BACKGROUND AND OBJECTIVES: Documenting obesity on the problem list has been shown to promote action about obesity and overweight, yet a majority of primary care providers do not record obesity on the medical problem list. With this in mind, our objectives were to determine the proportion of physicians' documentation of overweight (OW) or obesity on the problem list in our primary care teaching practice and to identify predictors of physician documentation of OW/obesity.

METHODS: De-identified health records of 6,195 adult patients with BMI ? 25 kg/m2 seen by a family physician over a 2-year period were included. Using multivariate logistic regression, patient age, BMI, gender, race, insurance, comorbidities, number of visits, physician gender or role, and practice site (suburban versus urban) were examined in relation to inclusion of OW/obesity on the medical problem list.

RESULTS: Few (21.1%) patients had OW/obesity on their problem list. In the multivariate model, female PCPs were significantly more likely to document OW/obesity (OR=1.39, 95% CI=1.18--1.63) compared to male PCPs, and faculty were 26% more likely to document obesity (95% CI=1.07--1.48) compared to residents. Female patients, those with hypertension, diabetes, hyperlipidemia, and those with six or more visits were significantly more likely to have obesity on their problem lists, while patients with Medicaid were less likely to have obesity recorded. No significant difference was seen by race.

CONCLUSIONS: Nearly 80% of OW and obese patients were not identified on the problem list. Patient gender, comorbidity, and number of visits were associated with documentation. Future research should examine automatic documentation of OW/obesity on the medical problem list.