Under-reporting remains a key limitation of self-reported dietary intake: an analysis of the 2008/09 New Zealand Adult Nutrition Survey

BACKGROUND/OBJECTIVES: The most recent New Zealand Adult Nutrition Survey 2008/09 (ANS 08/9) revealed a decrease in reported energy intakes (EIs) compared with the previous 1997 National Nutrition Survey (NNS97). Conversely, measured body weights increased over the same period. We conducted an analysis on the ANS 08/9 data sets to evaluate reported EIs. SUBJECTS/METHODS: Analysis was conducted on data from 3919 (1715 men and 2204 women aged >= 15 years) survey participants who completed the 24-h dietary recall in the ANS 08/9. Under-reporting was assessed using the ratio of reported EI to estimated resting metabolic rate (EI:RMRest), and a cutoff limit of <0.9 (EI:RMRest) was used to identify low energy reporters (LERs). Results were examined by gender, body size, age and ethnicity. RESULTS: The mean EI: RMRest (s.e.m.) was 1.34 (0.02) for men, and 1.23 (0.02) for women. Overall, 21% of men and 25% of women were classified as LERs. There was a greater prevalence of LERs among people with overweight (25%), or obesity (30%) than people with normal body weight (16%, P<0.001). The oldest age group (>= 65 years) had a greater prevalence of LERs (33%) compared with all other age groups (19-24%, P<0.001). Pacific people had a greater prevalence of LERs (33%) compared with Maori (26%, P = 0.007) and European (23%, P<0.001). Compared with the NNS97, a substantial increase in the prevalence of LERs was evident in most subgroups. CONCLUSIONS: Under-reporting of EI will continue to be a major limitation of nutrition surveys without technological innovation. Care should be taken when interpreting EI data.