Evaluation of Spending Differences Between Beneficiaries in Medicare Advantage and the Medicare Shared Savings Program.

Importance: The 2 primary efforts of Medicare to advance value-based care are Medicare Advantage (MA) and the fee-for-service-based Medicare Shared Savings Program (MSSP). It is unknown how spending differs between the 2 programs after accounting for differences in patient clinical risk. Objective: To examine how spending and utilization differ between MA and MSSP beneficiaries after accounting for differences in clinical risk using data from administrative claims and electronic health records. Design, Setting, and Participants: This retrospective economic evaluation used data from 15 763 propensity score-matched beneficiaries who were continuously enrolled in MA or MSSP from January 1, 2014, to December 31, 2018, with diabetes, congestive heart failure (CHF), chronic kidney disease (CKD), or hypertension. Participants received care at a large nonprofit academic health system in the southern United States that bears risk for Medicare beneficiaries through both the MA and MSSP programs. Differences in beneficiary risk were mitigated by propensity score matching using validated clinical criteria based on data from administrative claims and electronic health records. Data were analyzed from January 2019 to May 2022. Exposures: Enrollment in MA or attribution to an accountable care organization in the MSSP program. Main Outcomes and Measures: Per-beneficiary annual total spending and subcomponents, including inpatient hospital, outpatient hospital, skilled nursing facility, emergency department, primary care, and specialist spending. Results: The sample of 15 763 participants included 12 720 (81%) MA and 3043 (19%) MSSP beneficiaries. MA beneficiaries, compared with MSSP beneficiaries, were more likely to be older (median [IQR] age, 75.0 [69.9-81.8] years vs 73.1 [68.3-79.8] years), male (5515 [43%] vs 1119 [37%]), and White (9644 [76%] vs 2046 [69%]) and less likely to live in low-income zip codes (2338 [19%] vs 750 [25%]). The mean unadjusted per-member per-year spending difference between MSSP and MA disease-specific subcohorts was $2159 in diabetes, $4074 in CHF, $2560 in CKD, and $2330 in hypertension. After matching on clinical risk and demographic factors, MSSP spending was higher for patients with diabetes (mean per-member per-year spending difference in 2015: $2454; 95% CI, $1431-$3574), CHF ($3699; 95% CI, $1235-$6523), CKD ($2478; 95% CI, $1172-$3920), and hypertension ($2258; 95% CI, $1616-2,939). Higher MSSP spending among matched beneficiaries was consistent over time. In the matched cohort in 2018, MSSP total spending ranged from 23% (CHF) to 30% (CKD) higher than MA. Adjusting for differential trends in coding intensity did not affect these results. Higher outpatient hospital spending among MSSP beneficiaries contributed most to spending differences between MSSP and MA, representing 49% to 62% of spending differences across disease cohorts. Conclusions and Relevance: In this study, utilization and spending were consistently higher for MSSP than MA beneficiaries within the same health system even after adjusting for granular metrics of clinical risk. Nonclinical factors likely contribute to the large differences in MA vs MSSP spending, which may create challenges for health systems participating in MSSP relative to their participation in MA.

Assessment of Length and Readability of Informed Consent Documents for COVID-19 Vaccine Trials.

Importance: Informed consent is a fundamental element of research ethics. The COVID-19 vaccine trials are high profile trials that have enrolled more than 100 000 participants. Consent documents must be succinct and understandable to ensure informed voluntary participation. Objective: To assess how well informed consent documents of the COVID-19 vaccine trials achieve the ideal of being succinct and understandable, and to create a shorter, more readable document. Design, Setting, and Participants: This quality improvement study collected and analyzed the informed consent documents used in 4 COVID-19 vaccine phase III randomized clinical trials to quantitatively assess readability and length and, based on this analysis, created a measurably more accessible informed consent document. Analysis was conducted from October 2020 to January 2021. Main Outcomes and Measures: The main outcomes were number of words (measured as word count), time-to-read (measured at reading speeds of 175-300 words per minute), language complexity (measured using Flesch-Kincaid Grade Level assessment), and readability (measured using Flesch Reading Ease Score). Secondary outcomes included clarity of how the placebo group could access the vaccine if it is proven safe and effective. The study also examined the length and readability of an improved consent document. Results: The 4 informed consent documents were a mean (range) of 8333 (7821 to 9340) words long, with a mean (range) 35 (32.6 to 38.9) minutes to read at 240 words per minute. All documents exceeded grade 9 language complexity and scored lower than 60 in the formal reading ease metric, which constitutes difficult. Only 1 document specified that participants in the placebo group might receive vaccine. It was possible to write a document in fewer than 3000 words with a grade 7 to 8 reading level and a formal readability score that was not difficult. Conclusions and Relevance: These findings suggest that existing COVID-19 vaccine informed consent documents were too long, difficult to read, and exceeded grade 9 in language complexity. It was possible to create a shorter, more readable informed consent document for these trials.