Kenya's stroke deaths fall as new cases climb, county-level GBD data show

Ischaemic stroke — the kind caused by a clot cutting off blood to the brain — killed 11,328 people in Kenya in 2023 and accounted for 317,719 disability-adjusted life years, according to a county-level analysis of Global Burden of Disease Study 2023 data published in The Lancet Regional Health – Africa on 7 October 2026 [s1].

The headline pattern is one that recurs across middle-income countries as they age: deaths from the disease are falling, but the number of new cases is not. Over 1990 to 2023 the age-standardised death rate declined at an estimated annual percentage change of −0.69% and the DALY rate at −0.45%, while age-standardised incidence edged upward at 0.22% a year [s1]. The authors read that divergence as treatment catching up faster than prevention — more people surviving a stroke, but roughly as many still having one.

What was analysed

The study is a secondary analysis rather than new fieldwork. The authors extracted GBD 2023 estimates of ischaemic stroke incidence, prevalence, mortality and DALYs for each of Kenya's 47 counties, broken down by age and sex, from 1990 to 2023 [s1]. They used Joinpoint regression to characterise the long-run trends, with estimated annual percentage changes and 95% confidence intervals, and an ARIMA time-series model to project the burden forward to 2033 [s1]. A comparative risk assessment quantified how much of the burden could be attributed to modifiable risk factors [s1]. The work received no funding [s1].

That design carries the usual caveat for any GBD-derived paper: the estimates are only as good as the data feeding the model, which in many Kenyan counties means sparse vital registration and verbal-autopsy inputs. The uncertainty is widest exactly where surveillance is thinnest.

The geography

The national averages hide a wide spread between counties. Kwale, on the southern coast, recorded the highest age-standardised incidence at 108.97 per 100,000, while Siaya, in the west near Lake Victoria, recorded the lowest at 57.77 — nearly a twofold gap [s1]. Garissa, in the arid north-east, carried the highest mortality at 46.49 per 100,000 [s1]. Those are not small differences, and they are the kind of subnational detail a devolved health system is supposed to be able to act on.

The single largest attributable factor was blood pressure: high systolic blood pressure accounted for an estimated 35% of the ischaemic stroke burden [s1]. That matters for policy because blood pressure is cheap to screen for and treatable with generic drugs, which makes it the most tractable lever a county health department has.

Devolution as a turning point

The paper's framing device is Kenya's 2013 constitutional devolution, which moved responsibility for running health services from the centre to the 47 county governments. The authors report that declines in stroke mortality accelerated after 2013, and they present this as evidence that decentralised governance improved access and outcomes [s1]. That is an association drawn from trend data, not a controlled comparison, so it cannot by itself prove devolution caused the improvement — other things changed over the same decade, including the rollout of universal health coverage pilots and wider access to acute care.

The age and sex pattern

Working-age men, roughly 40 to 65, bore the highest mortality burden, a pattern consistent with earlier onset of hypertension and cardiovascular risk in men [s1]. Among the oldest old, from 85 years up, women's burden exceeded men's — largely a survivorship effect, as women live longer and so make up more of the very elderly population in which strokes concentrate [s1].

The forecast

Projecting to 2033, the model suggests mortality continuing to fall, to roughly 24.5 per 100,000, while incidence keeps rising, to around 85 to 86 per 100,000 [s1]. The authors read that as a warning rather than reassurance: better acute treatment will keep more stroke survivors alive with disability, which raises the long-term burden on rehabilitation and chronic care even as the death rate improves.

What to watch

The practical recommendation is unglamorous and familiar: strengthen primary prevention, especially blood-pressure control, rather than relying on acute treatment to mop up afterwards [s1]. The county-level estimates give health planners a map of where incidence is highest and where prevention is weakest; whether that translates into screening and treatment programmes on the ground is the question the next decade of surveillance will answer. The figure to watch is incidence — if prevention works, the upward trend the model projects to 2033 should bend down first.

Sources

• National and subnational burden of ischaemic stroke in Kenya from centralisation to devolution, 1990–2023, with forecasts to 2033: an analysis for Global Burden of Disease Study 2023 — The Lancet Regional Health – Africa, 7 October 2026