Dietary under-reporting is a common problem when using self-reported dietary assessment tools. However, there are few studies regarding under-reporting during pregnancy. This study aimed to explore the demographic and psychosocial characteristics related to dietary under-reporting in pregnant Japanese women. A cross-sectional study was conducted between 2010 and 2011 at a university hospital in Tokyo, Japan. Nutrient intake was assessed using a self-administered Diet History Questionnaire (DHQ), which had questions about the consumption frequency and portion size of selected food items. The 24-h urinary excretion levels of urea N and K were used as the dietary protein and K intake reference values, respectively. Under-reporting of protein and K was defined as the bottom 25 % of the reporting accuracy (the ratio of reported intake on the DHQ to the estimated intake based on urinary excretion). Under-reporters were defined as participants who under-reported both protein and K intake. Multiple logistic regression analysis was performed to examine the factors associated with under-reporters. Of 271 healthy women at 19–23 weeks of gestation, thirty-five participants (12·9 %) were identified as under-reporters. Under-reporters had a lower pre-pregnancy BMI (adjusted OR (AOR) = 0·81) and lower gestational weight gain (AOR = 0·82); they also reported managing their gestational weight gain with the aim to return to their pre-pregnancy weight soon after childbirth (AOR = 2·99). Healthcare professionals should consider the potential for dietary under-reporting and the possible related factors when assessing the dietary intakes of pregnant Japanese women using self-administered questionnaires.

An analysis of dietary survey data from Sweden and Ireland revealed that with a declining ratio of energy intake to estimated basal metabolic rate (EI : BMR), %energy from fat, saturated fat and carbohydrate fell while that of protein increased. However, where the definition of under-reporting is less strict, as in the Irish database (EI : BMR) < (1·27), the quantitative effect on macronutrient balance, while statistically significant, is not of such magnitude as to be of nutritional significance. However, with respect to foods, under-reporting was found to be of considerable significance in using such data to formulate food-based dietary guidelines. The Swedish data show that under-reporting of foods is associated with a decreasing percentage of consumers, decreasing intake per eating occasion and decreasing frequency of eating. The development of food-based dietary guidelines will involve comparisons of patterns of food intake among people with contrasting high or low intakes of target nutrients. The Swedish data compared food intakes for those with low or high intakes of saturated fatty acids (SFA) among under-reporters and acceptable reporters of energy intake. Whereas absolute values were always lower with the lower EI : BMR group (< 1·1) compared to the higher EI : BMR group (> 1·35) the ratio of intakes for the low vs. high SFA groups were broadly significant at either EI : BMR ratio. This paper highlights the difficulties that under-reporting will pose for food-based dietary guidelines.