The Utilization of Local Food Environments By Urban Seniors

INTRODUCTION

Over the past decade, researchers have investigated the associations between food environments, diet and health outcomes (Austin SB, 2005; Block JP, 2004; Cheadle A, 1991; Cheadle A, 1993; Cummins SC, 2005; Fisher BD, 1999; Hearn MD, 1998; Horowitz CR, 2004; Laraia BA, 2004; Lewis LB, 2005; Maddock J, 2004; Morland KB, 2002a; Morland KB, 2002b; Morland KB, 2006; Sooman A, 1993; Sturm R, 2005; Wechler H, 1995; Zenk SN, 2005a; Zenk SN, 2005b;). Studies have demonstrated that access to foods differs by neighborhood characteristics such as socioeconomic and race/ethnic composition (Austin SB, 2005; Block JP, 2004; Cummins SC, 2005; Horowitz CR, 2004; Lewis LB, 2005; Morland KB, 2002a; Sooman A, 1993; Wechler H, 1995; Zenk SN, 2005a;). In addition, proximity to supermarkets has been shown to be linked to diet and related health outcomes in some social groups (Cheadle A, 1991; Laraia BA, 2004; Maddock J, 2004; Morland K, 2002b;; Morland KB, 2006; Sturm R, 2005; Zenk SN, 2005b). There is also some evidence that changes in environments (e.g. school environments) has an influence on the purchasing of healthier food options (French SA, 2003; Gortmaker SL, 1999; Lytle LA, 2004). These studies have made important contributions to the public health literature by beginning to provide empirical evidence of the impact of environmental factors on individuals’ dietary behaviors. However, fewer studies have measured how individuals interact with and utilize their local food environments. Understanding the ways in which individuals navigate and make use of food purchasing options could have important implications for the design of interventions to improve access to healthy foods.

The elderly population is the fastest growing sub-population in the United States. Roughly one in eight Americans were over the age of 65 in 1994 and it is estimated that by 2030 one in five Americans will be 65 or older (Hobbs FB, 1995). This population is at greater risk for chronic disease such as heart disease, stroke, diabetes and hypertension, all of which can be controlled with diet and exercise (U.S Department of Health and Human Services, 1990). The majority of U.S. Seniors live independently and are often self reliant for obtaining their daily meals. It has been shown that grocery shopping among Seniors is restricted to their immediate neighborhood and therefore proximity to grocery stores contributes to the well being of Seniors (Smith GC, 1991; Smith GC, 1995).

This study aims to (a) describe food shopping patterns for urban Seniors and (b) measure the influence of neighborhood and individual factors on servings per day of fresh fruits and vegetables. Because previous research in the United States demonstrates that a greater proportion of supermarkets are located in white areas (Morland KB, 2002), and that these types of food stores carry the greatest variety of produce (Morland KB, 2007), we hypothesized that Seniors living and shopping in black or racially mixed areas would consume fewer servings of fruits and vegetables and travel further to obtain produce. Finally, we aimed to measure the relative contribution of neighborhood and individual-level factors on the variation of fruit and vegetable intake among Seniors.

METHODS

A cross sectional study was conducted from September 2005 to August 2006 in which 314 participants were recruited from ten Brooklyn, NY senior centers. Senior center events coordinators scheduled daytime recruitments at each site and an interview-administered questionnaire was conducted in English or Spanish by trained research staff. Respondents were queried about how and what types of produce were usually purchased for their household, satisfaction with selection, cost and quality of produce, intake of produce, as well as location of the primary food store used to purchase fruits and vegetables.

RESULTS

The mean age for participants was 72 years and the majority of participants were women (81.3%) and either black American or Latinos (77.4%) (Table 1). Most Seniors were either divorced, single or widowed (82.1%) and only a third lived in a household containing more than one adult. Nearly half of the Seniors lived in racially-mixed areas, while one-quarter lived in predominantly black areas. Most of the participants reported that they did their own food shopping (90.7%) and the mean distance traveled to a primary food store was 0.8 miles (with only 17% of seniors reported they shopped within their census tract of residence). A large proportion of participants living in predominately white areas reported that they also shopped in white areas (89.2%), whereas only about half of residents of black or racially mixed areas reported shopping in areas of similar race composition to their census tract of residence. Mean daily servings of fruits and vegetables were close to five or just above five for all race/ethnicity groups. On average, individuals living in predominately white areas reported the lowest intake.

Seniors’ produce purchasing patterns are presented in table 2 by race/ethnicity. The greatest variation in purchasing produce between the three race/ethnic groups were observed in the purchasing of eggplant, mushrooms, berries, green beans, corn, asparagus, squash, cauliflower and nectarines (where p-values were less than 0.05). Other produce such as oranges, broccoli, bananas, leafy greens, carrots, grapes and peas are reported to be purchased by over 80% of all Seniors during the past month Over one third of all Seniors reported not purchasing certain produce because they did not like the taste. Less than 16% of all participants reported that the availability of any type of produce influenced their purchasing patterns. One third of Blacks reported that some produce types were too low in quality to be purchased. However, when asked about overall satisfactions with quality of produce purchased, most seniors of all race/ethnic groups reported that they were satisfied. Roughly 25% of all groups reported being not very satisfied or completely unsatisfied with the price of produce purchased. A greater proportion of Blacks travel more than one mile to obtain produce, followed by Latinos and Whites (32.3%, 22% and 13.8% respectively).

Table 3 describes the influence of neighborhood and store location neighborhood characteristics; distance traveled to primary food store and individual level factors on servings per day of fruits and vegetables among urban Seniors. A 0.03 increase in servings per day of fruits and vegetables was observed for each 10^th of a mile traveled to Seniors’ primary food store (mean difference (MD) = 0.029, standard error (SE)=0.135). Moreover, individuals living in Black or racially-mixed areas report higher intake of produce compared to seniors living in predominately White areas (MD Black=0.668, standard error (SE)=0.618; MD mixed=0.523, SE=0.437). Seniors shopping in black neighborhoods also had a higher intake of produce (MD=0.770, SE=0.595) than those shopping in predominately white areas. Nevertheless, Blacks and Latinos reported fewer servings of fruits and vegetables (MD Black = −0.827, SE=0.474; MD Latino =−1.334, SE=0.429). Age is also inversely associated with produce intake, whereas being male and marital status other than married, were positively associated with intake. Only individual level factors (age, gender and race/ethnicity) were statistically significant (alpha <0.05).

DISCUSSION

This study illuminates a number of important factors about the delivery of foods to urban Seniors and how those Seniors navigate their local food environment to obtain healthy diets, measured here as intake of fruits and vegetables. For instance, 90% of the Seniors surveyed did their own shopping and traveled an average of 16 city blocks to get to the store. Furthermore, less than 20% of these Seniors shopped within their residential census tract. Secondly, while most Seniors do not shop within their residential census tract, nearly 90% of Seniors living in predominately white census tracts also shop in predominately white census tracts. Little differences between race/ethnic groups were observed in relation to satisfaction with the quality or price of produce purchased. However, black Americans more often reported quality to be a reason for not purchasing some produce types. Third, dietary intake of produce was high for these Seniors where average intakes were roughly five per day as recommended. Seniors’ reported dietary intake is supported by their equally high responses of purchases of fresh produce during the previous month. Out of the 28 types of produce queried, 80% or more of all Seniors reported buying nearly half of all varieties. Finally, although we did observe fewer servings of produce among Blacks and Latinos, the associations between racial makeup of residential and primary food store locations and intake did not support our hypothesis, suggesting that variation in dietary intake by race/ethnicity cannot be fully explained by neighborhood availability of foods. However, these findings may be influenced by the fact that most Seniors do not shop within residential census tracts (the geographic boundary used to describe their local food environment in this study) and hence differential misclassification by neighborhood type may have influenced our results.

Our findings are limited by a number of factors. First, because Seniors were recruited from Senior Centers, this group of Seniors may be healthier and more mobile than other Seniors of similar ages. Furthermore, the majority of these Seniors do their own shopping and cooking (where 68% report eating breakfast at home and 90% report eating dinner at home most often). Any effect observed regarding reliance on a local environment for the purchase of produce may be underestimated with this group. Second, these Seniors may also be unique in the fact that the reported servings per day of fruits and vegetables were high regardless of race or residential neighborhood type. This finding due may in part be due to the large portion sizes reported. As people age, their portion sizes tend to get smaller. Yet for the participants in this study, the mean serving size was greater than one for almost all categories of foods queried in the Fruit and Vegetables Screener. However, since portion sizes do not vary by race or neighborhood of residence, differential bias has not likely occurred. Third, although we asked Seniors where they did their primary shopping, we did not directly measure what types of foods were available in those stores, nor did we collect data on the types of food stores located in residential areas. Fourth, we are limited by our observational study design where we sampled Seniors from Seniors Centers, which resulted in our study population being distributed over an area in close proximity to the Center. Our sampling method did not allow us to over-sample specific race/ethnic groups within certain neighborhood types. Hence, we had small cells for some race/ethnic groups within specific neighborhood types and therefore low statistical power to detect effects of each race/ethnic group living and shopping in each of the neighborhood types. Finally, models did not include produce satisfaction measures because models would not converge which limits out ability to make conclusions on how these factors would effect our results.

CONCLUSIONS

The few studies that have investigated competing factors that influence dietary choices (cost, food quality, convenience, etc.) have shown proximity to food to be an important factor for determining diet (U.S. House of Representatives Select Committee on Hunger, 1987; Mooney C, 1990; U.S. House of Representatives Select Committee on Hunger, 1992; Turrell G, 1996; Glanz K, 1998; Cade J, 1999). However, the causal mechanisms involved in how Seniors maintain a healthy diet are a complex interplay between many individual level and environmental factors. A better understanding of how different groups of people go about these activities of daily living has potential importance in a clinical setting where Seniors are at particular risk for diet related illness, but also has significance as we further develop public health programs to address disparities in access to healthy and affordable foods. Despite the fact that statistical significance was not reached, even in densely urban environments, this study found that for every mile traveled to a primary food store there is a 0.3 increase in servings of fruits and vegetables (6% of the recommended amount). Although small, the increase in servings per day associated with distance traveled to primary food stores suggest that fruits and vegetables are not locally available and presents an opportunity for policy makers and city planners to develop areas where healthy food options are convenient for consumers. Further research is needed to evaluate how Seniors, as well as other groups of people living in other urban and non-urban settings, navigate their built environments to meet their basic needs.