Early experience with employee choice of consumer-directed health plans and satisfaction with enrollment

Health Services Research, August, 2004 by Jinnet Briggs Fowles, Elizabeth A. Kind, Barbara L. Braun, John Bertko

The Satisfaction with Enrollment Model Using a multivariate logistic regression, we modeled responses to each of four attributes of the enrollment process: helpfulness of communications in preparing for enrollment, having enough time to review enrollment information and enroll at work, understandability of benefits information, and ease of finding needed information. The independent variables used in these analyses were education, race, and health status.

FINDINGS

The response rate was 66.2 percent. Using administrative data to compare respondents with nonrespondents, we found that respondents were significantly different from nonrespondents on several characteristics: respondents were older (mean age 40 years versus 35 years), more likely to hold exempt positions (56 percent versus 37 percent), less likely to have employee-only coverage (38 percent versus 43 percent), or to enroll in the HMO option (29 percent versus 40 percent).

Question 1: How Did Those Who Chose the Consumer-Defined Health Plans Compare with Those Who Did Not?

Two-hundred-four employees selected one of the CDHP options (7.3 percent). At the bivariate level (Table 1), the employees who selected the CDHPs differed from those who selected other plan options in socio-demographic characteristics, health status, health care utilization, preferences for plan attributes, and responses to the enrollment process. Those selecting the CDHPs were more often college educated, white, male, and in exempt positions than employees who selected other plan options. They more frequently had employee-only coverage from Humana and also additional coverage from another source. Those selecting a CDHP option were significantly healthier on every dimension measured. They more often reported excellent health status, and less often had a covered member receiving regular medical treatment. They less often had a personal physician. Although they less frequently believed that the health plan decision was extremely important, they more often found the decision difficult, probably because of the novelty of the choice. Those who selected a CDHP option more frequently rated premiums as the most important feature of the plan. They more often used the decision support tool and agreed with how it ranked the plan options. An analysis of the comments made by those who selected the CDHPs reflected the widespread need to have more detailed information about these novel options. A common concern was how the initial $500 allowance would be calculated.

Results from Multivariate Logistic Regression Analysis. In the multivariate analysis, variables from each of the four domains (sociodemographic characteristics, health status, health care utilization, and perceived importance of plan attributes) were related to plan choice (Table 2). Among the sociodemographic characteristics, employees who were black were half as likely to select the CDHP options (OR 0.46). Those having only Humana Inc. coverage were also less likely to select the CDHP options (OR 0.30). In contrast, those with exempt job classifications and those electing employee-only coverage were more likely to select the CDHP options (OR 1.61 and 1.77, respectively).