Olena O Seminog, Peter Scarborough, F Lucy Wright, Mike Rayner, Michael J Goldacre

British Medical Journal (2019) 365

Objectives To study trends in stroke mortality rates, event rates, and case fatality, and to explain the extent to which the reduction in stroke mortality rates was influenced by changes in stroke event rates or case fatality.

Design Population based study.

Setting Person linked routine hospital and mortality data, England.

Participants 795 869 adults aged 20 and older who were admitted to hospital with acute stroke or died from stroke.

Main outcome measures Stroke mortality rates, stroke event rates (stroke admission or stroke death without admission), and case fatality within 30 days after stroke.

Results Between 2001 and 2010 stroke mortality rates decreased by 55%, stroke event rates by 20%, and case fatality by 40%. The study population included 358 599 (45%) men and 437 270 (55%) women. Average annual change in mortality rate was −6.0% (95% confidence interval −6.2% to −5.8%) in men and −6.1% (−6.3% to −6.0%) in women, in stroke event rate was −1.3% (−1.4% to −1.2%) in men and −2.1% (−2.2 to −2.0) in women, and in case fatality was −4.7% (−4.9% to −4.5%) in men and −4.4% (−4.5% to −4.2%) in women. Mortality and case fatality but not event rate declined in all age groups: the stroke event rate decreased in older people but increased by 2% each year in adults aged 35 to 54 years. Of the total decline in mortality rates, 71% was attributed to the decline in case fatality (78% in men and 66% in women) and the remainder to the reduction in stroke event rates. The contribution of the two factors varied between age groups. Whereas the reduction in mortality rates in people younger than 55 years was due to the reduction in case fatality, in the oldest age group (≥85 years) reductions in case fatality and event rates contributed nearly equally.

Conclusions Declines in case fatality, probably driven by improvements in stroke care, contributed more than declines in event rates to the overall reduction in stroke mortality. Mortality reduction in men and women younger than 55 was solely a result of a decrease in case fatality, whereas stroke event rates increased in the age group 35 to 54 years. The increase in stroke event rates in young adults is a concern. This suggests that stroke prevention needs to be strengthened to reduce the occurrence of stroke in people younger than 55 years.