Dementia Prevalence in a Rural Region of South Africa: A Cross-Sectional Community Study

• Published in: Journal of Alzheimer's disease Vol. 60; no. 3; p. 1087
• Main Authors: de Jager, Celeste A, Msemburi, William, Pepper, Katy, Combrinck, Marc I
• Format: Journal Article
• Language: English
• Published: Netherlands 01.01.2017
• Subjects: Age Factors; Aged; Aged, 80 and over; Cross-Sectional Studies; Dementia - epidemiology; Depression - epidemiology; Educational Status; Female; Humans; Likelihood Functions; Male; Middle Aged; Poverty; Prevalence; Risk Factors; Rural Population; Sex Factors; South Africa - epidemiology; low- and middle-income country; epidemiology; older people; Community Screening Instrument for Dementia; dementia screening; population
• ISSN: 1875-8908
• DOI: 10.3233/JAD-170325

Dementia is a growing concern for low- and middle-income countries where longevity is increasing and service provision is poor. Global prevalence estimates vary from 2% to 8.5% for those aged 60 years and older. There have been few dementia studies in sub-Saharan Africa, and prevalence data are lacking for South Africa. To conduct a large dementia prevalence study in a low income rural population in South Africa. 1,394 Xhosa-speaking community dwellers, aged ≥60 y (mean age±sd 71.3±8.3 y), in three clinic catchment areas, were screened at home. Trained community health workers administered the brief Community Screening Instrument for Dementia (CSID) to participants and informants to assess cognitive and functional capacity. Depressive symptoms were assessed with three questions from the EURO-D. The prevalence estimate using published CSID sensitivity/specificity values was 0.8 (95% CI: 0.06-0.09). Using CSID cut-off scores the estimated prevalence was 0.12 (95% CI: 0.10-0.13), with 161 screen-positives. Both methods gave a rate of 0.11 (95% CI: 0.09-0.13) for those over 65 years (n = 1051). 68.6% of participants were female and 69.8% had less than 7 years of education. Dementia risk was associated with older age and symptoms of depression, but not with sex. The association with education was not significant when controlled for by age. Dementia prevalence estimates were higher than expected for this low-income rural community. There is a need for increased dementia awareness and feasible support interventions. We also need further studies of regional prevalences, dementia subtypes, and modifiable risk factors in South Africa.