TY - JOUR T1 - GP documentation of obesity: what does it achieve? JF - British Journal of General Practice JO - Br J Gen Pract SP - 890 LP - 894 VL - 48 IS - 426 AU - P Little Y1 - 1998/01/01 UR - http://bjgp.org/content/48/426/890.abstract N2 - BACKGROUND: Obesity is a major and increasing health problem in the United Kingdom, and, until recently, the government health promotion package for general practice reimbursed general practitioners for documenting obesity. Despite poor evidence for effectiveness of interventions in primary care, documentation of obesity could possibly improve patient awareness and knowledge, or provide public health information. AIM: To assess patient perception of obesity and its health risk, and the accuracy of estimating obesity using patient information. METHOD: Subjects were consecutive attenders to a general practitioner (GP) at a single urban practice in the South and West Region. Outcome measures were 'measured' body mass index (BMI) calculated from measured weight and height, 'estimated' BMI using patient information, and patient perception of obesity and the health risks of obesity. RESULTS: There is good correlation between 'estimated' and 'measured' BMI (intraclass correlation 0.91). Estimated BMI is lower than measured BMI (mean 0.77 lower), and the difference increases with age and level of BMI: for BMIs of < 20, 20-24.99, 25-29.99, and > or = 30 the mean differences (estimated-measured) were -0.06, -0.46, -0.98 and -1.72 respectively. Estimated obesity (BMI > 30) is reasonably sensitive (70%), specific (99%), and predictive (93% positive predictive value) of measured obesity (kappa 0.78). All obese subjects are aware that they are overweight, and most of them (78%, 95% confidence interval 66-88%) are aware that their weight is a health risk. CONCLUSION: Obese patients attending GPs' surgeries are likely to know if they are overweight, or could easily estimate from their knowledge of height and weight that they were overweight with reasonable accuracy. Obese subjects also know that their weight carries health risks. Thus, measurement of obesity in the general population is not likely to improve risk assessment or patient knowledge significantly. Without evidence for effective intervention or improved decision-making in primary care, reimbursement guidelines linked to the documentation of obesity in the population are probably an inefficient use of resources. ER -