Healthcare provider accuracy at estimating women's BMI and intent to provide counseling based on appearance alone

Abstract

To assess healthcare providers' ability to estimate women's body mass index (BMI) based on physical appearance and determine the prevalence of, and barriers to, weight-related counseling. A web-based survey was distributed to healthcare providers ("participants") at a university-based hospital and contained photographs of anonymous women ("photographed women (PW)") as well as questions regarding participant demographics. Participants were asked to estimate BMI category based on physical appearance, state whether they would provide weight-loss counseling for each PW and identify barriers to counseling. One hundred forty-two participants completed the survey. BMI estimations were poor among all participants, with an overall accuracy of only 41% and a large proportion of underestimations. Standardization of PW clothing did not improve accuracy; 41% for own clothing versus 40% for scrubs, P = 0.2. BMI assessments were more accurate for Caucasian versus African American PW (45% versus 36%, P < 0.001) and PW with normal weight (84%) and obesity III (38%) compared to PW with mid-range BMI (P < 0.001). Although the frequency of weight loss counseling was positively associated with PW BMI, participants only intended to counsel 69% of overweight and obese PW. The most commonly cited reason for lack of counseling was time constraints (54%). Healthcare providers are inaccurate at appearance-based BMI categorization and thus, BMI should be routinely calculated in order to improve identification of those in need of counseling. When appropriately identified, time constraints may prevent practitioners from providing appropriate weight-loss counseling-further complicating the already difficult task of fighting obesity.