Retail Food Environment in an Urban Resettlement Colony of Delhi: A Community-Based Cross-Sectional Study

Abstract

Background:

The changing retail food environment contributes to increasing burden of obesity and noncommunicable diseases. The study assessed the density of retail stores along with availability, price, and placement of food options in a low-income urban community of Delhi, India.

Materials and Methods:

An audit of retail stores was conducted using an adapted Nutrition Environment Measure Survey tool covering nine locally relevant food items. For each item, healthier and less healthy options were identified based on global recommendations. Data were collected by personal observation and response of the shopkeeper. The study area was estimated using geospatial methods and population by electoral list. Results are expressed as proportions for availability, price ratio of less healthy to healthier food items, and density of stores per 100,000 population and per sq. km.

Results:

Among 70 stores, 88.2%, 7.4%, and 4.4% were small and large corner stores and small convenience stores, respectively. The density of food retail stores was 304 stores per sq. km and 99 stores per 100,000 population. A similar proportion of healthier and less healthy options were available for items like wheat flour, pulses, salt, milk, and bread. The price ratio was <1 for all items except for wheat flour and milk. The mean (95% confidence interval) number of less healthy items displayed at the cash counter, 7.38 (6.48–8.28), was significantly higher (P < 0.001) than that of healthier items, 2.74 (2.28–3.20).

Conclusions:

The food environment was characterized by being predominantly small corner stores based on high density, equal availability of less healthy and healthy options, and adverse pricing and placement for healthier options.

BACKGROUND

Worldwide, obesity has nearly tripled since 1975. In 2016, more than 1.9 billion adults were overweight; of these, over 650 million were obese.[1] Obesity, in turn, has a contributory role in the development of noncommunicable diseases (NCDs). The fundamental cause of increasing obesity and overweight is an energy imbalance between calories consumed and calories expended. The calories consumed are determined by the food choices, which in turn would be influenced, among other things, by the neighborhood food environment.[2] The World Health Organisation (WHO) and the Food and Agriculture Organisation (FAO) have recognized the impact of the food environment on society’s health.

ANH-FEWG has defined the food environment as the interface that mediates one’s food acquisition and consumption within the wider food system[3] and by the European Public Health Alliance as a space where food is made available, accessible, and affordable.[4]

The food environment includes community, consumer, organizational, and informational components.[5] Given the major role of the food environment in the development of NCDs,[6] it is important to assess the retail food environment. Some of the quantitative indicators of the retail environment include the density of food retail stores, availability and price of healthy food choices in the stores, and placement of healthy food choices in stores.[7] Nutrition Environment Measure Survey (NEMS) tool is one standard tool developed for objective measurement of the retail food environment. In contrast, other subjective tools measure perceived access and affordability to specific food items like fruits and vegetables or unhealthy food items like processed foods.[8,9] At present, most of the retail stores globally, especially in low- and middle-income communities, do not encourage healthy eating.[10,11]

In India, there is an increasing trend in the prevalence of obesity among adults, from 13.5% in 2005–2006[12] to 23.5% in 2019–2021.[13] Almost 3.4% of under-5 children in India are obese.[13] The number of food retail stores in India is estimated to be around 13 million as of 2022, and the retail food environment is in transition due to global influences.[14] This study aimed to assess the retail food environment of an urban resettlement colony of Delhi by estimating the density of stores and assessing the availability, price, and placement of healthy food options in these food retail stores. A food retail store was defined as a fixed store selling groceries with or without additional items and excluded roadside vendors and eateries.

MATERIALS AND METHODS

A community-based cross-sectional study was conducted in an urban resettlement colony in Dakshinpuri extension, Ambedkar Nagar, New Delhi. This study was part of a larger study that included a household-based component in the same area on their healthy food purchasing practices.[15] The study received ethical clearance from the Institute Ethics Committee. The household-based component of the larger study was performed to estimate the proportion of healthier food purchases made by the households and also determine the determinants of their healthier food purchases. The present study was conducted among the food retail stores in the study area.

The colony is divided into ten blocks, further divided into numerous lanes of 20 plots with 3–4 story flats in each plot. The population of the area was ascertained by the electoral roll published by the chief electoral officer, Delhi, for the year 2021[16] and considered under 18 years of population to be 30%. During the house-to-house visits for the healthier food purchase study, food retail stores visited by the households for food purchases were listed. The list was supplemented by physically touring the neighborhood to identify all food retail stores. Geographical mapping of these identified stores was done in Google My Maps. A total of 70 such food retail stores were identified and formed the sample for the retail food environment assessment. Therefore, we included all the food retail stores in the study area. The shopkeeper provided written informed consent, which was included in the study. The shops found closed after two visits were excluded from the study.

An adapted NEMS tool was used for the audit. The NEMS store tool[17] assesses the price and availability of 11 food items, including both healthier and less healthy options. The food items included in the original scale were milk, fresh fruits and vegetables, ground beef, hot dogs, frozen dinners, baked goods, beverages (soda/juice), whole grain bread, baked chips, and cereal. Since these were not appropriate to the Indian context, we replaced these with more appropriate healthier and less healthy options in different food groups. Food items retained were milk, baked goods, beverages, whole grain bread, and cereals, whereas other food items were replaced with grains, pulses, cooking oil, salt, and snacks. Healthier options for the identified food item were based on relevance for the prevention of NCDs and as per global and national healthy eating recommendations.[18,19,20]

A semistructured questionnaire with sections on general information about the store, including the availability of cold storage facility, products being stored, cash counter assessment, cost, availability, and placement in store of healthier options in comparison to their less healthy counterpart, was used. The tool was piloted in ten retail stores in a nearby area before finalization. The tool is given as a supplementary file.

Assessments included both personal observation and as reported by the storekeeper. Data collection was done by a visit to the store by the lead author. In each store, informed consent from the owner was taken. Assessments for a type of store, cold storage, and cash counter in the store were based on direct observation from the cash counter, whereas the rest of the information was obtained from the storekeeper. Availability and placement of food items were assessed by the storekeeper’s reported response if not visible by direct observation from the cash counter.

Food stores were classified as corner stores (stores where the shopkeeper hands out the food items), convenience stores (where customers can pick up the food items and move to the cash counter with filled baskets), or supermarkets (stores selling food and nonfood items, where the customer can pick up the food items and move to the cash counter with a filled basket). Small stores were those with one cash counter, whereas those with more than one cash counter were classified as large stores. Placement of food items within the store could be either prominent (i.e., can easily be seen from the purchase counter) or inconspicuous (i.e., that cannot be seen easily from the purchase counter).

Data entry was done in Epi Info, and analysis was carried out using Stata version 15.0 (College Station, Texas, USA). The key outcomes of the assessment were density of retail stores in the study area and availability, price, and placement of healthier food items as compared to less healthy options in food retail stores. The density of stores was expressed as the number of stores per 100,000 population and stores per sq.km. The total area of the study site was estimated using Google Earth software. The availability of food items was expressed in terms of frequency counts and percentages. The price of food items (in USD) was expressed in terms of mean and standard deviation. The conversion rate considered as of August 2021, that is, 1 USD, equals 74.2 INR. The mean price ratio of less healthy to healthier food items was calculated. Unpaired t-test was used to assess the statistical significance of the difference in number of healthier and less healthy food items at the cash counter. The Mann–Whitney U-test was used to assess the statistical significance of differences in the placement of healthier and less healthy food items in stores.

RESULTS

Of the 70 food retail stores mapped, 62 (88.2%) were small corner stores, 5 (7.4%) were large corner stores, and 3 (4.4%) were small convenience stores. No supermarket was identified. The total study site area was estimated as 0.23 sq km with a population of 70,756. The density of the food retail store was 304 stores per sq km and 99 stores per 100,000 population. The food retail stores were mostly situated at either end of the lane or very few inside the lanes [Figure 1].

Figure 1.

Geographical mapping of food retail stores in the study area

Two store owners refused their consent. Around 55 (80.9%) stores had cold storage facilities and 52 (76.5%) stores were storing carbonated drinks, 52 (76.5%) stores yoghurt (76.5%), 50 (73.5%) milk, and 37 (54.9%) fruit juice. Two (2.9%) stores were part of a franchise. All stores had a similar proportion of healthier and less healthy options for staple food items like wheat flour, pulses, salt, milk, and bread. However, less healthy options for sugar, beverages, and processed food were more commonly found. Healthier cooking oil options were more available. The mean price ratio of less healthy to healthier food items was less than 1, indicating that, in general, healthier options were costlier for almost all items except for wheat flour and milk [Table 1].

Table 1.

Comparison of availability and price of healthier and less healthy food items in retail stores (n=68)

Food items	Availability in stores [n (%)]		Price of options in stores expressed as USD/Kg
	Healthier option	Less healthy option	Healthier option (Mean±SD) (A)	Less healthy option (Mean±SD) (B)	Price Ratio (B/A)*
Wheat flour	67 (98.5)	67 (98.5)	0.4±0.1	0.6±0.1	1.5
Grains (Millets)	27 (39.7)	0	0.3±0.1	None	NA
Pulses	64 (94.1)	64 (94.1)	1.4±0.1	1.3±0.1	0.9
Cooking oil	66 (97.1)	4 (5.9)	2.3±0.2	1.5±0.2	0.7
Salt	52 (76.5)	52 (76.5)	0.5±0.1	0.3±0.0	0.6
Sugar	36 (52.9)	66 (97.1)	0.7±0.2	0.5±0.1	0.7
Processed food	6 (8.8)	55 (80.9)	2.1±0.9	0.9±0.3	0.4
Milk	56 (82.4)	53 (77.9)	0.7±0.1	0.8±0.1	1.1
Bread	51 (75.0)	51 (75.0)	1.7±0.2	1.1±0.2	0.6
Snacks	43 (63.2)	67 (98.5)	3.2±1.4	2.3±0.5	0.7
Beverages	7 (10.3)	55 (80.9)	1.3±0.8	0.9±0.2	0.7

*A ratio above 1 indicates that less healthier options are costlier

An almost equal proportion of stores had healthier (91.2%) and less healthy (98.5%) options at the cash counter, but less healthy (7.4; 95% CI: 6.5–8.3) options at the cash counter were many more as compared to healthier (2.7; 95% CI: 2.3–3.2) options. Similarly, for the shelves behind the cash counter, 55 (80.9%) stores had both healthier and 59 (86.8%) stores had less healthy options, but there was a significant difference (P value: 0.01) between the mean number of healthier (3.8; 95% CI: 3.3–4.3) and less healthy (6.4; 95% CI: 5.6–7.1) options. Overall promotional activities in all the stores were very poor. Only two stores (2.9%) had promotional materials near the cash counter, and three stores (4.4%) had promotional materials for healthier food items.

Wheat flour, bread, pulses, and cooking oils were placed prominently in the store. Both healthier and less healthy options of wheat flour (88.2%) and bread (60.3%) were equally likely to be placed prominently. Cooking oils available in stores were mostly healthy and were placed prominently in 43 (63.2%) stores. All sugar and salt options were mostly placed inconspicuously, while less healthy snacks were usually prominently placed in 60 (89.1%) stores [Table 2].

Table 2.

Prominent placement of healthier and less healthy food options in food retail stores (n=68)

Food items	Number of items (Mean±SD)	Stores with healthier options	Stores with less healthy options	P*
Wheat flour	3.3±1.3	60 (88.2)	60 (88.2)	0.29
Grains (millets)	1.0±0.0	9 (13.2)	0	0.00
Pulses	10.9±2.6	41 (60.3)	43 (63.2)	0.18
Cooking oil	4.2±2.5	43 (63.2)	0	0.00
Salt	3.3±0.6	6 (8.8)	20 (29.4)	0.01
Sugar	1.5±1.2	2 (2.9)	22 (32.4)	0.01
Processed food	9.3±4.5	4 (5.9)	45 (66.2)	0.00
Milk	2.7±1.6	15 (22.1)	15 (22.1)	0.25
Bread	2.9±1.3	41 (60.3)	41 (60.3)	0.19
Snacks	2.8±2.6	18 (26.5)	60 (88.2)	0.01
Beverages	6.2±1.3	2 (2.9)	49 (72.1)	0.00

*Mann–Whitney U-test was applied to find out statistical difference between prominent placement of healthier and less healthy options. NA: Not Applicable

DISCUSSION

The study done in a low-socioeconomic-status neighborhood with predominantly small corner stores estimated the density of food retail stores and assessed the availability, affordability, and placement of healthier options of food items in these stores. The study area was an urban resettlement colony where the majority of the households belonged to low socioeconomic status. The households in this area purchase food items two to three times a week from small corner stores compared to monthly purchases from convenience stores by households in more affluent urban areas.[21]

The food retail stores in the study area were densely located, that is, 304 stores per sq km of area, which was way beyond most of the reported retail food density in different areas.[22,23] These studies reported the retail store density in the range of 9–250 stores per sq km. Other studies on retail food store density usually report more convenience stores and supermarkets. Such stores are usually larger and can store more options and serve more people, thereby reducing the overall density of retail stores in the area. It is to be noted that the overall population density of the study area is itself high (i.e., 307,635 inhabitants per sq.km area).

While the food retail stores were storing healthier staple food items, the overall availability of healthier options was poor. The prices of healthier food items were consistently higher than the less healthy options, except for wheat and milk. This finding was similar to the one reported by Rao et al.[24] in a systematic review and meta-analysis in 2013 and Zorbas et al.[7] in 2022. Rao et al.[24] reported that healthier options in dairy were slightly less expensive per serving (-$0.004/serving; 95% CI -$0.005 to -$0.004). Zorbas et al.[7] reported that all the healthier food options compared to current options were consistently more expensive. Nonavailability of multigrain wheat and the availability of refined wheat flour at the stores in the present study accounted for the difference in prices. It is to be noted that the production and distribution costs of healthier options of milk are less as compared to the less healthy options.[25]

Barnes et al.[26] in 2016 reported that 77% of stores had at least one healthier option of food item featured at the cash counter. Barnes et al.[26] conducted the study in convenience stores which differs significantly in placement of food items as compared to the corner stores. Schultz et al.[27] in 2022 reported more than 50% of end-of-aisle displays included less healthy options of food items, which is similar to the present study. The placement of the food items in the stores is based on the demand and supply of the food items among the consumers. Less healthy options of food items like snacks, beverages, and processed food were prominently placed. Kholina et al.[28] in 2021 reported that less healthy snack foods and sugar-sweetened beverages were more prominently displayed in food retail stores than their healthier options with the exception of cereal. In the present study, both healthier and less healthy options of wheat were prominently placed. Promotion of any food items was hardly seen in the study area. Barnes et al.[26] in 2016 reported that promotions for healthier options of foods and beverages were found in less than half of stores and for less healthy items were found in approximately half of stores.

The strength of the study was that all the corner stores in the given geographical area were listed and surveyed, and a standard tool was adapted to the local context. The limitations of the study were that we adapted the tool from a global tool developed for a convenience store rather than for corner stores; there is a need to develop and validate a retail food environment tool suitable for the Indian context. Vendors and special weekly markets were not included in the study even, though most vegetables and fruits are probably purchased from these sources. It is likely that the stores studied cater to a larger population beyond the boundaries and people within the boundaries make their food purchases beyond these stores. Therefore, the density of stores may not have been properly assessed. Some of the data obtained were reported by the shopkeeper, which can lead to bias.

The findings of this study call for interventions aimed at improving the food retail environment in India. The Baltimore Healthy Stores (BHS) program in 2004–2005 focused on improving the food environment and educating store owners and adult customers.[29] The food environment was improved by increasing the availability of healthy food in stores. Owners of the stores were provided with stocking guidelines, promotional materials, incentive cards to wholesaler stores, and small supplies of promoted foods to increase the demand.

Simultaneously, our study among the households in that area showed deficiencies in the perceptions of healthy food and that awareness and affordability were important considerations for healthy food purchases.[15] Thus, interventions like posters and flyers which convey the benefits of healthy food have been used to increase consumer demand. The Minneapolis Healthy Corner Store program[30] in 2010 focused mainly on store enhancement, including store layout rearrangements that increase the visibility of healthier food options and hanging signage to advertise fresh produce. While such interventions have been shown to work in the context of convenience stores, we foresee challenges in their implementation in the Indian context, given the high-density corner store-based food retail environment measurement. While this study emphasizes the need to modify the retail food environment, we must develop workable models in the Indian context.

CONCLUSION

The study area was characterized as densely populated. Overall, the availability of healthier options was poorer. Healthier options of food items were costlier than less healthy ones, and less healthy options were more likely to be prominently placed at the store and also near the cash counter.

Conflicts of interest

There are no conflicts of interest.

Funding Statement

Nil.

SUPPLEMENTARY FILE

Supplementary Table 1.

Retail store environment assessment tool

Retail Store Environment Assessment Tool Store Definitions: Corner Sore – Shopkeeper hands out the products. Convenience Store – Customer can pick up the products and moves to cash counter with filled basket Retail Store/ Supermarket: Sells other products as well Small Versus Large – one cash counter versus more than one Name of the Store: GPS Co-ordinates Name of the Respondent: Age: Sex: Type of Store – Small Corner Store/large Corner Store/Small convenience store large Convenience Store/Big retail store or supermarket Part of a Franchisee: yes/No If yes Name____________________________ Time Open: ____ Presence of Cold storage facilities: Yes/No Number and kind of food products stored: Beverages - Carbonated Drinks_______________________Fruit Juices___________________ Flavored Milk________________ Milk__________________Milk products Yoghurt)________________________ Cheese____________ Fruits__________________________ Vegetables___________________ Cash Counter Assessment Eatables on the counter: _ Number of healthy-options: __________________ Unhealthy options:__________________ _____________________________________ Eatables shelf behind the counter: Number of healthy-options: __________________ Unhealthy options: _______________________________ Promotion Materials near cash counter: Number of healthy-options: __________________ Unhealthy options: ___________________________ Promotion Material on the Store: Number of healthy-options: __________________ Unhealthy options: _________________ How frequently do you get requests from customers for healthier food options? Lots of times/Many times/Sometimes/Rarely or never
Instructions to fill: 1. Fill for most commonly sold options. 2. While comparing price use the same unit for both – be it Kg or Litre or any other unit.
Healthy options	Available Yes/No	Price Per Kg or Unit		Number of varieties available		Promotion/advertisement or discount available		Store placement (Inconspicuous/Prominent)		Shelf Placement (Bottom/Prominent/Cash Counter)
		Healthy Option	Unhealthy Option	Healthy Option	Unhealthy Option	Healthy Option	Unhealthy Option	Healthy Option	Unhealthy Option	Healthy Option	Unhealthy Option
M1. Wheat flour with fibre
M2. Coarse grains (millets)
M3. Red or unpolished Rice
M4. Unpolished pulses (Non-dhuli)
M5. Unsaturated oil
M6. Low sodium salt,
M7. Sugar alternatives
M8. Honey
M9. Almonds, Kaju, Akhrot
M10. Low salt/fat containing products
M11. Diet Coke/Soda or other low sugar-based drinks
M12. Frozen fruits or vegetables
M13. Canned Fruit juices with no sugar
W1. Green leafy vegetables
W2. Fresh Fruits
W3. Toned and skimmed milk
W4. Fish/sea food
W5. Baked food products
W6. Brown/Multigrain Bread
O1.
02.
03.

Supplementary Table 2.

Healthier and less healthy options of food items considered for food retail store assessment

Food items	Healthier options	Less healthy options
Wheat flour	Wheat flour with fibre	Refined wheat flour
Millets	Major millets	No less healthy option identified
Pulses	Unpolished/unprocessed pulses	Polished/processed pulses
Cooking oil	Unsaturated and plant-based oils	Saturated or partially hydrogenated oil
Salt	Low sodium salt	Normal table salt
Sugar	Sugar substitutes	Refined white sugar
Processed food	Wholewheat noodles/pasta	Noodles/pasta made of refined flour
Milk	Low-fat/skimmed/toned milk	Full fat milk
Bread	Wholewheat/multigrain/brown bread	White Bread made of refined flour
Snacks	Baked food products	Fried food products
Beverages	Diet coke/soda	Sugar-rich carbonated drinks