Determining the Proportionality of Ischemic Stroke Risk Factors to Age.

Area of Science:

• Cardiovascular epidemiology
• Biostatistics
• Health informatics

Background:

• Age is a key stroke risk factor, but its inclusion in predictive models presents challenges.
• Including age can lead to over- or under-prediction, potentially due to non-proportional relationships between other risk factors and age.

Purpose of the Study:

• To develop and apply a framework for assessing the proportionality of stroke risk factors with age.
• To investigate whether common stroke risk factors exhibit non-proportionality across different age groups.

Main Methods:

• A novel framework was developed to test for proportionality between risk factors and age.
• The framework was applied to Framingham Heart Study data, utilizing data from the NHLBI Biologic Specimen and Data Repository Information Coordinating Center.
• Analysis focused on identifying age-dependent contributions of various risk factors to stroke risk.

Main Results:

• Evidence of non-proportionality was found for several stroke risk factors, including diastolic blood pressure, total cholesterol, body mass index (BMI), sex, and hypertension treatment.
• These findings indicate that the influence of these factors on stroke risk varies significantly with age.

Conclusions:

• The proportionality of risk factors with age should be a key consideration in stroke risk prediction modeling.
• Traditional modeling approaches may require adjustments to accurately capture age-related non-proportionality for improved stroke risk assessment.