NOT MY CHILD: DOES OPTIMISM BIAS INFLUENCE PARENT PREDICTIONS OF CHILDHOOD-OBESITY RELATED DISEASE RISKS?

Purpose:  Most parents believe childhood obesity is a problem for society, but not for their own children. We sought to understand whether parents’ risk assessment was skewed by optimism bias, the tendency to overestimate one’s chances of experiencing unlikely positive events.

Method: We administered a national web-based survey to parents of 5-12 year old children (n = 545) though the U.S. GfK KnowledgePanel®. We asked parents to estimate the chances that (a) their child and (b) “a typical child in their community” would be overweight or obese, and develop hypertension, heart disease, and/or diabetes in adulthood using a visual analog scale. We collected data on family demographic and health characteristics, and parent numeracy. We used a microsimulation model informed by survey responses and epidemiologic data to simulate the expected BMI class and probabilities of developing obesity-related co-morbidities in adulthood for the respondent’s child and the “typical child.” We used logistic and linear regression models with mixed effects to evaluate whether optimism bias was present in parent predictions of a child’s future health outcomes.

Results: We found 273 (50%) parents predicted their child would be the right weight, but that a typical child would be overweight in adulthood. Of parents who predicted their child would be overweight in adulthood (n= 109), 93% predicted the typical child would also be overweight in adulthood. Parents had 33 times lower adjusted odds (p< 0.001) of predicting that their child (versus a typical child) would be overweight or obese in adulthood. Current child obesity, parental obesity, high numeracy, among other factors increased the odds that a parent would predict future overweight or obesity for their own child (p < 0.05).  Controlling for health and demographic characteristics, parents estimated that their children’s chances of developing hypertension, heart disease, and diabetes would be 14.5, 12.8, and 14.8 percentage points lower than that of a typical child in their community, respectively (all p < 0.001).

Conclusion: Parent biases around estimating childhood obesity-related health risks are skewed by optimism bias and low numeracy, among other demographic characteristics. Accurate parental perception of a child’s objective obesity-related health risks could motivate parents to engage in behavior change. Researchers should develop risk communication approaches that can compensate for these factors and improve parent risk assessment.