Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

An Estimate of the Incidence and Prevalence of Stroke in Africa: A Systematic Review and Meta-Analysis

An Estimate of the Incidence and Prevalence of Stroke in Africa: A Systematic Review and Meta-Analysis

  • Davies Adeloye
PLOS
x

Abstract

Background

Stroke is increasingly becoming a challenging public health issue in Africa, and the non-availability of data has limited research output and consequently the response to this burden. This study aimed to estimate the incidence and prevalence of stroke in Africa in 2009 towards improved policy response and management of the disease in the region.

Methods

A systematic search of Medline, EMBASE and Global Health for original population-based or hospital-based studies on stroke was conducted. A random effect meta-analysis was conducted on crude stroke incidence and prevalence rates, and a meta-regression-like epidemiological model was applied on all data points. The fitted curve generated from the model was used to estimate incident cases of stroke and number of stroke survivors in Africa at midpoints of the United Nation population 5-year age groups for the year 2009.

Results

The literature search yielded a total of 1227 studies. 19 studies from 10 African countries were selected. 483 thousand new stroke cases among people aged 15 years or more were estimated in Africa in 2009, equivalent to 81.2 (13.2–94.9)/100,000 person years. A total of 1.89 million stroke survivors among people aged 15 years or more were estimated in Africa in 2009, with a prevalence of 317.3 (314.0–748.2)/100000 population. Comparable figures for the year 2013 based on the same rates would amount to 535 thousand (87.0–625.3) new stroke cases and 2.09 million (2.06–4.93) stroke survivors, suggesting an increase of 10.8% and 9.6% of incident stroke cases and stroke survivors respectively, attributable to population growth and ageing between 2009 and 2013.

Conclusion

The findings of this review suggest the burden of stroke in Africa is high and still increasing. There is need for more research on stroke and other vascular risk factors towards instituting appropriate policy, and effective preventive and management measures.

Introduction

The burden of stroke is increasing in many low- and middle-income countries (LMIC) [1], and due to high fatality rates and overwhelming resource incurred by the health systems, stroke and many non-communicable diseases (NCDs) are now targeted public health priorities in these regions [2], [3]. Globally, about 16 million new cases of stroke and 62 million stroke survivors were estimated in 2005, with deaths from stroke accounting for 9.7% of all global deaths, and this is expected to increase to over 23 million new stroke cases and 7.8 million stroke deaths by 2030 in the absence of significant global public health response [4], [5].

It has been estimated that LMIC account for over 87% disability adjusted life years (DALYs) from stroke, which is about seven times the DALYs lost in high-income countries (HIC) [6]. Africa is particularly worst hit, owing to population growth, unchecked industrialization and increased consumption of western diets, leading to a rise in many modifiable vascular disease risk factors including smoking, harmful use of alcohol, physical inactivity and unhealthy diets, and invariably resulting in increased prevalence of hypertension, diabetes and obesity [7], [8]. In 2000, two African countries, although recorded low stroke prevalences, had remarkably high stroke incidence rates [9]. According to GBD 2002 estimates, three African countries (Angola, Liberia and Sierra Leone) recorded the highest stroke mortalities and DALYs worldwide [5], [10]. Between 2002 and 2004, estimates further revealed an increasing prevalence with 8% of new stroke cases and 5% of stroke survivors occurring in Africa [5], [11]. Even with this increasing burden, the public health response, accesses to health services and treatment options in many African countries have been poor [7], [12]. Specifically, the lack of functional stroke units, neurologists, health workers, cranial computed tomography (CT) scans, magnetic resonance imaging (MRI) machines and echo-doppler machines, among many others, has negatively affected stroke outcomes [2], [12]. Moreover, the high cost of medical care in a relatively low-income African society could have resulted in high stroke fatalities, as some studies have indicated that stroke prevalence and deaths in Africa increased due to an overtly poor socioeconomic status [6]. For example, a recent study revealed the incidence of stroke in HIC decreased by over 40% between 1970 and 2008, but with actual number of stroke cases increasing due to ageing of the population [13], while in Africa and many LMIC, stroke incidence rose by over 100% over the same period [13]. Furthermore, due to the high proportion of undiagnosed hypertension in Africa especially among the younger population [14], stroke incidence has also been reported to be more severe and higher among the active and productive population age groups [15].

Meanwhile, the World Health Organization (WHO) technically supported her member countries with methods for improved data collation and registration of hospital stroke cases [16]. Notwithstanding, another set-back in the response to the management of stroke in Africa is the lack of data and low research output [7], [14]. Stroke case ascertainment and survey methodologies have not, in most cases, complied with international protocols [7]. Published research studies are characterized by poorly organized community-based studies, difficulties in making retrospective diagnosis, and overlapping cases of first and recurrent strokes [7], [17]. The few studies on stroke, therefore, could have been marked by under-estimation of the stroke burden in Africa. In view of this high burden of stroke, its public health importance, and the relatively low research output in Africa, this study aimed to estimate the incidence and prevalence rates of stroke in Africa in order to attempt to quantify the burden and inform decision regarding policy responses and health system interventions across many countries in the region.

Methods

Search strategy and selection criteria

After identification of relevant Medical Subject Headings (MESH) and keywords, a final search strategy was developed. Searches were conducted in three main databases: Medline, EMBASE and Global Health. The search date was set from January 1970 to December 2013. An additional search was conducted on Google Scholar and reference lists of relevant studies to identify publications that could have been omitted in the database searches. The search terms employed on Medline are shown in Table 1, while those employed on other databases are shown in Table S1 and Table S2 in File S1.

Studies included for further screening were mainly population/community- and hospital-based studies on stroke in Africa, conducted on or after 1970 and providing numerical estimates on the incidence and/or prevalence of stroke in the region. African countries were as defined by the World Bank list of economies (October 2013) [18]. Studies conducted before 1970, without numerical estimates, on non-human subjects, and that were mainly reviews were excluded. Studies with well-defined stroke diagnostic criteria and survey protocols were further retained. Due to the paucity of data, varying sources of information including demographic health surveys, community-based door-to-door surveys, hospital records and outpatient clinics were allowed. However, the final stroke case ascertainment complied with the standard WHO definition, defined as “rapidly developing clinical signs of focal (or global) disturbance of cerebral function lasting longer than 24 hour, unless interrupted by death, with no apparent cause other than that of vascular origin” [19], [20]. According to experts, new cases of stroke were defined as number of people presenting with first ever stroke in a given period, while stroke survivors were the total number of people who have had stroke or living with its sequelae at a given time [20], [21]

Data extraction and statistical analysis

An independent parallel search and double extraction was conducted and all extracted data was stored in a Microsoft Excel file format. Data were abstracted systematically on study location, study period, mean age or age range, person years or sample size, incident cases of stroke or number of stroke survivors, and their respective age- and sex-specific incidence or prevalence rates. These were sorted into population-based or hospital-based data separately for analysis. For studies conducted on the same study site, population or cohort, the first chronologically published study was selected, and all additional data from other studies were compared for consistency and included in the selected paper.

From reported overall crude incidence or prevalence of stroke in a given cohort, a random effect meta-analysis was conducted with pooled effect of stroke expressed per 100,000 person years or population respectively. The overall data estimates of age- and sex-specific prevalence and incidence from all studies were used in our modelling (see Table S3 in File S1). A meta-regression-like epidemiological model in the form of a bubble graph was applied (done separately for males and females), adjusted for mean ages and the crude prevalence and incidence rates of stroke from all studies, with the size of the bubble corresponding to the given sample size. The fitted curve explaining the largest proportion of variance (best fit) was applied. From all data points, the median year of study was estimated, and the equations generated from the modelled curves were then separately used to estimate the new cases of stroke and number of stroke survivors at midpoints of the United Nation (UN) population 5-year age groups for the estimated median year. Africa populations were determined from the 2012 United Nations population demographics [22]. All statistical analyses were conducted on Stata 13.1 (Copyright 1985-2013 StataCorp LP).

Results

Systematic review

The literature search returned 1227 publications from Medline (286), EMBASE (731) and Global Health (210). A further 5 studies were included from other sources (Google Scholar and reference lists of relevant publications). 927 studies remained after removing duplicates. On screening titles for relevance (stroke studies conducted primarily on African populations), 839 studies were excluded, giving a total of 88 full texts that were assessed. After applying the quality criteria, 69 studies were further excluded (32 articles did not provide numerical estimates on incidence and/or prevalence of stroke, and 37 articles did not clarify study designs and survey methodologies). A total of 19 studies were finally retained for the review (Figure 1).

Study characteristics

The retained 19 studies [23][41] were conducted across the main regions of Africa (east, north, west and south), but with Northern Africa having the highest output (7 studies). 10 African countries were represented; Egypt and Nigeria ranked highest with 4 studies each, Libya, South Africa and Tanzania had two studies each, while Benin, Ethiopia, Mozambique, Tunisia and Zimbabwe had one study each. Most studies (84.2%) were completed within one year period and the median year of study from all data points was 2009. About 57.9% of studies were conducted in predominantly urban settings. The total sample size from all retained studies was over 6.3 million, with a mean and median sample size of 332,276.5 and 60,820 respectively. 14 studies were population-based, of which 8 were community-based door-to-door surveys and 2 studies each were based on demographic health surveys, population/community-based stroke registries and cross-sectional population-based surveys. There were 5 hospital-based studies with only one from a hospital-based stroke registry. Studies comply with the WHO case ascertainment or a modified definition, while some studies employed cranial computed tomography (CT) or magnetic resonance imaging (MRI) to confirm diagnosis (Table 2). Most studies were conducted on the entire study population with an overall mean age of 55.9 years. Across retained studies, age determination of subjects were determined from documented age-verification records, and in the absence of such, historical landmarks were employed.

Pooled estimates of reported crude stroke incidence and prevalence rates

Across studies reporting crude incidences of stroke, there were 6 population/community-based and 5 hospital-based studies. Population-based incidence rates were generally higher ranging from 25.2/100,000 person years (py) and 26.0/100,000 py in Lagos and Ibadan Nigeria in 2007 and 1979 respectively [38], [40], to 250/100,000 py in Al-Kharga Egypt in 2007 [26]. The hospital-based studies reported lower incidence rates ranging from 30/100,000 py in Harare Zimbabwe in 1991 [36], to 148.7/100,000 py in Maputo Mozambique in 2006 [33] (Table 3). The random effect meta-analysis of population-based incidence rates was 112.94/100,000 py (95% CI = 90.7–135.17, I2 = 97.5%, p = 0.000) (Figure 2). The hospital-based meta-analysis was lower with a pooled estimate of 77.39/100,000 py (95% CI = 51.31–103.48, I2 = 99.1%, p = 0.000) (Figure 3).There were 11studies (all population/community-based) reporting crude prevalences of stroke survivors with prevalence rates ranging from 15/100,000 population in Ethiopia in 1988 [23], to 963/100,000 population in 2010 [28] (Table 4). Random effect meta-analysis yielded a pooled prevalence rate of 387.93/100,000 population (95% CI = 284.16–491.70, I2 = 98.8%, p = 0.000) (Figure 4). A Tanzanian study reported a prevalence of 2300/100,000 among people aged 70 years above in Hai district in 2010 [24], this was not included in the meta-analysis as other studies were mostly based on the general population with mean age ranging between 50 and 60 years (Table 4).

thumbnail
Figure 2. Pooled crude incidence rates of stroke from population-based studies.

https://doi.org/10.1371/journal.pone.0100724.g002

thumbnail
Figure 3. Pooled crude incidence rates of stroke from hospital-based studies.

https://doi.org/10.1371/journal.pone.0100724.g003

thumbnail
Figure 4. Pooled crude prevalence rates of stroke survivors from population-based studies.

https://doi.org/10.1371/journal.pone.0100724.g004

thumbnail
Table 3. Summary of data from studies reporting crude incidence of stroke.

https://doi.org/10.1371/journal.pone.0100724.t003

thumbnail
Table 4. Summary of data from studies reporting crude prevalence of stroke survivors (all population/community-based).

https://doi.org/10.1371/journal.pone.0100724.t004

Modelled estimates of stroke incidence and prevalence rates in Africa

Based on the UN population demographics and bubble graphs derived from all data points, incident cases and number of stroke survivors were estimated for the year 2009, which was our estimated median year of study.

There were over 483 thousand new cases of stroke in Africa in 2009 among people aged 15 years or more equivalent to 81.2 (13.2–94.9)/100,000 py, with about 305 thousand and over 178 thousand new cases of stroke equivalent to 103.3 (20.7–109.2)/100,000 py and 59.5 (6.9–84.3)/100,000 py among men and women, respectively (Table 5 and Figure 5). Comparable figures for the year 2010 and 2013 based on the same incidence rates would amount to 496 (80.6–579.7) and 535 (87.0–625.3) thousand new stroke cases respectively, suggesting an increase of 10.8% between 2009 and 2013 that is attributable to growth and ageing of the African population alone.

thumbnail
Figure 5. Bubble graph showing relationship between age and crude incidence of stroke, with size of bubble corresponding to sample size.

https://doi.org/10.1371/journal.pone.0100724.g005

thumbnail
Table 5. Estimated new stroke cases and incidence rates in Africa in 2009 (estimates derived from bubble graph model and United Nations population figures).

https://doi.org/10.1371/journal.pone.0100724.t005

The estimated number of stroke survivors in Africa in 2009 was 1.89 million among people aged 15 years or more with a prevalence of 317.3 (314.0–748.2)/100000 population. There were about 990 thousand and 898 thousand stroke survivors equivalent to 335.5 (302.3–702.7)/100,000 and 299.3 (268.4–579.0)/100,000 among men and women, respectively (Table 6 and Figure 6). Based on the same prevalence rates, comparable figures for the year 2010 and 2013 would amount to 1.94 (1.90–4.57) and 2.09 (2.06–4.93) million stroke survivors respectively, also suggesting an increase of 9.6% between 2009 and 2013 that is attributable to growth and ageing of the African population alone.

thumbnail
Figure 6. Bubble graph showing relationship between age and crude prevalence of stroke survivors, with size of bubble corresponding to sample size.

https://doi.org/10.1371/journal.pone.0100724.g006

thumbnail
Table 6. Estimated number of stroke survivors and prevalence in Africa in 2009 (estimates derived from bubble graph model and United Nations population figures).

https://doi.org/10.1371/journal.pone.0100724.t006

Discussion

Some systematic reviews have been published on the burden of stroke in Africa but without a continent-wide estimate of stroke incidence and prevalence rates [7], [42]. There are also global reviews of stroke with few studies on Africa population included [13], [43], [44]. For example, in a systematic review of 56 population-based studies globally, only one African site (Ibadan, Nigeria) was considered. The result from this survey may not necessarily reflect the overall burden of stroke in Africa [13]. However, to the best knowledge, this study provides the first continent-wide estimate of the incidence and prevalence rates of stroke in Africa. The estimates were strictly based on a simple statistical analysis with appropriate consideration of reported mean ages and sample sizes from individual studies. Moreover, having applied the UN population demographics in our final model, the current estimates fairly reflect the, albeit very limited, available published data on incidence and prevalence rates of stroke in Africa, and may help policymakers across several African countries institute effective public health response to the growing burden.

Generally, crude incidences of stroke from population/community based-studies were higher with the two low incidence rates recorded obtained from population-based stroke registries in Nigeria [38], [40] (Table 3). Despite over 3 decades of potential improvement in stroke registration between the two studies (1975–2007), the low incidence rates may still be indicative of incompleteness of stroke registries in many Africa settings, and that data obtained from these registries may be unreliable and inappropriate for estimation of stroke burden. In contrast, in many high income countries where there is active registration of stroke cases, population-based stroke registries have been reliable sources of data for estimation of stroke incidences [7], [45]. In this study, the pooled crude incidence rate from community-based studies was higher at 112.9 per 100,000 person years compared to 77.4 from hospital-based studies. The difference suggests a likely under-estimation of stroke incidences from hospital-based studies, which has also been observed by some studies, particularly due to the very few stroke cases presenting to standard health facilities [2]. The observed prevalence rates of stroke survivors were generally high (all prevalence studies were population-based) with a pooled crude prevalence rate of 387.9/100,000 population. The low prevalence rate recorded in Ethiopia in 1988 may not be unconnected with the high mortality rates from stroke, which has generally been reported in many parts of Africa [23], [46]. Moreover, the Ethiopian study was broadly a survey of neurological disorders in the community, which could possibly imply that active case recognition of specific stroke cases may be less rigorous.

The modelling showed a rising incidence and prevalence rates of stroke with increasing age and higher figures recorded among men, which is in keeping with several research findings on stroke burden [24]. Over 483 thousand new cases of stroke with an incidence rate of 81.2/100,000 py (men 103.3, women 59.5), and about 1.89 million stroke survivors with a prevalence rate of 317.3/100000 population (men 335.5, women 299.3), both among people aged 15 years or more were estimated in 2009. A report in 2004 suggested that that about 8% of all first-ever strokes (about 5 million) occurred in Africa and 5% of over 30 million stroke survivors worldwide were in Africa [5], [11], and this amounts to about 400 thousand new stroke cases and 1.5 million stroke survivors. A systematic review in sub-Saharan Africa on studies published between 1966–2006 showed age-standardized prevalence rates of 114–315/100,000 populations and 154–281/100,000 population among men and women, respectively [7]. Another review in 2006 showed that the prevalence of stroke survivors ranges from 200–300/100,000 population in sub-Saharan Africa, and incidence rates ranges from 15–68/100,000 py [42]. These figures are comparable with the current estimates, which further underpins a near representation of the burden of stroke in Africa. The minor differences may probably be due to the study periods, age groups, fewer data-points, focus on sub-Saharan Africa, and the fact that these were largely qualitative reviews and not based on a detailed statistical synthesis. However, according to the 2014 GBD estimates by Feigin and colleagues, over a 100% increase in the total number of new stroke cases and stroke survivors was recorded between 1990 and 2010 in LMIC, with an estimated incidence rate of 281.1/100,000 py and prevalence rate of 393.4/100,000 population in 2010 [1]. While it is understandable that not all LMIC have contextual similarities with African countries and direct comparisons may be inappropriate, it may however still be logical to conclude that the estimates from this study reflect the current stroke burden in many LMIC.

While this study aimed to provide an evidence-based continent-wide estimation of stroke incidence and prevalence rates in Africa through simple statistical analysis, the study has some important limitations. In particular the low research output and quality of selected studies from Africa constrained the overall analysis. There were 19 studies covering only 10 countries in Africa with an overall sample size of over 6.3 million. Moreover, stroke case ascertainment were not well defined across some studies, and this has been documented in some reviews [45]. Due to these limited data and with some full texts assessed showing evidence of a detailed epidemiological exercise, an inclusion of some of these studies in our final analysis was allowed. In addition, some reported stroke incidence data were based on both first stroke and recurrent stroke events, and not all reported prevalence rates were strictly based on stroke survivors. However, having ensured all studies showed evidence of a degree of epidemiological survey rigour, and all extracted data points included in our modelling (refer to Table 2, and Table S3 in File S1), the current estimates could still give a near representation of the burden of stroke in Africa.

Public health response to stroke in Africa

The prevention of stroke and many non-communicable diseases in Africa has been affected mainly by weak health systems and poor government response [46]. To date, the priorities of many African countries remain infectious diseases: mainly HIV/AIDS, malaria and tuberculosis [3], [47], despite the availability of affordable and cost effective stroke prevention initiatives [48], [49]. For example, Walker and colleagues reported that African countries do not have national strategies to address smoking, alcohol, physical inactivity and unhealthy diets including reducing salt and fat contents of processed foods [48]; and stroke units, where the awareness on these risk factors could have been raised, are rarely available [12], [50]. The INTERSTROKE study findings show that hypertension is the main risk factor of all stroke subtypes with odds of about 2.64 [8], this is more prominent among young Africans who present with stroke unaware of their high blood pressure status [15]. Truelsen argues that the prevalence of stroke in Africa might increase due to substantial changes in major stroke risk factors in the presence of a biased focus on the prevention and control of infectious diseases, at the expense of many NCDs [51].

The diagnosis of stroke in many African settings remain a huge challenge [42]. Some hospital surveys in sub-Saharan Africa have shown that CT scans are only conducted on less than half of patients presenting with stroke, and this is mainly among those that can afford it [42], [52]. In fact, experts have reported that the unavailability and/or high costs of cranial CT imaging in many parts of Africa have limited information on the pathologic profiles of different stroke types in the continent, with this often affecting the diagnosis, treatment and the overall management of the disease [52]. Reports further show that in some areas with better population-wide access to CT scans, as the case in Tanzania, Ghana, and the Medical University of Southern African (MEDUNSA) Stroke Data Bank (MSDB), there have been improvements in diagnosis of stroke, with varying cases of ischaemic and haemorrhagic strokes reported [53][55]. Moreover, the challenges of appropriately distinguishing first from recurrent stroke episodes have also affected stroke case ascertainment, especially during epidemiological surveys [7] Neurologists have noted the importance of proper planning during community surveys, active registration and follow-up of new stroke cases identified, and training and re-training of health workers on stroke diagnosis [17]; arguing that the absence of these in many African settings have resulted in increased number of stroke cases in the community who have never had contact with standard health facilities, and inability to categorize these as first or recurrent strokes, including relating such existing stroke cases to a particular period during surveys [7], [17].

As noted earlier, a systematic review has suggested that income is a strong predictor of stroke risk and fatalities [6]. For example, the average cost of a cranial CT in Ugandan was approximately $60 USD between 2000 and 2010 [2]. This is expensive in most African population groups where many still live below the poverty index of less than $1.25/day [56]. The high cost of health services in the absence of an effective health insurance schemes and adequate resources allocated for stroke prevention and management has affected healthcare seeking behaviour in some African settings [7], [57]. Many stroke patients have been managed at home due to lack of hospital funds, with only few presenting to standard health facilities several days after the onset of symptoms having tried low-priced under-resourced clinics [2]. For those who manage to get to standard health facilities, there are also challenges arising from poor quality of care, as several studies have reported massive gaps exist in the management of acute stroke in Africa compared to many high income countries [58], [59].

The unavailability of data with low research output has been a major setback in the management of stroke in Africa [51]. Experts have reported that reliable data from which evidence-based policy decisions can be made are sparse in Africa [60]. Many have argued that no study in Africa can be regarded as an ideal stroke study [17], adding that there were no proper stroke registries and demographic health surveys which invariably limit active registration, follow-up of cases and conduct of community-based studies [45]. Based on the current findings, hospital-based studies and door-to-door surveys were mainly conducted in Africa and despite the rigour of these few epidemiological surveys, gaps have been identified with regards to case ascertainment and study protocols [46].

Our findings suggest an increasing burden of stroke in Africa. However, with the current low availability of data, there is still need for more research on stroke, and related vascular disease risk factors to appropriately quantify this burden. An investment in research capacity, basically to conduct and fund higher quality research may help raise awareness on stroke burden in Africa. An awareness and fair understanding of stroke burden and disease pattern in Africa may further prompt appropriate policy response and scale up current intervention programmes.

Supporting Information

File S1.

Table S1. Search terms (EMBASE). Table S2. Search terms (Global Health). Table S3. All data points employed in modelling.

https://doi.org/10.1371/journal.pone.0100724.s001

(DOC)

Acknowledgments

The author wishes to thank Professors Harry Campbell and Igor Rudan for reviewing the models, and ‘Funke Davies-Adeloye for proof-reading the manuscript.

Author Contributions

Conceived and designed the experiments: DA. Performed the experiments: DA. Analyzed the data: DA. Contributed reagents/materials/analysis tools: DA. Wrote the paper: DA.

References

  1. 1. Feigin VL, Forouzanfar MH, Krishnamurthi R, Mensah GA, Connor M, et al. (2014) Global and regional burden of stroke during 1990–2010: findings from the Global Burden of Disease Study 2010. Lancet 383: 245–255.
  2. 2. Chin JH (2012) Stroke in Sub-Saharan Africa: An Urgent Call For Prevention. Neurology 78: 1007–1008.
  3. 3. Beaglehole R, Bonita R, Alleyne G, Horton R, Li L, et al. (2011) UN High-Level Meeting on Non-Communicable Diseases: addressing four questions. Lancet 378: 449–455.
  4. 4. Strong K, Mathers CD, Bonita R (2007) Preventing stroke: saving lives around the world. Lancet Neurol 6: 182–187.
  5. 5. World Health Organization (2004) Global Burden of Disease (GBD) 2002 estimates. World Health Report 2004. Geneva, Switzerland: WHO.
  6. 6. Johnston SC, Mendis S, Mathers CD (2009) Global variation in stroke burden and mortality: estimates from monitoring, surveillance, and modelling. Lancet Neurol 8: 345–354.
  7. 7. Connor MD, Walker R, Modi G, Warlow CP (2007) Burden of stroke in black populations in sub-Saharan Africa. Lancet Neurology 6: 269–278.
  8. 8. O'Donnell MJ, Xavier D, Liu L, Zhang H, Chin S, et al. (2010) Risk factors for ischaemic and intracerebral haemorrhagic stroke in 22 countries (the INTERSTROKE study): a case-control study. Lancet 376: 112–123.
  9. 9. Truelsen T, Beggs S, Mathers CD (2006) The global burden of cerebrovascular disease. Geneva, Switzerland: WHO.
  10. 10. World Health Organization (2002) The World Health Report 2002: reducing risks, promoting healthy life. Geneva, Switzerland: WHO.
  11. 11. Truelsen T (2010) Stroke incidence studies in Africa. The Lancet Neurology 9 (8): 755–757.
  12. 12. Wahab KW (2008) The burden of stroke in Nigeria. International Journal of Stroke 3: 290–292.
  13. 13. Feigin VL, Lawes CMM, Bennett DA, Barker-Collo SL, Parag V (2009) Worldwide stroke incidence and early case fatality reported in 56 population-based studies: a systematic review. Lancet Neurol 8: 355–369.
  14. 14. Mensah GA (2008) Epidemiology of stroke and high blood pressure in Africa. Heart 94: 697–705.
  15. 15. Walker R (1994) Hypertension and stroke in sub-saharan Africa. Transactions of the Royal Society of Tropical Medicine & Hygiene 88: 609–611.
  16. 16. Mathers CD (2005) Uncertainty and data availability for the global burden of disease estimates 2000–2002. Evidence and Information for Policy Working Paper. Geneva, Switzerland: WHO.
  17. 17. Sudlow CL, Warlow CP (1996) Comparing stroke incidence worldwide: what makes studies comparable? Stroke 27: 550–558.
  18. 18. World Bank (2014) World Bank list of economies (October 2013). Washington DC: World Bank.
  19. 19. Hatano S (1976) Experience from a multicentre stroke register: a preliminary report. Bull World Health Organ 54: 541–553.
  20. 20. WHO MONICA Project Investigators (1988) The World Health Organization MONICA Project (Monitoring trends and determinants in cardiovascular disease). J Clin Epidemiol 41: 105–114.
  21. 21. World Health Organization (1990) Section 2: Stroke event registration data component. WHO Monica Project: MONICA manual Part IV: Event Registration. Geneva, Switzerland: WHO.
  22. 22. United ations 2013) World Population Prospects: The 2012 Revision. New York City, United Nations.
  23. 23. Tekle-Haimanot R, Abebe M, Gebre-Mariam A, Forsgren L, Heijbel J, et al. (1990) Community-based study of neurological disorders in rural central Ethiopia. Neuroepidemiology 9: 263–277.
  24. 24. Dewhurst F, Dewhurst MJ, Gray WK, Aris E, Orega G, et al. (2013) The prevalence of neurological disorders in older people in Tanzania. Acta Neurologica Scandinavica 127 (3): 198–207.
  25. 25. Walker R, Whiting D, Unwin N, Mugusi F, Swai M, et al. (2010) Stroke incidence in rural and urban Tanzania: a prospective, community-based study. Lancet Neurology 9: 786–792.
  26. 26. Farghaly WMA, El-Tallawy HN, Shehata GA, Rageh TA, Abdel-Hakeem NM, et al. (2013) Epidemiology of nonfatal stroke and transient ischemic attack in Al-Kharga District, New Valley, Egypt. Neuropsychiatric Disease and Treatment 9: 1785–1790.
  27. 27. El Tallawy HNA, Farghaly WMA, Rageh TA, Shehata GA, Badry R, et al. (2013) Door-to-door survey of major neurological disorders (project) in Al Quseir City, Red Sea Governorate, Egypt. Neuropsychiatric Disease and Treatment 9: 767–771.
  28. 28. Khedr EM, Elfetoh NA, Al Attar G, Ahmed MA, Ali AM, et al. (2013) Epidemiological study and risk factors of stroke in assiut governorate, Egypt: Community-based study. Neuroepidemiology 40 (4): 288–294.
  29. 29. Kandil MR, El-Tallawy HN, Farawez HM, Khalifa G, Ahmed MA, et al. (2006) Epidemiology of Cerebrovascular Stroke and TIA in Upper Egypt (Sohag) – Relative Frequency of Stroke in Assiut University Hospital. Egypt J Neurol Psychiat Neurosurg 43: 593–602.
  30. 30. El Zunni S, Ahmed M, Prakash PS, Hassan KM (1995) Stroke: Incidence and pattern in Benghazi, Libya. Annals of Saudi Medicine 15 (4): 367–369.
  31. 31. Ashok PP, Radhakrishnan K, Sridharan R, el-Mangoush MA (1986) Incidence and pattern of cerebrovascular diseases in Benghazi, Libya. Journal of Neurology, Neurosurgery & Psychiatry 49: 519–523.
  32. 32. Attia Romdhane N, Hamida MB, Mrabet A, Larnaout A, Samoud S, et al. (1993) Prevalence study of neurologic disorders in Kelibia (Tunisia). Neuroepidemiology 12 (5): 285–299.
  33. 33. Damasceno A, Gomes J, Azevedo A, Carrilho C, Lobo V, et al. (2010) An epidemiological study of stroke hospitalizations in Maputo, Mozambique: a high burden of disease in a resource-poor country. Stroke 41: 2463–2469.
  34. 34. Connor M (2004) Prevalence of Stroke Survivors in Rural South Africa: Results from the Southern Africa Stroke Prevention Initiative (SASPI) Agincourt Field Site. Stroke 35 (3): 627–632.
  35. 35. Rosman KD (1986) The Epidemiology of Stroke In An Urban Black Population. Stroke 17: 667–669.
  36. 36. Matenga J (1997) Stroke incidence rates among black residents of Harare–a prospective community-based study. South African Medical Journal Suid-Afrikaanse Tydskrif Vir Geneeskunde 87: 606–609.
  37. 37. Cossi M-J, Gobron C, Preux P-M, Niama D, Chabriat H, et al. (2012) Stroke: prevalence and disability in Cotonou, Benin. Cerebrovascular Diseases 33: 166–172.
  38. 38. Osuntokun BO, Bademosi O, Akinkugbe OO, Oyediran AB, Carlisle R (1979) Incidence of stroke in an African City: results from the Stroke Registry at Ibadan, Nigeria, 1973–1975. Stroke 10: 205–207.
  39. 39. Osuntokun BO, Adeuja AOG, Schoenberg BS (1987) Neurological disorders in Nigerian Africans: A community-based study. Acta Neurologica Scandinavica 75 (1): 13–21.
  40. 40. Danesi MA, Okubadejo NU, Ojini FI, Ojo OO (2013) Incidence and 30-day case fatality rate of first-ever stroke in urban Nigeria: The prospective community based Epidemiology of Stroke in Lagos (EPISIL) phase II results. Journal of the Neurological Sciences 331 (1–2): 43–47.
  41. 41. Danesi M, Okubadejo N, Ojini F (2007) Prevalence of stroke in an urban, mixed-income community in Lagos, Nigeria. Neuroepidemiology 28: 216–223.
  42. 42. Kengne AP, Anderson CS (2006) The neglected burden of stroke in Sub-Saharan Africa. International Journal of Stroke 1: 180–190.
  43. 43. Feigin VL, Lawes CMM, Bennett DA, Anderson CS (2003) Stroke epidemiology: a review of populationbased studies of incidence, prevalence, and case-fatality in the late 20th century. Lancet Neurology 2: 43–53.
  44. 44. Aho K, Harmsen P, Hatano S, Marquardsen J, Smirnov VE, et al. (1980) Cerebrovascular disease in the community: results of a WHO Collaborative Study. Bulletin of the World Health Organization 58: 113–130.
  45. 45. The Global Burden of Disease Stroke Expert Group (2012) Methodology of the Global and Regional Burden of Stroke Study. Neuroepidemiology 38: 30–40.
  46. 46. Bonita R, Truelsen T (2003) Stroke in sub-Saharan Africa: a neglected chronic disease. Lancet neurol 2: 592.
  47. 47. Chan KY, Adeloye D, Grant L, Kolčić I, Marušić A (2012) How big is the 'next big thing'? Estimating the burden of non-communicable diseases in low- and middle-income countries. J Glob Health 2: 20101.
  48. 48. Walker RW, Jusabani A, Aris E, Gray WK, Mugusi F, et al. (2013) Correlates of short- and long-term case fatality within an incident stroke population in Tanzania. South African Medical Journal Suid-Afrikaanse Tydskrif Vir Geneeskunde 103: 107–112.
  49. 49. Mensah GA (2003) A heart-healthy and "stroke-free" world through policy development, systems change, and environmental supports: a 2020 vision for sub-Saharan Africa. Ethnicity & Disease 13: S4–12.
  50. 50. Kolapo KO, Vento S (2011) Stroke: a realistic approach to a growing problem in sub-Saharan Africa is urgently needed. Tropical Medicine & International Health 16: 707–710.
  51. 51. Truelsen T (2010) Stroke incidence studies in Africa. Lancet Neurology 9: 755–757.
  52. 52. Olubunmi O (2007) Recent Advances In The Management Of Cerebro-Vascular Accident. Benin Journal of Postgraduate Medicine 9.
  53. 53. Joubert J (1991) The MEDUNSA Stroke Data Bank. An analysis of 304 patients seen between 1986 and 1987. South African Medical Journal Suid-Afrikaanse Tydskrif Vir Geneeskunde 80: 567–570.
  54. 54. Matuja W, Janabi M, Kazema R, Mashuke D (2004) Stroke subtypes in Black Tanzanians: a retrospective study of computerized tomography scan diagnoses at Muhimbili National Hospital, Dar es Salaam. Tropical Doctor 34: 144–146.
  55. 55. Nyame PK, Bonsu-Bruce N, Amoah AG, Adjei S, Nyarko E, et al. (1994) Current trends in the incidence of cerebrovascular accidents in Accra. West African Journal of Medicine 13: 183–186.
  56. 56. World Bank (2014) Poverty and Equity data. New York, USA: World Bank.
  57. 57. Walker RW, McLarty DG, Kitange HM, Whiting D, Masuki G, et al. (2000) Stroke mortality in urban and rural Tanzania. Adult Morbidity and Mortality Project. Lancet 355: 1684–1687.
  58. 58. Lemogoum D, Degaute JP, Bovet P (2005) Stroke prevention, treatment, and rehabilitation in sub-saharan Africa. American Journal of Preventive Medicine 29: 95–101.
  59. 59. Adoukonou TA, Vallat JM, Joubert J, MacIan F, Kabore R, et al. (2010) Management of stroke in sub-Saharan Africa: Current issues. [French]. Revue Neurologique 166 (11): 882–893.
  60. 60. Owolabi M (2012) Taming the burgeoning stroke epidemic in Africa: Stroke quadrangle to the rescue. Neurology Conference: 64th American Academy of Neurology Annual Meeting New Orleans, LA United States Conference Start 78.