Effects of Hypothetical Interventions on Ischemic Stroke Using Parametric G-Formula

Area of Science:

• Cardiovascular Epidemiology
• Biostatistics
• Preventive Medicine

Background:

• Standard analytic methods struggle with time-dependent confounding in exposure-outcome analyses.
• Biased estimates of exposure effects on outcomes can result from inadequate control of time-dependent confounders.
• Accurate assessment of modifiable risk factors for ischemic stroke is crucial.

Purpose of the Study:

• To estimate the 11-year risk of ischemic stroke by sex using the parametric g-formula.
• To control for time-dependent confounders in assessing the effects of body mass index, physical activity, cholesterol levels, hypertension, and smoking.
• To evaluate the impact of hypothetical interventions on ischemic stroke risk.

Main Methods:

• Utilized data from the Multi-Ethnic Study of Atherosclerosis (MESA) cohort.
• Followed 6809 men and women aged 45-84 years.
• Employed the parametric g-formula to estimate stroke risk under six hypothetical interventions, including lifestyle modifications and risk factor control.

Main Results:

• Hypothetical interventions could reduce the 11-year risk of ischemic stroke by 85% in men.
• In women, the same interventions were estimated to reduce ischemic stroke risk by 55%.
• Joint hypothetical interventions were also simulated, showing potential for risk reduction.

Conclusions:

• The explored hypothetical interventions demonstrate significant potential for reducing ischemic stroke risk in both sexes.
• The parametric g-formula provides a robust method for analyzing time-dependent confounding in epidemiological studies.
• Modifying key risk factors holds substantial promise for primary prevention of ischemic stroke.