Dementia in sub-Saharan Africa: increasing evidence highlighting the current and future burden.

Dr Maëlenn Guerchet, King’s College London (Institute of Psychiatry, Psychology and Neuroscience, UK)

The worldwide dementia epidemic is now well recognised, and the burden brought by dementia symptoms is considerable for the people affected, their relatives and societies. At the same time, the African population is ageing at an unprecedented rate. In 2015, over 58% of all people with dementia were living in low and middle countries (also referred as developing countries), of whom 4.03 million were living in Africa ( The 2015 World Alzheimer Report highlighted increasing evidence of dementia in sub-Saharan Africa. Unfortunately, despite the magnitude of the situation, awareness about dementia remains low within the population and older people are rarely the target of specific health policies. This article will present a short review of the evidence for dementia burden in sub-Saharan Africa through epidemiological indicators (prevalence, incidence, mortality, and trends) as well as cost estimates.

The investigation of the dementia prevalence (ie the proportion of people with the condition in a given population at a certain time) in sub-Saharan Africa started in the early 1990s, with the comparative epidemiological Ibadan-Indianapolis Dementia Project in Nigeria. At that time, a few community-based studies of neurological disorders in Nigeria and Ethiopia were suggesting that Alzheimer’s disease was uncommon. Two decades later, the evidence on the prevalence of dementia in this region of the world has greatly improved, especially over the last 10 years. A number of studies have been carried out in different settings (healthcare centres, hospitals, or in the community), but in terms of population-based studies, sub-Saharan Africa remains the tropical region with the worst coverage. While the 2009 World Alzheimer Report only had population based studies from the West Sub-Saharan Africa , the 2015 update of the systematic review found studies from two additional regions: Central and East Sub-Saharan Africa, while Southern sub-Saharan Africa remains understudied.

In total, nine studies from Nigeria, Benin, Central African Republic, the Republic of Congo and Tanzania were identified. The majority were conducted after 2000 and had a two-stage design, with a number of participants included between 500 and 1499 older people. The DSM-IV/III-R remained the preferred diagnostic criteria in this region. One additional study from Nigeria has been published since (Ogunniyi et al., 2016). Overall, the initial and more recent studies from Nigeria as well as the studies in neighbouring Benin report a low prevalence of dementia (from 2.3% in Ibadan to 3.7% in Cotonou) and Alzheimer’s disease (1.4% to 2.8%) in West Africa. The difference in findings from two communities of similar ethnic origins in Ibadan and Indianapolis (lower prevalence in Nigeria compared to the USA) contributed to the evidence highlighting potential interactions between genes and environment. While more ‘traditional’ risk factors were also found to be associated with dementia in those studies (age, marital status, female sex, malnutrition, depression), the role of the most known genetic factor for AD, the APOE Ɛ4 allele, was not significant at first. Looking at other sub-Saharan African regions, the newest evidence from Central and East Africa shows much higher age-standardised prevalence of dementia from 4.5% in rural Congo to 6.4% in rural Tanzania and 7.2% in rural Central African Republic.

After collating this evidence for the World Alzheimer Report 2015 (all but one studies cited above), the standardised prevalence of dementia for sub-Saharan Africa was estimated to be 5.5%, which works out as an estimated 1.63 million people over the age of 60 living with dementia. The expected proportionate increase is quite similar to that of Latin America,which rose from 145% to 300% between 2015 and 2050.

While dementia prevalence studies have become more ‘common’ in Sub-Sahara Africa, incidence studies (which estimate the rate at which new cases occur in a population) can almost still be counted on the fingers of one hand. Again, Nigeria has been leading the way by providing the first estimates and also providing most of the evidence. Hence the 2015 World Alzheimer Report included evidence from Nigeria, the only country with data available at that time. In the Indianapolis-Ibadan Dementia Project, the age-standardised annual incidence of dementia was significantly lower among Yoruba than among African Americans (Yoruba, 1.3%; African Americans, 3.24%). The same pattern was shown for Alzheimer’s disease (Yoruba, 1.1% vs. African Americans, 2.5%). More recently, after a 3-year follow-up, the estimated incidence of dementia was 21.8 per 1,000 person-years in the Ibadan Study of Aging. This evidence from Nigeria led to the estimation of a total annual number of 446,569 new cases of dementia in Sub-Sahara Africa (World Alzheimer Report 2015). Since then, new evidence has been published from Nigeria after 5 years of follow-up in the Ibadan Study of Aging confirming the previous estimates with 20.9 per 1000 person-years (Ojagbemi et al., 2016) and from Central Africa with the EPIDEMCA-FU (Epidemiology of Dementia in Central Africa – Follow-Up) study, reporting a standardised incidence of 13.5 per 1000 person-years after a 2-year follow-up in rural and urban Congo (Samba et al., 2016). Overall, the incidence estimated in sub-Saharan African countries is comparable to the incidence of dementia reported for low and middle income countries, just slightly lower than the ones in high-income countries (World Alzheimer Report 2015).

Nigeria was again the first country to provide estimates of mortality among people with dementia. In the Indianapolis Ibadan Dementia Project, dementia was significantly associated with increased mortality in Ibadan with a relative risk of 2.8 (Perkins et al., 2002), which is higher than the hazard ratio reported later in the Ibadan Study of Aging (adjusted HR = 1.5) (Ojagbemi et al., 2016). The risk reported among older people living in the Republic of Congo was similar, more than 2.5 times higher among those with dementia (Samba et al., 2016), while a larger relative risk has been recorded in Tanzania after a 4-year follow-up (adjusted HR = 6.3) (Paddick et al., 2015). This higher mortality risk in people with dementia in sub-Saharan Africa is consistent with what has been observed in other low and middle income countries of Asia and Latin America in the 10/66 Dementia Research Group studies, where mortality hazards were 1.6 to 5.7 times higher in individuals with dementia at baseline (Prince et al., 2012). Those results support the fact that dementia was found to be a leading contributor to mortality in older population of countries with low or middle income.

For now, all the projections for the number of people living with dementia assume that the age- and sex- specific prevalence of dementia will remain constant (ie not variable over time) and that the increases are only driven by population ageing. This scenario is highly improbable and secular trends in dementia prevalence are very likely. The prevalence of dementia could be affected by a change in either its incidence or its average duration, or in both. Over time, societal changes occur, potentially affecting population health. During successive generations, risk (eg vascular diseases) and protective (eg education) factors have changed considerably. As a result of changes in life expectancy and risk profiles, secular trends may vary among world regions, and among different population subgroups within one country. Therefore, there has been an increased interest recently over a possible decline of age-specific dementia prevalence or incidence in high income countries. A global review of trends in dementia prevalence, incidence and mortality was conducted, in which studies applying constant methods to defined populations were included. Among the nine studies tracking dementia prevalence, seven tracked dementia incidence, four tracked mortality among people with dementia and all but one were conducted in high income countries. The evidence was considered too inconsistent to reach sound and generalizable conclusions regarding the existence of trends in dementia prevalence, incidence and mortality globally. It is even more true for low and middle income countries where the evidence on trends is almost non-existent for now. A single study assessing dementia incidence over time in Nigeria found no significant difference in dementia or Alzheimer’s disease incidence between the 1992 and 2001 Yoruba cohorts (Gao et al., 2016).

In the 2015 World Alzheimer Report, the costs of dementia globally and for each region have been estimated. The total costs for sub-Saharan Africa was US$ 1.4 billion, representing about 0.2% of the global cost of dementia. In sub-Saharan Africa, costs were mainly due to informal care (57.3% to 68.9%) rather than direct social (8.1% to 14.1%) or medical care (16.4% to 28.5%) costs. The African region presented one of the greatest relative increase in costs compared to the previous estimates in 2010, mainly driven by the upward revision of the prevalence in the area, and also the greater contribution of informal care to the costs. While these estimates are without doubt informative and useful, we need to keep in mind that they rely on the very scarce evidence available at the moment regarding the distribution of care, the organisation of care and the cost of care within those populations. The estimated costs are likely to increase over the years as the numbers of people affected increase but also as the evidence in the area is increasing for sub-Saharan African countries.

In the context of an on-going epidemiologic and demographic transition on the African continent, now more than ever, there is a need for new studies to estimate dementia prevalence, incidence, mortality and to monitor their changes in over time. These studies are necessary to underline to stakeholders, governments, local and international institutions the need to target health policies for older people and the development of strategies for dementia care in sub-Saharan Africa.

Most African countries are not ready to face such a heavy burden as the health systems are often weak and focused on caring for child and maternal health or common communicable and infectious diseases rather than non-communicable diseases and older populations. Traditional values and extended families, which were in the past part of a great support system, are fading somewhat with rapid urbanisation, and economic and industrial development. The lack of awareness of dementia is striking in populations and communities, but even extends to health workers who have neither education nor training about dementia. Representations and beliefs of dementia are often behind reports of witchcraft accusations, abuse and violence against older people in sub-Saharan Africa (Ndamba-Bandouzi et al., 2014; Kehoua et al., 2016). Our understanding of those aspects is also essential for the development of new studies but most importantly health policies and training programmes in this region in the future.