Differential Hospital Participation in Bundled Payments in Communities with Higher Shares of Marginalized Populations.

BACKGROUND: Medicare's voluntary bundled payment programs have demonstrated generally favorable results. However, it remains unknown whether uneven hospital participation in these programs in communities with greater shares of minorities and patients of low socioeconomic status results in disparate access to practice redesign innovations. OBJECTIVE: Examine whether communities with higher proportions of marginalized individuals were less likely to be served by a hospital participating in Bundled Payments for Care Improvement Advanced (BPCI-Advanced). DESIGN: Cross-sectional study using ordinary least squares regression controlling for patient and community factors. PARTICIPANTS: Medicare fee-for-service patients enrolled from 2015-2017 (pre-BPCI-Advanced) and residing in 2,058 local communities nationwide defined by Hospital Service Areas (HSAs). Each community's share of marginalized patients was calculated separately for each of the share of beneficiaries of Black race, Hispanic ethnicity, or dual eligibility for Medicare and Medicaid. MAIN MEASURES: Dichotomous variable indicating whether a given community had at least one hospital that ever participated in BPCI-Advanced from 2018-2022. KEY RESULTS: Communities with higher shares of dual-eligible individuals were less likely to be served by a hospital participating in BPCI-Advanced than communities with the lowest quartile of dual-eligible individuals (Q4: -15.1 percentage points [pp] lower than Q1, 95% CI: -21.0 to -9.1, p < 0.001). There was no consistent significant relationship between community proportion of Black beneficiaries and likelihood of having a hospital participating in BPCI-Advanced. Communities with higher shares of Hispanic beneficiaries were more likely to have a hospital participating in BPCI-Advanced than those in the lowest quartile (Q4: 19.2 pp higher than Q1, 95% CI: 13.4 to 24.9, p < 0.001). CONCLUSIONS: Communities with greater shares of dual-eligible beneficiaries, but not racial or ethnic minorities, were less likely to be served by a hospital participating in BPCI-Advanced Policymakers should consider approaches to incentivize more socioeconomically uniform participation in voluntary bundled payments.

Association Between Urbanicity and Outcomes Among Patients with Spinal Cord Ependymomas in the United States.

BACKGROUND: Spinal cord ependymomas (SCEs) represent the most common intramedullary spinal cord tumors among adults. Research shows that access to neurosurgical care and patient outcomes can be greatly influenced by patient location. This study investigates the association between the outcomes of patients with SCE in metropolitan and nonmetropolitan areas. METHODS: Cases of SCE between 2004 and 2019 were identified within the Central Brain Tumor Registry of the United States, a combined dataset including the Centers for Disease Control and Prevention's National Program of Cancer Registries and National Cancer Institute's Surveillance, Epidemiology, and End Results Program data. Multivariable logistic regression models were constructed to evaluate the association between urbanicity and SCE treatment, adjusted for age at diagnosis, sex, race and ethnicity. Survival data was available from 42 National Program of Cancer Registries (excluding Kansas and Minnesota, for which county data are unavailable), and Cox proportional hazard models were used to understand the effect of surgical treatment, county urbanicity, age at diagnosis, and the interaction effect between age at diagnosis and surgery, on the survival time of patients. RESULTS: Overall, 7577 patients were identified, with 6454 (85%) residing in metropolitan and 1223 (15%) in nonmetropolitan counties. Metropolitan and nonmetropolitan counties had different age, sex, and race/ethnicity compositions; however, demographics were not associated with differences in the type of surgery received when stratified by urbanicity. Irrespective of metropolitan status, individuals who were American Indian/Alaska Native non-Hispanic and Hispanic (all races) were associated with reduced odds of receiving surgery. Individuals who were Black non-Hispanic and Hispanic were associated with increased odds of receiving comprehensive treatment. Diagnosis of SCE at later ages was linked with elevated mortality (hazard ratio = 4.85, P < 0.001). Gross total resection was associated with reduced risk of death (hazard ratio = 0.37, P = 0.004), and age did not interact with gross total resection to influence risk of death. CONCLUSIONS: The relationship between patients' residential location and access to neurosurgical care is critical to ensuring equitable distribution of care. This study represents an important step in delineating areas of existing disparities.

How Physician Self-Perceptions Affect the Impact of Peer Comparison Feedback on Opioid Prescribing.

Peer comparison feedback is a promising strategy for reducing opioid prescribing and opioid-related harms. Such comparisons may be particularly impactful among underestimating clinicians who do not perceive themselves as high prescribers relative to their peers. But peer comparisons could also unintentionally increase prescribing among overestimating clinicians who do not perceive themselves as lower prescribers than peers. The objective of this study was to assess if the impact of peer comparisons varied by clinicians' preexisting opioid prescribing self-perceptions. Subgroup analysis of a randomized trial of peer comparison interventions among emergency department and urgent care clinicians was used. Generalized mixed-effects models were used to assess whether the impact of peer comparisons, alone or combined with individual feedback, varied by underestimating or overestimating prescriber status. Underestimating and overestimating prescribers were defined as those who self-reported relative prescribing amounts that were lower and higher, respectively, than actual relative baseline amounts. The primary outcome was pills per opioid prescription. Among 438 clinicians, 54% (n = 236) provided baseline prescribing self-perceptions and were included in this analysis. Overall, 17% (n = 40) were underestimating prescribers whereas 5% (n = 11) were overestimating prescribers. Underestimating prescribers exhibited a differentially greater decrease in pills per prescription compared to nonunderestimating clinicians when receiving peer comparison feedback (1.7 pills, 95% CI, -3.2 to -0.2 pills) or combined peer and individual feedback (2.8 pills, 95% CI, -4.8 to -0.8 pills). In contrast, there were no differential changes in pills per prescription for overestimating versus nonoverestimating prescribers after receiving peer comparison (1.5 pills, 95% CI, -0.9 to 3.9 pills) or combined peer and individual feedback (3.0 pills, 95% CI, -0.3 to 6.2 pills). Peer comparisons were more impactful among clinicians who underestimated their prescribing compared to peers. By correcting inaccurate self-perceptions, peer comparison feedback can be an effective strategy for influencing opioid prescribing.

Disparities in emergency department and urgent care opioid prescribing before and after randomized clinician feedback interventions.

OBJECTIVES: Racial and ethnic minorities receive opioid prescriptions at lower rates and dosages than White patients. Though opioid stewardship interventions can improve or exacerbate these disparities, there is little evidence about these effects. We conducted a secondary analysis of a cluster-randomized controlled trial conducted among 438 clinicians from 21 emergency departments and 27 urgent care clinics. Our objective was to determine whether randomly allocated opioid stewardship clinician feedback interventions that were designed to reduce opioid prescriptions had unintended effects on disparities in prescribing by patient race and ethnicity. METHODS: The primary outcome was likelihood of receiving a low-pill prescription (low ≤10 pills, medium 11-19 pills, high ≥20 pills). Generalized mixed-effects models were used to determine patient characteristics associated with low-pill prescriptions during the baseline period. These models were then used to determine whether receipt of a low-pill prescription varied by patient race or ethnicity during the intervention period between usual care and three opioid stewardship interventions: (1) individual audit feedback, (2) peer comparison feedback, and (3) combined (individual audit + peer comparison) feedback. RESULTS: Compared with White patients, Black patients were more likely to receive a low-pill prescription during the baseline (adjusted odds ratio [OR] 1.18, 95% confidence interval [CI] 1.06-1.31, p = 0.002) and intervention (adjusted OR 1.43, 95% CI 1.07-1.91, p = 0.015). While combined feedback was associated with an overall increase in low-pill prescriptions as intended (adjusted OR 1.89, 95% CI 1.28-2.78, p = 0.001), there were no significant differences in treatment effects of any of the interventions by patient race and ethnicity. CONCLUSIONS: Combined individual audit and peer comparison feedback was associated with fewer opioid pills per prescription equally by patient race and ethnicity. However, the intervention did not significantly close the baseline disparity in prescribing by race.

Molecular biomarker-defined brain tumors: Epidemiology, validity, and completeness in the United States.

BACKGROUND: Selected molecular biomarkers were incorporated into the US cancer registry reporting for patients with brain tumors beginning in 2018. We investigated the completeness and validity of these variables and described the epidemiology of molecularly defined brain tumor types. METHODS: Brain tumor patients with histopathologically confirmed diagnosis in 2018 were identified within the Central Brain Tumor Registry of the United States and NCI's Surveillance, Epidemiology, and End Results Incidence databases. The brain molecular markers (BMM) site-specific data item was assessed for coding completeness and validity. 1p/19q status, MGMT promoter methylation, WHO grade data items, and new ICD-O-3 codes were additionally evaluated. These data were used to profile the characteristics and age-adjusted incidence rates per 100 000 population of molecularly defined brain tumors with 95% confidence intervals (95% CI). RESULTS: BMM completeness across the applicable tumor types was 75%-92% and demonstrated favorable coding validity. IDH-wildtype glioblastomas' incidence rate was 1.74 (95% CI: 1.69-1.78), as compared to 0.14 for WHO grade 2 (95% CI: 0.12-0.15), 0.15 for grade 3 (95% CI: 0.14-0.16), and 0.07 for grade 4 (95% CI: 0.06-0.08) IDH-mutant astrocytomas. Irrespective of WHO grade, IDH mutation prevalence was highest in adolescent and young adult patients, and IDH-mutant astrocytomas were more frequently MGMT promoter methylated. Among pediatric-type tumors, the incidence rate was 0.06 for H3K27M-mutant diffuse midline gliomas (95% CI: 0.05-0.07), 0.03 for SHH-activated/TP53-wildtype medulloblastomas (95% CI: 0.02-0.03), and <0.01 for both C19MC-altered embryonal tumor with multilayered rosettes and RELA-fusion ependymomas. CONCLUSIONS: Our findings illustrate the success of developing a dedicated, integrated diagnosis variable, which provides critical molecular information about brain tumors related to accurate diagnosis.