Mortality in Incident Cognitive Impairment: Results of the Prospective AgeCoDe Study

Objectives: To investigate mortality risk and survival time in new-incident cases of cognitive impairment (CI) in old age.

Design: Prospective cohort study in six German cities.

Setting: German Study on Ageing, Cognition, and Dementia in Primary Care Patients (AgeCoDe).

Participants: Two thousand eighty-nine nondemented GP patients aged 75+.

Measurements: Every 18 months, trained psychologists and physicians conducted structured clinical interviews at the participants' homes. Dates of death were obtained from relatives, general practitioner (GP), or the local registry offices. We used the Kaplan-Meier survival method to estimate survival times of individuals with and without incident CI and multivariable Cox proportional hazards regressions to assess the association between CI and mortality risk, controlled for covariates.

Results: Out of the 2,089 included patients at follow-up I, 859 (41.1%) died during the subsequent mean observation period of 8.0 years. Patients with incident CI at follow-up I showed a significantly higher case-fatality rate per 1,000 person-years (74.2, 95% CI = 64.2-84.2 vs 47.8, 95% CI = 44.6-51.0) and a significantly shorter mean survival time in the observation period than those without (7.8 vs 9.1 years; P < .001). The association between incident CI and mortality remained significant in the multivariable Cox analyses-incident CI was associated with a 42% increased, incident severe CI with a 75% increased mortality risk.

Conclusion: Our findings suggest an elevated mortality risk in newly acquired cognitive deficits in old age. Even though further studies are required to analyze potential underlying mechanisms, our findings support the notion that such cognitive deficits should be taken seriously in clinical practice not only for an increased risk of developing dementia but also for a broader range of possible adverse health outcomes.