Missed opportunity: Clinician recognition of overweight and obesity 10 years after the publication of NIH guideline

Background: Obesity and being overweight cost the United States an estimated 418,000 lives and $117 billion, annually. Two of every three American adults are overweight or obese. In 1998, the National Institutes of Health published the first national guideline on the assessment and management of overweight and obesity, recommending that clinicians determine the degree of obesity for each patient. Ten years after the NIH guideline, this study determines how well clinicians are diagnosing overweight or obesity and which patient and clinician factors affect the diagnosis.

Methods: This is a retrospective cohort study of patients aged 20-59 years who had their first medical appointment at the treating sites between 1 January and 31 December 2007. Chart reviews were conducted at three primary care clinics, following each patient for 12 months, to abstract age, gender, race, insurance status, height, weight, diagnosis of overweight or obesity, treatment recommendations, and co-morbid conditions of hypertension or diabetes. To test the hypothesis that less than 50% of overweight or obese patients are correctly identified by clinicians, a proportional stratified random sample size of 260 patient charts was determined by assigning a pretest probability of .05, a power of 80%, and .08 effect size. The Student's t-test was utilized to test the hypothesis, while McNemar and X² tests were employed to determine the association of co-factors with obesity identification.

Results: Clinicians identified less than a fifth (19.6%, p<.001) of overweight or obese patients. Increasing gradations of weight were correlated with an increase in identification; 0%, 12%, and 62% of overweight, obese, and morbidly obese patients were diagnosed, respectively. Patients with hypertension or diabetes were diagnosed more frequently (27%) than patients without these co-morbidities (9%) for an OR=3.6. Patients with a diagnosis of overweight or obesity were given a treatment recommendation 53% of the time versus only 7% for those without a diagnosis (OR=14.5, CI=3.1-68.4).

Conclusions: A minority of patients were appropriately diagnosed as being obese or overweight. The diagnosis was more commonly made at the extremes of weight and in patients with hypertension or diabetes. Treatment recommendations are rare without a diagnosis of overweight or obesity. Identification of overweight and obesity is the first step in combating this pervasive disorder that is associated with many chronic disease states.