4. Discussion
The present study aimed to explore how the determinants of food choice shape the perceptions regarding supermarket-based nudging strategies among adults with low SEP. Overall, interviewees mentioned a wide range of factors on an individual level, socio-economic level and environmental level to influence their food choices. Food costs, convenience, healthiness, taste, and habits were frequently mentioned, and their relative importance seemed to be context-specific. Moreover, interviewees generally had positive perceptions towards nudges, especially when they aligned with their product preferences, information needs, and beliefs about the food environment. Still, a small number of interviewees also expressed distrust towards nudging strategies, suspecting ulterior motives of supermarkets or the food industry, irrespective of their determinants of food choice.
4.1. Determinants of Food Choice and Nudging
In the present study, we confirm earlier findings on the importance of determinants, such as food costs [14,15,16,17,18,26], convenience [15], healthiness [15,26], and taste [15,26]. Additionally, this study provides preliminary evidence that nudges that align with interviewee’s determinants of food choice, are perceived more positively. This was most notably observed for product preferences: if nudges targeted foods which interviewees liked (e.g., in terms of taste, convenience), then nudges were also perceived more positively. This may imply that the type of food which is nudged may be instrumental for nudge effectiveness in populations with low SEP. For example, following international dietary recommendations, the intake of red and processed meat should be reduced [27]. From the present study, it may be deduced, that in order to nudge people from meat to another substitute, this alternative should align with values such as such as taste and convenience. As a result, nudging a healthy meat substitute (e.g., similar taste and preparation method) may be a better option than, for example, tofu (e.g., notably different taste, potentially different preparation method). We also noted that nudges were perceived more positively when they tapped into certain information needs, which may imply that information nudges [28] may be of particular interest to populations with low SEP.
However, a number of interviewees also voiced concerns about the legitimacy of supermarket-driven health-promoting initiatives, suspecting ulterior motives. Distrust of information on health behaviours has previously been highlighted as a barrier for attaining behaviour change in community-based interventions among populations with low SEP [29] but has not been documented in the specific case of supermarket-based nudging strategies. As opponents of nudging often frame nudging as infringement on autonomous choice [30], one way to address the underlying feeling of distrust is for supermarkets to be transparent about applying nudging strategies using positive framing and to explain why, how, and with whom (e.g., governmental agencies) they participate in such initiatives. In particular, with regard to the application of information nudges, this would imply that the information provided is perceived to come from a credible source. Still, it must be noted that interviewees did not voice ethical concerns about nudging per se, and some even perceived nudges to actually provide them with autonomy to make in-store food choices.
Beliefs about the food environment also influenced perceptions of nudging strategies. We observed that interviewees often viewed the supermarket environment in a relatively negative way, describing it as a tempting environment for which the interviewees had developed strategies to withhold themselves from engaging in alluring unhealthy purchases (e.g., restricting supermarket visits). As such, interviewees perceived nudging to be a welcome strategy for making the supermarket food environment healthier. This contrasts heavily with the perception among retailers of customer demand for healthy foods to be low and customer interest in health to be limited, which is often cited as a barrier for retailer’s engagement in health promoting interventions [31]. This could indicate a mismatch between consumer needs, and retailers’ perceptions of them, and present a potential argument to convince retailers to invest more into health-promoting initiatives.
4.2. The Role of Habits
The role of habit as an important determinant of food choice in adults with low SEP, warrants some extra consideration in the context of nudging. Nearly all interviewees indicated that their food habits and/or shopping habits were largely habitual. Habits have been defined as behaviours that have become established through a history of systematic repetition and reinforcement, and therefore have become automatized [32]. Since habits and nudges share the same proposed working mechanism of automated unconscious cognitive processes, the question arises to what extent nudges are able to overrule hard-wired automated habitual food choices. Emerging literature suggests that strong habits or a priori preferences indeed may form boundary conditions for nudge effectiveness regarding food choice [33,34]. For example, a study conducted by Venema et al. showed that the effects of a portion size nudge aimed at decreasing sugar consumption in tea were less effective in participants with strong habits concerning the addition of sugar to tea beverages, compared to participants with less strong habits [34]. Similarly, in a study aiming to nudge participants to a smaller soda drink, it was found that a priori preferences for soda drinks (e.g., liking of those drinks or intentions to reduce consumption of soda drinks), were stronger predictors of the chosen drinks than the nudge itself [33]. Although these former studies were conducted in laboratory settings, these findings may suggest that also in real-life environments such as supermarkets, a priori preferences of customers for specific foods are likely to compete with nudges. Examples of strong preferences that were noted in the present study include the preference for fresh vegetables as opposed to canned vegetables, preference for certain brands, or preference for foods that are convenient. A practical implication that may follow from this, is making sure that the nudged foods are not subject to strong a priori preferences, but represent foods for which adults with low SEP are likely to be fairly indifferent.
4.3. Needs Concerning Other Public Health Interventions
The needs expressed by interviewees also underline the importance of other public health interventions other than nudging strategies that aim to promote healthy food choices. In particular, in the case of ill health, there was a need to obtain objective and individualized dietary information (e.g., information on salt content prompted by heart failure induced salt restriction), which may be fulfilled by the use of mobile apps that can be used in the supermarket environment. An example of such an app is the ‘Kies ik Gezond?’ (do I make a healthy choice?) application from the Dutch Nutrition Center, which allows users to scan barcodes in the supermarket and subsequently informs the user whether the scanned food item is contained in the Wheel of Five (the Dutch dietary recommendations), and also provides healthy alternative suggestions for unhealthy foods. The use of these apps could be promoted, especially given the fact that some interviewees perceived nutrition information to be conflicting, and even obtained very erroneous nutrition information from sources, such as the Internet. Lastly, the interviews highlighted that price of foods often guided food choice within the supermarket setting, indicating that pricing strategies are likely to be supported by populations with low SEP. Concluding, the needs expressed by the interviewees suggest that nudging is not likely to be a silver bullet for achieving healthy diets [35], but that in fact, we need a multitude of interventions (e.g., personalized nutrition advice, pricing strategies) to match the varying needs and preferences of adults with low SEP.
4.4. Limitations and Strengths
The present study should be interpreted in light of some limitations. First, we only used educational level as an indicator of SEP, as we deemed it stigmatizing to ask interviewees about other proxies of SEP such as income. As it is well-known that SEP is a multidimensional construct, the use of educational level only is likely to be a simplification of reality. Moreover, given the difficulties associated with recruiting individuals with low SEP [36], we adhered to a broader definition of lower educational than, for example, the definition used by the Dutch Central Bureau of Statistics, for pragmatic reasons. Third, our sample was characterized by a relatively high age, which may limit the generalizability of our findings to younger populations.
The study also had several strengths. The analysis and presentation of the data were thoroughly reviewed by and discussed with a second researcher, which strengthens the internal validity of the study. Although not systematically recorded, the interviewers judged the interviewees to constitute a representative sample of the population with regard to weight status, which may be of importance when exploring the topic of food choice. Lastly, although the sample size was relatively small, saturation was reached, which indicates that all major determinants for the interviewed group are likely to have been identified.
CC BY 4.0. "Determinants of Food Choice and Perceptions of Supermarket-Based Nudging Interventions among Adults with Low Socioeconomic Position: The SUPREME NUDGE Project" by Harbers MC, Middel CNH, Stuber JM et al., International journal of environmental research and public health (2021). https://doi.org/10.3390/ijerph18116175. Licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Peer-reviewed and open access; full text from PubMed Central (PMC8201246). Headings and layout adapted for reading; figures and tables are in the original.