Do demographic characteristics moderate differences in don’t know responding between cognitive and affective risk questions?

Abstract

Background: Risk perception is a central construct in many health behavior theories. However, many people report uncertainty about their risk of cancer and other diseases. This reported uncertainty is higher for cognitive than affective risk perceptions. We explored whether this difference varied in magnitude by sociodemographic characteristics that are associated with differential reliance on affect in decision making: sex, age, education, race, numeracy, and family history. Methods: We used secondary data analysis (N=835) to compare the frequency of “don’t know†responses (DKR) to items assessing cognitive and affective perceived risk (e.g. “I feel…â€). Absolute and comparative risk perceptions of both colon cancer and other “exercise-related diseases†were assessed for each of the items. All eight items included an explicit DK response option. Multilevel logistic regression analyses explored sociodemographic characteristics as potential moderators. Findings: The odds of DKR were higher for cognitive than affective perceived risk (OR=0,64, 95%CI=0.54-0.75, p<0.001). This difference occurred for absolute but not comparative risk perceptions. No interactions with sociodemographic characteristics were found (ps>.05). Conclusions: Participants responded “don’t know†less often when asked to indicate their affective (versus cognitive) absolute perceived risk. This effect was consistent in magnitude across many sociodemographic and family history characteristics. Thus, interventions that help people understand the cognitive aspects of risk perceptions (e.g., probability likelihoods) may target the whole population rather than certain subsets. To inform those interventions, the mechanisms driving the difference in cognitive vs. affective DKR need to be identified.