Choosing a Doctor: Does Presentation Format Affect the Way Consumers Use Health Care Performance Information?

Background Choosing a new health service provider can be difficult and is dependent on the type and clarity of the information available. This study examines if the presentation of service quality information affects the decisions of consumers choosing a general medical practice. Objectives The aim was to examine the impact of presentation format on attribute level interpretation and relative importance. Methods A discrete choice experiment eliciting preferences for a general medical practice was conducted using four different presentation formats for service quality attributes: (1) frequency and percentage with an icon array, (2) star ratings, (3) star ratings with a text benchmark, and (4) percentage alone. A total of 1208 respondents from an online panel were randomised to see two formats, answering nine choices for each, where one was a dominated choice. Logistic regression was used to assess the impact of presentation format on the probability of choosing a dominated alternative. A generalised multinomial logit model was used to estimate the relative importance of the attribute levels. Results The probability of incorrectly choosing a dominated alternative was significantly higher when the quality information was presented as a percentage relative to a frequency with icon array, star rating or bench-marked star rating. Preferences for a practice did not differ significantly by presentation format, nor did the probability of finding the information difficult to understand. Conclusions Quantitative health service quality information will be more useful to consumers if presented by combining the numerical information with a graphic, or using a star rating if appropriate for the context.