{"paper_id":"34e9881c-704d-46b7-a353-2f6502f1c2c0","body_text":"1\n1\n2\n3\n4\n5\n6\n7\n8\n9\n10\n11 Interprofessional Care Team, Staffing, and Setting Characteristics that Impact Patient Outcomes: \n12 A Review of Reviews\n13 Authors: Alix Pletcher, BA1*, Kyla Woodward, PhD, RN2, Natalie Hoge, MPH2, Nathaniel Blair-Stahn, \n14 PhD1, Paulina Lindstedt, MPH1, Zahra Gohari, MSGE2, Abraham Flaxman, PhD1, and Sarah Iribarren, \n15 PhD, RN2\n16\n17 1Institute for Health Metrics and Evaluation, University of Washington School of Medicine\n18 2University of Washington School of Nursing\n19 *Corresponding author: pletale@uw.edu\n20\n21\n22\n23\n24\n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted January 7, 2024. ; https://doi.org/10.1101/2024.01.04.24300868doi: medRxiv preprint \nNOTE: This preprint reports new research that has not been certified by peer review and should not be used to guide clinical practice.\n\n2\n25 Abstract\n26 Background. The purpose of this study was to identify research methods and evidence pertaining to the \n27 relationship of interprofessional acute care teams and hospital characteristics on patient outcomes in \n28 hospital-based acute care. \n29 Methods. A review was completed using the Preferred Reporting Items for Systematic Reviews and \n30 Meta-Analysis extension for Scoping Reviews guidelines. The search strategy was executed across \n31 PubMed, CINAHL, and Embase. The review included 12 systematic reviews from 2012 to 2023 that \n32 examine the impact of acute care staffing characteristics on patient outcomes. \n33 Results. Workforce characteristics primarily focused on nurse staffing, with a limited number of studies \n34 assessing the impact of interprofessional teams or non-clinical workers on care quality. There is limited \n35 data describing the context of care delivery via potential relationships between hospital characteristics, \n36 interprofessional team staffing levels, and patient outcomes. \n37 Conclusions. To promote comparability across studies, future workforce research should include a \n38 comprehensive analytic approach that includes clearly defined variables representing interprofessional \n39 care teams, community factors, and staffing and patient characteristics.\n40\n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted January 7, 2024. ; https://doi.org/10.1101/2024.01.04.24300868doi: medRxiv preprint \n\n3\n41 Keywords\n42 Healthcare; Workforce; Interprofessional Care Team \n43 Introduction and Background\n44 Shortages of healthcare workers have been a long-standing and pervasive issue in international \n45 acute care settings, but since the onset of the COVID-19 pandemic, maintaining adequate staffing levels \n46 has become an even greater challenge for acute care facilities. Internationally, healthcare workers left jobs \n47 or reduced hours for multiple reasons, including illness, fear of infecting themselves or loved ones with \n48 COVID-19, needing to care for children and other family members, and the effects of extended heavy \n49 workloads.(1-3) Those who stayed at work also experienced persistent negative impacts, with one \n50 international review identifying consequences such as losing hope or professional identity,(4) which have \n51 led to persistent workforce shortages.) \n52 Ongoing staffing shortages place a burden on all healthcare workers and affect their perceptions \n53 of safety in the workplace,(5) and also impact patient, particularly those with higher support needs related \n54 to individual. A number of studies have demonstrated that nurse staffing and adverse patient outcomes are \n55 inversely related, so that higher staffing ratios result in fewer adverse outcomes.(6, 7) However, despite \n56 the fact that nursing care does not occur in a void, there is little if any data exploring links between \n57 nursing workload, nurse burnout, and staffing of other interprofessional care team members. Nurse \n58 staffing data does not provide information about the staffing of other clinical and nonclinical care team \n59 members such as social workers, physical and respiratory therapists, nutrition services, and environmental \n60 services, collectively called the ‘interprofessional care team’, who help provide critical support and \n61 services for patients with higher needs. While workers in these roles may not provide direct care, their \n62 jobs include tasks often absorbed by nursing staff when there is insufficient staffing. This absorption of \n63 additional duties contributes to burnout and dilution of RN scope of practice, increasing the likelihood of \n64 missed care and other adverse events, and leading to missed opportunities to advance equity.(8, 9)\n65 Likewise, few studies account for specific features of communities and patients that impact acute \n66 care experiences and outcomes, particularly regarding health equity. Community level factors such as \n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted January 7, 2024. ; https://doi.org/10.1101/2024.01.04.24300868doi: medRxiv preprint \n\n4\n67 urbanicity, county socioeconomic status, and housing type may shape the patient population served by the \n68 organization and the resources available to patients after discharge. Patient outcomes also differ due to \n69 characteristics such as comorbidities, mental health problems, and social determinants of health (SDOH), \n70 a set of social and environmental factors underlying health inequities.(10) Increases in inpatient \n71 assessment of SDOH have improved the care teams’ ability to identify patients with higher social and \n72 support needs; those needs require interprofessional services and support alongside nursing care to \n73 improve equity and optimize patient outcomes.(11) With ongoing shortages of healthcare workers across \n74 multiple professions and documented burnout among RNs, it is critical to understand how care team \n75 composition in acute care settings affects outcomes for both patients and workers. \n76 In 2021, the WA legislature directed the Washington State Department of Health to contract an \n77 interdisciplinary team of University of Washington researchers led by the School of Nursing to conduct a \n78 workforce study examining the impacts of healthcare staffing characteristics on patient outcomes.(12)  As \n79 our first step in developing an appropriate analytic strategy for the study, this paper reports a review of \n80 reviews of existing international data on factors that may influence both care team staffing and patient \n81 outcomes. The purpose of this study was to identify what variables have been used to examine \n82 relationships between interprofessional team staffing and patient outcomes, including variables pertaining \n83 to community, hospital, care team, and patient characteristics that may influence either staffing or \n84 outcomes, and to develop a preliminary causal model for further use in our project.\n85 Methods\n86 This study focused on identifying variables used to assess the impact of inpatient care teams on \n87 patient outcomes. We mapped the types of variables, outcomes, and analyses used to evaluate \n88 relationships between interprofessional acute care teams and patient outcomes in existing literature. We \n89 chose to review systematic reviews to allow representation of a large pool of literature and get a high-\n90 level overview of what has been studied in this domain to ascertain the scope of the research and identify \n91 gaps in knowledge.(13) The stages of the review included establishing research question(s), identifying \n92 relevant studies, selecting studies, charting the data, and collating, summarizing, and reporting the \n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted January 7, 2024. ; https://doi.org/10.1101/2024.01.04.24300868doi: medRxiv preprint \n\n5\n93 results.(14) This study was designed and reported according to the Preferred Reporting Items for \n94 Systematic Reviews and Meta-Analysis (PRISMA) Guidelines.(15) \n95 Step 1. Establish the research questions\n96 Using our team’s expertise and directives from the legislature, we narrowed our research \n97 questions to the following:\n98 1. How were team members from different professions and roles (clinical and nonclinical) \n99 represented in studies assessing the impact of care teams on patient outcomes?\n100 2. What community, hospital/organizational, care team, and patient factors were included in \n101 description or analysis of staffing as it impacted patient outcomes, and how were these variables \n102 defined?\n103 3. What analytic strategies were used to examine the impact of teams or staffing on patient \n104 outcomes?\n105 Step 2. Identify relevant studies\n106 The search strategy was developed and executed in consultation with an experienced research \n107 librarian (CM). The original search occurred on March 3, 2022 and was most recently updated on March \n108 30, 2023. The search strategy was created for PubMed (which includes Medline, PubMed Central, and \n109 other resources) using a combination of Medical Subject Heading (MeSH) terms, keywords, and phrases \n110 (see the complete search strategy in Supporting Information). The search terms targeted health workers, \n111 health research, hospitals/hospital settings/hospitalization, and patient outcomes. The PubMed search \n112 terms were translated for Cumulated Index to Nursing and Allied Health Literature (CINAHL) and \n113 Embase using their respective thesaurus terms and advanced search features. A manual search using the \n114 reference lists of retrieved citations was conducted for other relevant studies. \n115 Table 1 summarizes the participants, interventions, comparison, outcomes, and study design \n116 (PICOS) used to define the inclusion and exclusion criteria strategy. Eligible studies included systematic \n117 reviews that examined interventions with a workforce component and their impact on adult acute care \n118 patient outcomes and were published within the last 10 years. We excluded all other study or publication \n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted January 7, 2024. ; https://doi.org/10.1101/2024.01.04.24300868doi: medRxiv preprint \n\n6\n119 types, such as cross-sectional studies, randomized control trials, commentaries, editorials, letters to the \n120 editor, abstract proceedings and reviews that did not include workforce personnel or patient outcomes. \n121 We restricted our search to only include reports that were published in English. \n122 Table 1. Study inclusion criteria\n123\n124\n125 3. Select Relevant Studies\n126 We compiled the search results in EndNote and removed duplicates, then transferred citations to \n127 Covidence software for screening. The screening process was composed of two phases: first, we reviewed \n128 the title and abstract for relevance using the inclusion criteria; and second, we reviewed the full text of \n129 remaining articles. In each of these phases, two reviewers (AP, ZG) independently screened the articles \n130 and a third reviewer (SI) assisted in resolving any discrepancies. \n131 4. Charting the Data\n132 Project directives informed the selection of variables to include analysis of the impact of the \n133 number, type, education, training, and experience of acute care hospital staffing personnel on patient \n134 mortality and patient outcomes. Control variables included factors such as access to equipment, patients’ \n135 underlying conditions and diagnoses, patients’ demographic information, the trauma level designation of \n136 the hospital, transfers from other hospitals, and external factors impacting hospital volumes.(12) Based on \n137 an initial review of papers to identify key components, the reviewers used a standard data extraction sheet \n138 to record the following categories: study design, data sources, inclusion criteria within that review, \n139 interprofessional team members, types of patients (e.g. acute versus intensive care), patient outcomes \n140 (e.g., length of stay, hospital acquired infection), additional factors considered in the review or analysis \nPICOS element Description\nPopulation Adults in acute care hospitals\nIntervention or \nComparison\nStaffing models or measures, including those focused on nurses or other \ninterprofessional care team members OR\nStaff training or education level\nOutcome Any patient outcome (e.g., mortality, pressure ulcers)\nStudy type Systematic review, with or without meta-analysis\n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted January 7, 2024. ; https://doi.org/10.1101/2024.01.04.24300868doi: medRxiv preprint \n\n7\n141 (e.g., equipment, technology, external factors), and analytic strategy or model. In addition, any reported \n142 challenges, limitations, strengths, or recommendations to each were extracted.\n143 5. Synthesis of Results \n144 We used a descriptive approach to report the data on the variables, approaches, and healthcare \n145 team composition. Given the expected heterogeneity of systematic reviews found in our search and our \n146 goal to simply identify the presence and use of variables rather than the direction of any relationships \n147 between them, a quantitative summary measure of results was not undertaken. We summarized frequency \n148 of use, definitions, and concordance of analytical strategies across the studies. We synthesized the \n149 findings into main concepts, grouped by variable categories (i.e., hospital characteristics, patient \n150 characteristics, staffing, or patient outcome). We also summarized reported challenges, limitations, \n151 strengths, and recommendations for each variable. Results informed the development of a baseline causal \n152 model that was used to guide subsequent stakeholder interviews and to develop and refine the analytic \n153 strategy for the state workforce study.   \n154 Results\n155 Search results yielded a total of 164 articles, of which 37 met the inclusion criteria for full text review \n156 (see Figure 1). We then identified a total of 12 systematic reviews for data extraction, which represented \n157 more than 575 individual studies (not counting duplicate instances of 13 studies among the reviews). \n158 Only 4 reviews included data from the most recent 5 years,(16-19) while the remainder included studies \n159 dating back to 1986. One review focused only on the United States,(8) and the other eleven included \n160 international data. Reviews predominantly included reports of primary research, with only one using \n161 secondary sources such as reviews or editorials.(19)  We organized results into interprofessional team and \n162 staffing characteristics, elements of communities, hospitals and patients that may interact with staffing \n163 characteristics, and patient outcomes (see Table 2). We highlighted elements that consider personnel or \n164 factors outside of nursing to understand the relationship between healthcare teams and patient outcomes. \n165\n166 Figure 1. PRISMA flow diagram.\n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted January 7, 2024. ; https://doi.org/10.1101/2024.01.04.24300868doi: medRxiv preprint \n\n8\n167 Table 2. Overview of included studies.\n168\nReview \n(# studies)\nStudy aim Staffing variables Patient outcome \nvariables\nOther variables\nBae et al. \n(2014)(26)\nn=24\nTo systematically \nevaluate the effect \nof nurse overtime \nand long work \nhours on nurse and \npatient outcomes\nRN work hours\nRN overtime (OT)\nAdverse events\nFailure-to-rescue \nMortality\nPatient satisfaction\nN/A\nBagnasco et al. \n(2019) (16)\nn=44\nTo review and \nsynthesize research \nstudies on surgical \nand medical \ninpatients' \nperceptions on \nunmet nursing care \nneeds\nNursing workload\nNHPPD\nNursing skill mix\nRN OT\nMissed patient care\nMortality\nTeam: work \nenvironment, RN \neducation, RN \nexperience \nHospital: size, \ntechnology level, \nteaching status\nPatient: age, sex, \nadmission type, \ncomorbidities\nBurke et al. (2021) \n(23)\nn=52\nTo investigate what \nclinically relevant \ninterventions have \nbeen shown to \nimprove \norganizational fail-\nto-rescue rates\nNPR Failure-to-rescue \nMortality\nTeam: work \nenvironment; RN \neducation\nCassarino et al. \n(2019)(20)\nn=6 \nTo synthesize the \ntotality of evidence \nrelating to the \nimpact of early \nassessment and \nintervention by \nhealthcare teams on \nquality, safety, and \neffectiveness of care \nin the Emergency \nDepartment\nPresence of \ninterdisciplinary \nteam members\nHospital admission\nLength of stay \nMortality\nPatient satisfaction\nReadmissions\nN/A \nDall’Ora et al. \n(2022)(17)\nn=27\nTo evaluate \nevidence for an \nassociation between \nnurse staffing levels \nand patient \noutcomes in acute \ncare settings\nNPR\nNHPPD\nAdverse events\nLength of stay\nMortality\nReadmissions\nTeam: RN skill mix\nPatient: age, sex, \nadmission type, \ncomorbidities, \nseverity of illness\nEvangelou et al. \n(2018)(22)\nTo identify quality \nindicators \nassociated with \nNPR; NHPPD\nNursing manpower \nuse score\nAdverse events \nCost\nLength of stay \nTeam: work \nenvironment, RN \neducation, RN \n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted January 7, 2024. ; https://doi.org/10.1101/2024.01.04.24300868doi: medRxiv preprint \n\n9\nn=13 nursing care for \nadult ICUs in \nliterature\nPatient care \nassistant-to-patient \nratio\nMortality\nPatient satisfaction \nReadmissions\nexperience, RN skill \nmix, RN group \ncompetence \nGriffiths et al. \n(2018)(8) \nn=18\nTo identify nursing \ncare most \nfrequently missed in \nacute adult inpatient \ncare wards and to \ndetermine evidence \nfor the association \nof missed care with \nnurse staffing \nNPR Missed patient care N/A\nKerlin et al. \n(2016)(21)\nn=18 \nTo review the \nassociation of \nnighttime intensivist \nstaffing with \noutcomes of ICU \npatients\nNighttime \nintensivist staffing\nNighttime:daytime \nstaffing differences\nStaff type\nComplications\nDuration of \nmechanical \nventilation\nLength of stay \nMortality\nReadmissions \nHospital: size\nPatient: severity of \nillness \nMcGahan et al. \n(2012)(24) \nn=19\nTo examine the \nrelationship \nbetween nurse \nstaffing levels and \nthe incidence of \nmortality and \nmorbidity in adult \nintensive care unit \npatients\nNPR; NHPPD\nTherapeutic \nIntervention \nScoring System-to-\nRN ratio\nBed-to-nurse ratio\nAdverse events\nMortality\nN/A\nPlotnikoff et al. \n(2021)(19)\nn=314\nTo investigate what \nelements facilitate a \nsuccessful, high-\nquality discharge \nfrom the ICU\nNPR\nProvider \nexperience; \nprovider training; \nprovider workload; \npresence of an \ninterdisciplinary \nteam\nAdverse events\nCosts\nLength of stay\nMortality\nPatient satisfaction\nReadmissions\nTeam: provider-to-\nprovider \ncommunication \nPatient: \ndemographics, \nprovider-to-patient \ncommunication, \nfamily engagement, \ndischarge \neducation; \nadmission type\nHospital: trauma \ndesignation, \ntechnology\nRae et al. \n(2021)(18)\nn=55\nTo determine \nassociations \nbetween variations \nin registered nurse \nstaffing levels in \nadult critical care \nNPR; NHPPD\nBed-to-nurse ratio\nNumber of RNs\nNursing Activities \nScore\nAdverse events \nMortality\nLength of stay\nDuration of \nmechanical \nventilation\nN/A\n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted January 7, 2024. ; https://doi.org/10.1101/2024.01.04.24300868doi: medRxiv preprint \n\n10\nunits and outcomes \nsuch as patient, \nnurse, \norganizational and \nfamily outcomes\nRNs' perception of \nstaffing adequacy\nLength of weaning\nPatient satisfaction\nRecio-Saucedo et \nal. (2017)(25)\nn=14\nTo systematically \nreview the impact \nof missed nursing \ncare on patient \noutcomes, in acute \nhospital wards and \nnursing homes \nNPR; NHPPD Adverse events\nLength of stay\nMortality\nPatient satisfaction\nPatient safety\nReadmissions\nTeam: work \nenvironment, RN \neducation, RN \nexperience, RN age\nHospital: size, \nteaching status, \nMagnet© \ndesignation, type, \ntechnology level, \nownership, \nMedicare cost-to-\ncharge ratio\nCommunity: \npopulation density, \nvolume of patients \nwith heart failure, \nhospital location, \nlanguage region, \nprofit status\nPatients: \ndemographics, \nadmission type, \ninsurance status, \nnumber of \nprocedures, patient \nhealth status, \ncomorbidities\n169 Note. RN = Registered Nurse; ICU = Intensive Care Unit; NPR = Nurse-to-Patient Ratio; NHPPD = \n170 Nursing Hours Per Patient Day\n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted January 7, 2024. ; https://doi.org/10.1101/2024.01.04.24300868doi: medRxiv preprint \n\n11\nInterprofessional Teams and Staffing\nThe reviews provide limited evidence on the impact of interprofessional healthcare teams on \npatient outcomes. Most studies either focused exclusively on nursing staff (10/12) or considered direct \ncare providers such as physicians or respiratory therapists (2/12) without mentioning the impact of \nindirect healthcare workers such as nutrition or environmental services. One study of interprofessional \nteams in the emergency department defined the healthcare team as an assortment of care professionals \nincluding occupational and physical therapists and medical social workers.(20) One review examined the \nrelationship between nighttime intensivist staffing and patient outcomes in intensive care.(21) Two \nreviews included a limited number of studies examining relationships between non-registered nurse team \nmembers and quality indicators or discharge.(19, 22)\nStudies that quantified the impact of nursing staff on patient outcomes used various measures to \nquantify the patient care workload. The most common measure of absolute staffing levels was nurse-to-\npatient ratio (8/12),(8, 17-19, 22-25) followed by Nursing Hours Per Patient Day (6/12).(16-18, 22, 24, \n25) Other less commonly used measures for nursing workload included nurse overtime (2/12),(16, 26) \nNursing Activities Score (1/12),(18) number of hours worked by nurses (1/12),(26) and Therapeutic \nIntervention Scoring System-to-nurse ratio (1/12).(24) \nTable 3. Nursing workforce measures and definitions. \nMeasure Definition\nNurse overtime Number of overtime hours worked by nurse. Least specific to \npatient acuity\nNurse-to-patient ratio \n(NPR)\nRatio indicating the number of patients that a nurse \nmust provide care for at a given time.\nNursing Hours Per \nPatient Day (NHPPD)\nCalculation used to quantify the number of nursing \nhours needed to provide care for a patient or group of \npatients.\nTherapeutic \nIntervention Scoring \nSystem-to-nurse ratio \n(TISS) (30)\nSet of therapeutic nursing actions performed daily in \nintensive care; used to compare nursing work between \ngroups of patients. \n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted January 7, 2024. ; https://doi.org/10.1101/2024.01.04.24300868doi: medRxiv preprint \n\n12\nNursing activities score \n(NAS)(31)\nScoring system that builds on the TISS by adding \ncritical care-specific activities and weighting those \nactivities according to actual time required for \ncompletion. \nMost specific to \npatient acuity\nNurse characteristics\nAs part of assessing staffing, reviews addressed characteristics of nursing staff and workplace \nfactors. Assessments of nursing work environment (as measured by the Nursing Work Scale) were \nincluded in 4/12 reviews.(16, 22, 23, 25) Another common characteristic was education (4/12), or the \npercentage of RNs with a baccalaureate or higher degree in nursing.(16, 22, 23, 25) Three of those \nreviews also looked at years of nursing experience,(16, 22, 25)  2/12 included specialty certification, (22, \n25) and 1/12 included tenure at the organization.(25) Job-related training or knowledge was included in \n3/12 reviews,(17, 19, 25) 3/12 looked at nursing skill mix, or the percentage of nursing staff who were \nregistered nurses,(16, 17, 22) and 1/12 included nurse reported quality measures.(25) Most studies used \nthese elements as independent variables, while 1/12 used them as controls for their main findings.(19)\nCommunity, Hospital, and Patient Characteristics\nIn general, community characteristics such as rurality or population demographics were not used \nas independent variables or controls when relationships between healthcare teams and patient outcomes \nwere assessed; only 1/12 reviews included community characteristics as external factors that may affect \nhospital volume.(25) Within hospitals, characteristics such as equipment and technology, size, and \nteaching status were examined in 4/12 reviews.(16, 17, 19, 25) Two reviews controlled for equipment and \ntechnology, with definitions ranging from counts of resources such as critical care beds to facilities for \nmajor organ transplant or open-heart surgery.(16, 25) One review included hospital-level structures for \ndischarge.(19) In all, community and hospital variables were not consistently represented or defined \nacross the included reviews. \nPatient-level characteristics were used as descriptive and control variables. 4/12 reviews \nconsidered measures related to patient health status at baseline or severity of illness during \n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted January 7, 2024. ; https://doi.org/10.1101/2024.01.04.24300868doi: medRxiv preprint \n\n13\nhospitalization.(17, 19, 21, 25) Only 2/12 reviews addressed other patient variables such as demographics \nand healthcare logistics such as insurance or finances(19, 25) and 1/12 looked at admission type (e.g., \nelective versus emergent admissions).(17) No included studies referenced or defined SDOH or discussed \nhealth equity.\nPatient Outcomes\nPatient outcomes were treated as a dependent variable in most of the included reviews (9/12), and were \ndefined as mortality, adverse events, and other care outcomes. We used the National Quality Forum \nclassification for adverse events, which defines 29 possible serious, reportable events that are considered \nlargely preventable and therefore indicative of inadequate healthcare safety mechanisms.(27) Patient \nmortality was included in some form in 11/12 reviews. Of those, 5/12 looked at inpatient mortality,(17, \n18, 21, 22, 25) 5/12 looked at post-discharge mortality at 1-month(17, 18, 20, 23, 26) or 1/12 at 1 \nyear,(20) 1/12 did not specify a timeframe,(24) and 1/12 noted the use of mortality in many of the \nincluded studies but did not treat it as a predictor or outcome.(19) Non-fatal adverse events related to \ninpatient care were used in 7/12 reviews as a staffing-related outcome. Reported adverse events included \nhealthcare associated infection (6/12),(17, 18, 22, 24-26) pressure injury (5/12),(17, 22, 24-26) patient \nfalls (3/12),(22, 25, 26) medication errors (2/12),(22, 25) and gastrointestinal bleeds (1/12).(26) Other \npatient outcomes were length of stay (7/12),(17-22, 25) hospital readmission (6/12), (17, 19-22, 25) \nsatisfaction (5/12), (18, 20, 22, 25, 26) and ventilator days (3/12).(17, 18, 21) Two studies focused on \nmissed care(8, 16) and another included health-related quality of life as a patient outcome.(20) \nStrategies for Analysis\nQuantitative strategies for analysis were only addressed in 5/12 reviews. Of the reviews that \nextracted information on quantitative analysis, multivariate logistic regression was the most used model \ntype in the evaluation of staffing impacts on patient outcomes (3/12),(17, 23, 26) followed by multivariate \nlinear regression (2/12)(19, 24) and negative binomial regression (2/12).(17, 26) One of 12 reviews \n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted January 7, 2024. ; https://doi.org/10.1101/2024.01.04.24300868doi: medRxiv preprint \n\n14\nreported use of other statistical analyses, including the Cox proportional hazard model and hierarchical \nmixed effects survival.(17)\nCausal Model\nStudy findings were used to develop an initial causal model that was used throughout the WA \nworkforce study the Washington Acute Care CHaracteristics and patient Outcomes (WACCHO) model \n(Figure 2). The model presents the different categories of variables present in the current literature. We \nused this model to guide our stakeholder interviews and focus groups and determine if any factors were \nmissing or inadequately represented in the model. \nFigure 2. Initial causal model based on review findings.\nDiscussion\nKey findings in this review reinforce the reliance on nursing in existing research assessing the \nimpact of staffing on patient outcomes. Our results also indicate a lack of attention to the contexts in \nwhich acute care occurs within organizations and the community, and which may impact patient needs \nand equitable care delivery. These findings suggest several critical areas for future research on staffing \nand patient outcomes, including interprofessional care team composition and characteristics of \ncommunities, hospitals, and patients that may influence staffing and/or patient outcomes. In particular, a \nlack of attention to patients’ SDOH limits our ability to understand implications of staffing on health \nequity. Findings also point to a need for clearly defined, consistently used variables and appropriate \nanalytic strategies in healthcare workforce research. \nCare team composition\nA key finding from this review was that interprofessional acute care team composition is \nundertheorized and underrepresented in current scientific literature. Because studies that included \ninterprofessional team members occurred in specific areas outside of acute care or focused on a specific \nevent such as discharge, we found no evidence describing the effects of a comprehensive, \ninterprofessional care team on patient outcomes in acute care. Furthermore, studies predominantly focus \n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted January 7, 2024. ; https://doi.org/10.1101/2024.01.04.24300868doi: medRxiv preprint \n\n15\non the effects of nurse staffing and infrequently consider the influence of non-nursing staff or indirect \ncare providers such as environmental services on patient outcomes. These findings reinforce trends in \ncurrent workforce analysis to focus exclusively on nursing personnel without acknowledgement of \nassociated factors that impact the RN workload.(7) While a focus on nursing is warranted due to their \nessential role in acute care, research that fails to account for other factors and team members impacting \nRN work will miss important elements that influence the relationship between staffing and patient \noutcomes. For example, in the absence of other care team members, RNs often  take on non-nursing roles, \ndiluting time that could be spent on activities that optimize their full scope of practice and increasing \nfrustration and burnout.(9) Understanding the impacts of other types of staffing is therefore essential to \nunderstanding how outcomes are affected in settings where RNs are able to exclusively focus on RN work \nand settings where they routinely take on other tasks. A national workplace-focused survey in the United \nStates found that among 5,461 acute care RN respondents, 27% rated availability of appropriate ancillary \nstaff as ‘seldom’ or ‘never’, 39% indicated availability as ‘sometimes’, and only 29% rated availability as \n‘often’ or ‘always’.(28) In the context of staffing shortages and nursing burnout, organizations need to \nunderstand the optimal care team composition for both patient and worker outcomes. \nCommunity, hospital, and patient characteristics\nNearly half of the reviews included in the study did not address community, hospital or \norganizational, or patient characteristics. When hospital characteristics were included in reviews, there \nwas little consensus on how to best define and control for characteristics that may influence the provision \nof high-quality care. For example, hospitals equipped to perform open heart surgeries were used as proxy \nfor more advanced equipment and technological capacity than those who do not perform open heart \nsurgery, but it is unclear whether this denotes a substantive difference in organizational capacity or \nresources compared to a hospital that does not perform open heart surgeries. Data suggests that \ncommunity level factors such as urbanicity, county socioeconomic status, and housing type may shape the \npatient population served by the organization, which impacts both care delivery and outcomes,(10) so \nthese factors need to be considered when examining the impacts of staffing and workforce issues.\n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted January 7, 2024. ; https://doi.org/10.1101/2024.01.04.24300868doi: medRxiv preprint \n\n16\nNone of the included reviews addressed SDOH as patient characteristics, nor did they use the \nterm ‘health equity’ in describing patient outcomes. This may have occurred in part because of the \ntemporality of the reviews and the papers they sourced, but is nonetheless a gap in our understanding of \nrelationships between staffing and patient outcomes. In general, patient characteristics such as \ndemographics were not consistently included in staffing models, and more nuanced patient-level metrics \nsuch as insurance status or admission type were even less commonly included. Measures intended to \nreflect patient acuity, such as an adjustment for comorbidities or severity of illness, were similarly sparse. \nIn the existing literature, some studies reported diagnostic codes or acuity ratings to quantify workload \nrelated to patient care, but even these measures failed to identify patients requiring extra time and \npersonnel resources to achieve similar outcomes.(29) This implies an over-reliance on patient \ndemographics such as age, sex, and race as proxies for SDOH and therefore limits the interpretation of \nresults and their impact on health equity. This pattern may be related to a lack of available data in primary \nresearch studies endeavoring to quantify these relationships. In Washington State, House Bill 1272 \nrequires hospitals to begin reporting additional patient demographic information in 2023. This data should \nimprove future researchers’ ability to assess health equity as a critical outcome. \nVariable definition and analytic strategy\nDefinitions of staffing metrics used by primary studies were rarely provided in the systematic \nreviews, suggesting that these variables were not consistently defined in workforce research. The lack of \nclear definitions or consistent measures makes further analysis and application of findings difficult. For \nexample, the most common staffing metric used in the study was nurse-to-patient ratio. This ratio does \nnot overtly reflect patient acuity as well as measurements such as Nursing Hours Per Patient Day, and \ntherefore hinders the comparison of staffing ratios and patient outcomes among different populations and \nin different acute care settings.\nAlong with inconsistent variable definitions, we found a lack of consensus regarding the analytic \nstrategy or model type used in studies exploring the impacts of healthcare staffing on patient outcomes. \nMost included reviews did not report the methods of quantitative analysis used by primary workforce \n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted January 7, 2024. ; https://doi.org/10.1101/2024.01.04.24300868doi: medRxiv preprint \n\n17\nstudies. Of those reporting analytic strategies, multivariate logistic regression was the most common \nmethod of analysis. Future studies in this field could explore which model types and methods best reflect \nthe relationship between healthcare staffing and patient outcomes, particularly when examining the \ninfluence of community, hospital, or patient characteristics. \nLimitations\nA review has inherent limitations, since findings generally rely on the accuracy and completeness \nof the review, and quality may vary across studies. For example, reviews may have poorly specified \ninclusion and exclusion criteria or an inadequate search process. This study relied on the information \npresented in the papers we reviewed, which may have missing information from primary studies or may \nbe interpreted differently in our findings. Additionally, despite careful development of a systematic search \nstrategy, we may have missed some relevant systematic reviews. Despite the various limitations, a key \nstrength of this report is the sheer volume of primary studies assessed. More than 575 studies were \nrepresented in the included reviews. Examination of reference lists showed that 13 primary studies were \nrepresented in multiple reviews, with 9/13 in 2 reviews, 3/13 in 3 reviews, and 1/13 in 4 reviews. \nHowever, assessment of each review showed that findings were not exclusively drawn from duplicated \nstudies.  Altogether, this review offers a comprehensive overview of current studies about \ninterprofessional team composition and staffing and their impacts on patient outcomes in the hospital \nsetting. \nConclusions\nThis review revealed that contextual factors such as healthcare team composition or hospital \nsetting were largely unexamined in current health services literature. Further research is needed to better \nunderstand how these factors impact hospital function, work environment, care quality, and staff and \npatient outcomes. Given the pervasive and ongoing shortages of workers in healthcare, we need to build \non our understanding of nurse staffing by examining how the availability of interprofessional care team \nmembers impact clinician workload and patient outcomes, and how hospitals in different settings and \nthose serving diverse patient populations may require different care team composition. This work can \n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted January 7, 2024. ; https://doi.org/10.1101/2024.01.04.24300868doi: medRxiv preprint \n\n18\ninform strategies to optimize team composition with the goal of improving patient outcomes and \nfurthering health equity. In addition, to promote comparability across studies, future workforce research \nshould include a comprehensive analytic approach that includes clearly defined variables representing \ninterprofessional care teams, community factors, and staffing and patient characteristics. More \ncomprehensive and applicable research can better inform practice and policy, improving outcomes for \npatients, workers, and communities.\nAbbreviations\nSDOH: Social determinants of health\nRN: Registered nurse \nPRISMA-ScR: Preferred Reporting Items for Systematic Reviews and Meta-Analysis\nPICOS: Population, Intervention, Comparison, Outcome, and Study type\nMeSH: Medical Subject Heading\nCINAHL: Cumulated Index to Nursing and Allied Health Literature\nNPR: Nurse-to-patient ratio\nNHPPD: Nurse Hours Per Patient Day\nNAS: Nursing Activities Score\nTISS: Therapeutic Intervention Scoring System\nDeclarations\nEthics approval and consent to participate\nNot applicable\nConsent for publication\nNot applicable\nAvailability of data and materials\nThe datasets used and/or analyzed during the current study are available from the corresponding author on \nreasonable request.\nCompeting interests\n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted January 7, 2024. ; https://doi.org/10.1101/2024.01.04.24300868doi: medRxiv preprint \n\n19\nThe authors declare that they have no competing interests.\nFunding\nThe study was funded by the Washington State Department of Health through a contract (HED26380) \nwith the University of Washington’s School of Nursing. The content is solely the responsibility of the \nauthors and does not necessarily represent the official views of the Washington State Department of \nHealth or the University of Washington. This work was also supported, in part, by the National Institutes \nof Health, National Institute of Nursing Research Training Program in Global Health Nursing at the \nUniversity of Washington (T32 NR019761). The content is solely the responsibility of the authors and \ndoes not necessarily represent the official views of the National Institutes of Health.\nAuthors’ contributions\nAP is the corresponding author for this study. AP and ZG designed and performed the literature \nextraction. CM executed the literature search. AP and KW drafted the manuscript and designed the \nfigures. NBS, NH, PL, AF, and SI were involved in planning and supervising the work. All authors \nreviewed and contributed to editing the manuscript.\nAcknowledgments\nNot applicable. \nReferences\n1. American Nurses Foundation. Pulse on the nation's nurses COVID-19 series. American Nurses \nFoundation.; 2020. https://www.nursingworld.org/practice-policy/work-environment/health-\nsafety/disaster-preparedness/coronavirus/what-you-need-to-know/mental-health-and-wellbeing-\nsurvey/. Accessed 20 Mar 2022.\n2. Brown KM, Robinson GE, Nadelson CC, Grigoriadis S, Mittal LP, Conteh N, et al. Psychological \nimpact of COVID-19 on minority women. J Nerv Mental Dis. 2021;209(10):695-6.\n3. Zipf AL, Polifroni EC, Beck CT. The experience of the nurse during the COVID ‐19 pandemic: A \nglobal meta‐synthesis in the year of the nurse. J Nurs Scholarship. 2022;54(1):92-103.\n4. Boone LD, Rodgers MM, Baur A, Vitek E, Epstein C. An integrative review of factors and \ninterventions affecting the well-being and safety of nurses during a global pandemic. Worldv Evid-\nBased Nu. 2023;20:107–115.\n5. Riman KA, Harrison JM, Sloan DM, McHugh MD. Work Environment and Operational Failures \nAssociated With Nurse Outcomes, Patient Safety, and Patient Satisfaction. Nursing Research. \n2023;72(1):p 20-29.\n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted January 7, 2024. ; https://doi.org/10.1101/2024.01.04.24300868doi: medRxiv preprint \n\n20\n6. Dierkes A, Do D, Morin H, Rochman M, Sloane DM, McHugh MD. The impact of California's \nstaffing mandate and the economic recession on registered nurse staffing levels: A longitudinal \nanalysis. Nurs Outlook. 2022;70(2):219-27.\n7. Pittman P. Evidence on hospital staffing and outcomes: implications for Washington. Washington, \nDC: George Washington University; 2022.\n8. Griffiths P, Recio‐Saucedo A, Dall'Ora C, Briggs J, Maruotti A, Meredith P, et al. The association \nbetween nurse staffing and omissions in nursing care: a systematic review. J Adv Nurs. \n2018;74(7):1474-87.\n9. Gottlieb LN, Gottlieb B, Bitzas V. Creating empowering conditions for nurses with workplace \nautonomy and agency: how healthcare leaders could be guided by strengths-based nursing and \nhealthcare leadership (SBNH-L). J Healthcare Leadership. 2021;13:169.\n10. Al-Amin M, Islam MN, Li K, Shiels N, Buresh J. Is there an association between hospital staffing \nlevels and inpatient-COVID-19 mortality rates? PLoS One. 2022;17(10):e0275500.\n11. Wammes JJG, van der Wees PJ, Tanke MAC, Westert GP, Jeurissen PPT. Systematic review of \nhigh-cost patients’ characteristics and healthcare utilisation. BMJ Open. 2018;8(9):e023113.\n12.  Washington state legislature. E2SHB 1272 bill report: Concerning health system transparency. \nhttps://lawfilesext.leg.wa.gov/biennium/2021-22/Pdf/Bill%20Reports/House/1272-\nS2.E%20HBR%20FBR%2021.pdf?q=20231219104654. Accessed 19 Dec 2023.\n13. Armstrong R, Hall BJ, Doyle J, Waters E. ‘Scoping the scope’of a cochrane review. J Pub Hlth. \n2011;33(1):147-50.\n14. Levac D, Colquhoun H, O'Brien KK. Scoping studies: advancing the methodology. Impl Sci. \n2010;5:1-9.\n15. Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al. The PRISMA \n2020 statement: an updated guideline for reporting systematic reviews. Sys Rev. 2021;10(1):89.\n16. Bagnasco A, Dasso N, Rossi S, Galanti C, Varone G, Catania G, et al. Unmet nursing care needs on \nmedical and surgical wards: A scoping review of patients’ perspectives. J Clin Nurs. 2020;29(3-\n4):347-69.\n17. Dall'Ora C, Saville C, Rubbo B, Turner L, Jones J, Griffiths P. Nurse staffing levels and patient \noutcomes: A systematic review of longitudinal studies. Int J Nurs Stud. 2022:104311.\n18. Rae PJ, Pearce S, Greaves PJ, Dall'Ora C, Griffiths P, Endacott R. Outcomes sensitive to critical \ncare nurse staffing levels: A systematic review. Intensive Crit Care Nurs. 2021;67:103110.\n19. Plotnikoff KM, Krewulak KD, Hernández L, Spence K, Foster N, Longmore S, et al. Patient \ndischarge from intensive care: an updated scoping review to identify tools and practices to inform \nhigh-quality care. Crit Care. 2021;25(1):1-13.\n20. Cassarino M, Robinson K, Quinn R, Naddy B, O’Regan A, Ryan D, et al. Impact of early \nassessment and intervention by teams involving health and social care professionals in the \nemergency department: A systematic review. PLoS One. 2019;14(7):e0220709.\n21. Kerlin MP, Adhikari NK, Rose L, Wilcox ME, Bellamy CJ, Costa DK, et al. An official American \nThoracic Society systematic review: the effect of nighttime intensivist staffing on mortality and \nlength of stay among intensive care unit patients. Am J Resp Crit Care Med. 2017;195(3):383-93.\n22. Evangelou E, Lambrinou E, Kouta C, Middleton N. Identifying validated nursing quality indicators \nfor the intensive care unit: an integrative review. Connect: The World of Crit Care Nurs. \n2018;12(2):28-39.\n23. Burke JR, Downey C, Almoudaris AM. Failure to rescue deteriorating patients: a systematic review \nof root causes and improvement strategies. J Patient Safety. 2022;18(1):e140-e55.\n24. McGahan M, Kucharski G, Coyer F. Nurse staffing levels and the incidence of mortality and \nmorbidity in the adult intensive care unit: a literature review. Australian Crit Care. 2012;25(2):64-77.\n25. Recio‐Saucedo A, Dall'Ora C, Maruotti A, Ball J, Briggs J, Meredith P, et al. What impact does \nnursing care left undone have on patient outcomes? Review of the literature. J Clin Nurs. \n2018;27(11-12):2248-59.\n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted January 7, 2024. ; https://doi.org/10.1101/2024.01.04.24300868doi: medRxiv preprint \n\n21\n26. Bae S-H, Fabry D. Assessing the relationships between nurse work hours/overtime and nurse and \npatient outcomes: systematic literature review. Nurs Outlook. 2014;62(2):138-56.\n27. National Quality Forum. Serious reportable events in healthcare-2011 update: a concensus report. \nWashington, DC: National Quality Forum; 2011.\n28. American Nurses Foundation. COVID-19 survey series: 2022 workplace survey. American Nurses \nFoundation; 2022. https://www.nursingworld.org/practice-policy/work-environment/health-\nsafety/disaster-preparedness/coronavirus/what-you-need-to-know/covid-19-survey-series-anf-2022-\nworkplace-survey/. Accessed 1 Nov 2022.\n29. Juvé-Udina M-E, Adamuz J, López-Jimenez M-M, Tapia-Pérez M, Fabrellas N, Matud-Calvo C, et \nal. Predicting patient acuity according to their main problem. J Nurs Manage. 2019;27(8):1845-58.\n30. Miranda DR, de Rijk A, Schaufeli W. Simplified Therapeutic Intervention Scoring System: the \nTISS-28 items--results from a multicenter study. Crit Care Med. 1996;24(1):64-73.\n31. Miranda DR, Nap R, de Rijk A, Schaufeli W, Iapichino G. Nursing Activities Score. Crit Care Med. \n2003;31(2):374-82.\n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted January 7, 2024. ; https://doi.org/10.1101/2024.01.04.24300868doi: medRxiv preprint \n\n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted January 7, 2024. ; https://doi.org/10.1101/2024.01.04.24300868doi: medRxiv preprint \n\n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted January 7, 2024. ; https://doi.org/10.1101/2024.01.04.24300868doi: medRxiv preprint","source_license":"CC-BY-4.0","license_restricted":false}