Polypharmacy and Mild Cognitive Impairment in Older Adults: A 3-year Study of DO-HEALTH

To examine the association between polypharmacy and mild cognitive impairment (MCI) at baseline and over 3 years in community-dwelling older adults.

Observational analysis of the DO-HEALTH trial.

Community-dwelling adults aged ≥70 years with good cognitive function and without major diseases at baseline.

Main exposure was polypharmacy at baseline (≥5 medications). The outcome was MCI (Montreal Cognitive Assessment [MoCA] score <26), assessed at baseline and years 1, 2, and 3. Logistic regression and generalized estimating equations (GEEs) for repeated binary outcomes were used for the cross-sectional and longitudinal analysis, respectively. Minimally adjusted models included age, sex, prior fall, study site, body mass index, and education. Fully adjusted models additionally adjusted for mental health and multimorbidity. GEE models also accounted for time and DO-HEALTH treatment effects. Sensitivity analysis used a stricter MCI definition (MoCA < 24).

A total of 2153 participants completed MoCA at baseline. Baseline MCI prevalence was higher in participants with polypharmacy compared with those with no polypharmacy (MCI < 26: 53.5% vs 46.5%; MCI < 24: 34.9% vs 17.2%). Polypharmacy was associated with greater MCI odds at baseline in the minimally adjusted models (MoCA < 26: odds ratio [OR], 1.32; 95% CI, 1.05-1.66; MoCA < 24: OR, 1.37; 95% CI, 1.06-1.79). Over 3 years, polypharmacy was associated with increased MCI odds in the minimally adjusted models (MoCA < 26: OR, 1.28; 95% CI, 1.08-1.52; MoCA < 24: OR, 1.33; 95% CI, 1.07-1.65). Notably, these associations were somewhat attenuated and nonsignificant after controlling for mental health and multimorbidity.

Our findings suggest that MCI is more prevalent in older adults with polypharmacy compared with those without polypharmacy. Although polypharmacy was associated with MCI at baseline and over 3 years, these associations were attenuated by mental health and multimorbidity, suggesting that mental health and multimorbidity contribute to both polypharmacy and MCI.