Validation of self-reported energy intake by a self-administered diet history questionnaire using the doubly labeled water method in 140 Japanese adults

Abstract

To validate reported energy intake (rEI) with a self-administered diet history questionnaire (DHQ) against total energy expenditure (TEE) by the doubly labeled water (DLW) method. A total of 140 healthy Japanese adults (67 men and 73 women) aged 20-59 years living in four areas in Japan. Energy intake was assessed twice with DHQ over a 1-month period before and after TEE measurement (rEI(DHQ1) and rEI(DHQ2), respectively). TEE was measured by DLW during 2 weeks (TEE(DLW)). Mean rEI(DHQ1) was lower than those of TEE(DLW) by 1.9+/-2.4 MJ/day (16.4%, P<0.001) for men and 0.6+/-1.9 MJ/day (6.0%, P<0.01) for women. In men and women together, 62 subjects (44%) were defined as underreporters (rEI(DHQ1)/TEE(DLW) <0.84), 58 (41%) as acceptable reporters (0.84-1.16) and 20 (14%) as over-reporters (>1.16). Pearson correlation coefficient was 0.34 for men and 0.22 for women. After adjustment for the dietary and non-dietary factors related to rEI(DHQ1)/TEE(DLW), the correlation coefficient improved to 0.42 and 0.37, respectively. The energy intake assessed with DHQ correlated low to modestly with TEE measured by DLW. In addition, DHQ underestimated energy intake at a group level. Caution is needed when energy intake was evaluated by DHQ at both individual and group levels.