Prediction of resting energy requirements in people taking weight-inducing antipsychotic medications

The purpose of the present study was to determine whether the regression equations published by Mifflin are suitable for the prediction of resting metabolic rate in people taking weight-inducing antipsychotic medications. Resting metabolic rate was measured using indirect calorimetry in 45 people with psychotic illnesses who had been taking atypical antipsychotic medications for more than six months. Predicted resting metabolic rate was calculated using the Mifflin, Harris-Benedict and Schofield equations. The limits of agreement method were used to compare measured and predicted resting metabolic rate. In men, the prediction equations significantly overestimated resting metabolic rate. There was a systematic variation in the bias using the Mifflin equation. The limits of agreement using the Harris-Benedict and Schofield equations were 2793 and 3014 kJ/day, respectively. In women, predictions of resting metabolic rate using the Harris-Benedict equation had the lowest limits of agreement (1008 kJ/day) compared with 1978 kJ for the Mifflin and 3157 kJ/day for the Schofield equations, respectively. The present study does not support the recent suggestion that the Mifflin equation is the most suitable prediction equation to determine resting metabolic rate in people taking antipsychotic medications. In men, it may be advisable to reduce the estimate provided using the Harris-Benedict or Schofield equations by 840 kJ/day. The Harris-Benedict equation provided the 'best estimate' of resting energy requirements in women taking antipsychotic medications.