2014 CRL Blood Pressure Study of Life Insurance Applicants

Abstract

Objective .- Define the relative mortality risk by systolic (SBP) and diastolic blood pressure (DBP) in a relatively healthy cohort split by age and sex with adjustment for smoking status, other findings and admitted heart disease history. Method .- Blood pressure (BP in mm Hg), build, laboratory studies and limited medical history are collected when people apply for individual life insurance. Information on 2,472,706 applicants tested by Clinical Reference Laboratory from 1993 to 2007 was utilized with follow-up for vital status using the September 2011 Social Security Death Master File identifying 31,033 deaths. Data was analyzed by SBP and DBP split by age and sex accounting for smoking and for BMI, urine protein/creatinine ratio and history of heart disease in a Cox multivariate survival analysis. Separate analysis by admitted hypertension history was also conducted. Results are presented by SBP and DBP for 4 age-sex groups with and without added covariates beyond age and smoking status. Results .- Relative mortality progressively increased by SBP level from the 90 to 119 band (down to 80 in younger women) upward with little additional impact by DBP. Addition of covariates beyond age and smoking resulted in a 5% to 10% reduction in relative risk. Although high DBP had limited impact, a pulse pressure/SBP ratio >½ identified 1% of applicants at high mortality risk, with little difference in risk for ratios ≤½. Hypertension history with current BP control was associated with a 10% to 25% increase in relative mortality risk as compared to those with similar BP but no such history. Conclusion .- Increasing SBP is closely associated with increasing relative mortality, starting from the lowest SBP. Increasing DBP has little additional impact, but a pulse pressure/SBP ratio >½ is a potent marker of increased risk as well. Accounting for build and other laboratory findings reduces risk modestly. A history of hypertension with current control increases risk.