Volume 24
Number 2/3
2003

[Table of Contents]

Do healthy food baskets assess food security? 

Tasnim Nathoo and Jean Shoveller 

Abstract 

Developing indicators to measure the different facets of food security presents numerous conceptual and methodological challenges. This paper adopts an ecological framework to reflect on these issues through an examination of the Healthy Food Basket (HFB) tool. The HFB tool is used to measure food security conditions by determining the cost and availability of a group of foods in a shopping basket across a range of stores in different regions and neighbourhoods. The paper discusses the ability of the HFB tool to describe micro-, meso- and macro-level influences on food security and the use of the ecological model in developing complementary and alternative strategies for understanding and monitoring food security. 

Key words: ecological model; food security; measurement; nutrition indicators 

Introduction 

Different understandings of the term “food security” can create serious challenges to adopting multisectoral approaches that address food security issues. In a recent survey, Power et al.¹ found significant variations among Canada's registered dieticians regarding the conceptualization of the term. The dieticians responded with a wide range of understandings of the term, including food safety, food as a basic human right, adequate food to maintain health, sustainable agricultural systems, affordability, charitable food distribution systems and individual choice of personally acceptable foods. Canada's Action Plan for Food Security² adopts the definition of food security that emerged from the World Food Summit: “Food security exists when all people, at all times, have physical and economic access to sufficient, safe and nutritious food to meet their dietary needs and food preferences for an active and healthy life.” 

Canada's Action Plan for Food Security also lists, as a priority, the development of a monitoring system for food insecurity. It identifies the need for “a comprehensive set of agreed-upon indicators to determine the nature, extent and evolution of food insecurity, both to develop appropriate responses and to monitor their effectiveness.” Developing indicators to measure the different facets of food security presents numerous conceptual and methodological challenges. 

This paper adopts an ecological framework to reflect on the different dimensions of food security and discusses complementary and alternative strategies for understanding and monitoring food security. It examines a commonly used tool to measure food security conditions: the Healthy Food Basket (HFB) tool (also called the Nutritious Food Basket), which has been used in Canada for nearly half a century. Although the HFB tool is used to describe food security conditions, it is not clear which dimensions of food security are best captured and understood through the use of HFB data. 

The HFB tool determines the cost and availability of a group of foods in a shopping basket across a range of stores in different regions and neighbourhoods (see Table 1). Although the results from HFB surveys have been used widely to inform many policies and programs, including those related to social welfare and nutrition, the adequacy and comprehensiveness of the HFB tool in describing food security conditions at the micro (individual), meso (community) and macro (population) levels remains to be determined. 

Applying an ecological framework to the HFB 

Ecological theory asserts that a reciprocal and dynamic interrelation exists between the individual and subsystems of the environment.^3–5 Ecological theory and approaches are not new, and they are widely used to examine the determinants of population health.^6,7 Using an ecological framework to examine the HFB provides an opportunity to examine micro-, meso- and macro-level influences on food security. At the most basic level, the micro level, the analysis pertains to individual activities, roles and interpersonal relationships in a given setting;⁴ the meso-level analysis examines relations among groups across settings; and the macro level pertains to socio-cultural and policy-level influences that may originate with institutions. 

The different components of food security, such as food availability in a community, individual food consumption and sustainable agricultural production, are determined by a combination of micro-, meso- and macro-level influences. The HFB tool has been used to answer questions at the micro level, such as “What does it cost to feed a family with food that is nutritious, acceptable and sufficient in quantity?” It has also been used to examine meso-level questions, such as “Are healthier foods readily available to people in different communities or regions?” as well as macro-level questions, like “How should the social security system be structured to enhance the ability of the poor to purchase healthy food?” 

Information from food basket surveys has been used in a wide range of program and policy applications, including the development of educational material on nutrition and dietary guidelines, to promote access to healthy foods in remote and rural areas, and to assess the adequacy of welfare food allowances. Many advocacy groups, professional organizations and public institutions also use such information as an indicator of food insecurity conditions in communities across Canada and to promote and support policy development to increase access to healthy foods. 

Reflecting on the HFB from an ecological perspective 

Micro level 

Individual and household incomes are important determinants of food choice. At the micro level, the HFB has been used primarily to assess the ability of individuals to afford healthy food. For example, the HFB tool can assess the cost of a diet that reflects basic nutrient and calorie needs, and this information can be used to demonstrate the (in)adequacy of income support allowances to provide individuals and families with the means to afford a healthy diet. Community nutritionists in British Columbia have used the HFB tool to produce an annual report, The Cost of Eating in B.C., which demonstrates how families of four on income assistance and single parent families (in which the parent works full time at a minimum wage job) have great difficulty in meeting their shelter and food needs and other basic necessities with their limited incomes.⁸

In interpreting HFB estimates at the micro level, important cautions should be considered. When used in this way, the HFB tool, like the U.S. Department of Agriculture household food security scale,⁹ focuses on financial constraints to individual and household food security. It also tends to equate food insecurity with hunger. It does not measure the other broad concepts included in definitions of food security (e.g., food safety, sustainable agriculture) nor does it identify individuals or groups who may be vulnerable to food insecurity for other reasons (e.g., many elderly or disabled people who may not have physical access to food). HFB assumes that people with higher incomes do not experience food insecurity, and it is not able to capture differences in food security levels that may exist between individuals and households with limited incomes (e.g., individuals receiving social assistance in an urban area may have higher housing costs and less money to allot to food). In addition, the HFB is not sufficiently flexible to capture understandings of food security that may be culturally specific, an important issue for consideration in Canada's highly multicultural society. 

The HFB is sometimes used to determine what it might cost an individual to eat healthy food (e.g., to estimate the costs as part of a student loan) or to feed people in group homes, extended care facilities or homeless shelters. However, it should be remembered that the contents of the HFB are generally derived from population-level food consumption patterns.^8,10–12 Although the cost of the HFB can be altered to reflect the nutritional requirements for different age and sex groups, it should be remembered that these values as well as the food consumption patterns that the basket is derived from are based on averages (e.g., they do not reflect individual differences in activity and metabolism or special dietary needs). By making inferences about individuals from a tool derived from population-level data, researchers and food security planners may be in danger of committing an ecological fallacy. Because food consumption patterns vary significantly among individuals, the statistical and theoretical assumptions underpinning population-level analyses do not necessarily translate directly to the individual or micro level. Thus, the HFB tool may be effective in assessing which populations may be economically vulnerable to food insecurity, but it provides, at best, only a proxy for individual or micro-level food security. 

Meso level 

At a micro level, we are concerned with the ability of individuals to afford a healthy diet. At a meso level, the analysis is focused more on whether affordable food is available to the individual in his or her community. For example, meso-level forces may include how communities respond to hunger (e.g., the presence or absence of food banks or soup kitchens in the community), the variety and mix of retail food outlets and the existence of municipal by-laws that support events such as farmers' markets. 

Physical access to food in local areas is a growing concern, and several researchers have used the HFB tool in an urban context to answer meso-level questions like “Does a healthy food basket cost more in poorer areas?” Several studies have shown that a healthy food basket is more expensive and the items in the basket are less available in poorer urban areas;^13–18 thus, groups of individuals or families with low incomes face both economic and physical barriers to a healthy diet. 

Many users have cautioned against comparing HFB estimates at the meso level across communities and regions. Although researchers have found differences between extremely different urban neighbourhoods, there is some evidence that the HFB tool may not be able to adequately capture the influence of local context on people's food choices.^11,16 For example, the HFB tool cannot account for the influence of different market share of stores in different communities or different buying patterns in different regions. As well, communities are not precisely defined and, in fact, subcommunities may overlap with each other – for instance, many communities may include a variety of economically diverse neighborhoods, which may be located adjacent to one another. In these cases, members of the different neighbourhoods may cross boundaries to shop in the same stores. This issue can be important when a comparison is made of different neighbourhoods in an urban region where food access is more strongly influenced by factors such as transportation and store location rather than by geographical boundaries. 

In this situation, researchers and food security planners may be in danger of committing an atomistic fallacy. A relation may exist at the micro level (e.g., individuals with low income may not have equal access to the foods in a healthy food basket); however, it may not necessarily hold at the meso level (e.g., the average cost and availability of food items in a healthy food basket may be equal across neighbourhoods). This subtle, yet important, flaw in the logic of HFB has the potential to mislead researchers and planners in their efforts to promote food security in Canada. Thus, the HFB tool has the capacity to indicate whether healthy foods are available to groups of individuals in their local environments, but it fails to provide sufficiently sensitive information regarding the impact of micro-level food consumption patterns on food security within and across communities. 

Macro level 

Macro-level determinants of food security include socio-cultural and policy-level influences. It is within this complex context that HFBs are implemented. The HFB tool does not directly measure the impact of specific socio-cultural or policy-level influences – rather, it describes the cost of a pre-specified list of foods across a number of stores, which represent one facet of food security that is influenced by macro-level forces. Thus, the HFB represents one indicator of the potential impact of macro-level changes. However, on its own, HFB provides an insufficient reflection of the overall complex myriad of factors affecting food security. 

Although the HFB tool may be able to provide limited information about why food security conditions are changing, data from HFB surveys can be used in conjunction with other indicators (e.g., prevalence of nutrition-related disease, income inequality, unemployment rates) to inform changes in social, health and agriculture systems. For example, the HFB tool has provided insights into the question of “How should the social security system be structured to enhance the ability of the poor to purchase healthy food?” and has been used to demonstrate the inadequacies of welfare allowances. 

Discussion 

The HFB tool appears to be an effective tool to monitor one of the key determinants of individual food security, food affordability. It has the capacity to identify local differences in cost and access to healthy food. However, the HFB does not appear to have the capacity to adequately or comprehensively monitor the nutritional health of the population. 

At present, we have several indicators that indirectly measure food security, and these measures focus on issues of hunger (e.g., growth of food banks and increasing use of emergency feeding programs). Although hunger is a growing concern in Canada, over-nutrition and malnutrition remain concerns for a large percentage of the population. Food costing techniques have long been used to monitor the affordability of an adequate diet to prevent hunger and malnutrition. Cancer, heart disease and strokes, all related to nutrition, are the three leading causes of death in Canada, and obesity is a concern of nearly half of the Canadian population.  Although epidemiology has been successful in demonstrating the relation between nutrition and individual health (e.g., vitamin deficiencies), at a macro level the causal pathways between nutrition and the health of the population are subtler and are less clearly understood. The HFB is an intrinsically appealing food-costing tool, but it is not sufficiently sophisticated to document the influence of institutional and sociological phenomena on food availability, cost and consumption patterns at the population level. 

Food affordability and access, as measured by the HFB, represent one piece of the food security puzzle. In order to understand the context of individual food consumption, we need to pay greater attention to the impact of macro-level influences, such as changes in global governance, methods of food production and the composition of the retail food industry, as well as to the role of cultural change and technological developments. 

Adopting an ecological framework to understand the influences of food security and population-level nutrition demands an examination of these structural and contextual influences. Although ecological approaches have typically been used to identify and address factors associated with each level of influence, the capacity to examine simultaneously the interactions among micro-, meso- and macro-level determinants has been underused. For example, environmental degradation, loss of biodiversity and the advancement of monoculture have resulted from complex interactions among micro-, meso- and macro-level forces. Thus, a more complete adoption and implementation of ecological approaches hold promise for developing new insights into the different facets and dimensions of food security. 

Conclusions 

This analysis describes the parameters within which the HFB tool should continue to be used. The HFB provides a proxy estimate of individual food affordability and accessibility. We recommend that it continue to be used to address these aspects of food security at the micro level. The tool has practical appeal because it can be implemented easily and quickly at low cost. We suggest that interpretation of the tool could be further improved by comparing the cost of the HFB to the proportion of individual income spent on food. This approach would account for fluctuations in food basket costs and income levels and would be another tool to monitor food security change over time. 

As food affordability and access is only one component of a comprehensive understanding of the food security question, we also wanted to use this examination of the HFB as an opportunity to stimulate thinking about complementary and alternative approaches to understanding food security. An ecological approach can clarify the appropriateness of using specific tools at each level of analysis (micro, meso, macro) as well as emphasize the importance of the interactions across levels. The field could benefit significantly from the development of new tools to assess both meso- and macro-level influences on food security. While traditional research approaches are unable to untangle the complex web of interactions across the micro, meso and macro levels, new statistical and methodological tools developed in other disciplines could be applied to the problem of food security. 

At the level of practice, an ecological framework may also reinforce an understanding that interventions must move beyond an exclusive focus on individual (micro level) food security. New approaches are required to achieve population-level improvements in food security. Policies and programs that are informed by research that captures the interactions between individuals and more “upstream” social, cultural and institutional influences hold promise for improving nutrition at the population level.^19–21 To increase our understanding of the reciprocal and dynamic interrelations that exist between individuals and various subsystems within their environments requires a more ecological approach to research and program/policy planning. 

Newly emerging mixed-method research approaches^6,22 and multi-level modelling techniques²³ hold promise for developing more ecological understandings of food security and nutritional health at the population level, because they allow us to examine interactions across levels of influence while controlling for both the atomistic and ecological fallacies. By adopting an ecological perspective, research agencies and other stakeholders dedicated to promoting population health should be encouraged to investigate these important issues using these innovative research approaches. Findings from such research could then be used to inform the development of more progressive, comprehensive and population-based policies and programs to promote equitable access to healthy food by all groups of the population. 

Acknowledgements 

The preparation of this manuscript was possible thanks to a British Columbia Medical Services Foundation Studentship to Ms. Nathoo and through a career award from the BC Health Research Foundation to Dr. Shoveller. The authors also would like to thank Dr. Aleck Ostry for his helpful comments and suggestions on an earlier draft of this manuscript. 

References 

1. Power EM, Sheeshka JD, Heron AL. Canadian dietitians' understanding of food security. J Nutr Educ 1998;30(1):45–9. 

2. Agriculture and Agri-Food Canada. Canada's action plan for food security: a response to the World Food Summit. Ottawa: Government of Canada, 1998. 

3. Bronfenbrenner U. The ecology of human development. Cambridge, MA: Harvard University Press, 1979. 

4. Bronfenbrenner U. Ecological systems theory. In: R V, ed. Six theories of child development: revised formulations and current issues. London, UK: Jessica Kingsley, 1992: 187–249. 

5. Bronfenbrenner U. Developmental ecology through space and time: a future perspective. In: Moen P, Elder GHJ, eds. Examining lives in context: perspectives on the ecology of human development. Washington, DC: American Psychological Association, 1995:619–47. 

6. Green LW, Richard L, Potvin L. Ecological foundations of health promotion. Am J Health Promotion 1996;10(4):270–80. 

7. McLeroy KR, Steckler AB, Simons-Morton B et al. Social science theory in health education: time for a new model? Health Educ Res 1993;8(3):305–12. 

8. Canadian Association of Food Banks, Dieticians of Canada, Community Nutritionists Council of BC. The cost of eating in BC. 2000. 

9. Bickel G, Nord M, Price C et al. Guide to measuring household food security, revised 2000. Alexandria, VA: U.S. Department of Agriculture, Food and Nutrition Service, 2000. 

10. Robbins IG, Robichon-Hunt L. The Agriculture Canada Nutritious Food Basket and the Thrifty Nutritious Food Basket, 1989. Food Market Commentary 1989;11(1): 31–41. 

11. Travers KD, Cogdon W, McDonald C et al. Availability and cost of heart healthy dietary changes in Nova Scotia. J Canadian Dietetic Association 1997;58:176–83. 

12. Crockett EG, Clancy KL, Bowering J. Comparing the cost of a Thrifty Food Plan Market Basket in the three areas of New York State. J Nutr Educ 1992;24(1):71S–78S. 

13. Chung C, Myers S. Do the poor pay more for food? An analysis of grocery store availability and food price disparities. J Consumer Affairs 1999;33(2):276–96. 

14. Curtis KA, McClellan S. Falling through the safety net: poverty, food assistance and shopping constraints in an American city. Urban Anthropology 1995;24(1–2):93–135. 

15. Reisig V, Hobbiss A. Food deserts and how to tackle them: a study of one city's approach. Health Educ J 2000;59:137–49. 

16. Travers KD. The social organization of nutritional inequities. Soc Sci Med 1996;43(4): 543–53. 

17. 17.    Mooney C. Cost and availability of healthy food choices in a London health district. J Hum Nutr Dietet 1990;3:111–20. 

18. Sooman A, Macintyre S, Anderson A. Scotland's health – a more difficult challenge for some? The price and availability of healthy foods in socially contrasting localities in the west of Scotland. Health Bull 1993;51(5):276–84. 

19. Glass TA. Psychosocial intervention. In: Berkman LF, Kawachi I, eds. Social epidemiology. New York: Oxford University Press, 2000: 267–305. 

20. Bosma H. Neighborhood socio-economic status and all cause mortality. Am J Epidemiol 2001;153:365. 

21. Yen IH, Syme SL. The social environment and health: a discussion of the epidemiologic literature. Annu Rev Public Health 1999;20:287–308. 

22. Tashakkori A, Teddlie C. Mixed methodology: combining qualitative and quantitative approaches. Thousand Oaks, CA: SAGE; 1998 

23. Marmot M. Multilevel approaches to understanding social determinants. In: Berkman LF, Kawachi I, eds. Social epidemiology. New York: Oxford University Press, 2000:349–67.

Author References 

Tasnim Nathoo, Department of Health Care and Epidemiology, University of British Columbia, Vancouver, British Columbia, Canada 

Jean Shoveller, Department of Health Care and Epidemiology, and the Centre for Community Health and Health Evaluation Research, University of British Columbia, Vancouver, British Columbia, Canada 

Correspondence: Tasnim Nathoo, Department of Health Care and Epidemiology, James Mather Building, 5804 Fairview Avenue, University of British Columbia, Vancouver, British Columbia, Canada V6T 1Z3; Fax: (604) 822-3724; E-mail: tasnim@interchange.ubc.ca 

[Previous] [Table of Contents] [Next]