Measurement issues in the evaluation of chronic disease self-management programs

Abstract

To provide an in-depth analysis of outcome measures used in the evaluation of chronic disease self-management programs consistent with the Stanford curricula. Based on a systematic review on self-management programs, effect sizes derived from reported outcome measures are categorized according to the quality of life appraisal model developed by Schwartz and Rapkin which classifies outcomes from performance-based measures (e.g., clinical outcomes) to evaluation-based measures (e.g., emotional well-being). The majority of outcomes assessed in self-management trials are based on evaluation-based methods. Overall, effects on knowledge--the only performance-based measure observed in selected trials--are generally medium to large. In contrast, substantially more inconsistent results are found for both perception- and evaluation-based measures that mostly range between nil and small positive effects. Effectiveness of self-management interventions and resulting recommendations for health policy makers are most frequently derived from highly variable evaluation-based measures, that is, types of outcomes that potentially carry a substantial amount of measurement error and/or bias such as response shift. Therefore, decisions regarding the value and efficacy of chronic disease self-management programs need to be interpreted with care. More research, especially qualitative studies, is needed to unravel cognitive processes and the role of response shift bias in the measurement of change.