Factors important in the purchase of partnership long-term care insurance

Health Services Research, June, 1998 by Nelda McCall, Steven Mangle, Ellen Bauer, James Knickman

NOTES

1. This percentage is based on data available from the National Association of Insurance Commissioners (NAIC). It should be noted that because the collection of this type of information by NAIC is relatively new, the data reported should be viewed as only a general indicator of the long-term care insurance market.

2. These percentages are calculated based on 1993 and 1994 data from the programs divided by state data reported yearly by the insurers to the National Association of Insurance Commissioners (NAIC). Sales for partial years during which Partnership policies were sold were estimated by multiplying total year sales by the proportion of the year Partnership policies were sold. The NAIC data, especially for the earlier years, should be taken only as an approximation, because it is acknowledged to be subject to known inaccuracies in data reporting due to the newness of the reporting forms.

3. Data from a survey of California purchasers of a long-term care insurance policy that was not a Partnership policy after the Partnership was offered indicated that almost 80 percent did not know about the Partnership. Of those who knew about the Partnership, almost half said they wanted lifetime coverage, which was not available through the Partnership product; more than one-third said that the link to Medicaid was an important reason not to purchase; and almost three of ten said they wanted a policy that had less extensive benefits than the Partnership minimums (i.e., they wanted lower daily benefit amounts or did not want inflation protection).

4. It could be argued that four of the variables in the model (i.e., believes that Medicare covers sufficient long-term care, talked with a financial planner, is willing to go to a nursing home for care, and believes that individuals should pay for their own long-term care services) should not be in the model because they are the results rather than the causes of the purchase. To satisfy this concern, we ran the model with and without these variables, and found no material differences in the coefficients of the remaining variables.

5. Of the 73 observations, 68 had a policy still in force, 32 purchased it during the Partnership, and 9 reported that it was a Partnership policy.

6. A Chow test indicates the -2 log likelihood of the difference between the restricted and the unrestricted (sum of the two equations) models is statistically significant at p [is less than] .0001.

REFERENCES

Arrow, K. 1963. "Uncertainty and the Welfare Economics of Medical Care." American Economic Review 33 (December): 384-416.

Atchley, R., and M. Dorman. 1994. "Gaining Market Insights from the Ohio Long-Term Care Insurance Survey." Journal of the American Society of CLU and ChFC 48 (5): 66-71.

Ball, R., and T. Bethell. 1989. Because We're All in This Together. Washington, DC: Families USA Foundation.

California Partnership for Long-Term Care. 1996. Quarterly Report No. 5, 1 October-31 December 1995. Sacramento, 18 March.

Cohen, M., N. Kumar, and S. Wallack. 1992. "Who Buys Long-Term Care Insurance?" Health Affairs 11 (spring): 208-23.