-0214-7 Segal JB, Sen AP, Glanzberg-Krainin E, Hutfless S. Factors Associated With Overuse of Health Care Within US Health Systems: A Cross-sectional Analysis of Medicare Beneficiaries From 2016 to 2018. JAMA Health Forum . 2022;3(1):e214543. doi: 10.1001/jamahealthforum.2021.4543 Cheng Y, Goodin AJ, Pahor M, Manini T, Brown JD. Healthcare Utilization and Physical Functioning in Older Adults in the United States. J Am Geriatr Soc . 2020;68(2):266–271. doi: 10.1111/jgs.16260 Gauthier JA, Widmer ED, Bucher P, Notredame C. 1. Multichannel Sequence Analysis Applied to Social Science Data. Sociol Methodol . 2010;40(1):1–38. doi: 10.1111/j.1467-9531.2010.01227.x Gabadinho A, Ritschard G, Müller NS, Studer M. Analyzing and Visualizing State Sequences in R with TraMineR. J Stat Softw . 2011;40(4):1–37. doi: https://doi.org/10.18637/jss.v040.i04 Maechler M, Rousseeuw P, Hubert M, Hornik K. cluster: Cluster Analysis Basics and Extensions. Published online 2025. Wiroonsri N, Preedasawakul O. A correlation-based fuzzy cluster validity index with secondary options detector. arXiv . Preprint posted online March 7, 2024. doi: 10.48550/arXiv.2308.14785 Wiroonsri N. Clustering performance analysis using a new correlation-based cluster validity index. Pattern Recognit . 2024;145:109910. doi: 10.1016/j.patcog.2023.109910 Lesnard L. Using Optimal Matching Analysis in Sociology: Cost Setting and Sociology of Time. In: Blanchard P, Bühlmann F, Gauthier JA, eds. Advances in Sequence Analysis: Theory, Method, Applications . Vol 2. Life Course Research and Social Policies. Springer International Publishing; 2014:39–50. doi: 10.1007/978-3-319-04969-4 Studer M, Ritschard G. What Matters in Differences Between Life Trajectories: A Comparative Review of Sequence Dissimilarity Measures. J R Stat Soc Ser A Stat Soc . 2016;179(2):481–511. doi: 10.1111/rssa.12125 Batool F, Hennig C. Clustering with the Average Silhouette Width. Comput Stat Data Anal . 2021;158:107190. doi: 10.1016/j.csda.2021.107190 Caliński T, Harabasz J. A dendrite method for cluster analysis. Commun Stat . 1974;3(1):1–27. doi: 10.1080/03610927408827101 Institute of Medicine (US) Committee on Quality of Health Care in America. Crossing the Quality Chasm: A New Health System for the 21st Century . National Academies Press (US); 2001. Accessed August 20, 2025. http://www.ncbi.nlm.nih.gov/books/NBK222274/ Choxi H, VanDerSchaaf H, Li Y, Morgan E. Telehealth and the Digital Divide: Identifying Potential Care Gaps in Video Visit Use. J Med Syst . 2022;46(9):58. doi: 10.1007/s10916-022-01843-x Coughlin SS, Williams LB, Hatzigeorgiou C. A systematic review of studies of web portals for patients with diabetes mellitus. mHealth . 2017;3:23–23. doi: 10.21037/mhealth.2017.05.05 Khoong EC, Butler BA, Mesina O, et al. Patient interest in and barriers to telemedicine video visits in a multilingual urban safety-net system. J Am Med Inform Assoc . 2021;28(2):349–353. doi: 10.1093/jamia/ocaa234 Lee G, Chang A, Pal A, Tran TA, Cui X, Quach T. Understanding and Addressing the Digital Health Literacy Needs of Low-Income Limited English Proficient Asian American Patients. Health Equity . 2022;6(1):494–499. doi: 10.1089/heq.2022.0045 Schoen C, Osborn R, Squires D, Doty MM. Access, Affordability, And Insurance Complexity Are Often Worse In The United States Compared To Ten Other Countries. Health Aff (Millwood) . 2013;32(12):2205–2215. doi: 10.1377/hlthaff.2013.0879 Tieu L, Lyles CR, Kim HC, et al. A self-report measure of digital skills needed to use digital health tools among older adults—the Skills Measurement and Readiness Training for Digital Health (SMART Digital Health) Scale. J Am Med Inform Assoc . Published online September 13, 2025:ocaf151. doi: 10.1093/jamia/ocaf151 Manuscripts submitted for publication: Rapson T, Ramirez M, He S, Wong J, Kim HC, Luna I, Yang A, Li J, Fishman P, Ralston JD, Lyles C, Khoong E. Identifying Skill and Usability Barriers to Digital Health Tool Use among Patients in Safety Net Clinics: A Mixed Methods Study. JMIR Preprints, Under Review. doi: 10.2196/preprints.78430 Rapson T, Naranjo-Arellano M, Ralston JD, Garcia F, Green BB, Gutierrez C, Guzman-Estrada S, Khoong E, Sarkar U, Shin J, Zhou C, Fishman P, Ramirez M, Lyles C. Impact of Digital Support Interventions on Patient Portal Use Among Patients in an Urban Safety-net Health System. Submitted for publication. 2025 Tables Table 1. Patient Sociodemographic and Clinical Characteristics at UCSF and SFHN UCSF SFHN n=3,860 n=6,811 Age Category (%) 18-34 91 (2%) 162 (2%) 35-49 472 (12%) 1,080 (16%) 50-64 1,234 (32%) 3,215 (47%) 65-74 1,118 (29%) 1,727 (25%) 75+ 945 (24%) 627 (9%) Sex (%) Female 2,091 (54%) 3,581 (53%) Male 1,769 (46%) 3,229 (47%) Joint Race Eth & Language (%) Asian & English Preference 1,050 (27%) 641 (9%) Asian & Non-English Preference 478 (12%) 1,753 (26%) Black/African American & English Preference 527 (14%) 1,027 (15%) Hispanic & English Preference 366 (9%) 449 (7%) Hispanic & Non-English Preference 114 (3%) 1,881 (28%) Other/Un & English Preference 317 (8%) 254 (4%) Other/Un & Non-English Preference 67 (2%) 84 (1%) White & English Preference 941 (24%) 722 (11%) Race or Ethnicity (%) Asian 1,528 (40%) 2,394 (35%) Black/African American 527 (14%) 1,027 (15%) Hispanic/Latino 480 (12%) 2,330 (34%) Other/Unknown 384 (10%) 338 (5%) White 941 (24%) 722 (11%) Language Preference (%) Chinese 296 (8%) 1,324 (19%) English 3,154 (82%) 2,931 (43%) Other 298 (8%) 655 (10%) Spanish 112 (3%) 1,901 (28%) Insurance (%) Commercial 1,708 (44%) 71 (1%) Healthy Workers 0 (0%) 1,049 (15%) Medicaid 502 (13%) 2,709 (40%) Medicare 1,644 (43%) 2,126 (31%) Missing 5 (0%) 28 (0%) Uninsured 1 (0%) 828 (12%) Neighborhood SES (1=lowest) (%) 1 (lowest) 651 (17%) 2,561 (38%) 2 689 (18%) 1,771 (26%) 3 752 (19%) 1,049 (15%) 4 886 (23%) 889 (13%) 5 (highest) 698 (18%) 405 (6%) Missing 184 (5%) 136 (2%) Charlson Score (%) 0-2 2,965 (77%) 5,818 (85%) 3+ 895 (23%) 993 (15%) Baseline A1c Control (%) Uncontrolled A1c ( 8) 818 (21%) 2,631 (39%) Controlled A1c (<8) 2,611 (68%) 3,307 (49%) Missing 431 (11%) 873 (13%) Baseline BP Control (%) BP <=120/80 658 (17%) 1,629 (24%) BP 140/90 930 (24%) 1,205 (18%) Missing 146 (4%) 379 (6%) Active MyChart (%) No 861 (22%) 6,217 (91%) Yes 2,999 (78%) 594 (9%) Table 2. Descriptions of UCSF Healthcare Trajectories by Cluster Pre-Pandemic Utilization Cluster N (%) How Each Care Pattern Changed Relative to Pre-Pandemic Post-Pandemic Care Trajectory In-person Telehealth Between Visits High Utilizers UCSF 1 177 (4.6%) Returned to pre levels High initial uptake, moderate sustained use Continued high usage with slight increase overall Added Telehealth High Utilizers UCSF 2 333 (8.6%) Returned to almost pre levels Sustained moderate uptake Moderate uptake with slight increase overall Added Telehealth and Between-Visit Care High Utilizers UCSF 3 212 (5.5%) Did not recover to pre levels Sustained high uptake Continued high usage (no sustained increase) Transitioned to Telehealth Moderate Utilizers UCSF 4 584 (15.1%) Did not recover to pre levels Sustained moderate uptake Moderate uptake with slight increase overall Transitioned to Telehealth and Between-Visit Care Moderate Utilizers UCSF 5 173 (4.5%) Did not recover to pre levels Moderate uptake not sustained Moderate usage, decreasing trends Lost Care Low Utilizers UCSF 6 769 (19.9%) Returned to almost pre-levels Sustained moderate uptake Low uptake with increase overall Added Telehealth Low Utilizers UCSF 7 1612 (41.8%) Did not recover to pre levels Low, consistent non usage Low, consistent non usage Lost Care We defined adding digital care as in-person visit rates returning to pre-pandemic levels, while telehealth and/or between-visit care remain elevated. Those who transitioned to digital care saw in-person visit rates remain below pre-pandemic levels, while telehealth and between-visit care increased to replace in-person care. Those who lost care had minimal uptake of between-visit care or telehealth, even though in-person care rates did not rebound. We also further specified whether those who added or transitioned to digital care primarily adopted Telehealth, Between-Visit Care, or both. Table 3. Sociodemographic Characteristics of Care Trajectories at UCSF Total N (%) Added Digital Care Transitioned to Digital Care Lost Care P-value Clusters 1, 2, 6 Clusters 3, 4 Clusters 5, 7 N (%) N=3,860 1,279 (33) 796 (21) 1,785 (46) Age Category (%) 0.079 18-34 91 (2) 28 (2) 22 (3) 41 (2) 35-49 472 (12) 155 (12) 113 (14) 204 (11) 50-64 1,234 (32) 419 (33) 267 (34) 548 (31) 65-74 1,118 (29) 347 (27) 227 (29) 544 (30) 75+ 945 (24) 330 (26) 167 (21) 448 (25) Sex (%) 0.020 Female 2,091 (54) 726 (57) 440 (55) 925 (52) Male 1,769 (46) 553 (43) 356 (45) 860 (48) Joint Race Ethnicity & Language (%) <0.001 Asian & English Preference 1,050 (27) 327 (26) 150 (19) 573 (32) Asian & Non-English Preference 478 (12) 145 (11) 55 (7) 278 (16) Black/African American 527 (14) 216 (17) 141 (18) 170 (10) Hispanic & English Preference 366 (9) 109 (9) 86 (11) 171 (10) Hispanic & Non-English Preference 114 (3) 52 (4) 17 (2) 45 (3) Other/Un & English Preference 317 (8) 104 (8) 66 (8) 147 (8) Other/Un & Non-English Preference 67 (2) 31 (2) 6 (1) 30 (2) White 941 (24) 295 (23) 275 (35) 371 (21) Race or Ethnicity (%) <0.001 Asian 1,528 (40) 472 (37) 205 (26) 851 (48) Black/African American 527 (14) 216 (17) 141 (18) 170 (10) Hispanic/Latino 480 (12) 161 (13) 103 (13) 216 (12) Other/Unknown 384 (10) 135 (11) 72 (9) 177 (10) White 941 (24) 295 (23) 275 (35) 371 (21) Language Preference (%) <0.001 Chinese 296 (8) 93 (7) 25 (3) 178 (10) English 3,154 (82) 1,029 (80) 714 (90) 1,411 (79) Other 298 (8) 105 (8) 42 (5) 151 (8) Spanish 112 (3) 52 (4) 15 (2) 45 (3) Insurance (%) 0.031 Commercial 1,708 (44) 559 (44) 329 (41) 820 (46) Medicaid 502 (13) 187 (15) 120 (15) 195 (11) Medicare 1,644 (43) 531 (42) 347 (44) 766 (43) Uninsured 1 (0) 0 (0) 0 (0) 1 (0) Missing 5 (0) 2 (0) 0 (0) 3 (0) Neighborhood SES (1=lowest) (%) 0.056 1 651 (17) 221 (17) 157 (20) 273 (15) 2 689 (18) 245 (19) 144 (18) 300 (17) 3 752 (19) 232 (18) 147 (18) 373 (21) 4 886 (23) 304 (24) 158 (20) 424 (24) 5 698 (18) 219 (17) 149 (19) 330 (18) Missing 184 (5) 58 (5) 41 (5) 85 (5) Charlson Score (%) <0.001 0-2 2,965 (77) 939 (73) 539 (68) 1,487 (83) 3+ 895 (23) 340 (27) 257 (32) 298 (17) Baseline A1c Control (%) =8) 818 (21) 289 (23) 192 (24) 337 (19) Controlled A1c (<8) 2,611 (68) 849 (66) 549 (69) 1,213 (68) Missing 431 (11) 141 (11) 55 (7) 235 (13) Baseline BP Control (%) 0.001 BP <=120/80 658 (17) 220 (17) 147 (18) 291 (16) BP 140/90 930 (24) 301 (24) 206 (26) 423 (24) Missing 146 (4) 42 (3) 13 (2) 91 (5) Patient Portal Enrollment (%) <0.001 No 861 (22) 293 (23) 97 (12) 471 (26) Yes 2,999 (78) 986 (77) 699 (88) 1,314 (74) Missing SES: socioeconomic status, A1c: hemoglobin A1c lab result, BP: blood pressure *P values compare patient characteristics by care trajectory based on Chi-squared test for categorical measures Table 4. Descriptions of SFHN Healthcare Trajectories by Cluster Pre-Pandemic Utilization Cluster N (%) How Each Care Pattern Changed Relative to Pre-Pandemic Post-Pandemic Care Trajectory In-person Telehealth Between Visits High Utilizers SFHN 1 319 (4.7%) Returned to pre levels Moderate initial uptake, some sustained use Continued high usage with slight increase overall Added Telehealth and Between-Visit Care High Utilizers SFHN 2 470 (6.9%) Returned to pre levels Low uptake not sustained Moderate uptake with slight increase overall Added Between-Visit Care High Utilizers SFHN 3 285 (4.2%) Did not recover to pre levels High initial uptake, moderate sustained use Continued high usage (no sustained increase) Transitioned to Telehealth Moderate Utilizers SFHN 4 364 (5.3%) Did not recover to pre levels High initial uptake, moderate sustained use Continued high usage with slight increase overall Transitioned to Telehealth and Between-Visit Care Moderate Utilizers SFHN 5 707 (10.4%) Returned to pre levels Low uptake not sustained Low uptake with increase overall Minimal Change Moderate Utilizers SFHN 6 973 (14.3%) Did not recover to pre levels Moderate uptake not sustained Moderate usage, decreasing trends Lost Care Low Utilizers SFHN 7 3693 (54.2%) Did not recover to pre levels Low uptake not sustained Low, consistent non usage Lost Care We defined adding digital care as in-person visit rates returning to pre-pandemic levels, while telehealth and/or between-visit care remain elevated. Those who transitioned to digital care saw in-person visit rates remain below pre-pandemic levels, while telehealth and between-visit care increased to replace in-person care. Those who lost care had minimal uptake of between-visit care or telehealth, even though in-person care rates did not rebound. Those with minimal change experienced minimal changes to their care utilization patterns from before the pandemic. We also further specified whether those who added or transitioned to digital care primarily adopted Telehealth, Between-Visit Care, or both. Table 5. Sociodemographic Characteristics of Care Trajectories at SFHN Total Added Digital Care Transitioned to Digital Care Lost Care Minimal Change P-value* Clusters 1,2 Clusters 3, 4 Clusters 6, 7 Cluster 5 N (%) N=6,811 N=789 (12) N=649 (10) N=4,666 (69) N=707 (10) Age Category (%) 0.037 18-34 162 (2) 10 (1) 16 (2) 123 (3) 13 (2) 35-49 1,080 (16) 118 (15) 101 (16) 723 (15) 138 (20) 50-64 3,215 (47) 378 (48) 308 (47) 2,201 (47) 328 (46) 65-74 1,727 (25) 201 (25) 156 (24) 1,214 (26) 156 (22) 75+ 627 (9) 82 (10) 68 (10) 405 (9) 72 (10) Sex (%) 0.094 Female 3,581 (53) 394 (50) 372 (57) 2,463 (53) 352 (50) Male 3,229 (47) 395 (50) 277 (43) 2,202 (47) 355 (50) Joint Race Ethnicity & Language (%) <0.001 Asian & English Preference 641 (9) 62 (8) 62 (10) 446 (10) 71 (10) Asian & Non-English Preference 1,753 (26) 94 (12) 85 (13) 1,442 (31) 132 (19) Black/African American 1,027 (15) 116 (15) 176 (27) 659 (14) 76 (11) Hispanic & English Preference 449 (7) 57 (7) 64 (10) 276 (6) 52 (7) Hispanic & Non-English Preference 1,881 (28) 340 (43) 103 (16) 1,137 (24) 301 (43) Other/Un & English Preference 254 (4) 24 (3) 33 (5) 173 (4) 24 (3) Other/Un & Non-English Preference 84 (1) 13 (2) 5 (1) 56 (1) 10 (1) White 722 (11) 83 (11) 121 (19) 477 (10) 41 (6) Race or Ethnicity (%) <0.001 Asian 2,394 (35) 156 (20) 147 (23) 1,888 (40) 203 (29) Black/African American 1,027 (15) 116 (15) 176 (27) 659 (14) 76 (11) Hispanic/Latino 2,330 (34) 397 (50) 167 (26) 1,413 (30) 353 (50) Other/Unknown 338 (5) 37 (5) 38 (6) 229 (5) 34 (5) White 722 (11) 83 (11) 121 (19) 477 (10) 41 (6) Language Preference (%) <0.001 Chinese 1,324 (19) 52 (7) 67 (10) 1,130 (24) 75 (11) English 2,931 (43) 335 (42) 437 (67) 1,908 (41) 251 (36) Other 655 (10) 55 (7) 44 (7) 480 (10) 76 (11) Spanish 1,901 (28) 347 (44) 101 (16) 1,148 (25) 305 (43) Insurance (%) <0.001 Commercial 71 (1) 6 (1) 6 (1) 53 (1) 6 (1) Healthy Workers 1,049 (15) 44 (6) 53 (8) 875 (19) 77 (11) Medicaid 2,709 (40) 364 (46) 274 (42) 1,806 (39) 265 (37) Medicare 2,126 (31) 283 (36) 278 (43) 1,359 (29) 206 (29) Uninsured 828 (12) 92 (12) 38 (6) 546 (12) 152 (21) Missing 28 (0) 0 (0) 0 (0) 27 (1) 1 (0) Neighborhood SES (1=lowest) (%) 0.053 1 2,561 (38) 306 (39) 261 (40) 1,728 (37) 266 (38) 2 1,771 (26) 227 (29) 147 (23) 1,199 (26) 198 (28) 3 1,049 (15) 122 (15) 88 (14) 737 (16) 102 (14) 4 889 (13) 78 (10) 92 (14) 634 (14) 85 (12) 5 405 (6) 40 (5) 48 (7) 280 (6) 37 (5) Missing 136 (2) 16 (2) 13 (2) 88 (2) 19 (3) Charlson Score (%) <0.001 0-2 5,818 (85) 575 (73) 477 (73) 4,148 (89) 618 (87) 3+ 993 (15) 214 (27) 172 (27) 518 (11) 89 (13) Baseline A1c Control (%) =8) 2,631 (39) 376 (48) 272 (42) 1,677 (36) 306 (43) Controlled A1c (<8) 3,307 (49) 364 (46) 337 (52) 2,273 (49) 333 (47) Missing 873 (13) 49 (6) 40 (6) 716 (15) 68 (10) Baseline BP Control (%) <0.001 BP <=120/80 1,629 (24) 198 (25) 143 (22) 1,134 (24) 154 (22) BP 140/90 1,205 (18) 147 (19) 149 (23) 783 (17) 126 (18) Missing 379 (6) 17 (2) 6 (1) 331 (7) 25 (4) Patient Portal Enrollment (%) <0.001 No 6,217 (91) 724 (92) 524 (81) 4,293 (92) 676 (96) Yes 594 (9) 65 (8) 125 (19) 373 (8) 31 (4) SES: socioeconomic status, A1c: hemoglobin A1c lab result, BP: blood pressure *P values compare patient characteristics by care trajectory based on Chi-squared test for categorical measures Additional Declarations No competing interests reported. Supplementary Files MethodologicalAppendix.docx Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 01 Feb, 2026 Reviews received at journal 14 Jan, 2026 Reviews received at journal 24 Dec, 2025 Reviewers agreed at journal 14 Dec, 2025 Reviewers agreed at journal 11 Dec, 2025 Reviewers invited by journal 09 Dec, 2025 Editor assigned by journal 29 Nov, 2025 Submission checks completed at journal 29 Nov, 2025 First submitted to journal 21 Nov, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8176924","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":559520773,"identity":"33dae69b-fc61-46e0-ac4a-630a66ed900c","order_by":0,"name":"Taylor Rapson","email":"","orcid":"","institution":"University of Washington","correspondingAuthor":false,"prefix":"","firstName":"Taylor","middleName":"","lastName":"Rapson","suffix":""},{"id":559520776,"identity":"d4e0c3ff-4ed2-4552-a248-69c7d6855c4a","order_by":1,"name":"Kathryn E Kemper-McIsaac","email":"","orcid":"","institution":"University of California San Francisco","correspondingAuthor":false,"prefix":"","firstName":"Kathryn","middleName":"E","lastName":"Kemper-McIsaac","suffix":""},{"id":559520778,"identity":"e247ad5a-4af7-429f-bd3a-fc1324b2de1d","order_by":2,"name":"Namuun Clifford","email":"","orcid":"","institution":"The University of Texas at Austin","correspondingAuthor":false,"prefix":"","firstName":"Namuun","middleName":"","lastName":"Clifford","suffix":""},{"id":559520781,"identity":"69264290-312c-4769-87c8-e3f84763f957","order_by":3,"name":"Elizabeth B Sherwin","email":"","orcid":"","institution":"University of California San Francisco","correspondingAuthor":false,"prefix":"","firstName":"Elizabeth","middleName":"B","lastName":"Sherwin","suffix":""},{"id":559520782,"identity":"75d720f5-9a49-4ccd-a6ce-2f528c3160c6","order_by":4,"name":"Lucia Pacca","email":"","orcid":"","institution":"University of California San Francisco","correspondingAuthor":false,"prefix":"","firstName":"Lucia","middleName":"","lastName":"Pacca","suffix":""},{"id":559520783,"identity":"a1b13be8-2999-404f-a951-83664ff56bd1","order_by":5,"name":"Anna Rubinsky","email":"","orcid":"","institution":"University of California San Francisco","correspondingAuthor":false,"prefix":"","firstName":"Anna","middleName":"","lastName":"Rubinsky","suffix":""},{"id":559520784,"identity":"73edc13b-6b9a-489b-b281-c78fb8fc6b35","order_by":6,"name":"Courtney Lyles","email":"","orcid":"","institution":"University of California, Davis","correspondingAuthor":false,"prefix":"","firstName":"Courtney","middleName":"","lastName":"Lyles","suffix":""},{"id":559520786,"identity":"53125b62-1454-4b6a-a961-078755a0a5b1","order_by":7,"name":"Elaine C Khoong","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA/0lEQVRIiWNgGAWjYDACdgaGA0DKwADMqwBiZiCfB58WZhQtZ4CYLYGwFga4FsY2IrTwMzM/PFxQwWBszt587OHXeYfl5dsYGB+8bcOtRbKZzeDwjDMMZpY9x9KNZbcdNtxwjIHZcC4eLQaHgYi3jcHG4EaOmbTktsMJBvINbNK8eLTYH2b/cJj3H0zLnMMJQIex/8anxYCZB2hLA4MZSIvkx4bDCQzHGNiY8WmROMxTcJjnmISxwZljadIMx9KBfmFslpxzDrcW/vb2zZ95amwMNxxvPib5o8YaGGLMBz+8KcOtBWYZmGSGRAdjA0H1cMD4g3i1o2AUjIJRMIIAALtjTEXVsDKcAAAAAElFTkSuQmCC","orcid":"","institution":"University of California San Francisco","correspondingAuthor":true,"prefix":"","firstName":"Elaine","middleName":"C","lastName":"Khoong","suffix":""}],"badges":[],"createdAt":"2025-11-21 23:23:10","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8176924/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8176924/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":98245679,"identity":"4a9d8afc-1cbc-42b8-9048-84b25dce9ca7","added_by":"auto","created_at":"2025-12-15 16:18:17","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1704008,"visible":true,"origin":"","legend":"","description":"","filename":"TrajectoriesofUtilizationASequenceAnalysisManuscript20251126CLEAN.docx","url":"https://assets-eu.researchsquare.com/files/rs-8176924/v1/6fb0a4d41b8381a99703121f.docx"},{"id":98245671,"identity":"0afd4bec-6bd6-4635-ae1b-67a1cc46b40d","added_by":"auto","created_at":"2025-12-15 16:18:16","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":8579,"visible":true,"origin":"","legend":"","description":"","filename":"7b5b6191bb23405db63c5bd8670c5ee3.json","url":"https://assets-eu.researchsquare.com/files/rs-8176924/v1/1048bf822597c7466708cfbb.json"},{"id":98245733,"identity":"51ba3f93-f202-419f-99a8-e518ce2b7671","added_by":"auto","created_at":"2025-12-15 16:18:20","extension":"xml","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":365134,"visible":true,"origin":"","legend":"","description":"","filename":"7b5b6191bb23405db63c5bd8670c5ee31enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-8176924/v1/ee88dedb0a0e22790a4f3cd6.xml"},{"id":98245808,"identity":"543eb76f-7e2a-4330-b15c-daf6b4ad3634","added_by":"auto","created_at":"2025-12-15 16:18:25","extension":"jpeg","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":231712,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8176924/v1/5e14072cb11a8651dd953746.jpeg"},{"id":98245704,"identity":"2c38e33e-b07e-4270-b71b-4dfc38eaf18a","added_by":"auto","created_at":"2025-12-15 16:18:18","extension":"jpeg","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":231460,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8176924/v1/e7ce875841c45315256ae8f8.jpeg"},{"id":98245817,"identity":"80535a70-87f0-416f-aaf5-64e3386eb5fa","added_by":"auto","created_at":"2025-12-15 16:18:26","extension":"jpeg","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1233812,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8176924/v1/33773881a9e03a431e22d5d2.jpeg"},{"id":98245794,"identity":"23140017-08f1-4b7c-8770-992c97e19fe2","added_by":"auto","created_at":"2025-12-15 16:18:24","extension":"jpeg","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":312908,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage4.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8176924/v1/70830ee5805f46f75f543fda.jpeg"},{"id":98245565,"identity":"1d6793c5-2073-4839-bcba-68340a046232","added_by":"auto","created_at":"2025-12-15 16:18:06","extension":"jpeg","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":286870,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage5.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8176924/v1/3f8f96e266413e4d00347e13.jpeg"},{"id":98245513,"identity":"25dd3a72-e8f5-4c19-baea-5f6a8c7dfce3","added_by":"auto","created_at":"2025-12-15 16:18:00","extension":"jpeg","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":287306,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage6.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8176924/v1/c86607bfc83f83eef5171029.jpeg"},{"id":98245702,"identity":"ec983339-56e2-4156-acea-74963e622ddf","added_by":"auto","created_at":"2025-12-15 16:18:18","extension":"jpeg","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":216126,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage7.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8176924/v1/28032646bffd8a72aa4234df.jpeg"},{"id":98245700,"identity":"a295e76d-d989-4de8-b390-bb65ebceb5d1","added_by":"auto","created_at":"2025-12-15 16:18:18","extension":"jpeg","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":215228,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage8.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8176924/v1/427a3e3e9ca90dfc1bc98097.jpeg"},{"id":98245559,"identity":"d13e5372-e837-471d-a4be-d74668855a91","added_by":"auto","created_at":"2025-12-15 16:18:05","extension":"png","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":79133,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8176924/v1/a5b88779f209c9d84a506c04.png"},{"id":98245547,"identity":"8811b85a-5a32-4730-b40a-e4d82f44a998","added_by":"auto","created_at":"2025-12-15 16:18:03","extension":"png","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":80481,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-8176924/v1/4b32832bde2af5e76002989b.png"},{"id":98245654,"identity":"124ffc5a-fc9e-475b-ab28-07aa39f83383","added_by":"auto","created_at":"2025-12-15 16:18:14","extension":"png","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":10257,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-8176924/v1/7a7d937bb9e486d3a31391b6.png"},{"id":98245739,"identity":"fa05f1bc-d960-427d-925c-15b80bab8170","added_by":"auto","created_at":"2025-12-15 16:18:20","extension":"png","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":57100,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-8176924/v1/b1a921745da9c4a90e4733e4.png"},{"id":98245815,"identity":"6d268497-6c81-4492-9a56-99022b0c4343","added_by":"auto","created_at":"2025-12-15 16:18:25","extension":"png","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":139531,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-8176924/v1/2d857d9d1ce4ae03eb4579a4.png"},{"id":98245737,"identity":"ce94cabb-982a-4c87-96bb-72a1ce1c7912","added_by":"auto","created_at":"2025-12-15 16:18:20","extension":"png","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":148686,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-8176924/v1/0d424a82ad60a4c10eb78857.png"},{"id":98245811,"identity":"311cbcb4-05a1-4fb3-a2e6-472315e8057d","added_by":"auto","created_at":"2025-12-15 16:18:25","extension":"png","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":71409,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-8176924/v1/97c679a8482f1f03766de7c6.png"},{"id":98245672,"identity":"da403a0b-df1f-40f5-a523-e49b59609150","added_by":"auto","created_at":"2025-12-15 16:18:17","extension":"png","order_by":18,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":71541,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage8.png","url":"https://assets-eu.researchsquare.com/files/rs-8176924/v1/67c6cd1097957a6f16df54ca.png"},{"id":98245568,"identity":"7482ace4-106f-417b-b1de-6747adea42b9","added_by":"auto","created_at":"2025-12-15 16:18:06","extension":"xml","order_by":19,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":371330,"visible":true,"origin":"","legend":"","description":"","filename":"7b5b6191bb23405db63c5bd8670c5ee31structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-8176924/v1/992f2ac96971b2b61c9abbe9.xml"},{"id":98245673,"identity":"5693310e-171f-4cf1-8640-13c7cbd6b7fb","added_by":"auto","created_at":"2025-12-15 16:18:17","extension":"html","order_by":20,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":382349,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-8176924/v1/bbfc02b8497b0f1e689707e3.html"},{"id":98245745,"identity":"5495d57f-f124-411f-a04e-c740b1fe17cf","added_by":"auto","created_at":"2025-12-15 16:18:20","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":514121,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eState distribution plots of In-Person, Telehealth, and Between-Visit Care Clusters at UCSF\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe state distribution plot shows the proportion of individuals in each care state at each time point. The x-axis represents time (e.g., months or encounters), and the stacked colored areas indicate the relative frequency of each state. Changes in the height of each colored band over time reflect shifts in the overall composition of care states across the cohort, allowing visualization of dominant modalities and temporal trends in care engagement.\u003c/p\u003e\n\u003cp\u003eHU: High Utilizer, MU: Moderate Utilizer, LU: Low Utilizer, TH: Telehealth, BVC: Between-Visit Care Interactions\u003c/p\u003e","description":"","filename":"Picture1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8176924/v1/b4211de516f9bf82a98131c8.jpg"},{"id":98245813,"identity":"afca2471-28e7-4f0c-ae8c-ff888a01cd1d","added_by":"auto","created_at":"2025-12-15 16:18:25","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":520274,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003ePlots of In-Person, Telehealth, and Between-Visit Care Clusters at SFHN\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe state distribution plot shows the proportion of individuals in each care state at each time point. The x-axis represents time (e.g., months or encounters), and the stacked colored areas indicate the relative frequency of each state. Changes in the height of each colored band over time reflect shifts in the overall composition of care states across the cohort, allowing visualization of dominant modalities and temporal trends in care engagement.\u003c/p\u003e\n\u003cp\u003eHU: High Utilizer, MU: Moderate Utilizer, LU: Low Utilizer, TH: Telehealth, BVC: Between-Visit Care Interactions\u003c/p\u003e","description":"","filename":"Picture2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8176924/v1/d36334ff55fa76f74489d9de.jpg"},{"id":98246498,"identity":"3de1e3e5-da2c-4a19-a3cd-b802a68f016d","added_by":"auto","created_at":"2025-12-15 16:19:08","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3172568,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8176924/v1/dd367788-ac54-447b-bbb0-337ee7d63dcc.pdf"},{"id":98245371,"identity":"cf21e070-87a5-427c-9998-45086d585790","added_by":"auto","created_at":"2025-12-15 16:17:50","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":3539081,"visible":true,"origin":"","legend":"","description":"","filename":"MethodologicalAppendix.docx","url":"https://assets-eu.researchsquare.com/files/rs-8176924/v1/66cbc6729d0b469112642f8b.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Trajectories of In-Person and Remote Care Utilization: A Sequence Analysis of Primary Care Modalities Pre- and Post-COVID-19","fulltext":[{"header":"Introduction","content":"\u003cp\u003eThe COVID-19 pandemic spurred the growth of telehealth (scheduled video and/or telephone encounters) and between-visit care, such as unscheduled telephone visits and secure messaging in the patient portal, as alternative modalities for receiving healthcare\u003csup\u003e\u003cspan additionalcitationids=\"CR2\" citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u003c/sup\u003e. Use of digital health and personal devices to engage in remote care has been linked to improved quality of care and clinical outcomes, particularly among individuals with chronic disease\u003csup\u003e\u003cspan additionalcitationids=\"CR5 CR6 CR7 CR8\" citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e. Evidence suggests that using the patient portal to send secure messages, request refills, and schedule appointments is linked to patient empowerment and engagement in their care, as well as improved patient-provider communication\u003csup\u003e\u003cspan additionalcitationids=\"CR7 CR8 CR9 CR10 CR11\" citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e. Similarly, use of the patient portal and telehealth has been shown to have a positive effect on self-management behaviors and clinical outcomes across chronic conditions, such as improved symptom control and medication adherence, which are critical drivers of health\u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e,\u003cspan additionalcitationids=\"CR7\" citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e,\u003cspan additionalcitationids=\"CR14 CR15\" citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eDespite the well-documented increase in the use of telehealth visits and between-visit care as additional access points for engagement with primary care clinical teams in recent years, the longitudinal healthcare utilization patterns comprising multiple modalities of care and their differences across sociodemographic groups are understudied. Prior research has documented unequal uptake of digital healthcare across racial and ethnic groups, language preference, age, and gender, as well as differences in use for those facing barriers to accessing broadband internet, devices, or data (e.g., due to cost)\u003csup\u003e\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e,\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e,\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e,\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e,\u003cspan additionalcitationids=\"CR18 CR19 CR20 CR21 CR22 CR23 CR24 CR25\" citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e. In particular, older adults, especially those with non-English language preferences or who identify as a racial or ethnic minority, experience significant disparities in digital health adoption due to disparities in device access, data access, or digital literacy\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e,\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e,\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e,\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u003c/sup\u003e. Together, these barriers create a \u0026ldquo;digital divide\u0026rdquo; that may worsen inequities in healthcare settings that increasingly rely on digital health tools, especially in systems serving low-income, diverse populations\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e,\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e,\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e,\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e,\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e. As digital health becomes further embedded in healthcare delivery, these barriers to the uptake and use of digital health tools may ultimately shape healthcare utilization and outcomes for marginalized populations\u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eThese concerns are particularly salient for patients with Type 2 Diabetes Mellitus (T2DM), one of the most common chronic conditions and one that disproportionately affects marginalized populations. Patients with uncontrolled diabetes and complications are disproportionately represented in minoritized racial and ethnic groups and lower socioeconomic neighborhoods, highlighting the need for equitable care modalities\u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e,\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e,\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e\u003c/sup\u003e. Because effective diabetes management requires ongoing monitoring, communication, and care coordination, patients with T2DM may benefit substantially from expanded modalities for accessing care. Remote tools can help address barriers to in-person care, support continuity between visits, and encourage better disease self-management. Yet, given persistent disparities in both digital health adoption and diabetes control, it remains uncertain whether the expansion of telehealth and patient portal use during the pandemic has narrowed or widened existing inequities.\u003c/p\u003e\u003cp\u003eUnderstanding healthcare utilization patterns is, therefore, critical\u0026mdash;not only to evaluate how patients with T2DM and other chronic conditions engage with multiple modalities of care, but also to identify where gaps in care delivery may exacerbate disparities. However, most prior studies rely on repeated cross-sectional data, examining a single care modality and classifying individuals as either \u0026ldquo;high\u0026rdquo; or \u0026ldquo;low\u0026rdquo; utilizers of care\u003csup\u003e\u003cspan additionalcitationids=\"CR34 CR35\" citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u003c/sup\u003e. Such designs capture snapshots rather than trajectories, limiting the ability to understand the multifaceted nature of healthcare utilization and how it evolves over time, particularly in an era of expanded digital health use\u003csup\u003e\u003cspan additionalcitationids=\"CR3\" citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u003c/sup\u003e. In contrast, a longitudinal, sequence-based approach can reveal how individuals transition between in-person, telehealth, and between-visit modalities, illuminating dynamic patterns that reflect real-world care experiences. Identifying patterns of underutilization across modalities can further pinpoint unmet needs and opportunities to improve equity. For example, a recent study highlighted that when telehealth, between-visit care, and in-person care were available to patients, there were fewer differences in utilization by race and ethnicity and preferred language, suggesting that multiple options for care utilization can reduce disparities\u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e. Still, research remains limited on how these alternative access points to care can alleviate barriers to care and interact over time to supplement traditional care-seeking patterns\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e,\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eSequence analysis is a data-driven approach that compares and groups longitudinal trajectories of events, making it well-suited for evaluating healthcare utilization patterns, as it preserves the timing, order, and duration of event\u003csup\u003e\u003cspan additionalcitationids=\"CR38\" citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e. While studies often summarize utilization as aggregate counts over fixed windows, multichannel sequence analysis can reveal when and in what order patients use in-person, telehealth, and between-visit care, identifying recognizable patterns (for example, \u0026ldquo;transitioned to digital\u0026rdquo; or \u0026ldquo;reduced care engagement\u0026rdquo;\u003csup\u003e39\u0026ndash;41\u003c/sup\u003e. This moves beyond simple categorizations of \u0026ldquo;high\u0026rdquo; or \u0026ldquo;low\u0026rdquo; utilizers and can surface nuanced variation in utilization patterns. The ability to cluster patients based on the timing, order, and duration of their healthcare encounters can provide a more granular understanding of patient heterogeneity and reveal hidden patterns and subgroups\u003csup\u003e\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e,\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eThis study advances the existing research by employing multichannel sequence analysis and cluster analysis to create clusters of patients with similar temporal sequences of in-person, between-visit, and telehealth interactions across two health systems. We examine sociodemographic differences between clusters of healthcare utilization to assess whether specific patient characteristics are disproportionately represented in certain clusters. By comparing clusters across two health systems, we aim to 1) identify heterogeneities in multi-modality healthcare utilization over time, 2) assess whether clusters of care utilization are patterned by socio-demographic factors, which could suggest potential inequities in access and utilization, and 3) assess whether care utilization patterns are consistent across health systems or vary by site, potentially reflecting differences in patient populations, resources, and care delivery models. Given the potential role of remote care engagement to either mitigate or perpetuate existing disparities, the evidence from these comparisons can shape policies that promote equitable integration of digital health across diverse healthcare settings.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003eStudy Setting and Data Source\u003c/p\u003e\u003cp\u003eWe conducted a 4-year retrospective longitudinal cohort study of adults with Type 2 Diabetes Mellitus (T2DM) receiving care at two health systems in San Francisco: the University of California, San Francisco (UCSF) and the San Francisco Health Network (SFHN). UCSF is an academic tertiary care center with four primary care practices. SFHN is an integrated, urban safety-net health system with 14 primary care practices. The study period was April 1, 2019 \u0026ndash; March 31, 2023 at UCSF and October 1, 2019 \u0026ndash; March 31, 2023 at SFHN. The difference in start date was due to the adoption of a new EHR software in August 2019 at SFHN.\u003c/p\u003e\u003cp\u003eWe used electronic health records (EHR) from the two health systems, capturing sociodemographic data, clinical characteristics, and encounter data. Since the study procedures involved only a review of data collected for routine clinical care, informed consent was waived due to minimal risk to patients in the cohort. All study procedures were approved by the UCSF Institutional Review Board (IRB #20-31253) in accordance with the Declaration of Helsinki. We report our procedures and results in accordance with the STROBE guidelines.\u003c/p\u003e\u003cp\u003eStudy Sample\u003c/p\u003e\u003cp\u003eWe identified a cohort of adult patients \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\ge\\:\\)\u003c/span\u003e\u003c/span\u003e18 years old with T2DM who were assigned a primary care provider as of April 1, 2019, at UCSF and SFHN. The creation of the cohort is described in detail in a previous study\u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e (Methodological Appendix). In brief, patients with T2DM were identified based on the following criteria before the study start date: an active DM health maintenance modifier; having a T2DM-related ICD-9/ICD-10 code on their active problem list; the most recent HbA1c test result \u0026ge;6.5%, and/or having an active diabetes medication on their medication list. To ensure active engagement in care, patients were only included if they had at least one encounter with their diabetes (primary care or endocrinology) team between April 1, 2019 \u0026ndash; March 31, 2021, and at least one healthcare interaction (including patient portal messages, phone calls) in the last 12 months of the study period (April 1, 2022 \u0026ndash; March 31, 2023).\u003c/p\u003e\u003cp\u003eExposure variables\u003c/p\u003e\u003cp\u003eWe included a comprehensive set of patient-level sociodemographic characteristics as of 4/1/2019 that were extracted from the EHR. We included age (as a continuous variable and categorized as 18\u0026ndash;34, 35\u0026ndash;49, 50\u0026ndash;64, 65\u0026ndash;74, or 75+), race and ethnicity (categorized as Hispanic/Latino(a); non-Hispanic Asian; non-Hispanic Black; non-Hispanic White/Caucasian; or other non-Hispanic races, which included American Indian or Alaskan Native, Native Hawaiian or Other Pacific Islander), preferred language (English, Chinese, Spanish, or Other), sex (male/female), health insurance as of April 2019 (Commercial/Private, Medicare, Medicaid, Healthy Workers [an SF-specific program that provides health insurance coverage to people working in In-Home Supportive Services, foster care, or other publicly funded roles who do not qualify for traditional Medicaid or employer-sponsored coverage], or Uninsured), and neighborhood socioeconomic status (nSES) based on geocoded addresses, and categorized as quintiles from low to high\u003csup\u003e\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e\u003c/sup\u003e. Further details about insurance status and nSES are included in the Methodological Appendix.\u003c/p\u003e\u003cp\u003eClinical characteristics included the Charlson comorbidity index (CCI, modified to exclude diabetes) from 03/01/2019-02/28/2020, baseline A1c as of 04/01/2019, and baseline blood pressure control as of 04/01/2019. We dichotomized CCI as 0\u0026ndash;2 or 3\u0026thinsp;+\u0026thinsp;to represent the severity of disease comorbidities\u003csup\u003e\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e\u003c/sup\u003e. Baseline A1c control was categorized as: controlled A1c (A1c \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\le\\:\\)\u003c/span\u003e\u003c/span\u003e8), uncontrolled A1c (A1c \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\u0026gt;\\)\u003c/span\u003e\u003c/span\u003e8), or missing. Baseline blood pressure was categorized as: \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\le\\:\\)\u003c/span\u003e\u003c/span\u003e120/80 (no hypertension), \u0026gt;120/80 and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\le\\:\\)\u003c/span\u003e\u003c/span\u003e140/90 (controlled hypertension), or \u0026gt;140/90 mmHg (uncontrolled hypertension). Patient portal enrollment was included as a proxy measure of digital access. Patient portal enrollment status was extracted from the EHR as of 4/1/2020 and documented as active or inactive (yes/no).\u003c/p\u003e\u003cp\u003eConstruction of Care Trajectories and Cluster Analysis\u003c/p\u003e\u003cp\u003eWe conducted sequence and cluster analysis to group patients into homogeneous clusters of healthcare utilization over time. This approach enabled us to measure the similarity of each patient\u0026rsquo;s sequence for each encounter modality across quarters and cluster similar sequences together\u003csup\u003e\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e. For example, a patient whose sequence includes frequent in-person visits early in the study period followed by a shift to telehealth visits may be grouped in a cluster representing patients who transitioned from traditional to digital care modalities. To account for multiple modalities (in-person visits, telehealth visits, and between-visit interactions), we employed multichannel (also called multidimensional) sequence analysis, which allows creating three distinct channels, one for each encounter type\u003csup\u003e\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e\u003c/sup\u003e. Analyses were conducted in R Version 4.2.1 using the \u0026ldquo;TraMineR,\u0026rdquo; \u0026ldquo;cluster,\u0026rdquo; and \u0026ldquo;UniversalCVI\u0026rdquo; packages\u003csup\u003e\u003cspan additionalcitationids=\"CR48 CR49\" citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\n\u003ch3\u003eStep 1: Building individual sequences across three channels\u003c/h3\u003e\n\u003cp\u003eWe summarized care utilization across three distinct types of patient encounters with the healthcare team (comprised of physicians, advanced practice providers, nurses, social workers, pharmacists, dietitians, medical assistants, or embedded behavioral health clinicians):\u003c/p\u003e\u003cp\u003e\u003col\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003e\u003cem\u003eIn-person visits\u003c/em\u003e: scheduled in-person visits with a member of the care team.\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003e\u003cem\u003eTelehealth visits\u003c/em\u003e: scheduled video or telephone visits with a member of the care team.\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003e\u003cem\u003eBetween-visit interactions\u003c/em\u003e: care that occurred between visits, documented in the EHR as unscheduled telephone calls or patient portal messages.\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003c/ol\u003e\u003cdiv class=\"BlockQuote\"\u003e\u003cp\u003eFor each patient, quarterly counts of encounters for each modality were grouped into ordered categories based on utilization levels. These categories were designed to capture meaningful differences in utilization while maintaining sufficient sample sizes within each category (details in Methodological Appendix). In both health systems, Q2 2020 (April\u0026ndash;June 2020) was excluded from the analysis due to the unique encounter patterns during the COVID-19 Shelter-in-Place period (further detailed in Appendix B). For each modality, quarterly counts were binned into the following ordered categories:\u003c/p\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e\u003col\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003e\u003cem\u003eIn-person visits\u003c/em\u003e: UCSF: 0, 1, 2, 3+; SFHN: 0, 1, 2, 3\u0026ndash;5, 6+\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003e\u003cem\u003eTelehealth visits\u003c/em\u003e: UCSF: 0, 1, 2+; SFHN: 0, 1, 2\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003e\u003cem\u003eBetween-visit interactions\u003c/em\u003e: UCSF: 0,1,2,3,4\u0026ndash;6,7+; SFHN: 0, 1, 2,3\u0026ndash;4, 5+\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003c/ol\u003e\u003c/p\u003e\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003eStep 2: Sequence analysis to quantify differences between trajectories\u003c/h2\u003e\u003cp\u003eAfter building the individual sequences, we then used multichannel sequence analysis to compare patients\u0026rsquo; trajectories over time, computing differences between sequences for each modality using the dynamic Hamming distances (DHD) method\u003csup\u003e\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e\u003c/sup\u003e (further detailed in Methodological Appendix). This approach allowed us to assess whole-person care patterns over time and identify clusters of patients with similar utilization trajectories.\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003eStep 3: Cluster analysis to group similar trajectories\u003c/h3\u003e\n\u003cp\u003eWe grouped patients into healthcare utilization clusters using agglomerative hierarchical clustering with Ward\u0026rsquo;s linkage. Cluster quality was evaluated using indicators such as the Average Silhouette Width (ASW), the Hubert\u0026rsquo;s C index (HC), and the Calinski\u0026ndash;Harabasz (CH) index to identify cluster solutions with high homogeneity within clusters and distinction between clusters (see Appendix C for more details). Since several cluster solutions may have similar quality, we also used expert judgement to make considerations on parsimony and salience and choose the final number of clusters\u003csup\u003e\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eTo ease our comparison of clusters within and between health systems, we conducted post-hoc analyses to categorize clusters based on similar overall trends in utilization. We grouped clusters into categories based on their trajectories after the shelter-in-place mandate was lifted, relative to their pre-pandemic behavior. We further characterized clusters by pre-pandemic in-person utilization levels.\u003c/p\u003e\u003cp\u003eStatistical analyses\u003c/p\u003e\u003cp\u003eWe described patient characteristics overall and by healthcare utilization trajectory. In addition to the exposure variables listed above, we included an interaction term between race and ethnicity and preferred language to capture the intersectional and integrated nature of these two variables. Categorical variables were compared using chi-squared tests. Analyses were conducted separately for each health system, with cross-system comparisons of care trajectories and pre-pandemic utilization presented descriptively.\u003c/p\u003e\u003cp\u003eSensitivity analyses\u003c/p\u003e\u003cp\u003eSensitivity analyses were conducted to assess the robustness of the clustering approach, ensuring the stability of cluster solutions under different assumptions and methodologies. We repeated the sequence and cluster analysis using an alternative sequence dissimilarity measure and compared the resulting cluster solutions and interpretations.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003eFirst, we separately outlined the visit trajectories within each healthcare system, given their unique patient populations and different uptake of telehealth and between-visit care during and post-pandemic.\u003c/p\u003e\n\u003ch3\u003eUCSF\u003c/h3\u003e\n\u003cp\u003ePatient Characteristics\u003c/p\u003e\u003cp\u003eAfter excluding patients without a visit in the last 12 months of the study period, we identified a cohort of 3,860 patients with T2DM at UCSF (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Over half (53%) of patients were aged 65+. Most patients were English-speaking (82%), and the cohort included high proportions of Asian (40%, including 12% with non-English preference) and White (24%) patients. Patients were relatively evenly distributed across nSES quintiles, with 17% residing in the lowest and 18% in the highest quintile. Most patients in the cohort had insurance coverage through Commercial/private (44%) and Medicare plans (43%). Almost a quarter (23%) of patients had high comorbidity scores (CCI 3+), and nearly a third had uncontrolled A1c at baseline. Digital engagement was high, with 78% of patients having an active MyChart account at baseline.\u003c/p\u003e\u003cp\u003eDescription of Care Utilization Clusters\u003c/p\u003e\u003cp\u003eWe identified seven distinct care utilization clusters as the final number of clusters within the UCSF diabetes cohort, representing diverse patterns of engagement with in-person, telehealth, and between-visit care modalities over time. Figure\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e displays the distribution of each care modality at each time point by cluster. Index plots showing all the individual trajectories in each cluster are displayed in Appendix E.1.\u003c/p\u003e\u003cp\u003e\u003col\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003e\u003cb\u003eHigh Utilizers who added Telehealth\u003c/b\u003e (n\u0026thinsp;=\u0026thinsp;177): These patients had very high in-person and between-visit care before the pandemic. About a year after the shelter-in-place mandate, they returned to very high in-person care and sustained high levels of between-visit care alongside moderate telehealth uptake.\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003e\u003cb\u003eHigh utilizers who added Telehealth and Between-Visit care\u003c/b\u003e (n\u0026thinsp;=\u0026thinsp;333): Patients in this cluster had high in-person and moderate between-visit care before the pandemic. Following the shelter-in-place mandate, they returned to prior in-person care levels, sustained moderate increases in telehealth, and increased between-visit care.\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003e\u003cb\u003eHigh utilizers who transitioned to Telehealth\u003c/b\u003e (n\u0026thinsp;=\u0026thinsp;212): These patients had high in-person, very high between-visit care, and some telehealth use before the pandemic. Post-shelter-in-place, they shifted to high telehealth and between-visit care without returning to prior in-person rates.\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003e\u003cb\u003eModerate utilizers who transitioned to Telehealth\u003c/b\u003e (n\u0026thinsp;=\u0026thinsp;584): These patients had moderate in-person and between-visit care before the pandemic. Post-shelter-in-place, they shifted to moderate telehealth and increased between-visit care without returning to prior in-person rates.\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003e\u003cb\u003eModerate utilizers who lost care\u003c/b\u003e (n\u0026thinsp;=\u0026thinsp;173): These patients had moderate in-person and between-visit care before the pandemic but did not recover either modality after the shelf-in-place period, and telehealth uptake was minimal.\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003e\u003cb\u003eLow utilizers who added Telehealth\u003c/b\u003e (n\u0026thinsp;=\u0026thinsp;769): These patients had low in-person and between-visit care before the pandemic. Post-shelter-in-place, they returned nearly to prior in-person rates, while adding telehealth and slightly increasing between-visit care.\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003e\u003cb\u003eLow utilizers who lost care\u003c/b\u003e (n\u0026thinsp;=\u0026thinsp;1,612): These patients had low utilization across all modalities before and after the shelter-in-place mandate. They did not return to pre-pandemic in-person rates, demonstrated minimal telehealth uptake, and had no between-visit care.\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003c/ol\u003e\u003c/p\u003e\u003cp\u003eSociodemographic Characteristics by Post-Pandemic Care Trajectories\u003c/p\u003e\u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e details the seven clusters classified by pre-pandemic utilization and post-pandemic changes in in-person, telehealth, and between-visit care trajectories over time. For comparisons of sociodemographic and clinical characteristics, clusters were grouped into three post-pandemic trajectories: added digital care (telehealth and/or between-visit care), transitioned to digital care (telehealth and/or between-visit care), and lost care. Among the 3,860 patients, 33% added digital care, 21% transitioned to digital care, and 46% lost care. Sociodemographic composition varied significantly across trajectories (Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e, full descriptive tables for each cluster are presented in Appendix D.1). Patients who added digital care were more likely to be female (p\u0026thinsp;=\u0026thinsp;.02) and Black/African American (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), or Hispanic with non-English language preference (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) compared to patients who lost care or transitioned to digital care. These patients were also slightly more likely to be covered by Medicaid (p\u0026thinsp;=\u0026thinsp;.031), have high comorbidity scores (CCI 3+) (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), and uncontrolled A1c at baseline (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). Patients who transitioned to digital care were more likely to be White, Black/African American, or Hispanic with an English-language preference (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), and be covered by Medicaid or Medicare (p\u0026thinsp;=\u0026thinsp;.031) compared to patients in other care trajectories. Individuals who transitioned to digital care also had higher comorbidity scores (CCI 3+) (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), were less likely to have missing A1c or blood pressure values (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001 and p\u0026thinsp;=\u0026thinsp;0.001, respectively) and were more likely to have an active patient portal at baseline (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). Patients who lost care were more likely to be male (p\u0026thinsp;=\u0026thinsp;0.02), Asian - particularly with a non-English language preference (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), and commercially insured (p\u0026thinsp;=\u0026thinsp;.031) compared to patients in other care trajectories. Patients who lost care were also more likely to have lower comorbidity scores (CCI 0\u0026ndash;2) (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), missing baseline A1c or blood pressure results (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001 and p\u0026thinsp;=\u0026thinsp;0.001, respectively), and low patient portal enrollment (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001).\u003c/p\u003e\n\u003ch3\u003eSFHN\u003c/h3\u003e\n\u003cp\u003ePatient Characteristics\u003c/p\u003e\u003cp\u003eAfter excluding patients without a visit in the last 12 months of the study period, we identified a cohort of 6,811 patients with diabetes at SFHN (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). The SFHN diabetes cohort was younger than the UCSF cohort, with two-thirds of patients under the age of 65. Most patients at this health system were non-English speakers (57%), including 28% identifying as Hispanic and Spanish preference and 26% identifying as Asian and Chinese preference. Most patients were insured through Medicaid (40%), Medicare (31%), or Healthy Workers (15%), and nearly two-thirds (64%) resided in neighborhoods within the lowest two nSES quintiles. While 85% of patients had lower comorbidity scores (CCI\u0026thinsp;\u0026lt;\u0026thinsp;3), 39% had uncontrolled A1c at baseline. Digital engagement was notably low, with less than 10% of patients having an active MyChart account at baseline.\u003c/p\u003e\u003cp\u003eDescription of Care Utilization Clusters\u003c/p\u003e\u003cp\u003eWe identified seven care utilization clusters as the final number of clusters within the SFHN cohort, described below. Figure\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e displays the distribution of each care modality at each time point by cluster. Index plots showing all the individual trajectories are displayed in Appendix E.2.\u003c/p\u003e\u003cp\u003e\u003col\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003e\u003cb\u003eHigh utilizers who added Telehealth and Between-Visit care\u003c/b\u003e (n\u0026thinsp;=\u0026thinsp;319): These patients had high in-person and between-visit care prior to the pandemic. Within one year of the shelter-in-place mandate, they returned to pre-pandemic in-person care levels while adding more between-visit care and moderate telehealth uptake.\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003e\u003cb\u003eHigh utilizers who added Between-Visit care\u003c/b\u003e (n\u0026thinsp;=\u0026thinsp;470): These patients had high in-person and moderate between-visit care before the pandemic. A year after the shelter-in-place mandate was lifted, they returned to prior in-person rates, while increasing between-visit care. Telehealth uptake was minimal and not sustained over time.\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003e\u003cb\u003eHigh utilizers who transitioned to Telehealth\u003c/b\u003e (n\u0026thinsp;=\u0026thinsp;285): These patients had high in-person and between-visit care with some telehealth use before the pandemic. Post-shelter-in-place, they did not return to pre-pandemic in-person care levels but maintained high between-visit care and had high adoption of telehealth.\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003e\u003cb\u003eModerate utilizers who transitioned to Telehealth and Between-Visit care\u003c/b\u003e (n\u0026thinsp;=\u0026thinsp;364): These patients had moderate in-person and high between-visit engagement prior to the pandemic. Post-shelter-in-place, they did not return to pre-pandemic in-person levels but had high telehealth uptake and increased between-visit care.\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003e\u003cb\u003eModerate utilizers with minimal change\u003c/b\u003e (n\u0026thinsp;=\u0026thinsp;707): These patients had moderate in-person and low between-visit care prior to the pandemic. After the shelter-in-place mandate lifted, their in-person care levels nearly returned to pre-pandemic levels, with slight telehealth uptake and between-visit engagement that was not sustained over time.\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003e\u003cb\u003eModerate utilizers who lost care\u003c/b\u003e (n\u0026thinsp;=\u0026thinsp;973): These patients had moderate in-person and between-visit care utilization prior to the pandemic. After the shelter-in-place mandate, they did not recover in-person care levels, which continued to decline over time, with any telehealth and between-visit care are also trending downward.\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003e\u003cb\u003eLow utilizers who lost care\u003c/b\u003e (n\u0026thinsp;=\u0026thinsp;3,693): These patients had low in-person and between-visit utilization before the pandemic. Post-shelter-in-place, they did not return to in-person levels and maintained minimal between-visit care, with any telehealth uptake was not sustained.\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003c/ol\u003e\u003c/p\u003e\u003cp\u003eSociodemographic Characteristics by Post-Pandemic Care Trajectories\u003c/p\u003e\u003cp\u003eWe classified the seven SFHN clusters by pre-pandemic utilization and post-pandemic changes in in-person, telehealth, and between-visit care (Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e) and grouped them into four trajectories: added digital care (telehealth and/or between-visit care), transitioned to digital care (telehealth and/or between-visit care), lost care, and sustained minimal change (Table\u0026nbsp;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e5\u003c/span\u003e, full descriptive tables for each cluster are presented in Appendix D.2). Patients who added digital care were more likely to be \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\ge\\:\\)\u003c/span\u003e\u003c/span\u003e50 years old (p = .037), Hispanic with a non-English language preference (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), and insured through Medicaid or Medicare (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) compared to other care trajectories. Clinically, these patients were also more likely to have high comorbidity scores (CCI 3+) (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) and uncontrolled A1c levels at baseline (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). Patients who transitioned to digital care were more likely to be 75+ years old (p = .037), Black/African American or White (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), and much less likely to have a non-English language preference (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) compared to other care trajectories. These patients were most likely to be covered by Medicare, had high comorbidity scores (CCI 3+), have controlled baseline A1c levels, and have an active patient portal at baseline (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001 for all characteristics). Patients who lost care were more likely to be Asian -particularly with a non-English language preference (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) and be enrolled in Healthy Workers (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) compared to other care trajectories. Clinically, these patients were more likely to have lower comorbidity scores (CCI 0\u0026ndash;2), missing baseline A1c and blood pressure results, and low patient portal enrollment (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001 for all characteristics). SFHN patients who had minimal change in care patterns were more likely to be 35\u0026ndash;49 years old (p\u0026thinsp;=\u0026thinsp;.037), Hispanic/Latino with a non-English language preference (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), and uninsured (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) compared to other care trajectories. Clinically, they were also more likely to have lower comorbidity scores (CCI 0\u0026ndash;2), slightly more likely to have uncontrolled A1c at baseline, and a very low likelihood of being enrolled in the patient portal (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001 for all characteristics).\u003c/p\u003e\u003cp\u003eKey Patterns in Care Trajectories Across Health Systems\u003c/p\u003e\u003cp\u003eCare Trajectories by Post-Pandemic Changes in Utilization\u003c/p\u003e\u003cp\u003eNext, we descriptively compared these trajectories across healthcare systems to identify important patterns and/or differences. UCSF saw higher proportions of patients who added or transitioned to digital care, with 33% of patients adding digital care and 21% transitioning to digital care (Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). In contrast, SFHN had lower proportions in these categories: 12% added digital care and 10% transitioned to digital care (Table\u0026nbsp;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e5\u003c/span\u003e). At UCSF, those who added digital care (Clusters UCSF 1, UCSF 2, UCSF 6) demonstrated sustained moderate to high telehealth uptake and varying rates of between-visit interactions, but with an increase in use overall (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). At SFHN, however, those who added digital care (Clusters SFHN 1, SFHN 2) showed low to moderate uptake of telehealth (that was not sustained) and higher volumes of between-visit care (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eAt both UCSF and SFHN, those who transitioned to digital care (Clusters UCSF 3, UCSF 4, SFHN 3, SFHN 4) showed moderate to high uptake of telehealth. Patients who transitioned to digital care had heterogeneous uptake of between-visit care, as some clusters increased their interactions (Clusters UCSF 4, SFHN 4) while others maintained moderate to high uptake (with no sustained increase in volume; Clusters UCSF 3, SFHN 3).\u003c/p\u003e\u003cp\u003eSFHN exhibited higher proportions of patients who either maintained prior care patterns with minimal changes or experienced care disengagement: 10% maintained in-person visit rates with minimal to no uptake of digital care, and 69% experienced care disruptions and lost care (compared to 46% of patients who lost care at UCSF). Patients who lost care (Clusters UCSF 5, UCSF 7, SFHN 6, SFHN 7) demonstrated low adoption of telehealth and between-visit care, often characterized by consistent non-usage or decreasing trends in use. In-person visit rates also did not recover to pre-pandemic levels among the clusters that lost care.\u003c/p\u003e\u003cp\u003eCare Trajectories by Pre-Pandemic Utilization\u003c/p\u003e\u003cp\u003eTo compare care trajectories across both health systems, we focused on how care patterns changed for participants who were high, moderate, or low utilizers in the pre-pandemic period (see Appendix F for a detailed description). In brief, high utilizers mostly returned to their pre-pandemic in-person rates of care and incorporated both telehealth and/or between-visit care with sustained use of between-visit care modalities. Moderate utilizers had more heterogeneity, with some transitioning to digital care, some losing care, and some showing minimal changes. Low utilizers were the largest cluster in both health systems; some low utilizers lost care, while some added telehealth care.\u003c/p\u003e\u003cp\u003eSensitivity Analysis\u003c/p\u003e\u003cp\u003eSensitivity analyses using Optimal Matching (OM) instead of Dynamic Hamming Distance (DHD) produced broadly similar trajectory types. While OM yielded some variation in the distribution of patients, the overall care patterns and substantive interpretations remained consistent with the DHD results (Appendix G).\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eIn a large, diverse sample of patients with T2DM across two large San Francisco health systems, we characterized clusters of multi-modal healthcare utilization patterns across three types of visits (in-person, telehealth visits, and between-visit encounters) from April 2019 to March 2023 using sequence and cluster analysis. Across both UCSF and SFHN, three distinct patterns emerged: adding digital care, where in-person visit volumes returned to baseline while telehealth and between-visit communications remained elevated; transitioning to digital care, characterized by sustained reductions in in-person visits alongside increased telehealth and/or between-visit use; and loss of care, with neither in-person nor digital services returning to pre-pandemic levels. At SFHN, a fourth pattern\u0026mdash;minimal change\u0026mdash;was observed, reflecting stable use of in-person visits with persistently low use of telehealth and between-visit care. Within these patterns, we noticed heterogeneity not only in the degree of telehealth uptake and sustainability of between-visit care, but also in the timing of these shifts\u0026mdash;differences that were captured through our use of sequence analysis\u0026mdash;highlighting nuanced variation both across and within health systems. We then examined whether sociodemographic and clinical factors were associated with distinct healthcare utilization trajectories. Our findings provide critical insights into the factors shaping transitions to digital care, sustained in-person use, and care loss.\u003c/p\u003e\u003cp\u003ePost-Pandemic Shifts in Care Trajectories\u003c/p\u003e\u003cp\u003eA key finding was the identification of characteristics among the clusters that adopted digital care. Several clusters reflected a transition from primarily in-person care to primarily digital engagement after the shelter\u0026ndash;in\u0026ndash;place mandate, which was more common among Black or White patients with an English-language preference, those covered by Medicare, and those with higher comorbidity burdens. These patterns suggest that digital modalities may be particularly attractive or accessible for managing complex health needs or for those facing barriers to consistent in-person care, such as transportation, work schedules, or mobility issues, which are more pronounced among marginalized groups\u003csup\u003e\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e,\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e\u003c/sup\u003e. These results align with the existing literature, which indicates the value of hybrid models of care in increasing accessibility and supporting chronic disease management\u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e,\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e,\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e,\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e,\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e,\u003cspan additionalcitationids=\"CR54\" citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e\u003c/sup\u003e. These findings are particularly insightful because they both corroborate prior literature on digital shifts in care delivery and extend it to a broader population, including both academic and safety-net settings. By analyzing two health systems serving demographically distinct populations, we demonstrate that digital care adoption is not limited to well-resourced populations but also emerges as a critical tool for sustaining care engagement in more diverse, underserved settings. Notably, UCSF clusters experienced a higher uptake of telehealth compared to those at SFHN, which is surprising since SFHN primarily relied on generally more accessible between-visit interactions or phone visits, whereas patients at UCSF had higher proportions of video visits\u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eHeterogeneity in Sociodemographic Characteristics\u003c/p\u003e\u003cp\u003eRace and ethnicity, preferred language, and health insurance were associated with each care trajectory, yet their distributions differed across systems. This heterogeneity implies that demographic factors do not determine care utilization but can vary depending on the context, such as the health system\u0026rsquo;s capacity or digital health outreach approach. Notably, race and ethnicity and preferred language were similar across both health systems, with patients who added digital care being more likely to be Spanish-speaking Hispanic, particularly at SFHN; patients who transitioned to digital care were more likely to be White or Black, and those who lost care were more likely to be Asian with a non-English language preference. Differences by race and ethnicity and language indicate that cultural and linguistic factors may shape how patients engage with digital care, reflecting persistent barriers as well as opportunities to enhance access. At the same time, system-level elements, such as language-concordant communication, digital literacy support, and targeted outreach, likely play a critical role in determining which patients are able to adopt and sustain digital care use, underscoring the need for equity-oriented implementation strategies\u003csup\u003e\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e,\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e,\u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e57\u003c/span\u003e\u003c/sup\u003e. The fact that individuals who added digital care were more likely to be Spanish-speaking suggests that both health systems were successful in offering and supporting the adoption of digital care options for Spanish speakers and can serve as a model for engaging diverse populations who may face barriers to digital engagement.\u003c/p\u003e\u003cp\u003eMoreover, there was notable heterogeneity in health insurance coverage across the two health systems, highlighting how structural context may shape care engagement. These differences reflect the broader roles of each system: UCSF as a tertiary referral center with a commercially insured base, and SFHN as a safety-net system serving populations with more limited financial and digital access. Insurance status corresponded closely to care trajectories. At UCSF, patients who lost care were more likely to be commercially insured, a group that may experience weaker continuity with a single health system due to broader care alternatives, potentially reflecting data loss rather than true change in care patterns. At SFHN, uninsured patients were overrepresented among those who lost care or exhibited minimal changes in utilization. For uninsured patients, financial barriers and limited access options likely constrain sustained engagement, while for commercially insured patients, greater flexibility in provider choice may diffuse continuity across systems. Insurance type functions as a structural determinant of healthcare engagement, influencing not only access but also the stability and modality of care over time. Health systems serving distinct payer mixes, therefore, face different equity challenges: safety-net systems must continue to address affordability and digital access, while commercially oriented systems may need to strengthen coordination and continuity within fragmented care networks. Addressing these gaps through targeted interventions\u0026mdash;such as insurance navigation support, proactive outreach, and integrated digital care supports\u0026mdash;may improve continuity and mitigate disparities across diverse insurance contexts. Additionally, examining how policy-level factors, such as insurance or network restrictions, influence care engagement could inform system-wide strategies to prevent care loss across diverse populations.\u003c/p\u003e\u003cp\u003eCare Trajectories Based on Pre-Pandemic Utilization\u003c/p\u003e\u003cp\u003eWhen examining care trajectories through the lens of patients\u0026rsquo; pre-pandemic utilization levels, clear trends emerged across both health systems. High utilizers prior to the pandemic were the most likely to adopt or transition to digital care, demonstrating moderate to high telehealth uptake and between-visit care engagement. This suggests that the most engaged patients were more likely to incorporate digital modalities into their care. Notably, there was one UCSF cluster of low utilizers that also added care, demonstrating some heterogeneity. No clusters among moderate utilizers showed consistent trends of adding care.\u003c/p\u003e\u003cp\u003eThose who transitioned to digital care tended to be moderate to high utilizers prior to the pandemic, adopting telehealth and between-visit care as a partial substitute for lost in-person visits. However, for both the trajectories that added digital care or transitioned to digital care, sustained between-visit growth was not universal, suggesting that engagement in care between visits may ebb and flow depending on patients\u0026rsquo; needs and interests. Future studies should expand upon this work to evaluate if and how between-visit care supports care access.\u003c/p\u003e\u003cp\u003ePatients who lost care constituted the largest segments at both systems, mainly comprised of low or moderate pre-pandemic utilizers with limited telehealth uptake. This pattern underscores how individuals with historically lower engagement may be particularly vulnerable to care disruption and highlights the limited capacity of digital modalities alone to re-engage these patients. Notably, SFHN had a larger group of low utilizers who added care, suggesting that targeted outreach or leveraging more accessible modalities of engagement (such as greater emphasis on telephone calls versus video visits) may mitigate disengagement.\u003c/p\u003e\u003cp\u003eFuture Directions\u003c/p\u003e\u003cp\u003eThese results underscore the complex interplay between digital health adoption, socioeconomic factors, and healthcare disparities in urban contexts. By examining how patients navigate and combine multiple care modalities, we can better understand how these options may create complexity, overwhelm, or mismatch with patient needs, ultimately leading to reduced engagement.\u003c/p\u003e\u003cp\u003eIn this study, those who lost care or sustained minimal changes in their care were more likely to speak Chinese or Spanish. Several recent studies demonstrated that Chinese and Spanish speakers were disproportionately more likely to struggle with specific tasks required to use digital health tools effectively, reinforcing that linguistic and cultural barriers influence how patients access and interact with different modalities and aspects of care\u003csup\u003e\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e,\u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e59\u003c/span\u003e,\u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e65\u003c/span\u003e\u003c/sup\u003e. However, a recent study at SFHN found that language-concordant digital skills training particularly benefits Spanish speakers and patients with lower digital engagement, highlighting a possible strategy to address barriers to and reduce disparities in engagement among those at risk of care disengagement\u003csup\u003e\u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e66\u003c/span\u003e\u003c/sup\u003e. Designing tailored interventions and strategies for moderate and low utilizers at risk for care disengagement, including language-concordant support, digital literacy programs, and outreach efforts, has the potential to address barriers such as perceived usefulness and digital skills.\u003c/p\u003e\u003cp\u003eThe differences in cluster patterns at UCSF and SFHN also indicate that digital engagement is not one-size-fits-all, and the importance of tailoring digital adoption strategies to the structural and patient-level realities of each organization, including digital preferences and needs\u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e,\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e,\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e\u003c/sup\u003e. Clusters at SFHN demonstrated moderate to low telehealth engagement, compared to UCSF, which had multiple clusters with high telehealth adoption; but SFHN patients were more likely to have sustained or increased their between-visit engagement. This highlights the continued importance of telephone calls (as opposed to patient portal messages) as an approach to engage historically excluded populations\u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eLastly, future studies should consider integrating clinical outcomes. Linking utilization care trajectories to longitudinal health outcomes (such as glycemic control, blood pressure, or hospitalizations), will further clarify whether digital care adoption translates into measurable health benefits, and for whom.\u003c/p\u003e\u003cp\u003eLimitations\u003c/p\u003e\u003cp\u003eThis study has some limitations. First, as this is an observational study using EHR data, we cannot rule out residual confounding, particularly related to unmeasured factors such as patient motivation, access to devices and internet, and lifestyle factors or social norms that may influence attitude towards diabetes and remote engagement in care. While we included a range of clinical and sociodemographic covariates, selection into different care modalities is non-random and likely reflects underlying health engagement, care-seeking behavior, or attitudes toward different care modalities. Inclusion in the study was restricted to those who had at least one interaction with the healthcare system both pre- and post-pandemic, and results are therefore only representative of an empaneled, care-seeking population. Lastly, results are likely not generalizable to rural settings due to differences in digital access and infrastructure.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eIn two urban health system cohorts of adults with Type 2 diabetes analyzed over multiple years, we found that care utilization trajectories differed by sociodemographic and clinical characteristics. Specifically, in the post-pandemic era, we saw a substantial proportion of the patient population at both health systems embrace digital care (both telehealth and between-visit care), suggesting that increased digital accessibility supplements and allows patients to engage in care in ways that better suit their needs. However, a significant proportion of patients either lost access to care and experienced care disruptions or did not adopt digital care, suggesting a need for future research and tailored strategies to ensure equity in digital engagement and to close gaps in care continuity. As healthcare systems continue to expand hybrid models of care, it is critical to ensure equitable access and effectiveness across diverse populations. Future work should build upon this research and explore ways to mitigate disparities in the use of all care modalities, particularly for diverse patients and healthcare systems.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAuthor Contributions:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eConceptualization: Elaine C Khoong\u003c/p\u003e\n\u003cp\u003eData curation: Kathryn E Kemper-McIsaac, Elizabeth B Sherwin,\u0026nbsp;Anna Rubinsky\u003c/p\u003e\n\u003cp\u003eFormal analysis: Taylor Rapson, Kathryn E Kemper-McIsaac, Elizabeth B Sherwin\u003c/p\u003e\n\u003cp\u003eFunding acquisition: Elaine C Khoong\u003c/p\u003e\n\u003cp\u003eMethodology: Elizabeth B Sherwin, Lucia Pacca,\u0026nbsp;Anna Rubinsky\u003c/p\u003e\n\u003cp\u003eProject administration: Kathryn E Kemper-McIsaac\u003c/p\u003e\n\u003cp\u003eSupervision: Elaine C Khoong\u003c/p\u003e\n\u003cp\u003eVisualization: Kathryn E Kemper-McIsaac, Elizabeth B Sherwin\u003c/p\u003e\n\u003cp\u003eWriting \u0026ndash; original draft: Taylor Rapson, Kathryn E Kemper-McIsaac\u003c/p\u003e\n\u003cp\u003eWriting \u0026ndash; review \u0026amp; editing: All authors\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe following authors received this funding to support part of this work. No funders played any role in the study procedures.\u003c/p\u003e\n\u003cp\u003e- Residual class settlement funds in the matter of April Krueger v. Wyeth, Inc., Case No. 03-cv-2496 (US District Court, SD of Calif. (Kemper-McIsaac; Rubinsky; Khoong)\u003c/p\u003e\n\u003cp\u003e- National Institute of Nursing Research, T32NR019035 (Clifford)\u003c/p\u003e\n\u003cp\u003e- National Institute of Diabetes and Digestive and Kidney Diseases, P30 DK092924 (Lyles; Khoong)\u003c/p\u003e\n\u003cp\u003e- National Heart, Lung, and Blood Institute, K23HL157750 (Khoong)\u003cbr\u003e\u0026nbsp;- National Center for Advancing Translational Sciences, KL2TR001870 (Khoong)\u003c/p\u003e\n\u003cp\u003eDeclarations:\u003c/p\u003e\n\u003cp\u003eNo authors have any competing interests as defined by Nature Portfolio, or other interests that might be perceived to influence the results and/or discussion reported in this paper.\u003c/p\u003e\n\u003ch2\u003eAcknowledgements\u003c/h2\u003e\n\u003cp\u003eAuthors of this paper for sharing their code:\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eEisenberg-Guyot J, Peckham T, Andrea SB, Oddo V, Seixas N, Hajat A. Life-course trajectories of employment quality and health in the U.S.: A multichannel sequence analysis. Soc Sci Med. 2020 Nov;264:113327. doi: 10.1016/j.socscimed.2020.113327.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eRodriguez JA, Saadi A, Schwamm LH, Bates DW, Samal L. Disparities In Telehealth Use Among California Patients With Limited English Proficiency. \u003cem\u003eHealth Aff Proj Hope\u003c/em\u003e. 2021;40(3):487\u0026ndash;495. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1377/hlthaff.2020.00823\u003c/span\u003e\u003cspan address=\"10.1377/hlthaff.2020.00823\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRamsetty A, Adams C. Impact of the digital divide in the age of COVID-19. \u003cem\u003eJ Am Med Inform Assoc\u003c/em\u003e. 2020;27(7):1147\u0026ndash;1148. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1093/jamia/ocaa078\u003c/span\u003e\u003cspan address=\"10.1093/jamia/ocaa078\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRobinson J, Borgo L, Fennell K, Funahashi TT. The Covid-19 Pandemic Accelerates the Transition to Virtual Care. Published online 2020.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eReed M, Huang J, Graetz I, et al. Outpatient Electronic Health Records and the Clinical Care and Outcomes of Patients With Diabetes Mellitus. \u003cem\u003eAnn Intern Med\u003c/em\u003e. 2012;157(7):482\u0026ndash;489. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.7326/0003-4819-157-7-201210020-00004\u003c/span\u003e\u003cspan address=\"10.7326/0003-4819-157-7-201210020-00004\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHerrin J, Da Graca B, Nicewander D, et al. The Effectiveness of Implementing an Electronic Health Record on Diabetes Care and Outcomes. \u003cem\u003eHealth Serv Res\u003c/em\u003e. 2012;47(4):1522\u0026ndash;1540. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1111/j.1475-6773.2011.01370.x\u003c/span\u003e\u003cspan address=\"10.1111/j.1475-6773.2011.01370.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHarris LT, Koepsell TD, Haneuse SJ, Martin DP, Ralston JD. Glycemic Control Associated With Secure Patient-Provider Messaging Within a Shared Electronic Medical Record. \u003cem\u003eDiabetes Care\u003c/em\u003e. 2013;36(9):2726\u0026ndash;2733. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.2337/dc12-2003\u003c/span\u003e\u003cspan address=\"10.2337/dc12-2003\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLyles C, Schillinger D, Sarkar U. Connecting the Dots: Health Information Technology Expansion and Health Disparities. \u003cem\u003ePLOS Med\u003c/em\u003e. 2015;12(7):e1001852. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1371/journal.pmed.1001852\u003c/span\u003e\u003cspan address=\"10.1371/journal.pmed.1001852\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKruse CS, Argueta DA, Lopez L, Nair A. Patient and Provider Attitudes Toward the Use of Patient Portals for the Management of Chronic Disease: A Systematic Review. \u003cem\u003eJ Med Internet Res\u003c/em\u003e. 2015;17(2):e40. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.2196/jmir.3703\u003c/span\u003e\u003cspan address=\"10.2196/jmir.3703\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGoldzweig CL, Orshansky G, Paige NM, et al. Electronic Patient Portals: Evidence on Health Outcomes, Satisfaction, Efficiency, and Attitudes: A Systematic Review. \u003cem\u003eAnn Intern Med\u003c/em\u003e. 2013;159(10):677\u0026ndash;687. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.7326/0003-4819-159-10-201311190-00006\u003c/span\u003e\u003cspan address=\"10.7326/0003-4819-159-10-201311190-00006\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRodriguez JA, Casillas A, Cook BL, Marlin RP. The Language of Equity in Digital Health: Prioritizing the Needs of Limited English Proficient Communities in the Patient Portal 2.0. \u003cem\u003eJ Health Care Poor Underserved\u003c/em\u003e. 2021;32:211\u0026ndash;219.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRalston JD, Rutter CM, Carrell D, Hecht J, Rubanowice D, Simon GE. Patient Use of Secure Electronic Messaging Within a Shared Medical Record: A Cross-sectional Study. \u003cem\u003eJ Gen Intern Med\u003c/em\u003e. 2009;24(3):349\u0026ndash;355. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s11606-008-0899-z\u003c/span\u003e\u003cspan address=\"10.1007/s11606-008-0899-z\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLyles CR, Nelson EC, Frampton S, Dykes PC, Cemballi AG, Sarkar U. Using Electronic Health Record Portals to Improve Patient Engagement: Research Priorities and Best Practices. \u003cem\u003eAnn Intern Med\u003c/em\u003e. 2020;172(11_Supplement):S123-S129. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.7326/M19-0876\u003c/span\u003e\u003cspan address=\"10.7326/M19-0876\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAntonio MG, Petrovskaya O, Lau F. Is research on patient portals attuned to health equity? A scoping review. \u003cem\u003eJ Am Med Inform Assoc\u003c/em\u003e. 2019;26(8\u0026ndash;9):871\u0026ndash;883. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1093/jamia/ocz054\u003c/span\u003e\u003cspan address=\"10.1093/jamia/ocz054\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSarkar U, Karter AJ, Liu JY, et al. Social disparities in internet patient portal use in diabetes: evidence that the digital divide extends beyond access. \u003cem\u003eJ Am Med Inform Assoc\u003c/em\u003e. 2011;18(3):318\u0026ndash;321. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1136/jamia.2010.006015\u003c/span\u003e\u003cspan address=\"10.1136/jamia.2010.006015\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDeshpande N, Arora VM, Vollbrecht H, Meltzer DO, Press V. eHealth Literacy and Patient Portal Use and Attitudes: Cross-sectional Observational Study. \u003cem\u003eJMIR Hum Factors\u003c/em\u003e. 2023;10:e40105. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.2196/40105\u003c/span\u003e\u003cspan address=\"10.2196/40105\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAncker JS, Barr\u0026oacute;n Y, Rockoff ML, et al. Use of an Electronic Patient Portal Among Disadvantaged Populations. \u003cem\u003eJ Gen Intern Med\u003c/em\u003e. 2011;26(10):1117\u0026ndash;1123. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s11606-011-1749-y\u003c/span\u003e\u003cspan address=\"10.1007/s11606-011-1749-y\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAncker JS, Hafeez B, Kaushal R. Socioeconomic Disparities in Adoption of Personal Health Records Over Time. \u003cem\u003eAm J Manag Care\u003c/em\u003e. 2016;22(8):539\u0026ndash;540.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKhoong EC, Rivadeneira NA, Hiatt RA, Sarkar U. The Use of Technology for Communicating With Clinicians or Seeking Health Information in a Multilingual Urban Cohort: Cross-Sectional Survey. \u003cem\u003eJ Med Internet Res\u003c/em\u003e. 2020;22(4):e16951. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.2196/16951\u003c/span\u003e\u003cspan address=\"10.2196/16951\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTieu L, Schillinger D, Sarkar U, et al. Online patient websites for electronic health record access among vulnerable populations: portals to nowhere? \u003cem\u003eJ Am Med Inform Assoc\u003c/em\u003e. 2017;24(e1):e47-e54. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1093/jamia/ocw098\u003c/span\u003e\u003cspan address=\"10.1093/jamia/ocw098\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGibbons MC. A Historical Overview of Health Disparities and the Potential of eHealth Solutions. \u003cem\u003eJ Med Internet Res\u003c/em\u003e. 2005;7(5):e50. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.2196/jmir.7.5.e50\u003c/span\u003e\u003cspan address=\"10.2196/jmir.7.5.e50\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eEberly LA, Kallan MJ, Julien HM, et al. Patient Characteristics Associated With Telemedicine Access for Primary and Specialty Ambulatory Care During the COVID-19 Pandemic. \u003cem\u003eJAMA Netw Open\u003c/em\u003e. 2020;3(12):e2031640. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1001/jamanetworkopen.2020.31640\u003c/span\u003e\u003cspan address=\"10.1001/jamanetworkopen.2020.31640\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRyskina KL, Shultz K, Zhou Y, Lautenbach G, Brown RT. Older adults\u0026rsquo; access to primary care: Gender, racial, and ethnic disparities in telemedicine. \u003cem\u003eJ Am Geriatr Soc\u003c/em\u003e. 2021;69(10):2732\u0026ndash;2740. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1111/jgs.17354\u003c/span\u003e\u003cspan address=\"10.1111/jgs.17354\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSarkar U, Karter AJ, Liu JY, et al. The Literacy Divide: Health Literacy and the Use of an Internet-Based Patient Portal in an Integrated Health System\u0026mdash;Results from the Diabetes Study of Northern California (DISTANCE). \u003cem\u003eJ Health Commun\u003c/em\u003e. 2010;15(sup2):183\u0026ndash;196. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1080/10810730.2010.499988\u003c/span\u003e\u003cspan address=\"10.1080/10810730.2010.499988\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDavies AR, Honeyman M, Gann B. Addressing the Digital Inverse Care Law in the Time of COVID-19: Potential for Digital Technology to Exacerbate or Mitigate Health Inequalities. \u003cem\u003eJ Med Internet Res\u003c/em\u003e. 2021;23(4):e21726. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.2196/21726\u003c/span\u003e\u003cspan address=\"10.2196/21726\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLyles CR, Harris LT, Jordan L, et al. Patient Race/Ethnicity and Shared Medical Record Use Among Diabetes Patients. \u003cem\u003eMed Care\u003c/em\u003e. 2012;50(5):434. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1097/MLR.0b013e318249d81b\u003c/span\u003e\u003cspan address=\"10.1097/MLR.0b013e318249d81b\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAnthony DL, Campos-Castillo C, Lim PS. Who Isn\u0026rsquo;t Using Patient Portals And Why? Evidence And Implications From A National Sample Of US Adults. \u003cem\u003eHealth Aff (Millwood)\u003c/em\u003e. 2018;37(12):1948\u0026ndash;1954. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1377/hlthaff.2018.05117\u003c/span\u003e\u003cspan address=\"10.1377/hlthaff.2018.05117\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLyles CR, Nguyen OK, Khoong EC, Aguilera A, Sarkar U. Multilevel Determinants of Digital Health Equity: A Literature Synthesis to Advance the Field. \u003cem\u003eAnnu Rev Public Health\u003c/em\u003e. 2023;44:383\u0026ndash;405. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1146/annurev-publhealth-071521-023913\u003c/span\u003e\u003cspan address=\"10.1146/annurev-publhealth-071521-023913\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAlkureishi MA, Choo ZY, Rahman A, et al. Digitally Disconnected: Qualitative Study of Patient Perspectives on the Digital Divide and Potential Solutions. \u003cem\u003eJMIR Hum Factors\u003c/em\u003e. 2021;8(4):e33364. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.2196/33364\u003c/span\u003e\u003cspan address=\"10.2196/33364\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNouri S, Khoong EC, Lyles CR, Karliner L. Addressing Equity in Telemedicine for Chronic Disease Management During the Covid-19 Pandemic. \u003cem\u003eCatal Non-Issue Content\u003c/em\u003e. 2020;1(3). doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1056/CAT.20.0123\u003c/span\u003e\u003cspan address=\"10.1056/CAT.20.0123\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNouri S, Lyles CR, Sherwin EB, et al. Visit and Between-Visit Interaction Frequency Before and After COVID-19 Telehealth Implementation. \u003cem\u003eJAMA Netw Open\u003c/em\u003e. 2023;6(9):e2333944. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1001/jamanetworkopen.2023.33944\u003c/span\u003e\u003cspan address=\"10.1001/jamanetworkopen.2023.33944\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLi P, Zhang P, Guan D, et al. Changes in racial and ethnic disparities in glucose-lowering drug utilization and glycated haemoglobin A1c in US adults with diabetes: 2005‐2018. \u003cem\u003eDiabetes Obes Metab\u003c/em\u003e. 2023;25(2):516\u0026ndash;525. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1111/dom.14894\u003c/span\u003e\u003cspan address=\"10.1111/dom.14894\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSheon AR, Bolen SD, Callahan B, Shick S, Perzynski AT. Addressing Disparities in Diabetes Management Through Novel Approaches to Encourage Technology Adoption and Use. \u003cem\u003eJMIR Diabetes\u003c/em\u003e. 2017;2(2):e16. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.2196/diabetes.6751\u003c/span\u003e\u003cspan address=\"10.2196/diabetes.6751\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLyles CR, Grothaus L, Reid RJ, Sarkar U, Ralston JD. Communication about diabetes risk factors during between-visit encounters. \u003cem\u003eAm J Manag Care\u003c/em\u003e. 2012;18(12):807\u0026ndash;815.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJorge A. Rodriguez MD, Joseph R. Betancourt MD, Thomas D. Sequist MD, Ishani Ganguli MD. Differences in the Use of Telephone and Video Telemedicine Visits During the COVID-19 Pandemic. 2021;27. Accessed August 12, 2025. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.ajmc.com/view/differences-in-the-use-of-telephone-and-video-telemedicine-visits-during-the-covid-19-pandemic\u003c/span\u003e\u003cspan address=\"https://www.ajmc.com/view/differences-in-the-use-of-telephone-and-video-telemedicine-visits-during-the-covid-19-pandemic\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNath B, Williams B, Jeffery MM, et al. Trends in Electronic Health Record Inbox Messaging During the COVID-19 Pandemic in an Ambulatory Practice Network in New England. \u003cem\u003eJAMA Netw Open\u003c/em\u003e. 2021;4(10):e2131490. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1001/jamanetworkopen.2021.31490\u003c/span\u003e\u003cspan address=\"10.1001/jamanetworkopen.2021.31490\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePatel SY, Mehrotra A, Huskamp HA, Uscher-Pines L, Ganguli I, Barnett ML. Trends in Outpatient Care Delivery and Telemedicine During the COVID-19 Pandemic in the US. \u003cem\u003eJAMA Intern Med\u003c/em\u003e. 2021;181(3):388\u0026ndash;391. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1001/jamainternmed.2020.5928\u003c/span\u003e\u003cspan address=\"10.1001/jamainternmed.2020.5928\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAbbott A, Tsay A. Sequence Analysis and Optimal Matching Methods in Sociology. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1177/0049124100029001001\u003c/span\u003e\u003cspan address=\"10.1177/0049124100029001001\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eEmery K, Berchtold A. Comparison of two approaches in multichannel sequence analysis using the Swiss Household Panel. \u003cem\u003eLongitud Life Course Stud\u003c/em\u003e. 2023;14(4):592\u0026ndash;623. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1332/175795921X16698302233894\u003c/span\u003e\u003cspan address=\"10.1332/175795921X16698302233894\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eEisenberg-Guyot J, Peckham T, Andrea SB, Oddo V, Seixas N, Hajat A. Life-course trajectories of employment quality and health in the U.S.: A multichannel sequence analysis. \u003cem\u003eSoc Sci Med\u003c/em\u003e. 2020;264:113327. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.socscimed.2020.113327\u003c/span\u003e\u003cspan address=\"10.1016/j.socscimed.2020.113327\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSavar\u0026eacute; L, Ieva F, Corrao G, Lora A. Mining and evaluation of patients\u0026rsquo; diagnostic therapeutic paths through state sequences analysis. \u003cem\u003earXiv\u003c/em\u003e. Preprint posted online September 9, 2022. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.48550/arXiv.2209.04384\u003c/span\u003e\u003cspan address=\"10.48550/arXiv.2209.04384\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKempa-Liehr AW, Lin CYC, Britten R, et al. Healthcare pathway discovery and probabilistic machine learning. \u003cem\u003eInt J Med Inf\u003c/em\u003e. 2020;137:104087. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.ijmedinf.2020.104087\u003c/span\u003e\u003cspan address=\"10.1016/j.ijmedinf.2020.104087\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMathew S, Peat G, Parry E, Sokhal BS, Yu D. Applying sequence analysis to uncover \u0026lsquo;real-world\u0026rsquo; clinical pathways from routinely collected data: a systematic review. \u003cem\u003eJ Clin Epidemiol\u003c/em\u003e. 2024;166:111226. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.jclinepi.2023.111226\u003c/span\u003e\u003cspan address=\"10.1016/j.jclinepi.2023.111226\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCezard G, Sullivan F, Keenan K. Understanding multimorbidity trajectories in Scotland using sequence analysis. \u003cem\u003eSci Rep\u003c/em\u003e. 2022;12(1):16485. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41598-022-20546-4\u003c/span\u003e\u003cspan address=\"10.1038/s41598-022-20546-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGomez SL, Glaser SL, McClure LA, et al. The California Neighborhoods Data System: a new resource for examining the impact of neighborhood characteristics on cancer incidence and outcomes in populations. \u003cem\u003eCancer Causes Control CCC\u003c/em\u003e. 2011;22(4):631\u0026ndash;647. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s10552-011-9736-5\u003c/span\u003e\u003cspan address=\"10.1007/s10552-011-9736-5\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eQuan H, Sundararajan V, Halfon P, et al. Coding algorithms for defining comorbidities in ICD-9-CM and ICD-10 administrative data. \u003cem\u003eMed Care\u003c/em\u003e. 2005;43(11):1130\u0026ndash;1139. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1097/01.mlr.0000182534.19832.83\u003c/span\u003e\u003cspan address=\"10.1097/01.mlr.0000182534.19832.83\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAgborsangaya CB, Lau D, Lahtinen M, Cooke T, Johnson JA. Health-related quality of life and healthcare utilization in multimorbidity: results of a cross-sectional survey. \u003cem\u003eQual Life Res\u003c/em\u003e. 2013;22(4):791\u0026ndash;799. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s11136-012-0214-7\u003c/span\u003e\u003cspan address=\"10.1007/s11136-012-0214-7\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSegal JB, Sen AP, Glanzberg-Krainin E, Hutfless S. Factors Associated With Overuse of Health Care Within US Health Systems: A Cross-sectional Analysis of Medicare Beneficiaries From 2016 to 2018. \u003cem\u003eJAMA Health Forum\u003c/em\u003e. 2022;3(1):e214543. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1001/jamahealthforum.2021.4543\u003c/span\u003e\u003cspan address=\"10.1001/jamahealthforum.2021.4543\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCheng Y, Goodin AJ, Pahor M, Manini T, Brown JD. Healthcare Utilization and Physical Functioning in Older Adults in the United States. \u003cem\u003eJ Am Geriatr Soc\u003c/em\u003e. 2020;68(2):266\u0026ndash;271. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1111/jgs.16260\u003c/span\u003e\u003cspan address=\"10.1111/jgs.16260\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGauthier JA, Widmer ED, Bucher P, Notredame C. 1. Multichannel Sequence Analysis Applied to Social Science Data. \u003cem\u003eSociol Methodol\u003c/em\u003e. 2010;40(1):1\u0026ndash;38. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1111/j.1467-9531.2010.01227.x\u003c/span\u003e\u003cspan address=\"10.1111/j.1467-9531.2010.01227.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGabadinho A, Ritschard G, M\u0026uuml;ller NS, Studer M. Analyzing and Visualizing State Sequences in R with TraMineR. \u003cem\u003eJ Stat Softw\u003c/em\u003e. 2011;40(4):1\u0026ndash;37. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.18637/jss.v040.i04\u003c/span\u003e\u003cspan address=\"10.18637/jss.v040.i04\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMaechler M, Rousseeuw P, Hubert M, Hornik K. cluster: Cluster Analysis Basics and Extensions. Published online 2025.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWiroonsri N, Preedasawakul O. A correlation-based fuzzy cluster validity index with secondary options detector. \u003cem\u003earXiv\u003c/em\u003e. Preprint posted online March 7, 2024. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.48550/arXiv.2308.14785\u003c/span\u003e\u003cspan address=\"10.48550/arXiv.2308.14785\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWiroonsri N. Clustering performance analysis using a new correlation-based cluster validity index. \u003cem\u003ePattern Recognit\u003c/em\u003e. 2024;145:109910. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.patcog.2023.109910\u003c/span\u003e\u003cspan address=\"10.1016/j.patcog.2023.109910\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLesnard L. Using Optimal Matching Analysis in Sociology: Cost Setting and Sociology of Time. In: Blanchard P, B\u0026uuml;hlmann F, Gauthier JA, eds. \u003cem\u003eAdvances in Sequence Analysis: Theory, Method, Applications\u003c/em\u003e. Vol 2. Life Course Research and Social Policies. Springer International Publishing; 2014:39\u0026ndash;50. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/978-3-319-04969-4\u003c/span\u003e\u003cspan address=\"10.1007/978-3-319-04969-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eStuder M, Ritschard G. What Matters in Differences Between Life Trajectories: A Comparative Review of Sequence Dissimilarity Measures. \u003cem\u003eJ R Stat Soc Ser A Stat Soc\u003c/em\u003e. 2016;179(2):481\u0026ndash;511. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1111/rssa.12125\u003c/span\u003e\u003cspan address=\"10.1111/rssa.12125\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBatool F, Hennig C. Clustering with the Average Silhouette Width. \u003cem\u003eComput Stat Data Anal\u003c/em\u003e. 2021;158:107190. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.csda.2021.107190\u003c/span\u003e\u003cspan address=\"10.1016/j.csda.2021.107190\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCaliński T, Harabasz J. A dendrite method for cluster analysis. \u003cem\u003eCommun Stat\u003c/em\u003e. 1974;3(1):1\u0026ndash;27. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1080/03610927408827101\u003c/span\u003e\u003cspan address=\"10.1080/03610927408827101\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eInstitute of Medicine (US) Committee on Quality of Health Care in America. \u003cem\u003eCrossing the Quality Chasm: A New Health System for the 21st Century\u003c/em\u003e. National Academies Press (US); 2001. Accessed August 20, 2025. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://www.ncbi.nlm.nih.gov/books/NBK222274/\u003c/span\u003e\u003cspan address=\"http://www.ncbi.nlm.nih.gov/books/NBK222274/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eChoxi H, VanDerSchaaf H, Li Y, Morgan E. Telehealth and the Digital Divide: Identifying Potential Care Gaps in Video Visit Use. \u003cem\u003eJ Med Syst\u003c/em\u003e. 2022;46(9):58. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s10916-022-01843-x\u003c/span\u003e\u003cspan address=\"10.1007/s10916-022-01843-x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCoughlin SS, Williams LB, Hatzigeorgiou C. A systematic review of studies of web portals for patients with diabetes mellitus. \u003cem\u003emHealth\u003c/em\u003e. 2017;3:23\u0026ndash;23. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.21037/mhealth.2017.05.05\u003c/span\u003e\u003cspan address=\"10.21037/mhealth.2017.05.05\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKhoong EC, Butler BA, Mesina O, et al. Patient interest in and barriers to telemedicine video visits in a multilingual urban safety-net system. \u003cem\u003eJ Am Med Inform Assoc\u003c/em\u003e. 2021;28(2):349\u0026ndash;353. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1093/jamia/ocaa234\u003c/span\u003e\u003cspan address=\"10.1093/jamia/ocaa234\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLee G, Chang A, Pal A, Tran TA, Cui X, Quach T. Understanding and Addressing the Digital Health Literacy Needs of Low-Income Limited English Proficient Asian American Patients. \u003cem\u003eHealth Equity\u003c/em\u003e. 2022;6(1):494\u0026ndash;499. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1089/heq.2022.0045\u003c/span\u003e\u003cspan address=\"10.1089/heq.2022.0045\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSchoen C, Osborn R, Squires D, Doty MM. Access, Affordability, And Insurance Complexity Are Often Worse In The United States Compared To Ten Other Countries. \u003cem\u003eHealth Aff (Millwood)\u003c/em\u003e. 2013;32(12):2205\u0026ndash;2215. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1377/hlthaff.2013.0879\u003c/span\u003e\u003cspan address=\"10.1377/hlthaff.2013.0879\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTieu L, Lyles CR, Kim HC, et al. A self-report measure of digital skills needed to use digital health tools among older adults\u0026mdash;the Skills Measurement and Readiness Training for Digital Health (SMART Digital Health) Scale. \u003cem\u003eJ Am Med Inform Assoc\u003c/em\u003e. Published online September 13, 2025:ocaf151. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1093/jamia/ocaf151\u003c/span\u003e\u003cspan address=\"10.1093/jamia/ocaf151\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e Manuscripts submitted for publication:\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRapson T, Ramirez M, He S, Wong J, Kim HC, Luna I, Yang A, Li J, Fishman P, Ralston JD, Lyles C, Khoong E. Identifying Skill and Usability Barriers to Digital Health Tool Use among Patients in Safety Net Clinics: A Mixed Methods Study. JMIR Preprints, Under Review. doi: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.2196/preprints.78430\u003c/span\u003e\u003cspan address=\"10.2196/preprints.78430\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRapson T, Naranjo-Arellano M, Ralston JD, Garcia F, Green BB, Gutierrez C, Guzman-Estrada S, Khoong E, Sarkar U, Shin J, Zhou C, Fishman P, Ramirez M, Lyles C. Impact of Digital Support Interventions on Patient Portal Use Among Patients in an Urban Safety-net Health System. Submitted for publication. 2025\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"},{"header":"Tables","content":"\u003cp\u003e\u003cstrong\u003eTable 1. Patient Sociodemographic and Clinical Characteristics at UCSF and SFHN\u003c/strong\u003e\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"576\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eUCSF\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eSFHN\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\n \u003cp\u003e\u003cstrong\u003en=3,860\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e\u003cstrong\u003en=6,811\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAge Category (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;18-34\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\n \u003cp\u003e91 (2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e162 (2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;35-49\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\n \u003cp\u003e472 (12%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e1,080 (16%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;50-64\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\n \u003cp\u003e1,234 (32%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e3,215 (47%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;65-74\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\n \u003cp\u003e1,118 (29%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e1,727 (25%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;75+\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\n \u003cp\u003e945 (24%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e627 (9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eSex (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Female\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\n \u003cp\u003e2,091 (54%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e3,581 (53%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Male\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\n \u003cp\u003e1,769 (46%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e3,229 (47%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eJoint Race Eth \u0026amp; Language (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 50%;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Asian \u0026amp; English Preference\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\n \u003cp\u003e1,050 (27%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e641 (9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 50%;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Asian \u0026amp; \u0026nbsp;Non-English Preference\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\n \u003cp\u003e478 (12%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e1,753 (26%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 50%;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Black/African American \u0026amp; English Preference\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\n \u003cp\u003e527 (14%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e1,027 (15%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 50%;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Hispanic \u0026amp; English Preference\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\n \u003cp\u003e366 (9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e449 (7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 50%;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Hispanic \u0026amp; Non-English Preference\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\n \u003cp\u003e114 (3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e1,881 (28%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 50%;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Other/Un \u0026amp; English Preference\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\n \u003cp\u003e317 (8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e254 (4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 50%;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Other/Un \u0026amp; Non-English Preference\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\n \u003cp\u003e67 (2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e84 (1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 50%;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;White \u0026amp; English Preference\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\n \u003cp\u003e941 (24%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e722 (11%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eRace or Ethnicity (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Asian\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\n \u003cp\u003e1,528 (40%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e2,394 (35%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 50%;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Black/African American\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\n \u003cp\u003e527 (14%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e1,027 (15%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Hispanic/Latino\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\n \u003cp\u003e480 (12%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e2,330 (34%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Other/Unknown\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\n \u003cp\u003e384 (10%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e338 (5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;White\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\n \u003cp\u003e941 (24%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e722 (11%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eLanguage Preference (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Chinese\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\n \u003cp\u003e296 (8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e1,324 (19%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;English\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\n \u003cp\u003e3,154 (82%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e2,931 (43%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Other\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\n \u003cp\u003e298 (8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e655 (10%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Spanish\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\n \u003cp\u003e112 (3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e1,901 (28%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eInsurance (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Commercial\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\n \u003cp\u003e1,708 (44%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e71 (1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003eHealthy Workers\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\n \u003cp\u003e0 (0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e1,049 (15%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Medicaid\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\n \u003cp\u003e502 (13%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e2,709 (40%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Medicare\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\n \u003cp\u003e1,644 (43%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e2,126 (31%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Missing\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\n \u003cp\u003e5 (0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e28 (0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Uninsured\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\n \u003cp\u003e1 (0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e828 (12%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eNeighborhood SES (1=lowest) (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;1 (lowest)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\n \u003cp\u003e651 (17%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e2,561 (38%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\n \u003cp\u003e689 (18%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e1,771 (26%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\n \u003cp\u003e752 (19%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e1,049 (15%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\n \u003cp\u003e886 (23%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e889 (13%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;5 (highest)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\n \u003cp\u003e698 (18%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e405 (6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Missing\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\n \u003cp\u003e184 (5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e136 (2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eCharlson Score (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;0-2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\n \u003cp\u003e2,965 (77%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e5,818 (85%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;3+\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\n \u003cp\u003e895 (23%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e993 (15%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBaseline A1c Control (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Uncontrolled A1c (\u003cimg width=\"8\" height=\"13\" src=\"https://myfiles.space/user_files/69519_bce2c0439cd956a6/69519_custom_files/img1765797971.png\" alt=\"image\"\u003e8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\n \u003cp\u003e818 (21%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e2,631 (39%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Controlled A1c (\u0026lt;8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\n \u003cp\u003e2,611 (68%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e3,307 (49%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Missing\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\n \u003cp\u003e431 (11%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e873 (13%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBaseline BP Control (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;BP \u0026lt;=120/80\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\n \u003cp\u003e658 (17%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e1,629 (24%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;BP \u0026lt;=140/90\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\n \u003cp\u003e2,126 (55%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e3,598 (53%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;BP \u0026gt;140/90\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\n \u003cp\u003e930 (24%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e1,205 (18%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Missing\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\n \u003cp\u003e146 (4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e379 (6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eActive MyChart (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;No\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\n \u003cp\u003e861 (22%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e6,217 (91%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Yes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 23.9583%;\"\u003e\n \u003cp\u003e2,999 (78%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 26.0417%;\"\u003e\n \u003cp\u003e594 (9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cstrong\u003eTable 2. Descriptions of UCSF Healthcare Trajectories by Cluster\u003c/strong\u003e\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"726\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"2\" style=\"width: 108px;\"\u003e\n \u003cp\u003e\u003cstrong\u003ePre-Pandemic Utilization\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" style=\"width: 60px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eCluster\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" style=\"width: 78px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eN (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" style=\"width: 354px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eHow Each Care Pattern Changed Relative to Pre-Pandemic\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" style=\"width: 126px;\"\u003e\n \u003cp\u003e\u003cstrong\u003ePost-Pandemic Care Trajectory\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 114px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eIn-person\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 108px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTelehealth\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 132px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBetween Visits\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 108px;\"\u003e\n \u003cp\u003eHigh Utilizers\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 60px;\"\u003e\n \u003cp\u003eUCSF 1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e177 (4.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 114px;\"\u003e\n \u003cp\u003eReturned to pre levels\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 108px;\"\u003e\n \u003cp\u003eHigh initial uptake, moderate sustained use\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 132px;\"\u003e\n \u003cp\u003eContinued high usage with slight increase overall\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 126px;\"\u003e\n \u003cp\u003eAdded Telehealth\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 108px;\"\u003e\n \u003cp\u003eHigh Utilizers\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 60px;\"\u003e\n \u003cp\u003eUCSF 2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e333 (8.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 114px;\"\u003e\n \u003cp\u003eReturned to almost pre levels\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 108px;\"\u003e\n \u003cp\u003eSustained moderate uptake\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 132px;\"\u003e\n \u003cp\u003eModerate uptake with slight increase overall\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 126px;\"\u003e\n \u003cp\u003eAdded Telehealth and Between-Visit Care\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 108px;\"\u003e\n \u003cp\u003eHigh Utilizers\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 60px;\"\u003e\n \u003cp\u003eUCSF 3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e212 (5.5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 114px;\"\u003e\n \u003cp\u003eDid not recover to pre levels\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 108px;\"\u003e\n \u003cp\u003eSustained high uptake\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 132px;\"\u003e\n \u003cp\u003eContinued high usage (no sustained increase)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 126px;\"\u003e\n \u003cp\u003eTransitioned to Telehealth\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 108px;\"\u003e\n \u003cp\u003eModerate Utilizers\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 60px;\"\u003e\n \u003cp\u003eUCSF 4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e584 (15.1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 114px;\"\u003e\n \u003cp\u003eDid not recover to pre levels\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 108px;\"\u003e\n \u003cp\u003eSustained moderate uptake\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 132px;\"\u003e\n \u003cp\u003eModerate uptake with slight increase overall\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 126px;\"\u003e\n \u003cp\u003eTransitioned to Telehealth and Between-Visit Care\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 108px;\"\u003e\n \u003cp\u003eModerate Utilizers\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 60px;\"\u003e\n \u003cp\u003eUCSF 5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e173 (4.5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 114px;\"\u003e\n \u003cp\u003eDid not recover to pre levels\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 108px;\"\u003e\n \u003cp\u003eModerate uptake not sustained\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 132px;\"\u003e\n \u003cp\u003eModerate usage, decreasing trends\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 126px;\"\u003e\n \u003cp\u003eLost Care\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 108px;\"\u003e\n \u003cp\u003eLow Utilizers\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 60px;\"\u003e\n \u003cp\u003eUCSF 6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e769 (19.9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 114px;\"\u003e\n \u003cp\u003eReturned to almost pre-levels\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 108px;\"\u003e\n \u003cp\u003eSustained moderate uptake\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 132px;\"\u003e\n \u003cp\u003eLow uptake with increase overall\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 126px;\"\u003e\n \u003cp\u003eAdded Telehealth\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 108px;\"\u003e\n \u003cp\u003eLow Utilizers\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 60px;\"\u003e\n \u003cp\u003eUCSF 7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e1612 (41.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 114px;\"\u003e\n \u003cp\u003eDid not recover to pre levels\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 108px;\"\u003e\n \u003cp\u003eLow, consistent non usage\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 132px;\"\u003e\n \u003cp\u003eLow, consistent non usage\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 126px;\"\u003e\n \u003cp\u003eLost Care\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eWe defined adding digital care as in-person visit rates returning to pre-pandemic levels, while telehealth and/or between-visit care remain elevated. Those who transitioned to digital care saw in-person visit rates remain below pre-pandemic levels, while telehealth and between-visit care increased to replace in-person care. Those who lost care had minimal uptake of between-visit care or telehealth, even though in-person care rates did not rebound. We also further specified whether those who added or transitioned to digital care primarily adopted Telehealth, Between-Visit Care, or both.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 3. Sociodemographic Characteristics of Care Trajectories at UCSF\u003c/strong\u003e\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"638\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 192px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd rowspan=\"2\" style=\"width: 102px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTotal N (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAdded Digital Care\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTransitioned to Digital Care\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eLost Care\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" style=\"width: 68px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eP-value\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 192px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 96px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eClusters 1, 2, 6\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 96px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eClusters 3, 4\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 84px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eClusters 5, 7\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 192px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eN (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 102px;\"\u003e\n \u003cp\u003eN=3,860\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 96px;\"\u003e\n \u003cp\u003e1,279 (33)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 96px;\"\u003e\n \u003cp\u003e796 (21)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 84px;\"\u003e\n \u003cp\u003e1,785 (46)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 68px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" style=\"width: 570px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAge Category (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\n \u003cp\u003e0.079\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 192px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;18-34\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 102px;\"\u003e\n \u003cp\u003e91 (2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e28 (2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e22 (3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e41 (2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 192px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;35-49\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 102px;\"\u003e\n \u003cp\u003e472 (12)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e155 (12)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e113 (14)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e204 (11)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 192px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;50-64\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 102px;\"\u003e\n \u003cp\u003e1,234 (32)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e419 (33)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e267 (34)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e548 (31)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 192px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;65-74\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 102px;\"\u003e\n \u003cp\u003e1,118 (29)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e347 (27)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e227 (29)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e544 (30)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 192px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;75+\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 102px;\"\u003e\n \u003cp\u003e945 (24)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e330 (26)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e167 (21)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e448 (25)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" style=\"width: 570px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eSex (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\n \u003cp\u003e0.020\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 192px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Female\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 102px;\"\u003e\n \u003cp\u003e2,091 (54)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e726 (57)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e440 (55)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e925 (52)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 192px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Male\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 102px;\"\u003e\n \u003cp\u003e1,769 (46)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e553 (43)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e356 (45)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e860 (48)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" style=\"width: 570px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eJoint Race Ethnicity \u0026amp; Language (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 192px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Asian \u0026amp; English Preference\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 102px;\"\u003e\n \u003cp\u003e1,050 (27)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e327 (26)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e150 (19)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e573 (32)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 192px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Asian \u0026amp; Non-English Preference\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 102px;\"\u003e\n \u003cp\u003e478 (12)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e145 (11)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e55 (7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e278 (16)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 192px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Black/African American\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 102px;\"\u003e\n \u003cp\u003e527 (14)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e216 (17)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e141 (18)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e170 (10)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 192px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Hispanic \u0026amp; English Preference\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 102px;\"\u003e\n \u003cp\u003e366 (9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e109 (9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e86 (11)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e171 (10)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 192px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Hispanic \u0026amp; Non-English Preference\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 102px;\"\u003e\n \u003cp\u003e114 (3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e52 (4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e17 (2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e45 (3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 192px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Other/Un \u0026amp; English Preference\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 102px;\"\u003e\n \u003cp\u003e317 (8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e104 (8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e66 (8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e147 (8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 192px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Other/Un \u0026amp; Non-English Preference\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 102px;\"\u003e\n \u003cp\u003e67 (2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e31 (2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e6 (1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e30 (2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 192px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;White\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 102px;\"\u003e\n \u003cp\u003e941 (24)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e295 (23)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e275 (35)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e371 (21)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" style=\"width: 570px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eRace or Ethnicity (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 192px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Asian\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 102px;\"\u003e\n \u003cp\u003e1,528 (40)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e472 (37)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e205 (26)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e851 (48)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 192px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Black/African American\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 102px;\"\u003e\n \u003cp\u003e527 (14)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e216 (17)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e141 (18)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e170 (10)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 192px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Hispanic/Latino\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 102px;\"\u003e\n \u003cp\u003e480 (12)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e161 (13)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e103 (13)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e216 (12)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 192px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Other/Unknown\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 102px;\"\u003e\n \u003cp\u003e384 (10)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e135 (11)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e72 (9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e177 (10)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 192px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;White\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 102px;\"\u003e\n \u003cp\u003e941 (24)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e295 (23)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e275 (35)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e371 (21)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" style=\"width: 570px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eLanguage Preference (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 192px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Chinese\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 102px;\"\u003e\n \u003cp\u003e296 (8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e93 (7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e25 (3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e178 (10)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 192px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;English\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 102px;\"\u003e\n \u003cp\u003e3,154 (82)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e1,029 (80)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e714 (90)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e1,411 (79)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 192px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Other\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 102px;\"\u003e\n \u003cp\u003e298 (8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e105 (8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e42 (5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e151 (8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 192px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Spanish\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 102px;\"\u003e\n \u003cp\u003e112 (3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e52 (4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e15 (2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e45 (3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" style=\"width: 570px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eInsurance (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\n \u003cp\u003e0.031\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 192px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Commercial\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 102px;\"\u003e\n \u003cp\u003e1,708 (44)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e559 (44)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e329 (41)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e820 (46)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 192px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Medicaid\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 102px;\"\u003e\n \u003cp\u003e502 (13)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e187 (15)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e120 (15)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e195 (11)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 192px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Medicare\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 102px;\"\u003e\n \u003cp\u003e1,644 (43)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e531 (42)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e347 (44)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e766 (43)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 192px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Uninsured\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 102px;\"\u003e\n \u003cp\u003e1 (0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 96px;\"\u003e\n \u003cp\u003e0 (0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 96px;\"\u003e\n \u003cp\u003e0 (0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 84px;\"\u003e\n \u003cp\u003e1 (0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 192px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Missing\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 102px;\"\u003e\n \u003cp\u003e5 (0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 96px;\"\u003e\n \u003cp\u003e2 (0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 96px;\"\u003e\n \u003cp\u003e0 (0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 84px;\"\u003e\n \u003cp\u003e3 (0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" style=\"width: 570px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eNeighborhood SES (1=lowest) (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\n \u003cp\u003e0.056\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 192px;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 102px;\"\u003e\n \u003cp\u003e651 (17)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e221 (17)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e157 (20)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e273 (15)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 192px;\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 102px;\"\u003e\n \u003cp\u003e689 (18)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e245 (19)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e144 (18)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e300 (17)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 192px;\"\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 102px;\"\u003e\n \u003cp\u003e752 (19)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e232 (18)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e147 (18)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e373 (21)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 192px;\"\u003e\n \u003cp\u003e4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 102px;\"\u003e\n \u003cp\u003e886 (23)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e304 (24)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e158 (20)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e424 (24)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 192px;\"\u003e\n \u003cp\u003e5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 102px;\"\u003e\n \u003cp\u003e698 (18)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e219 (17)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e149 (19)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e330 (18)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 192px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Missing\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 102px;\"\u003e\n \u003cp\u003e184 (5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e58 (5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e41 (5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e85 (5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" style=\"width: 570px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eCharlson Score (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 192px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;0-2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 102px;\"\u003e\n \u003cp\u003e2,965 (77)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e939 (73)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e539 (68)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e1,487 (83)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 192px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;3+\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 102px;\"\u003e\n \u003cp\u003e895 (23)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e340 (27)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e257 (32)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e298 (17)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" style=\"width: 570px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBaseline A1c Control (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 192px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Uncontrolled A1c (\u0026gt;=8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 102px;\"\u003e\n \u003cp\u003e818 (21)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e289 (23)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e192 (24)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e337 (19)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 192px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Controlled A1c (\u0026lt;8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 102px;\"\u003e\n \u003cp\u003e2,611 (68)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e849 (66)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e549 (69)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e1,213 (68)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 192px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Missing\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 102px;\"\u003e\n \u003cp\u003e431 (11)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e141 (11)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e55 (7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e235 (13)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" style=\"width: 570px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBaseline BP Control (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\n \u003cp\u003e0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 192px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;BP \u0026lt;=120/80\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 102px;\"\u003e\n \u003cp\u003e658 (17)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e220 (17)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e147 (18)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e291 (16)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 192px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;BP \u0026lt;=140/90\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 102px;\"\u003e\n \u003cp\u003e2,126 (55)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e716 (56)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e430 (54)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e980 (55)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 192px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;BP \u0026gt;140/90\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 102px;\"\u003e\n \u003cp\u003e930 (24)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e301 (24)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e206 (26)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e423 (24)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 192px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Missing\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 102px;\"\u003e\n \u003cp\u003e146 (4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e42 (3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e13 (2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e91 (5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" style=\"width: 570px;\"\u003e\n \u003cp\u003e\u003cstrong\u003ePatient Portal Enrollment (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 192px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;No\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 102px;\"\u003e\n \u003cp\u003e861 (22)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e293 (23)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e97 (12)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e471 (26)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 192px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Yes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 102px;\"\u003e\n \u003cp\u003e2,999 (78)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e986 (77)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e699 (88)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 84px;\"\u003e\n \u003cp\u003e1,314 (74)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 192px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Missing\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 102px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 96px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 96px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 84px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eSES: socioeconomic status, A1c: hemoglobin A1c lab result, BP: blood pressure\u003c/p\u003e\n\u003cp\u003e*P values compare patient characteristics by care trajectory based on Chi-squared test for categorical measures\u0026emsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 4. Descriptions of SFHN Healthcare Trajectories by Cluster\u003c/strong\u003e\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"726\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"2\" style=\"width: 108px;\"\u003e\n \u003cp\u003e\u003cstrong\u003ePre-Pandemic Utilization\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" style=\"width: 60px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eCluster\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" style=\"width: 78px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eN (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" style=\"width: 354px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eHow Each Care Pattern Changed Relative to Pre-Pandemic\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" style=\"width: 126px;\"\u003e\n \u003cp\u003e\u003cstrong\u003ePost-Pandemic Care Trajectory\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 114px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eIn-person\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 108px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTelehealth\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 132px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBetween Visits\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 108px;\"\u003e\n \u003cp\u003eHigh Utilizers\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 60px;\"\u003e\n \u003cp\u003eSFHN 1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e319 (4.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 114px;\"\u003e\n \u003cp\u003eReturned to pre levels\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 108px;\"\u003e\n \u003cp\u003eModerate initial uptake, some sustained use\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 132px;\"\u003e\n \u003cp\u003eContinued high usage with slight increase overall\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 126px;\"\u003e\n \u003cp\u003eAdded Telehealth and Between-Visit Care\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 108px;\"\u003e\n \u003cp\u003eHigh Utilizers\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 60px;\"\u003e\n \u003cp\u003eSFHN 2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e470 (6.9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 114px;\"\u003e\n \u003cp\u003eReturned to pre levels\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 108px;\"\u003e\n \u003cp\u003eLow uptake not sustained\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 132px;\"\u003e\n \u003cp\u003eModerate uptake with slight increase overall\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 126px;\"\u003e\n \u003cp\u003eAdded Between-Visit Care\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 108px;\"\u003e\n \u003cp\u003eHigh Utilizers\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 60px;\"\u003e\n \u003cp\u003eSFHN 3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e285 (4.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 114px;\"\u003e\n \u003cp\u003eDid not recover to pre levels\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 108px;\"\u003e\n \u003cp\u003eHigh initial uptake, moderate sustained use\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 132px;\"\u003e\n \u003cp\u003eContinued high usage (no sustained increase)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 126px;\"\u003e\n \u003cp\u003eTransitioned to Telehealth\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 108px;\"\u003e\n \u003cp\u003eModerate Utilizers\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 60px;\"\u003e\n \u003cp\u003eSFHN 4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e364 (5.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 114px;\"\u003e\n \u003cp\u003eDid not recover to pre levels\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 108px;\"\u003e\n \u003cp\u003eHigh initial uptake, moderate sustained use\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 132px;\"\u003e\n \u003cp\u003eContinued high usage with slight increase overall\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 126px;\"\u003e\n \u003cp\u003eTransitioned to Telehealth and Between-Visit Care\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 108px;\"\u003e\n \u003cp\u003eModerate Utilizers\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 60px;\"\u003e\n \u003cp\u003eSFHN 5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e707 (10.4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 114px;\"\u003e\n \u003cp\u003eReturned to pre levels\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 108px;\"\u003e\n \u003cp\u003eLow uptake not sustained\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 132px;\"\u003e\n \u003cp\u003eLow uptake with increase overall\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 126px;\"\u003e\n \u003cp\u003eMinimal Change\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 108px;\"\u003e\n \u003cp\u003eModerate Utilizers\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 60px;\"\u003e\n \u003cp\u003eSFHN 6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e973 (14.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 114px;\"\u003e\n \u003cp\u003eDid not recover to pre levels\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 108px;\"\u003e\n \u003cp\u003eModerate uptake not sustained\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 132px;\"\u003e\n \u003cp\u003eModerate usage, decreasing trends\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 126px;\"\u003e\n \u003cp\u003eLost Care\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 108px;\"\u003e\n \u003cp\u003eLow Utilizers\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 60px;\"\u003e\n \u003cp\u003eSFHN 7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e3693 (54.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 114px;\"\u003e\n \u003cp\u003eDid not recover to pre levels\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 108px;\"\u003e\n \u003cp\u003eLow uptake not sustained\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 132px;\"\u003e\n \u003cp\u003eLow, consistent non usage\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 126px;\"\u003e\n \u003cp\u003eLost Care\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eWe defined adding digital care as in-person visit rates returning to pre-pandemic levels, while telehealth and/or between-visit care remain elevated. Those who transitioned to digital care saw in-person visit rates remain below pre-pandemic levels, while telehealth and between-visit care increased to replace in-person care. Those who lost care had minimal uptake of between-visit care or telehealth, even though in-person care rates did not rebound. Those with minimal change experienced minimal changes to their care utilization patterns from before the pandemic. We also further specified whether those who added or transitioned to digital care primarily adopted Telehealth, Between-Visit Care, or both.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 5. Sociodemographic Characteristics of Care Trajectories at SFHN\u003c/strong\u003e\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"686\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"2\" valign=\"bottom\" style=\"width: 186px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" style=\"width: 78px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTotal\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 90px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAdded Digital Care\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 96px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTransitioned to Digital Care\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 90px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eLost Care\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 83px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eMinimal Change\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" style=\"width: 62px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eP-value*\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 90px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eClusters 1,2\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 96px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eClusters 3, 4\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 90px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eClusters 6, 7\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 83px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eCluster 5\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 186px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003cstrong\u003eN (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003eN=6,811\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 90px;\"\u003e\n \u003cp\u003eN=789 (12)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 96px;\"\u003e\n \u003cp\u003eN=649 (10)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 90px;\"\u003e\n \u003cp\u003eN=4,666 (69)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 83px;\"\u003e\n \u003cp\u003eN=707 (10)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"6\" style=\"width: 623px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAge Category (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 62px;\"\u003e\n \u003cp\u003e0.037\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 186px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;18-34\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 78px;\"\u003e\n \u003cp\u003e162 (2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e10 (1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e16 (2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e123 (3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 83px;\"\u003e\n \u003cp\u003e13 (2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 186px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;35-49\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 78px;\"\u003e\n \u003cp\u003e1,080 (16)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e118 (15)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e101 (16)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e723 (15)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 83px;\"\u003e\n \u003cp\u003e138 (20)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 186px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;50-64\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 78px;\"\u003e\n \u003cp\u003e3,215 (47)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e378 (48)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e308 (47)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e2,201 (47)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 83px;\"\u003e\n \u003cp\u003e328 (46)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 186px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;65-74\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 78px;\"\u003e\n \u003cp\u003e1,727 (25)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e201 (25)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e156 (24)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e1,214 (26)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 83px;\"\u003e\n \u003cp\u003e156 (22)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 186px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;75+\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 78px;\"\u003e\n \u003cp\u003e627 (9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e82 (10)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e68 (10)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e405 (9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 83px;\"\u003e\n \u003cp\u003e72 (10)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"6\" style=\"width: 623px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eSex (%)\u003c/strong\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 62px;\"\u003e\n \u003cp\u003e0.094\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 186px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Female\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 78px;\"\u003e\n \u003cp\u003e3,581 (53)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e394 (50)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e372 (57)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e2,463 (53)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 83px;\"\u003e\n \u003cp\u003e352 (50)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 186px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Male\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 78px;\"\u003e\n \u003cp\u003e3,229 (47)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e395 (50)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e277 (43)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e2,202 (47)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 83px;\"\u003e\n \u003cp\u003e355 (50)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"6\" style=\"width: 623px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eJoint Race Ethnicity \u0026amp; Language (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 62px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 186px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Asian \u0026amp; English Preference\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 78px;\"\u003e\n \u003cp\u003e641 (9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e62 (8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e62 (10)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e446 (10)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 83px;\"\u003e\n \u003cp\u003e71 (10)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 186px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Asian \u0026amp; Non-English Preference\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 78px;\"\u003e\n \u003cp\u003e1,753 (26)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e94 (12)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e85 (13)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e1,442 (31)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 83px;\"\u003e\n \u003cp\u003e132 (19)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 186px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Black/African American\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 78px;\"\u003e\n \u003cp\u003e1,027 (15)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e116 (15)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e176 (27)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e659 (14)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 83px;\"\u003e\n \u003cp\u003e76 (11)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 186px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Hispanic \u0026amp; English Preference\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 78px;\"\u003e\n \u003cp\u003e449 (7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e57 (7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e64 (10)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e276 (6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 83px;\"\u003e\n \u003cp\u003e52 (7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 186px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Hispanic \u0026amp; Non-English Preference\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 78px;\"\u003e\n \u003cp\u003e1,881 (28)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e340 (43)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e103 (16)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e1,137 (24)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 83px;\"\u003e\n \u003cp\u003e301 (43)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 186px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Other/Un \u0026amp; English Preference\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 78px;\"\u003e\n \u003cp\u003e254 (4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e24 (3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e33 (5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e173 (4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 83px;\"\u003e\n \u003cp\u003e24 (3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 186px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Other/Un \u0026amp; Non-English Preference\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 78px;\"\u003e\n \u003cp\u003e84 (1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e13 (2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e5 (1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e56 (1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 83px;\"\u003e\n \u003cp\u003e10 (1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 186px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;White\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 78px;\"\u003e\n \u003cp\u003e722 (11)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e83 (11)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e121 (19)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e477 (10)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 83px;\"\u003e\n \u003cp\u003e41 (6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"6\" style=\"width: 623px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eRace or Ethnicity (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 62px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 186px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Asian\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 78px;\"\u003e\n \u003cp\u003e2,394 (35)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e156 (20)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e147 (23)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e1,888 (40)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 83px;\"\u003e\n \u003cp\u003e203 (29)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 186px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Black/African American\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 78px;\"\u003e\n \u003cp\u003e1,027 (15)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e116 (15)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e176 (27)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e659 (14)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 83px;\"\u003e\n \u003cp\u003e76 (11)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 186px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Hispanic/Latino\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 78px;\"\u003e\n \u003cp\u003e2,330 (34)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e397 (50)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e167 (26)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e1,413 (30)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 83px;\"\u003e\n \u003cp\u003e353 (50)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 186px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Other/Unknown\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 78px;\"\u003e\n \u003cp\u003e338 (5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e37 (5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e38 (6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e229 (5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 83px;\"\u003e\n \u003cp\u003e34 (5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 186px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;White\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 78px;\"\u003e\n \u003cp\u003e722 (11)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e83 (11)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e121 (19)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e477 (10)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 83px;\"\u003e\n \u003cp\u003e41 (6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"6\" style=\"width: 623px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eLanguage Preference (%)\u003c/strong\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 62px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 186px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Chinese\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 78px;\"\u003e\n \u003cp\u003e1,324 (19)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e52 (7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e67 (10)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e1,130 (24)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 83px;\"\u003e\n \u003cp\u003e75 (11)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 186px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;English\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 78px;\"\u003e\n \u003cp\u003e2,931 (43)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e335 (42)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e437 (67)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e1,908 (41)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 83px;\"\u003e\n \u003cp\u003e251 (36)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 186px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Other\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 78px;\"\u003e\n \u003cp\u003e655 (10)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e55 (7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e44 (7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e480 (10)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 83px;\"\u003e\n \u003cp\u003e76 (11)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 186px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Spanish\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 78px;\"\u003e\n \u003cp\u003e1,901 (28)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e347 (44)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e101 (16)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e1,148 (25)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 83px;\"\u003e\n \u003cp\u003e305 (43)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"6\" style=\"width: 623px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eInsurance (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 62px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 186px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Commercial\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 78px;\"\u003e\n \u003cp\u003e71 (1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e6 (1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e6 (1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e53 (1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 83px;\"\u003e\n \u003cp\u003e6 (1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 186px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Healthy Workers\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 78px;\"\u003e\n \u003cp\u003e1,049 (15)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e44 (6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e53 (8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e875 (19)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 83px;\"\u003e\n \u003cp\u003e77 (11)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 186px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Medicaid\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 78px;\"\u003e\n \u003cp\u003e2,709 (40)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e364 (46)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e274 (42)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e1,806 (39)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 83px;\"\u003e\n \u003cp\u003e265 (37)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 186px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Medicare\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 78px;\"\u003e\n \u003cp\u003e2,126 (31)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e283 (36)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e278 (43)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e1,359 (29)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 83px;\"\u003e\n \u003cp\u003e206 (29)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 186px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Uninsured\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e828 (12)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 90px;\"\u003e\n \u003cp\u003e92 (12)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 96px;\"\u003e\n \u003cp\u003e38 (6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e546 (12)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 83px;\"\u003e\n \u003cp\u003e152 (21)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 186px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Missing\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e28 (0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 90px;\"\u003e\n \u003cp\u003e0 (0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 96px;\"\u003e\n \u003cp\u003e0 (0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 90px;\"\u003e\n \u003cp\u003e27 (1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 83px;\"\u003e\n \u003cp\u003e1 (0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"6\" style=\"width: 623px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eNeighborhood SES (1=lowest) (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 62px;\"\u003e\n \u003cp\u003e0.053\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 186px;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 78px;\"\u003e\n \u003cp\u003e2,561 (38)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e306 (39)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e261 (40)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e1,728 (37)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 83px;\"\u003e\n \u003cp\u003e266 (38)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 186px;\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 78px;\"\u003e\n \u003cp\u003e1,771 (26)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e227 (29)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e147 (23)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e1,199 (26)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 83px;\"\u003e\n \u003cp\u003e198 (28)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 186px;\"\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 78px;\"\u003e\n \u003cp\u003e1,049 (15)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e122 (15)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e88 (14)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e737 (16)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 83px;\"\u003e\n \u003cp\u003e102 (14)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 186px;\"\u003e\n \u003cp\u003e4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 78px;\"\u003e\n \u003cp\u003e889 (13)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e78 (10)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e92 (14)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e634 (14)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 83px;\"\u003e\n \u003cp\u003e85 (12)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 186px;\"\u003e\n \u003cp\u003e5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 78px;\"\u003e\n \u003cp\u003e405 (6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e40 (5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e48 (7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e280 (6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 83px;\"\u003e\n \u003cp\u003e37 (5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 186px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Missing\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 78px;\"\u003e\n \u003cp\u003e136 (2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e16 (2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e13 (2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e88 (2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 83px;\"\u003e\n \u003cp\u003e19 (3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"6\" style=\"width: 623px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eCharlson Score (%)\u003c/strong\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 62px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 186px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;0-2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 78px;\"\u003e\n \u003cp\u003e5,818 (85)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e575 (73)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e477 (73)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e4,148 (89)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 83px;\"\u003e\n \u003cp\u003e618 (87)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 186px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;3+\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 78px;\"\u003e\n \u003cp\u003e993 (15)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e214 (27)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e172 (27)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e518 (11)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 83px;\"\u003e\n \u003cp\u003e89 (13)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"6\" style=\"width: 623px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBaseline A1c Control (%)\u003c/strong\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 62px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 186px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Uncontrolled A1c (\u0026gt;=8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 78px;\"\u003e\n \u003cp\u003e2,631 (39)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e376 (48)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e272 (42)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e1,677 (36)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 83px;\"\u003e\n \u003cp\u003e306 (43)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 186px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Controlled A1c (\u0026lt;8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 78px;\"\u003e\n \u003cp\u003e3,307 (49)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e364 (46)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e337 (52)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e2,273 (49)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 83px;\"\u003e\n \u003cp\u003e333 (47)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 186px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Missing\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 78px;\"\u003e\n \u003cp\u003e873 (13)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e49 (6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e40 (6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e716 (15)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 83px;\"\u003e\n \u003cp\u003e68 (10)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"6\" style=\"width: 623px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBaseline BP Control (%)\u003c/strong\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 62px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 186px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;BP \u0026lt;=120/80\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 78px;\"\u003e\n \u003cp\u003e1,629 (24)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e198 (25)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e143 (22)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e1,134 (24)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 83px;\"\u003e\n \u003cp\u003e154 (22)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 186px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;BP \u0026lt;=140/90\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 78px;\"\u003e\n \u003cp\u003e3,598 (53)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e427 (54)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e351 (54)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e2,418 (52)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 83px;\"\u003e\n \u003cp\u003e402 (57)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 186px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;BP \u0026gt;140/90\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 78px;\"\u003e\n \u003cp\u003e1,205 (18)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e147 (19)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e149 (23)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e783 (17)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 83px;\"\u003e\n \u003cp\u003e126 (18)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 186px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Missing\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 78px;\"\u003e\n \u003cp\u003e379 (6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e17 (2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e6 (1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e331 (7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 83px;\"\u003e\n \u003cp\u003e25 (4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"6\" style=\"width: 623px;\"\u003e\n \u003cp\u003e\u003cstrong\u003ePatient Portal Enrollment (%)\u003c/strong\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 62px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 186px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;No\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 78px;\"\u003e\n \u003cp\u003e6,217 (91)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e724 (92)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e524 (81)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e4,293 (92)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 83px;\"\u003e\n \u003cp\u003e676 (96)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 186px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Yes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 78px;\"\u003e\n \u003cp\u003e594 (9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e65 (8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 96px;\"\u003e\n \u003cp\u003e125 (19)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 90px;\"\u003e\n \u003cp\u003e373 (8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 83px;\"\u003e\n \u003cp\u003e31 (4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eSES: socioeconomic status, A1c: hemoglobin A1c lab result, BP: blood pressure\u003c/p\u003e\n\u003cp\u003e*P values compare patient characteristics by \u0026nbsp;care trajectory based on Chi-squared test for categorical measures\u0026emsp;\u003c/p\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"npj-digital-medicine","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"npjdigitalmed","sideBox":"Learn more about [npj Digital Medicine](http://www.nature.com/npjdigitalmed/)","snPcode":"41746","submissionUrl":"https://submission.springernature.com/new-submission/41746/3","title":"npj Digital Medicine","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"NPJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Sequence analysis, health disparities, telehealth, between-visit care, healthcare utilization","lastPublishedDoi":"10.21203/rs.3.rs-8176924/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8176924/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThe COVID-19 pandemic propelled the growth of virtual care, but few studies have examined how it reshaped care utilization patterns longitudinally across diverse populations. We used multichannel sequence analysis to classify 10,671 diabetic adults at two San Francisco health systems by utilization patterns from April 2019-March 2023. Patients who transitioned to digital care were Black or White, were Medicare beneficiaries, or had higher comorbidity burdens. Individuals who increased their combined in-person and virtual healthcare utilization were Hispanic with Spanish language preference and had greater disease burden. Those who decreased care utilization overall were Asian with Chinese language preference and had low patient portal engagement. While technology-enabled care models have the potential to enhance access, certain populations may be at risk of reduced access to care. Health systems should consider targeted interventions to ensure equitable access to telehealth for populations at risk of digital exclusion with the growth of remote care.\u003c/p\u003e","manuscriptTitle":"Trajectories of In-Person and Remote Care Utilization: A Sequence Analysis of Primary Care Modalities Pre- and Post-COVID-19","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-12-15 16:10:20","doi":"10.21203/rs.3.rs-8176924/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-02-01T22:12:36+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-01-15T03:54:34+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-12-25T00:06:09+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"33521791823848697217371066972108862814","date":"2025-12-15T04:01:54+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"143943901637558803171960020770334069717","date":"2025-12-11T19:19:26+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-12-09T19:07:43+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-11-29T20:02:37+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-11-29T06:57:11+00:00","index":"","fulltext":""},{"type":"submitted","content":"npj Digital Medicine","date":"2025-11-21T23:15:59+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"npj-digital-medicine","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"npjdigitalmed","sideBox":"Learn more about [npj Digital Medicine](http://www.nature.com/npjdigitalmed/)","snPcode":"41746","submissionUrl":"https://submission.springernature.com/new-submission/41746/3","title":"npj Digital Medicine","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"NPJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"b4027187-dbf4-42a6-b67c-0e83c7fc1650","owner":[],"postedDate":"December 15th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":59569272,"name":"Health sciences/Diseases"},{"id":59569273,"name":"Health sciences/Health care"},{"id":59569274,"name":"Health sciences/Medical research"}],"tags":[],"updatedAt":"2026-05-06T23:23:19+00:00","versionOfRecord":[],"versionCreatedAt":"2025-12-15 16:10:20","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8176924","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8176924","identity":"rs-8176924","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}