IMPORTANT: there are later versions of this dataset. Please find updated versions at https://doi.org/10.17863/CAM.78207 (version 2) and https://doi.org/10.17863/CAM.91760 (version 3)

Longitudinal dietary data for the United Kingdom (UK) on food supply, provided by FAO food balance sheets (1961-2013) (FAO-FBS ), and food purchases, provided Defra household budget surveys (National Food Survey [1942-2000] and Family Food Module [2001-2018]) (Defra-HBS).

Studying dietary trends can shed light on progress towards healthier and more sustainable diets but longitudinal data are often confounded by lack of standardized methods. Two main data sources are often used for longitudinal analysis of diets: food balance sheets (per capita food supply estimated from production and trade data) and household budget surveys (household surveys on food purchased). The impact of these different collection methods has not been quantified for the UK.

The data provided here were used to assess how trends in dietary change compared between the two collection methods for calories, meat and fish, nuts and pulses, and dairy, and how disparities between FAO-FBS and Defra-HBS have changed over time. These food types are comparable between FAO-FBS and Defra-HBS and can be used to monitor consumption and protein intake.

The primary differences in quantities estimated by FAO-FBS and Defra-HBS occur in part due to inclusion of retail waste in FAO data and likely under-reporting of consumption in Defra data. FAO-FBS and Defra-HBS were significantly (p<0.001) different for all food types and calorie intake. For all except nuts and pulses, FAO estimates were higher than Defra, and differences increased over time. Both showed an increase in per capita meat supply (1961-2013) and purchases (1942-2018). However, for all other food types and calorie intake, temporal trends did not agree between the two sources.

FAO-FBS differed significantly from Defra-HBS. For most food types, this difference increased between 1961 and 2013. This raises questions over the robustness of both data sources for monitoring UK dietary change, especially when used for evidence-based decision making. Further research into reasons for increasing discrepancies is needed.

UK National Food Survey data for 1942-1973 were downloaded on 08/01/2020 from https://webarchive.nationalarchives.gov.uk/ukgwa/20130103024837/http://www.defra.gov.uk/statistics/foodfarm/food/familyfood/nationalfoodsurvey/

National Food Survey data are named 'Household nutrient data from 1940 to 2000 – 1940-2000' and 'Household consumption of selected foods from 1942 to 2000 – 1942-2000'.

UK Family Food data for 1974 onwards were first downloaded on 19/02/2020 from https://www.gov.uk/government/statistical-data-sets/family-food-datasets.

Datasets are named 'UK - household purchases', 'UK - eating out purchases' and 'UK - household and eating out nutrient intakes'.

FAO data between 1961 and 2013 were first downloaded on 08/01/2020 from

http://www.fao.org/faostat/en/#data/FBSH titled 'Food Balances (-2013, old methodology and population)'.

'FAOSTAT - food supply' was downloaded by selecting 'United Kingdom of Great Britain and Northern Island' from 'Countries' and 'Food supply quantity (kg/capita/yr)' from 'Elements' between 1961 and 2013 for the items listed in Table 1 of 'Discrepancies between two long-term dietary datasets in the United Kingdom (UK).' available here (insert Wellcome Open Research link).

'FAOSTAT - calorie supply' was downloaded by selecting 'United Kingdom of Great Britain and Northern Island' from 'Countries', 'Food supply quantity (kcal/capita/day)' from 'Elements', and 'Grand Total + (Total) from 'Items (aggregated)' between 1961 and 2013.

This data supports the publication titled 'Discrepancies between two long-term dietary datasets in the United Kingdom (UK).'. 'Figure 2 - data', 'Figure 3 - data', and 'Table 2 - data' contain data used for constructing the figures and tables of this publication.