Many factors hinder adoption of implementation strategies and / or target evidence-based interventions. A gap in implementation science is the paucity of methodology for precisely assessing which medical practices will benefit most from a particular implementation strategy, how much support a medical practice might need, and if some medical practices could adopt evidence-based interventions on their own. Using machine learning, we developed and validated predictive models of the adoption of electronic health record (EHR)-related implementation strategies. This retrospective study extracted data from an EHR-based cervical cancer screening (CVC) clinical decision tool for 310 community health centers from 5/1/2017 to 6/30/2019. Machine learning models (i.e., logistic lasso regression) were used to predict adoption (any tool use in the 12 months after tool rollout vs. none) and sustained use (≥ 1 tool use in the last quarter of the 12-month study period vs. < 1 tool use in this time period). Model performance was compared using the area under the receiver operating characteristics curve (AUC) from an external, withheld clinic sample. Prediction models showed good performance for adoption (AUC = 0.891, 95

This case-control study assesses the association of losing insurance with diabetes outcomes among people with low income served by community health centers.

Background Hypertension is a major risk factor for cardiovascular disease and stroke, with a growing burden in Nigeria. Limited access to blood pressure (BP) screenings, low awareness, and barriers to healthcare drive low diagnosis rates. Community-based screenings for other conditions have been successful and could improve diagnosis and care linkage. This study evaluates the effectiveness of a community-based screening and referral intervention to increase hypertension detection and facilitate connections to healthcare. Methods This cluster-randomized effectiveness-implementation hybrid type III trial in Nigeria tests the implementation and effectiveness of a community-based BP screening intervention. There are two arms: 1). Core: community screenings supported by mobile health technology with referral to a healthcare facility for participants with at least one elevated BP reading, and 2.) Core+ (enhanced): additional support from community health advisors to help participants link to healthcare facilities. Participants with elevated BP will be informed about medication vouchers if diagnosed with hypertension and prescribed medication. Community gathering sites hosting screenings leverage local engagement structures. The study will assess the proportion of individuals identified to have elevated BP, rates of linkage to healthcare facilities, and factors influencing intervention uptake and sustainability. Discussion To address gaps in hypertension detection and management utilizing community-based strategies, findings will provide insights into the feasibility and effectiveness of implementing a scalable, community-driven approach to hypertension screening and linkages to healthcare in Nigeria. If successful, this model could be adapted to other resource-limited settings in the United States and globally to improve detection and reduce hypertension-related complications. Name of registry: clinicaltrials.gov Trial Registration NCT06659900. Registered 26 October 2024, prospectively registered.https://clinicaltrials.gov/study/NCT06659900?id=NCT06659900&rank=1

Objective: This study evaluates whether gaining Medicaid following the Affordable Care Act (ACA) expansion led to changes in the rate of acute diabetes complications diagnosed in primary care settings, relative to in inpatient, emergency department (ED), or urgent care (UC) settings. Methods: This retrospective cohort study used Medicaid administrative claims data linked to electronic health records for 3767 patients, aged 19 to 64 years, who experienced acute preventable complications of diabetes between 2014 and 2019 diagnosed in inpatient, ED, UC, or primary care settings in the state of Oregon. These patients were classified as either continuously Medicaid-insured or having gained Medicaid. Results: Annual rates of acute complications diagnosed in primary care and inpatient/ED/UC settings increased for both continuously [Adjusted Rate Ratio (aRR) = 2.20, 95% CI = 1.65-2.91] and newly Medicaid-insured patients (aRR = 2.67, 95% CI = 2.05-3.47) after the ACA. Among newly Medicaid-insured, annual rates of abnormal blood glucose diagnosed in primary care settings significantly increased with time while those diagnosed in inpatient/ED/UC decreased (2014 vs 2016 aRR = 3.36, 95% CI = 1.60-7.09). Conclusion: We found a significantly greater rate of abnormal blood glucose diagnosed in primary care clinics among patients who gained Medicaid post-ACA and a corresponding decline in diagnosis in inpatient/ED/UC settings.

BACKGROUND:To evaluate insurance instability (churn) among adults with diabetes receiving care at community-based health centers (CHCs). METHODS:Retrospective cohort study using patients' electronic health records data for 300,158 adults aged 19 to 64 with ≥3 ambulatory visits between 2014 and 2019 of which 39,542 churned out of insurance. Generalized estimating equation-based (GEE) logistic regression models were fitted to assess the odds of churning. RESULTS:Among CHC patients, those with diabetes had 1.25 greater odds of churning than those without diabetes (aOR = 1.25; 95%CI = 1.18, 1.33). Among CHC patients with diabetes, the odds of churning were higher for those with uncontrolled diabetes, more complex medication regimens, and acute diabetes complication. CONCLUSIONS:CHC patients with diabetes are more likely to experience insurance instability than those without diabetes. Outreach efforts to reduce the impact of the postpandemic Medicaid disenrollment among patients with diabetes and lower income will be critical to reduce harmful health consequences.

Background: Neighborhood socioeconomic marginalization and racial residential segregation are associated with differential health outcomes in adulthood and pregnancy, but the intergenerational effects of these exposures on early childhood growth are underexplored. Our objective was to investigate racial and ethnic differences in the association between neighborhood deprivation and early childhood growth trajectories, with modification by neighborhood racial concentration. Methods: Using longitudinal clinical data among 58,860 children receiving care in community-based clinics in the ADVANCE Clinical Data Research Network, we identified four early childhood (0-24 months) body mass index (BMI) trajectories using group-based trajectory modeling: Low, Catch-Up, Moderate, and High. In race- and ethnicity-stratified multinomial logistic regression analyses, trajectory group membership was modeled as a function of neighborhood deprivation, neighborhood racial concentration, neighborhood deprivation*racial concentration interactions, and confounders. Results: Greater neighborhood deprivation was marginally associated with greater odds of Catch-Up trajectory for most racial and ethnic groups, with a null association observed among Assimilated Hispanic children. Conversely, neighborhood deprivation was not associated with Low trajectory for non-Hispanic Black or White children; however, in Less Assimilated Hispanic children, higher neighborhood deprivation was marginally associated with higher odds of Low trajectory, most strongly in neighborhoods with higher vs. lower Hispanic concentration. Associations between neighborhood deprivation and High trajectories varied substantially by race and ethnicity, ranging from inverse among Less Assimilated Hispanic children to a positive association among non-Hispanic White children that was attenuated in neighborhoods with higher White concentration. Conclusion: Greater neighborhood deprivation was generally associated with greater or similar odds of each alternative growth trajectory, most consistently for non-Hispanic White and Black children. Associations were largely similar across levels of neighborhood racial concentration. Further research is needed to understand contextual or behavioral factors that contribute to the observed racial and ethnic differences in the association between neighborhood deprivation and early childhood growth.

This Viewpoint discusses the findings of a recent National Academies of Sciences, Engineering, and Medicine report suggesting that current health care delivery and accountability structures perpetuate, rather than reduce, health inequities and details several changes needed to address these structural problems.

Machine learning (ML) approaches could expand the usefulness and application of implementation science methods in clinical medicine and public health settings. The aim of this viewpoint is to introduce a roadmap for applying ML techniques to address implementation science questions, such as predicting what will work best, for whom, under what circumstances, and with what predicted level of support, and what and when adaptation or deimplementation are needed. We describe how ML approaches could be used and discuss challenges that implementation scientists and methodologists will need to consider when using ML throughout the stages of implementation.

Background. Policy leaders and researchers have identified a range of primary care spending conceptualizations, developed frameworks and methods for measuring primary care spending, and documented the pros and cons of different approaches. However, these efforts have not been comprehensive, particularly as the number of estimates has grown. We continue this work by identifying the definitions, data sources, and approaches used to estimate primary care spending in the United States. Our objective was to identify where there is and is not consensus across methods, and how initial steps toward a standardized approach to estimating primary care spending might be achieved. We approached this comparison from a societal economic perspective. Methods. Searches were conducted in Ovid MEDLINE® and Cochrane CENTRAL databases (inception to May 2, 2023), and were supplemented by manual reviews of reference lists, Scopus searches of key articles, gray literature searches of State and organization websites, and responses to a Federal Register Notice, as well as recommendations from Key Informants. Websites of States and organizations that produced reports were reviewed in November 2023 to identify updates. Publicly available estimates and reports of methods were supplemented by discussions with experts who have supported States’ estimates. Findings. We identified 67 primary care spending estimates for 2010 to 2021: 42 of these were produced by 11 State Governments for their State, 2 were published by the Veterans Health Administration, and 23 were published by researchers or other organizations, which include foundations and policy organizations. Forty-four estimates reported on primary care spending for a single State, one estimate reported spending for the New England States, and 22 reported national spending. To date, 13 State Governments have developed and/or are implementing measurements of primary care spending. When State Governments measure primary care spending, they produce regular, often yearly, estimates. States have produced one to eight estimates, demonstrating some States have more experience with this task than others. Primary care spending estimates in our sample ranged from 3.1 to 10.3 percent. These estimates started with definitions of primary care, which are often labeled narrow or broad. Estimates may use these same labels to mean different things. Narrow definitions of primary care usually include fewer providers, locations, or service types, while broad definitions include more. State, regional, or national estimates are either reported as two estimates, one using a narrow and one using a broad definition of primary care, or as a single estimate labeled neither narrow nor broad. Variations in what providers, services, and locations are included in definitions of primary care are significant and likely contribute to variation in primary care spending estimates. However, it is difficult to distinguish differences in definitions and measurement from differences in actual primary care spending. Conclusions. While there are some core similarities in how primary care spending is measured across State, regional, and national estimates, there are more differences. While there may be rationale behind some of these variations, this variation limits comparisons and what could be understood about the impact of policies. Furthermore, lack of clear, detailed reporting of methods can obscure precisely how and why estimates differ. Research is needed that quantifies the impact different decisions and measurement methods have on spending estimates. To assure the validity and reliability of estimates of primary care spending, and facilitate comparisons and links to health outcomes, Federal, State, and policy leaders need to: (1) collaborate to create a primary care clinician database that can function as a public utility for States to allow for more precise identification of primary care clinics and clinicians, and reduce reliance on Current Procedural Terminology/Healthcare Common Procedure Coding System codes; (2) develop a template for transparent reporting of methods used to estimate primary care spending; (3) foster collaboration among Federal agencies and State leaders to develop a consensus definition of primary care and process for estimating primary care spending, with consideration of methods that are easy to understand and transparent; and (4) support the development and ongoing maintenance of State All-Payer Claims Databases, expand to include nonclaims payments, and supply Medicare and Medicaid estimates for every State.

To investigate the association between maternal cervical cancer (CC) screening status and child human papillomavirus (HPV) vaccination uptake. To understand if child sex or social deprivation index (SDI) modify this association. We used a national cohort of children linked to at least one parent using electronic health record (EHR) data from a network of community health centers across the United States. We used SDI scores and child sex as moderating variables. We performed the analysis (1) for the whole sample (with SDI and child sex added as covariates), (2) stratified by SDI quartile (with child sex added as a covariate), and (3) stratified by SDI quartile and child sex, to examine whether associations vary by SDI quartile and by child sex. N = 52,919 linked mother–child pairs. Mother’s receipt of CC screening was positively associated with the linked child’s odds of receiving HPV vaccination [adjusted odds ratio (AOR) 1.39, 95

This case series identifies states’ estimates of primary care spending and recommends steps policymakers can take toward standardizing these estimates.

The Implementation Science Centers in Cancer Control (ISC3) initiative, funded by the National Cancer Institute, called for the development of implementation laboratories to bolster implementation science, create research-ready environments, and expedite adoption and implementation of evidence-based interventions (EBIs) into practice. The Building Research in Implementation and Dissemination to close Gaps and achieve Equity in Cancer Control (BRIDGE-C2) Center is one of seven ISC3 centers. BRIDGE-C2 aims to identify strategies to improve implementation of cancer prevention EBIs and conduct research / develop pragmatic methods to tailor, enhance, and support the adoption and sustainability of these strategies; advance implementation science; and build capacity and training opportunities. Since its inception, the BRIDGE-C2 Center has been conducting research and training activities to advance knowledge on how to effectively implement strategies to improve cancer prevention EBIs in primary care clinics serving socioeconomically disadvantaged patients. The translational science benefits model (TSBM) provides a useful framework for organizing a description of the BRIDGE-C2 Center's activities. In this paper, we describe examples of BRIDGE-C2 activities and the specific impact indicators within each relevant domain/subdomain of the TSBM, demonstrating that a single activity or project has multiple impacts on methods and capacity building, clinical domains, and community health.

Aims: Children of parents with substance use and/or other mental health (SU/MH) diagnoses are at increased risk for health problems. It is unknown whether these children benefit from receiving primary care at the same clinic as their parents. Thus, among children of parents with >1 SU/MH diagnosis, we examined the association of parent-child clinic concordance with rates of well-child checks (WCCs) and childhood vaccinations. Design: Retrospective cohort study using electronic health record (EHR) data from the OCHIN network of community health organizations (CHOs), 2010-2018. Setting: 280 CHOs across 17 states. Participants/Cases: 41,413 parents with >1 SU/MH diagnosis, linked to 65,417 children aged 0 to 17 years, each with >1 visit to an OCHIN clinic during the study period. Measurements: Dependent variables: rates of WCCs during (1) the first 15 months of life, and (2) ages 3 to 17 years; vaccine completeness (3) by the age of 2, and (4) before the age of 18. Estimates were attained using generalized estimating equations Poisson or logistic regression. Findings: Among children utilizing the same clinic as their parent versus children using a different clinic (reference group), we observed greater WCC rates in the first 15 months of life [adjusted rate ratio (aRR) = 1.06; 95% confidence interval (CI) = 1.02-1.10]; no difference in WCC rates in ages 3 to 17; higher odds for vaccine completion before age 2 [adjusted odds ratio (aOR) = 1.12; 95% CI = 1.03-1.21]; and lower odds for vaccine completion before age 18 (aOR = 0.88; 95% CI = 0.81-0.95). Conclusion: Among children whose parents have at least one SU/MH diagnosis, parent-child clinic concordance was associated with greater rates of WCCs and higher odds of completed vaccinations for children in the youngest age groups, but not the older children. This suggests the need for greater emphasis on family-oriented healthcare for young children of parents with SU/MH diagnoses; this may be less important for older children.

Storylines of Family Medicineis a 12-part series of thematically linked essays with accompanying illustrations that explore the many dimensions of family medicine, as interpreted by individual family physicians and medical educators in the USA and elsewhere around the world. In ‘V: ways of thinking—honing the therapeutic self’, authors present the following sections: ‘Reflective practice in action’, ‘The doctor as drug—Balint groups’, ‘Cultivating compassion’, ‘Towards a humanistic approach to doctoring’, ‘Intimacy in family medicine’, ‘The many faces of suffering’, ‘Transcending suffering’ and ‘The power of listening to stories.’ May readers feel a deeper sense of their own therapeutic agency by reflecting on these essays.

Patient-centered outcomes research (PCOR) is designed to generate high-quality evidence important to patients and families, clinicians, and policymakers about treatments, services, and other health care interventions. PCOR studies have traditionally focused on understanding causal relationships among the use of health care treatments and services, treatment effects, and clinical outcomes. That focus has broadened to include a more holistic understanding of health and well-being, including the significant economic impacts of health care use on individuals, families, and their communities, such as out-of-pocket spending and informal caregiving needs. In many cases, data to study the economic impacts of health care from the perspective of individuals, families, and communities are unmeasured, not routinely collected, or unavailable for research. The growing recognition that economic factors often impact health outcomes, decision-making, and equity in health care is the impetus for the articles in this special issue of Medical Care. The Affordable Care Act established the Patient-Centered Outcomes Research Trust Fund (PCORTF) in 2010 and structured it for use in strengthening the evidence base for decision-making.1 The trust fund provides resources to build data capacity, conduct research, disseminate research findings, train investigators on PCOR methods, and engage people with lived experience in carrying out these activities. Funding is provided by the PCORTF to the Secretary of Health and Human Services (HHS), the Agency for Healthcare Research and Quality, and the Patient-Centered Outcomes Research Institute, with each entity having a distinct yet related set of responsibilities for evidence generation, dissemination, and implementation. The Assistant Secretary for Planning and Evaluation (ASPE) has the unique responsibility of coordinating across relevant federal health programs to build data capacity for PCOR. ASPE works in partnership with all HHS agencies on a portfolio of projects that support the collection, linkage, and analysis of data for PCOR studies. This body of work is collectively referred to as the Office of the Secretary Patient-Centered Outcomes Research Trust Fund portfolio (https://aspe.hhs.gov/collaborations-committees-advisory-groups/os-pcortf/explore-portfolio) and has led to the production of a variety of data products, including standards, algorithms, and linked datasets that would not have been possible without the PCORTF. These products support evidence generation in health and health care by HHS agencies and for departmental priorities such as reducing maternal mortality and substance use, increasing emergency preparedness, and improving the equitable delivery of health care. The 2019 reauthorization of the PCORTF underscored the importance of assessing a full range of outcomes, explicitly expanding the scope of patient outcomes that should be considered in PCOR studies to include the potential burdens and economic impacts of the utilization of medical treatments, items, and services on different stakeholders and decision-makers respectively. These potential burdens and economic impacts include medical out-of-pocket costs, including health plan benefit and formulary design, nonmedical costs to the patient and family, including caregiving, effects on future costs of care, workplace productivity and absenteeism, and healthcare utilization.2 Following the reauthorization of the PCORTF, and in response to this new priority, HHS developed a new strategic plan for the Office of the Secretary Patient-Centered Outcomes Research Trust Fund3 based on input from a National Academies of Sciences, Engineering, and Medicine study committee4 and representatives across HHS.5 The fourth objective of the OS-PCORTF Strategic Plan focuses on addressing data capacity limitations to support a more comprehensive view of health outcomes, including improving the availability, quality, and relevance of data on economic outcomes.3 As part of its work to implement and achieve this objective, ASPE sponsored a symposium and this special issue of Medical Care to bring together multiple perspectives on building data capacity for economic outcomes in PCOR.6 The symposium was designed not only to review and discuss current efforts and challenges related to data capacity and infrastructure, as set forth in a set of invited manuscripts, but also to identify areas of importance not addressed in this work, interrogate underlying assumptions, and develop a foundation for initiating and sustaining efforts to advance data capacity for economic outcomes in PCOR studies. The work featured in this special issue draws on the themes that emerged from the symposium. These articles highlight specific and important challenges in building PCOR data capacity on economic outcomes. However, as noted by symposium attendees, particularly patient stakeholders, these articles (and the discussions around them) often lack an explicitly patient-centered focus. That is, what are the information needs of patients related to economic impacts, and how can we ensure that efforts to build data capacity on economic outcomes align with those needs? First, the measurement and collection of data on economic outcomes should reflect the types of questions and decisions patients, caregivers, clinicians, and policymakers face. Unfortunately, the evidence to inform these decisions and improve patient outcomes—and the data and data infrastructure needed to generate it—are generally lacking in most data sources. For example, as articles published in this issue of Medical Care have described, data are needed to understand household economic impacts7 and family decisions and trade-offs.8 The data needs for specific populations should also be considered. Palatucci and colleagues9 provide an overview of appropriate data collection on economic outcomes for individuals with intellectual disabilities, while efforts to build data capacity for economic outcomes among cancer patients and embed data collection within oncology practices are discussed by Halpern et al10 and Williams et al,11 respectively. Beyond improving the measurement and capture of data on economic outcomes, efforts are needed to build more comprehensive data resources through data linkage and improved data sharing and access. Brown et al12 report on a review of federally funded administrative and survey data sources linked or linkable to Medicare fee-for-service claims that can be used to increase the range of outcomes included in PCOR studies. Jones and colleagues13 describe a novel effort to link Medicare and Medicaid data in North Carolina and discuss how to support the development and use of patient-centered utilization measures, the development of integrated Medicare and Medicaid programs, and the evaluation of health equity impacts. Zhang and Meltzer14 discuss their work in developing an integrated dataset of Medicare beneficiaries to better understand cost-related medication nonadherence. Moving beyond specific linked datasets to integrated data infrastructure, Bradley et al15 and Waitman et al16 provide overviews of efforts to link a state-level All-Payer Claims Database in Colorado with a cancer registry and strengthen PCORnet, the National Patient-Centered Clinical Research Network, respectively, to support the inclusion of economic outcomes in PCOR studies. The final article in this collection responds to this imperative, synthesizing the robust discussions among symposium attendees to arrive at a set of cross-cutting considerations to guide efforts to build data capacity and identify initial opportunities to expand the availability and use of relevant, high-quality economic outcomes data in PCOR.17 Although the articles in this issue highlight the potential benefits of improved data capacity for economic outcomes in PCOR, they also make it clear that much work needs to be done—particularly in supporting the paradigm shift within health economics research to include the perspectives of patients and families. The significance of this shift mirrors the initial sea change in efforts to engage patients and other stakeholders in all aspects of clinical comparative effectiveness research following the establishment of the PCORTF in 2010. With the expanded scope of outcomes in the 2019 reauthorization, engaging patients and families in identifying questions, data, and research on economic impacts will be crucial to expanding the evidence about the outcomes and effectiveness of health care and providing equitable health care. More broadly, continued collaboration within the PCOR community around these and other opportunities to advance the collection, linkage, and analysis of economic outcomes data for PCOR will be needed to realize the gains from the nation’s investment in the PCORTF and support decision-makers in their efforts not only to improve health and well-being but also to limit the economic burdens of health care.

Context: In 2018, AHRQ developed staffing models with panel sizes, functions, ratios and financing approaches for 3 types of comprehensive primary care clinics. We used this model in an academic health system serving people of differing ages, medical complexity and social risk. Objective: Determine the usability and update the model for post-pandemic academic primary care. Study Design and Analysis: Mixed methods cross-sectional observational study; comparative analysis. Setting: 9 clinics: 2 safety-net, 1 internal medicine, 4 family medicine, 2 pediatric. Population studied: Clinic faculty and staff. Intervention/Instrument: Panel size, full time equivalents (FTEs) by function, encounter volume; interviews with a sample of each clinics’ members, with representation across functions. Outcome Measures: Identification of staff functions, panel sizes, staffing ratios, encounter numbers. Results: AHRQ’s model was usable in academic primary care, but needed to be modified to align with the blended populations served by clinics. A supplementary tool was needed to identify FTE gaps by function and support planning among clinic and system administration. Using this tool, we found that clinician panel sizes were similar to AHRQ model recommendations, but clinics were short staffed by an average of 9.5 FTE/clinic (range 1-22 FTE). Functional gaps were identified in complex care/care transitions, care coordination, and behavioral health (BH), the latter of which was an increased need since the pandemic. Non-visit-based telephone and portal encounters grew by 73,000 (32%) from 2019 to 2021 and are now approximately double the number of visit-based encounters. These communications take multiple touches and team members to complete, not all of which were counted. The explosion of non-visit-based work, according to staff, contributed to a spiral of work, burnout, and attrition. Conclusion: AHRQ’s staffing model is useful in primary care, with the addition of a tool to operationalize this model for leaders and decision makers. The model, however, requires expansion for pediatrics, where not all functions are equally needed, to account for non-visit-based work, and patients expanded BH needs. This expansion requires careful consideration of financing, as clinics are experiencing a double-hit (short-staffed and seeing an explosion of work) and examination of the impact of the expanded staffing model on meaningful outcomes (e.g., patient experience of comprehensive care).

Introduction Latino adolescents may face numerous barriers) to recommended vaccinations. There is little research on the association between Latino adolescent-mother preferred language concordance and vaccination completion and if it varies by neighborhood. To better understand the social/family factors associated with Latino adolescent vaccination, we studied the association of adolescent-mother language concordance and neighborhood social deprivation with adolescent vaccination completion. Methods We employed a multistate, electronic health record (EHR) based dataset of community health center patients to compare three Latino groups: (1) English-preferring adolescents with English-preferring mothers, (2) Spanish-preferring adolescents with Spanish-preferring mothers, and (3) English-preferring adolescents with Spanish-preferring mothers with non-Hispanic white adolescent-mother pairs for human papilloma virus (HPV), meningococcal, and influenza vaccinations. We adjusted for mother and adolescent demographics and care utilization and stratified by the social deprivation of the family’s neighborhood. Results Our sample included 56,542 adolescent-mother dyads. Compared with non-Hispanic white dyads, all three groups of Latino dyads had higher odds of adolescent HPV and meningococcal vaccines and higher rates of flu vaccines. Latino dyads with Spanish-preferring mothers had higher vaccination odds/rates than Latino dyads with English-preferring mothers. The effects of variation by neighborhood social deprivation in influenza vaccination rates were minor in comparison to differences by ethnicity/language concordance. Conclusion In a multistate analysis of vaccinations among Latino and non-Latino adolescents, English-preferring adolescents with Spanish-preferring mothers had the highest completion rates and English-preferring non-Hispanic white dyads the lowest. Further research can seek to understand why this language dyad may have an advantage in adolescent vaccination completion.

As recent extreme weather events demonstrate, climate change presents unprecedented and increasing health risks, disproportionately so for disadvantaged communities in the U.S. already experiencing health disparities. As patients in these frontline communities live through extreme weather events, socioeconomic and health stressors are compounded; thus, their healthcare teams will need tools to provide precision ecologic medicine approaches to their care. Many primary care teams are taking actionable steps to bring community-level socioeconomic data ("community vital signs") into electronic medical records, to facilitate tailoring care based on a given patient's circumstances. This work can be extended to include environmental risk data, thus equipping healthcare teams with an awareness of clinical and community vital signs and making them better positioned to mitigate climate impacts on health. For example, if healthcare teams can easily identify patients who have multiple chronic conditions and live in an urban heat island, they can proactively arrange to "prescribe" an air conditioner, heat pump, and/or air purifier. Or, when a severe storm/heat event/poor air quality event is predicted, they can take preemptive steps to get help to patients at high medical and socioeconomic risk, rather than waiting for them to arrive in the emergency department. Advances in health information technologies now make it technically feasible to integrate a wealth of publicly-available community-level data into EMRs. Efforts to bring this contextual data into clinical settings must be accelerated to equip healthcare teams to provide precision ecologic medicine interventions to their patients.

Electronic health record (EHR) tools such as documentation shortcuts and ordering templates are designed to enhance both the efficiency and quality of care delivery. Use of such tools vary widely in primary care. Understanding patterns of use as a means to identify priority areas for improvement are lacking. Objective: This study is designed to identify and understand patterns of EHR efficiency use among primary care clinicians as a means to inform training support needs. Study Design: Retrospective observational study. Setting or Dataset: 340 Community Health Centers (CHCs) across 48 US states from the OCHIN Network. Population Studied: The OCHIN’s network and 2033 primary care clinicians engaged in patient care at least 2 days a week during DEC 2019-FEB 2020. Methods: Six EHR efficiency indicators were drawn from the Epic EHR Signal efficiency indicators which are automatically generated monthly for clinicians. Indicators include: use of quick actions, preference lists, level of service and diagnosis speed buttons and chart search functions and notes written with smart tools. A weighted score of these indicators represented overall EHR proficiency with a possible range of 0-10 with higher scores suggesting higher efficiency. Latent profile analysis (LPA) was used to identify clusters of clinicians defined by distinct patterns of efficiency indicators. Results: The median EHR proficiency score was 4 and ranged from 0 to 9.5. The 9 cluster LPA solution the best overall fit. Clusters represented different patterns with notable strengths and deficiencies in EHR use efficiency. Cluster 2 was large (n=518) and characterized by being low on all indicators except for chart search. In contrast, cluster 9 was very small (n=12) and characterized by high scores for 4/6 indicators and low on quick actions. Conclusions: Substantial variation in EHR use efficiency among primary care clinicians was observed and can be represented by nine different patterns of indicators. These patterns show strengths and deficiencies in EHR use efficiency and could guide a tailored training approach to improve EHR efficiency.