Stable isotopes are non-radioactive, safe and are applied for various purposes in human health assessment in trace amounts that minimally disturb normal physiology. The International Atomic Energy Agency supports the use of stable isotope techniques to design and evaluate interventions addressing malnutrition in all its forms with focus on infant and young child feeding; maternal and adolescent nutrition; diet quality; prevention and control of non-communicable diseases; healthy ageing and gut function. These techniques can be used to objectively measure: (1) amount of human milk consumed and whether an infant is exclusively breastfed; (2) body composition in the context of re-feeding programmes for moderate and severe acute malnutrition and as an indicator of the risk for obesity; (3) bioavailability and bioconversion of pro-vitamin A and vitamin A body stores following vitamin A intervention programmes; (4) absorption and retention of iron, zinc and protein; (5) total energy expenditure for validation of physical activity measurement and dietary assessment tools and (6) diagnosis of Helicobacter pylori. Stable isotope techniques will be invaluable in the tracking of global targets on exclusive breast-feeding childhood obesity and anaemia among women. Efforts are underway to make nuclear techniques more affordable, field-friendly and less invasive, and to develop less sophisticated but precise equipment. Advocacy for the wide adoption of the techniques is needed.

Population exposure to food and nutrients can be estimated from household food purchases, but store surveys of foods and their composition are more available, less costly and might provide similar information. Our aim was to compare estimates of nutrient exposure from a store survey of packaged food with those from household panel food purchases. A cross-sectional store survey of all packaged foods for sale in two major supermarkets was undertaken in Auckland, New Zealand, between February and May 2012. Longitudinal household food purchase data (November 2011 to October 2012) were obtained from the nationally representative, population-weighted New Zealand Nielsen HomeScan® panel. Data on 8440 packaged food and non-alcoholic beverage products were collected in the store survey. Food purchase data were available for 1229 households and 16 812 products. Store survey data alone produced higher estimates of exposure to Na and sugar compared with estimates from household panel food purchases. The estimated mean difference in exposure to Na was 94 (95 % CI 72, 115) mg/100 g (20 % relative difference; P<0·01), to sugar 1·6 (95 % CI 0·8, 2·5) g/100 g (11 %; P<0·01), to SFA −0·3 (95 % CI −0·8, 0·3) g/100 g (6 %; P=0·3) and to energy −18 (−71, 35) kJ/100 g (2 %; P=0·51). Compared with household panel food purchases, store survey data provided a reasonable estimate of average population exposure to key nutrients from packaged foods. However, caution should be exercised in using such data to estimate population exposure to Na and sugar and in generalising these findings to other countries, as well as over time.