Relationships between treated hypertension and subsequent mortality in an insured population.

OBJECTIVE To investigate if a mortality differential exists between insurance policyholders with treated hypertension and policyholders who are not under such treatment, where both groups are noted to have the same blood pressure at the time of policy issue. BACKGROUND Hypertension is a known mortality risk factor in the insured and general population. Treatment for hypertension is very common in the insured population, especially as age increases. At the time of insurance application, a subset of individuals with treated hypertension will have blood pressures that are effectively controlled and are in the normal range. These individuals often meet established preferred underwriting criteria for blood pressure. In some life insurance companies, they may be offered insurance at the same rates as individuals who are not hypertensive with the same blood pressure. Such companies make the assumption that the pharmacologically induced normotensive state confers no excess risk relative to the natural normotensive state. Given the potential pricing implications of this decision, we undertook an investigation to test this hypothesis. METHODS We studied internal data on direct and reinsurance business between 1975 and 2001 followed through anniversaries in 2002 or prior termination with an average duration of 5.2 years per policy. Actual-to-expected analyses and Cox proportional hazards models were used to assess if a mortality differential existed between policyholders coded for hypertension and policyholders with the same blood pressure that were not coded as hypertensive. RESULTS Eight thousand six hundred forty-seven deaths were observed during follow-up in the standard or preferred policy cohort. Within the same blood pressure category, mortality was higher in policyholders identified as treated hypertensives compared with those in the subset of individuals who were not coded for hypertension. This finding was present in males and females and persisted across age groups in almost all age-gender-smoking status subsets examined. The differential in mortality was 125% to 160% of standard mortality based on the ratio of actual-to-expected claims. CONCLUSION In this insured cohort, a designation of treated hypertension is associated with increased relative mortality compared to life insurance policyholders not so coded.