Decline in dementia patients during the COVID-19 pandemic: A cohort study in Taiwan.

BACKGROUND: The cognitive consequences of COVID-19 in patients with dementia remain uncertain. This study examined whether COVID-19 infection was associated with cognitive decline. METHODS: From October 2020 to October 2022, COVID-19-naïve patients with dementia were recruited from a hospital-based memory clinic in Taipei. During follow-up, patients who developed COVID-19 were classified as the infected group, while others served as controls. For controls, baseline and follow-up were defined as two consecutive annual cognitive assessments. For infected patients, baseline and follow-up were the nearest assessments before and after COVID-19 infection, respectively. The mean assessment interval was approximately one year in both groups. Outcomes included the Mini-Mental State Examination (MMSE), Clinical Dementia Rating Sum of Boxes (CDR-SB), and Cognitive Abilities Screening Instrument (CASI). Linear mixed-effects models assessed whether cognitive trajectories differed by COVID-19 status. RESULTS: After excluding patients without eligible follow-up data, infected patients were matched with controls by age and sex. Overall, 44 patients with dementia who developed COVID-19 and 22 who remained uninfected were included in the study. Baseline demographic characteristics and cognitive scores were comparable. The mean age was 84.1 years, with 36% (24/66) male. At follow-up, cognitive performance did not differ significantly between infected and control groups. Within-group analyses showed significant declines in MMSE, CDR-SB, and CASI in the infected group, whereas controls declined significantly only in CASI. In adjusted mixed-effects models, the COVID-19 × time interaction was not significant (β = -0.427, p = 0.676), suggesting no significant difference in MMSE decline trajectory between groups. Higher education was associated with higher MMSE scores (β = 0.553, p = 0.002). Sensitivity analyses supported these findings. CONCLUSION: After accounting for age, sex, education, and assessment interval, COVID-19 infection was not associated with faster cognitive decline. Participants with more education generally had higher MMSE scores, but education did not significantly alter the rate of decline.