Portion size estimation in dietary assessment: a systematic review of existing tools, their strengths and limitations.

Abstract

Context: Over or under estimation of portion size leads to measurement error during dietary assessment. Objective: To identify portion size estimation elements (PSEEs) and evaluate their relative efficacy in relation to dietary assessment; and assess the quality of studies validating PSEEs. Data selection and extraction: Electronic databases, internet sites and cross-references of published records were searched, generating 16,801 initial records, from which 334 records were reviewed and 542 PSEEs were identified comprising: 5% one- dimensional (e.g. food guides), 46% two- dimensional (e.g. photographic atlases) and 49% three- dimensional tools (e.g. household utensils). Out of 334 studies, 21 validated a PSEE (compared PSEE to actual food amounts) and 13 compared PSEEs against other PSEEs. Conclusion: Quality assessment showed that only a few validation studies were of high quality. According to the findings of validation and comparison studies, food image-based PSEEs were more accurate compared to food models and household utensils. Key factors to consider when selecting a PSEE include efficiency of the PSEE and its applicability to targeted settings and populations.