Context, composition and heterogeneity: using multilevel models in health research

This paper considers the use of multilevel models in health research. Attention focuses on the structure and potential of such models and particular consideration is given to their use in elucidating the importance of contextual effects in relation to individual level social and demographic factors in understanding health outcomes, health-related behaviour and health service performance. Four graphical typologies are used to outline the questions that multilevel models can address and the paper illustrates their potential by drawing on published examples in a number of different research areas.

Dietary inequalities: what is the evidence for the effect of the neighbourhood food environment?

This review summarises the evidence for inequalities in community and consumer nutrition environments from ten previous review articles, and also assesses the evidence for the effect of the community and consumer nutrition environments on dietary intake. There is evidence for inequalities in food access in the US but trends are less apparent in other developed countries. There is a trend for greater access and availability to healthy and less healthy foods relating to better and poorer dietary outcomes respectively. Trends for price show that higher prices of healthy foods are associated with better dietary outcomes. More nuanced measures of the food environment, including multidimensional and individualised approaches, would enhance the state of the evidence and help inform future interventions.

Do places matter? A multi-level analysis of regional variations in health-related behaviour in Britain

A number of commentators have argued that there is a distinctive geography of health-related behaviour. Behaviour has to be understood not only in terms of individual characteristics, but also in relation to local cultures. Places matter, and the context in which behaviour takes place is crucial for understanding and policy. Previous empirical research has been unable to operationalize these ideas and take simultaneous account of both individual compositional and aggregate contextual factors. The present paper addresses this shortcoming through a multi-level analysis of smoking and drinking behaviours recorded in a large-scale national survey. It suggests that place, expressed as regional differences, may be less important than previously implied.

Smoking and deprivation: are there neighbourhood effects?

Debate has centred on whether the character of places plays an independent role in shaping individual smoking behaviour. At the small-area scale, particular attention has focused on whether measures of neighbourhood deprivation predict an individual's smoking status independent of their own personal characteristics. This study applies multilevel modelling techniques to data from the British Health and Lifestyle Survey and ward (local neighbourhood) level deprivation scores based on four variables from the national Census. Results suggest that after taking account of a large range of individual characteristics, both as main effects and interactions, together with complex structures of between-individual variation, measures of neighbourhood deprivation continue to have an independent effect on individual smoking status. In addition, significant between-ward differences in smoking behaviour remain which cannot be explained either by population composition or ward-level deprivation. The study suggests that the character of the local neighbourhood plays a role in shaping smoking behaviour.

Health-related behaviour in context: a multilevel modelling approach

Recent attempts to place individual health-related behaviour in context have been judged largely unsuccessful. This paper examines how this situation might be improved and is especially concerned with the role of quantitative methodologies. It is argued that, whilst recent developments in social theory help provide important theoretical guidelines, they can only be implemented with difficulty in empirical health-related behaviour research if traditional quantitative methodologies are used. It is suggested that the best way to implement social theory within a quantitative framework is to apply the newly developed technique of multilevel modelling. This paper offers an overview of the multilevel approach and outlines its significance for health-related behaviour research. In addition, it details a number of ways in which the multilevel framework can be extended so as to achieve further improvements in the conceptualization of health-related behaviour. To illustrate the value of the technique, the paper finishes by considering one of these extensions in detail and applying it to data recording smoking behaviour in the United Kingdom.

Psychiatric morbidity: a multilevel approach to regional variations in the UK

STUDY OBJECTIVE

To establish whether regional variations in psychiatric morbidity in Britain constitute a distinctive geography of mental health arising from factors that are context-specific at area level or whether these variations are an artifact generated by sampling fluctuations and differing population compositions in areas.

DESIGN

Multilevel modelling techniques were applied to data from the 1984-85 health and lifestyle survey. The outcome was the prevalence of psychiatric morbidity as recorded by the application of the general health questionnaire in this survey.

SETTING

The analysis was undertaken simultaneously at the individual level, electoral ward level, and regional level for England, Wales, and Scotland.

PARTICIPANTS

A total of 6572 adults were selected from the electoral register.

MAIN RESULTS

Regional variations were detected in crude aggregate general health questionnaire scores but these were found to be the result of sampling fluctuations and varying regional population compositions rather than higher level contextual effects. There was certainly no evidence of a clear north-south distinction in psychiatric morbidity as was suggested by earlier work. In addition, the local neighbourhood did not seem to have any importance beyond the type of people who lived there. A number of individual characteristics was shown to be associated with mental wellbeing but a large degree of individual variation remained unexplained.

CONCLUSIONS

In terms of low level psychiatric disturbance it seems that the characteristics of individuals have greater importance than the characteristics of areas, although the latter may still operate as important mediating factors. Multilevel modelling represents a robust statistical method of examining area variations in health outcomes and further work needs to be conducted, particularly on more serious psychiatric conditions.

Predicting small-area health-related behaviour: a comparison of smoking and drinking indicators

Health-related behaviours are of central importance to health promotion and to the promotion of enhanced population health. In the UK, localised knowledge of the quantitative dimensions of health-related behaviours is traditionally attained by conducting a costly sample survey. Such surveys seldom generate reliable data at scales more local than that of the health authority, they also need to be repeated regularly. This paper outlines an alternative framework for generating statistics on small-area health related behaviours using routinely available data from the annual Health Survey for England (N = 17,000) and the decennial Population Census. Using a multilevel modelling approach nesting individuals within postcode sectors within health authorities, and focusing on the prevalence of smoking and 'problem' drinking, the paper comprises four sections: a consideration of the modelling strategy, a comparison of the smoking and drinking models, an outline of the estimation strategy, and the presentation and discussion of ward-level estimates of smoking and drinking behaviour for England. The paper concludes that the method is better at estimating smoking than drinking but that it offers a feasible, cheap and more informative alternative to the survey approach to the generation of information on smoking and drinking behaviour.

Risk and protection: the discourse of confinement in contemporary mental health policy

Confinement has regained respectability in the discourses of contemporary UK mental health policy. This development reflects concern about violent offences by people with mental health problems and is rooted in claims about the 'failure' of community care. Confinement is presented as a strategic response to the risks and dangers posed by particular fractions of the population of mental health service users. Using two key policy statements issued by the Department of Health and associated discussions in the health services management press, the confinement theme is explored and assessed. The paper notes its emergence as a consequence of the spatial impacts of deinstitutionalization and its specific origins in response to violent offences by people with mental health problems. The notion that the growing emphasis on confinement presages a return to the asylum is considered and rejected. Rather, the paper stresses the importance of discourses of protection, safety, risk and dangerousness in understanding the turn to confinement.

The neighbourhood effects of geographical access to tobacco retailers on individual smoking behaviour

OBJECTIVE

To investigate whether neighbourhood measures of geographical accessibility to outlets selling tobacco (supermarkets, convenience stores and petrol stations) are associated with individual smoking behaviour in New Zealand.

METHODS

Using geographical information systems, travel times from the population-weighted centroid of each neighbourhood to the closest outlet selling tobacco were calculated for all 38,350 neighbourhoods across New Zealand. These measures were appended to the 2002/03 New Zealand Health Survey, a national survey of 12, 529 adults. Two-level logistic regression models were fitted to examine the effects of neighbourhood locational access upon individual smoking behaviour after controlling for potential individual- and neighbourhood-level confounding factors, including deprivation and urban/rural status.

RESULTS

After controlling for individual-level demographic and socioeconomic variables, individuals living in the quartiles of neighbourhoods with the best access to supermarkets (OR 1.23, 95% CI 1.06 to 1.42) and convenience stores (OR 1.19, 95% CI 1.03 to 1.38) had a higher odds of smoking compared with individuals in the worst access quartiles. However, the association between neighbourhood accessibility to supermarkets and convenience stores was not apparent once other neighbourhood-level variables (deprivation and rurality) were included.

CONCLUSIONS

At the national level, there is little evidence to suggest that, after adjustment for neighbourhood deprivation, better locational access to tobacco retail provision in New Zealand is associated with individual-level smoking behaviour.

Measuring the healthfulness of food retail stores: variations by store type and neighbourhood deprivation

Background

The consumer nutrition environment has been conceptualised as in-store environmental factors that influence food shopping habits. More healthful in-store environments could be characterised as those which promote healthful food choices such as selling good quality healthy foods or placing them in prominent locations to prompt purchasing. Research measuring the full-range of in-store environmental factors concurrently is limited.

Purpose

To develop a summary score of ‘healthfulness’ composed of nine in-store factors that influence food shopping behaviour, and to assess this score by store type and neighbourhood deprivation.

Methods

A cross-sectional survey of 601 retail food stores, including supermarkets, grocery stores and convenience stores, was completed in Hampshire, United Kingdom between July 2010 and June 2011. The survey measured nine variables (variety, price, quality, promotions, shelf placement, store placement, nutrition information, healthier alternatives and single fruit sale) to assess the healthfulness of retail food stores on seven healthy and five less healthy foods that are markers of diet quality. Four steps were completed to create nine individual variable scores and another three to create an overall score of healthfulness for each store.

Results

Analysis of variance showed strong evidence of a difference in overall healthfulness by store type (p < 0.001). Large and premium supermarkets offered the most healthful shopping environments for consumers. Discount supermarkets, ‘world’, convenience and petrol stores offered less healthful environments to consumers however there was variation across the healthfulness spectrum. No relationship between overall healthfulness and neighbourhood deprivation was observed (p = 0.1).

Conclusions

A new composite measure of nine variables that can influence food choices was developed to provide an overall assessment of the healthfulness of retail food stores. This composite score could be useful in future research to measure the relationship between main food store and quality of diet, and to evaluate the effects of multi-component food environment interventions.

Defining localities for health planning: a GIS approach

The National Health Service in Britain is undergoing far-reaching changes. While District and Regional Health Authorities are currently merging, professionals agree that primary health care is most efficiently managed at the local level. This paper uses geographical information systems (GIS) capabilities to identify a nested hierarchy of localities for the management of primary health care in West Sussex, England. GIS coverages were developed which contained key criteria for defining local areas, including nodes or focal points of service provision, edges which act as physical or psychological barriers to movement, districts such as official administrative areas and interaction criteria such as journey to work, school and family doctor (GP) surgeries. Central to the derivation of the localities was a large matrix of patient to GP flows based on postcoded data. Once managed, these data revealed clear geographical patterns of patient to GP allegiance. A large-scale field survey obtained supporting information on the perception of areas from local residents.

Change in prevalence of chronic kidney disease in England over time: comparison of nationally representative cross-sectional surveys from 2003 to 2010

OBJECTIVES

To determine whether the prevalence of chronic kidney disease (CKD) in England has changed over time.

DESIGN

Cross-sectional analysis of nationally representative Health Survey for England (HSE) random samples.

SETTING

England 2003 and 2009/2010.

SURVEY PARTICIPANTS

13,896 adults aged 16+ participating in HSE, adjusted for sampling and non-response, 2009/2010 surveys combined.

MAIN OUTCOME MEASURE

Change in prevalence of estimated glomerular filtration rate (eGFR)<60 mL/min/1.73 m2 (as proxy for stage 3-5 CKD), from 2003 to 2009/2010 based on a single serum creatinine measure using an isotope dilution mass spectrometry traceable enzymatic assay in a single laboratory; eGFR derived using Modified Diet in Renal Disease (MDRD) and Chronic Kidney Disease Epidemiology Collaboration (CKDEPI) eGFR formulae.

ANALYSIS

Multivariate logistic regression modelling to adjust time changes for sociodemographic and clinical factors (body mass index, hypertension, diabetes, lipids). A correction factor was applied to the 2003 HSE serum creatinine to account for a storage effect.

RESULTS

National prevalence of low eGFR (<60) decreased within each age and gender group for both formulae except in men aged 65-74. Prevalence of obesity and diabetes increased in this period, while there was a decrease in hypertension. Adjustment for demographic and clinical factors led to a significant decrease in CKD between the surveyed periods. The fully adjusted OR for eGFR<60 mL/min/1.73 m2 was 0.75 (0.61 to 0.92) comparing 2009/2010 with 2003 using the MDRD equation, and was similar using the CKDEPI equation 0.73 (0.57 to 0.93).

CONCLUSIONS

The prevalence of a low eGFR indicative of CKD in England appeared to decrease over this 7-year period, despite the rising prevalence of obesity and diabetes, two key causes of CKD. Hypertension prevalence declined and blood pressure control improved but this did not appear to explain the fall. Periodic assessment of eGFR and albuminuria in future HSEs is needed to evaluate trends in CKD.

Fat nation: deciphering the distinctive geographies of obesity in England

Much attention is focused on obesity by both the media and by public health. As a health risk, obesity is recognised as a contributing factor to numerous health problems. Recent evidence points to a growth in levels of obesity in many countries and particular attention is usually given to rising levels of obesity among younger people. England is no exception to these generalisations with recent studies revealing a clear geography to what has been termed an 'obesity epidemic.' This paper examines the complexities inherent in the geography of adult obesity in England. Existing knowledge about the sub-national geography of obesity is examined and assessed. Multilevel synthetic estimation is then used to construct an age-sex-ethnicity disaggregated geography of obesity. These differing geographies are compared and contrasted with pre-existing findings and explored at multiple scales. A complex picture of the geography of obesity in England is revealed.

Predicting small area health-related behaviour: a comparison of multilevel synthetic estimation and local survey data

A recent paper in Social Science and Medicine (Twigg et al. 50 (2000) 1109) outlined an approach to the estimation of prevalences of small-area health-related behaviour using multilevel models. This paper compares results from the application of the multilevel approach with those derived using the more traditional strategy of the local 'lifestyle' survey. Estimations of smoking prevalence and high alcohol consumption are examined and critical assessments made of both estimation approaches. It is concluded that the alternative method is more suited towards the prediction of smoking rates as opposed to unsafe alcohol consumption.

Chronic kidney disease, albuminuria and socioeconomic status in the Health Surveys for England 2009 and 2010

BACKGROUND

Renal replacement therapy rates are inversely related to socioeconomic status (SES) in developed countries. The relationship between chronic kidney disease (CKD) and SES is less clear. This study examined the relationships between SES and CKD and albuminuria in England.

METHODS

Data from the Health Survey for England 2009 and 2010 were combined. The prevalence of CKD 3-5 and albuminuria was calculated, and logistic regression used to determine their association with five individual-level measures and one area-level measure of SES.

RESULTS

The prevalence of CKD 3-5 was 5.2% and albuminuria 8.0%. Age-sex-adjusted CKD 3-5 was associated with lack of qualifications [odds ratio (OR) 2.27 (95% confidence interval 1.40-3.69)], low income [OR 1.50 (1.02-2.21)] and renting tenure [OR 1.36 (1.01-1.84)]. Only tenure remained significant in fully adjusted models suggesting that co-variables were on the causal pathway. Albuminuria remained associated with several SES measures on full adjustment: low income [OR 1.55 (1.14-2.11)], no vehicle [OR 1.38 (1.05-1.81)], renting [OR 1.31 [1.03-1.67)] and most deprived area-level quintile [OR 1.55 (1.07-2.25)].

CONCLUSIONS

CKD 3-5 and albuminuria were associated with low SES using several measures. For albuminuria this was not explained by known measured causal factors.

Activation of caspase cascades in Korean mistletoe (Viscum album var. coloratum) lectin-II-induced apoptosis of human myeloleukemic U937 cells

Mistletoe lectins are of high biological activity and exert cytotoxic effects. We have previously shown that Korean mistletoe, Viscum album var. coloratum, lectin-II specifically induces apoptotic cell death in cancer cells, not normal lymphocytes. The destructive mechanism by mistletoe lectins on tumor cells was mediated by activation of c-JUN N-terminal kinase (JNK)/stress-activated protein kinase. Herein, we investigated the involvement of caspase cascade and its proteolytic cleavage effects on biosubstrates of human myeloleukemic U937 cells by D-galactoside and N-acetyl-galactosamine-specific Korean mistletoe lectin-II. Mistletoe lectin-II induced ladder pattern DNA fragmentation and activation of caspase-3, -8, and -9 of U937 cells, but not caspase-1 protease, in a time- and dose-dependent manner. Consistent with catalytic activation of protease, both poly(ADP-ribose) polymerase (PARP) and protein kinase C-delta (PKC-delta) are also cleaved in mistletoe lectin-II-treated U937 cells. An inhibitor of caspase-3-like protease, DEVD-CHO peptide, significantly inhibited mistletoe lectin-II-induced apoptosis, PARP cleavage, and fragmentation of DNA. These results provide the evidence that Korean mistletoe lectin-II induces apoptotic death of U937 cells via activation of caspase cascades.

VLSI implementation of synaptic weighting and summing in pulse coded neural-type cells

Presents the hardware realization for synaptic weighting and summing using pulse-coded neural-type cells (NTCs). The basic information processing element (NTC) encodes the information into the form of pulse duty cycles using voltage-controlled resistors, for which a pulse duty cycle modulation technique is proposed. Summation is executed by a simple capacitor circuit as a current integrator. Layouts and measurements on a fabricated integrated design are included.

The relationship between dietary quality and the local food environment differs according to level of educational attainment: A cross-sectional study

There is evidence that food outlet access differs according to level of neighbourhood deprivation but little is known about how individual circumstances affect associations between food outlet access and diet. This study explored the relationship between dietary quality and a measure of overall food environment, representing the balance between healthy and unhealthy food outlet access in individualised activity spaces. Furthermore, this study is the first to assess effect modification of level of educational attainment on this relationship. A total of 839 mothers with young children from Hampshire, United Kingdom (UK) completed a cross-sectional survey including a 20-item food frequency questionnaire to measure diet and questions about demographic characteristics and frequently visited locations including home, children’s centre, general practitioner, work, main food shop and physical activity location. Dietary information was used to calculate a standardised dietary quality score for each mother. Individualised activity spaces were produced by creating a 1000m buffer around frequently visited locations using ArcGIS. Cross-sectional observational food outlet data were overlaid onto activity spaces to derive an overall food environment score for each mother. These scores represented the balance between healthy and unhealthy food outlets using weightings to characterise the proportion of healthy or unhealthy foods sold in each outlet type. Food outlet access was dominated by the presence of unhealthy food outlets; only 1% of mothers were exposed to a healthy overall food environment in their daily activities. Level of educational attainment moderated the relationship between overall food environment and diet (mid vs low, p = 0.06; high vs low, p = 0.04). Adjusted stratified linear regression analyses showed poorer food environments were associated with better dietary quality among mothers with degrees (β = -0.02; 95%CI: -0.03, -0.001) and a tendency toward poorer dietary quality among mothers with low educational attainment, however this relationship was not statistically significant (β = 0.01; 95%CI: -0.01, 0.02). This study showed that unhealthy food outlets, like takeaways and convenience stores, dominated mothers’ food outlet access, and provides some empirical evidence to support the concept that individual characteristics, particularly educational attainment, are protective against exposure to unhealthy food environments. Improvements to the imbalance of healthy and unhealthy food outlets through planning restrictions could be important to reduce dietary inequalities.

Social inequality and ethnic differences in smoking in New Zealand

This study tests a generalisation of the 'Wilkinson' thesis that the greater a nation's income inequality, the poorer the average national health status. We consider the effect of socio-economic inequality upon ethnic variations in smoking in New Zealand. Analysis of Maori and Pakeha (New Zealanders of European descent) smoking rates from the 1996 Census is conducted for 73 Territorial Local Authority areas in New Zealand, disaggregated by gender and rural-urban location. Partial correlation is used to control for absolute levels of deprivation and examine the independent effect of ethnic social inequality upon smoking rates. The level of social inequality between Maori and Pakeha has an independent effect on Maori smoking rates. Pakeha smoking rates by contrast are more sensitive to variations in absolute rather than relative deprivation. The effect of inequality is greatest for Maori women, especially among urban residents. By contrast, among Maori men the effects are greatest in rural areas. The results provide some qualified support for the Wilkinson thesis and suggest that policies which address fundamental issues of social inequality will play a small, but significant, role in helping to reduce high smoking rates amongst Maori.

Conceptions of space and community in British health policy

Several recent United Kingdom health policy initiatives include ideas implicitly or even explicitly involving geographical space as a central theoretical construct. Neighbourhood nursing is perhaps the best known recent example. These initiatives are not without common features, and the paper commences with a typology of the roles which geographical space plays in health policy. A second section gives specific consideration to neighbourhood nursing. Thirdly the paper outlines the social theoretic debates surrounding geographic inputs to health policy: ideas such as community and locality. It is suggested that spatial conceptions in health policy reflect a complex amalgam of sociological assumptions which might fruitfully be considered in the light of Giddens' concept of locale.

Does social inequality matter? Changing ethnic socio-economic disparities and Maori smoking in New Zealand, 1981-1996

This paper builds on an earlier study of the effects of inequality on smoking by explicitly incorporating a temporal dimension. We examine the effects of changing levels of inequality upon ethnic variations in smoking rates in New Zealand for the period 1981 to 1996. This was a period of rapid structural change in New Zealand's economy and welfare state, changes which had a disproportionate effect on Maori. While Maori smoking rates declined during this period, the gap in smoking levels between Maori and Pakeha (persons of European descent) increased. The results suggest that levels of social inequality between Maori and Pakeha have an independent effect on Maori smoking rates and that communities which experienced increased social inequality during both the 1980s and 1990s were more likely to have higher Maori smoking rates. Controlling for confounders, the effect of increased ethnic inequality on smoking was particularly evident for Maori women (net R(2) = 0.150) compared to Maori men (net R(2) = 0.079). Nevertheless, absolute rather than relative socio-economic deprivation remains the most important predictor of smoking.

Assessing the effects of the introduction of the New Zealand Smokefree Environment Act 2003 on acute myocardial infarction hospital admissions in Christchurch, New Zealand

OBJECTIVE

To examine trends in Acute Myocardial Infarction (AMI) hospital admissions in Christchurch, New Zealand before and after the implementation of the New Zealand Smokefree Environments Act 2003 in December 2004.

METHODS

Data on AMI hospital admissions to Christchurch Public Hospital were extracted for the period 2003 to 2006. Poisson regression was used to calculate rate ratios by comparing for AMI rates of hospital admissions before (2003/04) and after (2005/06) the introduction of the Smokefree legislation, and to assess whether there was a significant change over time.

RESULTS

The introduction of the smokefree legislation was associated with a 5% reduction in AMI admissions. The 55-74 age group recorded the greatest decrease in admissions (9%) and this figure rose to 13% among never smokers in this group. Reductions were more marked for men. Adding the effects of area deprivation increased the reduction to 21% among 55-74 year olds living in more affluent (quintile 2) areas. Overall however, the statistical association of changing levels of AMI admissions with smoking status and with deprivation was not consistently significant.

CONCLUSION

At this early stage following the smokefree legislation, there are hints emerging of a positive impact on AMI admissions but these suggestions cannot yet be treated with certainty. Further research could usefully evaluate the longer-term effects of smoking legislation on the prevalence of smoking and exposure to second hand smoke, especially in more deprived urban communities.

Geographies of need and the new NHS: methodological issues in the definition and measurement of the health needs of local populations

For a variety of purposes it is increasingly necessary to establish the health needs of local populations. Following a critique of existing proxies of need, this paper presents and evaluates an alternative which draws upon epidemiological evidence concerning the age, sex and social class distribution of morbidity in order to estimate the prevalence of specific conditions in designated populations. By way of demonstrating the insights to be gained through the use of these indicative prevalence rates, the paper considers the treatment of ischaemic heart disease in a sample of 539 practices and presents evidence regarding the significance of deprivation and rurality in determining health service use relative to needs.

Education and the Relationship Between Supermarket Environment and Diet

INTRODUCTION

Supermarkets are a major source of food for many families. Knowledge of how educational attainment affects the relationship between in-store environments of supermarkets and diet is needed. This study examined the relationship between maternal dietary quality and overall in-store supermarket environment, and assessed the effect modification of educational attainment.

METHODS

Dietary quality z-scores were calculated for 829 mothers with young children using cross-sectional data collected in 2010-2011 from a 20-item food frequency questionnaire. Information about nine in-store factors (variety, price, quality, promotion, shelf placement, store placement, nutrition information, healthier alternatives, and single fruit sale) on 12 foods known to discriminate between better and poorer dietary quality were collected to create a standardized "healthfulness" z-score for each supermarket where mothers shopped.

RESULTS

Multilevel unadjusted linear regression analysis completed in 2014-2015 showed that shopping at more-healthful supermarkets was associated with better dietary quality (β=0.39 SD/SD, p=0.01, 95% CI=0.10, 0.68). However, the relationship differed according to educational attainment (interaction, p=0.006). Among mothers who left school at age 16 years, those who shopped at less healthful supermarkets had poorer dietary quality (β=0.31 SD/SD, 95% CI=0.07, 0.55). Among mothers with degrees, those who shopped at less healthful supermarkets had better dietary quality (β=-0.59 SD/SD, 95% CI=-1.19, 0.00).

CONCLUSIONS

Mothers with low educational attainment show greater susceptibility to less healthful in-store environments than mothers with higher educational attainment who may be protected by greater psychological and financial resources. Policy initiatives to improve supermarket environments may be necessary to address dietary inequalities.