Prevalence of online food delivery platforms, meal kit delivery, and online grocery
use in five countries: an analysis of the 2022 International Food Policy Study

doi:10.21203/rs.3.rs-4987799/v1

Background: Online food retail is increasing in popularity, with user-friendly apps, new delivery models, and the influence of COVID-19 lockdowns changing consumer expectations. We aimed to quantify the prevalence of online food retail platform use (defined as online food delivery platforms, online grocery retailers, and online meal kit delivery services) across Australia, Canada, Mexico, the United Kingdom and the United States. We also assessed the associations between sociodemographic and behavioural factors and use of online food retail.

Methods: We conducted a cross-sectional online survey with adults as part of the 2022 International Food Policy Survey (n=19 877). We described the frequency of use and number of meals ordered using different online food retail and delivery platforms. Logistic regression models were fitted to assess associations between the use of online food retail and delivery platforms, and sociodemographic and behavioural factors (including age, sex, household composition, BMI, income adequacy, ethnicity, cooking skills, nutrition knowledge, and frequency of food preparation). All estimates were weighted to age and sex-specific national populations.

Results: Overall, 58% of participants used online retail and delivery platforms. Of these, 36% of participants used online restaurants and 28% used online supermarkets, in the previous 30 days. The odds of using online restaurants was significantly higher for men (OR: 1.23, 95% CI: 1.14 -1.33) and participants aged 18-29 (compared to those 60 years or older)(OR: 6.10, 95% CI: 5.34 – 7.00). Participants aged 18-29 also had the highest odds of using online convenience stores (OR: 7.51, 95% CI: 5.71 – 9.88). Participants living with primary school aged children had higher odds of using online supermarkets compared to those without children (OR: 2.56, 95% CI: 2.22 – 2.94).

Conclusions: A substantial proportion of people are buying food online. Efforts to improve population diets need to ensure that online food retail platforms support good health and nutrition.

Frequent consumption of foods and beverages high in added sugars, saturated fats and sodium is a leading risk factor for weight gain and the development of diet-related non-communicable diseases such as type 2 diabetes, cardiovascular disease, and some cancers (1). Globally, the prevalence of overweight and obesity has risen dramatically since the 1970s, particularly in high income countries such as Australia, the United Kingdom (UK), the United States of America (USA), and Canada (2). Nations undergoing a transition towards diets that are increasingly dominated by unhealthy foods and beverages, such as Mexico, also experience a similar high prevalence of overweight and obesity, contributing to growing health concerns (3)

Online food retail, including websites and smart phone applications (apps), provide customers with easy access to food and beverages for either delivery or pick-up. Online food retail has been conceptualised into three types: online grocery retailers, online meal kit retailers, and online food delivery (OFD) platforms (4). Online grocery retail is where customers can order groceries from supermarkets for either “click and collect” or delivery (5). This category includes supermarkets that have physical stores, as well as “online only” stores, such as Amazon and smaller independent stores catering to specific dietary needs. Online meal kit retailers offer a subscription service for delivery of pre-measured quantities of ingredients along with recipes for meal preparation (6). OFD platforms such as Uber Eats, DoorDash, Just Eat, and SkipTheDishes, allow customers to order ready-to-eat meals from places like restaurants, cafes, and convenience stores (7). Online food retail platforms are designed to be user friendly, featuring intuitive interfaces to facilitate transactions, and are becoming increasingly popular (8). Globally, sales from online grocery have increased from US$110 billion in 2017, to US$390 billion in 2023 (9). Meal kit subscription services made US$12.47 billion in revenue worldwide in 2022, with this predicted to increase to US$18.47 billion by 2027 (10). In 2023 the global revenue of OFD platforms was US$390 billion(9). The number of users of online food retail is also increasing rapidly, with OFD platform users rising from 806 million in 2018 to 1.9 billion in 2023 (11), and the combined number of users of online grocery and meal kit subscription services rising from 547 million in 2018 to 1.3 billion in 2023 (11).

Given their widespread use, online food retail is likely to have significant implications for diets and health. OFD platforms have been criticised for promoting unhealthy food more frequently than healthy options, with most of the food outlets that sell through them considered unhealthy (12, 13). Other studies examining online grocery retail have reported potential dietary health risks from increasing shifts away from purchases at physical stores, such as a reluctance to order fresh or perishable products (such as meat, and fruit and vegetables) online (14, 15). However, online grocery retail may offer some health benefits related to increased access to healthy food for people with limited transport (16), and reducing purchases of unhealthy food online (17). Online meal kit subscriptions have been identified as potentially health promoting because they may increase cooking knowledge and vegetable consumption, although equity concerns have been raised due to the relatively high cost of meals offered (6).

Despite the popularity of online food retail, and their potential implications for population health, there is yet to be a comprehensive analysis of how the different online platforms are used, and how this differs across sociodemographic and behavioural characteristics. In this study, we aimed to:

1. Quantify the prevalence of online food retail platform use (OFD platforms, online grocery retailers, and online meal kit retailers) across Australia, Canada, Mexico, the United Kingdom, and the United States.
2. Examine how online food retail use varies according to sociodemographic and behavioural factors.

Study design and data collection

We used cross sectional data collected annually as part of the International Food Policy Survey (IFPS), in Australia, Canada, Mexico, the UK, and USA in 2022. The IFPS data collection methods have been previously described (18). After being screened for eligibility, and providing consent, adults aged 18 years and older completed an online survey. Participants were recruited through the Nielsen Consumer Insights Global Panel, and their partner panels, in selected countries. Participants were compensated for their time with point based or monetary rewards. The IFPS study received ethics clearance from a University of Waterloo Research Ethics Committee (ORE #30829).

Measures

Meals purchased away from home

All participants were asked about their meals prepared at home, and meals prepared away from home in the previous 7 days. For foods prepared away from home, participants were given a list of options to choose where the food was purchased from, including restaurants, fast food or takeaways, vending machines, etc. Similarly, for food prepared at home, participants were asked where they purchased the food from: instore, online, purchased by someone outside their home, a specialty store (e.g. Asian grocery store), grown by them or someone they know, or from a food bank.

Online food delivery platform use

Use of OFD platforms was assessed through questions that queried if participants had ordered meals/food/drinks from a restaurant or takeaway store, or snacks/food/drinks from a convenience store, online or using an app in the previous 30 days. Participants who responded affirmatively to these questions were then asked how often they had ordered from a restaurant or takeaway, and/or convenience store online or using an app in the previous 30 days, respectively, with the response options including “Less than once a week”, “Once a week”, “A few times a week”, “Every day”, “Don’t know”, and “Refuse to answer”. Participants who answered “Less than once a week”, “Once a week”, “A few times a week”, or “Every day” were considered users of online restaurants, or online convenience stores, respectively. Non-users were classified as those who responded that they had not ordered meals, food, or drinks from a restaurant or take-away online or using an app in the previous 30 days.

Online grocery retail use

To assess use of online grocery retail, participants were asked whether they had ordered groceries online or using an app from a supermarket or groceries from an online-only store (e.g. Amazon) in the previous 30 days. Participants who responded affirmatively were then asked how often they ordered in the past 30 days. Those who responded “Less than once a week”, “Once a week”, “A few times a week” or “Every day” were considered online grocery customers. Non-users were classified as those who responded that they had not ordered groceries from an online supermarket or online-only store, respectively.

Online meal kit delivery service use

Participants were asked if any of their food prepared at home included food purchased in the past seven days from a meal kit delivery service (e.g. Hello Fresh) (yes/no), and were classified accordingly as users and non-users.

Sociodemographic, behavioural, and body weight variables

We included participant sex at birth, age, presence of children in the home aged less than 18 years, perceived income adequacy, cooking skills, nutrition knowledge, frequency of food preparation at home (by asking “How often do you prepare a main meal for yourself or others?”), ethnicity, and self-reported body mass index (BMI) as independent variables.

Age of participants was classified according to generation, with those aged over 60 years considered “Silent Generation and Baby Boomers”, those aged 45–59 “Generation X”, those aged 30–44 “Millennials”, and those aged 18–29 “Generation Z”(19). Children in the home were classified according to the age of the youngest child. Participants whose youngest child was aged 4 years or younger were considered to have “preschool” aged children at home, those with youngest children aged 5 to 12 years were considered to have “primary school” aged children, and those with youngest children aged 13 to 18 years were considered “secondary school” aged.

Perceived income adequacy was assessed based on responses to a question asking how difficult or easy it was for their total monthly income to allow them to make ends meet. Those who reported that it was “difficult” or “very difficult” were considered to have “inadequate income”. Those who responded “very easy”, “easy” or “neither easy nor difficult” were considered to have “adequate income”.

Nutrition knowledge was dichotomised as knowledgeable (those reporting “extremely knowledgeable”, “very knowledgeable” or “somewhat knowledgeable”) or not knowledgeable (“a little knowledgeable” or “not at all knowledgeable”). Cooking skills were defined as “skilled” (those reporting “excellent”, “very good” or “good”) or “unskilled” (“fair” or “poor”). Participants who prepared food at home more than 3–4 times per week were classified as “frequent” with those who prepared food at home two or less times per week classified as “not frequent”. Ethnicity was self-reported using national country-specific measures, with categorisation as ethnic minority or majority for comparison across countries. BMI (kg/m²) was calculated from self-reported weight and height, and participant responses were classified as “obesity (BMI > 30kg/m²)”, “overweight (BMI 25.5-29kg/m²)” and “underweight/normal BMI (BMI < 25kg/m²)” and included a “missing” category for survey participants who did not disclose their weight (20).

Study sample

A total of 35 214 participants completed the full IFPS online survey. Respondents were excluded for the following reasons: ineligible region, invalid response to data quality question; below minimum survey completion time based on median survey time; and/or invalid responses to at least three of 20 open-ended measures (n = 8,941). Among the remaining 26 273 participants, n = 5289 were part of an oversample of participants with lower educational attainment from Mexico and Mexican Americans from the United States and were not part of our analysis (18). Additional participants were excluded from the sample because their responses included inappropriate language (n = 5), if they responded “don’t know” or “refuse to answer” to key variables of interest including those relating to use of online food retail platforms (n = 284) and sociodemographic or behavioural questions (n = 818), leaving a final analytic sample of 19 877 participants.

Statistical analysis

Descriptive statistics were used to summarise the prevalence of online food retail platform use by each country. Data were weighted with post-stratification sample weights, constructed using a raking algorithm with population estimates in each country, based on age group, sex at birth, region, ethnicity (except in Canada due to inconsistent collection methods used in national censuses) and education (except in Mexico due to difficulties recruiting low education participants) to improve the population representativeness of the results.

Logistic regression models were fitted to assess the association between sociodemographic, behavioural and body weight variables above and online food retail platform use for all countries combined. Separate logistic regression models were used for each of the different online food retail platforms, and sample weights were applied to all analyses. Odds ratios and 95% confidence intervals were used to determine effect size and variance. Significance was determined at p < 0.05. All analyses were conducted using Stata SE 17 software.

Overall, 51% of the sample was female, 32% of participants had children living at home aged under 18 years, and 27% of the sample belonged to the Millennial generation (aged 30–44). Mexico had the highest prevalence of children aged under 18 years living at home (50%), and the highest number of Millennial participants (31%). Canada had the lowest prevalence of children aged under 18 living at home (21%) and the UK had the lowest proportion of Millennial participants (25%).

Prevalence of online food retail platform use

Overall, 58% of participants reporting using at least one online food retail platform (Table 1). Restaurants and supermarkets were the most common source of food that was ordered online or using an app (36% and 28% of participants reported ordering food online from restaurants and supermarkets in the past 30 days, respectively). Overall, 11% indicated they had ordered groceries from an online-only store in the previous 30 days. In total, 14% of participants had consumed food purchased from meal kit delivery services in the past 7 days.

Table 1

weighted prevalence of using online food retail platforms and other sociodemographic characteristics, by country
Variables		Country
Variables		Australia (n = 4003)	Canada (n = 4174)	Mexico (n = 3979)	United Kingdom (n = 3990)	United States (n = 3731)	TOTAL (n = 19 877)
Prevalence of use of online ordering by food retail source (%)	Overall use of any retail or delivery platform	55.43	45.25	71.58	56.61	62.13	58.03
	Meals/food/drinks ordered online from restaurants (including through online food delivery platforms)	34.66	28.76	52.31	30.73	34.36	36.12
	Snacks/food/drinks ordered online from convenience stores (including through online food delivery platforms)	9.50	5.90	20.99	7.89	12.54	11.30
	Groceries ordered online from supermarkets with physical stores	28.26	18.72	28.18	31.90	33.10	27.90
	Groceries from online-only stores	9.20	6.90	15.08	10.73	16.00	11.49
	Meal kit delivery services	13.41	9.21	16.58	12.30	17.44	13.70
Sociodemographic characteristics (%)	Female	51.07	50.73	51.72	51.70	50.60	51.17
	Millennial generation	27.15	25.18	30.94	24.99	25.03	26.67
	Children under 18 years at home	30.68	21.28	50.03	28.90	28.28	31.80
	Adequate income	71.43	69.30	55.07	69.41	66.68	66.41
	Knowledgeable about nutrition	63.02	65.70	67.55	56.18	68.15	64.08
	Skilled at cooking	75.65	74.15	66.25	70.13	78.46	72.87
	Frequent food preparers	71.32	70.25	63.56	71.46	67.26	68.81
	Minority ethnic group	29.54	23.33	19.11	13.59	36.25	24.21

Across the five countries, the highest prevalence of online ordering from restaurants (52%), and online convenience stores (21%) was in Mexico. The US had the highest prevalence of online orders from supermarkets (33%), online-only stores (16%), and online meal kit delivery services (17%). Participants in Canada had the lowest prevalence of all online food retail platform uses (Table 1).

When comparing the participants’ use of online retail platforms across all countries (Table 2), participants in Canada generally (but not always) had lower odds of using online retail platforms compared to all other countries. The participants living in the USA had increased odds of ordering online from most retailers when compared to all countries, with the exception of online restaurants and online convenience stores in Mexico.

Table 2

Associations between country and prevalence of online food retail platform use
Variable	Categories	Online orders from restaurants, OR (95% CI)	Online orders from convenience stores OR (95% CI)	Groceries ordered online from physical supermarkets OR (95% CI)	Groceries ordered from online only stores, OR (95% CI)	Meal kits delivery OR (95% CI)
Country	Australia	Reference n = 1296	Reference n = 345	Reference n = 1087	Reference n = 360	Reference n = 534
	Canada	0.82 (0.73–0.92) * n = 1188	0.62 (0.49–0.77) * n = 212	0.62 (0.55–0.71) * n = 757	0.87 (0.72–1.06) n = 295	0.66 (0.56–0.80) * n = 337
	Mexico	1.70 (1.51–1.90) * n = 2172	2.13 (1.81–2.50) * n = 881	0.80 (0.71–0.90) * n = 1197	1.54 (1.31–1.82) * n = 657	1.18 (1.02–1.38) * n = 666
	UK	0.91 (0.81–1.01) n = 1256	0.89 (0.74–1.08) n = 315	1.30 (1.16–1.47) * n = 1277	1.43 (1.20–1.69) * n = 463	0.93 (0.80–1.08) n = 473
	USA	0.99 (0.88–1.12) n = 1313	1.31 (1.09–1.56) * n = 476	1.25 (1.11–1.41) * n = 1238	2.00 (1.69–2.36) * n = 632	1.40 (1.21–1.61) * n = 651
	Canada	Reference	Reference	Reference	Reference	Reference
	Mexico	2.07 (1.83–2.34) *	3.45 (2.81–4.25) *	1.28 (1.12–1.47) *	1.77 (1.47–2.13) *	1.78 (1.48–2.13) *
	UK	1.10 (0.98–1.25)	1.45 (1.15–1.82) *	2.10 (1.84–2.39) *	1.64 (1.35–1.98) *	1.40 (1.17–1.68) *
	USA	1.21 (1.07–1.37) (	2.12 (1.71–2.63) *	2.01 (1.77–2.30) *	2.29 (1.91–2.75) *	2.10 (1.75–2.49) *
	Mexico	Reference	Reference	Reference	Reference	Reference
	UK	0.53 (0.48–0.60) *	0.42 (0.35–0.50) *	1.63 (1.45–1.84) *	0.92 (0.79–1.09)	0.79 (0.68–0.92) *
	USA	0.59 (0.52–0.66) *	0.61 (0.52–0.72) *	1.57 (1.39–1.78) *	1.30 (1.11–1.51) *	1.18 (1.01–1.37) *
	UK	Reference	Reference	Reference	Reference	Reference
	USA	1.10 (0.97–1.24)	1.47 (1.22–1.77) *	0.96 (0.85–1.08)	1.40 (1.19–1.64) *	1.50 (1.29–1.74) *

*significant at < 0.05

Associations between sociodemographic variables and online food retail platform use

Sociodemographic characteristics

Logistic regression models among the entire sample revealed that females had significantly lower odds of making online orders from restaurants, convenience stores, and online-only stores, compared to males (Table 3). No significant difference in use of online meal kit delivery and online orders from supermarkets was observed between males and females.

Millennial and Generation Z age groups had significantly greater odds of using all online food retail platforms compared to the Baby Boomer/Silent Generation participants. As an example, for online restaurants, Generation Z had the greatest odds of use (OR: 6.10, 95% CI:5.34–7.00), followed by Millennials (OR: 4.83 95% CI 4.25–5.50) when compared to the Baby Boomer/Silent Generation participants.

Participants with primary school aged children had higher odds of ordering groceries online from supermarkets (OR: 2.56, 95% CI: 2.22–2.94) or online-only stores (OR: 2.56, 95% CI 2.22–2.94) compared to those without children. Participants with adequate income had greater odds of making online orders from restaurants (OR: 1.21, 95% CI: 1.10–1.32), convenience stores (OR: 1.31, 95% CI: 1.15–1.50), supermarkets (OR: 1.20, 95% CI: 1.09–1.31), and online-only stores (OR: 1.50, 95% CI: 1.31–1.71) than those who reported inadequate income.

Participants who reported that they identified as belonging to an ethnic minority group had greater odds of making online orders from restaurants (OR: 1.19, 95% CI: 1.09–1.31), convenience stores (OR: 1.26, 95% CI: 1.11–1.43), supermarkets (OR: 1.17, 95% CI: 1.06–1.28), and online-only stores (OR: 1.24, 95% CI: 1.10–1.39) than those identifying as an ethnic majority (Table 3). Participants with self-reported BMI greater than 25kg/m² had a lower odds of using online-only stores (OR: 0.74, 95% CI:0.65–0.84, and OR: 0.70, 95% CI 0.58–0.83, respectively).

Behavioural characteristics

Participants classified as frequently preparing meals at home had lower odds of making online orders from convenience stores (OR: 0.87, 95% CI: 0.77–0.98), supermarkets (OR: 0.88, 95% CI: 0.81–0.97), and online-only stores (OR: 0.71, 95% CI: 0.63–0.80) compared to those who did not prepare meals at home frequently. Participants who reported they were skilled at cooking had greater odds of ordering food online from restaurants (OR: 1.10, 95% CI: 1.01–1.22), convenience stores (OR: 1.41, 95% CI: 1.22–1.63), supermarkets, (OR: 1.35, 95% CI: 1.23–1.49), and online-only stores (OR: 1.67, 95% CI: 1.44–1.93) than participants who reported they were unskilled.

Table 3

All countries analyses of associations between use of online food retail platforms, and sociodemographic characteristics
Variable	Categories	Online restaurants OR (95% CI)	Online convenience stores OR (95% CI)	Online supermarkets OR (95% CI)	Online only stores OR (95% CI)	Meal kit delivery OR (95% CI)
Sex at birth	Male	Reference n = 3759	Reference n = 1252	Reference n = 2769	Reference n = 1405	Reference n = 1381
Sex at birth	Female	0.81 (0.75–0.88) * n = 3466	0.71 (0.63–0.79) * n = 977	0.94 (0.87–1.02) n = 2787	0.70 (0.63–0.79) * n = 1002	0.91 (0.82-1.00) n = 1280
BMI (kg/m²)	Underweight/ normal weight (BMI < 25 kg/m²)	Reference n = 2919	Reference n = 954	Reference n = 2233	Reference n = 1111	Reference n = 1050
	Overweight (BMI 25–29.9 kg/m²)	0.96 (0.87–1.06) n = 1808	0.91 (0.79–1.05) n = 517	0.93 (0.85–1.03) n = 1330	0.74 (0.65–0.84) * n = 520	1.00 (0.88–1.12) n = 706
	Obesity (BMI > 30 kg/m²)	1.05 (0.94–1.18) n = 1193	0.89 (0.74–1.06) n = 288	1.04 (01.93–1.16) n = 949	0.70 (0.58–0.83) * n = 275	0.95 (0.82–1.10) n = 450
	Missing value	1.04 (0.93–1.17) n = 1305	0.98 (0.84–1.14) n = 470	1.78 (1.05–1.32) * n = 1044	1.07 (0.92–1.23) n = 501	1.02 (0.88–1.19) n = 455
Children at home	No children	Reference n = 3720	Reference n = 897	Reference n = 2882	Reference n = 920	Reference n = 1618
	Preschool aged	1.55 (1.38–1.75) * n = 1229	1.78 (1.52–2.09) * n = 491	1.78 (1.57–2.01) * n = 958	1.91 (1.63–2.23) * n = 470	1.13 (0.97–1.33) n = 369
	Primary school aged	1.87 (1.68–2.09) * n = 1649	1.73 (1.49–2.00) * n = 619	1.78 (1.59–1.99) * n = 1239	2.56 (2.22–2.94) * n = 791	1.13 (0.98–1.31) n = 505
	High school aged	1.66 (1.43–1.93) * n = 627	2.01 (1.64–2.45) * n = 222	1.71 (1.47–1.98) * n = 477	2.23 (1.83–2.71) p = 0 n = 26	0.90 (0.73–0.11) p = 0.33 n = 169
Generation	Silent and Baby Boomer (aged 60+)	Reference n = 695	Reference n = 90	Reference n = 800	Reference n = 171	Reference n = 575
	Generation X (aged 45–59)	2.33 (2.05–2.64) * n = 1635	2.58 (1.95–3.41) * n = 381	1.40 (1.24–1.59) * n = 1294	1.80 (1.44–2.24) * n = 458	1.03 (0.88–1.19) n = 595
	Millennial (aged 30–44)	4.83 (4.25–5.50) * n = 2692	5.53 (4.21–7.26) * n = 940	2.13 (1.87–2.42) * n = 2009	3.41 (2.75–4.21) * n = 1118	1.71 (1.46–1.99) * n = 885
	Generation Z (aged 18–29)	6.10 (5.34–7.00) * n = 2204	7.51 (5.71–9.88) * n = 818	2.24 (1.96–2.56) * n = 1444	3.07 (2.47–3.83) * n = 660	1.45 (1.23–1.72) * n = 606
Food preparation at home	Not frequent	Reference n = 2468	Reference n = 842	Reference n = 1831	Reference n = 936	Reference n = 921
Food preparation at home	Frequent	0.96 (1.01–1.22) n = 4757	0.87 (0.77–0.98) * n = 1387	0.88 (0.81–0.97) * n = 3725	0.71 (0.63–0.80) * n = 1471	0.93 (0.83–1.03) n = 1740
Cooking skills	Unskilled	Reference n = 1763	Reference n = 462	Reference n = 1220	Reference n = 404	Reference n = 732
Cooking skills	Skilled	1.10 (1.01–1.22) * n = 5462	1.41 (1.22–1.63) * n = 1767	1.35 (1.23–1.49) * n = 4336	1.67 (1.44–1.93) * n = 2003	0.96 (0.85–1.08) n = 1929
Nutrition knowledge	Not knowledgeable	Reference n = 1763	Reference n = 462	Reference n = 1220	Reference n = 404	Reference n = 896
Nutrition knowledge	Knowledgeable	1.14 (1.05–1.24) * n = 5462	1.28 (1.13–1.45) * n = 1767	1.24 (1.14–1.34) * n = 4336	1.39 (1.22–1.57) * n = 2003	1.05 (0.94–1.17) n = 1765
Perceived income	Inadequate	Reference n = 1763	Reference n = 462	Reference n = 1220	Reference n = 404	Reference n = 890
Perceived income	Adequate	1.21 (1.10–1.32) * n = 5462	1.31 (1.15–1.50) * n = 1767	1.20 (1.09–1.31) * n = 4336	1.50 (1.31–1.71) * n = 2003	0.92 (0.82–1.03) n = 1771
Ethnicity	Majority	Reference n = 5473	Reference n = 1621	Reference n = 4231	Reference n = 1720	Reference n = 2107
Ethnicity	Minority	1.19 (1.09–1.31) * n = 1752	1.26 (1.11–1.43) * n = 608	1.17 (1.06–1.28) * n = 1325	1.24 (1.10–1.39) * n = 687	1.01 (0.89–1.14) n = 554

* significant p < 0.05

Participants classified as knowledgeable about nutrition had greater odds of making online orders for restaurants (OR: 1.14, 95% CI: 1.05–1.24), convenience stores (OR: 1.28, 95% CI: 1.13–1.45), supermarkets (OR: 1.24, 95% CI: 1.14–1.34), and online-only stores (OR: 1.39, 95% CI: 1.22–1.57) than those classified as not knowledgeable.

To our knowledge, this is the first paper to describe the use of multiple different online food retail platforms across multiple countries. We found that OFD platforms were used most frequently, for online orders from restaurants (36%). This was followed by online grocery orders from supermarkets with physical stores (28%). By country, Mexico had the highest prevalence of online ordering from restaurants (52%) and convenience stores (21%). The US had the highest prevalence of online ordering from supermarkets (33%), online-only stores (16%), and meal kit delivery services (18%). Significant associations were found between use of online food retail platforms and being male, younger age, and having children present in the home. Adequate income was generally associated with greater odds of using all online food retail. Those who frequently prepared meals at home were significantly less likely to use online convenience stores, online supermarkets, and online-only stores.

Our results are similar to other analyses focusing solely on OFD platforms, using 2018 IFPS data, which found that men (OR: 1.50, 95% CI: 1.35–1.66), those who identify as an ethnic minority (OR: 1.57, 95% CI: 1.38–1.78), and those who live with children aged under 18 years (OR: 2.71, 95% CI: 2.44–3.01) had higher odds of using OFDPs (for restaurants and takeaways)(21). Similarly, a study examining adults aged 18–25 in the United States found that more frequent use of OFD platforms was associated with identifying as an ethnic minority, experiencing food insecurity, and having higher perceived social status (22). Similar to our findings, a UK study examining online grocery shopping found that females, more affluent households, and those aged 25–44 were the most likely to shop online for groceries to be delivered (23). However, a previous study of over 34 000 Italian respondents found a positive relationship between purchasing groceries online and being well-educated, female, and experiencing obesity, contrary to our results which found no association between being female or BMI above 30kg/m² and ordering from online supermarkets (Dominici et al., 2021).

Our study found a significant relationship between ordering from online food retail platforms and being a Millennial (30–44 years old) or member of Generation Z (18–29 years old). Both of these generations are characterised by their willingness to try new technology, considering themselves “early adopters”(24, 25). Additionally, these age groups find convenience worth paying more for, compared with older generations (24). Some evidence suggests that Millennials parents are more time-poor than previous generations (26), due in part to the higher likelihood of both adults in a typical household working outside the home. This could mean that options that save time (such as ordering groceries online) are more appealing to them than older generations who are no longer raising young children. Some have suggested that online food retail platforms may also allow for greater opportunities to try new foods, rather than being confined to the products available at retail outlets within a neighbourhood food environment (4). Millennials (27)and Generation Z (28) have been described as more adventurous than previous generations in terms of trying new foods, which could also make online ordering attractive.

Prevalence of use of all online food retail was highest in Mexico, which may be explained, at least in part, by government COVID-19 related policies (29). Earlier in the COVID-19 pandemic, the Mexican government limited the hours that businesses such as restaurants were able to operate and introduced curfews to limit movement and spread of the virus. It is thought that these restrictions encouraged the use of services such as OFD platforms, with existing platforms “thriving” during this period (30). As the data for this study were collected in 2022, it is possible that behaviours adopted earlier in the COVID-19 pandemic were still influencing participants’ food purchasing methods. Further research to explore the high prevalence of online food retail use in Mexico is warranted.

Prevalence of online ordering from supermarkets, online-only stores, and meal kit delivery services were highest in the US. This may be explained by the food environment characteristics in the US, which have been described as often lacking in retailers that sell fresh foods, especially in neighbourhoods that are predominantly non-white or lower socioeconomic position (31). Associations have been found between characteristics of the US food environment (such as access to fast food restaurants and convenience stores, or lack of access to supermarkets) and obesity (32). Moreover, the online grocery landscape in the US is dominated by large corporations, such as Walmart and Amazon (33). These retailers have vast logistic networks and marketing strategies to facilitate deliveries across wide geographic areas (34), potentially increasing access to groceries for those in areas that are not well served by physical stores. Although operating globally, meal kit brand Hello Fresh reported 66% of its 2023 revenue from North America (35). Meal kits can fulfill a similar niche as online grocery in that they are able to provide fresh ingredients as an alternative to takeaway foods (6). Interestingly, in our study ordering from meal kit delivery services was not found to be associated with participant income, despite their increased cost relative to normal groceries (36). This could be due to the marketing strategies of meal kit companies (advertising benefits including reduced food waste and stress surrounding meal planning) which may appeal to a broad range of customers, regardless of income. Further research is needed to better understand the factors that drive meal kit usage across different income groups, particularly to understand how marketing may influence consumer perceptions.

Results from our study suggest that users of online food retail platforms were more likely to perceive that they have good cooking skills, and are knowledgeable about nutrition, but they typically prepared food at home less than 2 times per week. This may be related to potential implications of online food retail platforms on diet and health as home cooked meals are beneficial to health, shown in a UK study examining adults (n = 11396) who ate home cooked meals at least five times per week and found that they had a greater consumption of fruits and vegetables, and were more likely to have a normal range BMI (less than 25kg/m²) and body fat percentage, compared to those who prepared meals three times per week or less (37). A cross sectional study of US adults (n = 12842) found that those who ate meals that were entirely home cooked had a 26% lower odds of obesity than those who ate some or no home cooked meals (OR 0.74 95% CI = 0.62–0.88)(38). However, the determinants of preparing food at home have been theorised as being more complex than simply having the required cooking skills (39). Adults and families are more time-poor now than previously, which reduces available time for food preparation and may see them relying on prepared foods (40), such as those available through online food retail platforms. Generally, females are responsible for food purchasing and preparation (39), particularly in families (41). Our findings show that males are more likely than females to use online food retail platforms to order from restaurants, convenience stores, and online-only stores.

Policy and research implications

Given that diet is a leading risk factor for chronic disease (1), and the relatively high prevalence of online food retail platform use, efforts to improve population diets need to ensure that online food retail platforms support good health and nutrition. For example, providing nutritional information in online settings that are equivalent to physical food retail stores can allow customers to make an informed choice about their purchases (42). Ensuring that platforms are not promoting unhealthy options through algorithmic boosting, or using pricing strategies can make healthy options more visible and appealing may also be useful health promoting strategies. More detailed research understanding how families are using online food retail platforms to meet their needs could be explored in future research, as it is important to ensure that unhealthy food purchasing behaviours are not being normalised through use of these platforms.

Policies that aim to improve the healthiness of online food retail platforms may also benefit people belonging to ethnic minority groups, as we found they have a higher-odds of making online orders from restaurants, convenience stores, supermarkets, and online-only stores, compared to ethnic majority groups. Ethnic minority groups are often priority populations for public health, as they can experience poorer nutrition outcomes, have a greater prevalence of obesity, and are more likely to live in neighbourhoods with lower availability of healthy food (43, 44). Given that online food retail platforms have been shown to promote unhealthy food options (12), and reduce the desire to purchase fresh items like fruit due to quality concerns (15), their potential role in perpetuating or exacerbating dietary inequalities is of concern. Identifying factors that influence people from ethnic minority groups to use online food retail platforms, and their perceptions as to how to improve the healthiness of offerings available could be studied in further research.

Strengths and limitations

This study has several strengths, including the use of data from multiple countries which allows for comparisons between population groups to determine if a unique social or political circumstance has caused greater demand for online food retail platforms in one country. Additionally, the sample size is large, and our analysis used weighted samples to ensure the results most closely represent those of each country’s general population.

However, this study is also subject to limitations. Some measures were self-reported and relied on participants’ perceptions. These data are not objectively measured and may be influenced by social desirability bias, though this may be reduced somewhat by the online nature of the IFPS. Finally, the survey is conducted online so the sample population is likely familiar with digital technology and may represent a group that is more technologically savvy, which may affect their likelihood of previously using online ordering to purchase food. Future studies using non-online data collection methods would be of interest as a comparator.

The overall prevalence of online ordering from restaurants and supermarkets use was substantial across the five countries surveyed. Associations were found between use of online food retail and sex at birth, age, presence of children in the home, self-reported cooking skills and nutritional knowledge, frequency of food prepared at home, and identifying as being from an ethnic minority. Efforts to improve population diets need to ensure that online food retail platforms support good health and nutrition.

Conflict of interest:

We have no conflict of interest to declare.

Popkin BM, Ng SW. The nutrition transition to a stage of high obesity and noncommunicable disease prevalence dominated by ultra-processed foods is not inevitable. Obesity Reviews. 2022;23(1):e13366.
Campos-Nonato I, Galván-Valencia Ó, Hernández-Barrera L, Oviedo-Solís C, Barquera S. Prevalence of obesity and associated risk factors in Mexican adults: results of Ensanut 2022. Salud Publica Mex. 2023;65.
Bennett R, Keeble M, Zorbas C, Sacks G, Driessen C, Grigsby-Duffy L, et al. The potential influence of the digital food retail environment on health: A systematic scoping review of the literature. Obesity Reviews. 2023;n/a(n/a):e13671.
Brand C, Schwanen T, Anable J. ‘Online Omnivores’ or ‘Willing but struggling’? Identifying online grocery shopping behavior segments using attitude theory. Journal of Retailing and Consumer Services. 2020;57:102195.
Fraser K, Love P, Laws R, Campbell KJ, Spence A. Meal kit subscription services and opportunities to improve family vegetable consumption. Health Promotion International. 2023;38(6):daad155.
Bates S, Reeve B, Trevena H. A narrative review of online food delivery in Australia: challenges and opportunities for public health nutrition policy. Public health nutrition. 2020:1–11.
Gupta A, Sacks G, Cameron AJ, Huggins CE, Peeters A, Backholer K, et al. Use of online food delivery services among adults in five countries from the International Food Policy Study 2018–2021. Prev Med Rep. 2024;43:102766.
Statista. Revenues of the online food delivery market worldwide from 2017 to 2028, by segment 2023 [Available from: https://www.statista.com/statistics/1170631/online-food-delivery-market-size-worldwide/.
Statista. Revenue of the meal kit delivery industry worldwide 2018–2028 2023 [Available from: https://www-statista-com.ezproxy-f.deakin.edu.au/forecasts/1346146/global-revenue-in-the-meal-kits-market.
Statista. Number of users of the online food delivery market worldwide from 2018 to 2028, by segment 2024 [Available from: https://www.statista.com/forecasts/891088/online-food-delivery-users-by-segment-worldwide.
Partridge SR, Gibson AA, Roy R, Malloy JA, Raeside R, Jia SS, et al. Junk Food on Demand: A Cross-Sectional Analysis of the Nutritional Quality of Popular Online Food Delivery Outlets in Australia and New Zealand. Nutrients. 2020;12(10):3107.
Wang C, Korai A, Jia SS, Allman-Farinelli M, Chan V, Roy R, et al. Hunger for Home Delivery: Cross-Sectional Analysis of the Nutritional Quality of Complete Menus on an Online Food Delivery Platform in Australia. 2021;13(3):905.
Jilcott Pitts SB, Ng SW, Blitstein JL, Gustafson A, Kelley CJ, Pandya S, et al. Perceived Advantages and Disadvantages of Online Grocery Shopping among Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) Participants in Eastern North Carolina. Current Developments in Nutrition. 2020;4(5):nzaa076.
Bennett R, Driessen C, Zorbas C, Sacks G, Gupta A, Cameron A, et al. “Healthier options tend to get lost in the noise of online” - Australian shoppers’ experiences with online grocery platforms. Public Health Nutrition. 2024:1–29.
Dillahunt TR, Simioni S, Xu X. Online Grocery Delivery Services: An Opportunity to Address Food Disparities in Transportation-scarce Areas. Proceedings of the 2019 CHI Conference on Human Factors in Computing Systems; Glasgow, Scotland Uk: Association for Computing Machinery; 2019. p. Paper 649.
Zatz LY, Moran AJ, Franckle RL, Block JP, Hou T, Blue D, et al. Comparing Online and In-Store Grocery Purchases. Journal of Nutrition Education and Behavior. 2021;53(6):471–9.
Hammond D, White C, Rynard V. INTERNATIONAL FOOD POLICY STUDY: TECHNICAL REPORT – 2022 SURVEY (WAVE 6). University of Waterloo; 2024.
University of Southern California. Demographics - Age groups 2024 [Available from: https://libguides.usc.edu/busdem/age#:~:text=The%20Baby%20Boomer%20Generation%20%E2%80%93%20born,Z%20%E2%80%93%20born%201995%2D2012.
Van de Mortel TF. Faking it: social desirability response bias in self-report research. Australian Journal of Advanced Nursing, The. 2008;25(4):40–8.
Keeble M, Adams J, Sacks G, Vanderlee L, White CM, Hammond D, et al. Use of Online Food Delivery Services to Order Food Prepared Away-From-Home and Associated Sociodemographic Characteristics: A Cross-Sectional, Multi-Country Analysis. 2020;17(14):5190.
Buettner SA, Pasch KE, Poulos NS. Factors Associated with Food Delivery App use Among Young Adults. Journal of Community Health. 2023;48(5):840–6.
Hood N, Urquhart R, Newing A, Heppenstall A. Sociodemographic and spatial disaggregation of e-commerce channel use in the grocery market in Great Britain. Journal of Retailing and Consumer Services. 2020;55:102076.
Lantos G. Marketing to Millennials: Reach the Largest and Most Influential Generation of Consumers Ever. Journal of Consumer Marketing. 2014;31(5):401–3.
Priporas C-V, Stylos N, Fotiadis AK. Generation Z consumers' expectations of interactions in smart retailing: A future agenda. Computers in Human Behavior. 2017;77:374–81.
Moore A. They've never taken a swim and thought about jaws: Understanding the millennial generation. College and University. 2007;82(4):41.
Kraus S, Sen S, Savitskie K, Kumar SK, Brooks J. Attracting the millennial customer: the case of food trucks. British Food Journal. 2022;124(13):165–82.
Kılıç B, Bekar A, Yozukmaz N. The new foodie generation: gen Z. Generation Z Marketing and Management in Tourism and Hospitality: The Future of the Industry: Springer; 2021. p. 223–47.
Acton RB, Vanderlee L, Cameron AJ, Goodman S, Jáuregui A, Sacks G, et al. Self-Reported Impacts of the COVID-19 Pandemic on Diet-Related Behaviors and Food Security in 5 Countries: Results from the International Food Policy Study 2020. J Nutr. 2022;152(Suppl 1):35s-46s.
United States Department of Agriculture. Mexico: Mexican E-Commerce Surges and It Is Here to Stay. 2021.
Powell LM, Slater S, Mirtcheva D, Bao Y, Chaloupka FJ. Food store availability and neighborhood characteristics in the United States. Preventive Medicine. 2007;44(3):189–95.
Gamba RJ, Schuchter J, Rutt C, Seto EYW. Measuring the Food Environment and its Effects on Obesity in the United States: A Systematic Review of Methods and Results. Journal of Community Health. 2015;40(3):464–75.
Statista. Online food grocery market share in the United States in 2020 and 2021, by company 2021 [Available from: https://www-statista-com.ezproxy-b.deakin.edu.au/statistics/1147672/coronavirus-online-food-grocery-market-share-by-company-us/.
Cleary S, McLarney C. How Logistics Is Evolving: Why It Is Appropriate for Today and the Next Decade. IUP Journal of Supply Chain Management. 2021;18(1):7–24.
Statista. Revenue of HelloFresh from 2016 to 2023, by region 2023 [Available from: https://www-statista-com.ezproxy-b.deakin.edu.au/statistics/948073/hellofresh-net-revenue-by-region/.
McKay FH. What’s in a commercial meal kit? Structured review of Australian meal kits. Public Health Nutrition. 2023;26(6):1284–92.
Mills S, Brown H, Wrieden W, White M, Adams J. Frequency of eating home cooked meals and potential benefits for diet and health: cross-sectional analysis of a population-based cohort study. International Journal of Behavioral Nutrition and Physical Activity. 2017;14(1):109.
Tumin R, Anderson SE. Television, Home-Cooked Meals, and Family Meal Frequency: Associations with Adult Obesity. Journal of the Academy of Nutrition and Dietetics. 2017;117(6):937–45.
Mills S, White M, Brown H, Wrieden W, Kwasnicka D, Halligan J, et al. Health and social determinants and outcomes of home cooking: A systematic review of observational studies. Appetite. 2017;111:116–34.
Jabs J, Devine CM. Time scarcity and food choices: An overview. Appetite. 2006;47(2):196–204.
Power K. The COVID-19 pandemic has increased the care burden of women and families. Sustainability: Science, Practice and Policy. 2020;16(1):67–73.
Sacks G, Vanderlee L. Recommended nutrition-related practices for online food delivery companies. Public Health Nutrition. 2023;26(12):3343–8.
Odoms-Young AM, Zenk SN, Karpyn A, Ayala GX, Gittelsohn J. Obesity and the Food Environment Among Minority Groups. Current Obesity Reports. 2012;1(3):141–51.
Brown TH, O'Rand AM, Adkins DE. Race-Ethnicity and Health Trajectories: Tests of Three Hypotheses across Multiple Groups and Health Outcomes. Journal of Health and Social Behavior. 2012;53(3):359–77.

There is NO conflict of interest to disclose

Prevalence of online food delivery platforms, meal kit delivery, and online grocery use in five countries: an analysis of the 2022 International Food Policy Study

Status:

Version 1

Abstract

Background

Methods