Performance of Body Mass Index in Identifying Obesity Defined by Body Fat Percentage and Hypertension Among Malaysian Population: A Retrospective Study

PURPOSE: Body mass index (BMI) is used universally to define obesity. Many studies have indicated that the current BMI cutoff value for obesity may be inaccurate in identifying individuals with excess body fat (BF) and at risk for cardiovascular diseases (CVD). This study aims to assess the performance of BMI in diagnosing obesity defined by BF percentage (BF%). PATIENTS AND METHODS: A total of 136 participants who attended an annual health screening programme were recruited. The subjects completed the health examinations, including BMI, BF% and blood pressure measurement. A receiver operating curve (ROC) analysis was conducted to determine the optimal cutoff value of BMI in classifying obesity based on BF% (>25%). RESULTS: The ROC analysis revealed that the optimal BMI cutoff value in classifying subjects with obesity based on BF% was 24.8 kg/m(2). The agreement between the classification scheme based on the new BMI cutoff (>24.8 kg/m(2)) and BF% was higher (κ=0.722) compared to the standard BMI cutoff (>27.5 kg/m(2)) (κ=0.532). BMI 24.8 kg/m(2) also had higher sensitivity (80.0%) than 27.5 kg/m(2) (56.0%) in detecting subjects with high adiposity. The new BMI cutoff also showed a sensitivity of 63.9% in identifying subjects with hypertension compared to the standard cutoff (36.1%). CONCLUSION: The current definition of obesity based on BMI value needs to be reassessed by taking BF% into account. A new BMI cutoff point, 24.8 kg/m(2) for obesity, can identify a higher percentage of Malaysian at risk for CVD.