OBJECTIVES: The Medicare Prescription Drug, Improvement, and Modernization Act of 2003 will provide drug coverage to seniors through a variety of mechanisms, including stand-alone prescription drug plans. Although these stand-alone plans must provide a standard minimum benefit package, variation in cost-sharing and utilization controls is permitted, leading potentially to a wide variety of plans from which beneficiaries might choose. Thus, it is important for both policy makers and health insurance firms to better understand the processes seniors use to choose among drug insurance plans. The objectives of this study were to assess and compare the use of a compensatory judgment model with two configural (conjunctive and disjunctive) judgment models in older adults' evaluations of drug benefit plans with varying attributes. METHODS: Three focus groups conducted with a total of 19 members of a local senior advocacy group suggested that copayment, premium, deductible, formulary use, and mail-service use were relevant plan attributes. A separate group of 32 seniors judged the suitability of forty-eight drug benefit profiles based on these attributes using a 5-point Likert-type scale. Three regression models, approximating compensatory, conjunctive, and disjunctive judgments, were fit by least-squares to each participant's judgments using plan attributes as predictors. Reliability of predictions from each model was assessed by comparison of multiple correlations. RESULTS: Mean multiple correlations were strong and ranged from 0.782 for the compensatory model to 0.789 for both configural models and were statistically indistinguishable. Multiple correlations were marginally higher for conjunctive or disjunctive models for a majority (62.5%) of subjects. CONCLUSIONS: Results suggest that both compensatory and configural judgment models may provide plausible models of information integration in older adults' assessments of prescription drug plans. Future research could use more fine-grained techniques (e.g. process tracing) to better understand this process.