A process examination of the framing effect in younger and older adult medical decision making

Older and younger adults are susceptible to a decisional bias when faced with medical decisions, which results in different treatment decisions when presented with survival or mortality data. Although this bias, termed the framing effect, has been demonstrated in multiple studies, no published studies have attempted to determine the decisional process older and younger adults engage in when presented with different information frames. The current study used a think-aloud procedure to examine decisional process differences in younger and older adults who did and did not demonstrate a framing effect. All participants were presented with two data formats (interval and cumulative probabilities) in both survival and mortality wording. Data were analyzed quantitatively to determine the presence of a framing effect. Think-aloud data were analyzed qualitatively to determine decisional process differences stratified by age and demonstration of the framing effect. Interpretation of results for the interval probability format was limited by typographical errors that were found to exist in the scenario after data collection was complete. No statistically significant framing effects were found for either format among both younger and older adults. Older and younger adults had relatively equal personal and vicarious experience with the decision. Qualitative analyses performed on a subset of the data suggest that, among those older adults who did not demonstrate the framing effect, there was a reliance on pertinent experience with the decision, and little reliance on the presented data. In contrast, among younger adults who demonstrated the framing effect, there was a tendency toward incomplete analysis of the data when younger adults had knowledge about the decision. Across age groups, those that demonstrated the framing effect were significantly more likely to reference the presented data. Older adults were significantly less likely to reference the presented data than younger adults. Younger and older adults did not differ significantly on time-to-decision or word count for the think-aloud transcripts. Results are discussed in terms of age-related differences in decision making processes