Grocery purchase data in the study of alcohol use - A validity study

Tensorial Principal Component Analysis in Detecting Temporal Trajectories of Purchase Patterns in Loyalty Card Data: Retrospective Cohort Study

BACKGROUND

Loyalty card data automatically collected by retailers provide an excellent source for evaluating health-related purchase behavior of customers. The data comprise information on every grocery purchase, including expenditures on product groups and the time of purchase for each customer. Such data where customers have an expenditure value for every product group for each time can be formulated as 3D tensorial data.

OBJECTIVE

This study aimed to use the modern tensorial principal component analysis (PCA) method to uncover the characteristics of health-related purchase patterns from loyalty card data. Another aim was to identify card holders with distinct purchase patterns. We also considered the interpretation, advantages, and challenges of tensorial PCA compared with standard PCA.

METHODS

Loyalty card program members from the largest retailer in Finland were invited to participate in this study. Our LoCard data consist of the purchases of 7251 card holders who consented to the use of their data from the year 2016. The purchases were reclassified into 55 product groups and aggregated across 52 weeks. The data were then analyzed using tensorial PCA, allowing us to effectively reduce the time and product group-wise dimensions simultaneously. The augmentation method was used for selecting the suitable number of principal components for the analysis.

RESULTS

Using tensorial PCA, we were able to systematically search for typical food purchasing patterns across time and product groups as well as detect different purchasing behaviors across groups of card holders. For example, we identified customers who purchased large amounts of meat products and separated them further into groups based on time profiles, that is, customers whose purchases of meat remained stable, increased, or decreased throughout the year or varied between seasons of the year.

CONCLUSIONS

Using tensorial PCA, we can effectively examine customers' purchasing behavior in more detail than with traditional methods because it can handle time and product group dimensions simultaneously. When interpreting the results, both time and product dimensions must be considered. In further analyses, these time and product groups can be directly associated with additional consumer characteristics such as socioeconomic and demographic predictors of dietary patterns. In addition, they can be linked to external factors that impact grocery purchases such as inflation and unexpected pandemics. This enables us to identify what types of people have specific purchasing patterns, which can help in the development of ways in which consumers can be steered toward making healthier food choices.

Assessing household lifestyle exposures from consumer purchases, the My Purchases cohort

Consumer purchase data (CPD) is a promising instrument to assess the impact of purchases on health, but is limited by the need for manual scanning, a lack of access to data from multiple retailers, and limited information on product data and health outcomes. Here we describe the My Purchases cohort, a web-app enabled, prospective collection of CPD, covering several large retail chains in Denmark, that enables linkage to health outcomes. The cohort included 459 participants as of July 03, 2023. Up to eight years of CPD have been collected, with 2,225,010 products purchased, comprising 223,440 unique products. We matched 88.5% of all products by product name or item number to one generic food database and three product databases. Combined, the databases enable analysis of key exposures such as nutrients, ingredients, or additives. We found that increasing the number of retailers that provide CPD for each consumer improved the stability of individual CPD profiles and when we compared kilojoule information from generic and specific product matches, we found a median modified relative difference of 0.23. Combined with extensive product databases and health outcomes, CPD could provide the basis for extensive investigations of how what we buy affects our health.

Who buys non-alcoholic beer in Finland? Sociodemographic characteristics and associations with regular beer purchases

BACKGROUND

The consumption of non-alcoholic beer and other non-alcoholic and low-alcohol beverages has grown significantly in recent years. Due to a lack of suitable datasets, there have been few studies conducted on the forerunners of the non-alcoholic beer consumption trend. This study examined the associations of sociodemographic characteristics with non-alcoholic beer purchase, and of non-alcoholic beer purchases with regular beer purchases.

METHODS

The data consisted of longitudinal individual purchases of non-alcoholic and regular beer from grocery stores in 2017 and 2018. The study participants were loyalty cardholders from the largest food retailer in Finland (n = 47,066). The level of education, household income and occupational status were analyzed as determinants of non-alcoholic beer purchase using logistic regression models. The changes in the regular and non-alcoholic beer purchases from 2017 to 2018 and the distributions of non-alcoholic beer purchase by regular beer purchase, by gender and by age were described.

RESULTS

Between 2017 and 2018, the total volume of non-alcoholic beer purchases increased from 2.3% to 3.7% of the total volume of all beer purchases. Men and older people purchased non-alcoholic beer more often than women and younger people did. Non-alcoholic beer purchases were most common among the highly educated and high-income consumers. Non-alcoholic beer purchases were most prevalent in the groups with the highest volumes of regular beer purchase.

CONCLUSIONS

Educated and affluent consumers have been the forerunners of non-alcoholic beer consumption in Finland. In order to promote the substitution of regular beer with non-alcoholic beer the shift towards lower-strength beverages should be facilitated across social strata.

Changes in alcohol purchases from grocery stores after authorising the sale of stronger beverages: The case of the Finnish alcohol legislation reform in 2018

Aims: The Finnish alcohol law was reformed in January 2018. The availability of alcoholic beverages in grocery stores increased as the legal limit for retail sales of alcoholic drinks was raised from 4.7% to 5.5% alcohol, and the requirement of production by fermentation was abolished. We analysed how the inclusion of strong beers, ciders, and ready-to-drink beverages in grocery stores was reflected in alcohol purchases, and how these changes differed by age, sex, level of education and household income. Design: The study sample included 47,066 loyalty card holders from the largest food retailer in Finland. The data consisted of longitudinal, individual-level information on alcohol purchases from grocery stores, covering the time period between 1 January 2017 and 31 December 2018. The volumes of absolute alcohol during a calendar year from beers, ciders, ready-to-drink beverages, and in total, were calculated. Alcohol purchases in 2017 and 2018 were compared. Results: There was no overall change in the total alcohol (0.04 [95% CI -0.03, 0.11] litres/year) or beer purchases (-0.05 [95% CI -0.11, 0.02] litres/year). Purchases of ready-to-drink beverages increased by 0.10 [95% CI 0.09, 0.11] litres/year (+ 84%). Total alcohol purchases increased in the three highest income groups, whereas they decreased in the two lowest groups (p for the interaction < 0.0001). Conclusions: The increased purchases of alcohol as ready-to-drink beverages were, on the average, compensated for by a decrease in purchases of other alcoholic beverages. Higher prices probably limited the purchases among lower income groups and younger consumers, while the increase was sharper in higher income groups.

Do we eat what we buy? Relative validity of grocery purchase data as an indicator of food consumption in the LoCard study

The validity of grocery purchase data as an indicator of food consumption is uncertain. This paper investigated (1) the associations between food consumption and grocery purchases using automatically accumulated purchase data and (2) whether the strength of the associations differed in certain sub-populations. The participants filled in a FFQ, and a major Finnish retailer issued us with their loyalty-card holders' grocery purchase data covering the 1- and 12-month periods preceding the FFQ. We used gamma statistics to study the association between thirds/quarters of FFQ and grocery purchase data (frequency/amount) separately for eighteen food groups among the 11 983 participants. Stratified analyses were conducted for subgroups based on sex, family structure, educational level, household income and self-estimated share of purchases from the retailer. We also examined the proportion of participants classified into the same, adjacent, subsequent and opposite categories using the FFQ and purchase data. The gammas ranged from 0·12 (cooked vegetables) to 0·75 (margarines). Single households had stronger gammas than two-adult families, and participants with > 60 % of purchases from the retailer had stronger gammas. For most food groups, the proportion of participants classified into the same or adjacent category was > 70 %. Most discrepancies were observed for fresh/cooked vegetables, berries and vegetable oils. Even though the two methods did not categorise all food groups similarly, we conclude that grocery purchase data are able to describe food consumption in an adult population, and future studies should consider purchase data as a resource-saving and moderately valid measure in large samples.

Food purchase behaviour in a Finnish population: patterns, carbon footprints and expenditures

OBJECTIVE

To identify food purchase patterns and to assess their carbon footprint and expenditure.

DESIGN

Cross-sectional.

SETTING

Purchase patterns were identified by factor analysis from the annual purchases of 3435 product groups. The associations between purchase patterns and the total purchases' carbon footprints (based on life-cycle assessment) and expenditure were analysed using linear regression and adjusted for nutritional energy content of the purchases.

PARTICIPANTS

Loyalty card holders (n 22 860) of the largest food retailer in Finland.

RESULTS

Eight patterns explained 55 % of the variation in food purchases. The Animal-based pattern made the greatest contribution to the annual carbon footprint, followed by the Easy-cooking, and Ready-to-eat patterns. High-energy, Traditional and Plant-based patterns made the smallest contribution to the carbon footprint of the purchases. Animal-based, Ready-to-eat, Plant-based and High-energy patterns made the greatest contribution, whereas the Traditional and Easy-cooking patterns made the smallest contribution to food expenditure. Carbon footprint per euros spent increased with stronger adherence to the Traditional, Animal-based and Easy-cooking patterns.

CONCLUSIONS

The Animal-based, Ready-to-eat and High-energy patterns were associated with relatively high expenditure on food, suggesting no economic barrier to a potential shift towards a plant-based diet for consumers adherent to those patterns. Strong adherence to the Traditional pattern resulted in a low energy-adjusted carbon footprint but high carbon footprint per euro. This suggests a preference for cheap nutritional energy rather than environment-conscious purchase behaviour. Whether a shift towards a plant-based pattern would be affordable for those with more traditional and cheaper purchase patterns requires more research.