Cognitive and behavioural strategies employed to overcome “lapses” and prevent “relapse” among weight‐loss maintainers and regainers: A qualitative study

While many behavioural weight management programmes are effective in the short‐term, post‐programme weight regain is common. Overcoming “lapses” and preventing “relapse” has been highlighted as important in weight‐loss maintenance, but little is known on how this is achieved. This study aimed to compare the cognitive and behavioural strategies employed to overcome “lapses” and prevent “relapse” by people who had regained weight or maintained weight‐loss after participating in a weight management programme. By investigating differences between groups, we intended to identify strategies associated with better weight‐loss maintenance. Semi‐structured interviews were conducted with 26 participants (58% female) recruited from the 5‐year follow‐up of the Weight Loss Referrals for Adults in Primary Care (WRAP) trial (evaluation of a commercial weight‐loss programme). Participants who had lost ≥5% baseline weight during the active intervention were purposively sampled according to 5‐year weight trajectories (n = 16 'Regainers', n = 10 'Maintainers'). Interviews were audio‐recorded, transcribed verbatim, and analysed thematically. Key differences in strategies were that Maintainers continued to pay attention to their dietary intake, anticipated and planned for potential lapses in high‐risk situations, and managed impulses using distraction techniques. Regainers did not report making plans, used relaxed dietary monitoring, found distraction techniques to be ineffective and appeared to have difficulty navigating food within interpersonal relationships. This study is one of the longest qualitative follow‐ups of a weight loss trial to date, offering unique insights into long‐term maintenance. Future programmes should emphasize strategies focusing on self‐monitoring, planning and managing interpersonal relationships to help participants successfully maintain weight‐loss in the longer‐term.


Summary
While many behavioural weight management programmes are effective in the short-term, post-programme weight regain is common. Overcoming "lapses" and preventing "relapse" has been highlighted as important in weight-loss maintenance, but little is known on how this is achieved. This study aimed to compare the cognitive and behavioural strategies employed to overcome "lapses" and prevent "relapse" by people who had regained weight or maintained weight-loss after participating in a weight management programme. By investigating differences between groups, we intended to identify strategies associated with better weight-loss maintenance. Semi-structured interviews were conducted with 26 participants (58% female) recruited from the 5-year follow-up of the Weight Loss Referrals for Adults in Primary Care (WRAP) trial (evaluation of a commercial weight-loss programme). Participants who had lost ≥5% baseline weight during the active intervention were purposively sampled according to 5-year weight trajectories (n = 16 'Regainers', n = 10 'Maintainers'). Interviews were audio-recorded, transcribed verbatim, and analysed thematically. Key differences in strategies were that Maintainers continued to pay attention to their dietary intake, anticipated and planned for potential lapses in high-risk situations, and managed impulses using distraction techniques. Regainers did not report making plans, used relaxed dietary monitoring, found distraction techniques to be ineffective and appeared to have difficulty navigating food within interpersonal relationships. This study is one of the longest qualitative follow-ups of a weight loss trial to date, offering unique insights into long-term maintenance. Future programmes should emphasize strategies focusing on self-monitoring, planning and managing interpersonal relationships to help participants successfully maintain weight-loss in the longer-term. A major weakness of previous qualitative studies has been the short-term follow-up; participants have typically been managing their weight for less than 12 months, therefore little is known about strategies employed for long-term weight management. Indeed, many participants were likely to be still aiming for weight-loss rather than maintenance, and may have limited experience of lapses and relapse.
Evidence suggests that implementation of strategies and challenges in weight management change over time. [26][27][28] For example, in a recent study, Pedersen et al 27 explored whether self-regulation of food intake differed between short and long-term weight loss maintainers, finding key differences around planning, shopping/storing and preparing/cooking behaviours. However, this study also defined long-term as >12 months, and used self-reported history to categorize participants' weight trajectories.
To address the limitations of previous research, we have conducted one of the longest qualitative follow-ups of a behavioural weight management programme to date. Further, we have used multiple objective weight measurements to categorize participants by their weight trajectories. Our study aimed to compare the cognitive and behavioural strategies employed to overcome "lapses" and prevent "relapse" by people who had regained or maintained weight-loss 5 years after participating in a weight management programme. By comparing groups, we aimed to identify strategies associated with better long-term weight-loss maintenance in order to inform the What is already known about this subject?
• Behavioural weight-management programmes are effective in the short-term, but weight regain is common • Dealing with lapses and preventing relapse is a particular threat during the weight-loss maintenance phase • Identifying strategies that support successful long-term weight maintenance could inform better treatment but is an under-researched area.
What this study adds?
• Maintainers employed self-regulation techniques, anticipated potential lapses and made plans to compensate for these although few engaged in regular self-weighing.
• Regainers made some efforts to self-regulate their behaviour, but they did not tend to make plans, used relaxed dietary monitoring, found distraction strategies ineffective and struggled with navigating interpersonal relationships in relation to food.
• Monitoring, planning and managing interpersonal relationships appeared to be crucial strategies for overcoming "lapses" and preventing "relapse" after participating in a weight management programme and should be incorporated in future programmes. development or refinement of future weight management programmes.

| Sampling and participants
This qualitative study was embedded within the 5-year follow-up of the Weight Loss Referrals for Adults in Primary Care (WRAP) randomized controlled trial (RCT). 29,30 Detailed information on intervention and control content can be found in previously published articles. 29,30 In brief, the trial recruited adults with a BMI over 28 kg/m 2 and randomized them to either: referral to a commercial open-group behavioural weight management programme (WW, formerly Weight Watchers) for 12 weeks; referral to the same programme for 52 weeks; or the control group. The WW intervention consisted of weekly in-person group meetings, including a weigh-in and a 30-minute interactive education session led by the coach. The education sessions provided advice on diet, physical activity (PA), positive mind-set, using behavioural strategies (eg, goal-setting, self-monitoring, problem-solving, modifying the personal food environment and relapse prevention), and facilitated peer support from coaches and other group members. Participants also received access to WW online tools for the duration of the intervention. The control intervention was a standardized brief intervention: recognition of the problem by the general practitioner (letter of invitation), basic written information on self-help weight loss strategies provided by a member of the research team at baseline and weighing at follow up time points.
Twenty six participants were purposively sampled from the 12 and 52 week behavioural weight management programme arms, using trial data on weight trajectories (weight was objectively measured at 0, 3, 12, 24 and 60 months) and demographic variables to gain a diverse sample. Participants from the 12 and 52 week intervention arms were interviewed as we were specifically interested in the experiences of weight management after losing weight as part of a behavioural weight management programme. We selected participants who had lost ≥5% baseline weight during the first 12 months of the trial, to ensure that they had experience of weight loss and of trying to maintain this loss. We aimed for variation in gender, age, education and income. To facilitate this, we recruited participants from the Cambridge (n = 15) and Liverpool (n = 11) centres of the trial.
Using objectively measured data from their 5-year follow up from baseline, participants were categorized into those who had maintained their weight-loss (+ − 3 kg) or lost more weight (n = 10; 'Maintainers') since the end of the WRAP study intervention, and those who had regained >3 kg of weight since the end of the WRAP study intervention (n = 16; 'Regainers') (end of intervention was either 12 or 52 weeks from baseline, dependant on group allocation). During this time period, participants did not attend the weight management programme offered in the study, unless they decided to continue of their own volition at their own financial cost.
Baseline weight for Maintainers ranged from 65.9 to 159.9 kg, with a mean of 103.6 kg (SD 26.6), and regainers ranged from 77.1 kg to 110.5 kg, with a mean of 94.9 kg (SD 11.9). From the end of the intervention to time of interview, Maintainers had lost 4.1% of their body weight (−3.9 kg) and Regainers had gained 11.3% (+9.8 kg) (Tables 1 and 2).

| Interview schedule
The interview schedule was developed through reviewing relevant literature 7 and consulting with experts (a specialist bariatric general practitioner and academics specializing in behaviour change research from psychology, sociology and medical backgrounds). It was reviewed by members of a specialist multidisciplinary weight management programme to ensure coverage of important topics and appropriate language. Questions focused on key personal, social and environmental challenges to weight-loss maintenance and strategies used in managing lapses and relapse. The semistructured schedule ensured that key topics were addressed in all interviews and allowed for further probing and discussion to be guided by participant responses. The schedule was piloted and revised after three interviews (these interviews were included in the main analysis). Those who responded positively to the invitation were telephoned to arrange an appointment for interview. Non-responders were followed up with a reminder telephone call. All participants approached agreed to be interviewed.

| Data collection
The lead author (ERL) conducted individual face-to-face interviews with participants using a semi-structured interview schedule.
Participants had previously participated in the WRAP study, but had not had any previous contact with ERL. Participants were given a choice of location for the interviews, either within their homes or in a private room at the University of Cambridge or the University of Liverpool. Only the researcher and the participant were present for all except two interviews, where the participant's spouse was also present. Participants provided written consent to participate and for their interview to be digitally audio-recorded. Interviews were then transcribed verbatim. Each interview was anonymised by implementation of a number coding system prior to transcription.
Interviews were conducted between May to September 2018, with interviews lasting between 25 and 87 minutes (mean of 53 minutes).

| Data analysis
Verbatim transcripts were analysed using a thematic approach to provide a detailed and data driven account of participant's experiences. 31 Given the limited knowledge of experiences of weight maintenance beyond 12 months, the aim of the current study was not to test a specific theory, but rather to take an inductive approach that identified points of particular salience in participants' own accounts of their experience. NVivo software (version 12) was used to manage and store data. The first five interviews were coded independently by two authors (ERL and CAH) and then discussed to ensure consistency and appropriateness of categories before continuing analysis of remaining interviews. ERL is a postdoctoral researcher with training and experience in conducting and analysing qualitative interviews on behaviour  change; CAH is a general practitioner and bariatric physician with experience in qualitative research who runs a specialist weight management service. ERL and CAH met frequently to discuss themes, and used a reflective code book to record decisions and to assist with theme refinement. A public representative (GDP) read and provided feedback on the same initial five interviews. They identified key points that they felt to be particularly important, and which resonated with their weight-loss journey. A comparison between the data for Maintainers and Regainers was then conducted, identifying differences in strategies. Throughout the process, findings were discussed with ALA and RD to gain additional insight. Findings were discussed with GDP and with a patient user group panel to assist with interpretation of results.

| Theme 1: Monitoring
Sub-themes identified were: (i) self-monitoring of behaviours, (ii) selfmonitoring of weight and (iii) triggers for action.

| Sub-theme i): Self-monitoring of behaviours
Many participants referred to self-monitoring of dietary behaviours as a key part of their weight-loss maintenance efforts. Maintainers continued to pay close attention to their dietary intake. They reported efforts to balance types and quantity of food throughout the day to facilitate an energy balance, particularly recognizing the evening to be a time when they are prone to lapses. Some Maintainers described instances of utilizing formal strategies to monitor their food intake, such as writing down or inputting foods into a tracking app. A few participants used pre-planned strategies to limit their exposure to tempting foods. One Maintainer tried to avoid certain shopping aisles, whereas a Regainer revealed a lack of foresight by not preparing shopping lists despite noticing an increase in impulse-buying unhealthy foods.
"… I'm that shopper that doesn't make a list… I'm an impulse buyer and it's always then the bad stuff…" (P30298, Regainer, female, aged 49, 52 weeks) No participants reported avoiding any social occasions at which they knew food would be present.

| Sub-theme iv): Flexibility in dietary behaviours
Rather than adhering to prescriptive diets, most participants had flexibility in their dietary behaviours. Swapping carbohydrates, particularly bread, to a perceived lighter or less calorific alternative was common, especially by Maintainers.

| Theme 3: Managing interpersonal relationships
Sub-themes identified were: (i) Capitalizing on social support and (ii) Navigating the role of food.

| Sub-theme i): Capitalizing on social support
Some participants from both groups who were living with family members had changed the whole family's eating behaviours, with everyone eating healthier meals.
A few participants from both groups were dependent on their partners for shopping and food preparation. Generally partners were supportive, providing healthy meals and reducing portion sizes. This also reduced the participants' personal responsibility to make these changes, which they may not have made themselves.
"My wife was good at cutting back on portion size… I mean that the fact that she does the cooking, and can sort of control that without me sort of cheating…" (P10493, Regainer, male, aged 56, 52 weeks) A few participants spoke favourably of medically-referred exercise classes they had attended, as members were of a similar ability and had the same aim.

| DISCUSSION
The aim of this study was to compare the cognitive and behavioural strategies employed to overcome "lapses" and prevent "relapse" by people who had regained or maintained weight-loss 5 years after participating in a weight-management programme. Maintainers employed self-regulation techniques, anticipated potential lapses and made plans to compensate for these. However, few Maintainers reported self-weighing and this varied in frequency. Regainers also made some efforts to self-regulate their behaviour, but they did not tend to make plans, used relaxed dietary monitoring and struggled with navigating interpersonal relationships in relation to food. For both groups, PA appeared to have limited salience for managing lapses, there was no evidence of avoidance of 'high-risk' social situations and distraction strategies to overcome impulses were implemented by both groups but Regainers found them ineffective.
Our findings echo many elements of Greaves et al review, 7 in particular, recognizing self-regulation and managing internal and external influences as important themes for Maintainers. However, the idea of "Regainers not self-regulating" and that "Regainers do not manage influences" is not entirely supported here. Rather, our findings found many Regainers implemented some relaxed dietary monitoring and made some healthier swaps. Regainers also employed some distraction techniques to manage impulses but either used them inconsistently or found them ineffective. More salient was that they experienced problems around interpersonal relationships and food.
Potential reasons for differences between our findings and those of Greaves et al 7 are that only a few studies in their review made a comparison between weight trajectory groups, reducing the strength of evidence. Further, as previous studies had a shorter-term follow-up, potentially their participants had a more vivid recollection of the programme or may not yet be in the 'true' weight-loss maintenance phase as differing strategies may be used in the weight loss and weight loss maintenance phases. 26 Some studies also interviewed participants who had not previously participated in a weight-management programme.
Maintainers continued to pay attention to their dietary intake and it appeared to be more salient to them, compared to Regainers. Dietary self-monitoring is known to facilitate weight-loss maintenance, 32 increase feelings of responsibility, and heighten awareness of food intake. 23 [18][19][20] The only Regainers to mention monitoring change in weight or shape did so without reflection. This is important as a previous study 40 found that reflection was a key process in moving from self-weighing to self-regulation.
Participants desired continual monitoring by a someone else, echoing evidence that soliciting assistance has been found to increase accountability, enhance motivational support and is acceptabile. 21,25,28,39 However, these reports may also be attributable to the memory of successful weight-loss during WW when they were weighed at each meeting.
A distinguishing feature between groups was that Maintainers engaged more in anticipatory planning than Regainers (eg, planning for lapses at social gatherings and preparing for eating occasions).
They may have a heightened awareness of lapse potential in 'high-risk' situations or more self-efficacy to recover. Indeed, relapse prevention theory proposes that previous experience of applying effective responses in 'high-risk' situations may actually increase an individual's self-efficacy and positive outcome expectancy, and decrease the likelihood of future relapse. 9 Our findings suggest that future weight management programmes may benefit from placing additional emphasis on self-regulatory planning strategies (eg, weekly meal planning or goal setting), and should develop people's self-efficacy to manage in potential relapse situations.
Both groups implemented similar strategies to overcome impulses to eat food, but Regainers found them ineffective. This suggests that more in-depth exploration of other underlying factors and mechanisms is required. This could also help find ways to support subgroups needing additional assistance to manage their weight.
Reflecting previous studies, 22,23,41,42 our data showed awareness of the benefits and need for long-term sustainability and flexibility of behaviours. Indeed, restrictive dietary behaviours can potentially increase feelings of deprivation, inducing temptation and lapses. 10,23 Contrary to other studies, 22 Generally, participants' families showed an acceptance of the individual making dietary changes but our data also revealed instances of them challenging weight management, similar to previous studies. 43,44 A new insight was that Regainers had greater difficulty than Maintainers in navigating this challenge. Strategies supporting healthy interpersonal relationships could strengthen future programmes, such as information on communicating with partners/families.

| Strengths and limitations
A key strength of our study was the 5-year follow-up post programme; our findings offer unique insights into long-term mainte- those in the shorter intervention may have received less support and information on these strategies, therefore we acknowledge that this may have impacted their ability to implement them in the longer-term. Our findings should be understood in that context.
This study provided an in-depth exploration of the cognitive and behavioural strategies employed to overcome "lapses" and prevent "relapse" by people who had regained weight or maintained weightloss after participating in a weight management programme. Comparison of findings between the two groups suggest that monitoring, planning and managing interpersonal relationships are crucial strategies for successfully maintaining weight loss maintenance.
New important insights into long-term weight maintenance include the absence of PA self-monitoring strategies or using PA to counteract lapses, little avoidance of 'high-risk' social occasions, and that few Maintainers engaged in regular self-weighing. Further, both groups implemented distraction strategies for managing impulses but Regainers found them ineffective and also struggled to navigate interpersonal relationships. The findings of our study have already been successfully put into practice; they have been used to guide the content of an online acceptance-based intervention for weight loss maintenance. Subsequent studies should also test how these findings can be successfully integrated into programmes and through different delivery modes. Healthcare providers should incorporate these strategies in future programmes to equip participants with the skills to enact these strategies in the longer-term and prevent weight regain.

CONFLICTS OF INTEREST
ALA is the Chief Investigator on two publically funded (MRC, NIHR) trials where the intervention is provided by WW (formerly Weight Watchers) at no cost outside the submitted work. AJH reports receiving personal fees from Slimming World, outside the submitted work.