Objective: To provide a framework for use of the doubly labelled water method to measure energy expenditure in order to validate dietary instruments for the assessment of energy. Design: Review and description of the use of doubly labelled water method for use as a biomarker for habitual energy intake. Results: The doubly labelled water method has a relative accuracy of 1% and withinsubject precision of 5 to 8%. Comparison of self-reported energy intake with energy expenditure demonstrated that over one-third of individuals may underreport energy intake by more than 25%. Conclusions: The doubly labelled water method, although expensive and dependent on non-routine laboratory instrumentation, is an excellent biomarker of energy intake.
Keywords
Energy metabolism Doubly labelled water Stable isotopes Nutritional epidemiologyThe assessment of dietary intake plays a vital role in many aspects of nutritional science and, not surprisingly, a variety of dietary assessment instruments have been developed; including weighed food records, diet histories, 24-hour recalls and food-frequency questionnaires, each with many variations to suit particular investigative situations 1 . As these survey instruments have been developed and modified, investigators have tested their validity. These have included measurements of repeatability and accuracy. The latter, however, have usually involved comparison with a second survey instrument that has a known history of use and an assumed level of accuracy. Thus, while the testing procedure provided estimates of precision, the accuracy was not absolute, but rather a relative accuracy that could not detect biases that might be inherent in both methods. Such biases could include errors in the methodology used to convert a food to its nutrient values or systematic reporting bias by the participant.There have, however, been a few attempts to validate accuracy using more objective measures of dietary intake. As reviewed by Bingham 2 , urinary nitrogen has been used as a biomarker to test the accuracy of self-reported protein intake for over 75 years. Although some dietary instruments, primarily diet histories, have demonstrated good agreement between reported protein intake and urinary nitrogen, most validations have demonstrated modest underreporting. Some validations, however, have reported large discrepancies, with the largest underreport found amongst obese women, who underreported their protein intake by 50% 3 . Other validations have been performed using direct observation as the criterion method for validating intake instruments. Krall and Dwyer 4 asked participants to complete 3-day diaries and a 1-week food-frequency questionnaire (FFQ) during an in-patient period in which the food provided was monitored carefully and found that energy, macronutrients and vitamins A and C were underreported on the FFQ. Very recently, Schaefer et al. 5 had subjects complete an FFQ during trials in which participants were provided with high-or low-fat diets. The FFQ was found to underestimate fa...