Related Experiment Video
Updated: Aug 12, 2026

Inverse Probability of Treatment Weighting (Propensity Score) using the Military Health System Data Repository and National Death Index
Published on: January 8, 2020
The West Point Study: 40 years of follow-up
D A Clark1, G D Tolan, R Johnson
1Armstrong Laboratory (AFMC), Clinical Sciences Division, Brooks AFB, TX 78235-5117.
Abstract:
Completion of cardiovascular evaluations of 387 members marked the end of 40 years of follow-up in the West Point Study. Coronary artery disease (CAD) caused 4 cases of sudden death, 14 cases of myocardial infarction (MI), 13 cases of angina, and 17 cases of silent CAD. Using risk factors (serum cholesterol, estimated HDL-cholesterol, systolic blood pressure, and smoking status) measured before age 28, we derived a multivariate regression formula for predicting which members of the study, had they been pilots, would have been grounded for CAD before age 55. This derivation used data from only those subjects with CAD or with no evidence of CAD. We then used the formula to compute a risk-related score for each member of the study. In the tertile group with the highest risk-related scores, 17% manifested CAD by age 55 and the first event occurred at age 39. In the tertile group of lowest scores, 2% experienced CAD by age 55 and the first event occurred at age 51. We conclude that it is possible to select pilot candidates with the lowest risk for CAD.
Related Concept Videos
Longitudinal Research
Regression Toward the Mean
Longitudinal Studies