A Comparison of Home Health Utilization, Outcomes, and Cost Between Medicare Advantage and Traditional Medicare.

BACKGROUND: Home health use is rising rapidly in the United States as the population ages, the prevalence of chronic disease increases, and older Americans express their desire to age at home. Enrollment in Medicare Advantage (MA) plans rather than Traditional Medicare (TM) has grown as well, from 13% of total Medicare enrollment in 2004 to 39% in 2020. Despite these shifts, little is known about outcomes and costs following home health in MA as compared with TM. OBJECTIVE: The objective of this study was to measure the association of MA enrollment with outcomes and costs for patients using home health. DESIGN: This was a retrospective cohort study. PARTICIPANTS: Patients enrolled in plans offered by 1 large, national MA organization and patients enrolled in TM, with at least 1 home health visit between January 1, 2017, and June 30, 2018. EXPOSURE: MA enrollment. MAIN MEASURES: We compared the intensity of home health services and types of care delivered. The main outcome measures were hospitalization, the proportion of days in the home, and total allowed costs during the 180-day period following the first qualifying home health visit during the study period. KEY RESULTS: Among patients who used home health, our models demonstrated enrollment in MA was associated with 14%, and 6% decreased odds of 60- and 180-day hospitalization, respectively, a 12.8% and 14.7% decrease in medical costs exclusive and inclusive of home health costs, respectively, and a 0.27% increase in the proportion of days at home during the 180-day follow-up, equivalent to an additional half-day at home. There were few differences in home health care delivered for MA and TM [mean number of visits in the first episode of care (17.1 vs. 17.3) and mean visits per week (3.2 vs. 3.3)]. The mean number of visits by visit type and percent of patients with each type was similar between MA and TM as well. CONCLUSIONS: Compared with enrollment in TM, enrollment in MA was associated with improved patient-centered outcomes and lower cost and utilization, despite few differences in the way home health was delivered. These findings might be explained by structural components of MA that encourage better care management, but further investigation is needed to clarify the mechanisms by which MA enrollment may lead to higher value home health care.

Post-SNF outcomes and cost comparison: Medicare Advantage vs traditional Medicare.

OBJECTIVES: To compare outcomes and costs following skilled nursing facility (SNF) discharge for patients within a Medicare Advantage (MA) organization vs traditional Medicare (TM). STUDY DESIGN: Retrospective analysis of adults with a postacute SNF admission identified from MA claims (MA cohort: n = 56,228) and the Medicare 5% Limited Data Sets (TM cohort: n = 67,859). METHODS: Outcomes included hospitalization, proportion of days at home, and total medical costs during the 180 days post SNF discharge, and successful community discharge. Regression models accounted for patient characteristics and health care utilization in the 180 days prior to the proximal hospitalization and characteristics of the proximal hospitalization using backward variable selection and fixed effects for MA enrollment. To control for observable differences between individuals who selected MA vs TM, inverse probability of treatment weighting (IPTW) was conducted. RESULTS: The MA cohort was younger than the TM cohort (median age, 77 vs 81 years), more likely to have qualified for Medicare based on disability (29% vs 20%), and less likely to have dual Medicare/Medicaid eligibility (16% vs 23%). After adjustment, MA was associated with 22% decreased odds of hospitalization during the 180 days post SNF discharge, 19% increased odds of successful community discharge, a 4% increase in the proportion of days at home (equating to 6.7 additional days), and a 24% decrease in medical costs post SNF discharge. Results using IPTW were similar. CONCLUSIONS: MA was associated with better outcomes and lower costs post SNF discharge, suggesting efficiencies in care for SNF patients with MA. Further research is needed to evaluate specific MA features that may lead to better value.

Large-scale influenza vaccination promotion on a mobile app platform: A randomized controlled trial.

While health-care providers have used incentives in an attempt to motivate patients to obtain vaccinations, their effect on vaccination rates has not been systematically evaluated on a large scale. In this study, we examined whether mobile applications may improve population vaccination rates through enhanced communication and incentives education. Our study is the first randomized controlled trial assessing the effect of large-scale messaging combined with individualized incentives on influenza-vaccination rates. In this trial, we delivered messages regarding influenza vaccinations to 50,286 adults, aged 18 through 65, then compared the subsequent vaccination rate, the effectiveness of the message content and the timing. Multiple rounds of messaging occurred over a seven-week period during the 2016 flu season, after which vaccination rates were observed for one week. Participants were randomly assigned to one of three messaging approaches: conspicuous (highlighting the amount of rewards to be received for obtaining a flu shot); generic (promoting vaccinations with no mention of rewards); or no-message. Evidence of vaccination obtainment was indicated by medical and pharmacy claims, augmented by patients self-reporting through the mobile wellness app during the study period. Of the people assigned to receive messaging, 23.2% obtained influenza vaccination, compared to 22.0% of people who obtained vaccination in the no-messaging control arm. This difference was statistically significant (p < 0.01). The research revealed that messaging effectiveness decreased after each successive batch sent, suggesting that most participants responsive to messaging would become activated immediately after receiving one alert. Interestingly, in this large-scale study, there were no significant differences between conspicuous incentives and generic messaging, suggesting an important area for future research. Trial Registration: clinicaltrials.gov identifier: NCT02908893.

Impact of potential pregabalin or duloxetine drug-drug interactions on health care costs and utilization among Medicare members with fibromyalgia.

PURPOSE: To examine the impact of newly initiated pregabalin or duloxetine treatment on fibromyalgia (FM) patients' encounters with potential drug-drug interactions (DDIs), the health care cost and utilization consequences of those interactions, and the impact of treatment on opioid utilization. PATIENTS AND METHODS: Subjects included those with an FM diagnosis, a pregabalin or duloxetine prescription claim (index event), ≥1 inpatient or ≥2 outpatient medical claims, and ≥12 months preindex and ≥6 postindex enrollment. Propensity score matching was used to help balance the pregabalin and duloxetine cohorts on baseline demographics and comorbidities. Potential DDIs were defined based on Micromedex 2.0 software and were identified by prescription claims. RESULTS: No significant differences in baseline characteristics were found between matched pregabalin (n=794) and duloxetine cohorts (n=794). Potential DDI prevalence was significantly greater (P<0.0001) among duloxetine subjects (71.9%) than among pregabalin subjects (4.0%). There were no significant differences in all-cause health care utilization or costs between pregabalin subjects with and without a potential DDI. By contrast, duloxetine subjects with a potential DDI had higher mean all-cause costs ($9,373 versus $7,228; P<0.0001) and higher mean number of outpatient visits/member (16.0 versus 13.0; P=0.0009) in comparison to duloxetine subjects without a potential DDI. There was a trend toward a statistically significant difference between pregabalin and duloxetine subjects in their respective pre- versus post-differences in use of ≥1 long-acting opioids (1.6% and 3.4%, respectively; P=0.077). CONCLUSION: The significantly higher prevalence of potential DDIs and potential cost impact found in FM duloxetine subjects, relative to pregabalin subjects, underscore the importance of considering DDIs when selecting a treatment.