Health Services Research最新文献

Objective: To assess changes in publicly reported, potentially avoidable hospital visits after chemotherapy since the introduction of a Medicare quality measure.

Study setting and design: Retrospective analysis of avoidable emergency department (ED) and inpatient admission (ADM) rates after chemotherapy between 2018 and 2022, across absolute visit rates and relative hospital performance ("better than", "no different than", "worse than" the national rate). We stratified hospitals into quartiles of visit rates in 2018 and used this to model the change in visit rates from 2018 to 2022 with generalized linear regression.

Data sources and analytic sample: A longitudinal cohort of hospitals from the Medicare Outpatient Quality Reporting Program.

Principal findings: We analyzed 1179 hospitals (94.3% non-profit, 22.9% teaching). National avoidable ED visit rates were 6.0% in 2018, 5.4% in 2022; ADM rates were 12.5% in 2018, 10.3% in 2022. Nearly all hospitals were deemed to have performed "no different" than the national rate each year in ED (≥ 95.3%) and ADM (≥ 91.1%). In adjusted analyses, visit rates for hospitals in the lowest 2018 visit rate quartiles declined the least by 2022 (ED: -0.44% 95% CI: -0.58 to -2.94; ADM: -0.91%, 95% CI: -1.14 to -0.69), and declined the most for hospitals in the highest 2018 quartiles (ED: -1.72%, 95% CI: -1.85 to -7.73; ADM: -3.03%, 95% CI: -3.27 to -2.81). We estimated that the tendency for extreme baseline values to approach the average over time accounted for up to one-tenth of the decline among the worst-performing 2018 quartiles (ED: 10.6% of rate change, 95% CI: 9.8 to 11.5; ADM: 9.0%, 95% CI: 8.2 to 9.8).

Conclusion: Hospitals reduced their potentially avoidable hospital visit rates, though Medicare deemed that nearly all hospitals performed "no different" than the national average each year. It remains unclear if the reductions were driven by this quality measure.

Objective: Develop a causal machine learning (causal ML) framework for estimating how a diagnosis (cancer in this study) affects the likelihood of receiving a specific health care service (advance care planning in this study) and associated heterogeneity.

Study setting and design: Our proposed framework leverages the causal forest method, combined with a population-weighted resampling and averaging over estimations strategy, to estimate average treatment effects (ATEs) and conditional average treatment effects (CATEs). Post hoc, we used best linear projections to identify covariates associated with variation in the CATEs. We illustrate the framework by applying it to a stratified random sample of patients, where the strata are defined by the crosstabulation of cancer diagnosis (diagnosed vs. not diagnosed) and ACP receipt (documented vs. not documented).

Data sources and analytic sample: We extracted deidentified patient data from October 2019 to October 2024 (n = 87,772) with explanatory variables in three categories: demographics, morbidity, and health care system utilization.

Principal findings: In application of the causal ML framework, we found that patients diagnosed with cancer at this health care system to be at least 17.2% more likely to have documented ACP than similar patients not diagnosed with cancer. We also found significant heterogeneity. For instance, a one standard deviation increase in in-person outpatient visits was associated with an on-average increase in the CATE estimate (by 6.1 percentage points), while a one standard deviation increase in hospital admissions, inpatient days, and surgical duration in minutes was associated with an on-average decrease in the CATE estimate (by -1.3, -5.6, and -0.5 percentage points, respectively).

Conclusions: The proposed causal ML framework enables estimation of the effect of a diagnosis on receiving a relevant health care service. In the cancer diagnosis context, it can identify patient groups less likely to receive ACP, thus informing service allocation strategies.

Objective: To examine the association between COVID-19-related hardship and 1-year adjuvant endocrine therapy (AET) adherence among women with early-stage hormone-receptor-positive breast cancer.

Study setting and design: This post hoc analysis utilized data from the THRIVE trial, which tested a 6-month remote monitoring intervention on 1-year AET adherence, measured using an electronic pillbox. The 1-year follow-up survey included questions about pandemic-related hardship, including financial loss, changes/gaps in health insurance, and difficulty accessing basic needs. Participants reporting any of these were categorized as experiencing pandemic-related hardship. Logistic regressions estimated the association between patient characteristics and pandemic-related hardship, and between hardship and AET adherence (≥ 80% proportion of days covered), controlling for patient characteristics and randomization group.

Data sources and analytic sample: We included 217 women diagnosed with early-stage breast cancer prescribed AET at a large cancer center who enrolled in THRIVE between April 2019 and June 2021.

Principal findings: Overall, 39.6% of participants reported any pandemic-related hardship: 34.6% reported financial loss, 10.6% reported changes/gaps in insurance, and 11.1% reported difficulty accessing basic needs. In adjusted analyses, having an income ≤ 100% of federal poverty level or prior chemotherapy or radiation was associated with a 41.4 (95% CI: 9.8-73.0) and 13.8 (95% CI: 0.3-27.2) percentage-point higher likelihood, respectively, of having any pandemic-related hardship. Over half (52%) of participants were AET adherent. In adjusted analyses, 40.1% of those with any pandemic-related hardship were AET adherent, compared with 59.5% of those without hardship, a 19.3 percentage-point lower likelihood (95% CI: -33.0 to -5.7).

Conclusions: Pandemic-related hardship was more common among individuals with lower income or prior radiation or chemotherapy, and was associated with lower AET adherence, with possible impacts on cancer progression and survival. These findings highlight the need for routine financial screening and targeted support, particularly among lower-income patients on long-term AET.

Trial registration: NCT03592771.

Objective: To investigate whether the acquisition of physician practices by Optum, a subsidiary of United Health Group (UHG), influences patient volume and service prices, particularly, for patients enrolled in health insurance plans competing with UHG.

Study setting and design: We employed a novel database cataloging health insurer acquisitions of physician practices to identify those acquired by Optum-the nation's largest payvider (vertically integrated payer-provider)-from 2007 to 2023. These data were integrated with non-UHG commercial health insurance claims for practices acquired between 2015 and 2019. Using a stacked difference-in-differences design, we analyzed relative changes in prices and office visits across 12 Optum-acquired practices compared to a control group. Adjustments were made for physician profiles, practice characteristics, and calendar-year fixed effects to ensure robust estimates.

Principal findings: From 2007 to 2023, Optum acquired 44 physician practices, employing 7828 physicians by 2023. Postacquisition, we found no statistically significant average change in prices for most acquired practices relative to controls. However, the single largest acquisition was associated with a relative price increase of 4.5% (95% CI: [1.2%, 7.8%]; p = 0.02) for established patient visits. Preacquisition trends showed prices at acquired practices rising faster than controls. Additionally, Optum acquisitions were linked to suggestive declines in claim volume 1-1.5 years postacquisition, though this shift was predominantly driven by the largest acquired practice, indicating variability in outcomes across the sample.

Conclusions: Optum's acquisition of physician practices did not broadly result in significant price changes for evaluation and management services provided to patients with competing insurance plans, despite higher baseline prices at acquired practices. Suggestive reductions in patient volume emerged postacquisition, but effects were inconsistent. Extended follow-up research is warranted to evaluate whether these acquisitions reshape local healthcare market dynamics over time.