Hostname: page-component-78c5997874-m6dg7 Total loading time: 0 Render date: 2024-11-03T04:04:41.032Z Has data issue: false hasContentIssue false

Suggested explanations for the (in)effectiveness of nutrition information interventions among adults with a low socioeconomic status: a scoping review

Published online by Cambridge University Press:  23 June 2022

Tim van Meurs*
Affiliation:
Department of Public Administration and Sociology, Erasmus University Rotterdam, Burgemeester Oudlaan 50, 3062PA Rotterdam, the Netherlands
Joost Oude Groeniger
Affiliation:
Department of Public Administration and Sociology, Erasmus University Rotterdam, Burgemeester Oudlaan 50, 3062PA Rotterdam, the Netherlands Department of Public Health, Erasmus MC, Doctor Molewaterplein 40, 3015GD Rotterdam, the Netherlands
Willem de Koster
Affiliation:
Department of Public Administration and Sociology, Erasmus University Rotterdam, Burgemeester Oudlaan 50, 3062PA Rotterdam, the Netherlands
Jeroen van der Waal
Affiliation:
Department of Public Administration and Sociology, Erasmus University Rotterdam, Burgemeester Oudlaan 50, 3062PA Rotterdam, the Netherlands
*
*Corresponding author: Tim van Meurs, email [email protected]

Abstract

Unhealthy diets are a major threat to population health and are especially prevalent among those with a low socioeconomic status (SES). Health promotion initiatives often rely on nutrition information interventions (NIIs), but are usually less effective among adults with a low SES than in their high-SES counterparts. Explanations for this lower effectiveness are set out in extant studies. These have been conducted across a wide range of disciplines and subject fields and using a variety of methodological approaches. We have therefore conducted a scoping review to identify and synthesise the following: (1) explanations suggested in studies carried out in high-income countries for why NIIs are (in)effective among adults with a low SES and (2) whether these suggested explanations were studied empirically. Eight databases were searched for relevant studies published since 2009 across various disciplines. This identified 4951 papers, 27 of which were included in our review after screening. Only fifteen of these proposed an explanation for the (in)effectiveness of NIIs among adults with a low SES. The following four main themes were uncovered: health literacy, economic resources, social resources and convenience. Ten studies tested their explanations empirically, but the results were inconsistent. The reasons why NIIs are (in)effective among low-SES adults are therefore still largely unclear. Also, current literature predominantly relies on individualistic explanations, most notably focusing on psychological and economic attributes. Consequently, if the effectiveness of NIIs among low-SES populations is to be improved, future studies should examine a wider range of explanations and test them systematically and empirically.

Type
Review Article
Creative Commons
Creative Common License - CCCreative Common License - BY
This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted re-use, distribution, and reproduction in any medium, provided the original work is properly cited.
Copyright
Copyright © The Author(s), 2022. Published by Cambridge University Press on behalf of The Nutrition Society

Introduction

Notwithstanding the substantial efforts being made by governmental, scientific and health institutions to promote good health, a significant difference remains in this regard between those in higher and lower socioeconomic status (SES) groups(Reference Mackenbach, Stirbu and Roskam1). The persistence of this gap is partly the result of dietary inequalities(Reference Darmon and Drewnowski2,Reference Plessz and Gojard3) . These reflect the reality that health interventions aimed at improving what we eat are either ineffective among adults in the lower social strata, or less effective than among their higher-SES counterparts(Reference Hartmann-Boyce, Bianchi and Piernas4,Reference Sarink, Peeters and Freak-Poli5) . Consequently, nutritional interventions are often, and inadvertently, failing to narrow this SES health disparity. In particular, nutrition information interventions (NIIs) that encourage healthier eating by informing people ‘how to choose nutritious foods in order to follow guidelines for healthy eating’(Reference Glanz, Coulston, Rock and Monsen6) are frequently less successful at achieving their intended goals among those with a low SES(Reference Beauchamp, Backholer and Magliano7,Reference Koç and van Kippersluis8) . Nevertheless, NIIs are still popular(Reference Schiavo9), mainly because they are easy to execute and not particularly dependent on governmental decisions and the enactment of legislation.

It is, therefore, clear that the development of more effective and, specifically, more equitable NIIs requires an understanding of why current NIIs are (in)effective among low-SES groups. Nevertheless, there has been no comprehensive overview of the explanations suggested for why NIIs are (in)effective among these groups, nor of the extent to which these explanations have been studied empirically. Yet, since this reasoning probably differs across disciplines and research fields, it is important to synthesise this knowledge base. Consequently, we performed a scoping review to identify intervention studies conducted in high-income countries that examined the effectiveness of NIIs among low-SES groups (either specifically in these populations or that included a subgroup analysis). The review encompassed research conducted in a variety of fields, using various study designs, and with different types of NIIs. In particular, we carried out a thematic analysis to uncover the explanations suggested in the studies for the (in)effectiveness of these interventions among low-SES adults. We also examined whether these explanations were studied empirically. Our review was guided by the following research questions: (1) What are the key explanations suggested in health intervention studies for why NIIs are (in)effective at improving health knowledge and achieving (intended) behavioural change among low-SES adults? and (2) Have these explanations been studied empirically?

Methods

We conducted a scoping review to answer our research questions(Reference Arksey and O'Malley10,Reference Davis, Drey and Gould11) . Scoping reviews are commonly used to summarise, rather than evaluate, a particular field, which enabled us to ‘examine the extent, range and nature of research activity’(Reference Arksey and O'Malley10) relevant to the issue at hand. We conducted the review following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) checklist extension for scoping reviews(Reference Tricco, Lillie and Zarin12).

Identifying relevant studies

The studies in our sample were obtained after a search of a variety of electronic databases, in particular by translating into a search syntax the concepts concerning (un)healthy diets that were closely related to the research questions. These were then combined with the terms ‘health information interventions’ and ‘socioeconomic status’, as well as their equivalent medical subject headings (MeSH). Abbreviations, synonyms and indicators were added to widen the search.

A variety of databases was used (Web of Science, Embase, Medline Ovid, Cochrane, Psyc INFO, Econ Lit, Abi/inform and Google Scholar) to ensure the inclusion of studies from diverse disciplines and research fields (Supplementary Appendix lists the search queries). Subsequently, references of the included studies were scanned to identify papers that may have been missed in the initial search, but none were detected.

Study selection

All of the studies’ titles, abstracts and keywords, as well as the full texts, were screened independently by the first and last authors based on pre-set inclusion and exclusion criteria. Studies were included when they contained NIIs that: (1) were empirically discussed; (2) were produced by official institutions (e.g., governmental, scientific and health institutions); (3) had the aims of improving knowledge of health issues and/or changing (intended) behaviour; (4) concerned (un)healthy diets and (5) were examined on the basis of their effectiveness among low-SES groups. Studies were excluded when the intervention did not take place in a high-income Organisation for Economic Co-operation and Development (OECD) country or was not targeted exclusively at adults. The search period covered papers published from January 2009 to April 2019. The decision to use 2009 as a starting point was based on the landmark publication of the report ‘Closing the Gap in a Generation’. This was produced by the World Health Organisation's Commission on Social Determinants of Health(13), and caused an upsurge in research focusing on reducing socioeconomic health inequalities.

Charting data

Quotes were extracted from the papers concerning the studies’ designs, locations, outcome measures, intervention target groups, types of NII and identified effects on the low-SES participants. We also extracted details on the equitability of the intervention examined and the (suggested) explanations for why it was (in)effective. These quotes were then coded inductively by the first author.

Collating, summarising and reporting findings

Coding was conducted based on a thematic analysis (the themes of the explanations proposed). Each study was assigned a theme (and potential additional themes) based on the reasons used to explain why the NII was (in)effective among the low-SES participants. These themes were then categorised using higher-level coding to facilitate the synthesis of the studies. Our analysis thus produced a scoping review of the different explanations currently suggested for the (in)effectiveness and their empirical value. To the best of our knowledge, this is the first review to focus on why NIIs are said to be (in)effective among adults with a low SES, rather than on whether they have an impact and, if so, to what extent.

Results

Descriptive numerical summary

The initial database search produced 22 985 entries, reduced to 15 172 after the removal of duplicates. Of these, 10 581 were pre-excluded based on the publication date (pre-2009), study population (not adults) and the country where the study was carried out (not a high-income OECD member). This reduced the sample to 4951 studies. Titles and abstracts were then reviewed, producing fifty-eight full texts for screening. This led to a sample comprising twenty-nine studies. Two of these were then excluded after careful consideration during the data-extraction phase, as they proved to be ineligible after all, leaving a final sample of twenty-seven studies for use in the thematic analysis. The inter-coder reliability for the full-text phase was 82⋅1 %. The coders subsequently reached a consensus by discussing whether to include or exclude the remaining studies that had initially been regarded as eligible by only one of them. Fig. 1 contains a detailed overview of the selection process.

Fig. 1. Flowchart of inclusions and exclusions.

Studies discussed nutritional or calorie-labelling (n 10); tailored (web-based) health information (n 6); general guidelines or recommendations (e.g., state-level guidelines, educational poster) (n 5); mass/multimedia campaigns (n 2); on-location information (e.g., point-of-purchase merchandising, grocery store interventions) (n 2); combined interventions with informational meetings (n 2); multicomponent education interventions (n 1) and online interventions (n 1). The NIIs examined in the studies were produced by the government (n 12); health institutions (n 7); science/academia (n 6) and non-governmental organisations (n 1). Two studies (both systematic literature reviews) did not specify the institution(s) that provided the health information.

Most of the studies concerned a single country, in particular: the United States (n 13); the Netherlands (n 5); New Zealand (n 2); the United Kingdom (n 2); Australia (n 1); France (n 1) and Norway (n 1). Two studies – both literature reviews – involved multiple countries. A cross-country comparison showed that the studies from the US had a disproportionately strong focus on (fast food menu) calorie-labelling, which was a feature in five of thirteen studies. The only other studies on labelling (one in the UK and one in Norway) focused more broadly on a variety of forms of labelling.

The crucial findings of the studies included in our thematic analysis are described in columns seven (effect among low SES) and eight (equity effect) in Table 1. Fifteen studies identified a positive intervention effect on their low-SES participants, five did not, and four had mixed results (a positive effect was found for some of the outcome measures). In three studies, the effect for low-SES participants was unclear. In terms of equity effects, only one study reported that the NII was completely equity-positive (i.e., inequalities decreased), whereas five identified negative equity effects, and seven highlighted similar effects across SES groups. Seven studies described mixed equity results, which generally meant that the equity findings differed per outcome. In one case(Reference Øvrum, Alfnes and Almli28), these effects varied per SES indicator, while another was a systematic literature review that identified different equity effects in the studies included in its sample. Seven studies from the US were the only ones to focus exclusively on those with a low SES, with those conducted elsewhere merely examining entire populations. As a result, equity effects were not reported in over half of the US-based studies.

Table 1. Summary table of included studies

Thematic analysis

The analysis uncovered four main themes, which were supplemented with a category given the name ‘other’. The subsections below are ordered according to the number of times a theme was identified, starting with the most common.

Health literacy

Twelve studies used the issue of ‘health literacy’ (i.e., being (un)able to understand the information contained in an NII) to explain the (in)effectiveness of the interventions among those with a low SES(Reference Sarink, Peeters and Freak-Poli5,Reference Blake, Lancsar and Peeters15,Reference Elbel, Kersh and Brescoll19,Reference Gans, Risica and Strolla20,Reference Gorton, Ni Mhurchu and Chen22,Reference Mancino and Kuchler25,Reference Masic, Christiansen and Boyland26,Reference Schindler, Kiszko and Abrams30Reference Springvloet, Lechner and Candel33,Reference Walsh, Kattelmann and White37) . As an example, NIIs were described as succeeding in this group because they were ‘clear and simple’(Reference Blake, Lancsar and Peeters15) or had an approach that was ‘low literate [in] nature’(Reference Gans, Risica and Strolla20). In both cases, the NII was effective at reducing inequalities in nutritional health between the studies’ low- and high-SES participants. In other cases, too, simpler design elements were reported to be the reason for their effectiveness among those with a low SES(Reference Gorton, Ni Mhurchu and Chen22,Reference Masic, Christiansen and Boyland26,Reference Walsh, Kattelmann and White37) , although no increases in equity were identified.

Other studies argued that NIIs are less effective because low-SES groups find it harder to process the materials provided to them, reaching the conclusion that simpler interventions could improve equity effects(Reference Sarink, Peeters and Freak-Poli5). The outcomes of the intervention tested by Springvloet et al. (Reference Springvloet, Lechner and De Vries31Reference Springvloet, Lechner and Candel33) caused them to suggest that people with a low SES may be overwhelmed by the scope (in terms of content or quantity) of the NIIs provided to them. Meanwhile, the studies by Elbel et al. (Reference Elbel, Kersh and Brescoll19) and Schindler et al. (Reference Schindler, Kiszko and Abrams30), both of which examined the effectiveness of calorie-labelling, proposed that this kind of information is currently not clear enough and requires improvement if this type of NII is to succeed.

Mancino and Kuchler(Reference Mancino and Kuchler25) used a slightly different argument to highlight the importance of health literacy. In particular, they suggested that the dietary guidelines concerning wholegrain bread might have less of an impact on low-income consumers, since they may find it difficult to distinguish between wholegrain and non-wholegrain products (even if they understand the message that the former is the healthier option).

This theme was also the only one in which cross-country differences were observable. Although health literacy was discussed in the studies conducted in a number of countries (Australia, New Zealand, the Netherlands, the UK and the US), it was relatively less prominent in US-based studies: while health literacy was discussed in five of seven US-based studies that suggested an explanation for the (in)effectiveness of an NII, the concept mostly arose in combination with other themes. Conversely, save for a single UK-based study, non-US-based studies focused exclusively on health literacy as the explanation for the (in)effectiveness identified.

Economic resources

The second-most common theme was economic resources (n 5)(Reference Sarink, Peeters and Freak-Poli5,Reference Capacci and Mazzocchi18,Reference Hersey, Cates and Blitstein23,Reference Mancino and Kuchler25,Reference Schindler, Kiszko and Abrams30) , which was given as a main reason for the limited effectiveness of NIIs among those in the low-SES group. Two studies(Reference Hersey, Cates and Blitstein23,Reference Mancino and Kuchler25) suggested that the chief cause of this was the (un)affordability of healthy food. Indeed, even though the NII employed in the study by Hersey et al. (Reference Hersey, Cates and Blitstein23) did lead to an increase in the intake of fruit and vegetables among low-SES individuals, the daily amounts consumed did not accord with the NII's recommendations. The possible explanation suggested for this outcome was the high cost of the relevant products. Mancino and Kuchler(Reference Mancino and Kuchler25) echoed these findings, arguing that many in the low-SES group live in ‘areas with limited access to affordable and nutritious foods’, signalling the concept of ‘food deserts’. They also reported that finding inexpensive food was more important to their economically deprived participants than consuming healthy options. This was also highlighted as an issue by the interviewees in Schindler et al.'s study(Reference Schindler, Kiszko and Abrams30).

Capacci and Mazzocchi(Reference Capacci and Mazzocchi18) likewise found that having limited economic resources impacted the effectiveness of their study's multimedia campaign among those with a low SES in their sample, making the argument that additional income support or raising the prices of unhealthy food could improve this. The final study to refer to this theme was conducted by Sarink et al. (Reference Sarink, Peeters and Freak-Poli5), although it offered no further explanation beyond identifying deprivation as a potential factor.

Social resources

The third main theme identified was mentioned by two studies and focuses on the limited social resources of the low-SES group(Reference Sarink, Peeters and Freak-Poli5,Reference Hersey, Cates and Blitstein23) . Hersey et al. (Reference Hersey, Cates and Blitstein23) suggested that social support might be relevant after noting that their participants discussed the information in the material provided to them with their peers. The authors thus proposed that such discussions could increase the backing for healthy food options, implying that NIIs might be less effective for those with a limited social support.

Meanwhile, Sarink et al. (Reference Sarink, Peeters and Freak-Poli5) also suggest that there is greater uptake of the advice in NIIs if the intended target has more social resources. Although they do not elaborate on the underlying causal mechanisms, these authors do argue that adults with a low SES have relatively fewer such resources and, as a result, demonstrate only limited support for menu labelling.

Convenience

Two studies pointed to the convenience of unhealthy food as a potential reason for why the advice in NIIs is not adopted(Reference Mancino and Kuchler25,Reference Schindler, Kiszko and Abrams30) . According to Mancino and Kuchler(Reference Mancino and Kuchler25), health was a lower priority than convenience for the lower-income consumers in their sample, who were unmoved by the intervention examined in the study. This was also an argument made by Schindler et al. (Reference Schindler, Kiszko and Abrams30), who found that convenience was a factor in the decisions made about food options, even when a menu provided information on the calories in each dish. In particular, despite the recommendations in the NII employed in the research, their low-income participants continued to buy from fast food chains, stressing that they did so because the convenience of not having to buy and prepare food was more important to them than the issue of the price of fast food. The participants also pointed to a lack of time and, as a consequence, their disregard for information about calories. Nonetheless, neither of these studies identified why issues of convenience were more prevalent in their particular low-SES sample.

Others

Rameshbabu(Reference Rameshbabu29) found that study participants who scored highly for self-regulation and self-efficacy were affected more by the NII being examined. She consequently argued that ‘teaching self-regulation skills serves to involve the individual in making the behaviour change rather than passively providing them with the information to do so’(Reference Rameshbabu29). However, her study contained no comparisons of those in different SES groups.

Gans et al. (Reference Gans, Risica and Strolla20), meanwhile, noted that the NIIs used in their study, in part, worked better among the low-SES respondents, because the material was ‘more personally relevant’: these interventions addressed the participants directly by using their name and tailoring the health information message to them as individuals, contributing to them being ‘more positive about how interesting and personally relevant’ this tailored information was. Their study was the only one to mention the tailored nature of NIIs as a possible reason for their effectiveness among low-SES groups, even though other studies have also examined these effects. However, these other studies do not specifically suggest that the tailoring of such interventions – and their resulting personal appeal – is an explanation for the extent of their effectiveness in these individuals. As a consequence, we have not included them within this theme.

No explanation

A key finding of our review was that twelve of the twenty-seven studies analysed did not include any explanation at all for why NIIs are (in)effective among low-SES groups(Reference Hartmann-Boyce, Bianchi and Piernas4,Reference Backman, Cheung and Francis14,Reference Blakely, Mhurchu and Jiang16,Reference Breck, Cantor and Martinez17,Reference Gans, Risica and Keita21,Reference Irz, Leroy and Réquillart24,Reference McGeary27,Reference Øvrum, Alfnes and Almli28,Reference Taksler and Elbel34Reference Verheijden, van Dommelen and van Empelen36,Reference Walthouwer, Oenema and Lechner38) ; instead, these studies simply reported their results, with no discussion of possible reasons for these outcomes.

Empirical scrutiny of the suggested explanations

Of the twenty-seven studies included in the analysis, ten contained some empirical scrutiny of the explanations suggested for the (in)effectiveness of an NII among their low-SES participants. The most direct evidence came from focus group interviews(Reference Schindler, Kiszko and Abrams30), which discussed various factors in order to identify why calorie information in fast food restaurants is rarely considered. The issues considered above all others by the participants were, most notably, clarity (i.e., health literacy) and convenience.

Nine studies referred to the design of an NII to account for its (in)effectiveness. Two tested different types of nutritional labels(Reference Gorton, Ni Mhurchu and Chen22,Reference Masic, Christiansen and Boyland26) finding that simplified versions were more effective among low-SES groups. These included traffic-light labelling (using colours to indicate the healthiness of an item) and physical-activity labels (indicating the amount of exercise required to burn off the calories in the product). These approaches thus reduce the need for health literacy to ensure the effectiveness of an intervention. This is in contrast to numerical labels, which simply state the nutritional content in absolute numbers or as a percentage of daily intake. It should be noted that this simplified form of nutritional information also resonated with those in high-SES groups and, as a result, did not lead to any equity changes.

Six studies made the claim that the analysed NII had been developed with the specific goal that the information should be comprehensible to those with lower health literacy(Reference Blake, Lancsar and Peeters15,Reference Gans, Risica and Strolla20,Reference Springvloet, Lechner and De Vries32,Reference Springvloet, Lechner and Candel33,Reference Walsh, Kattelmann and White37,Reference Springvloet, Willemsen and Mons39) , implying that any disparity in relation to this factor is the reason for the relative ineffectiveness of NIIs among low-SES groups. Nonetheless, the outcomes of these studies are inconsistent, with some producing equity-positive results and others equity-neutral or equity-negative outcomes. Moreover, none of the studies based on NIIs that were easy to understand were compared to versions containing less digestible information.

As well as taking health literacy into account in the design of their study, the tailored nature of the NII examined by Gans et al. (Reference Gans, Risica and Strolla20) was also claimed to be a reason for its effectivity, as it was deemed to be more personally appealing to those with a low SES. It should be noted, however, that an equity effect was only identified for one of the four outcome measures (change in the intake of fruit and vegetables at 7 months follow-up); for the other three (change in the fruit and vegetable intake at 4 months, and in the intake of fats at 4 and 7 months) no such effects were uncovered. Nevertheless, this combination of equity-neutral and equity-positive results does suggest that the use of tailored nutrition information takes us a step closer to reducing inequality. This seems to be endorsed in the studies by Springvloet et al. (Reference Springvloet, Lechner and De Vries32,Reference Springvloet, Lechner and Candel33,Reference Springvloet, Willemsen and Mons39) and Walthouwer et al. (Reference Walthouwer, Oenema and Lechner38), whose mainly equity-neutral or equity-positive results were achieved with tailored NIIs. Nonetheless, it should be noted that neither study attributes the effectiveness of the interventions examined to their tailored nature.

Finally, Rameshbabu(Reference Rameshbabu29) found that promoting self-regulatory skills within an informational message had a positive impact on the extent to which the material was absorbed and acted upon. While the intervention alone also seemed to influence the study's respondents positively, adding information about self-regulation increased this effect significantly. Nonetheless, with participants exclusively from low-SES groups (non-academic employees at one university), it was not possible for the study to make claims about equitability, i.e., it was unable to determine whether the endorsement of self-regulatory skills alongside an informational message would affect high-SES groups to a different extent.

Conclusion and discussion

Our scoping review identified twenty-seven studies that examined the effectiveness of NIIs among adults with a low SES. While most of the interventions investigated were shown to be effective among low-SES groups, they were often just as, or more, effective among those whose SES was high. Our thematic analysis revealed that almost half of the twenty-seven studies offered no explanation for the (lack of) impact on low-SES groups. In those that did, four main themes were identified: health literacy, economic resources, social resources and convenience; there were two further explanations that did not fit within these themes: self-regulation/self-efficacy and personal appeal.

Ten of the studies examined included some form of empirical research on the tenability of the explanations proposed. These predominately targeted the issue of ‘health literacy’ and the provision of simplified and easily digestible information. However, since these studies had inconsistent outcomes (equity-positive, equity-neutral and equity-negative) and most did not include a clear comparison group (e.g., information not adjusted to the level of literacy), it was not possible to determine the empirical tenability of this explanation. This suggests that an intervention that is only easy to understand is no panacea when it comes to reducing the nutritional health inequalities that exist today.

The most direct empirical evidence came from focus group interviews. Nonetheless, the qualitative nature of this type of study does not enable findings to be generalised to the population at large. Moreover, the participants were exclusively from low-SES groups, with most also having an ethnic minority background. This makes it difficult to determine whether the attitudes and actions reported were the results of a low SES, a particular ethnic background or a combination thereof.

Overall, therefore, it is still unclear why the equity effects of NIIs are inconsistent. In large part, this is because many of the studies examined were unable to test for differential effects (e.g., they were not powered to test the moderating effects of SES). However, even those that did include such an analysis did not always offer an explanation of their findings, perhaps because no between-group differences were identified. Nevertheless, even these limited interventions may provide valuable insights, as it is clearly more common for NIIs to be less effective among those in lower-SES groups. Accordingly, if the impact of such information is to be improved, it is important for intervention studies to focus more on why – instead of just on whether – some NIIs are effective and equitable and others are not.

Implications

This is the first review to focus on why NIIs are (in)effective among those in low-SES groups. Our findings emphasise the need to add an explanatory perspective to a field that primarily focuses on impact assessments. Using research designs that enable determining why an NII is (in)effective in specific target groups could, however, provide the crucial information required to develop more effective – and more equitable – interventions. Moreover, to achieve a better understanding of the mechanisms that explain why an NII is more or less impactful among lower-SES groups, it is crucial that valid arguments are provided about why a specific approach would affect the impact of an intervention and why this mechanism may be distributed differentially between socioeconomic groups. Each pathway determines the effectiveness of an NII and should, as a result, be detailed enough to enable the design of more effective campaigns. Moreover, this approach should, perhaps, not only be limited to NIIs, since any health promotion intervention would benefit from its creators knowing why and how it is likely to be (in)effective, both generally and per SES group.

Our scoping review revealed a dearth of studies that conducted a rigorous, empirical examination of whether the explanations could actually account for the effects observed. This means it is impossible to make any empirically substantiated claims about why NIIs are (in)effective among low-SES groups. Nevertheless, our review has identified the types of explanation proposed, most notably health literacy and economic resources. If these explanations do, in fact, have an empirical basis, NIIs could become more equitable by relying less on the information-processing capacity of the intended recipient, and more on increasing the affordability and availability of the touted healthy food products. Removing any cognitive and financial barriers should then lead to the creation of interventions that are more successful among lower-SES groups.

Finally, it is important to note that the explanations suggested in the studies we examined are predominantly individualistic accounts, most notably focusing on psychological and economic attributes. As such, the possible relevance of sociocultural conditions in shaping the uptake of health knowledge has not yet been covered systematically. The field may, therefore, benefit from adding non-individualistic explanations, e.g., from fields like sociology and anthropology(Reference Abel40Reference Lambert and McKevitt42).

Limitations

This research has some limitations. First, only papers printed in English were considered for inclusion, meaning that a considerable number of studies written in other languages were excluded. The inclusion of more languages may have given a more complete picture of the field. Moreover, various studies in our thematic analysis focused on actual behavioural change rather than information uptake. This could have led to the relatively high number of times that economic factors were suggested as the reason for the (in)effectiveness of the interventions investigated: acting on information probably depends more on economic resources than is the case for its uptake. Furthermore, the relationship between information uptake and behavioural change has been contested(Reference Plessz, Kesse-Guyot and Zins43), which may partly explain the relatively low impact of the analysed interventions on behavioural change.

Conclusion

Our scoping review has highlighted that only about half of the studies to examine NIIs suggest any explanations for their (in)effectiveness among those with a low SES; focusing mainly on cognitive and financial factors. Moreover, only about a third of these studies investigated empirically whether those explanations did actually account for the (in)effectiveness identified. This makes it difficult to learn lessons from past interventions. Future intervention studies should therefore focus more on establishing empirically why NIIs do, or do not, work as intended. This information is essential if we are to confront and reduce the considerable health inequalities that persist across the globe.

Supplementary material

The supplementary material for this article can be found at https://doi.org/10.1017/jns.2022.42.

Acknowledgements

The authors thank Elise Krabbendam and Wichor Bramer (Biomedical Information Specialists, Medical Library Erasmus MC) for their help with the literature search.

This work was supported by the Erasmus Initiative ‘Smarter Choices for Better Health’. The study sponsor had no role in study design, writing the report or the decision to submit the report for publication.

All the authors conceived and designed the research. TvM and JvdW conducted the study selection independently. TvM analysed the data and drafted the manuscript, while JvdW provided detailed feedback on the process. The manuscript was comprehensively and meticulously revised by all the authors. All authors approved the final manuscript.

References

Mackenbach, JP, Stirbu, I, Roskam, AJR, et al. (2008) Socioeconomic inequalities in health in 22 European countries. N Engl J Med 358, 24682481.CrossRefGoogle ScholarPubMed
Darmon, N & Drewnowski, A (2008) Does social class predict diet quality? Am J Clin Nutr 87, 11071117.CrossRefGoogle ScholarPubMed
Plessz, M & Gojard, S (2015) Fresh is best? Social position, cooking, and vegetable consumption in France. Sociology 49, 172190.CrossRefGoogle Scholar
Hartmann-Boyce, J, Bianchi, F, Piernas, C, et al. (2018) Grocery store interventions to change food purchasing behaviors: a systematic review of randomized controlled trials. Am J Clin Nutr 107, 10041016.CrossRefGoogle ScholarPubMed
Sarink, D, Peeters, A, Freak-Poli, R, et al. (2016) The impact of menu energy labelling across socioeconomic groups: a systematic review. Appetite 99, 5975.CrossRefGoogle ScholarPubMed
Glanz, K (2001) Current theoretical bases for nutrition intervention and their uses. In Nutrition in the Prevention and Treatment of Disease, pp. 8093 [Coulston, AM, Rock, CL and Monsen, ER Eds.,]. London: Academic Press.Google Scholar
Beauchamp, A, Backholer, K, Magliano, D, et al. (2014) The effect of obesity prevention interventions according to socioeconomic position: a systematic review. Obes Rev 15, 541554.CrossRefGoogle ScholarPubMed
Koç, H & van Kippersluis, H (2017) Thought for food: nutritional information and educational disparities in diet. J Hum Cap 11, 508522.CrossRefGoogle Scholar
Schiavo, R (2011) Health equity and health communication: a new dawn? J Healthc Commun 4, 6769.CrossRefGoogle Scholar
Arksey, H & O'Malley, L (2005) Scoping studies: towards a methodological framework. Int J Soc Res Methodol: Theory Pract 8, 1932.CrossRefGoogle Scholar
Davis, K, Drey, N & Gould, D (2009) What are scoping studies? A review of the nursing literature. Int J Nurs Stud 46, 13861400.CrossRefGoogle ScholarPubMed
Tricco, AC, Lillie, E, Zarin, W, et al. (2018) PRISMA extension for scoping reviews (PRISMA-ScR): checklist and explanation. Ann Intern Med 169, 467473.CrossRefGoogle ScholarPubMed
CSDH (2008) Closing the Gap in a Generation. Health Equity Through Action on the Social Determinants of Health. Geneva: World Health Organization.Google Scholar
Backman, D, Cheung, J, Francis, D, et al. (2011) Catering trucks in California promote healthful eating in low-wage worksites. J Nutr Educ Behav 43, S155S157.CrossRefGoogle ScholarPubMed
Blake, MR, Lancsar, E, Peeters, A, et al. (2018) The effect of sugar-sweetened beverage price increases and educational messages on beverage purchasing behavior among adults. Appetite 126, 156162.CrossRefGoogle ScholarPubMed
Blakely, T, Mhurchu, CN, Jiang, Y, et al. (2011) Do effects of price discounts and nutrition education on food purchases vary by ethnicity, income and education? Results from a randomised, controlled trial. J Epidemiol Community Health 65, 902908.CrossRefGoogle ScholarPubMed
Breck, A, Cantor, J, Martinez, O, et al. (2014) Who reports noticing and using calorie information posted on fast food restaurant menus? Appetite 81, 3036.CrossRefGoogle ScholarPubMed
Capacci, S & Mazzocchi, M (2011) Five-a-day, a price to pay: an evaluation of the UK program impact accounting for market forces. J Health Econ 30, 8798.CrossRefGoogle Scholar
Elbel, B, Kersh, R, Brescoll, VL, et al. (2009) Calorie labeling and food choices: a first look at the effects on low-income people in New York city: calorie information on menus appears to increase awareness of calorie content, but not necessarily the number of calories people purchase. Health Affairs 28, w1110w1121.CrossRefGoogle Scholar
Gans, KM, Risica, PM, Strolla, LO, et al. (2009) Effectiveness of different methods for delivering tailored nutrition education to low income, ethnically diverse adults. Int J Behav Nutr Phys Act 6, 111.CrossRefGoogle ScholarPubMed
Gans, KM, Risica, PM, Keita, AD, et al. (2018) Multilevel approaches to increase fruit and vegetable intake in low-income housing communities: final results of the “live well, Viva bien” cluster-randomized trial. Int J Behav Nutr Phys Act 15, 118.CrossRefGoogle ScholarPubMed
Gorton, D, Ni Mhurchu, C, Chen, MH, et al. (2009) Nutrition labels: a survey of use, understanding and preferences among ethnically diverse shoppers in New Zealand. Public Health Nutr 12, 13591365.CrossRefGoogle ScholarPubMed
Hersey, JC, Cates, SC, Blitstein, JL, et al. (2015) Eat smart, live strong intervention increases fruit and vegetable consumption among low-income older adults. J Nutr Gerontol Geriatr 34, 6680.CrossRefGoogle ScholarPubMed
Irz, X, Leroy, P, Réquillart, V, et al. (2015) Economic assessment of nutritional recommendations. J Health Econ 39, 188210.CrossRefGoogle ScholarPubMed
Mancino, L & Kuchler, F (2012) Demand for whole-grain bread before and after the release of dietary guidelines. Appl Econ Perspect Policy 34, 76101.CrossRefGoogle Scholar
Masic, U, Christiansen, P & Boyland, EJ (2017) The influence of calorie and physical activity labelling on snack and beverage choices. Appetite 112, 5258.CrossRefGoogle ScholarPubMed
McGeary, KA (2013) The impact of state-level nutrition-education program funding on BMI: evidence from the behavioral risk factor surveillance system. Soc Sci Med 82, 6778.CrossRefGoogle Scholar
Øvrum, A, Alfnes, F, Almli, VL, et al. (2012) Health information and diet choices: results from a cheese experiment. Food Policy 37, 520529.CrossRefGoogle Scholar
Rameshbabu, A (2013) Self-regulation of Saturated Fat Intake in Blue-Collar Employees: A Randomized Intervention Study. Milwaukee: University of Wisconsin-Milwaukee.Google Scholar
Schindler, J, Kiszko, K, Abrams, C, et al. (2013) Environmental and individual factors affecting menu labeling utilization: a qualitative research study. J Acad Nutr Diet 113, 667672.CrossRefGoogle ScholarPubMed
Springvloet, L, Lechner, L, De Vries, H, et al. (2015) Long-term efficacy of a web-based computer-tailored nutrition education intervention for adults including cognitive and environmental feedback: a randomized controlled trial. BMC Public Health 15, 116.CrossRefGoogle ScholarPubMed
Springvloet, L, Lechner, L, De Vries, H, et al. (2015) Short-and medium-term efficacy of a web-based computer-tailored nutrition education intervention for adults including cognitive and environmental feedback: randomized controlled trial. J Med Internet Res 17, e23.CrossRefGoogle ScholarPubMed
Springvloet, L, Lechner, L, Candel, MJJM, et al. (2016) Exploring individual cognitions, self-regulation skills, and environmental-level factors as mediating variables of two versions of a web-based computer-tailored nutrition education intervention aimed at adults: a randomized controlled trial. Appetite 98, 101114.CrossRefGoogle ScholarPubMed
Taksler, GB & Elbel, B (2014) Calorie labeling and consumer estimation of calories purchased. Int J Behav Nutr Phys Act 11, 18.CrossRefGoogle ScholarPubMed
Thunström, L (2019) Welfare effects of nudges: the emotional tax of calorie menu labeling. Judgm Decis Mak 14, 1125.Google Scholar
Verheijden, MW, van Dommelen, P, van Empelen, P, et al. (2012) Changes in self-reported energy balance behaviours and body mass index during a mass media campaign. Family Practice 29.CrossRefGoogle ScholarPubMed
Walsh, J, Kattelmann, K & White, A (2017) Stage-based healthy lifestyles program for non-college young adults. Health Educ 117, 148161.CrossRefGoogle Scholar
Walthouwer, MJL, Oenema, A, Lechner, L, et al. (2015) Comparing a video and text version of a web-based computer-tailored intervention for obesity prevention: a randomized controlled trial. J Med Internet Res 17, i75i85.Google ScholarPubMed
Springvloet, L, Willemsen, MC, Mons, U, et al. (2015) Educational differences in associations of noticing anti-tobacco information with smoking-related attitudes and quit intentions: findings from the international tobacco control Europe surveys. Health Educ Res 30, 719730.CrossRefGoogle ScholarPubMed
Abel, T (2008) Cultural capital and social inequality in health. J Epidemiol Community Health 62, e13.CrossRefGoogle ScholarPubMed
Bunton, R, Nettleton, S & Burrows, R (1995) The Sociology of Health Promotion. London: Routledge.Google Scholar
Lambert, H & McKevitt, C (2002) Anthropology in health research: from qualitative methods to multidisciplinarity. Br Med J 325, 210213.CrossRefGoogle ScholarPubMed
Plessz, M, Kesse-Guyot, E, Zins, M, et al. (2019) Poverty does not modify the association between perceived diet healthiness and adherence to nutritional guidelines in the Constances cohort (France). Appetite 138, 190197.CrossRefGoogle Scholar
Figure 0

Fig. 1. Flowchart of inclusions and exclusions.

Figure 1

Table 1. Summary table of included studies

Supplementary material: File

van Meurs et al. supplementary material

Appendix A

Download van Meurs et al. supplementary material(File)
File 26.1 KB