Economic access to fruits and vegetables in the greater Quebec City: do disparities exist?.
Drouin, Sarah ; Hamelin, Anne-Marie ; Ouellet, Denise 等
Social inequalities in health are the focus of an increasing number
of studies trying to determine the elements and processes involved.
(1-4) Among these studies, those addressing the economic accessibility
of food suggest that unequal access can contribute to health disparities
by putting low-income people at a disadvantage. (5-7) With respect to
consumption of fruits and vegetables (FV), 58% of Canadian households
with low income (<$30,000) reported eating fewer than five portions
of FV per day in 2004, and this proportion fell to 41% for privileged
households (>$60,000). (8) Significant associations were found
between low consumption of FV and the risk of chronic diseases such as
obesity, cancer and diabetes. (9-11)
Studies have analyzed the variation in healthy food costs across
different socio-economic contexts. (12-15) In Canada, a few studies have
shown no variation in the cost of a Nutritious Food Basket (NFB)
according to socio-economic status. (16-18) In these studies, food store
type was the only predictor of cost of an NFB, which was lowest in large
grocery stores.
A pilot study conducted in Quebec City on the variation in food
costs found an uneven distribution of the various food store types and
suggested the need for a closer investigation of the variation in costs
with respect to socio-economic level. (19) The present study was thus
undertaken to respond to the following research question: Are there
differences in the cost of FV in different types of food stores
according to 1) the level of urbanization (urban or rural) and 2) the
socio-economic levels of the urban centres in the greater Quebec City?
METHODS
Area selection
Four areas in the greater Quebec City with various socio-economic
contexts were selected for this study. They are identified either as
urban (Quebec City) or rural (three regional county municipalities).
(20) Each rural centre had to be located in a different municipality,
and all rural centres were selected to reflect the diversity of the
metropolitan-influenced zone. This concept is used in order to better
differentiate regions of Canada that lie outside the major metropolitan
areas. (21,22) To examine variations in the cost of FV according to
socio-economic level, the Pampalon and Raymond material deprivation
index was used to select neighbourhoods reflecting different
socio-economic contexts of urban area. (23) Deprivation is expressed as
low, medium or high. The index is based on three indicators (education,
employment and income) that determine the socio-economic conditions
under which people live in small geographic areas with a population
between 400 and 700 and for which data are available.
Food stores census
All areas' food stores likely to offer FV were identified and
classified by type (grocery, greengrocer, convenience) using Business
411 software (Tamec Inc., version 7.2, 28th ed., Montreal) and a 2005
database from Quebec's Ministere de l'Agriculture, des
Pecheries et de l'Alimentation. A record was provided by the City
of Quebec (2007) to help us sort the type of grocery store into three
categories by surface area: small (<800 [ft.sup.2], 74.3 [m.sup.2]),
conventional (<30,000 [ft.sup.2], 2,787.1 [m.sup.2]) and large
(<100,000 [ft.sup.2], 9,290.3 [m.sup.2]).
A random stratified sample of food stores was drawn according to
type of store (5 types), urbanization level (2 levels), as well as the
material deprivation index (3 levels). As determined by a statistician,
the aim was to include a minimum of six food stores for each stratum to
produce sufficient statistical power, but for certain categories it was
deemed impossible to reach this number because of the absence of stores
located in the designated area. In all, the study sample had 85 stores.
Fruits and vegetables basket content
A list of the 80 most consumed FV (fresh, canned and frozen) was
first compiled using statistics on consumption patterns of the Quebec
population. (24) This list was refined according to the products, brands
and sizes most commonly found in some food stores of the study's
areas. Final list adjustment was made on the basis of pretests in 46
stores. Because the products offered were different from one type of
food store to another, the list was split into three baskets: a grocery
basket composed of fresh, canned and frozen FV; a fresh FV basket
specifically created for greengrocers containing fresh FV also found in
the grocery basket; and finally, a convenience basket for convenience
stores with canned products also in the grocery basket. Table 1
illustrates the contents of each basket.
Data collection
Four investigators were trained, and data were collected during the
week of 17-23 September, 2007. Product description, brand, variety, unit
of reference and the regular marked price were collected by each
investigator using audio recorders and then transcribed into SPSS
(Statistical Package for the Social Sciences).
Organization and analysis of data
Prices were examined for outliers and missing data. To retain a
maximum of stores and provide as detailed a list of FV as possible, it
was necessary to omit items not available in some stores, with the
exception of the convenience basket. To avoid price imputation problems,
a store was removed from the analysis when at least one of the items
listed in the basket was absent. Therefore, a total of 11 convenience
stores (46% of all convenience stores), 11 small grocery stores (55%),
six conventional grocery stores (27%), and one greengrocer (11%) were
not considered.
To determine whether differences in the cost of FV baskets are
related to the level of urbanization and the material deprivation index,
a series of t tests and analysis of variance were conducted using SPSS
software version 13.0. The distribution of food stores included in the
initial sample was examined by area and by population using chi-square
tests.
RESULTS
Data from 56 food stores (66%) were analyzed (Table 2). Small
grocery stores were placed in the convenience store category because of
their limited supply and variety of FV compared with the other two types
of grocery stores. The convenience store category was thus renamed
"convenience+". Only one small grocery store in our study
provided a supply similar to that of the other types of grocery stores,
and it was thus placed in the conventional grocery category.
Neither the level of material deprivation nor the level of
urbanization influenced the cost of FV baskets (Table 3). The type of
food store was the only factor having a significant effect on the cost.
Figure 1 illustrates that the average price was lower in large grocery
stores and greengrocers when compared with other store types. The price
for a convenience basket was also lower in large and conventional
stores. For the fresh FV basket prices in greengrocers were
significantly lower than in conventional stores and showed a tendency to
be lower than in large stores. Finally, the grocery basket was less
expensive in large than in conventional stores.
Table 4 shows that for all types of store, the total number of
stores per 100 [km.sup.2] was significantly greater in urban settings
than rural ones. However, the distribution of stores per 10,000
inhabitants was the same in both settings. The distribution of stores in
urban settings was also examined in terms of the material deprivation
index, and no differences were found except for the convenience+
category. In this case, the number of convenience+ stores was
significantly greater in areas where material deprivation was high (low
level of deprivation=3 stores, intermediate =17, high=24; p<0.001).
DISCUSSION
The results of this study corroborate those of previous studies
conducted in Canada demonstrating that the store type alone can
significantly influence FV basket cost. (16-18) Large stores and
greengrocers were, on average, cheaper than any other type. This finding
is not surprising given that large grocery stores have more room
available to keep low-price bulk food items and are in a position to
offer a much larger variety of FV to consumers at very competitive
prices. (25) Although a significant difference in the price of just one
of the four items of the convenience store basket could account for all
differences found in the cost of the basket, the reality is that these
four products were the only ones common to all convenience stores.
Moreover, these products were exactly the same (brand names, reference
unit), so we assume they represented the real offer in these stores. The
fact that brand names were the same for each product across the stores
reinforces our result that the variation of cost found among food stores
can be imputed to differences among store types.
In contrast to our results, some previous studies that have
monitored food prices in various settings showed that low-income urban
households and rural inhabitants face greater disadvantages in terms of
economic access to food. (13-15) In the majority of these studies,
discussions emphasize the difficulties in accessing supermarkets or
large stores, resulting in increased food costs. In this study, the fact
that every type of food store was established in each kind of setting
may have accounted for the finding that food costs did not vary with
respect to the level of urbanization and the material deprivation index.
Although basket prices were not influenced by urbanization and
deprivation, the distribution of store types across settings could
contribute to health disparities in the greater Quebec City. In fact, a
recent Montreal study found that 40% of the population did not have
access to fresh FV (food stores <640 [ft.sup.2], 59.5 [m.sup.2])
within walking distance (500 m) from their home. (4) The present study
observed a significantly lower number of food stores of all types per
100 [km.sup.2] in rural settings compared with urban settings. Travel
distances could affect the total cost of FV and expose rural inhabitants
and indeed less privileged households without a car to greater
difficulties in accessing FV. (26)
The absence of large grocery stores could contribute to higher food
prices in a given setting. (14) In the poor urban settings sampled in
this study, convenience stores dominated the food landscape and were
more likely to serve low-income households. Such households are faced
with a very poor supply of FV, which is limited to whatever these stores
are selling, namely canned foods priced much higher than in other types
of stores. In spite of this finding, large stores were also located in
areas where the material deprivation index is high. This could also
contribute to explaining why differences in the cost of FV baskets in
relation to the level of deprivation were not found. To provide stronger
evidence of social inequalities in economic access to FV, more research
is needed to better understand perceptions of low-income people with
regard to their access to large food stores, FV purchase patterns and
available means of transportation.
Methodological considerations confirmed the representativeness of
the food environment of the area under study. For instance, FV baskets
were defined according to the target population's consumption
frequency of food items, as well as the availability of the food items
in the various stores. Greengrocers were included since they are the
second most popular place throughout Quebec to buy fresh FV. (25)
The absence of stores in some strata in the planned sample and the
loss of stores because of missing data have reduced the sample size and
may limit the ability to generalize the results. The reduced sample was,
nevertheless, representative of each area's particular food
environment. While other studies have made use of the price substitution
method (27) in order to prevent loss of data, it is not without bias. If
the price substitution method had been used, the final sample would have
been bigger but not as representative of the area's reality.
CONCLUSION
This study revealed no evidence of price disparity in FV across
different socio-economic contexts. The type of store, however, does
influence cost and, given the distribution of stores, economic access to
FV could become a major issue for households living in a disenfranchised
urban setting or for rural households without a car.
Economic accessibility to healthy food is but one component of the
food environment capable of having an effect on social inequalities in
health. To ensure that people have access to high-quality food, everyone
must have the physical and material means of acquiring the foods
recommended by nutritionists. (27,28)
To meet the goal of improving the food environment and thus
reaching a better overall health status, it is imperative that research
help us interpret and understand the components of the food environment
and how they relate to social inequalities in health in places.
Acknowledgements: We are indebted to l'Agence de la Sante et
des Services Sociaux de la Capitale-Nationale and the funds of Jean-Paul
Houde for study. We also thank the Statistical Consulting Service of the
Department of Mathematics and Statistics of Universite Laval.
Received: December 17, 2008
Accepted: August 6, 2009
REFERENCES
(1.) Pampalon R, Hamel D, De Koninck M, Disant M-J. Perception of
place and health: Differences between neighbourhoods in the Quebec City
region. Soc Sci Med 2007;65(1):95-111.
(2.) Pearce J, Hiscock R, Blakely T, Witten K. The contextual
effects of neighbourhood access to supermarkets and convenience stores
on individual fruit and vegetable consumption. J Epidemiol Community
Health 2008;62(3):198-201.
(3.) James WPT, Nelson M, Ralph A, Leather S. Socioeconomic
determinants of health: The contribution of nutrition to inequalities in
health. BMJ 1997;314(7093):1545.
(4.) Bertrand L, Therien F, Cloutier M. Measuring and mapping
disparities in access to fresh fruits and vegetables in Montreal. Can J
Public Health 2008;99(1):6-11.
(5.) Power E. Les determinants de la saine alimentation chez les
Canadiens a faible revenu. Can J Public Health 2005;96(suppl 3):S42-S48.
(6.) Raine K. Les determinants de la saine alimentation au Canada.
Can J Public Health 2005;96(suppl 3):S8-S15.
(7.) Krebs-Smith S, Kantor S. Choose a variety of fruits and
vegetables daily: Understanding the complexities. J Nutrition
2001;131:487S-501S.
(8.) Garriguet D. Vue d'ensemble des habitudes alimentaires
des Canadiens. Ottawa, ON: Statistics Canada, 2004.
(9.) World Health Organization. Diet, Nutrition and the Prevention
of Chronic diseases: Report of a joint WHO/FAO Expert Consultation. WHO
Technical Report Series 916. Geneva: WHO, 2003.
(10.) Henry H, Reicks M, Smith C, Reimer K, Atwell J, Thomas R.
Identification of factors affecting purchasing and preparation of fruit
and vegetables by stage of change for low-income African American
mothers using the think-aloud method. J Am Diet Assoc
2003;103(12):1643-46.
(11.) Pincemail J, Degrune F, Voussure S, Malherbe C, Paquot N,
Defraigne J-O. Effet d'une alimentation riche en fruits et legumes
sur les taux plasmatiques en antioxydants et des marqueurs des dommages
oxydatifs. Nutrition clinique et metabolisme 2007;21:66-75.
(12.) Block D, Kouba J. A comparison of the availability and
affordability of a market basket in two communities in the Chicago area.
Public Health Nutr 2005;9(7):837-45.
(13.) Frank S. Higher prices, same benefits: An exploration of WIC
food prices in rural and urban Kentucky. Rural Sociological Society
Annual Meeting, Louisville, KY, 2006.
(14.) Fellowes M. From poverty, opportunity: Putting the market to
work for lower income families. 2006. Available online at:
http://www.brookings.edu/ (Accessed June 5, 2008).
(15.) Crockett EG, Clancy K, Bowering J. Comparing the cost of a
thrifty food plan market basket in three areas of New York state. J Nutr
Educ 1992;24:72S-79S. dietary changes in Nova Scotia. J Can Diet Assoc
1997;58:176-83.
(16.) Travers K, Cogdon A, McDonald W. Availability and cost of
heart healthy dietary changes in Nova Scotia. J Can Diet Assoc
1997;58:176-83.
(17.) Duquette MP, Demmers T, Demers J. Etude sur le cout du panier
a provisions nutritif dans divers quartiers de Montreal. Montreal:
Dispensaire dietetique de Montreal, 2006.
(18.) Latham J, Moffat T. Determinants of variation in food cost
and availability in two socioeconomically contrasting neighbourhoods of
Hamilton, Ontario, Canada. Health Place 2007;13:273-87.
(19.) Hamelin A, Ouellet D, Lamontagne C, Dugal C. Etude pilote sur
la variation du cout des aliments dans la region de Quebec. Cahiers de
nutrition publique, GENUP, Universite Laval 2007;5(Aout).
(20.) Statistics Canada. Classification des secteurs statistiques
(CSS). 2002. Available online at: http://www.statcan.ca (Accessed
February 21, 2007).
(21.) McNiven C, Puderer H, Janes D. Zones d'influence des
regions metropolitaines de recensement et des agglomerations de
recensement (ZIM) : une description de la methodologie. Ottawa, ON:
Statistics Canada, 2000.
(22.) Statistics Canada. Zone d'influence metropolitaine
(ZIM). 2003. Available online at: http://geodepot.statcan.ca/ (Accessed
February 15, 2007).
(23.) Pampalon R, Raymond G. Un indice de defavorisation pour la
planification de la sante et du bien-etre au Quebec. Chron Dis Can
2000;21(3).
(24.) Ministere de l'Agriculture, Pecheries et Alimentation du
Quebec. Les depenses alimentaires des Quebecois: constats et limites de
comparaisons. Bio clips+ 2005;8(5).
(25.) Groupe AGECO. Portrait des reseaux de distribution des fruits
et legumes frais du Quebec. Quebec, 2007.
(26.) Hendrickson D, Smith C, Eikenberry N. Fruit and vegetable
access in four low-income food deserts communities in Minnesota. Agric
Human Values 2006;23:371-83.
(27.) Sante Canada. Plan d'action du Canada pour la securite
alimentaire: une reponse au Sommet mondial de l'alimentation.
Ottawa, ON: Agriculture et agroalimentaire Canada, 1998.
(28.) Pouliot N, Hamelin A-M. Disparities in fruit and vegetable
supply: A reason for health concern in the greater Quebec City. Public
Health Nutr 2009; Advance access published April 2009.
Sarah Drouin, MSc, RD, Anne-Marie Hamelin, PhD, RD, Denise Ouellet,
PhD, RD
Groupe d'etudes en nutrition publique (GENUP), Departement des
sciences des aliments et de nutrition, Universite Laval, Sainte-Foy, QC
Correspondence and reprint requests: Sarah Drouin, GENUP, 2412,
Pavillon Paul-Comtois, Universite Laval, Quebec, Tel: 418-656-2131
#4140, E-mail: sarahdrouin@ hotmail.com
Table 1. Contents of Fruit and Vegetable Baskets
Baskets
Reference
Product Unit Grocery
Fresh apples 1kg X
Fresh bananas 1kg X
Fresh oranges 1kg X
Fresh pears 1kg X
Fresh white potatoes 4.54 kg X
Fresh tomatoes 1kg X
Fresh carrots 1kg X
Fresh yellow onions 1.36 kg X
Fruit salad canned 796 mL X
Peaches canned 796 mL X
Stewed apples canned 796 mL X
Whole tomatoes canned 796 mL X
Corn kernels canned 341 mL X
Cream corn canned 398 mL X
Small sweet peas canned 398 ml X
Frozen strawberries 600 g X
Baskets
Product Fresh FV * Convenience
Fresh apples X
Fresh bananas X
Fresh oranges X
Fresh pears X
Fresh white potatoes X
Fresh tomatoes X
Fresh carrots
Fresh yellow onions X
Fruit salad canned
Peaches canned
Stewed apples canned
Whole tomatoes canned X
Corn kernels canned X
Cream corn canned X
Small sweet peas canned X
Frozen strawberries
* Fruits and vegetables.
Table 2. Number of Food Stores by Type
and Socio-economic Context
Urbanization Level
Urban
Index of material deprivation
Low Intermediate High
Convenience+ * 2 5 7
Conventional grocery 3 4 5
Large grocery 2 3 2
Greengrocer 2 3 2
Total (%) 9 (16.1) 15 (26.8) 16 (28.5)
Urbanization Level
Rural
Total (%)
Convenience+ * 7 21 (38)
Conventional grocery 5 17 (30)
Large grocery 3 10 (18)
Greengrocer 1 8 (14)
Total (%) 16 (28.5) 56 (100)
* Convenience+ type includes small
grocery stores and convenience stores
distributed as follows: in urban areas 5
small grocery stores vs. 9 convenience
stores; in rural areas 3 vs. 4.
Table 3. Cost Comparison of FV
Baskets by Level of Urbanization,
Material Deprivation and Food Store Type
Grocery Basket *
n t / F p
Urbanization level 27 0.80 0.43
Index of material deprivation 19 0.06 0.94
Food store type 27 2.56 <0.01
Fresh FV Basket *
Urbanization level n t / F p
Index of material deprivation 35 0.77 0.45
Food store type 26 1.67 0.22
35 4.09 0.04
Convenience Store
Basket *
Urbanization level n t / F p
Index of material deprivation 48 0.53 0.60
Food store type 33 0.41 0.67
48 16.19 <0.001
* n = number of food stores included in
the analysis; t/F = results of t tests
(t) or analysis of variance (F); p =
level of significance
Table 4. Food Store Distribution by
Urbanization Level in Four Areas of
Greater Quebec City
Convenience Small Conventional
Store Grocery Grocery
Urban
N 132 36 20
N/10,000 res. * 7.87 2.15 1.19
N/100 [km.sup.2] 181.1. 49 27
([dagger]) ([dagger]) ([dagger])
Rural
N 34 8 6
N/10,000 res. * 9.54 2.24 1.68
N/100 [km.sup.2] 1.54 0.36 0.27
Large Greengrocer Total
Grocery
Urban
N 7 10 205
N/10,000 res. * 0.42 0.6 12.23
N/100 [km.sup.2] 9.6 14 280.7
([dagger]) ([dagger])
Rural
N 3 2 53
N/10,000 res. * 0.84 0.56 14.86
N/100 [km.sup.2] 0.14 0.09 2.4
* Residents
([dagger]) p<0.05
Figure 1. Comparison of the mean costs of fruit and
vegetable baskets by food store types
Large Conventional Greengrocer Convenience+
grocery grocery
Grocery 38.95 42.77
basket ([double ([double
dagger]) dagger])
Fresh
FV
basket 19,27 21,28 17,49
([double ([double ([double
dagger]) dagger]) dagger])
Convenience
basket 4,59 5,16 5,65
([dagger]) ([dagger]) ([dagger])
* p<0.001 ([dagger] ) p<0.05 ([double dagger]) p=0.01
Note: Table made from bar graph.
