This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in the Journal of Medical Internet Research, is properly cited. The complete bibliographic information, a link to the original publication on https://www.jmir.org/, as well as this copyright and license information must be included.
There is an urgent need to reduce society’s meat consumption to help mitigate climate change and reduce noncommunicable diseases.
This study aimed to investigate changes in meat intake after participation in an online, multicomponent, self-regulation intervention.
We conducted a pre-post observational study among adult meat eaters in the United Kingdom who signed up to a website offering support based on self-regulation theory to reduce meat consumption. The program lasted 9 weeks (including a 1-week baseline phase, a 4-week active intervention phase, and a 4-week maintenance phase), comprising self-monitoring, goal setting, action planning, and health and environmental feedback. Meat intake was estimated during weeks 1, 5, and 9 using a 7-day meat frequency questionnaire. We analyzed the change in mean daily meat intake from baseline to week 5 and week 9 among those reporting data using a hierarchical linear mixed model. We assessed changes in attitudes toward meat consumption by questionnaire and considered the acceptability and feasibility of the intervention.
The baseline cohort consisted of 289 participants, of whom 77 were analyzed at week 5 (26.6% of the baseline sample) and 55 at week 9 (71.4% of the week 5 sample). We observed large reductions in meat intake at 5 and 9 weeks: –57 (95% CI –70 to –43) g/day (
Among people motivated to engage, this online self-regulation program may lead to large reductions in meat intake for more than 2 months, with promising signs of a change in meat-eating identity toward more plant-based diets. This digital behavior change intervention could be offered to complement population-level interventions to support reduction of meat consumption.
Population-level changes in meat consumption are needed to help mitigate climate change and reduce noncommunicable diseases. The livestock sector is a leading contributor to environmental degradation [
Individual-level interventions (targeting our conscious and reflective decision-making processes) can complement interventions at a population level (targeting automatic, nonconscious processes) [
We recently developed an online multicomponent intervention, OPTIMISE (Online Programme to Tackle Individual’s Meat Intake Through Self-regulation), based on self-regulation theory to support individuals in reducing their meat consumption. The intervention guides individuals through a self-regulation process of self-monitoring, goal setting, learning about the health and environmental impact of their meat intake, action planning, and regular reflection. We tested its effectiveness in a randomized controlled trial (RCT) [
This population-based cohort study builds upon our previous RCT and aims to investigate whether this online self-regulation intervention is effective in helping the general population in the United Kingdom who eat at least some meat to reduce their meat intake. A secondary aim was to investigate the adherence to and acceptability of the intervention.
We conducted a cohort study among UK adults using OPTIMISE, an online program to support meat reduction based on self-regulation theory. All aspects of the study were delivered remotely through a website developed specifically for the intervention, through which all data collection took place between May 28, 2021, and December 13, 2021.
This study was granted ethical approval by the Central University Research Ethics Committee of the University of Oxford (R71430/RE003).
We made the website publicly available and signposted it to people over 6 months through public engagement events across the United Kingdom via our research team’s website and social media presence, as well as online newsletters and volunteer databases (eg, Research for the Future) [
People interested in taking part in the research completed a screening questionnaire to assess eligibility (participants were aged 18 years or older, were resident in the United Kingdom, were meat eaters, and wanted to reduce their meat intake), and they provided consent for participation in the study before registering with the OPTIMISE website using their email address. Participants who completed the program were entered into a raffle to win a £100 (US $122.53) digital gift card (1 gift card was available for every 50 participants).
The study lasted 9 weeks (including a baseline week of self-monitoring meat consumption, a 4-week active intervention phase, and a 4-week maintenance phase;
Procedure of the OPTIMISE (Online Programme to Tackle Individual’s Meat Intake Through Self-regulation) study.
The full intervention has been described in detail previously [
The main outcome measure was the change in mean daily meat consumption from baseline to week 5, measured by the daily meat frequency questionnaires [
We also explored the predictors of change in meat intake and change in attitudes toward meat consumption from baseline to weeks 5 and 9. We assessed adherence to the intervention as the proportion of the 42 sessions participants completed and the acceptability of the intervention based on responses to the intervention evaluation questionnaire.
All statistical analyses were conducted in Stata/IC (version 14.1).
For each participant and time point (baseline, week 5, and week 9), we calculated mean total daily intakes of all meat and meat subtypes (ie, red meat and processed meat). The main analysis used a hierarchical linear mixed model with fixed effects for “time point” and random effects for “participant” to investigate whether meat consumption at weeks 5 and 9 differed significantly from baseline. As prespecified in our statistical analysis plan, days in which reported meat intake exceeded 1.5 kg were excluded, as we deemed this implausible. We identified no confounding variables through univariable regressions and so the model was unadjusted.
To analyze the predictors of change in mean daily meat consumption from baseline to week 5, we used a multivariable linear regression model with change in meat consumption as the dependent variable and possible predictors included in one single model. The predictors were age, gender, ethnic group, highest educational qualification, household size, annual household income, the response to “currently trying to lose weight” (yes/no), dietary restrictions, baseline meat consumption, baseline attitudes toward meat (ie, meat-eating identity, including non–meat eater, reduced-meat eater, and meat eater; mean meat-free self-efficacy; meat reduction motivation; mean meat consumption social norms; and meat reduction social support), tertiles of engagement (based on the percentage of sessions participants completed throughout the active intervention phase), and the number of action categories tried at least once.
We used hierarchical linear mixed models to investigate changes in attitudes toward meat (ie, meat-free self-efficacy, meat reduction motivation, meat consumption social norms, and meat reduction social support), between baseline and weeks 5 and 9. Due to multicollinearity between tertiles of engagement and meat-eating identity changes, we used the chi-square goodness-of-fit test to explore the proportions of each meat-eating identity at both follow-ups compared to baseline. Written feedback collected from participants as part of the intervention evaluation questionnaire was analyzed qualitatively using inductive thematic analysis in NVivo 12 (QSR International) [
Sensitivity analyses were performed using 2-tailed independent
To explore barriers to adherence to participants’ chosen meat reduction actions, we analyzed the free-text responses to the daily action completion question when participants indicated they had not managed to perform their action using inductive thematic analysis [
A total of 566 individuals signed up to the study website, 59 of whom requested their account (and subsequently all their data) be deleted before the end of the study. We were unable to establish which of these 59 individuals were study participants and which were independent users. Of the remaining 507 individuals for whom we had data, 120 registered as independent users and 387 registered as study participants.
Of the study participants, 82 did not complete any baseline sessions, 3 did not complete the baseline demographics questionnaire, and 7 reported no meat consumption during the baseline week. Six participants were excluded as they self-reported eating >1.5 kg of meat per day in every meat frequency questionnaire they completed. The total baseline cohort, therefore, consisted of 289 of the 387 registered participants (74.7%). Participants were aged 18 to 84 years (mean 46.8, SD 13.8 years), 72.3% (209/289) were female, and 57.1% (165/289) were White British (
Eleven participants did not complete their goal setting, preselect their actions, or both, and a further 201 participants did not complete any sessions in week 5, leaving 77 participants in our first follow-up sample (week 5; 26.6% of the baseline sample of 289 participants). Of these participants, 22 did not complete any sessions in week 9, leaving 55 participants in our second follow-up sample (week 9; 71.4% of the first follow-up sample of 77 participants).
In the baseline cohort, the most important motivating factor to reduce meat intake on a scale from 1 (not at all important) to 10 (extremely important) was to help the environment (mean score 8.6, SD 1.5), followed by health benefits (mean score 7.9, SD 1.8) and animal welfare concerns (mean score 7.6, SD 2.3). The mean meat-consumption reduction goal shows participants on average challenged themselves to reduce their meat consumption by nearly a quarter (–23%, SD 13%; range 5%-90%).
Participants who dropped out before week 5 were more likely to be trying to lose weight (
Baseline characteristics (N=289).
Characteristics | Values | ||
Agea (years), mean (SD) | 46.8 (13.8) | ||
|
|||
|
Female | 209 (72.3) | |
|
Male | 78 (27) | |
|
Other/prefer not to say | 2 (0.7) | |
|
|||
|
White British | 165 (57.1) | |
|
White other | 84 (29.1) | |
|
Asian or Asian British | 17 (5.9) | |
|
Black or Black British | 4 (1.4) | |
|
Mixed/other | 18 (6.2) | |
|
Prefer not to say | 1 (0.4) | |
|
|||
|
University degree, NVQb level 4-5 or equivalent, and above | 242 (83.7) | |
|
Other post–high school qualifications | 15 (5.2) | |
|
A-levelsc, NVQ level 2-3 or equivalent | 21 (7.3) | |
|
Apprenticeship | 1 (0.4) | |
|
GCSEd, NVQ level 1, or equivalent | 2 (0.7) | |
|
Other vocational, work-related qualifications | 3 (1) | |
|
No formal qualifications | 1 (0.4) | |
|
Prefer not to say | 4 (1.4) | |
|
|||
|
1 person | 55 (19) | |
|
2 people | 115 (39.8) | |
|
3 people | 57 (19.7) | |
|
4 people | 48 (16.6) | |
|
5 people | 10 (3.5) | |
|
≥6 people | 4 (1.4) | |
|
|||
|
<£15,000 (US $18,418) | 10 (3.5) | |
|
£15,000-£24,999 (US $18,418-$30,695) | 24 (8.3) | |
|
£25,000-£39,999 (US $30,695-$49,113) | 45 (15.6) | |
|
£40,000-£75,000 (US $49,113-$92,090) | 99 (34.3) | |
|
>£75,000 (>US $92,090) | 90 (31.1) | |
|
Prefer not to say | 21 (7.3) | |
|
|||
|
Yes | 198 (68.5) | |
|
No | 91 (31.5) | |
|
|||
|
Dairy-free | 14 (4.9) | |
|
Gluten-free | 19 (6.6) | |
|
Fish and shellfish allergy | 3 (1) | |
|
None | 259 (89.6) | |
|
|||
|
Public engagement events | 4 (1.4) | |
|
Research team’s website/social media | 6 (2.1) | |
|
Friends or family members | 17 (5.9) | |
|
Social media | 48 (16.6) | |
|
Radio or newspaper | 181 (62.6) | |
|
Volunteer databases | 9 (3.1) | |
|
Other | 24 (8.3) |
aAge ranged from 18 to 84 years.
bNVQ: National Vocational Qualification.
cAdvanced level (A-level) qualifications are subject-based qualifications for students aged 16 or older.
dGCSE: General Certificate of Secondary Education.
eParticipants could select multiple answers to this question.
Meat consumption and attitudes at baseline and both follow-ups. Estimates are from mixed effects models with fixed effects for “time point” and random effects for “participant.” The models were unadjusted, as we identified no potential confounders in univariate analyses. Data on meat-eating identity are shown in
|
Baseline, mean (SD) | First follow-up (week 5) | Second follow-up (week 9) | ||||||
|
|
Mean (SD) | Mean difference (95% CI) | Mean (SD) | Mean difference (95% CI) | ||||
|
|||||||||
|
Total meat | 146 (162) | 61 (50) | –57 (–70 to –43) | <.001 | 68 (51) | –49 (–64 to –34) | <.001 | |
|
Red meat | 53 (65) | 27 (33) | –22 (–32 to –12) | <.001 | 28 (31) | –21 (–32 to –10) | <.001 | |
|
Processed meat | 40 (80) | 17 (23) | –13 (–19 to –8) | <.001 | 18 (22) | –12 (–18 to –5) | <.001 | |
|
Red and processed meat | 92 (121) | 44 (52) | –35 (–49 to –22) | <.001 | 46 (48) | –33 (–48 to –18) | <.001 | |
|
|||||||||
|
Meat-free self-efficacy scorea | 3.2 (1.2) | 2.8 (1.4) | –0.3 (–0.6 to 0.0) | .09 | 2.4 (1.1) | –0.8 (–1.2 to –0.5) | <.001 | |
|
Meat reduction motivation scoreb | 7.5 (1.6) | 8.1 (1.6) | 0.2 (–0.3 to 0.8) | .45 | 7.7 (2.1) | –0.1 (–0.7 to 0.5) | .72 | |
|
Meat consumption social norms scorec | 4.4 (1.0) | 4.2 (1.2) | –0.2 (–0.4 to 0.0) | .13 | 4.0 (1.1) | –0.4 (–0.6 to –0.1) | .001 | |
|
Meat reduction social support scored | 6.6 (2.5) | 7.0 (2.6) | –0.1 (–0.8 to 0.5) | .66 | 6.3 (2.5) | –0.8 (–1.5 to –0.1) | .02 |
aMean score of 3 self-efficacy questions (“I lack the cooking skills to prepare meat-free meals,” “I don’t know what to eat instead of meat,” and “I don’t have enough willpower to not eat meat”), measured on a scale from 1 (strongly disagree) to 7 (strongly agree).
bParticipants were asked to respond to the question “How motivated are you to reduce your meat intake beyond the context of this programme?” on a scale from 1 (not at all motivated) to 10 (extremely motivated).
cMean score of responses to 4 social norm questions using the 4
dParticipants were asked, “How willing are the people you share your meals with to reduce their meat consumption?” on a scale from 1 (not open at all) to 10 (very open to it).
Flow chart of participants.
Total mean consumption of meat decreased from baseline to week 5 by –57 (95% CI –70 to –43) g/day (
Higher baseline meat consumption was associated with a greater reduction in meat intake at week 5, with every 1 g of greater baseline intake predicting a 0.9 g/day greater reduction (95% CI –1.1 to –0.7;
There was a significant change in reported meat-eating identities toward reduced-meat and non–meat-eating identities from baseline to both follow-ups (
The most commonly reported barriers for not performing meat reduction actions were (1) other people (most frequently friends and family), (2) being too busy and not having enough time, (3) eating out and the lack of meat-free options available or the temptation to opt for a meat dish, and (4) eating meat leftovers and wanting to avoid food waste. Representative quotes are as follows:
I was not cooking yesterday, and when you’re a guest I think it’s polite to eat what’s been served.
I was very exhausted today and didn’t have the energy to make two dishes.
I ate leftover food, my partner had cooked more meat than the children wanted or needed.
More than 7 out of 10 participants dropped out before week 5 (73.4%, 212/289), but thereafter, 71% (55/77) completed the study. Of the participants who completed week 5 and week 9, 78% (60/77) and 98% (54/55) completed at least 80% (34/42) of the sessions, respectively. Fifty-five participants (71%, 55/77) completed the intervention evaluation questionnaire at week 5, rating the usefulness of the intervention components and additional resources on a scale from 1 (not useful) to 10 (very useful). Mean scores ranged from 7.2 (SD 2.7) to 9 (SD 1.7) (
In general I’ve found the study interesting and important. It has been effective to chart and reflect on my meat consumption, plan for change and see my evidence of change progressively.
It has helped me to confirm what my personal stumbling blocks are.
To me, tracking the meat consumption and planning activities was the best way to help me out eating less meat, because I’m a naturally planned person.
While some participants said the daily action planning was helpful, others said they would have preferred weekly actions to make planning meals in advance for the week easier. Some participants said they would have liked both social and competitive elements, allowing them to share their progress with others and compare their intake with other users or the UK average, or both.
Intervention evaluation questionnaire results. Participants were asked how useful they found the items on a scale of 1 (not useful) to 10 (very useful). The additional resources were optional and only evaluated by those who reported using them throughout the study.
Questionnaire items | Mean score (SD) | Respondents, n | |
|
|
|
|
|
Tracking your meat consumption on a daily basis | 8.7 (1.6) | 55 |
|
Feedback on the environmental and health impact of your meat consumption | 7.6 (2.4) | 55 |
|
Planning an action on a daily basis to reduce your meat consumption | 7.2 (2.7) | 55 |
|
|
|
|
|
Weekly action evaluation | 7.6 (2.4) | 55 |
|
Downloadable action diary | 9.0 (1.7) | 3 |
|
Downloadable action overview | 8.0 (2.0) | 3 |
|
Links to other resources | 8.3 (1.5) | 10 |
|
Ability to review your journey | 8.2 (1.9) | 22 |
We observed significant reductions in meat consumption when UK adult meat eaters engaged with a bespoke meat-consumption reduction website and were guided through a process of self-regulation. Participants reported marked changes in meat-eating identity toward reduced-meat and non–meat-eating identities, their meat-free self-efficacy increased, and their perception of meat consumption as the social norm decreased. There was a high dropout rate from registration to first follow-up, but the quarter of participants who provided outcome data had high engagement with the intervention and rated it highly, particularly the self-monitoring aspect.
Strengths of this study were that baseline meat intake was similar to that of the general UK population [
The absolute reduction in meat intake reported here was both large and remarkably similar to the reduction observed in the intervention group in our previous RCT (–58% vs –57% at week 5 and –53% vs –52% at week 9 in the current study and the RCT, respectively) [
Importantly, participants reported an increased meat-free self-efficacy, a marked shift toward reduced meat-eating and non–meat-eating identities, and a reduction in perception of meat consumption as the social norm. This reflects findings from a United Kingdom–based RCT that tested the effectiveness of a multicomponent behavioral intervention to reduce meat consumption [
The results of our two OPTIMISE studies, taken together, suggest this online self-regulation program may be effective for helping motivated individuals to reduce their meat intake and closing the intention-behavior gap. In comparison to in-person interventions, there is preliminary evidence to support the scalability [
An online program to encourage self-monitoring of meat consumption, together with goal setting, educative feedback, action planning, and reflection may help individuals seeking to reduce their meat intake to change their diet and foster a reduced-meat or non–meat-eating identity. This type of support could be offered at scale with minimal cost and could complement other environmental interventions to help people eat less meat.
Health and environmental benefits presented to participants.
Meat consumption attitude questions.
Meat consumption reduction actions.
Example of weekly health and environmental feedback.
Participant characteristics of week 5 drop-outs vs completers.
Meat-eating identity proportions.
Predictors of change in total meat intake.
National Diet and Nutrition Survey
Online Programme to Tackle Individual’s Meat Intake Through Self-regulation
Randomized Controlled Trial
The authors gratefully acknowledge One Ltd, who created the Online Programme to Tackle Individual’s Meat Intake Through Self-regulation (OPTIMISE) website, and we thank Dr Rachel Pechey (University of Oxford) for her statistical advice regarding the analysis of meat-eating identity changes. We would also like to thank all those who helped us test the website throughout its development and provided us with feedback. This research was funded by the Wellcome Trust Our Planet Our Health program (Livestock, Environment and People [LEAP]; 205212/Z/16/Z). SAJ is funded by the National Institute for Health and Care Research (NIHR) Oxford Biomedical Research Centre. SAJ and CP are funded by the Oxford and Thames Valley NIHR Applied Research Centre. The funders of the study were not involved in the study design, conduct of the study, data collection, data analysis, data interpretation, or any aspect pertinent to the study. The funders had no role in the writing of the manuscript or in the decision to submit it for publication.
Data described in the manuscript will be made available upon request pending application and approval.
CS, CP, KF, BC, and SAJ designed the research; CS conducted the research; CS analyzed the data; CS drafted the manuscript; and CS, CP, KF, and SAJ critically reviewed the manuscript. CS had primary responsibility for the final content. All authors read and approved the final manuscript.
None declared.