Abstract Background and aims Ethnic inequalities in stroke incidence are well-established, but long-term trends and the impact of socioeconomic status(SES) remain understudied.We estimated incidence rates by ethnicity and SES over 30-years using the South London Stroke Register. Methods We conducted a population-based -study including all residents of a geographically defined area of South-London, calculating standardized incidence rates using census-derived denominators. Area-level Index of Multiple Deprivation was the SES indicator. Poisson regression estimated incidence rate ratios comparing ethnic-minority vs White groups, adjusting for age, sex, and SES. Then, ethnicity-SES interaction testing and marginal rates were derived by ethnicity-SES subgroups. Results Among 7,726 people with first-ever stroke, overall standardized incidence decreased 34% between 1995-1999 and 2010-2014 (198;187.8-208 to 131;122.4-138.9 per 100,000), then increased 13% in 2020-2024. Black-African and Black-Caribbean populations had higher baseline incidence persisting across all periods. Ethnic inequalities narrowed initially and then widened, with incidence twice as high in Black-African (IRR = 2.31;2.03-2.62) and Black-Caribbean (IRR = 2.00;1.73-2.31) vs White groups in 2020-2024. Patterns were consistent across subtypes, with the largest inequalities in intracerebral hemorrhage. Ethnic inequalities in incidence attenuated on SES adjustment but remained significant. Within ethnic groups, most deprived Black-African and Black-Caribbean populations showed highest incidence. Conclusions Stroke incidence has risen sharply in the past five years, driven by rising incidence rates in Black-African and Black-Caribbean populations, with reduction in the White-population. Achieving equity requires ensuring uptake of cardiovascular risk programmes across all socioeconomic and ethnic groups, with earlier risk management for groups who develop stroke 10-12 years younger than the White population. Conflict of interest Camila Pantoja-Ruiz: Nothing to disclose.
Pantoja-Ruiz et al. (Fri,) studied this question.