Article Text

Download PDFPDF

OP41 Evaluation of a natural experiment to increase availability of healthier snack foods in vending machines using interrupted time series analysis
  1. Charlotte Evans
  1. Food Science and Nutrition, University of Leeds, Leeds, UK


Background The food environment plays a key role in access and availability of healthier food and drink choices. Community based vending machines are one source of energy dense snack foods high in saturated fats, sugars and salts, particularly for young people. The aim of this research was to evaluate the increased availability of healthier options of different types of snack products sold in vending machines situated in English leisure centres. The natural experiment was managed by Leeds city council with vending provision by Wilkes vending company and support from Public Health England (PHE).

Methods The intervention sought to increase the availability of healthier food items within 18 vending machines in Leeds leisure centres over three iterative phases during 39 weeks between September 2018 and May 2019. Products were altered to meet agreed nutrition criteria at each phase based on the Government Buying Standards for Food and Catering Services (GBSF) and taking into consideration calorie guidelines within PHE’s sugar reduction programme, and product availability. A quasi experimental trial design was used with interrupted time series analysis and segmented regression techniques adjusting for underlying trends, correlation between data points and holiday weeks when footfall is generally lower. The primary outcomes were the changes in level and trend of weekly purchased energy in phase 1 and phase 2. Secondary outcomes included phase changes in levels and trends for saturated fats, sugars, salt and purchased units.

Results Total energy, saturated fats, sugars and salt from products purchased all significantly reduced in phase 1 compared with baseline and although did not reduce further at phase 2 when more stringent standards were implemented, nutrients purchased continued to be lower compared with baseline. Trend changes during phase 1 were non-significant for all nutrients; however, during phase 2 there were upward trends for energy, saturated fats and salt. The total weekly amount of energy purchased by all machines reduced by 244,256kcal (95% CI -399,456 to -89,057 kcal), between baseline and phase 1. The reductions for saturated fats, sugars and salt were 4,917g (95%CI -8884 to -950g), 23,868g (95%CI -38,369 to -9,366g) and 425g(95%CI -644 to -207g) respectively. Weekly sales varied over time and were lower in phase 1 but recovered in phase 2.

Conclusion It is feasible to introduce standards to increase the availability of healthier snack products in vending machines and assess changes in purchased snacks without substantially affecting sales.

  • food environment
  • energy dense snacks
  • interrupted time series analysis

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.