Out of Home Food Intake and Behaviours Among Adults Living in Scotland, 2025

The project had two phases. In Phase 1, participants completed a web questionnaire (Annexe 1) followed by three 24-hour dietary recalls completed within a two-week period on randomly assigned days. In Phase 2, a subset of Phase 1 participants completed a short pre-task questionnaire (Annexe 2) on their most recent OOH purchase followed by an in-depth interview.

Participants were recruited from the Taylor McKenzie Research Panel (TM Panel) by sending an email introducing the survey, emphasising the importance of participating, describing what they will be asked to do and the incentive, as well as a link to the Participant Information Sheet (Annexe 3). The target sample size was 1,000 total participants, with specific quotas set for age groups, sex, and Scottish Index of Multiple Deprivation (SIMD) quintiles. Age quotas were: under 35 years, ~40%, 35–54 years, ~40%, 55 years and over, ~20%. Equal representation was sought across all SIMD quintiles. 

The target sample size for in-depth qualitative interviews in Phase 2 was 35 participants. Similar to Phase 1, specific quotas were applied to ensure coverage across OOH consumption levels (approximately 20 high, 10 moderate, and 5 low consumers), sex (around 50% male and 50% female), age groups (around 40% under 35 years, 40% aged 35–54 years, and 20% aged 55 years and older), and deprivation (at least 20% from both the most deprived and least deprived quintiles).

A pilot study was conducted in order to:

• Ensure all study documents (e.g., invitation letter, Participant Information Sheets, web questionnaire, and interview schedules) were easy to understand.
• Refine instructions to ensure they were understood in the same way by all participants.
• Ensure the survey and the online dietary intake tool (Intake24) were working in the same way across all platform types (e.g., smartphone versus computer versus tablet).
• The interview guide was pilot tested to ensure a smooth flow of discussion and that the questions and language were clear and understandable to participants.

Information from the pilot study was used to update study documents, the survey and Intake24. Changes included, for example:

• Amendments to the introduction of the survey, including adding a definition of OOH food and giving a list of the different options (e.g., food “on the go” from a supermarket, quick service restaurant, sit-down restaurant, etc.).
• A reordering of survey questions to improve flow.
• Added a clear summary of what is meant by ‘OOH food' at the start of the interview to ensure consistent understanding at each phase of the survey. 

Data collection occurred from September to November 2025. 99% of participants who completed the web questionnaire also completed dietary recalls. 

Participants received a unique URL in their email which led them to complete the web questionnaire, after which they were directed to complete their first dietary recall using Intake24. 

All participants self-administered a brief (<10 minutes) web questionnaire (Annexe 1). Participants were permitted to skip any question they did not feel comfortable answering.

Dietary recalls were conducted between 24th September and 18th November 2025. Three 24-hour dietary recalls were collected within a two-week period (97% within a one-week period) using Intake24 (https://intake24.co.uk/). Intake24 was developed by Newcastle University, originally with funding from FSS and is licenced under the Open Government Licence. The version of Intake24 used for this survey was provided and adapted by the University of Cambridge, based on the original, with technical advisory input from Newcastle University. 

At the start of Intake24, participants were prompted to watch a brief video (<5 minutes) explaining how to complete their recall. Intake24 is based on the multiple pass method, which guides participants through the previous 24 hours, asking them to recall all food- and drink-consumption occasions. Participants received an invitation to complete their second recall the day after completing their first recall. This invitation was sent via email. The same procedure was followed for the third recall. Participants were able to unsubscribe from receiving messages from the project team at any time by clicking an ‘unsubscribe’ link in the email or responding ‘STOP’ to the text message. 

For each eating occasion, participants were asked where they had bought or obtained most of the food for that occasion. Options included: 

• Household grocery shopping (mine or someone else’s),
• Eating out, takeaways (including Deliveroo, Just Eat, etc) and food “on the go” (including sandwich outlets and supermarket meal deals), and
• Don’t know. 

For each of the above three options, there were branching questions. The branching question for the ‘Eating out, takeaways and food on the go’ option was:

• Fast food/takeaway outlet, café, coffee or sandwich shop (including order from counter, drive thru, online, delivery apps, phone). Includes e.g. McDonalds, fish and chip shop etc, Costa, Pret, Greggs, Subway
• Sit down restaurant or pub e.g. Pizza Express, Wetherspoons
• Supermarket, convenience stores, corner shops
• School, nursery, college, work canteen

For each of these branching options (except ‘School, nursery, college, work canteen’), participants could choose which specific food or retail outlet they purchased from as listed below. These options were chosen based on previously published Worldpanel by Numerator data for Scotland (1).

What was the name of the [Fast food/takeaway outlet, café, coffee or sandwich shop]?

1. McDonald’s

2. KFC

3. Burger King

4. Domino’s Pizza

5. Papa John’s

6. Independent fish and chips shop

7. Independent kebab shop

8. Greggs

9. Subway

10. Pret

11. Costa Coffee

12. Caffe Nero

13. Starbucks

14. Department store cafés

15. Independent café

16. Other (specify)

What was the name of the [Sit down restaurant or pub]?

1. Pizza Hut

2. Frankie & Benny’s

3. Nando’s

4. Pizza Express

5. Wagamama

6. Wetherspoons

7. Brewers Fayre

8. Harvester

9. Toby Carvery

10. Beefeater Grill

11. Other (specify)

What was the name of the [Supermarket / convenience store/ corner shop]?

1.     Tesco

2.     Asda

3.     Sainsbury’s

4.     Aldi

5.     Morrisons

6.     Co-op

7.     Lidl

8.     M&S Food

9.     Waitrose

10.  Other (specify)

Participants agreeing to be contacted for the in-depth interview were contacted within one week of completing their dietary recalls to schedule the interview and provided with the Participant Information Sheet (Annexe 4). Interviews were conducted between 10th November and 5th December 2025. They completed a pre-task before the interview designed to capture a ‘typical OOH purchase or visit’. Individuals  purchased and consumed OOH food and non-alcoholic drinks as part of a naturally occurring occasion, defined as an OOH eating occasion that they would have undertaken as normal, independent of and without influence from the project. Where possible, they captured photos and/or videos of the food and/or non-alcoholic drinks purchased, as well as the outlet. Following the purchase, individuals completed a short online survey on their experience. This captured three stages of the journey: before the purchase, during the purchase, and after the purchase. Participants then took part in a remote qualitative interview (Annexe 5), during which they discussed the experience in more depth. Interviews also explored individuals’ broader habits and routines around OOH food. These interviews were conducted on Microsoft Teams. Interviews lasted, on average, 45 minutes (range 30-55 minutes). 

In Phase 1, participants who complete just the web questionnaire received £5. Those who also completed three 24-hour dietary recalls received an additional £15 (£20 total incentive). In Phase 2, participants who complete the in-depth interviews received £50. Consistent with other TM Panel surveys, the incentive was provided as points which participants could use to choose a voucher. 

All study documents were reviewed and received a favourable opinion from the Human Ethical Review Committee at the University of Edinburgh (protocol number: HERC_2025_079_2).

The following cleaning was performed on the dietary recall data:

• Participants who did not complete the required 3 recalls were removed from the analysis (n=48).
• Vitamin and mineral supplements and alcoholic drinks (largely beer, wine and spirits; 573 alcoholic drinks in total were reported by 303 participants) were excluded from the analysis and therefore all values reflect intakes from food and non-alcoholic drinks only.
• 432 reported food items (0.9% of the 47,628 items reported) could not be matched to a food composition code from the UK Nutrient Databank and were excluded from the analysis. Of the reported food items that could not be matched to a food composition code that had the food source reported (n=309), 254 (82%) were household grocery shopping and 55 (18%) were OOH.
• Recalls completed in <2 minutes, with <5 items or energy intake <400 kcal or >4,000 kcal were evaluated and where deemed improbable, the participants were excluded from the analysis (n=4).

Under the variable “food source”, entries lalabelled “Other” (n=2,748) were manually reviewed and reclassified into either “Eating out” or “Household shopping” by one researcher. Within the “Eating out” category, additional cleaning was performed both within subcategories (e.g., standardising entries such as “Other: Greggs” to “Greggs” under the Fast Food subcategory) and across subcategories (e.g., reassigning “Other: McDonald” that had been incorrectly recorded under Sit-down Restaurant to “McDonald’s” under the Fast Food subcategory).

• Dairy-free items were removed from “Milk and milk products” and assigned a new category of “Dairy-free”.

Each food and drink product in Intake24 is linked to a food composition code from the UK Nutrient Databank which provides the nutrient values associated with that food. UK Nutrient Databank extract version 3 (unpublished) underpins the version of Intake24 used for this study. 

To improve the accuracy of nutrient information for OOH food, we obtained nutritional information from publicly available information via online menus. We analysed a total of 6,562 food items reported by participants in Intake24 as sourced from OOH across three outlet categories: (1) fast food/takeaway, (2) supermarkets, and (3) sit-down restaurants or pubs. Within the fast food/takeaway category, the most common sources were independent takeaways (31% of food items sourced from the fast food/takeaway category), followed by others such as Chinese takeaways (19%), Greggs (13%), and McDonald’s (11%). For supermarkets, Tesco was the most frequently reported source (34% of food items sourced from supermarkets), followed by Sainsbury’s (12%), and other supermarkets or corner shops (11%). Among meals sourced from sit-down restaurants or pubs, 87% were classified as “other” and most of these were independent outlets or non-specific outlets such as “hotel” or “golf club”. 

If 5% or more of all food items sourced from OOH outlets were sourced from a particular OOH outlet, we obtained nutritional information from their website. A total of 6,562 items were sourced from OOH outlets, so this threshold corresponded to 328 food items. For example, 427 reported food items were purchased from McDonald’s (6.5% of all OOH food items) and so we obtained nutritional information from the McDonald’s website. Based on this threshold, nutritional data were extracted for Greggs, McDonald’s, and Tesco. As the majority of sit-down restaurant meals were from independent outlets, nutritional information could not be reliably obtained for this category. At Greggs, 257 unique menu items were reported as consumed. Nutritional information was identified and extracted for 121 of these items after removing duplicate entries and non-Greggs items (e.g., water or soft drinks). Where item descriptions were incomplete (e.g., soup without further detail), assumptions were made (e.g., soup was assumed to be Tomato soup and Chicken salad roll was assumed to be a Roast Chicken Salad Roll). At McDonald’s, 372 unique menu items were reported as consumed. Nutritional information was identified and extracted for 169 items after removing duplicate entries and non-McDonald’s items (e.g., soft drinks or juices). Assumptions were similarly applied where item descriptions were incomplete (e.g., “Quarter pounder no cheese” was assumed to be McDonald’s quarter pounder). At Tesco, 232 unique items were reported. Nutritional information was identified and extracted for 45 of these items after removing duplicates and non-Tesco items (e.g., biscuits or milk). Assumptions were similarly applied where item descriptions were incomplete (e.g., “Sushi, prawn” was assumed to be Tesco Spicy Prawn Dragon Roll). 

Nutritional data for Greggs and Tesco were available per 100 g. These values were adjusted to reflect individual consumption by multiplying the per-100 g nutrient values by the reported portion size divided by 100. For McDonald’s, nutritional information was provided per menu item rather than per 100 g. For most items (e.g., burgers and wraps), a standard portion was assumed (nutrition information for one Big Mac was assumed Big Mac in a meal). However, for other items such as chicken nuggets, fries, hash browns, and chicken selects, reported portion weights were used to select the most appropriate portion size from the McDonald’s website. For example, nutritional information on the McDonald’s website is provided for 6, 9, and 20 piece portions and therefore a reported intake of 72 g was assumed to correspond to 6 nuggets, while 180 g was assumed to correspond to 9 nuggets.

Conversion factors for grams of macronutrients to calories and sodium to salt are presented in Table 1.

Table  1. Conversion factors used for energy and nutrients.

Statistical analysis was conducted using R and is largely descriptive (i.e., means, SDs, percentages, etc.). All tables are presented according to frequency of OOH consumption. Differences by OOH consumption frequency were tested using Kruskal-Wallis rank sum tests for continuous variables and Pearson's Chi-squared tests for categorical variables. Multivariable linear regression was used to determine if energy and nutrient intake differed by frequency of OOH consumption after adjusting for sex. Differences in nutrient intake were further adjusted for energy intake. 

All R script files are available on GitHub: https://github.com/Runions2/Out-of-Home/.

For qualitative data analysis, interview recordings were uploaded to Coloop (Genie Technology, Limited, London, UK), a generative AI–assisted analysis tool, which was used to generate verbatim transcripts. A data-merger session involving three TMcK researchers was then conducted to identify similarities and differences in themes across interviews and participant types, resulting in the development of a consolidated set of key themes. Coloop was subsequently used to support thematic analysis across all interviews, improving the efficiency and consistency of the analytical process.