Jack Hadley, James D. Reschovsky
his analysis examined the relationship between health status and the cost of nongroup insurance by estimating a model of the relationship between the premium for nongroup insurance, the ages and health status of the people covered by the policy, and measures of the cost of insurance in the are and state regulation of nongroup insurance. Since people who buy nongroup insurance may be in better health than those who dont, the premium model was estimated both with and without a correction for possible selection bias. We used data on all non-elderly adults from the 1996-97 and 1998-99 Community Tracking Study. We excluded full-time students and people covered by military insurance or Medicare. We found that the probability of buying nongroup insurance was significantly lower for people in fair or poor health, by 15-17%, compared to a person in excellent health. With the selection correction, premiums were higher for people in less than excellent health, by 10-14% for people in very good to fair health, and by 34% for someone in poor health. Without the selection correction, none of the health status variables approach statistical significance, and suggest at most a 13% higher premium for someone in poor health compared to a person in excellent health.
For a full copy of this article, please Jack Hadley or James Reschovsky.