Risk factor changes account for 57% of decrease in ischaemic stroke incidence
Declining Incidence of Ischemic Stroke: What Is the Impact of Changing Risk Factors?
The Tromsø Study 1995 to 2012Vangen-Lønne AM, Wilsgaard T, Johnsen SH, et al. - Stroke. 2017; 48: published online ahead of print
Stroke mortality rates have declined during the last decades due to a declining stroke incidence and a reduced case fatality . A limited proportion of decreased stroke incidence can be explained by the improvement of risk factor control between 1982 and 1995 [2-4].
In this study, the impact of modifiable risk factors on the changing incidence of ischaemic stroke (IS) between 1995 and 2012 was estimated, using individual person data from repeated surveys in a general population (n=27 936).
- Individuals with incident IS had significantly higher systolic (SBP) and diastolic (DBP) at baseline, higher BMI and total cholesterol, lower HDL-C and were more often daily smokers. Women with IS were associated with a higher prevalence of diabetes mellitus (DM), were more often obese and reported less vigorous physical activity than women without IS.
- In Cox proportional hazards regression analysis, hypertension was the strongest risk factor, with a 92% increased multi-adjusted hazard for IS in hypertensive participants.
- DM was associated with 80% higher risk, daily smoking was associated with 71% higher risk, hyperlipidaemia was associated with 28% increased hazard and HDL-C was associated with a 22% reduced hazard rate per 1 mmol/L increase. Obese individuals had 28% higher hazard for IS compared with those with normal weight.
- The SBP reduction was the largest single contributor to the decreased risk of stroke, accounting for 26% of the observed decline. The decreasing prevalence of daily smoking accounted for 17% and changes in HDL-C for 2%. Reduction of total cholesterol and increase in physical activity were associated with 12% and 5% of the declining IS incidence, respectively (not significant).
- The elevated DM prevalence contributed negatively with 4% increase in risk and the change in BMI over time was associated with 5% increasing risk (not significant).
- In the fully adjusted model, all risk factors together explained 57% of the decline in incidence of IS from 1995 to 2012.
- For the youngest age group (baseline age <60 years), all risk factors together explained 47% (95% CI 17–100%) of the decline in incidence of IS from 1995 to 2012, while the corresponding estimate for the eldest age group (baseline age ≥60 years) was 50% (95% CI 13–100%).
- The incidence rate of IS among persons aged ≥30 years decreased from 363 per 100 000 person-years in 1995 (95% CI 267–459) to 306 per 100 000 person-years in 2012 (95% CI 234–379).
- According to Poisson regression models, there was an average annual decline of IS incidence rate of 1.7% (95% CI 0.6–2.8%, P for linear trend 0.0022).
In a general population analysis with data from 1995 to 2012, risk factor changes accounted for 57% of the decrease in IS incidence. The most important contributors were the declines in mean SBP and prevalence of smoking, whereas the increasing DM prevalence contributed negatively to the IS incidence. These findings confer to the need to further reduce modifiable risk factors and encourage future research of less well-recognized risk factors.