Qualitative study of perceptions of factors contributing to success or failure among participants in a US weight loss trial of financial incentives and environmental change strategies

Abstract

Background

The use of financial incentives and environmental change strategies to encourage health behaviour change is increasingly prevalent. However, the experiences of participants in incentive interventions are not well characterised. Examination of participant perceptions of financial incentives and environmental strategies can offer insights about how these interventions are facilitating or failing to encourage behaviour change.

Objective

This study aimed to learn how participants in a randomised trial that tested financial incentives and environmental interventions to support weight loss perceived factors contributing to their success or failure in the trial.

Design

Qualitative study with one-time interviews of trial participants with high and low success in losing weight, supplemented by study records of incentive payments and weight loss.

Participants

24 trial participants (12 with substantial weight loss and 12 with no weight loss) stratified equally across the 4 trial arms (incentives, environmental strategies, combined and usual care) were interviewed.

Analytical approach

Transcribed interviews were coded and interpreted using an iterative process. Explanation development was completed using an abductive approach.

Results

Responses of trial participants who were very successful in losing weight differed in several ways from those who were not. Successful participants described more robust prior attempts at dietary and exercise modification, more active engagement with self-limitations, more substantial social support and a greater ability to routinise dietary and exercise changes than did participants who did not lose weight. Successful participants often stated that weight loss was its own reward, even without receiving incentives. Neither group could articulate the details of the incentive intervention or consistently differentiate incentives from study payments.

Conclusions

A number of factors distinguished successful from unsuccessful participants in this intervention. Participants who were successful tended to attribute their success to intrinsic motivation and prior experience. Making incentives more salient may make them more effective for participants with greater extrinsic motivation.

Trial registration number

NCT02878343

Correspondence to Dr Karen Glanz; [email protected]

STRENGTHS AND LIMITATIONS OF THIS STUDY

• Participants were eager to share their experiences participating in the study which facilitated with data collection.

• Interviewees’ ability to recall trial details and experience may have been limited since these interviews were conducted >6 months after incentives ended.

• The survey response rate was 41.9%, which was sufficient for the purpose of this study but may have led to unmeasured bias.

Introduction

Identifying effective strategies for treating obesity and managing weight loss is of both clinical and public health significance.¹ Defined as a body mass index (BMI) of more than 30 kg/m², obesity among US adults increased from 11.0% to 35.0% for men and from 16.0% to 40.4% for women between 1960 and 2014.² Adult obesity is associated with higher rates of cardiovascular risk factors, disability, hospitalisation, healthcare expenditures and mortality risk.^2–4 Modest weight loss of about 10 lbs (4.5 kg) can reduce the incidence of diabetes and decrease risk of hypertension and hyperglycaemia.^{5 6}

Numerous approaches have been successful for achieving initial weight loss^{7 8} but maintaining weight loss over time is challenging.^9–12 Financial incentives can effectively modify health behaviours in a variety of contexts^13–16 and have been found effective in inducing initial weight loss.^14–16 It has become increasingly common to use financial incentives—with their roots in behavioural economics and specifically prospect theory¹⁷ for encouraging health behaviour change, especially in employee populations, often in combination with other strategies such as environmental changes. However, the experiences of participants in incentive-based interventions are not well characterised, although relevant lessons can be learnt from previous studies of facilitators and barriers to successful weight management.^18–20 A greater understanding of these experiences can offer insights about how these interventions are facilitating or failing to facilitate behaviour change and weight outcomes. This study was conducted to help interpret variability in results of a previously reported^{21 22} randomised trial of weight loss interventions and to gain insight into why some study participants benefited to a greater degree than others from the interventions.

Methods

Study design

This qualitative study was conducted at the conclusion of a four-arm randomised controlled trial that compared the effectiveness of financial incentives, environmental strategies and a combination of these interventions for increasing successful weight loss among employed adults who are obese.²² The trial enrolled 344 adults. Financial incentives were awarded at an expected value of US$3.00 per day or US$90 per month if the person’s weight loss goal was met.²¹ At the primary study endpoint, those in the incentive group had lost an average of 5.4 lbs more than the control group and those in the environmental strategies and combination groups lost 2.2 and 2.4 lbs, respectively, more than the controls. Financial incentives, environmental change strategies and the combination were not statistically significantly more effective than usual care.²²

In this interview study, we sought to learn more about participants’ perceptions of factors contributing to their success or failure at achieving significant weight loss during their study participation. To do so, we compared the experiences of trial participants across all four study arms who experienced the most weight loss (most successful) with those who experienced no weight loss (least successful), in an effort to identify factors that facilitated weight loss and those that hindered it. This kind of sequential explanatory mixed-methods approach is increasingly being used to help understand the results of clinical trials.^{23 24} The interview study was directed by an expert who was not involved in the weight loss trial (JC), and the lead researchers (KG and KGV) were involved in developing the interview questions and analysing the findings after they were coded. The Standards for Reporting Qualitative Research form provides details of all coauthors’ participation in both the trial and the interview study.

Participants

We used extreme case sampling (a type of purposive sampling approach) to recruit respondents who were most successful (top 10% wt loss) and least successful (bottom 10%, did not lose weight) at losing weight during the study period, stratified by study arm. This approach facilitated comparisons between cases to explore causes of substantial quantitative differences in trial outcome. Within each trial arm, we created a list of participants rank-ordered according to their absolute weight loss from baseline to the 18-month time point. We then recruited two groups of interviewees: for the first, we began with the participant with the most absolute weight loss and contacted participants by working down the list (to the second-most weight loss, third-most weight loss, etc); for the second, we began with the participant with the least absolute weight loss and worked up the list (to the second-least weight loss, etc). Because the sample was stratified by trial arm, it was possible to explore whether and how the different intervention strategies and usual care/control supported or limited participants’ weight loss efforts. Eligible participants were contacted by phone. If a participant could not be reached after three calls, the next individual in the rank order was contacted and the person was considered a non-respondent.

Patient and public involvement

All patients involved in this study agreed to participate and were properly consented, compensated and made aware of the purposes of the research. There was no public involvement.

Data collection

All interviews were conducted by JRC from June to August of 2018. Individuals who were reached and gave verbal consent completed one-time audiorecorded semistructured phone interviews. Participants received US$25 compensation for the interview.

Interviewees were asked about their general attitudes towards obesity and weight loss, the role of family and friends in their weight loss efforts, their dietary and exercise habits, their performance in the trial, the durability of any behavioural changes made during the trial and their summative assessments of the trial. Additionally, trial participants who had been randomised to one of the three intervention arms (financial incentives, environmental strategies, combined) were asked about their perceptions of the incentives, environmental strategies or both (see online supplemental appendix A for interview guide). Interviews were conducted until theoretical saturation: when additional data did not alter our coding schema or the explanations, we were developing to explain the patterns that coding made apparent (see Analysis).²⁵

Analysis

Interview recordings were transcribed professionally. Transcripts were coded using NVivo V.11 (QSR International). Supervised by JC, JRC and AC annotated the same four randomly selected transcripts to generate themes, which were then discussed, defined and organised into a formal codebook (a taxonomy for thematically categorising data).²⁶ Using this codebook, JRC and AC carried out two rounds of double coding. In each round, they double coded three randomly selected transcripts, performed intercoder comparison through the calculation of kappa values, and discussed and rectified any instance of coding with a kappa of less than 0.6. After each round, the codebook was revised to refine ambiguous categories, eliminate categories lacking relevance and create new categories to capture important themes previously missing from the taxonomy. Having refined the codebook and reached consensus about its application, the remaining transcripts were single coded by JRC and AC. All codebook revisions were applied to previously coded transcripts.

Once coding was complete, the two coders independently compared the responses of interviewees in the most successful weight loss group versus those in the least successful weight loss group across codebook categories in order to identify thematic differences between groups. They met to reach consensus on these differences. Blinded to their exact nature, JC then verified these differences by reviewing the relevant thematic areas. Group discussion involving all investigators was then used to develop explanations that could account for the thematic differences between the groups. Explanation development used an abductive approach^{25 27–29} in which we posited explanations for unexpected trends in the data, evaluated them against the data to assess their level of empirical support, and revised them until eventually arriving at a theoretical account that best fit our data.

Results

We interviewed 24 trial participants: 12 of the most successful participants and 12 of the least successful participants. Each performance group was equally stratified across the four trial arms (three interviewees per arm in each group). 12 of the 25 most successful participants (48%) and 12 of 35 least successful participants (34%) who were contacted were ultimately interviewed. See table 1 for demographic and physical characteristics of the interviewees. There were no significant differences between most and least successful interviewees in terms of age, baseline weight and BMI, gender, race/ethnicity, marital status, education, household income or household size. Below, we describe differences and commonalities between the most successful and least successful participants based on their interview responses.

Table 1

Demographic differences between most successful and least successful participants (interviewed only; n=24)

Characteristics	Least successful (N=12)	Most successful (N=12)	P value, Wilcoxon test (two sided)
Age			0.89
Mean	44.1	43.6
SD	10.09	10.87
Baseline–weight			0.62
Mean	238.5	250.2
SD	41.68	48.34
Baseline–BMI			0.79
Mean	38.0	38.6
SD	4.81	4.78
Weight–change			<0.0001
Mean	15.1	−35.1
SD	4.19	18.74
Gender, n (%)			0.62
Male	2 (16.7)	3 (25.0)
Female	10 (83.3)	9 (75.0)
Race/ethnicity, n (%)			0.57
Black	7 (58.3)	7 (58.3)
White	4 (33.3)	5 (41.7)
Hispanic/multirace/other/unknown	1 (8.3)	0 (0.0)
Marital status, n (%)			0.84
Single	4 (33.3)	3 (25.0)
Married/unmarried partners	6 (50.0)	6 (50.0)
Divorced/widowed	2 (16.7)	3 (25.0)
Education, n (%)			0.41
Less than college or some college	7 (58.3)	5 (41.7)
College graduate	5 (41.7)	7 (58.3)
Household income, n (%)			0.48
US$49 999 or less	3 (25.0)	5 (41.7)
US$50 000–US$74 999	6 (50.0)	3 (25.0)
US$75 000–US$99 999	0 (0.0)	1 (8.3)
US$100 000 or more	3 (25.0)	3 (25.0)
Household size, n (%)			0.67
1 person	3 (25.0)	2 (16.7)
2 people	2 (16.7)	1 (8.3)
3–4 people	5 (41.7)	8 (66.7)
5+ people	2 (16.7)	1 (8.3)

BMI, body mass index.

Attempts at weight loss

Interviewees were asked to describe their attempts, prior to enrolment in the trial, at trying to lose weight (see table 2). Most of the responses focused on changes in food intake, with occasional mentions of exercise. Nearly all of the most successful respondents described specific and purposeful attempts, often multiple, to implement healthier dietary habits prior to participation in the study (2.1). In contrast, just two of the least successful interviewees said they had undertaken significant dietary modification before the trial, and many characterised themselves as generally inattentive to the healthiness of what they ate (despite not being prompted to sum up their attitudes toward diet) (2.2).

Table 2

Attempts at weight loss

	Illustrative quotes
2.1 Attempts at weight loss pretrial: most successful	INTERVIEWER: And prior to your participation in the Healthy Weight Study, were you doing anything else to try and lose weight? MOST SUCCESSFUL INTERVIEWEE: I would try to watch and balance what I eat and exercise moderately—probably exercise and just watch what I take in, how many—limit my calories and watch my intakes of sweets and fat. I: And were those methods—were you able to successfully lose weight? H:(…)At certain times it would seem like it’s working and then other times I would just seem like it wasn’t. It wasn’t nothing that was long-term or maintained throughout a long period of time. No. But I mean, there were times when I just made sure I didn’t overeat. I’m not a overeater person. I just probably—I don’t eat a lot of between or anything like that. It’s probably my portions when I eat, because I’m not a person who snacks all day and grabs stuff. I mean, that’s rare when I do. I do, but it’s rare. But mostly probably my portion size at my meals and my meals in general probably know there was too many calories I would think. So it was more like that when I sit down to be mindful of my plate—portion size and what’s on the plate. I: And did this differ from your previous diet? R: Well, it depends, because what I would do when I didn’t want to cook or worry about food I would just get like the Weight Watchers thing and eat that meal—that meal size, because I need a portion size that’s smaller and the right size that we eat. And I would do that opposed to fixing it for myself.
	H: I believe I was doing Weight Watchers the second time I did it, along with the Healthy Weigh study. I was very successful the first time, and really watched what I ate. Actually, I—believe it or not, I was eating more during Weight Watchers than I was prior. And, I think that’s what helped me lose weight over the course of ten weeks, I lost about 30 pounds. And, in reality, looking back at it, I wasn’t eating enough during the week, running around at work and stuff. And then, on the weekends, I would overeat. So, it was kinda like a vicious cycle. And, I started eating that same amount of food every day. And, it is true, you have to eat to lose weight, if you’re eating the right things.(…)When you’re on Weight Watchers, you’re making up separate meals for the program and counting the same amount of points. You can only have, maybe, 12 points for lunch because you had 10 points for breakfast, and you wanna have like 20 points for dinner. So, you’re kinda managing what you eat so you have to make different choices of what you’re gonna take with you. Healthy snacks.
	H: Right before [participation in the trial] is when I got really serious about losing weight. I think that, right when the [trial] started, I might have lost 15 pounds. I went to see a nutritionist, got some information about just reading nutrition labels and things that I should be having in my diet every day. I was tracking everything I ate. I knew how many calories I should be having every day, how many calories I need to burn every day.
2.2 Attempts at weight loss pretrial: least successful	I: And how did this differ from your previous diet from before the study? LEAST SUCCESSFUL INTERVIEWEE: There really wasn’t no diet. I just ate—to me, it seemed like every time I said, let me cut down, I ate more. Every time I tried to diet I ate more, so that wasn’t a diet. You know what I mean? It seemed like when I made an effort to lose weight, I ate more, and it didn’t make any sense to me. I’m like, is this a mind thing? Like, this is crazy. So there really wasn’t no diet before.
	I: And prior to joining the Healthy Weigh Study were you doing anything else to try and lose weight? N: Not at that moment. No. I was just kind of trying to find fast stuff to cook and wasn’t really putting any thought into the kind of food it was or the impact it had on me.
	I: And prior to participating in the Healthy Weigh Study were you doing anything else to try and lose weight? N: I would try to eat healthy every now and then and exercise every now and then, but it wasn’t consistent.

Addressing self-limitations

Interviewees from both groups described a number of perceived self-limitations that they said made it difficult to lose weight, for example, laziness or lack of discipline. The groups’ approaches to these limitations differed (see table 3). The most successful interviewees described self-limitations and recounted specific strategies or approaches they used to address and overcome them (3.1), while the least successful respondents were considerably less likely to articulate such strategies. 10 of 12 least successful interviewees described physical limitations they attributed to their weight including chronic pain, lack of mobility and chronic fatigue, while just 4 of the most successful interviewees described physical limitations (3.2).

Table 3

Addressing self-limitations

	Illustrative quotes
3.1 Most successful address self-limitations	MOST SUCCESSFUL INTERVIEWEE: For me, it’s mind over matter. If I really wanted it bad enough, the only way it was gonna happen is if I did it. So I just did it.(…)It was something that I wanted to do. And I knew the only way that I was gonna do it is if I did it, and not procrastinate or have an excuse as to why I couldn’t do it.
	H: I hated going [to the gym]. I hated getting up. And I felt the best time for me to go [was] in the morning, so I was up at 5:00. I hated that. There was nothing I liked about the gym, so that was one reason I stopped going to that. I found that if I just did my own—like a walk, a three-mile walk—I was better and more likely to do it.
	H: See for me, because I know how I am. That’s why I got a membership next door to my job that I could work out on my lunch. Because if I joined Planet Fitness I was not gonna go after work because I’m tired. But if I use my lunch to go to the gym and it’s right next door—I just have to walk next door. I did things based on my personality. I could be a little lazy, so I had to do what I knew would be convenient for me by using my lunch to go exercise.
3.2 Weight-related functional limitations in least successful participants	I: And how does your weight affect your daily life? LEAST SUCCESSFUL INTERVIEWEE: I guess it affects my ability like if I feel like I have a lot of energy or if I can’t do a lot. If I feel, I guess, if I’m feeling literally—well not overweight like that, but I feel like it impacts how I feel, what I get done in a day, what kind of energy level I have for me.(…)I guess the level of activity, the level of exercise. I think if I feel too overweight or too overwhelmed it would limit me because I wouldn’t feel like it. It would just make me more dormant than more active.
	I: And how does your weight affect your daily life? N: Movement a little bit slower. My movement is a little bit slower.(…)Sometimes, just a little bit of difficulty going up the steps or down the steps.
	N: So, I haven’t been able to run as much as I used to because of pain. I: Okay. Aside from running, are there other ways in which your weight impacts what you’re able to do? N: Not really. I mean, I try to do more low-impact workouts now but, that’s basically it.

Social reinforcement and accountability to others

In discussing the involvement of family, friends and colleagues in their weight-loss efforts, the most successful respondents placed great emphasis on the importance of these relationships in holding them accountable for pursuing these efforts as well as positively reinforcing healthy behaviours. This social accountability and reinforcement could be direct—for example, help or advice from friends and family about cooking, eating and exercise:

MOST SUCCESSFUL GROUP INTERVIEWEE: [M]y mom and my fiancé [go] to the gym with me or just we go on walks or prepare my meals for me—the healthier meals for me. […] And my fiancé, he makes sure that we keep salmon in the house, because he knows that that’s something that I like that’s healthy. So my family is very supportive of it, because they see how happy I have been since I’ve lost weight. […] I got my first FitBit in May of 2016, and it was because my cousin—she had one and when she was visiting she was just bouncing all around. I’m like, why are you not sitting down? And she’s like, I gotta get my steps in. I gotta keep moving. I’m in a challenge. So then I was like, oh, well, then I wanna get a FitBit too, because I was already in the study and I was like, that’ll help me keep moving and get motivated to exercise.

Accountability to others could also be important, as one high-improvement interviewee recounted: ‘I have a co-worker who sits behind me who brings her lunch every day and she makes me mindful to bring one, because it’s just so easy. I said, ‘well, oh I’ll go out and get a salad or something.’ Non-improvement interviewees gave fewer and less detailed accounts of the role of social relationships in their attempts to lose weight. When asked how the study could have been improved to better facilitate weight loss, half of the least successful interviewees (as opposed to only one of the most successful respondents) said that they would have benefited from more in-person contact with study personnel, suggesting that having another person to report to and get support from, this would create a social environment that was more reinforcing of their weight loss efforts:

NON-IMPROVEMENT INTERVIEWEE: I think that if they did more interaction with the participants or at least set up something where participants weren’t left to themselves every six months [I would be more successful]. … [with] more interaction and buddying, I think people would do better. […] I think that would be more helpful […]. There’s nobody to talk to them and some people may be completely clueless on how to do it.

Behavioural routinisation and maintenance

In describing their diet and exercise behaviours, the groups exhibited differences in the degree to which these behaviours were routinised or regimented (see table 4). The most successful participants often described successful attempts to develop dietary and exercise routines and were detailed in their accounts of the attention they paid to implementing and maintaining such routines (4.1). Least successful respondents rarely described success in routinising diet or exercise: in fact, more non-improvement respondents (3) reflected explicitly on the difficulties they had in routinising diet and exercise than described successful routinisation (2) (4.2). When characterising their diet and exercise after the trial, the most successful interviewees self-reported higher maintenance of behavioural change than did the least successful interviewees, and their accounts of how their approaches to diet and exercise have changed were also more detailed.

Table 4

Behavioural routinisation and durability

	Illustrative quotes
4.1 Routinisation of diet and exercise among most successful	MOST SUCCESSFUL INTERVIEWEE: [During the trial] it became an everyday occurrence that I knew I had to be weighed, and I didn’t want to go get weighed heavier, so I would stick to the diet and what they recommended to do.(…)Even before I did the study, every once in a while, I would ride my bike to work, but during the study it became more of an incentive to ride it more. Sometimes I would take my car, but during the study—since I had to report and get weighed—then I made it a habit.(…)A regular routine—not a habit. I made it a routine.
	H: I just had to figure out what worked for my schedule and, instead of going after work when I was already exhausted, I started going to the gym before work. It was hard to figure out a schedule that would work, and then figuring out how to stop—how to fight cravings and to meal prep, so I had food with me, instead of making bad choices when I was out, and always having snacks – healthy snacks with me. So, it was difficult but, after a while, it got normal.(…)Yeah, I mean, I had to find out what worked.(…)It was like a trial and error kinda thing. I had to learn that if I cooked on Sunday, I could only prep up until Wednesday because, by Friday, I would hate the food and I didn’t wanna eat it. So, I would—I learned to just prep for Monday to Wednesday. And then, I would prep Wednesday night, for the rest of the week. But I didn’t learn it until later. But, I was throwing a lot of food away because I would cook on Sunday and cook all the way to Friday, and then, by Friday, I would—Thursday, I would just be like, I hate this. So, it was difficult just trying to figure out what worked. And, like I said, I work in a courtroom. There’s not—there’s nowhere to store things. I would have to figure out, where am I gonna put this food? How am I gonna sneak away and eat this—eat the food? But, I had to figure that out. But, once you get into a rhythm and, after some trial and error, you figure it out.
4.2 Least successful difficulties with routinisation	LEAST SUCCESSFUL INTERVIEWEE: I think if I miss a day I just give up, and just have to regroup and kinda reconfigure. Because I wanna be regimented again, and it’s very hard for me to manage getting regimented.(…)(I)t was about changing my behavior and actually being able to stick to a regiment. And for me—that was where the struggle was for me. I think the study was helpful. I think they sent supportive information, the friendly reminder. It was a very well set up study. But just for me, I would just have to just kinda say I didn’t achieve my success because I really—didn’t really focus in on prioritizing the weight loss.(…)I’ve gone to see probably three additional nutritionists over that time and afterwards, and I just feel like I’m still struggling to try to get regimented, because nobody can jump in my body and tell me, you have to change.
	N: I know that I don’t have self-discipline. So I would just start off doing good and then I would relapse.

Understanding and impact of incentives

Among interviewees who were in one of the two trial arms that involved incentives, both the most and least successful participants had difficulty recalling the details of the incentive intervention (see table 5, 5.1). They frequently conflated the incentives with the study payments that all individuals received for participation in the trial (eg, for completing surveys and weigh-ins) because the amounts were often similar. Interviewees’ responses were mixed when asked whether incentives were a major motivator in their weight loss efforts (5.2). Some interviewees described the incentives as effective motivators, while others stated that their central motivation for participating in the study was more intrinsic—simply to lose weight and that this accomplishment would be its own reward, rendering incentives less relevant—a sentiment more common among most successful interviewees than among the least successful respondents. Additionally, some interviewees across both groups considered the financial rewards not large enough to adequately incentivise their weight loss efforts.

Table 5

Understanding and impact of incentives

	Illustrative quotes
5.1 Difficulties recalling incentive design	INTERVIEWER: And were you eligible to receive financial bonuses if you met your weight loss goals? LEAST SUCCESSFUL INTERVIEWEE: I do not remember. I: Okay, that’s fine. Can you recall how the bonuses worked, just generally speaking? N: Yeah. If you—I’m trying to remember. Was it the weigh-ins? You got weighed in and I think that was it, just according to the weigh-ins. I: Okay. So if you completed the weigh-ins, you’d get the financial bonus? N: Yes. Right. Right. Right.
	I: Do you recall how these bonuses worked? N: It was like, if you were under a certain amount of weight, then you would get a certain amount of money, I believe, or something like that.
	I: Can you recall how these bonuses worked? H: No. I don’t. I’m sorry.(…)They’d say that you’ve met your goals or over it. You achieved it or something. No. I don’t. I’m sorry. I don’t.
	I: Can you recall how these bonuses worked? H: I think when I went to the weigh-in, if I met my goal I got like 50 dollars or 100—dependent on how many months or something and if you got weighed within the timeframe. Is that right? I think—yeah.
5.2 Incentives as motivation	I: And when you received one of these bonuses, what was your reaction? N: I was happy because it made me feel like, okay, I’m really doing something, and I’d really feel my work was great. I: Mm-hmm. And alternatively, when you didn’t qualify for receiving a bonus because you didn’t reach your weight loss goal, what was your reaction then? N: It made me a little sad because I thought I was working hard or I knew that I wasn’t working as hard as I could have, so it just pushed me to want to just do better next time.
	And when you received one of these bonuses, what was your initial reaction? H: Oh, happy. It was nice. I: Right. Alternatively, when you didn’t qualify for receiving a bonus because you didn’t reach your weight loss goal, what was your reaction then? H: Wished I would have met the goal, but I wasn’t heartbroken over it. I: Okay. And did you find that the financial bonuses were a major motivating factor for taking steps to lose weight? H: No, not really. A major—it was a motivator. I wouldn’t say major motivator. Yeah, I think the main motivator was just to make sure you stayed on track for what you said your goal was. That was more the main motivator.
	I: Right. And when you didn’t qualify for a bonus because you didn’t reach your weight-loss goal, what was your reaction then? H: Well, I mean the monetary thing—that wasn’t even why I did it—for real. So even if it was just a regular study, I would have joined it. So yeah. I wasn’t upset. I was upset because I didn’t make my goal—because my weight didn’t go down. I wasn’t upset because I didn’t make my goal and I wasn’t going to be rewarded.
	H:(M)y goal is to lose weight, not to make money. All right. My goal is to lose weight and get healthier.(…)If the study was offered again to participate and there’s able to be—no reward—I would still join it. I: And how much greater do you think the financial rewards would have to be to make a difference in your effort to lose weight? H: To me? None. Like I said, it doesn’t matter to me. I mean, health is more important to me than any financial thing that you can offer.
	I: And when you received one of these bonuses, what was your initial reaction? H: Nothing, really. It wasn’t a significant amount, just—it wasn’t like it made me a millionaire. It was very minute.

Discussion

This study was conducted to help interpret variability in results of a randomised trial of weight loss interventions and to gain insight into why some study participants benefited to a greater degree than others from the interventions. We found several differences between the most successful and least successful participants. Successful participants described much greater awareness of limitations and strategies for dealing with barriers, a greater sense of social accountability, and greater success in developing routines and sustaining those routines for dietary intake and/or exercise. Interviewees’ responses were mixed with respect to whether the financial incentives were central to their efforts to lose weight. This is consistent with previous research which found that, even in contexts in which smokers quit smoking at higher rates in an incentives group versus a control group, successful participants attributed their success to intrinsic motivation and not the incentives.^{30 31}

Our interview findings also highlight the importance of the broader trajectory of participants’ efforts to lose weight. The most successful interviewees described repeated, specific efforts to lose weight (eg, through participation in programmes such as weight catchers) and increased exercise prior to participation in the trial, while the least successful interviewees rarely did. It may have been the case that some participants were more fully prepared to successfully engage in these interventions, which were low-intensity and did not involve strategies such as group classes or individual coaching and counselling.

Related, the most successful participants were more likely to describe specific strategies for overcoming self-limitations. As noted in previous research on facilitators to successful weight loss,^{18 19} this suggests that those who did well in the trial were better able to engage in overcoming their tendencies that made it difficult to lose weight. That said, participants who were least successful may have had more frequent and severe functional limitations that may have prohibited active engagement, and future initiatives should consider how to design programmes that facilitate greater ongoing engagement with these individuals.

Social support and accountability to others were reported to be very important to weight loss for the most successful participants. The least successful described less support, as well as social environments less conducive to reaffirming healthy eating and exercise. In contrast to more successful participants, they also frequently cited lack of face-to-face support as a central limitation of the trial, which is consistent with findings of McVay et al.²⁰ While regular interaction via phone/internet/email reminders and tips was sufficient for some, an important question for future research is how to (1) identify participants who do not have sufficient built-in social support to do well in an intervention that does not provide dedicated support and (2) how to provide enhanced social support for those participants. Some of the comments indirectly allude to the environmental strategies that were tested, but respondents in the two study arms with environmental change strategies made few direct comments about how they experienced those strategies.

According to the interviews, the most successful and least successful participants reported similar levels of health literacy, engagement with nutrition and exercise information, and beliefs in the importance of losing weight, suggesting that it is difficult to identify ex ante who might be a responder or non-responder. However, the most successful people were better able to routinise diet and exercise changes than non-improvers. This may have been due to increased ability to form routines, greater day-to-day stability in schedules, or more willingness to prioritise these health promoting changes, as was found in Greaves et al’s review of 26 qualitative studies.¹² Similarity in participants’ enrolment characteristics make predicting success difficult, but future initiatives may consider polling participants about previous weight loss attempts, as prior studies have shown this to be a successful predictive factor.^{32 33}

Further, the fact that successful participants did not attribute their success to incentives should be noted. All participants were eligible to receive participation compensation of US$300 over the course of the study for enrolment, in-person weigh-ins and completing surveys. The financial incentives that were conditional on weight loss received by interviewed participants in the incentive and combined groups ranged from US$265 to US$940 with a median of US$400. Few participants in either group were able to describe the incentive structure and they often confused incentive payments with study participation payments, suggesting that the incentives conditional on weight loss were insufficiently salient. Better communication and reinforcement of information about the specifics of the incentive intervention might have made them more impactful. Also, planning for the incentives to be markedly higher than routine study payments may have achieved different results, though we used high participation incentives to avoid differential attrition in the arms of the study, receiving conditional financial incentives and not, and that was successful.

This study has limitations. The design of the study, which involved purposive sampling of high-success and low-success participants, has both strengths and weaknesses. This design gained efficiency by making it possible to contrast the participants who were most successful and least successful in losing weight. Participants were asked the same questions without regard to their weight loss success (see online supplemental appendix), to avoid leading questions. However, this design may have missed information about the ‘moderately successful’ participants. In addition, interviewees’ ability to recall the trial details and their experiences participating may have been limited since the interviews were conducted more than 6 months after their trial participation ended. The response rate was 41.9%; however, qualitative participants were comparable to eligible participants.

Conclusion

This study was conducted to help explain why some participants were more successful than others in a randomised trial of weight loss interventions^{21 22} and to increase understanding of why some study participants achieved better results from the interventions than did others. We found a number of interesting attributions of why the most successful participants in a weight loss intervention trial were better able to lose weight than the least successful participants. For example, the most successful participants articulated specific strategies for overcoming obstacles to their weight loss, reported having more support networks and were better able to form new habits of healthier eating and activity. These explanations were fairly consistent across participants assigned to the three study intervention arms and consistent with findings of previous qualitative studies of participants in weight loss programmes.^{12 18 19} Some of this may reflect differences in the psychological predispositions of people who respond better to a given intervention¹² as opposed to effects of the intervention, and we cannot readily distinguish between these two possibilities. Nonetheless, this study provides helpful insights in considering ways to further enhance the effectiveness of such interventions.

Data availability statement

Data are available on reasonable request.

Ethics statements

Patient consent for publication

Not applicable.

Ethics approval

This study involves human participants and this study was reviewed and approved by the University of Pennsylvania Institutional Review Board, IRB Protocol # 821428. Participants gave informed consent to participate in the study before taking part.

Footnote

Contributors KG and KGV conceptualised and directed the primary study. JRC conceived and led data collection and analysis for the qualitative study. KG, CK, AC, JRC, KGV and JC contributed to the study execution and analysis of the results. KG, CK and JC drafted the manuscript and KGV, JRC and AC helped to edit the manuscript. KG serves as overall guarantor.

Funding The research reported here was supported by Cooperative Agreement 1U48DP005053 from the US Centers for Disease Control and Prevention. Its contents are solely the responsibility of the authors and do not necessarily reflect the official views of the Centers for Disease Control and Prevention or DHHS.

Competing interests KGV is a part-owner of the consulting firm, VALHealth. He has received research funding from Hawaii Medical Services Association, not related to the current research.

Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

Provenance and peer review Not commissioned; externally peer reviewed.

Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.

References

¹ Obesity and overweight fact sheet. 2018. Available: www.who.int/mediacentre/factsheets/fs311/en/ [Accessed 18 Feb 2018 ].

² Flegal KM, Kruszon-Moran D, Carroll MD, et al. Trends in obesity among adults in the United States. JAMA 2016; 315: 2284. doi:10.1001/jama.2016.6458

³ Adams KF, Schatzkin A, Harris TB, et al. Overweight, obesity, and mortality in a large prospective cohort of persons 50 to 71 years old. N Engl J Med 2006; 355: 763–78. doi:10.1056/NEJMoa055643

⁴ Yan LL, Daviglus ML, Liu K, et al. Midlife body mass index and hospitalization and mortality in older age. JAMA 2006; 295: 190–8. doi:10.1001/jama.295.2.190

⁵ Finkelstein EA, Trogdon JG, Brown DS, et al. The lifetime medical cost burden of overweight and obesity: implications for obesity prevention. Obesity (Silver Spring) 2008; 16: 1843–8. doi:10.1038/oby.2008.290

⁶ Delahanty LM. Weight loss in the prevention and treatment of diabetes. Prev Med 2017; 104: 120–3. doi:10.1016/j.ypmed.2017.07.022

⁷ Nicklas JM, Huskey KW, Davis RB, et al. Successful weight loss among obese U.S. adults. AM J Prev Med 2012; 42: 481–5. doi:10.1016/j.amepre.2012.01.005

⁸ Spreckley M, Seidell J, Halberstadt J. Perspectives into the experience of successful, substantial long-term weight-loss maintenance: a systematic review. Int J Qual Stud Health Well-Being 2021; 16: 1862481. doi:10.1080/17482631.2020.1862481

⁹ Gardner CD, Trepanowski JF, Del Gobbo LC, et al. Effect of low-fat vs low-carbohydrate diet on 12-month weight loss in overweight adults and the association with genotype pattern or insulin secretion: the DIETFITS randomized clinical trial. JAMA 2018; 319: 667–79. doi:10.1001/jama.2018.0245

¹⁰ Thomas JG, Bond DS, Phelan S, et al. Weight-loss maintenance for 10 years in the National weight control registry. Am J Prev Med 2014; 46: 17–23. doi:10.1016/j.amepre.2013.08.019

¹¹ MacLean PS, Wing RR, Davidson T, et al. NIH working group report: innovative research to improve maintenance of weight loss. Obesity (Silver Spring) 2015; 23: 7–15. doi:10.1002/oby.20967

¹² Greaves C, Poltawski L, Garside R, et al. Understanding the challenge of weight loss maintenance: a systematic review and synthesis of qualitative research on weight loss maintenance. Health Psychol Rev 2017; 11: 145–63. doi:10.1080/17437199.2017.1299583

¹³ Volpp KG, Gurmankin Levy A, Asch DA, et al. A randomized controlled trial of financial incentives for smoking cessation. Cancer Epidemiol Biomarkers Prev 2006; 15: 12–8. doi:10.1158/1055-9965.EPI-05-0314

¹⁴ Volpp KG, Troxel AB, Pauly MV, et al. A randomized, controlled trial of financial incentives for smoking cessation. N Engl J Med 2009; 360: 699–709. doi:10.1056/NEJMsa0806819

¹⁵ John LK, Loewenstein G, Troxel AB, et al. Financial incentives for extended weight loss: a randomized, controlled trial. J Gen Intern Med 2011; 26: 621–6. doi:10.1007/s11606-010-1628-y

¹⁶ Kullgren JT, Troxel AB, Loewenstein G, et al. Individual- versus group-based financial incentives for weight loss: a randomized, controlled trial. Ann Intern Med 2013; 158: 505–14. doi:10.7326/0003-4819-158-7-201304020-00002

¹⁷ Kahneman D, Tversky A. Prospect theory: an analysis of decision under risk. Econometrica 1979; 47: 263. doi:10.2307/1914185

¹⁸ Garip G, Yardley L. A synthesis of qualitative research on overweight and obese people’s views and experiences of weight management. Clin Obes 2011; 1: 110–26. doi:10.1111/j.1758-8111.2011.00021.x

¹⁹ Teixeira PJ, Carraça EV, Marques MM, et al. Successful behavior change in obesity interventions in adults: a systematic review of self-regulation mediators. BMC Med 2015; 13: 84. doi:10.1186/s12916-015-0323-6

²⁰ McVay MA, Yancy WS, Bennett GG, et al. Perceived barriers and facilitators of initiation of behavioral weight loss interventions among adults with obesity: a qualitative study. BMC Public Health 2018; 18: 854. doi:10.1186/s12889-018-5795-9

²¹ Glanz K, Shaw PA, Hoffer K, et al. The healthy weigh study of lottery-based incentives and environmental strategies for weight loss: design and baseline characteristics. Contemp Clin Trials 2019; 76: 24–30. doi:10.1016/j.cct.2018.10.013

²² Glanz K, Shaw PA, Kwong PL, et al. Effect of financial incentives and environmental strategies on weight loss in the healthy weigh study: a randomized clinical trial. JAMA Netw Open 2021; 4: e2124132. doi:10.1001/jamanetworkopen.2021.24132

²³ O’Cathain A, Thomas KJ, Drabble SJ, et al. What can qualitative research do for randomised controlled trials? A systematic mapping review. BMJ Open 2013; 3: e002889. doi:10.1136/bmjopen-2013-002889

²⁴ Ivankova NV, Creswell JW, Stick SL. Using mixed-methods sequential explanatory design: from theory to practice. Field Methods 2006; 18: 3–20. doi:10.1177/1525822X05282260

²⁵ MacQueen KM, McLellan E, Kay K, et al. Codebook development for team-based qualitative analysis. CAM Journal 1998; 10: 31–6. doi:10.1177/1525822X980100020301

²⁶ Hennink MM, Kaiser BN, Marconi VC. Code saturation versus meaning saturation: how many interviews are enough Qual Health Res 2017; 27: 591–608. doi:10.1177/1049732316665344

²⁷ Glaser BG, Strauss A. The discovery of grounded theory: strategies for qualitative research. Chicago, IL: Aldine Publishing Co, 1967.

²⁸ Timmermans S, Tavory I. Theory construction in qualitative research: from grounded theory to abductive analysis. Sociol Theory 2012; 30: 167–86. doi:10.1177/0735275112457914

²⁹ Tavory I, Timmermans S. Abductive analysis: theorizing qualitative research. University of Chicago Press, 2014: 176.

³⁰ Van den Brand FA, Dohmen LME, Van Schayck OCP, et al. 'Secretly, it’s a competition': a qualitative study investigating what helped employees quit smoking during a workplace smoking cessation group training programme with incentives. BMJ Open 2018; 8: e023917. doi:10.1136/bmjopen-2018-023917

³¹ Kim A, Kamyab K, Zhu J, et al. Why are financial incentives not effective at influencing some Smokers to quit? results of a process evaluation of a worksite trial assessing the efficacy of financial incentives for smoking cessation. J Occup Environ Med 2011; 53: 62–7. doi:10.1097/JOM.0b013e31820061d7

³² Teixeira PJ, Palmeira AL, Branco TL, et al. Who will lose weight? A reexamination of predictors of weight loss in women. Int J Behav Nutr Phys Act 2004; 1: 12. doi:10.1186/1479-5868-1-12

³³ Kerrigan SG, Clark M, Convertino A, et al. The association between previous success with weight loss through dietary change and success in a lifestyle modification program. J Behav Med 2018; 41: 152–9. doi:10.1007/s10865-017-9883-6