Dementia-associated mortality and its predictors among older adults in sub-Saharan Africa: results from a 2-year follow-up in Congo (the EPIDEMCA-FU study).

Background: between 2011 and 2012, we carried out a study of dementia prevalence in central Africa throughout the EPIDEMCA (Epidemiology of Dementia in Central Africa) programme.

Objective: to assess dementia-related mortality among Congolese older people from the EPIDEMCA study after 2 years of follow-up.

Design: longitudinal population-based cohort study.

Setting: Gamboma and Brazzaville, Republic of Congo.

Methods: older participants were traced and interviewed in rural and urban Congo annually between 2012 and 2014. DSM-IV and NINCDS-ADRDA criteria were required for dementia diagnosis. Data on vital status were collected throughout the follow-up. Cox proportional hazards model was used to assess the link between baseline dementia diagnosis and mortality risk.

Results: of 1,029 participants at baseline, 910 (88.4%) have a complete cognitive diagnosis. There were 791 participants (76.87%) with normal cognition, 56 (5.44%) with MCI and 63 (6.12%) with dementia. After 2 years of follow-up, 101 (9.8%) participants had died. Compared with participants with normal cognition, patients with dementia had 2.5 times higher mortality risk (HR = 2.53, 95% CI 1.42-4.49, P = 0.001). Among those with dementia, only clinical severity of dementia was associated with an additional increased mortality risk (HR = 1.91; CI 95%, 1.23-2.96; P = 0.004). Age (per 5-year increase), male sex and living in an urban area were independently associated with increased mortality risk across the full cohort.

Conclusion: among Congolese older adults, dementia is associated with increased mortality risk. Our results highlight the need for targeted health policies and strategies for dementia care in sub-Saharan Africa (SSA).