Cross-sectional study of cognitive impairment and visual impairment among the elderly population in residential care in India: the Hyderabad Ocular Morbidity in Elderly Study (HOMES)

Objective To report the relationship between visual impairment (VI) and cognitive impairment (CI) among the older population living in residential care homes in Hyderabad, India. Study design Cross-sectional study. Setting 41 homes for the aged centres in the Hyderabad region. Participants 965 participants aged >= 60 years from homes for the aged centres. Primary outcome measures Visual impairment and cognitive impairment. Methods The Hindi mini-Mental Status Examination (HMSE) questionnaire was used to assess the cognitive function. The final HMSE score was calculated after excluding vision-dependent tasks (HMSE-VI). A detailed eye examination was conducted, including visual acuity (VA) measurement for distance and near vision, using a standard logarithm of the minimum angle of resolution chart under good illumination. CI was defined as having a HMSE-VI score of <= 17. VI was defined as presenting VA worse than 6/12 in the better-seeing eye. Near VI (NVI) was defined as binocular presenting near vision worse than N8 and distance VA of 6/18 or better in the better-seeing eye. Multiple logistic regression was done to assess the association between VI and CI. Results The mean age (+/- SD) was 74.3 (+/- 8.3) years (range: 60-97 years). There were 612 (63.4%) women, and 593 (61.5%) had a school education. In total, 260 (26.9%; 95% confidence intervals: 24.2 to 29.9) participants had CI. The prevalence of CI among those with VI was 40.5% compared with 14.6% among those without VI (p<0.01). The logistic regression analysis showed that the participants with VI for distance vision had three times higher odds of having CI (OR 3.09; 95% confidence intervals: 2.13 to 4.47; p<0.01). Similarly, participants with NVI had two times higher odds of having CI (OR 2.11; 95% confidence intervals: 1.36 to 3.29; p<0.01) after adjusting for other covariates. Conclusions CI was highly prevalent among those with distance and near VI. VI was independently and positively associated with CI after adjusting for potential confounders. Interventions can be planned to address VI in this vulnerable population which could have a ripple effect in preventing cognitive decline.