NPSV-deep: a deep learning method for genotyping structural variants in short read genome sequencing data.

MOTIVATION: Structural variants (SVs) play a causal role in numerous diseases but can be difficult to detect and accurately genotype (determine zygosity) with short-read genome sequencing data (SRS). Improving SV genotyping accuracy in SRS data, particularly for the many SVs first detected with long-read sequencing, will improve our understanding of genetic variation. RESULTS: NPSV-deep is a deep learning-based approach for genotyping previously reported insertion and deletion SVs that recasts this task as an image similarity problem. NPSV-deep predicts the SV genotype based on the similarity between pileup images generated from the actual SRS data and matching SRS simulations. We show that NPSV-deep consistently matches or improves upon the state-of-the-art for SV genotyping accuracy across different SV call sets, samples and variant types, including a 25% reduction in genotyping errors for the Genome-in-a-Bottle (GIAB) high-confidence SVs. NPSV-deep is not limited to the SVs as described; it improves deletion genotyping concordance a further 1.5 percentage points for GIAB SVs (92%) by automatically correcting imprecise/incorrectly described SVs. AVAILABILITY AND IMPLEMENTATION: Python/C++ source code and pre-trained models freely available at https://github.com/mlinderm/npsv2.

Long-Term Stability of Coverage Among Michigan Medicaid Beneficiaries : A Cohort Study.

BACKGROUND: Medicaid, the primary source of insurance coverage for disadvantaged Americans, was originally designed as a temporary safety-net program. No studies have used long-run data to assess the recent use of the program by beneficiaries. OBJECTIVE: To assess patterns of short- and long-term enrollment among beneficiaries, using a 10-year longitudinal panel of Michigan Medicaid eligibility data. DESIGN: Primary analyses assessing trends in Medicaid enrollment among cohorts of existing and new beneficiaries. SETTING: Administrative records from Michigan Medicaid for the period 2011 to 2020. PARTICIPANTS: 3.97 million Medicaid beneficiaries. MEASUREMENTS: Short- and long-term enrollment in the program. RESULTS: The sample includes 3.97 million unique beneficiaries enrolled at some point between 2011 and 2020. Among a cohort of 1.23 million beneficiaries enrolled in 2011, over half (53%) were also enrolled in Medicaid in June 2020, spending, on average, two-thirds of that period (67%) on Medicaid. These beneficiaries, however, experienced substantial lapses in coverage, as only 25% were continuously enrolled throughout the period. Enrollment was less stable when assessed from the perspective of newly enrolled beneficiaries, of whom only 37% remained enrolled at the end of the study period. LIMITATION: Primary estimates from a single state. CONCLUSION: For many beneficiaries, Medicaid has served as their primary source of coverage for at least a decade. This pattern would justify increasing investments in the program to improve long-term health outcomes. PRIMARY FUNDING SOURCE: Self-funded.

Conditional cross-design synthesis estimators for generalizability in Medicaid.

While much of the causal inference literature has focused on addressing internal validity biases, both internal and external validity are necessary for unbiased estimates in a target population of interest. However, few generalizability approaches exist for estimating causal quantities in a target population that is not well-represented by a randomized study but is reflected when additionally incorporating observational data. To generalize to a target population represented by a union of these data, we propose a novel class of conditional cross-design synthesis estimators that combine randomized and observational data, while addressing their estimates' respective biases-lack of overlap and unmeasured confounding. These methods enable estimating the causal effect of managed care plans on health care spending among Medicaid beneficiaries in New York City, which requires obtaining estimates for the 7% of beneficiaries randomized to a plan and 93% who choose a plan, who do not resemble randomized beneficiaries. Our new estimators include outcome regression, propensity weighting, and double robust approaches. All use the covariate overlap between the randomized and observational data to remove potential unmeasured confounding bias. Applying these methods, we find substantial heterogeneity in spending effects across managed care plans. This has major implications for our understanding of Medicaid, where this heterogeneity has previously been hidden. Additionally, we demonstrate that unmeasured confounding rather than lack of overlap poses a larger concern in this setting.

Stretch-Induced Uncrimping of Equatorial Sclera Collagen Bundles.

Stretch-induced collagen uncrimping underlies the nonlinear mechanical behavior of the sclera according to what is often called the process of recruitment. We recently reported experimental measurements of sclera collagen crimp and pressure-induced uncrimping. Our studies, however, were cross-sectional, providing statistical descriptions of crimp with no information on the effects of stretch on specific collagen bundles. Data on bundle-specific uncrimping is necessary to better understand the effects of macroscale input on the collagen microscale and tissue failure. Our goal in this project was to measure bundle-specific stretch-induced collagen uncrimping of sclera. Three goat eyes were cryosectioned sagittally (30 µm). Samples of equatorial sclera were isolated, mounted to a custom uni-axial stretcher and imaged with polarized light microscopy at various levels of clamp-to-clamp stretch until failure. At each stretch level, local strain was measured using image tracking techniques. The level of collagen crimping was determined from the bundle waviness, defined as the circular standard deviation of fiber orientation along a bundle. Eye-specific recruitment curves were then computed using eye-specific waviness at maximum stretch before sample failure to define fibers as recruited. Nonlinear mixed effect models were used to determine the associations of waviness to local strain and recruitment to clamp-to-clamp stretch. Waviness decreased exponentially with local strain (p < 0.001), whereas bundle recruitment followed a sigmoidal curve with clamp-to-clamp stretch (p < 0.001). Individual bundle responses to stretch varied substantially, but recruitment curves were similar across sections and eyes. In conclusion, uni-axial stretch caused measurable bundle-specific uncrimping, with the sigmoidal recruitment pattern characteristic of fiber-reinforced soft tissues.

Residual Confounding in Health Plan Performance Assessments: Evidence From Randomization in Medicaid.

BACKGROUND: Risk adjustment is used widely in payment systems and performance assessments, but the extent to which it distinguishes plan or provider effects from confounding due to patient differences is typically unknown. OBJECTIVE: To assess the degree to which risk-adjusted measures of health plan performance adequately adjust for the variation across plans that arises because of differences in patient characteristics (residual confounding). DESIGN: Comparison between plan performance estimates based on enrollees who made plan choices (observational population) and estimates based on enrollees assigned to plans (randomized population). SETTING: Natural experiment in which more than two thirds of a state's Medicaid population in 1 region was randomly assigned to 1 of 5 plans. PARTICIPANTS: 137 933 enrollees in 2013 to 2014, of whom 31.1% selected a plan and 68.9% were randomly assigned to 1 of the same 5 plans. MEASUREMENTS: Annual total spending (that is, payments to providers), primary care use, dental care use, and avoidable emergency department visits, all scored as plan-specific deviations from the "average" plan performance within each population. RESULTS: Enrollee characteristics were appreciably imbalanced across plans in the observational population, as expected, but were not in the randomized population. Annual total spending varied across plans more in the observational population (SD, $147 per enrollee) than in the randomized population (SD, $70 per enrollee) after accounting for baseline differences in the observational and randomized populations and for differences across plans. On average, a plan's spending score (its deviation from the "average" performance) in the observational population differed from its score in the randomized population by $67 per enrollee in absolute value (95% CI, $38 to $123), or 4.2% of mean spending per enrollee (P = 0.009, rejecting the null hypothesis that this difference would be expected from sampling error). The difference was reduced modestly by risk adjustment to $62 per enrollee (P = 0.012). Residual confounding was similarly substantial for most other performance measures. Further adjustment for social factors did not materially change estimates. LIMITATION: Potential heterogeneity in plan effects between the 2 populations. CONCLUSION: Residual confounding in risk-adjusted performance assessments can be substantial and should caution policymakers against assuming that risk adjustment isolates real differences in plan performance. PRIMARY FUNDING SOURCE: Arnold Ventures.

Racial/Ethnic Disparities in Substance Use Treatment in Medicaid Managed Care in New York City: The Role of Plan and Geography.

OBJECTIVE: The aim was to assess the magnitude of health care disparities in treatment for substance use disorder (SUD) and the role of health plan membership and place of residence in observed disparities in Medicaid Managed Care (MMC) plans in New York City (NYC). DATA SOURCE: Medicaid claims and managed care plan enrollment files for 2015-2017 in NYC. RESEARCH DESIGN: We studied Medicaid enrollees with a SUD diagnosis during their first 6 months of enrollment in a managed care plan in 2015-2017. A series of linear regression models quantified service disparities across race/ethnicity for 5 outcome indicators: treatment engagement, receipt of psychosocial treatment, follow-up after withdrawal, rapid readmission, and treatment continuation. We assessed the degree to which plan membership and place of residence contributed to observed disparities. RESULTS: We found disparities in access to treatment but the magnitude of the disparities in most cases was small. Plan membership and geography of residence explained little of the observed disparities. One exception is geography of residence among Asian Americans, which appears to mediate disparities for 2 of our 5 outcome measures. CONCLUSIONS: Reallocating enrollees among MMC plans in NYC or evolving trends in group place of residence are unlikely to reduce disparities in treatment for SUD. System-wide reforms are needed to mitigate disparities.

Interplay between intraocular and intracranial pressure effects on the optic nerve head in vivo.

Intracranial pressure (ICP) has been proposed to play an important role in the sensitivity to intraocular pressure (IOP) and susceptibility to glaucoma. However, the in vivo effects of simultaneous, controlled, acute variations in ICP and IOP have not been directly measured. We quantified the deformations of the anterior lamina cribrosa (ALC) and scleral canal at Bruch's membrane opening (BMO) under acute elevation of IOP and/or ICP. Four eyes of three adult monkeys were imaged in vivo with OCT under four pressure conditions: IOP and ICP either at baseline or elevated. The BMO and ALC were reconstructed from manual delineations. From these, we determined canal area at the BMO (BMO area), BMO aspect ratio and planarity, and ALC median depth relative to the BMO plane. To better account for the pressure effects on the imaging, we also measured ALC visibility as a percent of the BMO area. Further, ALC depths were analyzed only in regions where the ALC was visible in all pressure conditions. Bootstrap sampling was used to obtain mean estimates and confidence intervals, which were then used to test for significant effects of IOP and ICP, independently and in interaction. Response to pressure manipulation was highly individualized between eyes, with significant changes detected in a majority of the parameters. Significant interactions between ICP and IOP occurred in all measures, except ALC visibility. On average, ICP elevation expanded BMO area by 0.17 mm2 at baseline IOP, and contracted BMO area by 0.02 mm2 at high IOP. ICP elevation decreased ALC depth by 10 µm at baseline IOP, but increased depth by 7 µm at high IOP. ALC visibility decreased as ICP increased, both at baseline (-10%) and high IOP (-17%). IOP elevation expanded BMO area by 0.04 mm2 at baseline ICP, and contracted BMO area by 0.09 mm2 at high ICP. On average, IOP elevation caused the ALC to displace 3.3 µm anteriorly at baseline ICP, and 22 µm posteriorly at high ICP. ALC visibility improved as IOP increased, both at baseline (5%) and high ICP (8%). In summary, changing IOP or ICP significantly deformed both the scleral canal and the lamina of the monkey ONH, regardless of the other pressure level. There were significant interactions between the effects of IOP and those of ICP on LC depth, BMO area, aspect ratio and planarity. On most eyes, elevating both pressures by the same amount did not cancel out the effects. Altogether our results show that ICP affects sensitivity to IOP, and thus that it can potentially also affect susceptibility to glaucoma.

Association of Provider Performance with Changes in Insurance Networks.

BACKGROUND: Medicaid managed care plans change provider networks frequently, yet there is no evidence about the performance of exiting providers relative to those that remain. OBJECTIVES: To investigate the association between provider cost and quality and network exit. DESIGN: Observational study with provider network directory data linked to administrative claims from managed care plans in Tennessee's Medicaid program during the period 2010-2016. PARTICIPANTS: 1,966,022 recipients assigned to 9593 unique providers. MAIN MEASURES: Exposures were risk-adjusted total costs of care and nine measures from the Healthcare Effectiveness Data and Information Set (HEDIS) were used to construct a composite annual indicators of provider performance on quality. Outcome was provider exit from a Medicaid managed care plan. Differences in quality and cost between providers that exited and remained in managed care networks were estimated using a propensity score model to match exiting to nonexiting providers. KEY RESULTS: Over our study period, we found that 21% of participating providers exited at least one of the Medicaid managed care plans in Tennessee. As compared with providers that remained in networks, those that exited performed 3.8 percentage points [95% CI, 2.3, 5.3] worse on quality as measured by a composite of the nine HEDIS quality metrics. However, 22% of exiting providers performed above average in quality and cost and only 29% of exiting providers had lower than average quality scores and higher than average costs. Overall, exiting providers had lower aggregate costs in terms of the annual unadjusted cost of care per-member-month - $21.57 [95% CI, - $41.02, - $2.13], though difference in annual risk-adjusted cost per-member-month was nonsignificant. CONCLUSIONS: Providers exiting Medicaid managed care plans appear to have lower quality scores in the year prior to their exit than the providers who remain in network. Our study did not show that managed care plans disproportionately drop high-cost providers.

Crimp around the globe; patterns of collagen crimp across the corneoscleral shell.

Our goal was to systematically quantify the collagen crimp morphology around the corneoscleral shell, and test the hypothesis that collagen crimp is not uniform over the globe. Axial longitudinal cryosections (30 µm) of three sheep eyes, fixed at 0 mmHg IOP, were imaged using polarized light microscopy to quantify the local collagen in 8 regions: two corneal (central and peripheral) and six scleral (limbus, anterior-equatorial, equatorial, posterior-equatorial, posterior and peripapillary). Collagen crimp period (length of one wave), tortuosity (path length divided by end-to-end length), waviness (SD of orientation), amplitude (half the peak to trough distance), and conformity (width of contiguous similarly oriented bundles) were measured in each region. Measurements were obtained on 8216 collagen fiber bundles. When pooling measurements across the whole eye globe, the median crimp values were: 23.9 µm period, 13.2 µm conformity, 0.63 µm amplitude, 1.006 tortuosity, and 12.7° waviness. However, all parameters varied significantly across the globe. Median bundle periods in the central cornea, limbus, and peripapillary sclera (PPS) were 14.1 µm, 29.5 µm, and 22.9 µm, respectively. Median conformities were 20.8 µm, 14.5 µm, and 15.1 µm, respectively. Median tortuosities were 1.005, 1.007, and 1.007, respectively. Median waviness' were 11.4°, 13.2°, and 13.2°, respectively. Median amplitudes were 0.35 µm, 0.87 µm, and 0.65 µm, respectively. All parameters varied significantly across the globe. All regions differed significantly from one another on at least one parameter. Regions with small periods had large conformities, and bundles with high tortuosity had high waviness and amplitude. Waviness, tortuosity, and amplitude, associated with nonlinear biomechanical behavior, exhibited "double hump" distributions, whereas period and conformity, representing tissue organization, were substantially different between sclera and cornea. Though the biomechanical implications and origin of the patterns observed remain unclear, our findings of well-defined patterns of collagen crimp across the corneoscleral shell, consistent between eyes, support the existence of mechanisms that regulate collagen characteristics at the regional or smaller levels. These results are experimental data necessary for more realistic models of ocular biomechanics and remodeling.

Medicare Eligibility and Changes in Coverage, Access to Care, and Health by Sexual Orientation and Gender Identity.

Importance: Medicare provides nearly universal insurance coverage at age 65 years. However, how Medicare eligibility affects disparities in health insurance coverage, access to care, and health status among individuals by sexual orientation and gender identity is poorly understood. Objective: To assess the association of Medicare eligibility with disparities in health insurance coverage, access to care, and self-reported health status among individuals by sexual orientation and by gender identity. Design, Setting, and Participants: This cross-sectional study used the age discontinuity for Medicare eligibility at age 65 years to isolate the association of Medicare with health insurance coverage, access to care, and self-reported health status, by their sexual orientation and by their gender identity. Data were collected from the Behavioral Risk Factor Surveillance System for respondents from 51 to 79 years old from 2014 to 2021. Data analysis was performed from September 2022 to April 2023. Exposures: Medicare eligibility at age 65 years. Main Outcomes and Measures: Proportions of respondents with health insurance coverage, usual source of care, cost barriers to care, influenza vaccination, and self-reported health status. Results: The study population included 927 952 individuals (mean [SD] age, 64.4 [7.7] years; 524 972 [56.6%] females and 402 670 [43.4%] males), of whom 28 077 (3.03%) identified as a sexual minority-lesbian, gay, bisexual, or another sexual minority identity (LGB+) and 3286 (0.35%) as transgender or gender diverse. Respondents who identified as heterosexual had greater improvements at age 65 years in insurance coverage (4.2 percentage points [pp]; 95% CI, 4.0-4.4 pp) than those who identified as LGB+ (3.6 pp; 95% CI, 2.3-4.8 pp), except when the analysis was limited to a subsample of married respondents. For access to care, improvements in usual source of care, cost barriers to care, and influenza vaccination were larger at age 65 years for heterosexual respondents compared with LGB+ respondents, although confidence intervals were overlapping and less precise for LGB+ individuals. For self-reported health status, the analyses found larger improvements at age 65 years for LGB+ respondents compared with heterosexual respondents. There was considerable heterogeneity by state in disparities by sexual orientation among individuals who were nearly eligible for Medicare (close to 65 years old), with the US South and Central states demonstrating the highest disparities. Among the top-10 highest-disparities states, Medicare eligibility was associated with greater increases in coverage (6.7 pp vs 5.0 pp) and access to a usual source of care (1.4 pp vs 0.6 pp) for LGB+ respondents compared with heterosexual respondents. Conclusions and Relevance: The findings of this cross-sectional study indicate that Medicare eligibility was not associated with consistently greater improvements in health insurance coverage and access to care among LGBTQI+ individuals compared with heterosexual and/or cisgender individuals. However, among sexual minority individuals, Medicare may be associated with closing gaps in self-reported health status, and among states with the highest disparities, it may improve health insurance coverage, access to care, and self-reported health status.

Alternative Payment Models and Performance in Federally Qualified Health Centers.

Importance: Several state Medicaid agencies have transitioned from traditional fee-for-service to a value-centric alternative payment model (APM) to reimburse federally qualified health centers (FQHCs). Little is known about the effects of this shift on FQHC performance. Objective: To assess the association between APMs and the clinical performance, payer mix, risk profile, and financial sustainability of FQHCs. Design, Setting, and Participants: This retrospective cohort study was performed in 684 FQHCs (representing 37 states plus the District of Columbia) that continuously operated between January 2009 and December 2021. Data on payer mix (eg, type of insurance) and risk profile (eg, proportion of patients with chronic conditions) of FQHC patients were obtained from the Uniform Data System, and clinic-level financial data (eg, revenue) were obtained from Internal Revenue Service form 990 tax documents. Data were analyzed between November 2022 and October 2023. Exposure: Initial rollout of a value-based payment model (ie, an APM) for FQHCs, as offered by state Medicaid program, between January 2013 and December 2021. Main Outcomes and Measures: The main outcomes were 4 audited process measures of health care quality (cervical and colorectal cancer screening and body mass index [BMI] assessment for adults and children) and 2 intermediate health outcome measures (hypertension control and diabetes control). A difference-in-differences design was used with staggered implementation comparing FQHCs before and after the initial APM rollout vs contemporaneous changes in FQHCs in states without APMs. Results: A total of 684 FQHCs (8892 FQHC-years) that served 17 823 959 patients in 2021 (57.3% female) were included in the study. Among FQHCs in states implementing APMs, significant differential increases in 3 of the 4 process quality measures were observed compared with FQHCs in states that did not implement an APM: colorectal cancer screening (3.24 percentage points [pp]; 95% CI, 1.40-5.08 pp), adult BMI (3.19 pp; 95% CI, 0.70-5.68 pp), and child BMI (4.50 pp; 95% CI, 1.83-7.17 pp). There were also modest differential improvements in blood pressure control for individuals with hypertension (1.02 pp; 95% CI, 0.04-2.00 pp) and blood glucose control for individuals with type 2 diabetes (1.02 pp; 95% CI, 0.02-2.02 pp) compared with FQHCs in states without an APM. There was no evidence that the APM rollout was associated with clinics selecting healthier patients (-0.01 pp; 95% CI, -0.21 to 0.19 pp) or stinting on care (-0.02 visits; 95% CI, -0.08 to 0.04 visits). Conclusions and Relevance: In this cohort study, introduction of Medicaid APM options for FQHCs was associated with modest, statistically significant increases in quality concentrated among FQHCs with APM models that explicitly incentivized quality. This finding suggests that APMs can be both a financially viable and a health-promoting model for reimbursement in the health care safety net.

Adverse selection and network design under regulated plan prices: Evidence from Medicaid.

Health plans for the poor increasingly limit access to specialty hospitals. We investigate the role of adverse selection in generating this equilibrium among private plans in Medicaid. Studying a network change, we find that covering a top cancer hospital causes severe adverse selection, increasing demand for a plan by 50% among enrollees with cancer versus no impact for others. Medicaid's fixed insurer payments make offsetting this selection, and the contract distortions it induces, challenging, requiring either infeasibly high payment rates or near-perfect risk adjustment. By contrast, a small explicit bonus for covering the hospital is sufficient to make coverage profitable.

What Difference Does a Health Plan Make? Evidence from Random Plan Assignment in Medicaid.

Exploiting the random assignment of Medicaid beneficiaries to managed care plans, we find substantial plan-specific spending effects despite plans having identical cost sharing. Enrollment in the lowest-spending plan reduces spending by at least 25%-primarily through quantity reductions-relative to enrollment in the highest-spending plan. Rather than reducing "wasteful" spending, lower-spending plans broadly reduce medical service provision-including the provision of low-cost, high-value care-and worsen beneficiary satisfaction and health. Consumer demand follows spending: a 10 percent increase in plan-specific spending is associated with a 40 percent increase in market share. These facts have implications for the government's contracting problem and program cost growth.

Excess Death Rates for Republican and Democratic Registered Voters in Florida and Ohio During the COVID-19 Pandemic.

Importance: There is evidence that Republican-leaning counties have had higher COVID-19 death rates than Democratic-leaning counties and similar evidence of an association between political party affiliation and attitudes regarding COVID-19 vaccination; further data on these rates may be useful. Objective: To assess political party affiliation and mortality rates for individuals during the initial 22 months of the COVID-19 pandemic. Design, Setting, and Participants: A cross-sectional comparison of excess mortality between registered Republican and Democratic voters between March 2020 and December 2021 adjusted for age and state of voter registration was conducted. Voter and mortality data from Florida and Ohio in 2017 linked to mortality records for January 1, 2018, to December 31, 2021, were used in data analysis. Exposures: Political party affiliation. Main Outcomes and Measures: Excess weekly deaths during the COVID-19 pandemic adjusted for age, county, party affiliation, and seasonality. Results: Between January 1, 2018, and December 31, 2021, there were 538 159 individuals in Ohio and Florida who died at age 25 years or older in the study sample. The median age at death was 78 years (IQR, 71-89 years). Overall, the excess death rate for Republican voters was 2.8 percentage points, or 15%, higher than the excess death rate for Democratic voters (95% prediction interval [PI], 1.6-3.7 percentage points). After May 1, 2021, when vaccines were available to all adults, the excess death rate gap between Republican and Democratic voters widened from -0.9 percentage point (95% PI, -2.5 to 0.3 percentage points) to 7.7 percentage points (95% PI, 6.0-9.3 percentage points) in the adjusted analysis; the excess death rate among Republican voters was 43% higher than the excess death rate among Democratic voters. The gap in excess death rates between Republican and Democratic voters was larger in counties with lower vaccination rates and was primarily noted in voters residing in Ohio. Conclusions and Relevance: In this cross-sectional study, an association was observed between political party affiliation and excess deaths in Ohio and Florida after COVID-19 vaccines were available to all adults. These findings suggest that differences in vaccination attitudes and reported uptake between Republican and Democratic voters may have been factors in the severity and trajectory of the pandemic in the US.

After 50 Years, Health Professional Shortage Areas Had No Significant Impact On Mortality Or Physician Density.

Since 1965, the US federal government has incentivized physicians to practice in high-need areas of the country through the designation of Health Professional Shortage Areas (HPSAs). Despite its being in place for more than half a century and directing more than a billion dollars annually, there is limited evidence of the HPSA program's effectiveness at reducing geographic disparities in access to care and health outcomes. Using a generalized difference-in-differences design with matching, we found no statistically significant changes in mortality or physician density from 1970 to 2018 after a county-level HPSA designation. As a result, we found that 73 percent of counties designated as HPSAs remained physician shortage areas for at least ten years after their inclusion in the program. Fundamental improvements to the program's design and incentive structure may be necessary for it to achieve its intended results.

Risk Adjustment And Promoting Health Equity In Population-Based Payment: Concepts And Evidence.

The objective of risk adjustment is not to predict spending accurately but to support the social goals of a payment system, which include equity. Setting population-based payments at accurate predictions risks entrenching spending levels that are insufficient to mitigate the impact of social determinants on health care use and effectiveness. Instead, to advance equity, payments must be set above current levels of spending for historically disadvantaged groups. In analyses intended to guide such reallocations, we found that current risk adjustment for the community-dwelling Medicare population overpredicts annual spending for Black and Hispanic beneficiaries by $376-$1,264. The risk-adjusted spending for these populations is lower than spending for White beneficiaries despite the former populations' worse risk-adjusted health and functional status. Thus, continued movement from fee-for-service to population-based payment models that omit race and ethnicity from risk adjustment (as current models do) should result in sizable resource reallocations and incentives that support efforts to address racial and ethnic disparities in care. We found smaller overpredictions for less-educated beneficiaries and communities with higher proportions of residents who are Black, Hispanic, or less educated, suggesting that additional payment adjustments that depart from predictive accuracy are needed to support health equity. These findings also suggest that adding social risk factors as predictors to spending models used for risk adjustment may be counterproductive or accomplish little.

Association between Medicare eligibility at age 65 years and in-hospital treatment patterns and health outcomes for patients with trauma: regression discontinuity approach.

OBJECTIVE: To determine whether health systems in the United States modify treatment or discharge decisions for otherwise similar patients based on health insurance coverage. DESIGN: Regression discontinuity approach. SETTING: American College of Surgeons' National Trauma Data Bank, 2007-17. PARTICIPANTS: Adults aged between 50 and 79 years with a total of 1 586 577 trauma encounters at level I and level II trauma centers in the US. INTERVENTIONS: Eligibility for Medicare at age 65 years. MAIN OUTCOME MEASURES: The main outcome measure was change in health insurance coverage, complications, in-hospital mortality, processes of care in the trauma bay, treatment patterns during hospital admission, and discharge locations at age 65 years. RESULTS: 1 586 577 trauma encounters were included. At age 65, a discontinuous increase of 9.6 percentage points (95% confidence interval 9.1 to 10.1) was observed in the share of patients with health insurance coverage through Medicare at age 65 years. Entry to Medicare at age 65 was also associated with a decrease in length of hospital stay for each encounter, of 0.33 days (95% confidence interval -0.42 to -0.24 days), or nearly 5%), which coincided with an increase in discharges to nursing homes (1.56 percentage points, 95% confidence interval 0.94 to 2.16 percentage points) and transfers to other inpatient facilities (0.57 percentage points, 0.33 to 0.80 percentage points), and a large decrease in discharges to home (1.99 percentage points, -2.73 to -1.27 percentage points). Relatively small (or no) changes were observed in treatment patterns during the patients' hospital admission, including no changes in potentially life saving treatments (eg, blood transfusions) or mortality. CONCLUSIONS: The findings suggest that differences in treatment for otherwise similar patients with trauma with different forms of insurance coverage arose during the discharge planning process, with little evidence that health systems modified treatment decisions based on patients' coverage.