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.

Characterization and Correction of Bias Due to Nonparticipation and the Degree of Loyalty in Large-Scale Finnish Loyalty Card Data on Grocery Purchases: Cohort Study.

BACKGROUND: To date, the evaluation of diet has mostly been based on questionnaires and diaries that have their limitations in terms of being time and resource intensive, and a tendency toward social desirability. Loyalty card data obtained in retailing provides timely and objective information on diet-related behaviors. In Finland, the market is highly concentrated, which provides a unique opportunity to investigate diet through grocery purchases. OBJECTIVE: The aims of this study were as follows: (1) to investigate and quantify the selection bias in large-scale (n=47,066) loyalty card (LoCard) data and correct the bias by developing weighting schemes and (2) to investigate how the degree of loyalty relates to food purchases. METHODS: Members of a loyalty card program from a large retailer in Finland were contacted via email and invited to take part in the study, which involved consenting to the release of their grocery purchase data for research purposes. Participants' sociodemographic background was obtained through a web-based questionnaire and was compared to that of the general Finnish adult population obtained via Statistics Finland. To match the distributions of sociodemographic variables, poststratification weights were constructed by using the raking method. The degree of loyalty was self-estimated on a 5-point rating scale. RESULTS: On comparing our study sample with the general Finnish adult population, in our sample, there were more women (65.25%, 30,696/47,045 vs 51.12%, 2,273,139/4,446,869), individuals with higher education (56.91%, 20,684/36,348 vs 32.21%, 1,432,276/4,446,869), and employed individuals (60.53%, 22,086/36,487 vs 52.35%, 2,327,730/4,446,869). Additionally, in our sample, there was underrepresentation of individuals aged under 30 years (14.44%, 6,791/47,045 vs 18.04%, 802,295/4,446,869) and over 70 years (7.94%, 3,735/47,045 vs 18.20%, 809,317/4,446,869), as well as retired individuals (23.51%, 8,578/36,487 vs 31.82%, 1,414,785/4,446,869). Food purchases differed by the degree of loyalty, with higher shares of vegetable, red meat & processed meat, and fat spread purchases in the higher loyalty groups. CONCLUSIONS: Individuals who consented to the use of their loyalty card data for research purposes tended to diverge from the general Finnish adult population. However, the high volume of data enabled the inclusion of sociodemographically diverse subgroups and successful correction of the differences found in the distributions of sociodemographic variables. In addition, it seems that food purchases differ according to the degree of loyalty, which should be taken into account when researching loyalty card data. Despite the limitations, loyalty card data provide a cost-effective approach to reach large groups of people, including hard-to-reach population subgroups.

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.

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

BACKGROUND: Alcohol use epidemiology is facing challenges as survey response rates decline. In addition, population surveys fail to capture a large proportion of alcohol consumed and are expensive to conduct. This study aims to aid in complementing traditional epidemiological methods by validate grocery purchase data in the research on population alcohol use. METHODS: The LoCard study subjects were loyalty card holders of a grocery retail co-operative, which possessed more than 45 % market share in Finland. One third of those who consented to the analyses of their grocery purchases were presented a questionnaire including a Food Frequency Questionnaire on the web; N = 11,818 responded. The relationship between beer purchase frequency and self-reported beer drinking frequency was studied for association and agreement in different subgroups using crosstabulations and Poisson regression modeling. RESULTS: The association between beer purchase frequency and self-reported beer drinking frequency was good (Gamma = .556). The agreement between beer purchase frequency and drinking frequency was only fair (Kappa = .189). Limiting the data to those single adult households that reported making at least 61 % of their grocery purchases from this grocery retailer and collapsing the frequency categories to three instead of six increased the agreement to good (Kappa = .463). CONCLUSIONS: Information on beer purchase frequency from the loyalty card database can be used to rank people according to their drinking frequency and to estimate beer drinking frequency with fair to good accuracy, depending on what share of grocery purchases they make from the grocery retailer in question.

Large-scale loyalty card data in health research.

OBJECTIVE: To study the characteristics of large-scale loyalty card data obtained in Finland, and to evaluate their potential and challenges in health research. METHODS: We contacted the holders of a certain loyalty card living in a specific region in Finland via email, and requested their electronic informed consent to obtain their basic background characteristics and grocery expenditure data from 2016 for health research purposes. Non-participation and the characteristics and expenditure of the participants were mainly analysed using summary statistics and figures. RESULTS: The data on expenditure came from 14,595 (5.6% of those contacted) consenting loyalty card holders. A total of 68.5% of the participants were women, with an average age of 46 years. Women and residents of Helsinki were more likely to participate. Both young and old participants were underrepresented in the sample. We observed that annual expenditure represented roughly two-thirds of the nationally estimated annual averages. Customers and personnel differed in their characteristics and expenditure, but not so much in their most frequently bought items. CONCLUSIONS: Loyalty card data from a major retailer enabled us to reach a large, heterogeneous sample with fewer resources than conventional surveys of the same magnitude. The potential of the data was great because of their size, coverage, objectivity, and long periods of dynamic data collection, which enables timely investigations. The challenges included bias due to non-participation, purchases in other stores, the level of detail in product grouping, and the knowledge gaps in what is being consumed and by whom. Loyalty card data are an underutilised resource in research, and could be used not only in retailers' activities, but also for societal benefit.