Health Services Research最新文献

{"title":"Evaluating Methods for Imputing Race and Ethnicity in Electronic Health Record Data.","authors":"Sarah Conderino, Jasmin Divers, John A Dodson, Lorna E Thorpe, Mark G Weiner, Samrachana Adhikari","doi":"10.1111/1475-6773.14649","DOIUrl":"https://doi.org/10.1111/1475-6773.14649","url":null,"abstract":"<p><strong>Objective: </strong>To compare anonymized and non-anonymized approaches for imputing race and ethnicity in descriptive studies of chronic disease burden using electronic health record (EHR)-based datasets.</p><p><strong>Study setting and design: </strong>In this New York City-based study, we first conducted simulation analyses under different missing data mechanisms to assess the performance of Bayesian Improved Surname Geocoding (BISG), single imputation using neighborhood majority information, random forest imputation, and multiple imputation with chained equations (MICE). Imputation performance was measured using sensitivity, precision, and overall accuracy; agreement with self-reported race and ethnicity was measured with Cohen's kappa (κ). We then applied these methods to impute race and ethnicity in two EHR-based data sources and compared chronic disease burden (95% CIs) by race and ethnicity across imputation approaches.</p><p><strong>Data sources and analytic sample: </strong>Our data sources included EHR data from NYU Langone Health and the INSIGHT Clinical Research Network from 3/6/2016 to 3/7/2020 extracted for a parent study on older adults in NYC with multiple chronic conditions.</p><p><strong>Principal findings: </strong>Under simulation analyses, the non-anonymized BISG imputation provided the most accurate classification of race and ethnicity, ranging from 66% to 73% across missing data mechanisms. Anonymized imputation methods were more sensitive to the missing data mechanism, with agreement dropping when race and ethnicity was missing not at random (MNAR) (κ_single = 0.25, κ_MICE = 0.25, κ_randomforest = 0.33). When these methods were applied to the NYU and INSIGHT cohorts, however, racial and ethnic distributions and chronic disease burden were consistent across all imputation methods. Slight improvements in the precision of estimates were observed under all imputation approaches compared to a complete case analysis.</p><p><strong>Conclusions: </strong>BISG imputation may provide a more accurate racial and ethnic classification than single or multiple imputation using anonymized covariates, particularly if the missing data mechanism is MNAR. Descriptive studies of disease burden may not be sensitive to methods for imputing missing data.</p>","PeriodicalId":55065,"journal":{"name":"Health Services Research","volume":" ","pages":"e14649"},"PeriodicalIF":3.1,"publicationDate":"2025-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144152730","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Time on Your Side: Aggregating Data in Difference-In-Differences Studies.","authors":"Summer Rak, Laura A Hatfield, Carrie E Fry","doi":"10.1111/1475-6773.14636","DOIUrl":"https://doi.org/10.1111/1475-6773.14636","url":null,"abstract":"<p><strong>Objective: </strong>To compare the performance of difference-in-differences estimators fit to data aggregated to different time scales.</p><p><strong>Study setting and design: </strong>In simulations, we generated monthly observations for 50-100 units over 6 years from both a parametric model and a resampling simulation. The simulation scenarios varied panel balance, treatment timing, and true treatment effects. Our target parameters were static and dynamic average effects of treatment on the treated (ATT) estimated via linear regression (for common timing scenarios) and Callaway and Sant'Anna (2021) estimators (for staggered timing scenarios). We compared estimates from monthly, quarterly, and yearly data using bias, standard error, root mean squared error (RMSE), power, and Type I error. We also conducted a case study to illustrate the real-world impacts of these decisions.</p><p><strong>Data sources and analytic sample: </strong>We used data from a study of police retraining for the resampling simulations and case study. These data included counts of use-of-force incidents and dates of training enrollment for 8614 officers each month from 2011 to 2016.</p><p><strong>Principal findings: </strong>Results from the simulation varied across performance metrics, estimation methods, target estimands, and data structures. In general, the choice of time aggregation was more consequential when estimating dynamic (versus static) treatment effects, in unbalanced (versus balanced) panel data, and in the resampling simulations (where data had less autocorrelation). Although time aggregation mattered little in many scenarios, coarser aggregation was preferable in resampling simulations of staggered timing scenarios. The re-analysis of police training data was sensitive to time aggregation.</p><p><strong>Conclusions: </strong>In many scenarios, time aggregation has little impact on difference-in-differences estimators. However, when estimating dynamic effects, especially in staggered timing settings and unbalanced data, we found a tradeoff between precision and power, with finer aggregations being more powerful but less precise. In addition, estimators that use a single reference time point are more sensitive to noise in data measured at finer time scales.</p>","PeriodicalId":55065,"journal":{"name":"Health Services Research","volume":" ","pages":"e14636"},"PeriodicalIF":3.1,"publicationDate":"2025-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144152792","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"From the Editors' Desk: HSR's Report on the Demographics, Education, Employment, and Publication Experiences of Our Authors, Reviewers, and Editors","authors":"Austin Frakt,&nbsp;Chris Tachibana","doi":"10.1111/1475-6773.14644","DOIUrl":"10.1111/1475-6773.14644","url":null,"abstract":"","PeriodicalId":55065,"journal":{"name":"Health Services Research","volume":"60 3","pages":""},"PeriodicalIF":3.1,"publicationDate":"2025-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144133201","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Impact of Community Health Center Losses on County-Level Mortality: A Natural Experiment in the United States, 2011-2019.","authors":"Sanjay Basu, Robert Phillips, Hank Hoang","doi":"10.1111/1475-6773.14648","DOIUrl":"https://doi.org/10.1111/1475-6773.14648","url":null,"abstract":"<p><strong>Objective: </strong>To estimate the effect of Community Health Center (CHC) site losses on county-level mortality rates.</p><p><strong>Study setting and design: </strong>We conducted a natural experiment study using difference-in-differences analysis of propensity score-matched US counties from 2011 through 2019.</p><p><strong>Data sources and analytic sample: </strong>The study included 3142 US counties, with 177 counties experiencing CHC site losses in 2014, per data from the health resources and services administration.</p><p><strong>Principal findings: </strong>Loss of CHC sites was associated with an increase in age-adjusted all-cause mortality of 3.54 deaths per 100 000 population (95% CI: 1.19, 5.90; p = 0.003) in the year following the loss. The largest increase was observed for cancer mortality (2.61 per 100 000; 95% CI: 0.59, 4.62; p = 0.011). Primary care physician density and patient volume loss both mediated the relationship.</p><p><strong>Conclusions: </strong>CHC site losses were associated with increases in mortality. Preserving CHC access may be important for maintaining population health, particularly in underserved areas.</p>","PeriodicalId":55065,"journal":{"name":"Health Services Research","volume":" ","pages":"e14648"},"PeriodicalIF":3.1,"publicationDate":"2025-05-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144129546","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Telehealth Use by Home Health Agencies Before, During, and After COVID-19.","authors":"Dana B Mukamel, Debra Saliba, Heather Ladd, Melissa A Clark, Michelle L Rogers, Cheryl Meyer Nelson, Marisa L Roczen, Dara H Sorkin, Jacqueline S Zinn, Peter Huckfeldt","doi":"10.1111/1475-6773.14645","DOIUrl":"https://doi.org/10.1111/1475-6773.14645","url":null,"abstract":"<p><strong>Objective: </strong>To examine telehealth adoption and discontinuation by home health agencies (HHAs) during the COVID-19 pandemic in the context of telehealth pre-pandemic diffusion into the industry and its continued use once the pandemic abated.</p><p><strong>Study setting and design: </strong>HHAs nationally, serving the most patients with dementia (averaging 33% of the agency's patients) were surveyed during October 2023 to November 2024. Key variables included the agency's adoption and discontinuation of specific telehealth technologies by year, the reasons for discontinuation, and the reasons for not adopting any telehealth technology, either before or during the pandemic.</p><p><strong>Data sources and analytic sample: </strong>Data were collected via a web-based survey with telephone follow-ups. We received 791 responses (37% response-rate) and provide descriptive statistics of responses and graphics.</p><p><strong>Principal findings: </strong>By 2019, prior to COVID-19, 183 (23%) of HHAs used telehealth, increasing to 446 (56%) by 2021. Growth occurred mainly in virtual visits. Of those HHAs adopting telehealth, 96 (19%) discontinued use later in the pandemic. Key concerns were about the appropriateness of the patient population and reimbursement.</p><p><strong>Conclusions: </strong>Patterns of adoption and discontinuation suggest that COVID-19 interrupted the innovation diffusion process of telehealth into home health. Telehealth's future will depend on information about cost-effectiveness and Medicare reimbursement policies.</p>","PeriodicalId":55065,"journal":{"name":"Health Services Research","volume":" ","pages":"e14645"},"PeriodicalIF":3.1,"publicationDate":"2025-05-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144121487","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The 2005 TennCare Disenrollments Increased Rates of Intimate Partner Violence: Insights for the Post-COVID Medicaid Unwinding.","authors":"Wei Fu, Melissa B Eggen, Qi Zheng","doi":"10.1111/1475-6773.14647","DOIUrl":"https://doi.org/10.1111/1475-6773.14647","url":null,"abstract":"<p><strong>Objective: </strong>To quantify the impact of 2005 TennCare mandatory disenrollment on intimate partner violence (IPV).</p><p><strong>Study setting and design: </strong>Using 2005 TennCare mandatory disenrollment as a quasi-natural experiment, we classified counties with pre-reform TennCare enrollment rates above the median as high-exposure (treated group) and those below the median as low-exposure (control group). We employed a doubly robust Difference-in-Differences design that compares the high-exposure counties with the low-exposure counties before and after the TennCare disenrollment, adjusting for various county-level characteristics. We performed a dose-response DID model using the pre-reform TennCare enrollment rates as a continuous treatment for robustness checks.</p><p><strong>Data sources and analytic sample: </strong>We extracted IPV incidents from 2003 to 2008 National Incident-Based Reporting System, supplemented with data from Surveillance, Epidemiology, and End Results program and Area Health Resource Files. We aggregated IPV incidents by police agency, year, and quarter to create an aggregate-level count of IPV per 10,000 population. The analytic sample consists of 6780 agency-year-quarter observations.</p><p><strong>Principal findings: </strong>Our findings revealed that TennCare disenrollment led to an increase of 4.8 (95% CI: 0.90-8.6) IPV incidents per 10,000 population, approximately a 25.8% (95% CI: 4.9%-46.7%) increase. This impact was more salient among younger adults and women. Importantly, the disenrollment led to a significantly larger increase in drug-related IPV incidents. Our welfare calculation shows that the disenrollment resulted in a total economic burden of approximately $232.34 (95% CI: 43.95-420.94) million, with the government bearing $86.66 (95% CI: 16.39-157) million of this cost.</p><p><strong>Conclusions: </strong>This study underscores the importance of considering broader social costs associated with abrupt loss of insurance coverage. These findings support the need for Medicaid reforms that incorporate safety measures to prevent unintended social harms and support the well-being of economically vulnerable populations.</p>","PeriodicalId":55065,"journal":{"name":"Health Services Research","volume":" ","pages":"e14647"},"PeriodicalIF":3.1,"publicationDate":"2025-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144113072","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Site-Level Differences in the Provision of Telemental Health Care Within the Department of Veterans Affairs.","authors":"Samantha L Connolly, Rebecca A Raciborski, Hassen Abdulkerim, Jennifer L Sullivan, Leonie K Heyworth, Kendra R Weaver, Lisa C Eisele, Stephanie L Shimada, Jan A Lindsay, Timothy P Hogan, Christopher J Miller","doi":"10.1111/1475-6773.14639","DOIUrl":"https://doi.org/10.1111/1475-6773.14639","url":null,"abstract":"<p><strong>Objective: </strong>To examine site-level differences in telemental health use within the Department of Veterans Affairs (VA). Findings aim to identify barriers to telemental health use to improve access to care.</p><p><strong>Study setting and design: </strong>122 VA facilities were classified into three groups: sites with higher levels of in-person (n = 55), video (n = 40), and phone mental health (MH) care (n = 27). We used Pearson's chi-squared and F-tests to assess for group differences on organizational characteristics and patient population variables.</p><p><strong>Data sources and analytic sample: </strong>This was an observational study using VA administrative data from July 2021 to October 2022; analyses were conducted from June 2024 to March 2025.</p><p><strong>Principal findings: </strong>Sites in the video group tended to be larger, high-complexity, urban facilities that served more women, younger patients, and patients with greater broadband access. Sites in the in-person group served more patients of lower socioeconomic status and treated the highest percentage of rural patients. The phone group served the next highest percentage of rural patients, followed by the video group.</p><p><strong>Conclusions: </strong>Larger, higher-complexity sites may have stronger telehealth infrastructures, and urban areas have stronger broadband connectivity to support video visits. Smaller, rural sites may benefit from targeted support to increase video use.</p>","PeriodicalId":55065,"journal":{"name":"Health Services Research","volume":" ","pages":"e14639"},"PeriodicalIF":3.1,"publicationDate":"2025-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144082194","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Comparing Outcomes Among Medicaid Home and Community-Based Service Users: Private Homes vs. Assisted Living Facilities.","authors":"Hyunjee Kim, Katherine Courchaine, Angela Senders, Clint Sergi, Hiroko H Dodge, R Tamara Konetzka","doi":"10.1111/1475-6773.14643","DOIUrl":"https://doi.org/10.1111/1475-6773.14643","url":null,"abstract":"<p><strong>Objective: </strong>To compare outcomes of Medicaid home and community-based services (HCBS) users residing in a private home vs. in an assisted living facility among dually eligible individuals aged 65 or older.</p><p><strong>Study setting and design: </strong>Medicaid HCBS occurred either in private homes or assisted living facilities in 883 counties in 39 states in the United States from 2016 to 2019. We used an instrumental variable (IV) approach to account for unmeasured factors that might confound the association between HCBS settings and outcomes. Our IV was the monthly proportion of HCBS users in a private home (vs. in an assisted living facility) in surrounding counties within the same state. We examined four outcomes: hospitalizations, emergency department visits, days at home, and Medicare spending.</p><p><strong>Data sources and analytic sample: </strong>We used Medicare and Medicaid claims linked at the individual level. Our sample included dually eligible individuals aged 65 or older, enrolled in Medicare fee-for-service plans, who used HCBS either in a private home or an assisted living facility, between 2016 and 2019.</p><p><strong>Principal findings: </strong>Our sample included 8,140,213 person-months from 383,607 individuals, of whom 85% lived at home and 15% in assisted living facilities. Compared to those in assisted living facilities, in-home HCBS users were 1.3 percentage points more likely to be hospitalized each month (95% CI: 0.8-1.8) and had 0.3 fewer days at home per month (95% CI: -0.4 to -0.1). In an analysis using a sample matched on observed person-months, HCBS users at home had higher likelihoods of hospitalizations and emergency department visits, fewer days at home, and higher Medicaid spending.</p><p><strong>Conclusions: </strong>Overall, in-home HCBS users were more likely to experience adverse health events than those in assisted living facilities, suggesting that policymakers should consider improving care for HCBS users in private homes.</p>","PeriodicalId":55065,"journal":{"name":"Health Services Research","volume":" ","pages":"e14643"},"PeriodicalIF":3.1,"publicationDate":"2025-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144082191","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Correction to “Co-Creating a Theory of Change to Advance COVID-19 Testing and Vaccine Uptake in Underserved Communities”","authors":"","doi":"10.1111/1475-6773.14641","DOIUrl":"10.1111/1475-6773.14641","url":null,"abstract":"<p>\u0000 \u0000 <span>Stadnick, NA</span>, <span>Cain, KL</span>, <span>Oswald, W</span>, et al. “ <span>Co-Creating a Theory of Change to Advance COVID-19 Testing and Vaccine Uptake in Underserved Communities</span>.” <i>Health Serv Res</i> <span>2022</span>; <span>57</span>(<span>Suppl. 1</span>): <span>149</span>–<span>157</span>. https://doi.org/10.1111/1475-6773.13910.\u0000 </p><p>In the published version of this article, the following funding acknowledgment should be added to the Funding section:</p><p>“This research was, in part, funded by the National Institutes of Health (NIH) Agreement OT2HL158287. The views and conclusions contained in this document are those of the authors and should not be interpreted as representing the official policies, either expressed or implied, of the NIH.”We apologize for this error.</p>","PeriodicalId":55065,"journal":{"name":"Health Services Research","volume":"60 3","pages":""},"PeriodicalIF":3.1,"publicationDate":"2025-05-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1111/1475-6773.14641","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144019374","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Does Family-Centered Care Coordination Improve Outcomes for Children Enrolled in Early Intervention Services? An Analysis of Colorado's GO4IT Initiative.","authors":"Richard C Lindrooth, Natalie J Murphy, Rick Devoss, Elizabeth Boggs, Mary A Khetani, Gregory J Tung, Juliane Daniel, Beth McManus","doi":"10.1111/1475-6773.14634","DOIUrl":"https://doi.org/10.1111/1475-6773.14634","url":null,"abstract":"<p><strong>Objective: </strong>To measure the effect of the adoption of a family-centered care coordination approach in Colorado's early intervention program on family engagement and goal attainment outcomes.</p><p><strong>Study setting and design: </strong>We estimated difference-in-difference models to estimate the effect of Colorado's approach to family-centered care coordination, using Massachusetts as a control group. Massachusetts was chosen because there were no major changes to family-centered care coordination during our sample period. The primary outcome measures one of four exit reasons: disengagement (measured as lost to follow-up or withdrawal), achievement of care plan goals before Age 3, no longer eligible due to Age 3, or moved. Multinomial probit models estimated the effect of Colorado's program on the probability of each exit reason. Additional analyses were stratified by child race and ethnicity groups.</p><p><strong>Data sources and analytic sample: </strong>We obtained administrative early intervention data from Massachusetts and Colorado for children who were referred to early intervention in fiscal years 2013-2019 and had a documented exit reason. The full sample includes 103,635 children (n = 20,045 in Colorado; n = 83,590 in Massachusetts).</p><p><strong>Principal findings: </strong>Participation in Colorado's program was associated with a 4.6 percentage point reduction in family disengagement (p < 0.01) and a 3.5-4.6 percentage point higher likelihood of meeting early intervention care plan goals before Age 3 (p < 0.01). Estimates by race and ethnicity revealed improved engagement of White, Black, and Hispanic families, though only White and Hispanic children were more likely to achieve care plan goals before Age 3.</p><p><strong>Conclusions: </strong>Colorado's family-centered care coordination intervention improved family engagement with early intervention services, with fewer families lost to follow-up or withdrawn following early intervention care plan development. These findings can inform evidence-based recommendations to other states considering family-centered care coordination solutions to optimize early intervention service delivery and equity.</p>","PeriodicalId":55065,"journal":{"name":"Health Services Research","volume":" ","pages":"e14634"},"PeriodicalIF":3.1,"publicationDate":"2025-05-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144013017","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}