Advertisement

Accessibility and Affordability of Supermarkets: Associations With the DASH Diet

  • Joreintje D. Mackenbach
    Affiliations
    Department of Epidemiology and Biostatistics, Amsterdam Public Health Research Institute, VU Medical Center Amsterdam, Amsterdam, the Netherlands
    Search for articles by this author
  • Thomas Burgoine
    Affiliations
    UKCRC Centre for Diet and Activity Research (CEDAR), MRC Epidemiology Unit, University of Cambridge School of Clinical Medicine, Institute of Metabolic Science, Cambridge, United Kingdom
    Search for articles by this author
  • Jeroen Lakerveld
    Affiliations
    Department of Epidemiology and Biostatistics, Amsterdam Public Health Research Institute, VU Medical Center Amsterdam, Amsterdam, the Netherlands
    Search for articles by this author
  • Nita G. Forouhi
    Affiliations
    UKCRC Centre for Diet and Activity Research (CEDAR), MRC Epidemiology Unit, University of Cambridge School of Clinical Medicine, Institute of Metabolic Science, Cambridge, United Kingdom
    Search for articles by this author
  • Simon J. Griffin
    Affiliations
    UKCRC Centre for Diet and Activity Research (CEDAR), MRC Epidemiology Unit, University of Cambridge School of Clinical Medicine, Institute of Metabolic Science, Cambridge, United Kingdom

    Primary Care Unit, Department of Health and Primary Care, University of Cambridge, Cambridge, United Kingdom
    Search for articles by this author
  • Nicholas J. Wareham
    Affiliations
    UKCRC Centre for Diet and Activity Research (CEDAR), MRC Epidemiology Unit, University of Cambridge School of Clinical Medicine, Institute of Metabolic Science, Cambridge, United Kingdom
    Search for articles by this author
  • Pablo Monsivais
    Correspondence
    Address correspondence to: Pablo Monsivais, PhD, MPH, UKCRC Centre for Diet and Activity Research (CEDAR), MRC Epidemiology Unit, University of Cambridge School of Clinical Medicine, Box 285, Institute of Metabolic Science, Cambridge Biomedical Campus, Cambridge CB2 0QQ United Kingdom
    Affiliations
    UKCRC Centre for Diet and Activity Research (CEDAR), MRC Epidemiology Unit, University of Cambridge School of Clinical Medicine, Institute of Metabolic Science, Cambridge, United Kingdom
    Search for articles by this author
Open AccessPublished:March 20, 2017DOI:https://doi.org/10.1016/j.amepre.2017.01.044

      Introduction

      It is unknown whether there is an interplay of affordability (economic accessibility) and proximity (geographic accessibility) of supermarkets in relation to having a Dietary Approaches to Stop Hypertension (DASH)-accordant diet.

      Methods

      Data (collected: 2005–2015, analyzed: 2016) were from the cross-sectional, population-based Fenland Study cohort: 9,274 adults aged 29–64 years, living in Cambridgeshire, United Kingdom. Dietary quality was evaluated using an index of DASH dietary accordance, based on recorded consumption of foods and beverages in a validated 130-item, semi-quantitative food frequency questionnaire. DASH accordance was defined as a DASH score in the top quintile. Dietary costs (£/day) were estimated by attributing a food price variable to the foods consumed according to the questionnaire. Individuals were classified as having low-, medium-, or high-cost diets. Supermarket affordability was determined based on the cost of a 101-item market basket. Distances between home address to the nearest supermarket (geographic accessibility) and nearest economically-appropriate supermarket (economic accessibility) were divided into tertiles.

      Results

      Higher-cost diets were more likely to be DASH-accordant. After adjustment for key demographics and exposure to other food outlets, individuals with lowest economic accessibility to supermarkets had lower odds of being DASH-accordant (OR=0.59, 95% CI=0.52, 0.68) than individuals with greatest economic accessibility. This association was stronger than with geographic accessibility alone (OR=0.85, 95% CI=0.74, 0.98).

      Conclusions

      Results suggest that geographic and economic access to food should be taken into account when considering approaches to promote adherence to healthy diets for the prevention of cardiovascular diseases and other chronic disease.

      Introduction

      The consumption of a healthy diet is a priority for reducing obesity and risk of chronic diseases.
      • Lock K.
      • Pomerleau J.
      • Causer L.
      • Altmann D.R.
      • McKee M.
      The global burden of disease attributable to low consumption of fruit and vegetables: implications for the global strategy on diet.
      • Mozaffarian D.
      • Appel L.J.
      • Van Horn L.
      Components of a cardioprotective diet: new insights.
      The Dietary Approaches to Stop Hypertension (DASH) diet emphasizes fruits, vegetables, whole grains, and low-fat dairy products, and is relatively low in red meats and refined carbohydrates.
      • Karanja N.
      • Erlinger T.P.
      • Pao-Hwa L.
      • Miller E.R.
      • Bray G.A.
      The DASH diet for high blood pressure: from clinical trial to dinner table.
      The DASH diet has shown to lower blood pressure and decrease blood lipids in clinical trials, and is associated with lower risk of cardiovascular disease, including coronary heart disease, stroke, and heart failure.
      • Karanja N.
      • Erlinger T.P.
      • Pao-Hwa L.
      • Miller E.R.
      • Bray G.A.
      The DASH diet for high blood pressure: from clinical trial to dinner table.
      • Chiu S.
      • Bergeron N.
      • Williams P.T.
      • Bray G.A.
      • Sutherland B.
      • Krauss R.M.
      Comparison of the DASH (Dietary Approaches to Stop Hypertension) diet and a higher-fat DASH diet on blood pressure and lipids and lipoproteins: a randomized controlled trial.
      • Salehi-Abarqouei A.
      • Maghsoudi Z.
      • Shirani F.
      • Azadbakht L.
      Effects of Dietary Approaches to Stop Hypertension (DASH)-style diet on fatal or nonfatal cardiovascular disease-incidence: a systematic review and meta-analysis on observational prospective studies.
      Despite recommendations, widespread adoption of, and long-term adherence to, DASH have been limited.
      Diet substantially contributes to socioeconomic inequalities in cardiovascular morbidity and mortality.
      • Méjean C.
      • Droomers M.
      • van der Schouw Y.T.
      • et al.
      The contribution of diet and lifestyle to socioeconomic inequalities in cardiovascular morbidity and mortality.
      It is therefore important to get a better understanding of the barriers for adherence to the DASH diet. Reported individual-level barriers include: increased preparation time required to produce healthier meals, low quality of fruits and vegetables available to consumers, perceived lack of availability of (access to) healthy foods, and preferences for foods that are inconsistent with DASH.
      • Bertoni A.G.
      • Foy C.G.
      • Hunter J.C.
      • Quandt S.A.
      • Vitolins M.Z.
      • Whitt-Glover M.C.
      A multilevel assessment of barriers to adoption of Dietary Approaches to Stop Hypertension (DASH) among African Americans of low socioeconomic status.
      • Young C.M.
      • Batch B.C.
      • Svetkey L.P.
      Effect of socioeconomic status on food availability and cost of the dietary approaches to stop hypertension (DASH) dietary pattern.
      Healthy, DASH-accordant foods are also perceived to be expensive,
      • Young C.M.
      • Batch B.C.
      • Svetkey L.P.
      Effect of socioeconomic status on food availability and cost of the dietary approaches to stop hypertension (DASH) dietary pattern.
      and several studies confirm that adherence to DASH-accordant diets is associated with higher dietary costs.
      • Monsivais P.
      • Scarborough P.
      • Lloyd T.
      • et al.
      Greater accordance with the Dietary Approaches to Stop Hypertension dietary pattern is associated with lower diet-related greenhouse gas production but higher dietary costs in the United Kingdom.
      • Monsivais P.
      • Rehm C.D.
      • Drewnowski A.
      The DASH diet and diet costs among ethnic and racial groups in the United States.
      There is also evidence that neighborhood food environments, in particular supermarket access, may be important for diet and health. Cross-sectional observational studies, mostly conducted in the U.S., have found positive associations among the geographic accessibility of supermarkets, diet quality, and weight.
      • Laraia B.A.
      • Siega-Riz A.M.
      • Kaufman J.S.
      • Jones S.J.
      Proximity of supermarkets is positively associated with diet quality index for pregnancy.
      • Morland K.
      • Diez Roux A.V.
      • Wing S.
      Supermarkets, other food stores, and obesity: the atherosclerosis risk in communities study.
      • Robinson P.L.
      • Dominquez F.
      • Teklehaimanot S.
      • Lee M.
      • Brown A.
      • Goodchild M.
      Does distance decay modelling of supermarket accessibility predict fruit and vegetable intake by individuals in a large metropolitan area?.
      However, apart from an Irish study showing an association between DASH accordance and supermarket proximity,
      • Layte R.
      • Harrington J.
      • Sexton E.
      • Perry I.J.
      • Cullinan J.
      • Lyons S.
      Irish exceptionalism? Local food environments and dietary quality.
      studies from outside the U.S. have not found that geographic access plays a significant role in diet and weight.
      • Cummins S.
      • Flint E.
      • Matthews S.A.
      New neighborhood grocery store increased awareness of food access but did not alter dietary habits or obesity.
      • Cobb L.K.
      • Appel L.J.
      • Franco M.
      • Jones-Smith J.C.
      • Nur A.
      • Anderson C.A.M.
      The relationship of the local food environment with obesity: a systematic review of methods, study quality, and results.
      The lack of consistent findings may be due to the importance of economic factors, which are often unaccounted for. Although absence of supermarkets may affect diet and health, where supermarkets are present, economic inaccessibility of healthful foods may still be critical, especially for low socioeconomic groups.
      • Short A.
      • Guthman J.
      • Raskin S.
      Food deserts, oases, or mirages? Small markets and community food security in the San Francisco Bay area.
      • Breyer B.
      • Voss-Andreae A.
      Food mirages: geographic and economic barriers to healthful food access in Portland, Oregon.
      Recent studies suggest that taking affordability into account dramatically changes the definition of access to healthy foods.
      • Breyer B.
      • Voss-Andreae A.
      Food mirages: geographic and economic barriers to healthful food access in Portland, Oregon.
      • Jiao J.
      • Moudon A.V.
      • Ulmer J.
      • Hurvitz P.M.
      • Drewnowski A.
      How to identify food deserts: measuring physical and economic access to supermarkets in King County, Washington.
      Those with lower SES, or with lower food budgets, often use lower-cost supermarket chains
      • Pechey R.
      • Monsivais P.
      Supermarket choice, shopping behavior, socioeconomic status, and food purchases.
      • Drewnowski A.
      • Aggarwal A.
      • Hurvitz P.M.
      • Monsivais P.
      • Moudon A.V.
      Obesity and supermarket access: proximity or price?.
      and may have to travel beyond their nearest supermarket to reach an affordable supermarket.
      • Aggarwal A.
      • Cook A.J.
      • Jiao J.
      • Seguin R.A.
      • Vernez Moudon A.
      • Hurvitz P.M.
      Access to supermarkets and fruit and vegetable consumption.
      This may partly explain findings from natural experiments that improving geographic access to supermarkets, without addressing economic access, had little impact on dietary behaviors or obesity.
      • Cummins S.
      • Petticrew M.
      • Higgins C.
      • Findlay A.
      • Sparks L.
      Large scale food retailing as an intervention for diet and health: quasi-experimental evaluation of a natural experiment.
      • Wrigley N.
      • Warm D.
      • Margetts B.
      • Whelan A.
      Assessing the impact of improved retail access on diet in a “food desert”: a preliminary report.
      • Sharkey J.R.
      • Horel S.
      • Han D.
      • Huber J.C.
      Association between neighborhood need and spatial access to food stores and fast food restaurants in neighborhoods of colonias.
      Yet, empirical evidence on the relative importance of geographic and economic access to supermarkets in relation to individual dietary patterns is scarce.
      The aim of this study was to assess the importance of geographic and economic access to supermarkets and their interplay for the DASH diet, hypothesizing that geographic accessibility to supermarkets would be more strongly associated with diet when taking the price of supermarket foods into account.

      Methods

      Study Sample

      The Fenland Study is a population-based cohort study of adults born between 1950 and 1975 and registered with general practices in Cambridgeshire, United Kingdom.
      • De Lucia Rolfe E.
      • Loos R.J.F.
      • Druet C.
      • et al.
      Association between birth weight and visceral fat in adults.
      The study was conducted by the Medical Research Council Epidemiology Unit. Participants attended one of three testing sites in Cambridgeshire for detailed anthropometric measurements by trained research staff. Participants completed a general lifestyle questionnaire and a food frequency questionnaire (FFQ) for assessment of habitual diet. Exclusion criteria for the Fenland Study included pregnancy, previously diagnosed diabetes, inability to walk unaided, psychosis, or terminal illness. Recruitment for the study started in 2005 and continued until 2015. At the time of this analysis, data on 11,857 participants were available (27% response rate). All study procedures were approved by the Health Research Authority National Research Ethics Service Committee East of England-Cambridge Central, and participants provided written informed consent.
      Sex, age, and household size were captured in the Fenland general lifestyle questionnaire. Highest educational attainment was also captured, and stratified into low education (≤11 years of education), medium education (12–13 years of education), and high education (>13 years of education). Household income per year was measured in three groups: <£20,000, £20,000–40,000, and >£40,000. Participants self-reported smoking status as current, former, or never.

      Measures

      Participants recorded the frequency and portions of consumption of foods and beverages by completing a validated 130-item, semi-quantitative FFQ.
      • Bingham S.A.
      • Gill C.
      • Welch A.
      • et al.
      Validation of dietary assessment methods in the UK arm of EPIC using weighed records, and 24-hours urinary nitrogen and potassium and serum vitamin C and carotenoids as biomarkers.
      Overall dietary quality was evaluated using an index of dietary accordance with the DASH diet, adapted from that of Fung et al.
      • Monsivais P.
      • Scarborough P.
      • Lloyd T.
      • et al.
      Greater accordance with the Dietary Approaches to Stop Hypertension dietary pattern is associated with lower diet-related greenhouse gas production but higher dietary costs in the United Kingdom.
      • Fung T.T.
      • Chiuve S.E.
      • McCullough M.L.
      • Rexrode K.M.
      • Logroscino G.
      • Hu F.B.
      Adherence to a DASH-style diet and risk of coronary heart disease and stroke in women.
      The index consists of eight dietary components (grains/grain products, vegetables, fruits, low-fat/fat-free dairy, red and processed meat, nuts/seeds/dry beans, dietary sodium, and foods high in added sugar).
      • Monsivais P.
      • Scarborough P.
      • Lloyd T.
      • et al.
      Greater accordance with the Dietary Approaches to Stop Hypertension dietary pattern is associated with lower diet-related greenhouse gas production but higher dietary costs in the United Kingdom.
      Energy-adjusted residuals
      • Willett W.C.
      • Howe R.
      Adjustment for total energy intake in epidemiologic studies.
      of the eight components were divided into quintiles. By summing individual quintile scores, the overall DASH scores ranged between 8 (least healthy) and 40 (most healthy). DASH accordance was defined as a DASH score in the top quintile (>28 in this cohort).
      • Fung T.T.
      • Chiuve S.E.
      • McCullough M.L.
      • Rexrode K.M.
      • Logroscino G.
      • Hu F.B.
      Adherence to a DASH-style diet and risk of coronary heart disease and stroke in women.
      • Sherzai A.
      • Heim L.T.
      • Boothby C.
      • Sherzai A.D.
      Stroke, food groups, and dietary patterns: a systematic review.
      To rank individuals based on their overall dietary costs, individual dietary costs were measured by attributing a food price variable to the foods consumed according to the FFQ. This measure of dietary cost is positively correlated with actual food expenditures, which has been described in detail elsewhere.
      • Monsivais P.
      • Perrigue M.M.
      • Adams S.L.
      • Drewnowski A.
      Measuring diet cost at the individual level: a comparison of three methods.
      • Aaron G.J.
      • Keim N.L.
      • Drewnowski A.
      • Townsend M.S.
      Estimating dietary costs of low-income women in California: a comparison of 2 approaches.
      • Timmins K.A.
      • Morris M.A.
      • Hulme C.
      • Edwards K.L.
      • Clarke G.P.
      • Cade J.E.
      Comparability of methods assigning monetary costs to diets: derivation from household till receipts versus cost database estimation using 4-day food diaries.
      Briefly, retail prices for each of the foods in the FFQ were obtained from five major UK supermarket chains in 2012, which together had a majority market share in that year.

      BBC News. Tesco market share dips below 30%. www.bbc.co.uk/news/business-16817254. Published January 31, 2012. Accessed October 16, 2015.

      Food prices were combined with the Fenland FFQ food and nutrient database, which allowed for the derivation of dietary costs (£/day, crude diet cost) for each participant. The variable “energy-adjusted dietary costs” was created by energy-adjusting daily (crude) diet costs on the method of residuals,
      • Willett W.
      • Stampfer M.
      Implications for total energy intake for epidemiologic analyses.
      a standard energy adjustment in epidemiologic studies.
      • Monsivais P.
      • Perrigue M.M.
      • Adams S.L.
      • Drewnowski A.
      Measuring diet cost at the individual level: a comparison of three methods.
      Dietary costs were divided into tertiles, with those in the third tertile having the most expensive diets.
      Participants’ home addresses were mapped by postcode using GIS in ArcGIS, version 10. Accurate data on food outlet locations were sourced from ten local councils covering the Cambridgeshire study area in December 2011
      • Lake A.A.
      • Burgoine T.
      • Greenhalgh F.
      • Stamp E.
      • Tyrrell R.
      The foodscape: classification and field validation of secondary data sources.
      and mapped by postcode. The distance between home address and nearest supermarket of any type was calculated along the street network using the Ordnance Survey Integrated Transport Network. This measure of geographic access to supermarkets was divided into tertiles, with those in Tertile 1 having the shortest distance to their nearest supermarket.
      Then, supermarkets were classified into one of four economic tiers, based on a 101-item market basket (details described in Appendix Methods and Appendix Table 1, available online). Data on food prices from seven chain supermarkets in Cambridge were collected between December 2013 and January 2014. General Linear Model repeated measures pairwise comparisons identified four supermarket tiers. Given the small number of lowest-cost (n=5) and second lowest-cost supermarkets (n=6), these categories were combined. Using tertiles of individual dietary costs and the three supermarket cost tiers, individuals with the highest dietary costs were matched to any supermarket, and individuals with the lowest dietary costs were matched with lowest-cost supermarkets. Distance between home address and the nearest economically-matched supermarket was calculated along the street network. This measure of economic access was also divided into tertiles, with those in Tertile 1 having the shortest distance to their nearest economically-matched supermarket. Appendix Figure 2 (available online) illustrates how distance to the nearest supermarket may differ from distance to the nearest economically-matched supermarket.

      Statistical Analysis

      Data were analyzed in 2016. Characteristics of individuals in three tertiles of diet cost were assessed using ANOVA and chi-square tests. DASH accordance was compared between men and women, lower and higher education, and lower and higher income.
      Associations between tertiles of dietary costs and DASH accordance were examined using multiple logistic regression models. Similarly, associations between tertiles of distance to the nearest and nearest economically-matched supermarket and DASH accordance were examined using multiple logistic regression models. The addition of covariates into the models was theoretically informed a priori and included the following variables: age, sex, highest educational qualification, car ownership, and total energy intake. Associations between tertiles of dietary costs and DASH accordance were additionally adjusted for income and distance to the nearest supermarket. As both Breyer and Voss-Andreae
      • Breyer B.
      • Voss-Andreae A.
      Food mirages: geographic and economic barriers to healthful food access in Portland, Oregon.
      and Jiao and colleagues
      • Jiao J.
      • Moudon A.V.
      • Ulmer J.
      • Hurvitz P.M.
      • Drewnowski A.
      How to identify food deserts: measuring physical and economic access to supermarkets in King County, Washington.
      conducted their studies in low-income groups, educational attainment and income were tested for as potential moderators, but evidence for such interaction was not found. In the third model, analyses were additionally adjusted for exposure to other food outlets (restaurants, convenience stores, cafes, entertainment venues selling food, specialist retailers [e.g., butchers, fishmongers], fast food outlets, and other supermarkets) within a 1-mile (1.6-km) radius Euclidean buffer of the nearest supermarket. Adjustment for the availability of other food outlets was to account for food environment “context” as often different types of food outlets (that allow healthy and unhealthy foods to be purchased) are co-located. Sensitivity analyses additionally adjusted for marital status, ethnicity, mode of transport to work, vegetarianism/veganism, alcohol consumption, and year of attendance at the clinical testing site, but this did not affect the results (data not shown).
      Given the low prevalence of missing data, complete case analyses were conducted. Those living outside the Cambridgeshire study area (not allowing for supermarket proximity to be calculated, n=1,656), participants with extreme values for energy intake based on sex-specific cut offs as suggested by Willett
      • Willett W.
      (n=273), and participants with missing data on covariates (n=204) were excluded. This resulted in an analytic sample of 9,724 participants (Appendix Figure 3, available online, provides a participant flow chart). The analytic sample was slightly younger (aged 48.2 years vs 49.5 years, p<0.001) and less likely to be male (48.1% vs 51.9%, p=0.001), but did not differ in terms of DASH scores (24.0 vs 24.0, p=0.97) compared to the full sample. A two-sided α level of 0.05 was used to test for statistical significance. All analyses were conducted using SPSS, version 22.

      Results

      Mean (SD) age of the participants was 48.2 (7.3) years and 48.1% were men (Table 1). The mean (SD) estimated dietary costs were £4.21 (£1.25)/day. Participants with lower dietary costs had lower incomes and lower educational attainment and were less likely to have DASH-accordant diets, relative to participants with higher dietary costs. Mean distance to any supermarket was similar for those with low and high dietary costs (3.8 km), but mean distance to the nearest economically-matched supermarket was substantially larger for those with low dietary costs (10.8 km). Overall, 17.3% of the sample consumed diets that were DASH-accordant (DASH score >28), with the percentage higher for women than men (22.1% vs 12.1%, respectively). Individuals with lower income and lower educational attainment were less likely to have DASH-accordant diets (Table 2). Individuals with DASH-accordant diets had lower prevalence of hypertension and lower mean BMI compared with those with less DASH-accordant diets (Appendix Table 2, available online).
      Table 1Characteristics of Individuals Across Tertiles of Energy-Adjusted Dietary Costs in the Fenland Study (n=9,724)
      VariableDietary costs
      T1 (n=3,241)T2 (n=3,242)T3 (n=3,241)Overall (n=9,724)
      Age, years, M (SD)47.3 (7.3)48.0 (7.2)49.2 (7.2)48.2 (7.3)
      Sex (% male)58.146.639.548.1
      Dietary costs, £
      In 2012, GBP£1=USD$1.61.
      /day, M (SD)
      3.40 (0.91)4.02 (0.85)5.22 (1.19)4.21 (1.25)
      Household income, <£20,000
      In 2012, GBP£1=USD$1.61.
      /year, %
      14.110.910.311.7
      Educational attainment, ≤11 years of education, %23.019.119.720.6
      Smoking, current smoker, %13.210.411.611.7
      Car ownership, %92.094.695.394.0
      Fruit intake, g/day, M (SD)175.3 (135.2)233.3 (156.0)303.6 (259.7)246.4 (202.0)
      Vegetable intake, g/day, M (SD)206.3 (93.3)273.0 (107.2)365.3 (177.3)280.0 (145.1)
      Energy intake, kcal/day, M (SD)1,971 (597)1,859 (529)1,952 (537)1,927 (557)
      DASH score,
      DASH scores range from 8 to 40.
      M (SD)
      22.7 (4.5)24.0 (4.5)25.3 (4.5)24.0 (4.7)
      Achieving DASH accordance,
      DASH score of >28 (fifth quintile); T1=tertile 1 (lowest dietary costs) and T3=tertile 3 (highest dietary costs).
      %
      10.017.424.517.3
      Geographic access supermarket,
      Defined as street network distance to the nearest supermarket of any type.
      km, M (SD)
      3.8 (3.6)3.8 (3.7)3.8 (3.6)3.8 (3.6)
      Economic access supermarket,
      Defined as street network distance to the nearest economically-matched supermarket, based on supermarket price and dietary costs. DASH, Dietary Approaches to Stop Hypertension; T, tertile.
      km, M (SD)
      10.8 (8.9)5.1 (3.9)3.8 (3.6)6.6 (6.7)
      a In 2012, GBP£1=USD$1.61.
      b DASH scores range from 8 to 40.
      c DASH score of >28 (fifth quintile); T1=tertile 1 (lowest dietary costs) and T3=tertile 3 (highest dietary costs).
      d Defined as street network distance to the nearest supermarket of any type.
      e Defined as street network distance to the nearest economically-matched supermarket, based on supermarket price and dietary costs.DASH, Dietary Approaches to Stop Hypertension; T, tertile.
      Table 2DASH Accordance by Sex, Educational Attainment, and Household Income in the Fenland Study (n=9,724)
      Variablen (%)DASH score, M (SD)% DASH-accordant
      DASH score >28 (fifth quintile).
      Achieving DASH accordance
      OR (95% CI)p-value
      Sex
       Male4,673 (48.1)23.0 (4.6)12.10.47 (0.42, 0.52)<0.001
       Female5,051 (51.9)25.0 (4.5)22.11
      Household income
      In 2012, GBP£1=USD$1.61. DASH, Dietary Approaches to Stop Hypertension.
       <£20,000 per year1,119 (11.7)23.5 (4.9)17.70.83 (0.70, 0.98)0.030
       £20,000–40,000 per year3,409 (35.8)23.7 (4.7)16.40.83 (0.74, 0.94)0.002
       >£40,000 per year5,002 (52.5)24.3 (4.5)17.91
      Educational attainment
       ≤11 years of education2,004 (20.6)23.1 (4.6)12.80.43 (0.37, 0.51)<0.001
       12–13 years of education4,571 (47.0)23.6 (4.7)15.50.59 (0.53, 0.66)<0.001
       >13 years of education3,149 (32.4)25.2 (4.4)22.81
      Note: Boldface indicates statistical significance (p<0.05). ORs for sex are adjusted for age and educational attainment; ORs for income and educational attainment are adjusted for age and sex.
      a DASH score >28 (fifth quintile).
      b In 2012, GBP£1=USD$1.61.DASH, Dietary Approaches to Stop Hypertension.
      Table 3 shows that, after maximum covariate adjustment (Model 4), individuals in the middle and lowest tertile of dietary costs had 29% (95% CI=0.62, 0.81) and 58% (95% CI=0.36, 0.49) lower odds of having a DASH-accordant diet.
      Table 3Energy-Adjusted Dietary Costs Associated With Likelihood of Having a DASH-Accordant Diet (n=9,724)
      Exposure measureModel 1Model 2Model 3Model 4
      Dietary costs
       T10.34 (0.30, 0.40)0.40 (0.35, 0.46)0.40 (0.35, 0.46)0.40 (0.35, 0.46)
       T20.65 (0.58, 0.73)0.70 (0.62, 0.79)0.71 (0.63, 0.80)0.71 (0.63, 0.81)
       T31111
      Note: Values are OR (95% CI). Achieving a DASH-accordant diet was defined as a DASH score >28. Coefficients were derived from logistic regression analyses. Boldface indicates statistical significance (p<0.05). T1 is the tertile with lowest dietary costs, while T3 is the tertile with highest dietary costs (reference group). Model 1 is an unadjusted model. In Model 2 associations are adjusted for individual-level covariates (age, sex, educational level, and energy intake). In Model 3, associations are additionally adjusted for income. In Model 4, associations are additionally adjusted for distance to the nearest supermarket.
      DASH, Dietary Approaches to Stop Hypertension; T, tertile.
      Table 4 presents the odds of having a DASH-accordant diet according to distance to the nearest and nearest economically-matched supermarket. In the unadjusted model (Model 1), individuals living farthest away from their nearest supermarket had 24% lower odds of having a DASH-accordant diet (OR=0.76, 95% CI=0.66, 0.86), relative to those living closest, with evidence of a dose−response association. Adjustment for sociodemographic factors (Model 2) attenuated the association to non-significance for individuals in Tertile 2. Additional adjustment for exposure to other food outlets attenuated the association further (Model 3), but those living farthest away from their nearest supermarket still had 15% lower odds of having a DASH-accordant diet (OR=0.85, 95% CI=0.74, 0.98), relative to those living closest.
      Table 4Geographic and Economic Supermarket Access in Relation to Likelihood of Having a DASH-Accordant Diet (n=9,724)
      Exposure measureModel 1Model 2Model 3
      Geographic access to supermarkets
       T1 (ref; 0–1.12 km)111
       T2 (1.13–5.00 km)0.83 (0.73, 0.94)0.93 (0.82, 1.07)0.96 (0.84, 1.10)
       T3 (5.01–15.08 km)0.76 (0.66, 0.86)0.81 (0.71, 0.93)0.85 (0.74, 0.98)
      Economic access to supermarkets
       T1 (ref; 0–2.03 km)111
       T2 (2.04–7.35 km)0.67 (0.59, 0.76)0.71 (0.63, 0.81)0.73 (0.64, 0.83)
       T3 (7.36–32.16 km)0.53 (0.46, 0.60)0.57 (0.50, 0.66)0.59 (0.52, 0.68)
      Note: Values are OR (95% CI). Accordance to DASH was defined as a DASH score >28. Coefficients were derived from logistic regression analyses. Boldface indicates statistical significance (p<0.05). T1 is the tertile with shortest distance to the nearest supermarket (reference group), while T3 is the tertile with the longest distance to the nearest supermarket. Model 1 is an unadjusted model; in Model 2 associations are adjusted for individual-level covariates (age, sex, car ownership, educational level, and energy intake); in Model 3 associations are additionally adjusted for exposure to other food outlets within a 1-mile Euclidean buffer of the nearest supermarket.
      DASH, Dietary Approaches to Stop Hypertension; T, tertile.
      Similar associations, but with stronger effect sizes, were observed for distance to the nearest economically-matched supermarket. In the unadjusted model (Model 1), individuals living farthest away from their nearest economically-matched supermarket had 47% lower odds of having a DASH-accordant diet (OR=0.53, 95% CI=0.46, 0.60), relative to those living closest. Additional adjustment for sociodemographic factors and exposure to other food outlets (Model 3) showed that individuals living farthest away from their nearest economically-matched supermarket still had 41% lower odds of having a DASH-accordant diet (OR=0.59, 95% CI=0.52, 0.68), compared with individuals living closest, again with evidence of a dose−response association. Coefficients for other covariates included in these models are presented in Appendix Tables 3 and 4 (available online).
      Sensitivity analyses, adjusting for a variety of other factors (Methods), did not materially affect the results (data not shown).

      Discussion

      Given the beneficial effects of the DASH diet on risk of hypertension, heart failure, and coronary heart disease,
      • Karanja N.
      • Erlinger T.P.
      • Pao-Hwa L.
      • Miller E.R.
      • Bray G.A.
      The DASH diet for high blood pressure: from clinical trial to dinner table.
      • Chiu S.
      • Bergeron N.
      • Williams P.T.
      • Bray G.A.
      • Sutherland B.
      • Krauss R.M.
      Comparison of the DASH (Dietary Approaches to Stop Hypertension) diet and a higher-fat DASH diet on blood pressure and lipids and lipoproteins: a randomized controlled trial.
      • Salehi-Abarqouei A.
      • Maghsoudi Z.
      • Shirani F.
      • Azadbakht L.
      Effects of Dietary Approaches to Stop Hypertension (DASH)-style diet on fatal or nonfatal cardiovascular disease-incidence: a systematic review and meta-analysis on observational prospective studies.
      understanding the barriers in adherence to this diet is of public health importance. This population-based UK study demonstrated that the likelihood of consuming a DASH-accordant diet was dependent on both economic and geographic factors. The finding that both the local food environment and individual dietary budgets contribute to the accessibility of healthy food options suggests that improving adherence to a DASH diet may require structural approaches that take into account both affordability and proximity.
      As suggested previously, economic factors such as dietary costs were found to be strongly associated with diet quality.
      • Monsivais P.
      • Scarborough P.
      • Lloyd T.
      • et al.
      Greater accordance with the Dietary Approaches to Stop Hypertension dietary pattern is associated with lower diet-related greenhouse gas production but higher dietary costs in the United Kingdom.
      Having low dietary costs was associated with a 58% lower likelihood of consuming a DASH-accordant diet, comparable to the odds in participants with lowest education. Taking into account distance to the nearest supermarket did not influence these estimates, concordant with research that showed when modeled jointly, only price and not distance to stores was important for diet.
      • Aggarwal A.
      • Cook A.J.
      • Jiao J.
      • Seguin R.A.
      • Vernez Moudon A.
      • Hurvitz P.M.
      Access to supermarkets and fruit and vegetable consumption.
      Studying the importance of geographic accessibility of supermarkets revealed that individuals living farthest away from any supermarket were 15% less likely to consume DASH-accordant diets. This finding is in agreement with previous studies showing associations of supermarket proximity with dietary quality generally,
      • Laraia B.A.
      • Siega-Riz A.M.
      • Kaufman J.S.
      • Jones S.J.
      Proximity of supermarkets is positively associated with diet quality index for pregnancy.
      • Morland K.
      • Wing S.
      • Roux A.D.
      The contextual effect of the local food environment on residents’ diets: the Atherosclerosis Risk in Communities study.
      and accordance to DASH in particular
      • Layte R.
      • Harrington J.
      • Sexton E.
      • Perry I.J.
      • Cullinan J.
      • Lyons S.
      Irish exceptionalism? Local food environments and dietary quality.
      but for the first time shown in a UK context.
      Finally, supermarket access was defined according to both economic and geographic considerations, using a more nuanced metric consistent with theory described in two U.S. studies,
      • Breyer B.
      • Voss-Andreae A.
      Food mirages: geographic and economic barriers to healthful food access in Portland, Oregon.
      • Jiao J.
      • Moudon A.V.
      • Ulmer J.
      • Hurvitz P.M.
      • Drewnowski A.
      How to identify food deserts: measuring physical and economic access to supermarkets in King County, Washington.
      and for the first time in relation to dietary quality in the United Kingdom. As hypothesized, economic accessibility of supermarkets was more strongly related to dietary quality than geographic accessibility alone, with those living farthest away from an economically-matched supermarket having 41% lower likelihood of having a DASH-accordant diet. As such, price differences may be making some supermarkets more or less accessible for some than others, as a function of individual food budgets.
      • Jiao J.
      • Moudon A.V.
      • Ulmer J.
      • Hurvitz P.M.
      • Drewnowski A.
      How to identify food deserts: measuring physical and economic access to supermarkets in King County, Washington.
      By taking into account economics, latent (otherwise invisible) barriers to the uptake and maintenance of a DASH-accordant diet may have been captured that have not been captured in previous geographic studies.

      Implications

      Enhancing geographic access to supermarkets may constitute an effective supply-side solution to improving poor diets and health,
      • Treuhaft S.
      • Karpyn A.
      The Grocery Gap: Who Has Access to Healthy Food and Why it Matters.
      • Hawkes C.
      Dietary implications of supermarket development: a global perspective.
      as international research indicates that proximity to supermarkets remains highly variable.
      • Beaulac J.
      • Kristjansson E.
      • Cummins S.
      A systematic review of food deserts.
      However, as shown in a number of natural experiments, improving geographic access to supermarkets alone may be insufficient to promote behavior change.
      • Cummins S.
      • Petticrew M.
      • Higgins C.
      • Findlay A.
      • Sparks L.
      Large scale food retailing as an intervention for diet and health: quasi-experimental evaluation of a natural experiment.
      • Wrigley N.
      • Warm D.
      • Margetts B.
      • Whelan A.
      Assessing the impact of improved retail access on diet in a “food desert”: a preliminary report.
      • Sharkey J.R.
      • Horel S.
      • Han D.
      • Huber J.C.
      Association between neighborhood need and spatial access to food stores and fast food restaurants in neighborhoods of colonias.
      Public health gains could be maximized through neighborhood-level interventions focused on healthy and affordable food retail access, which is likely to be especially important for price-sensitive low-income consumers.
      As affordability depends on both food price and food budgets, accessibility of healthy foods should take into account the purchasing power of individuals. Demand-side solutions could be in the form of financial incentives for the purchase of healthy foods.

      Bartlett S, Jacob K, Lauren O, et al. Evaluation of the Healthy Incentives Pilot (HIP): Final Report. Prepared by Abt Associates. Washington, DC: U.S. Department of Agriculture, Food and Nutrition Service; September 2014

      As an example of supply- and demand-side interventions in parallel, U.S. farmers markets established to provide healthy food retail in the absence of supermarkets are also increasingly accepting Supplemental Nutrition Assistance Program coupons from low-income residents.
      • Young C.R.
      • Aquilante J.L.
      • Solomon S.
      • et al.
      Improving fruit and vegetable consumption among low-income customers at farmers markets: Philly Food Bucks, Philadelphia, Pennsylvania, 2011.
      To prevent widening of socioeconomic inequalities in diet, it may be important to target such food price policies specifically at low socioeconomic groups.
      • Darmon N.
      • Lacroix A.
      • Muller L.
      • Ruffieux B.
      Food price policies may improve diet but increase socioeconomic inequalities in nutrition.

      Limitations

      A number of factors may limit this study. Dietary intakes and dietary cost estimates were derived from an FFQ, an instrument subject to error and known biases.
      • Willett W.
      This study lacked information on participants’ actual food spending and the origin of foods consumed, which could have resulted in misclassification of dietary cost. However, dietary costs (as derived with reported dietary intakes and a fixed database of food prices) are modestly but positively correlated with actual food spending,
      • Monsivais P.
      • Perrigue M.M.
      • Adams S.L.
      • Drewnowski A.
      Measuring diet cost at the individual level: a comparison of three methods.
      • Aaron G.J.
      • Keim N.L.
      • Drewnowski A.
      • Townsend M.S.
      Estimating dietary costs of low-income women in California: a comparison of 2 approaches.
      • Timmins K.A.
      • Morris M.A.
      • Hulme C.
      • Edwards K.L.
      • Clarke G.P.
      • Cade J.E.
      Comparability of methods assigning monetary costs to diets: derivation from household till receipts versus cost database estimation using 4-day food diaries.
      therefore, being suitable for our purpose of ranking individuals into tertiles of dietary cost. There was no information on the actual shopping location of the participants. Instead, access to supermarkets was defined as proximity, based on the decreased likelihood for environments to influence individuals as a function of distance.
      • Cromley E.K.
      • McLafferty S.
      As the identification of supermarket tiers was based on full retail product prices, and collected over 2 consecutive months, supermarkets with more discounts on average, and at this time of year, may have been misclassified. Combining this with a lack of information on actual spending, participants may have been mismatched to supermarket tiers. Lastly, further exposure misclassification may have resulted from: calculating supermarket proximity only from the home address; neglecting other environmental settings; and temporal mismatch, as exposures (supermarket locations, 2011) were only measured at one time point, whereas outcomes (diet, 2005–2014) were measured over 9 years.
      This study did, however, benefit from the combination of objective information on the geographic location of supermarkets, the costs of food in these supermarkets on the basis of a 101-item market basket, and detailed information on the diet quality of more than 9,000 UK adults. Although originating from the U.S., the DASH diet has shown to be congruent with UK food preferences.
      • Harnden K.E.
      • Frayn K.N.
      • Hodson L.
      Dietary Approaches to Stop Hypertension (DASH) diet: applicability and acceptability to a UK population.
      A further strength is the derivation of individual-level dietary costs as a measure of economic access, providing an insight into the monetary value of individuals’ diets. Incorporating economic accessibility into our measure of physical supermarket access allowed for a more comprehensive definition of access. Finally, adjusting the analyses for exposure to other food outlets controlled for food environment “context,” specifically allowing for the fact that many healthy and unhealthy foods can be purchased at other food outlets.

      Conclusions

      This study provides novel evidence that geographic access to supermarkets is particularly important for accordance to the DASH diet once economic access to food is taken into account. The fact that higher dietary costs and supermarket proximity were associated with DASH accordance suggests that interventions to improve healthy eating should include structural changes involving both affordability and proximity of healthy food options.

      Acknowledgments

      We thank all the volunteers who participated in the Fenland Study, as well as Fenland Study Coordination, Field Epidemiology, and data management.
      This work was undertaken by the Centre for Diet and Activity Research, a UK Clinical Research Collaboration Public Health Research Centre of Excellence. Funding from the British Heart Foundation, Cancer Research UK, Economic and Social Research Council, Medical Research Council (MRC), National Institute for Health Research, and Wellcome Trust, under the auspices of the UK Clinical Research Collaboration, is gratefully acknowledged. Core MRC Epidemiology Unit support through programs MC_UU_12015/1 and MC_UU_12015/5 is acknowledged. Funders had no role in the design or conduct of the study; collection, management, analysis, or interpretation of the data; or preparation, review, or approval of the manuscript.
      The study analysis was devised by JDM, TB, and PM. NGF, SB, SJG, and NJW (Principal Investigators of the Fenland Study) conceived of the Fenland Study and were responsible for its overall design, management, and supervision of the data collection. TB was responsible for data collection from local councils. JDM led on data analysis, in consultation with TB, JL, NGF, SB, SJG, NJW, and PM. JDM, TB, JL, and PM drafted the manuscript together. All authors read and approved the final manuscript. JDM had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
      No financial disclosures were reported by the authors of this paper.

      Appendix A. Supplementary material

      References

        • Lock K.
        • Pomerleau J.
        • Causer L.
        • Altmann D.R.
        • McKee M.
        The global burden of disease attributable to low consumption of fruit and vegetables: implications for the global strategy on diet.
        Bull World Health Organ. 2005; 83: 100-108
        • Mozaffarian D.
        • Appel L.J.
        • Van Horn L.
        Components of a cardioprotective diet: new insights.
        Circulation. 2011; 123: 2870-2891https://doi.org/10.1161/CIRCULATIONAHA.110.968735
        • Karanja N.
        • Erlinger T.P.
        • Pao-Hwa L.
        • Miller E.R.
        • Bray G.A.
        The DASH diet for high blood pressure: from clinical trial to dinner table.
        Cleve Clin J Med. 2004; 71: 745-753https://doi.org/10.3949/ccjm.71.9.745
        • Chiu S.
        • Bergeron N.
        • Williams P.T.
        • Bray G.A.
        • Sutherland B.
        • Krauss R.M.
        Comparison of the DASH (Dietary Approaches to Stop Hypertension) diet and a higher-fat DASH diet on blood pressure and lipids and lipoproteins: a randomized controlled trial.
        Am J Clin Nutr. 2016; 103: 341-347https://doi.org/10.3945/ajcn.115.123281
        • Salehi-Abarqouei A.
        • Maghsoudi Z.
        • Shirani F.
        • Azadbakht L.
        Effects of Dietary Approaches to Stop Hypertension (DASH)-style diet on fatal or nonfatal cardiovascular disease-incidence: a systematic review and meta-analysis on observational prospective studies.
        Nutrition. 2013; 29: 611-618https://doi.org/10.1016/j.nut.2012.12.018
        • Méjean C.
        • Droomers M.
        • van der Schouw Y.T.
        • et al.
        The contribution of diet and lifestyle to socioeconomic inequalities in cardiovascular morbidity and mortality.
        Int J Cardiol. 2013; 168: 5190-5195https://doi.org/10.1016/j.ijcard.2013.07.188
        • Bertoni A.G.
        • Foy C.G.
        • Hunter J.C.
        • Quandt S.A.
        • Vitolins M.Z.
        • Whitt-Glover M.C.
        A multilevel assessment of barriers to adoption of Dietary Approaches to Stop Hypertension (DASH) among African Americans of low socioeconomic status.
        J Health Care Poor Underserved. 2011; 22: 1205-1220https://doi.org/10.1353/hpu.2011.0142
        • Young C.M.
        • Batch B.C.
        • Svetkey L.P.
        Effect of socioeconomic status on food availability and cost of the dietary approaches to stop hypertension (DASH) dietary pattern.
        J Clin Hypertens. 2008; 10: 603-611https://doi.org/10.1111/j.1751-7176.2008.08199.x
        • Monsivais P.
        • Scarborough P.
        • Lloyd T.
        • et al.
        Greater accordance with the Dietary Approaches to Stop Hypertension dietary pattern is associated with lower diet-related greenhouse gas production but higher dietary costs in the United Kingdom.
        Am J Clin Nutr. 2015; 102: 138-145https://doi.org/10.3945/ajcn.114.090639
        • Monsivais P.
        • Rehm C.D.
        • Drewnowski A.
        The DASH diet and diet costs among ethnic and racial groups in the United States.
        JAMA Intern Med. 2013; 173: 1922-1924https://doi.org/10.1001/jamainternmed.2013.9479
        • Laraia B.A.
        • Siega-Riz A.M.
        • Kaufman J.S.
        • Jones S.J.
        Proximity of supermarkets is positively associated with diet quality index for pregnancy.
        Prev Med (Baltim). 2004; 39: 869-875https://doi.org/10.1016/j.ypmed.2004.03.018
        • Morland K.
        • Diez Roux A.V.
        • Wing S.
        Supermarkets, other food stores, and obesity: the atherosclerosis risk in communities study.
        Am J Prev Med. 2006; 30: 333-339https://doi.org/10.1016/j.amepre.2005.11.003
        • Robinson P.L.
        • Dominquez F.
        • Teklehaimanot S.
        • Lee M.
        • Brown A.
        • Goodchild M.
        Does distance decay modelling of supermarket accessibility predict fruit and vegetable intake by individuals in a large metropolitan area?.
        J Health Care Poor Underserved. 2013; 24: 172-185https://doi.org/10.1353/hpu.2013.0049
        • Layte R.
        • Harrington J.
        • Sexton E.
        • Perry I.J.
        • Cullinan J.
        • Lyons S.
        Irish exceptionalism? Local food environments and dietary quality.
        J Epidemiol Community Health. 2011; 65: 881-888https://doi.org/10.1136/jech.2010.116749
        • Cummins S.
        • Flint E.
        • Matthews S.A.
        New neighborhood grocery store increased awareness of food access but did not alter dietary habits or obesity.
        Health Aff (Millwood). 2014; 33: 283-291https://doi.org/10.1377/hlthaff.2013.0512
        • Cobb L.K.
        • Appel L.J.
        • Franco M.
        • Jones-Smith J.C.
        • Nur A.
        • Anderson C.A.M.
        The relationship of the local food environment with obesity: a systematic review of methods, study quality, and results.
        Obesity (Silver Spring). 2015; 23: 1331-1344https://doi.org/10.1002/oby.21118
        • Short A.
        • Guthman J.
        • Raskin S.
        Food deserts, oases, or mirages? Small markets and community food security in the San Francisco Bay area.
        J Plan Educ Res. 2007; 26: 352-364https://doi.org/10.1177/0739456X06297795
        • Breyer B.
        • Voss-Andreae A.
        Food mirages: geographic and economic barriers to healthful food access in Portland, Oregon.
        Health Place. 2013; 24: 131-139https://doi.org/10.1016/j.healthplace.2013.07.008
        • Jiao J.
        • Moudon A.V.
        • Ulmer J.
        • Hurvitz P.M.
        • Drewnowski A.
        How to identify food deserts: measuring physical and economic access to supermarkets in King County, Washington.
        Am J Public Health. 2012; 102: e32-e39https://doi.org/10.2105/AJPH.2012.300675
        • Pechey R.
        • Monsivais P.
        Supermarket choice, shopping behavior, socioeconomic status, and food purchases.
        Am J Prev Med. 2015; 49: 868-877https://doi.org/10.1016/j.amepre.2015.04.020
        • Drewnowski A.
        • Aggarwal A.
        • Hurvitz P.M.
        • Monsivais P.
        • Moudon A.V.
        Obesity and supermarket access: proximity or price?.
        Am J Public Health. 2012; 102: 74-80https://doi.org/10.2105/AJPH.2012.300660
        • Aggarwal A.
        • Cook A.J.
        • Jiao J.
        • Seguin R.A.
        • Vernez Moudon A.
        • Hurvitz P.M.
        Access to supermarkets and fruit and vegetable consumption.
        Am J Public Health. 2014; 104: 917-923https://doi.org/10.2105/AJPH.2013.301763
        • Cummins S.
        • Petticrew M.
        • Higgins C.
        • Findlay A.
        • Sparks L.
        Large scale food retailing as an intervention for diet and health: quasi-experimental evaluation of a natural experiment.
        J Epidemiol Community Health. 2005; 59: 1035-1040https://doi.org/10.1136/jech.2004.029843
        • Wrigley N.
        • Warm D.
        • Margetts B.
        • Whelan A.
        Assessing the impact of improved retail access on diet in a “food desert”: a preliminary report.
        Urban Stud. 2002; 39: 2061-2082https://doi.org/10.1080/0042098022000011362
        • Sharkey J.R.
        • Horel S.
        • Han D.
        • Huber J.C.
        Association between neighborhood need and spatial access to food stores and fast food restaurants in neighborhoods of colonias.
        Int J Health Geogr. 2009; 8: 9https://doi.org/10.1186/1476-072X-8-9
        • De Lucia Rolfe E.
        • Loos R.J.F.
        • Druet C.
        • et al.
        Association between birth weight and visceral fat in adults.
        Am J Clin Nutr. 2010; 92: 347-352https://doi.org/10.3945/ajcn.2010.29247
        • Bingham S.A.
        • Gill C.
        • Welch A.
        • et al.
        Validation of dietary assessment methods in the UK arm of EPIC using weighed records, and 24-hours urinary nitrogen and potassium and serum vitamin C and carotenoids as biomarkers.
        Int J Epidemiol. 1997; 26: S137https://doi.org/10.1093/ije/26.suppl_1.S137
        • Fung T.T.
        • Chiuve S.E.
        • McCullough M.L.
        • Rexrode K.M.
        • Logroscino G.
        • Hu F.B.
        Adherence to a DASH-style diet and risk of coronary heart disease and stroke in women.
        Arch Intern Med. 2008; 168: 713-720https://doi.org/10.1001/archinte.168.7.713
        • Willett W.C.
        • Howe R.
        Adjustment for total energy intake in epidemiologic studies.
        Cancer. 1997; 65: 1220S-1228S
        • Sherzai A.
        • Heim L.T.
        • Boothby C.
        • Sherzai A.D.
        Stroke, food groups, and dietary patterns: a systematic review.
        Nutr Rev. 2012; 70: 423-435https://doi.org/10.1111/j.1753-4887.2012.00490.x
        • Monsivais P.
        • Perrigue M.M.
        • Adams S.L.
        • Drewnowski A.
        Measuring diet cost at the individual level: a comparison of three methods.
        Eur J Clin Nutr. 2013; 67: 1220-1225https://doi.org/10.1038/ejcn.2013.176
        • Aaron G.J.
        • Keim N.L.
        • Drewnowski A.
        • Townsend M.S.
        Estimating dietary costs of low-income women in California: a comparison of 2 approaches.
        Am J Clin Nutr. 2013; 97: 835-841https://doi.org/10.3945/ajcn.112.044453
        • Timmins K.A.
        • Morris M.A.
        • Hulme C.
        • Edwards K.L.
        • Clarke G.P.
        • Cade J.E.
        Comparability of methods assigning monetary costs to diets: derivation from household till receipts versus cost database estimation using 4-day food diaries.
        Eur J Clin Nutr. 2013; 67: 1072-1076https://doi.org/10.1038/ejcn.2013.157
      1. BBC News. Tesco market share dips below 30%. www.bbc.co.uk/news/business-16817254. Published January 31, 2012. Accessed October 16, 2015.

        • Willett W.
        • Stampfer M.
        Implications for total energy intake for epidemiologic analyses.
        in: Willett W. Nutritional Epidemiology. Oxford University Press, New York, NY1998: 273-301https://doi.org/10.1093/acprof:oso/9780195122978.003.11
        • Lake A.A.
        • Burgoine T.
        • Greenhalgh F.
        • Stamp E.
        • Tyrrell R.
        The foodscape: classification and field validation of secondary data sources.
        Health Place. 2010; 16: 666-673https://doi.org/10.1016/j.healthplace.2010.02.004
        • Willett W.
        Nutritional Epidemiology. 3rd ed. Oxford University Press, New York, NY1998https://doi.org/10.1093/acprof:oso/9780195122978.001.0001
        • Morland K.
        • Wing S.
        • Roux A.D.
        The contextual effect of the local food environment on residents’ diets: the Atherosclerosis Risk in Communities study.
        Am J Public Health. 2002; 92: 1761-1767https://doi.org/10.2105/AJPH.92.11.1761
        • Treuhaft S.
        • Karpyn A.
        The Grocery Gap: Who Has Access to Healthy Food and Why it Matters.
        PolicyLink, Oakland, CA2010 (Accessed March 10, 2017)
        • Hawkes C.
        Dietary implications of supermarket development: a global perspective.
        Dev Policy Rev. 2008; 26: 657-692https://doi.org/10.1111/j.1467-7679.2008.00428.x
        • Beaulac J.
        • Kristjansson E.
        • Cummins S.
        A systematic review of food deserts.
        Prev Chronic Dis. 2009; 6: A105
      2. Bartlett S, Jacob K, Lauren O, et al. Evaluation of the Healthy Incentives Pilot (HIP): Final Report. Prepared by Abt Associates. Washington, DC: U.S. Department of Agriculture, Food and Nutrition Service; September 2014

        • Young C.R.
        • Aquilante J.L.
        • Solomon S.
        • et al.
        Improving fruit and vegetable consumption among low-income customers at farmers markets: Philly Food Bucks, Philadelphia, Pennsylvania, 2011.
        Prev Chronic Dis. 2013; 10: E166https://doi.org/10.5888/pcd10.120356
        • Darmon N.
        • Lacroix A.
        • Muller L.
        • Ruffieux B.
        Food price policies may improve diet but increase socioeconomic inequalities in nutrition.
        World Rev Nutr Diet. 2016; 115: 36-45https://doi.org/10.1159/000442069
        • Cromley E.K.
        • McLafferty S.
        GIS and Public Health. 2nd ed. Guilford Press, New York, NY2012
        • Harnden K.E.
        • Frayn K.N.
        • Hodson L.
        Dietary Approaches to Stop Hypertension (DASH) diet: applicability and acceptability to a UK population.
        J Hum Nutr Diet. 2010; 23: 3-10https://doi.org/10.1111/j.1365-277X.2009.01007.x