Abstract TMP104: The Effect of Hypothetical Lifestyle Strategies on the Long-term Risk of Stroke

Abstract

Randomized trials have examined the short-term effects of lifestyle interventions on the risk of stroke. Using observational data to estimate the long-term effect of such strategies requires valid adjustment for time-varying confounders- such as hypertension- that are affected by prior lifestyle risk factors. We estimated the long-term effect of hypothetical lifestyle strategies on risk of stroke using data from the Nurses’ Health Study, a prospective cohort. We followed 60,390 women, aged 52 years on average at baseline, from 1986 to 2010. We used a Monte Carlo simulation to estimate the 24-year risk of stroke had all participants followed three non-dietary lifestyle strategies: smoking cessation, exercising (≥30 minutes/day), reducing body mass index (BMI) by 5% every two years if overweight/obese; as well as several dietary strategies including: eating ≥ 3 servings/week of fish, eating ≤ 3 servings/week of unprocessed red meat, eating no processed red meat, eating ≥ 5 servings/day of fruits and vegetables, among others. We used the parametric g-formula to adjust for baseline and time-varying confounders, comparing the estimated risk of stroke under our specified strategies of interest to the same had all participants followed a “do-nothing” strategy. The observed 24-year risks of total, ischemic and hemorrhagic stroke in this cohort were 2.9%, 2.2% and 0.6%. The estimated risk of total stroke was 25% lower (95% CI: 5% to 46%) under a combination of all three non-dietary strategies, the most effective being exercise (19% lower risk) and smoking cessation (5% lower risk). Of the dietary strategies, eating ≤ 3 servings of unprocessed red meat reduced the estimated risk of total stroke by 4% (1% to 7%). The risk of ischemic stroke under all three non-dietary strategies was 36% lower (15% to 59%), but risk of hemorrhagic stroke did not materially change. Exercise reduced the estimated risk of ischemic stroke by 29% (7% to 56%) and lowering BMI by 4% (0% to 8%). Additionally, ischemic stroke risk was lowered by 9% (0% to 19%) by decreasing intake of processed red meat and 6% (2% to 10%) by decreasing intake of unprocessed red meat. In summary, lifestyle modifications were estimated to reduce the 24-year risk of total stroke by up to 25% in this population of middle-aged women.