Cognitive impairment and mortality in older healthy Mexican subjects: a population-based 10-year follow-up study.

To estimate the incidence of cognitive impairment (CI) among cognitively healthy, Mexican subjects, and to evaluate the impact of demographic and vascular factors on the conversion to CI and mortality. 734 eligible subjects (aged 55 to >90 years) from a population-based sample were examined. The cognitive function of participants was assessed using the Mini-Mental State Examination (MMSE) every 2 years. The subjects were followed for an average of 3.2 years. The CI was defined using two sets of criteria: (i) moderate CI, as a drop to 25-21 on the MMSE at 2-year follow-up or a decrease of at least four points and (ii) severe CI, defined as a drop of 21 or less in MMES at follow-up. The incidence density and period prevalence were determined as epidemiological measures as well as the cumulative incidence as a risk measure. Kaplan-Meier survival curves were used to analyse the main points of interest: CI, dementia and mortality. The period prevalence of moderate CI was 20%, and 10% for severe CI. During 1959 person-years of follow-up, severe CI developed in 33 of the 361 participants. While during 2096 person-years of follow-up; 80 of 361 participants developed moderate CI. The rate of progression to severe CI in moderate CI subjects gradually increases with follow-up. Both, moderate and severe CI were associated with low educational level, higher age and higher mortality. Elderly people with moderate CI have an increased risk of severe CI. Moderate and severe CI are both predictive of higher mortality in Mexican subjects.