Relative Validity, Reliability, and Acceptability of an Image-Voice Dietary Assessment System for Estimating Individual Nutrient Intakes Intake in Cambodian Women and Children. (Preprint)

Data on individual level food and nutrient intakes is fundamental to informing development of nutrition policy and programs. However, in low- and lower-middle-income countries (LLMICs), proxy measures of individual intake (e.g. household consumption and expenditure surveys, food balance sheets) are often used, with limited implementation of new technology-assisted applications such as image-based food records. The Voice-Image Solution for Individual Dietary Assessment (VISIDA) is a new system designed specifically for use in LLMICs. To determine the relative validity, test-retest reliability and acceptability of the VISIDA system in a sample of Cambodian women and children (aged ≤5 years). Mother-child dyads were recruited from three locations (rural, semi-rural and urban) in Siem Reap province in Cambodia. Over a ~4-week period, dietary intake data was collected for each participant using two methods over three recording periods. In the first week, intake was recorded using the VISIDA image-voice smartphone app for three non-consecutive days, followed by three 24-hour recalls collected in weeks 2-3, and the VISIDA app used again to collect a 3-day record in week 4. Following completion of the third intake recording period, the participating mother completed a feedback survey. Differences between the estimated nutrient intakes for the three recordings periods for mothers and children were examined using a linear mixed model approach. 210 participants comprised of 119 mothers and 91 children were included in the analysis. Estimated mean nutrient intakes reported in both VISIDA recording periods were mostly lower compared to intakes reported using the 24-hour recall. Compared to 24-hour recalls, statistically significant differences were found for the VISIDA recording periods for 16 nutrients (out of 20 nutrients) for mothers and six nutrients (out of the 19 nutrients) for children. Nutrient intakes estimated from both VISIDA recording periods were not statistically significantly different for mothers and children. The majority of mothers reported the VISIDA smartphone app was ‘easy to use’ (63.0%) or ‘very easy to use’ (21.3%) for collecting dietary intake data. The VISIDA system produced lower estimates of nutrient intakes when compared to 24-hour recalls in a sample of mothers and children in Siem Reap province, Cambodia. However, estimated nutrient intakes for the two VISIDA recording periods were similar. Participating mothers reported high acceptability towards using the VISIDA smartphone app to collect intake data for themselves and their participating child.