Association of Energy Intake With the Lack of in-Person Review of Household Dietary Records: Analysis of Japan National Health and Nutrition Surveys From 1997 to 2011.

Background: National surveys have demonstrated a long-term decrease in mean energy intake in Japan, despite the absence of a decrease in the prevalence of overweight and obesity. We aimed to examine whether total energy intake of survey respondents is associated with completion of an in-person review of dietary records and whether it affects the trend in mean energy intake.
Methods: We pooled data from individuals aged 20-89 years from the National Nutrition Surveys of 1997-2002 and the National Health and Nutrition Surveys of 2003-2011. We conducted a linear mixed-effects regression to estimate the association between total energy intake and the lack of an in-person review of semi-weighed household dietary records with interviewers. As some respondents did not have their dietary data confirmed, we used regression coefficients to correct their total energy intake.
Results: Compared with respondents completing an in-person review, total energy intake was significantly inversely associated with respondents not completing a review across all sex and age groups (P < 0.001). After correction of total energy intake for those not completing a review, mean energy intake in each survey year significantly increased by 2.1%-3.9% in men and 1.3%-2.6% in women (P < 0.001), but the decreasing trend in mean energy intake was sustained.
Conclusions: Total energy intake may be underestimated without an in-person review of dietary records. Further efforts to facilitate completion of a review may improve accuracy of these data. However, the increasing proportion of respondents missing an in-person review had little impact on the decreasing mean caloric intake.