Objective: The aim of the study was to examine the daily variations in sales data for individual food items in a supermarket and to assess the usefulness of the computerized sales data of supermarkets for reliable monitoring and evaluation of shopping behaviour.
Design: Longitudinal observational study.
Setting: The study was carried out in one supermarket in Mikkeli, Finland. Seventy-nine packed food items from food groups important for salt and fat intake were monitored. In all food groups both 'healthier' and 'reference' products were included for assessment of both direct sales and proportional sales. The sales data were collected daily for 2 months in May and September 1996 by reading the European Article Numbering (EAN) codes of the packed foods.
Results: The proportional sales turned out to be a more stable and useful measure than the direct sales data and the variation remained the same when the monitoring time was increased from 1 week to 1 month.
Conclusion: Proportional sales data are proposed as a tool for measuring the effect of nutrition interventions and also as a possible indirect assessment for population salt and fat intake.