Investigating Racial Disparities in Drug Prescriptions for Patients with Endometriosis

Background: Endometriosis is a chronic disease with a long time to diagnosis and several known comorbidities that requires a range of treatments including of pain management and hormone-based medications. Racial disparities specific to endometriosis treatments are unknown.

We aim to investigate differences in patterns of drug prescriptions specific to endometriosis management in Black and White patients prior to diagnosis and after diagnosis of endometriosis and compare these differences to racial disparities established in the general population. Study Design: We conduct a retrospective cohort study using observational health data from the IBM MarketScan® Multi-state Medicaid dataset. We identify a cohort of endometriosis patients consisting of women between the ages of 15 and 49 with an endometriosis-related surgical procedure and a diagnosis code for endometriosis within 30 days of this procedure. Cohort is further restricted to patients with at least 3 years of continuous observation prior to diagnosis. We identify a non-endometriosis cohort of women between the ages of 15 and 49 with no endometriosis diagnosis and at least 1 year of continuous observation. We compare prevalence of prescriptions across selected drug classes for Black vs. White endometriosis patients. We further examine prevalence differences in the non-endometriosis cohort and prevalence differences pre- and post-diagnosis in the endometriosis cohort.

The endometriosis cohort comprised 16,372 endometriosis patients (23.3% Black, 66.0% White). Of the 28 drug classes examined, 17 were prescribed significantly less in Black patients compared to 21 in non-endometriosis cohort (n=3,663,904), and 4 were prescribed significantly more in Black patients compared to 6 in the non-endometriosis cohort. Of the 17 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted October 3, 2023. ; https://doi.org/10.1101/2023.10.02.23296435doi: medRxiv preprint 3 drugs prescribed more often in White patients, 16 have larger disparities pre-diagnosis than post- diagnosis.

Our analysis identified significant differences in medication prescriptions between White and Black patients with endometriosis, notably in hormonal treatments, pain management, and treatments for common endometriosis co-morbidities. Racial disparities in drug prescriptions are well established in healthcare, and better understanding these disparities in the specific context of chronic reproductive conditions and chronic pain is important for increasing equity in drug prescription practices.

Health disparities, Reproductive healthcare, observational study, biomedical informatics, prescription patterns, pain medication, chronic disease, temporal analysis All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted October 3, 2023. ; https://doi.org/10.1101/2023.10.02.23296435doi: medRxiv preprint 4

Endometriosis is a chronic, inflammatory disease characterized by growth of endometrial-like tissue outside the uterus. Identified symptoms of endometriosis vary widely and include dysmenorrhea, fatigue, non-menstrual abdominopelvic pain, and heavy menstrual bleeding1,2. While the precise prevalence of endometriosis is unknown, it affects an estimated 6- 10% of women of reproductive age2–4. Current medical interventions for endometriosis depend on patient prognosis and include analgesics, combined hormonal contraceptives, progestogens, gonadotropin-releasing hormone (GnRH) agonists, GnRH antagonists, aromatase inhibitors, and surgical treatments5. Racial health disparities in the United States are pervasive, and research shows that Black and Hispanic patients are less likely to receive a diagnosis of endometriosis compared to White patients6, as well as less likely to receive surgical treatment for endometriosis7. Racial bias is observed beyond endometriosis diagnosis, such as in pain management; Black and Hispanic patients are less likely to be treated for pain and other symptoms that impact quality of life8–11. In this work we characterize the drug prescription patterns for endometriosis patients and potential differences in prevalence across Black and White patients. Because endometriosis is a condition with delays in diagnosis, patients with suspected endometriosis are often prescribed first-line treatments prior to their diagnosis. We thus characterize treatment patterns in three ways: prior to diagnosis, after diagnosis, and overall. Furthermore, to account for multiple co- morbidities of endometriosis, we extend our analysis to all medication classes that are significantly more prescribed in our endometriosis cohort than in a non-endometriosis comparison cohort. To further contextualize this analysis, we compare treatment patterns of endometriosis patients with those without endometriosis and measure differences in treatment All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted October 3, 2023. ; https://doi.org/10.1101/2023.10.02.23296435doi: medRxiv preprint 5 patterns across White and Black patients in a non-endometriosis cohort, namely all women in reproductive age without any diagnosis of endometriosis.

The analysis and use of the de-identified dataset presented in this work were carried out under Research Protocol AAAO7805 approved by the Columbia University IRB. Dataset In this retrospective cohort study, we focus our analysis to patients with Medicaid, the U.S. government program that provides health insurance for people with limited income. The data comes from the IBM MarketScan® Multi-state Medicaid dataset, which draws Medicaid data from several states12. The Medicaid dataset contains de-identified longitudinal records of patients and includes inpatient and outpatient services, diagnostic history, and drug prescriptions for over 48 million enrollees12. We leverage the data under the OMOP Common Data Model (CDM) format, which follows standardized conventions for drugs, therapies, and other medical vocabularies13. The database has been used for a variety of observational health studies due to its flexibility and robustness14–16. Cohort Identification We select patients with endometriosis using a validated cohort definition17. The cohort definition includes women ages 15 to 49 years who have an endometriosis-related surgical procedure (e.g., laparoscopic surgery) as well as a diagnosis of endometriosis 30 days before or after this procedure. The phenotype definition was validated through manual chart review of 1,400 endometriosis patients yielding 70% sensitivity, 93% specificity, 85% positive predictive All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted October 3, 2023. ; https://doi.org/10.1101/2023.10.02.23296435doi: medRxiv preprint 6 value. Age is calculated at time of diagnosis. Only patients with endometriosis with at least three years of continuous observation prior to diagnosis are considered for the endometriosis cohort. The non-endometriosis cohort comprises of any woman aged 15-49 with no diagnostic code for endometriosis at any point in their medical history and at least one year of continuous observation. Age is calculated using the most recent visit date. In the claims database, it is unknown for individual patients whether the gender marker represents sex, gender, or both. Identification of medications The Anatomical Therapeutic Chemical (ATC) classification system separates drugs in a standardized manner; we create drug classes using ATC level 3 classifications because they are associated with specific pharmacological functions without specifying chemical structures18,19. We use the OMOP CDM which explicitly maps medications to their ATC level 3 classification. The ATC classifications create a “bridge” between the clinical guidelines for endometriosis and large-scale observational health data. We narrow the set of drug classes by identifying drug classes that are prescribed significantly more often in the endometriosis cohort than the non-endometriosis one. For each drug class, the proportion of patients in each cohort with at least one prescription belonging to that drug class is calculated. The statistical significance of the difference between these relative prevalence measurements is calculated using a two-sided Z-test with a 0.01 significance level. This procedure identified 28 drug classes that are significantly more prevalent in the endometriosis cohort than the non-endometriosis cohort (see Supplemental Table 1); we then measure prescription prevalence for each of these drug classes. To understand the clinical relevance of these 28 drug classes, we checked them against the drugs listed in the 2022 ESHRE guidelines for treating endometriosis5. The only All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted October 3, 2023. ; https://doi.org/10.1101/2023.10.02.23296435doi: medRxiv preprint 7 recommended treatments that are not included in our analysis are GnRH antagonists and aromatase inhibitors. GnRH antagonists were not prescribed in the Medicaid dataset, and aromatase inhibitors were excluded because the corresponding ATC 3rd level drug class (“hormone antagonists and related agents”) was not prescribed significantly more in patients with endometriosis than in the non-endometriosis cohort. Several other drug classes more prevalently prescribed in the endometriosis population are associated with known comorbidities or symptoms of endometriosis including gastrointestinal distress, depression, and anxiety23,24. Temporal Analysis Endometriosis is known to have extended delays between symptom onset and diagnosis20,21. We therefore aim to better understand the patterns of drug prescription prior to diagnosis (which occurs via laparoscopic surgery) and whether they differ from prescription practices after diagnosis. Prescription practices from the pre-diagnosis period can be thought of as prescription practices for undiagnosed endometriosis patients, which is an important area of study given that Black patients are less likely to be diagnosed with endometriosis compared to White patients6. We create two temporal subgroups of drug prescriptions in the endometriosis cohort (see Figure 1). Subgroup 1 consists of all prescriptions occurred prior to surgical diagnosis. Subgroup 2 consists of all prescriptions post-diagnosis. Note that neither subgroup 1 nor subgroup 2 contain prescriptions given on the day of diagnosis; these prescriptions are only counted in the “overall” prescriptions. Statistical Methods For each drug class of interest, the relative prevalence of prescriptions within that drug class was calculated for Black and White endometriosis patients. A person is counted as part of All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted October 3, 2023. ; https://doi.org/10.1101/2023.10.02.23296435doi: medRxiv preprint 8 the drug-positive group if they received at least one prescription that fell into the given drug class, and relative prevalence is calculated based on the total number of Black (n = 3,814) and White (n = 10,805) patients in the endometriosis cohort. The difference in prescription prevalence is calculated as 100% ∗ (𝑝!"#$% − 𝑝&'()* ), where 𝑝!"#$% and 𝑝&'()* relative prevalence (see Table 2, Table 3). To identify significant differences in the Black and White sub-populations, we conduct a two-sided Z-test from the difference in prevalence and the associated standard error. We apply the Bonferroni correction to account for multiple comparisons; we consider the difference significant when the adjusted p-value < 0.01. This procedure for statistical analysis was repeated for all prescription classes and both temporal subgroups (pre-diagnosis and post-diagnosis). Analysis was carried out using the scipy package in Python22.

Cohorts The non-endometriosis cohort comprises 3,663,904 patients (see Table 1). The endometriosis cohort (n=16,372) comprised more White (66%) than the non-endometriosis cohort (23%) (of note, 50% of patients in the non-endometriosis cohort have no recorded race). The endometriosis cohort also was slightly older (mean age 32.4 ± 8.0 years) than the non- endometriosis cohort (mean age 28.7 ± 10.3 years). The median length of observational period was 6.08 (IQR: 5.08 to 8.84) years for the endometriosis cohort and 4.00 (IQR: 2.00 to 6.25) years for the non-endometriosis cohort. For the endometriosis cohort, the median length of observation prior to diagnosis was 4.20 (IQR: 3.26 to 5.73) years, and the median length of observation post-diagnosis was 1.68 (IQR: 0.82 to 2.89) years. Differences in overall prevalence of drug prescriptions for Black and White patients All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted October 3, 2023. ; https://doi.org/10.1101/2023.10.02.23296435doi: medRxiv preprint 9 Of the 28 drug classes of interest, 17 (61%) are prescribed significantly more often for White endometriosis patients compared to Black endometriosis patients (see Figure 2). In the non-endometriosis cohort, 21 (75%) are prescribed significantly more for White patients than for Black patients and 6 (21%) are prescribed significantly more for Black patients. The largest racial differences in endometriosis patients occur for antidepressants (prescription prevalence higher by 20.7% in White patients than Black patients, 95% CI [19.8%, 21.6%]), anxiolytics (20.4%, 95% CI [19.5%, 21.3%]), antiepileptics (18.7%, 95% CI [17.8%, 19.6%]), and estrogens (10.2%, 95% CI [9.5%, 11.0%]). Additionally, 4 drug classes are prescribed significantly more often in Black patients than White patients, with the largest differences occurring for iron preparations (7.3%, 95% CI [6.6%, 8.0%]), hormones and related agents (7.1%, 95% CI [6.2%, 8.1%]), and antiinfectives/antiseptics (7.0%, 95% CI [6.1%, 7.9%]). Prescription prevalence differences are more pronounced in the endometriosis cohort than the non-endometriosis cohort for anxiolytics, antiepileptics, estrogens, iron preparations, and antiinfectives/antiseptics but are less pronounced in the endometriosis cohort for antidepressants and hormones and related agents (see Table 2). Out of the drug classes related to established endometriosis treatment options, three (opioids, other analgesics and antipyretics, and hormonal contraceptives for systemic use) are significantly more prevalent amongst White endometriosis patients; the remaining two (hormones and related agents, progestogens) are significantly more prevalent amongst Black endometriosis patients (see Table 2). These differences are generally consistent with the non- endometriosis cohort, where analgesics are more commonly prescribed for White women and hormones and related agents are more commonly prescribed in Black women. One major difference between the endometriosis and non-endometriosis cohorts, however, is in hormonal All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted October 3, 2023. ; https://doi.org/10.1101/2023.10.02.23296435doi: medRxiv preprint 10 contraceptives for systemic use, which are prescribed significantly more in White women with endometriosis (3.3%) and significantly more for Black women without endometriosis (2.0%). Additionally, the difference in prescriptions of estrogens is much higher in the endometriosis cohort (10.23%) than in the non-endometriosis cohort (1.83%). Temporal differences in drug prescription differences When observing drug prescription differences across White and Black endometriosis patients across temporal subgroups, there is a pattern of pre-diagnosis differences (temporal subgroup 1) exceeding post-diagnosis differences (temporal subgroup 2), especially for drug classes where White patients are prescribed at a higher rate (see Figure 2 and Table 3). Of the 17 drug classes that fall into this category, 16 (94%) have larger differences prior to diagnosis compared to post-diagnosis. For drugs that are significantly more prevalently prescribed in Black patients, the difference in prevalence decreases from pre- to post-diagnosis for iron preparations and antiinfectives/antiseptics; the difference in prevalence is greater post-diagnosis for hormones and related agents and progestogens. Comment Principal Findings Our analysis of drug prescription patterns amongst White and Black patients with endometriosis in the Medicaid population shows racial disparities in disease treatment and pain management within endometriosis. Differences in prescription prevalence tend to be larger pre- diagnosis compared to post-diagnosis. We also find that drugs associated with comorbidities of endometriosis (including gastrointestinal distress, irritable bowel syndrome, thyroid disorders, and lower urinary tract symptoms25–27) are prescribed significantly less prevalently in Black patients compared to White patients, often with larger disparities in the endometriosis population All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted October 3, 2023. ; https://doi.org/10.1101/2023.10.02.23296435doi: medRxiv preprint 11 than the non-endometriosis population; this suggests disparities in holistic endometriosis management.

in the Context of What is Known Several disparities identified in this dataset are well-documented in patient populations not specific to endometriosis (e.g. prescriptions of antidepressants28,29, select anxiolytics30, and pain medication8–11). Prior studies have also observed racial disparities in endometriosis diagnosis6 and surgical treatments7; therefore, it becomes important to consider how lack of treatment and lack of diagnosis may collectively impact patients’ healthcare experiences. Additionally, when considering that most prescription disparities are larger pre-diagnosis compared to post-diagnosis, we conceptualize the pre-diagnosis period as one where the patient may require treatment but has no formal diagnosis to explain the multi-system, chronic symptoms they may be facing. Clinical Implications The most notable differences in prescription prevalence occur for antidepressants and anxiolytics, which are generally used to treat depression and anxiety. Depression and anxiety are known comorbidities of endometriosis and are associated with worse endometriosis symptoms23,24. Prior studies have found that Black patients experience less rapport-building during visits31 and are additionally less likely to communicate their symptoms (especially those related to mental health) to healthcare providers31,32; this suggests a gap in physician-patient communication which could be due to lack of cultural competency and/or perceived racism33. Perceived racism34 includes acts or attitudes that may or may not “objectively” considered racist but do contribute to the (subjective) experience of racial prejudice and is known to contribute to healthcare-related stress35–37. Future qualitative work should investigate how endometriosis All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted October 3, 2023. ; https://doi.org/10.1101/2023.10.02.23296435doi: medRxiv preprint 12 patients (who typically have more frequent interactions with the healthcare system) experience racism and the impact this has on their healthcare-related stress. We observe disparities in the prescription of pain medication (both opioid and non- opioid) where Black patients are less likely to receive this treatment than White patients. False racial assumptions and stereotypes such as the idea that Black patients asking for pain medications are “drug-seeking”38,39 or that Black patients feel less pain than their White counterparts8 may contribute to these disparities. Perceived racism can also impact how Black patients experience pain management: in addition to exacerbating healthcare-related stress, the racism experienced by Black patients when seeking chronic pain treatment may also contribute to hopelessness (negative expectations about one’s present life and future), which can negatively impact pain management10. Additionally, misconceptions and lack of effective communication between patients and physicians surrounding medication tolerance, side effects, efficacy, and addiction has been found to disproportionately impact racial and ethnic minorities37. Our finding that the disparities in pain medication prescriptions is greater in the pre-diagnosis cohort than the post-diagnosis subgroups suggests that the reasons for lack of prescriptions are exacerbated pre- diagnosis when patients have symptoms of, but are not explicitly diagnosed with, endometriosis compared to post-diagnosis, when their endometriosis has been formally diagnosed. Finally, when comparing racial prevalence disparities across the endometriosis and non- endometriosis cohorts, there was broad consistency; however, several of the drug classes associated with comorbidities of endometriosis (anxiolytics, urologicals, drugs for functional gastrointestinal disorders, thyroid preparations) had disparities that were larger in the endometriosis than the non-endometriosis cohort. There may be inequitable treatment of Black women specific to endometriosis, especially during their journey to diagnosis (when all these All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted October 3, 2023. ; https://doi.org/10.1101/2023.10.02.23296435doi: medRxiv preprint 13 disparities are larger). Notably, the medication class with the highest difference was Estrogens (see Table 2). There are potentially multiple indications to drugs under that medication class that may differ in the endometriosis and non-endometriosis cohorts, such as hormone replacement therapy for post-menopausal women vs. fertility treatments for endometriosis patients. Research Implications We identify several classes of drugs where there may be racial disparities in prescriptions specifically for endometriosis treatment; however, this work does not dive into the context of these drug prescription patterns, nor does it investigate these disparities in the context of patient experiences and quality of life. Future work should focus on better understanding the context of drug prescription patterns, especially co-occurring conditions associated with specific prescriptions or self-reported data about pain and how that relates to analgesic prescriptions for White and Black patients. Additionally, qualitative work should be conducted to understand the multifactorial reasons for our findings, to assess the impact of these disparities on patient lives, and to identify methods for mitigating these disparities. Finally, this analysis should be extended to other patient populations to identify disparities in other racial and ethnic minorities, as well as uninsured patients (who may face even more barriers to care than the Medicaid patients in this dataset). Strengths and Limitations One major strength of this study was the use of the Multi-state Medicaid Dataset. Medicaid covers 72.5 million Americans, making it the largest source of healthcare in the United States40. Furthermore, in this study we aimed to compare groups of patients with similar access to care (e.g., patients with at least three years of continuous observation prior to endometriosis diagnosis) to ensure that this was not a confounding factor in analysis. Due to the wide variety of All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted October 3, 2023. ; https://doi.org/10.1101/2023.10.02.23296435doi: medRxiv preprint 14 healthcare settings represented within the Medicaid dataset, we believe our findings will be generalizable across insured populations within the United States; however, international generalizability may be limited. This work is also unlikely to accurately represent the experiences of patients without insurance, who experience different (and generally more severe) barriers to care, especially in the case of chronic conditions41. Another general limitation of using Medicaid data is the amount of missing data regarding racial and ethnic identities. data was only recorded as White, Black/African American, or missing, meaning that other potential racial disparities could not be measured. Similarly, high missingness in ethnicity data rendered us unable to measure possible ethnic disparities. In this work, we use a validated phenotype17 that requires laparoscopic surgery to define our endometriosis cohort. However, recent clinical guidelines recommend that treatment be explored concurrently with further diagnostic exploration if a clinical (vaginal) or imaging test (ultrasound, MRI) indicates endometriosis5. Therefore, some prescriptions that occur in the “pre- diagnostic” subgroup are likely meant to treat suspected endometriosis. Another limitation of this phenotype is the use of a “female” gender marker as inclusion criteria, which may exclude some transgender men with endometriosis from cohort inclusion. The use of ATC level 3 drug classes enables us to quickly and accurately pull relevant drugs based on their pharmacological functions while remaining robust to the fact that providers in different states may vary in their precise prescription patterns42,43. However, ATC medication classes are intended to reflect the drug indication, which may not always be consistent with why the drug was prescribed.

All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted October 3, 2023. ; https://doi.org/10.1101/2023.10.02.23296435doi: medRxiv preprint 15 In this work, we document racial disparities in medication prescription practices for Black endometriosis patients; these disparities exist in several aspects of treatment and are reflective of Black patients not receiving holistic endometriosis care. We compare these disparities to a non-endometriosis population and find several drug classes where disparities are larger in the endometriosis population, indicating that these disparities are unique to chronic disease management (and potentially to endometriosis). When coupled with the fact that these prescription disparities are larger pre-diagnosis compared to post-diagnosis, it becomes important for clinicians who treat chronic, complex patients to ensure that they are communicating with their patients and prescribing medications consistently and equitably. Data Sharing Statement The IBM MarketScan® Multi-state Medicaid dataset is available to license at https://www.ibm.com/watson-health/about/truven-health-analytics (see ref. 12). Code for cohort definitions is available through the OHDSI Phenotype Library and at https://github.com/elhadadlab/endochar. The underlying code for this study is available on Github at https://github.com/elhadadlab/endo_disparities.

1. Sharma S, Tripathi A, Sharma S, Tripathi A. Endometriosis: The Enigma That It Continues to Be. IntechOpen; 2022. doi:10.5772/intechopen.108774 2. Acién P, Velasco I. Endometriosis: A Disease That Remains Enigmatic. ISRN Obstet Gynecol. 2013;2013:242149. doi:10.1155/2013/242149 3. Agarwal SK, Chapron C, Giudice LC, et al. Clinical diagnosis of endometriosis: a call to action. Am J Obstet Gynecol. 2019;220(4):354.e1-354.e12. doi:10.1016/j.ajog.2018.12.039 4. Morassutto C, Monasta L, Ricci G, Barbone F, Ronfani L. Incidence and Estimated Prevalence of Endometriosis and Adenomyosis in Northeast Italy: A Data Linkage Study. PLOS ONE. 2016;11(4):e0154227. doi:10.1371/journal.pone.0154227 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted October 3, 2023. ; https://doi.org/10.1101/2023.10.02.23296435doi: medRxiv preprint 16 5. Becker CM, Bokor A, Heikinheimo O, et al. ESHRE guideline: endometriosis†. Hum Reprod Open. 2022;2022(2):hoac009. doi:10.1093/hropen/hoac009 6. Bougie O, Yap MaI, Sikora L, Flaxman T, Singh S. Influence of race/ethnicity on prevalence and presentation of endometriosis: a systematic review and meta-analysis. BJOG Int J Obstet Gynaecol. 2019;126(9):1104-1115. doi:10.1111/1471-0528.15692 7. Orlando MS, Luna Russo MA, Richards EG, et al. Racial and ethnic disparities in surgical care for endometriosis across the United States. Am J Obstet Gynecol. 2022;226(6):824.e1- 824.e11. doi:10.1016/j.ajog.2022.01.021 8. Hoffman KM, Trawalter S, Axt JR, Oliver MN. Racial bias in pain assessment and treatment recommendations, and false beliefs about biological differences between blacks and whites. Proc Natl Acad Sci U S A. 2016;113(16):4296-4301. doi:10.1073/pnas.1516047113 9. Ghoshal M, Shapiro H, Todd K, Schatman ME. Chronic Noncancer Pain Management and Systemic Racism: Time to Move Toward Equal Care Standards. J Pain Res. 2020;13:2825- 2836. doi:10.2147/JPR.S287314 10. Ezenwa MO, Fleming MF. Racial Disparities in Pain Management in Primary Care. J Health Disparities Res Pract. 2012;5(3):12-26. 11. Schoenthaler A, Williams N. Looking Beneath the Surface: Racial Bias in the Treatment and Management of Pain. JAMA Netw Open. 2022;5(6):e2216281. doi:10.1001/jamanetworkopen.2022.16281 12. Truven Health Analytics. Published June 22, 2022. Accessed February 23, 2023. https://www.ibm.com/watson-health/about/truven-health-analytics 13. Blacketer C, Defalco FJ, Ryan PB, Rijnbeek PR. Increasing trust in real-world evidence through evaluation of observational data quality. J Am Med Inform Assoc. 2021;28(10):2251- 2257. doi:10.1093/jamia/ocab132 14. Reps JM, Rijnbeek PR, Ryan PB. Identifying the DEAD: Development and Validation of a Patient-Level Model to Predict Death Status in Population-Level Claims Data. Drug Saf. 2019;42(11):1377-1386. doi:10.1007/s40264-019-00827-0 15. Nash D, Katcoff H, Faerber J, et al. Impact of Device Miniaturization on Insertable Cardiac Monitor Use in the Pediatric Population: An Analysis of the MarketScan Commercial and Medicaid Databases. J Am Heart Assoc. 2022;11(16):e024112. doi:10.1161/JAHA.121.024112 16. Adamson DM, Chang S, Hansen LG. Health Research Data for the Real World: The MarketScan Databases. 17. Elhadad N, McKillop MM, Schwartz JM, et al. Selection and Characterization of an Endometriosis Cohort Across an Observational Health Database Network. MedArxiv. Published online 2023. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted October 3, 2023. ; https://doi.org/10.1101/2023.10.02.23296435doi: medRxiv preprint 17 18. WHOCC - Structure and principles. Accessed February 23, 2023. https://www.whocc.no/atc/structure_and_principles/ 19. Anatomical Therapeutic Chemical (ATC) Classification. Accessed February 23, 2023. https://www.who.int/tools/atc-ddd-toolkit/atc-classification 20. Tewhaiti-Smith J, Semprini A, Bush D, et al. An Aotearoa New Zealand survey of the impact and diagnostic delay for endometriosis and chronic pelvic pain. Sci Rep. 2022;12(1):4425. doi:10.1038/s41598-022-08464-x 21. Nnoaham KE, Hummelshoj L, Webster P, et al. Impact of endometriosis on quality of life and work productivity: a multicenter study across ten countries. Fertil Steril. 2011;96(2):366-373.e8. doi:10.1016/j.fertnstert.2011.05.090 22. Virtanen P, Gommers R, Oliphant TE, et al. SciPy 1.0: fundamental algorithms for scientific computing in Python. Nat Methods. 2020;17(3):261-272. doi:10.1038/s41592-019-0686-2 23. Mirkin D, Murphy-Barron C, Iwasaki K. Actuarial Analysis of Private Payer Administrative Claims Data for Women With Endometriosis. J Manag Care Pharm. 2007;13(3):262-272. doi:10.18553/jmcp.2007.13.3.262 24. Missmer SA, Tu FF, Agarwal SK, et al. Impact of Endometriosis on Life-Course Potential: A Narrative Review. Int J Gen Med. 2021;14:9-25. doi:10.2147/IJGM.S261139 25. Holdsworth-Carson SJ, Ng CHM, Dior UP. Editorial: Comorbidities in Women With Endometriosis: Risks and Implications. Front Reprod Health. 2022;4. Accessed February 21, 2023. https://www.frontiersin.org/articles/10.3389/frph.2022.875277 26. Gabriel I, Vitonis AF, Missmer SA, et al. Association between endometriosis and lower urinary tract symptoms. Fertil Steril. 2022;117(4):822-830. doi:10.1016/j.fertnstert.2022.01.003 27. Peyneau M, Kavian N, Chouzenoux S, et al. Role of thyroid dysimmunity and thyroid hormones in endometriosis. Proc Natl Acad Sci. 2019;116(24):11894-11899. doi:10.1073/pnas.1820469116 28. Remmert JE, Guzman G, Mavandadi S, Oslin D. Racial Disparities in Prescription of Antidepressants Among U.S. Veterans Referred to Behavioral Health Care. Psychiatr Serv. 2022;73(9):984-990. doi:10.1176/appi.ps.202100237 29. González HM, Croghan TW, West BT, et al. Antidepressant Use among Blacks and Whites in the United States. Psychiatr Serv Wash DC. 2008;59(10):1131-1138. doi:10.1176/appi.ps.59.10.1131 30. Cook B, Creedon T, Wang Y, et al. Examining racial/ethnic differences in patterns of benzodiazepine prescription and misuse. Drug Alcohol Depend. 2018;187:29-34. doi:10.1016/j.drugalcdep.2018.02.011 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted October 3, 2023. ; https://doi.org/10.1101/2023.10.02.23296435doi: medRxiv preprint 18 31. Ghods BK, Roter DL, Ford DE, Larson S, Arbelaez JJ, Cooper LA. Patient–Physician Communication in the Primary Care Visits of African Americans and Whites with Depression. J Gen Intern Med. 2008;23(5):600. doi:10.1007/s11606-008-0539-7 32. Probst JC, Laditka SB, Moore CG, Harun N, Powell MP. Race and ethnicity differences in reporting of depressive symptoms. Adm Policy Ment Health. 2007;34(6):519-529. doi:10.1007/s10488-007-0136-9 33. Kim M. Racial/Ethnic Disparities in Depression and Its Theoretical Perspectives. Psychiatr Q. 2014;85(1):1-8. doi:10.1007/s11126-013-9265-3 34. McConahay JB, Hardee BB, Batts V. Has Racism Declined in America? It Depends on Who Is Asking and What Is Asked. J Confl Resolut. 1981;25(4):563-579. 35. Clark R, Anderson NB, Clark VR, Williams DR. Racism as a stressor for African Americans: A biopsychosocial model. Am Psychol. 19991101;54(10):805. doi:10.1037/0003-066X.54.10.805 36. Anderson NB, Bulatao RA, Cohen B, National Research Council (US) Panel on Race E. Significance of Perceived Racism: Toward Understanding Ethnic Group Disparities in Health, the Later Years. In: Critical Perspectives on Racial and Ethnic Differences in Health in Late Life. National Academies Press (US); 2004. Accessed May 19, 2023. https://www.ncbi.nlm.nih.gov/books/NBK25531/ 37. Mpofu JJ. Perceived Racism and Demographic, Mental Health, and Behavioral Characteristics Among High School Students During the COVID-19 Pandemic — Adolescent Behaviors and Experiences Survey, United States, January–June 2021. MMWR Suppl. 2022;71. doi:10.15585/mmwr.su7103a4 38. Cuevas AG, O’Brien K, Saha S. African American experiences in healthcare: “I always feel like I’m getting skipped over.” Health Psychol Off J Div Health Psychol Am Psychol Assoc. 2016;35(9):987-995. doi:10.1037/hea0000368 39. Cuevas AG, O’Brien K, Saha S. What is the key to culturally competent care: Reducing bias or cultural tailoring? Psychol Health. 2017;32(4):493-507. doi:10.1080/08870446.2017.1284221 40. Medicaid Eligibility | Medicaid. Accessed February 23, 2023. https://www.medicaid.gov/medicaid/eligibility/index.html 41. McWilliams JM. Health Consequences of Uninsurance among Adults in the United States: Recent Evidence and Implications. Milbank Q. 2009;87(2):443-494. doi:10.1111/j.1468- 0009.2009.00564.x 42. Russo V, Orlando V, Monetti VM, et al. Geographical Variation in Medication Prescriptions: A Multiregional Drug-Utilization Study. Front Pharmacol. 2020;11:418. doi:10.3389/fphar.2020.00418 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted October 3, 2023. ; https://doi.org/10.1101/2023.10.02.23296435doi: medRxiv preprint 19 43. Rosenberg A, Fucile C, White RJ, et al. Visualizing nationwide variation in medicare Part D prescribing patterns. BMC Med Inform Decis Mak. 2018;18(1):103. doi:10.1186/s12911- 018-0670-2 Tables Table 1: Descriptive statistics of the endometriosis and non-endometriosis cohorts. This table contains information about the age, race and ethnicity of patients in the endometrsiosi and non-endometriosis cohorts. Age is calculated as the date of cohort entry (date of diagnosis for endometriosis patients, date of most recent visit for non-endometriosis patients). Endometriosis Cohort (n = 16,372) Non-endometriosis Cohort (n = 3,663,904) Age 15-21 9.9% (n = 1,616) 33.7% (n = 1,233,489) 22-28 22.2% (n = 3,634) 20.6% (n = 755,365) 29-35 32.5% (n = 5,323) 18.3% (n = 671,403) 35-42 23.2% (n = 3,804) 13.6% (n = 499,965) 42-49 12.2% (n = 1,995) 13.7% (n = 503,682) Race Black or African American 23.3% (n = 3,814) 17.3% (n = 1,199,859) White 66.0% (n = 10,805) 32.7% (n = 1,829,312) No Matching Concept 10.7% (n = 1,753) 50.0% (n = 634,733) Ethnicity Hispanic or Latino 1.6% (n = 264) 3.8% (n = 138,634) No Matching Concept 98.4% (n = 16,108) 96.2% (n = 3,525,270) All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted October 3, 2023. ; https://doi.org/10.1101/2023.10.02.23296435doi: medRxiv preprint 20 Table 2: Overall prescription prevalence in the endometriosis and the non-endometriosis cohort. This table shows the differences in prescription prevalence (mean prevalence difference) for White and Black patients in the endometriosis and non-endometriosis cohorts. The standard error and adjusted p-value associated with the difference in prevalence is also reported. Significant p- values (p < 0.01) are emphasized in bold. Drug Class Endometriosis Cohort Non-endometriosis Cohort Mean Prevalence Difference (%) Standard Error (%) Adjusted p-value Mean Prevalence Difference (%) Standard Error (%) Adjusted p-value Antidepressants 20.7092674 0.8779937 7.33E-122 22.9035692 0.05304433 0 Anxiolytics 20.3798907 0.9113823 1.31E-109 15.6527752 0.05042332 0 Antiepileptics 18.6863978 0.92753865 4.07E-89 12.9321962 0.04436682 0 Estrogens 10.2351817 0.76278335 6.62E-40 1.8295592 0.02425678 0 Cough suppressants, excl. Combinations with expectorants 10.2306415 0.7721378 6.34E-39 5.98557177 0.03181319 0 Urologicals 8.41948621 0.73720344 4.61E-29 2.94880655 0.02595341 0 Drugs for functional gastrointestinal disorders 8.11512033 0.83853302 5.25E-21 2.98066657 0.02871509 0 Corticosteroids for systemic use, plain 7.61934586 0.67395744 1.73E-28 12.0253099 0.05781293 0 Hypnotics and sedatives 6.73798789 0.88230912 3.12E-13 7.11677719 0.04465059 0 Thyroid preparations 6.72891005 0.44941487 1.55E-49 4.43773974 0.02185524 0 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted October 3, 2023. ; https://doi.org/10.1101/2023.10.02.23296435doi: medRxiv preprint 21 Quinolone antibacterials 6.51072657 0.90775091 1.03E-11 6.41267606 0.04888034 0 Other analgesics and antipyretics 6.35556622 0.7432395 1.71E-16 7.47203815 0.05709184 0 Drugs for peptic ulcer and gastro-oesophageal reflux disease (GORD) 6.22370589 0.91284682 1.29E-10 7.15618371 0.05234589 0 Antiemetics and antinauseants 4.93577208 0.63104266 7.30E-14 7.03386181 0.05758966 0 Belladonna and derivatives, plain 4.5541342 0.74166983 1.15E-08 2.25201094 0.02541702 0 Antihistamines for systemic use 4.28910075 0.69098553 7.55E-09 3.91808881 0.05868259 0 Hormonal contraceptives for systemic use 3.26120892 0.88940093 0.00343917 -1.9807205 0.05820126 1.04E-252 Other systemic drugs for obstructive airway diseases 1.75543863 0.64534074 0.0913468 1.57064744 0.02842173 0 Other diagnostic agents 1.52705867 0.62911922 0.21296566 0.67426432 0.0239525 3.35E-173 Antiinflammatory and antirheumatic products, non-steroids 1.35203434 0.43286802 0.02502562 1.17594879 0.0579422 1.98E-90 Opioids 1.00063649 0.3378741 0.04285102 4.93323827 0.05817166 0 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted October 3, 2023. ; https://doi.org/10.1101/2023.10.02.23296435doi: medRxiv preprint 22 Selective calcium channel blockers with direct cardiac effects -0.1429571 0.36793811 9.76667522 0.14467697 0.01170009 5.63E-34 Antifibrinolytics -0.7107767 0.28571871 0.18001115 0.00790219 0.00613992 2.77322788 Drugs for constipation -2.4573705 0.94071259 0.12593019 -1.0426026 0.04477045 8.26E-119 Progestogens -4.0913806 0.93452218 0.00016765 -7.8743071 0.05176947 0 Antiinfectives and antiseptics, excl. Combinations with corticosteroids -6.9788283 0.87299686 1.83E-14 -6.3718204 0.0524945 0 Hormones and related agents -7.1121568 0.9398425 5.33E-13 -8.9228181 0.04865653 0 Iron preparations -7.2843687 0.71144019 1.86E-23 -4.4471093 0.03375195 0 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted October 3, 2023. ; https://doi.org/10.1101/2023.10.02.23296435doi: medRxiv preprint 23 Table 3: Disparities in Prescription Prevalence Pre-Diagnosis and Post-Diagnosis. This table contains all 28 drug classes and the difference in prevalence between White patients and Black patients. The associated standard error and p-value are also reported. A positive mean prevalence difference indicates that White patients were prescribed a drug within that class more often. Significant p-values (p < 0.01) are emphasized in bold. Drug Class Pre-Diagnosis Post-Diagnosis Mean Prevalence Difference (%) Standard Error (%) Adjusted p-value Mean Prevalence Difference (%) Standard Error (%) Adjusted p-value Antidepressants 24.2136409 0.91910921 8.27E-152 19.1033773 0.93308442 5.21E-92 Anxiolytics 21.6790021 0.92689508 7.77E-120 17.5023318 0.91112685 4.31E-81 Antiepileptics 19.0576737 0.8894292 1.05E-100 14.6465251 0.89299566 2.61E-59 Corticosteroids for systemic use, plain 13.4112531 0.88354112 6.82E-51 7.09665138 0.90348168 5.61E-14 Other analgesics and antipyretics 10.647111 0.89842351 2.98E-31 5.48276992 0.9261711 4.51E-08 Hypnotics and sedatives 10.0331305 0.92767679 4.08E-26 4.99233052 0.94689986 1.89E-06 Antiemetics and antinauseants 9.36654584 0.88938431 8.66E-25 6.54327582 0.87627634 1.15E-12 Cough suppressants, excl. Combinations with expectorants 8.58926626 0.64989361 9.88E-39 4.5777826 0.60477066 5.25E-13 Drugs for functional gastrointestinal disorders 7.06331071 0.72801319 4.13E-21 3.11520408 0.68196701 6.89E-05 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted October 3, 2023. ; https://doi.org/10.1101/2023.10.02.23296435doi: medRxiv preprint 24 Antihistamines for systemic use 6.94421682 0.82975723 8.14E-16 2.29503444 0.93040888 0.19091332 Drugs for peptic ulcer and gastro-oesophageal reflux disease (GORD) 6.79679748 0.94385519 8.36E-12 4.34071438 0.92647094 3.92E-05 Quinolone antibacterials 6.63919703 0.94358424 2.77E-11 3.88632803 0.91046271 0.00027548 Urologicals 5.58218552 0.56071233 3.34E-22 5.18144241 0.62964183 2.64E-15 Thyroid preparations 5.44076752 0.40182738 1.27E-40 5.21899767 0.40262683 2.80E-37 Hormonal contraceptives for systemic use 4.08447171 0.93484337 0.00017462 1.75921784 0.92248195 0.79120025 Opioids 3.83223567 0.64179058 3.30E-08 2.73050637 0.66811974 0.00061217 Belladonna and derivatives, plain 3.71041927 0.64993182 1.59E-07 1.78282791 0.53391252 0.01176382 Estrogens 3.70365484 0.5754246 1.71E-09 8.54471765 0.65909065 2.73E-37 Antiinflammatory and antirheumatic products, non-steroids 2.21986375 0.65859413 0.01050029 0.90768334 0.76028642 3.25540217 Other systemic drugs for obstructive airway diseases 1.3305698 0.54676884 0.20934261 1.38698549 0.52434092 0.11430042 Other diagnostic agents 1.24341988 0.51596245 0.22339418 0.37362264 0.45886946 5.81723098 Selective calcium channel blockers with direct cardiac effects -0.0672012 0.31416557 11.628708 -0.1512507 0.26786714 8.01238248 Antifibrinolytics -0.2432193 0.22567947 3.9362227 -0.5235776 0.19670476 0.10883203 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted October 3, 2023. ; https://doi.org/10.1101/2023.10.02.23296435doi: medRxiv preprint 25 Drugs for constipation -1.1445405 0.87089075 2.6428188 -3.0364077 0.8782654 0.00763926 Progestogens -3.9202865 0.93975007 0.00042341 -4.4411163 0.87031541 4.68E-06 Iron preparations -5.5722235 0.64193868 5.53E-17 -3.3586467 0.4702192 1.28E-11 Hormones and related agents -6.1206816 0.92499823 5.13E-10 -6.4695481 0.8338995 1.21E-13 Antiinfectives and antiseptics, excl. combinations with corticosteroids -7.8604257 0.9433205 1.11E-15 -6.7822939 0.94208523 8.48E-12 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted October 3, 2023. ; https://doi.org/10.1101/2023.10.02.23296435doi: medRxiv preprint 26 Figures Figure 1: Temporal subgroups for patient prescriptions. Three example longitudinal records are shown, with prescriptions marked as “x” and a laparoscopic diagnosis of endometriosis marked with a vertical line. As shown in the figure, prescriptions ordered prior to diagnosis fall into subgroup 1 (pre-diagnosis, orange) and prescriptions ordered after diagnosis fall into subgroup 2 (post-diagnosis, blue). Prescriptions from the day of diagnosis are not counted in either subgroup. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted October 3, 2023. ; https://doi.org/10.1101/2023.10.02.23296435doi: medRxiv preprint 27 Figure 2: Difference in drug class prescription prevalence for White and Black patients. This graph shows the percent difference in drug class prescription prevalence between White and Black patients. Error bars indicate standard error, and a difference greater than 0 indicates that White patients are prescribed a drug from that class with higher prevalence than Black patients. The percent difference is shown across all visits, visits prior to diagnosis (subgroup 1), and visits after diagnosis (subgroup 2). All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted October 3, 2023. ; https://doi.org/10.1101/2023.10.02.23296435doi: medRxiv preprint 28 Supplementary Material Supplemental Table 1: Difference in relative prevalence between endometriosis and comparison cohort. This table contains the difference in prevalence for the 29 ATC level 3 drug classes we analyze in our study. Associated p-values are also shared. Drug Class Endometriosis Cohort: Prevalence (%) Comparison Cohort: Prevalence (%) Difference in Prevalence (%) Adjusted p-value Other analgesics and antipyretics 43.00 36.28 6.71 0 Opioids 42.09 29.84 12.24 0 Antiinflammatory and antirheumatic products, non-steroids 38.32 34.44 3.88 0 Cough suppressants, excl. combinations with expectorants 30.46 29.50 0.96 1.269E-89 Antiemetics and antinauseants 30.18 22.40 7.77 0 Antidepressants 28.33 22.58 5.74 0 Antihistamines for systemic use 26.62 32.47 -5.84 0 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted October 3, 2023. ; https://doi.org/10.1101/2023.10.02.23296435doi: medRxiv preprint 29 Hormonal contraceptives for systemic use 26.61 22.59 4.02 1.863E-205 Drugs for peptic ulcer and gastro-oesophageal reflux disease (GORD) 22.87 22.06 0.81 0.0 Hypnotics and sedatives 22.79 18.39 4.39 0.0 Anxiolytics 22.48 13.28 9.20 0.0 Quinolone antibacterials 21.50 17.22 4.27 0.0 Corticosteroids for systemic use, plain 19.23 15.14 4.08 0.0 Antiepileptics 18.64 16.51 2.12 0.0 Progestogens 14.79 11.01 3.77 0.0 Drugs for constipation 13.45 8.98 4.47 4.848E-134 Hormones and related agents 13.42 13.98 -0.56 8.297E-272 Drugs for functional gastrointestinal disorders 10.79 9.89 0.90 0.0 Estrogens 9.99 5.05 4.94 7.640E-177 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted October 3, 2023. ; https://doi.org/10.1101/2023.10.02.23296435doi: medRxiv preprint 30 Antiinfectives and antiseptics, excl. combinations with corticosteroids 8.46 4.49 3.97 0.0 Thyroid preparations 7.77 5.31 2.45 7.029E-66 Urologicals 7.28 5.50 1.77 2.907E-103 Other systemic drugs for obstructive airway diseases 6.33 5.43 0.89 0.000859 Belladonna and derivatives, plain 6.15 5.64 0.50 4.858E-269 Iron preparations 5.18 5.45 -0.27 3.175E-05 Other diagnostic agents 4.25 4.21 0.041 2.454E-64 Antifibrinolytics 3.64 2.013 1.63 7.487E-50 Selective calcium channel blockers with direct cardiac effects 3.38 2.72 0.65 4.449E-45 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprintthis version posted October 3, 2023. ; https://doi.org/10.1101/2023.10.02.23296435doi: medRxiv preprint