Stroke survivors in low- and middle-income countries: a meta-analysis of prevalence and secular trends

Martinsixtus Ezejimofor, Yen-Fu Chen, Ngianga-Bakwin Kandala, Benedeth Ezejimofor, Aloysius Ezeabasili, Saverio Stranges, Olalekan Uthman

Research output: Contribution to journalArticlepeer-review

36 Citations (Scopus)
6 Downloads (Pure)


Purpose To provide an up-to-date estimate on the changing prevalence of stroke survivors, and examines the geographic and socioeconomic variations in low and middle-income countries (LMICs). Methods We searched MEDLINE, EMBASE, SCOPUS and Web of Science databases and systematically reviewed articles reporting stroke prevalence and risk factors from inception to July 2015. Pooled prevalence estimates and secular trends based on random-effects models were conducted across LMICs, World Bank regions and income groups. Results Overall, 101 eligible community-based studies were included in the meta-analysis. The pooled crude prevalence of stroke survivors was highest in Latin America and Caribbean (21.2 per 1000, 95% CI 13.7 to 30.29) but lowest in sub-Saharan Africa (3.5 per 1000, 95% CI 1.9 to 5.7). Steepest increase in stroke prevalence occurred in low-income countries, increasing by 14.3% annually while the lowest increase occurred in lower-middle income countries (6% annually), and for every 10 years increase in participants' mean age, the prevalence of stroke survivors increases by 62% (95% CI 6% to 147%). Conclusion The prevalence estimates of stroke survivors are significantly different across LMICs in both magnitude and secular trend. Improved stroke surveillance and care, as well as better management of the underlying risk factors, primarily undetected or uncontrolled high blood pressure (HBP) are needed.
Original languageEnglish
Pages (from-to)68-76
JournalJournal of the Neurological Sciences
Early online date10 Mar 2016
Publication statusPublished - 15 May 2016


Dive into the research topics of 'Stroke survivors in low- and middle-income countries: a meta-analysis of prevalence and secular trends'. Together they form a unique fingerprint.

Cite this