OBJECTIVE: To assess the initial impact of offering consumer-defined health plan (CDHP) options on employees. DATA SOURCES/STUDY SETTING: A mail survey of 4,680 employees in the corporate offices of Humana Inc. in June 2001. STUDY DESIGN: The study was a cross-sectional mail survey of employees aged 18 and older who were eligible for health care benefits. The survey was conducted following open enrollment. The primary outcome is the choice of consumer-directed health plan or not; the secondary outcome is satisfaction with the enrollment process. Important covariates include sociodemographic characteristics (age, gender, race, educational level, exempt or nonexempt status, type of coverage), health status, health care utilization, and plan design preferences. DATA COLLECTION METHODS: A six-page questionnaire was mailed to the home of each employee, followed by a reminder postcard and two subsequent mailings to nonrespondents. PRINCIPAL FINDINGS: The response rate was 66.2 percent. Seven percent selected one of the two new plan options. Because there were no meaningful differences between employees choosing either of the two new options, these groups were combined in multivariate analysis. A logistic regression modeled the likelihood of choosing the novel plan options. Those selecting the new plans were less likely to be black (odds ratio [OR] 0.46), less likely to have only Humana coverage (OR 0.30), and more likely to have single coverage (OR 1.77). They were less likely to have a chronic health problem (OR 0.56) and more likely to have had no recent medical visits (OR 3.21). They were more likely to believe that lowest premiums were the most important plan attribute (OR 2.89) and to think there were big differences in the premiums of available plans (OR 5.19). Employees in fair or poor health were more likely to have a difficult time during the online enrollment process. They were more likely to find the communications very helpful (OR 0.42) and the benefits information very understandable (OR 0.38). They were less likely to feel that they had enough time to make their enrollment decision (OR 0.47). CONCLUSIONS: Employees who were attracted to the new CDHP plan options valued the attributes that distinguished these plans from other choices. The shift to consumer-defined plans and to the electronic provision of information, however, requires a significant increase in the communication support for all employees, but particularly for those in fair or poor health whose information needs are the most complex and individualized.