Background
Polypharmacy (≥5 non-ART medications) is associated with falls. People with HIV (PWH) have higher polypharmacy prevalence due to comorbidity burden, but data is still lacking on its association to falls, particularly compared in non-PWH. We retrospectively examined whether high polypharmacy was associated with fall risk in PWH and non-PWH using a novel measure for both polypharmacy and regimen complexity known as the medication regimen complexity index (MRCI).
Methods
We retrospectively abstracted data from randomly selected PWH from the Boston Medical Center (BMC) HIV clinic. Inclusion criteria were age ≥40, ≥2 visits in ID clinic post-3/1/22, and HIV diagnosis for ≥1 year. Non-PWH participants from BMC Geriatrics aged ≥40 and with ≥2 geriatric visits post-3/1/22 were also randomly selected as a comparator. MRCI and polypharmacy were calculated using University of Colorado’s MRCI tool. Past year’s falls were manually abstracted. We stepwise and forward selected multivariable logistic regression models predicting fall occurrence by MRCI and polypharmacy. Chi-square tests assessed how MRCI and polypharmacy relate to fall category (0/1/>1 fall).
Results
Of the combined PWH and non-PWH cohort (N=600), 268 (44.7%) were male, 476 (81.2%) were minorities (non-White and/or Hispanic), 58 (10.1%) had recent substance use, and 215 (35.8%) were prescribed psychotropics (Table 1). Non-PWH (80.5) tended to be older than PWH (59.0). Overall MRCI was 35.8 indicating high regimen complexity (>15). 556 (92.7%) had ≥5 medications, 151 (25.2%) had fall occurrence, and 66 (11.0% had >1 fall. MRCI was significantly associated with fall occurrence in PWH (OR=1.03, 95% CI 1.01-1.05) and worse fall category in PWH and combined cohorts (p<0.05). Polypharmacy by medication count was not associated with fall occurrence. Prescribed psychotropic (OR=2.34, 95% CI 1.03-5.32) and recent substance use (OR=2.92, 95% CI 1.24-6.89) were also strongly associated with falls in PWH.
Conclusions
PWH with high MRCI have higher risk for fall occurrence and multiple falls, highlighting need for fall monitoring and polypharmacy interventions. MRCI was significantly associated with falls whereas polypharmacy by med count was not, possibly due to its high prevalence in both cohorts, implying MRCI may be a more sensitive screener. Overall, these data indicate the need for geriatric care approaches to screen for and manage polypharmacy and fall risk.
Table or Figure