JMIR Publications

ABSTRACT

Background:

Fall-risk screening is required in the Centers for Medicare & Medicaid Services Health Outcomes Survey but is inconsistently administered in primary care, partly because it often depends on a single annual wellness visit. Digital and telehealth care models may provide additional opportunities to capture fall risk outside that encounter.

Objective:

This study evaluated whether a virtual musculoskeletal integrated practice unit (V-IPU) could systematically screen for fall risk among Medicare Advantage (MA) beneficiaries and whether screening responses and care response differed by age, sex, and race.

Methods:

This retrospective cohort study included 513 MA beneficiaries enrolled in a V-IPU through a single MA plan. During intake, patients completed a four-item fall-risk questionnaire administered by a physical therapist. Bivariate associations between screening responses and age, sex, and race were assessed using chi-square or Fisher exact tests. Four multivariable logistic regression models, one per screening item, adjusted for baseline age, sex, race, body mass index, pain score, PHQ-2, GAD-2, three patient-reported outcome measure domains, and Patient Activation score. Screening items were not used as predictors of one another.

Results:

Of 513 patients, 197 (38.4%) reported a fall in the past 12 months and 204 (39.8%) reported balance or walking problems, whereas 175 (34.1%) had discussed fall risk with a physician and 119 (23.2%) reported provider preventive action. The gap between risk identification and documented care response was significant (McNemar P<.001). White race was independently associated with higher odds of reporting a fall than Black race (adjusted OR 2.81, 95% CI 1.64-4.82, P<.001), whereas older age (OR 0.967, 95% CI 0.937-0.999, P=.04) and male sex (OR 0.611, 95% CI 0.390-0.959, P=.03) were associated with lower odds. Better baseline physical health was associated with lower odds of reporting a fall (OR 0.93, 95% CI 0.89-0.98, P=.006) and was the only covariate independently associated with all four screening items. Race was not associated with provider fall-prevention action (OR 0.88, 95% CI 0.50-1.55, P=.66).

Conclusions:

A V-IPU systematically captured fall-risk screening data in an MA population and identified a significant gap between fall risk and documented care response, as well as an independent racial disparity in self-reported fall history. Digital care models may be able to use routinely collected screening data to trigger provider follow-up or targeted outreach; future work should evaluate whether such interventions improve fall-related outcomes.