Long-term weight loss following a randomised controlled trial of a weight management programme for men delivered through professional football clubs: the Football Fans in Training follow-up study

Outcome and predictor assessment

Stadia measurements were conducted by a team of fieldworkers, trained to standardised protocols, during 29 sessions across the 13 football clubs between March and May 2015. Home visits took place between June and August 2015. Most home measurements were conducted at the participant’s home, but one participant attended at the University of Edinburgh and another requested a visit to his workplace. Percentage body fat was not recorded among these participants. In addition, some men who were contacted for home visits elected to provide only their weight (which was either objectively measured by a fieldworker visiting them at home or self-reported to a fieldworker over the telephone). Men who declined a home visit or giving self-reported weight by telephone and men whom the fieldworkers had been unable to reach by telephone or e-mail during the follow-up study were sent a weight-only self-report form and prepaid return envelope following the completion of home visits. All outcome data collection ended on 25 September 2015.

The fieldwork team

Data were collected by teams of fieldworkers. Each measurement session team had a designated fieldwork team leader and a fieldwork nurse responsible for BP measurement. Fieldworkers were trained to standard measurement protocols by experienced research and survey staff in the Medical Research Council (MRC)/Chief Scientist Office (CSO) Social and Public Health Sciences Unit. Training took place over 2 days for stadia measurement staff. This included a half-day of training for fieldwork team leaders only, and 1.5 days of training for fieldworkers, with training sessions led jointly by the fieldwork team leaders and members of the research team. A further day of training for home visit fieldworkers was provided to emphasise strict adherence to protocol during home visits to minimise detection bias. All fieldworkers wore T-shirts branded with FFIT research logos at in-stadia measurement sessions (Figure 2) and during home visits.

FIGURE 2.

Fieldwork team arriving at a FFIT follow-up 3.5 years in-stadia measurement session.

Measurement protocols

Physical activity

Following standard procedures described in the International Physical Activity Questionnaire scoring protocol,54 we calculated and reported metabolic equivalent of task (MET) minutes per week for self-reported walking, vigorous, moderate and total PA, and time in minutes for daily sedentary time over the last 7 days. Activity times that exceed 180 minutes (3 hours) per day were truncated to 180 minutes to normalise the distribution of levels of activity, which are usually skewed in large population data sets. 68

Diet

We used the adapted DINE from the FFIT RCT43 to collect self-reported frequency of intake of various fatty and sugary food types, and fruit and vegetables over the previous 7 days (see questions 27–32, Report Supplementary Material 4). From these data, we calculated a fatty food score, a sugary food score, and a fruit and vegetables score (see Report Supplementary Material 5).

Alcohol intake

Following Emslie et al. ,58 we converted responses to the 7-day recall diary for alcohol to standard units that are equivalent to 8 g of pure alcohol (e.g. half a pint of ordinary beer, lager or cider, a small glass of wine and one measure of spirit each contain 1 unit of alcohol). We calculated the total number of units reported in the past week.

Psychological outcomes

Scores for both the RSE and PANAS short form were normalised so that values could be calculated for participants who had missed one or two items contributing to each scale. The PANAS normalised scales scores range from 1 to 5, with higher scores indicating higher negative affect and higher positive affect. Higher scores on the RSE (normalised range 0–3) indicate better self-esteem. SF-12v2 scores were summarised separately for mental and physical HRQoL following standard algorithms. 61

Mediators

Diet

In our protocol we said that we would measure change from baseline to 3.5 years in self-reported frequency of intake of key contributors to weight gain (e.g. fast foods, chocolate bars, chips, pies and sugary drinks) and of breakfast using questions adapted from DINE. However, to be consistent with the reporting of the FFIT RCT,43 we summarised the 17 original variables into three scores (fatty food, sugary food, and fruit and vegetables; see Preparation of self-reported variables for analysis, Diet) indicative of healthy changes that men could have made to their diets.

Satisfaction with behaviour and weight changes

In our protocol we said that we would measure perceived satisfaction with current PA and dietary behaviours. However, given evidence from reviews that suggests that satisfaction with short-term outcomes (including weight) may predict long-term outcomes,19,23 we asked men how satisfied they were with what they had experienced as a result of the changes they had made to their diet, PA and weight at the end of the FFIT programme, as well as currently.

Long-term weight trajectories

Table 6 shows that men in the FFIT-FU-I group gained 2.59 kg (95% CI 1.61 to 3.58 kg; p < 0.001) between the 12-month and the 3.5-year measures (i.e. 2.44%, 95% CI 1.61% to 3.27%, of their baseline weight). This equates to a weight gain of 1.04 kg per year. Nevertheless, 3.5 years after baseline measurement, men in the FFIT-FU-I group still weighed 2.90 kg less on average, demonstrating a sustained weight benefit from taking part in the FFIT programme. Meanwhile, men in the FFIT-FU-C group (who had the opportunity to take part in the FFIT programme under ‘routine delivery’ conditions immediately after the 12-month measures) lost 2.03 kg (95% CI 1.08 to 2.98 kg; p < 0.001) or 1.79% (95% CI 0.92% to 2.65%) of their baseline weight during the same time period. The mean between-group difference in weight trajectories was 4.62 kg (95% CI 3.26 to 5.99 kg; p < 0.001) or 4.23% (95% CI 3.02% to 5.43%; p < 0.001).

Figure 4 illustrates the data shown in Table 4, and clearly demonstrates that both groups weighed less at 3.5 years than at baseline.

FIGURE 4.

Mean weight (kg, 95% CI) of participants in the FFIT-FU-I and FFIT-FU-C groups at 12 months and 3.5 years after the FFIT RCT baseline measurements. Note that the y-axis [weight (kg)] does not start at zero.

Sensitivity analyses

Sensitivity analyses on the primary outcome (change in weight) were conducted to assess the sensitivity of the main analyses.

• Loss to follow-up sensitivity analyses used a variety of assumptions about the long-term weight outcomes of men who had not taken part in follow-up measures at 3.5 years.

• Baseline time points sensitivity analyses assessed the fact that both intervention and comparison groups had the opportunity to take part in the FFIT programme, but at different times.

In the loss to follow-up sensitivity analyses, the return-to-baseline and last-value-carried-forward methods were used to impute missing weight data at 12 months and 3.5 years for men who were not followed up. As men who were not followed up at 3.5 years (the not followed-up cohort) were, on average, heavier than the total followed-up cohort at baseline [112.8 kg (SD 17.2) vs. 107.8 kg (SD 17.1)] and 12 months [109.8 kg (SD 18.3) vs. 104.8 kg (SD 17.3)], the return-to-baseline sensitivity analysis is the most conservative and is reported here. The results of the last-value-carried-forward sensitivity analyses are shown in tables i and ii in Report Supplementary Material 7.

Table 7 shows that, in the return-to-baseline sensitivity analysis, the mean weight loss at 3.5 years was 1.81 kg (95% CI 1.09 to 2.52 kg) in the RCT intervention group (including imputed values) and 1.85 kg (1.12 to 2.58 kg) in the RCT comparison group (including imputed values). As in the main analyses, both figures were still significantly different from baseline, but there was no between-group difference.

Table 8 shows that, between 12 months and 3.5 years, men in the RCT intervention group (including imputed values) gained 3.15 kg (95% CI 2.37 to 3.92 kg; p < 0.001) and those in the RCT comparison group (including imputed values) lost 1.30 kg (95% CI 0.59 to 2.01 kg; p < 0.001). Although, as expected, there were slight differences in these values from the main analyses, the change in weight over time in each group remained significant, as did the between-group difference in weight trajectories (4.45 kg, 95% CI 3.40 to 5.50 kg; p < 0.001).

Full details of the baseline time points sensitivity analyses are provided in table iii in Report Supplementary Material 7 and confirm sustained weight loss in the FFIT-FU-I group at 3.5 years post intervention.

Objectively measured physical outcomes

Table 9 shows that there were sustained improvements in waist circumference, BMI, percentage body fat, and systolic and diastolic BP at 3.5 years and 2.5 years after taking part in the programme (for the FFIT-FU-I group and FFIT-FU-C group, respectively). There were no between-group differences.

Self-reported physical activity

Table 10 shows that the total PA was significantly higher at 3.5 years than at baseline in both the FFIT-FU-I group [800.0 MET-minutes per week, interquartile range (IQR) –120 to 2514 MET-minutes per week] and the FFIT-FU-C group (919.0 MET-minutes per week, IQR –186 to 2909 MET-minutes per week), and that there were no significant between-group differences. A similar pattern of results was observed for vigorous and moderate PA, and for walking. Table 10 also demonstrates a sustained reduction in time spent sitting in both groups.

Self-reported eating and alcohol intake

Table 11 shows that the consumption of fatty food and sugary food scores at 3.5 years were significantly reduced from baseline in both groups, and that there were no between-group differences. Fruit and vegetables consumption was significantly higher at 3.5 years in both groups, and, again, there were no between-group differences. Similar sustained improvements were also evident for portion sizes of cheese, meat, pasta and chips, and for alcohol consumption; all showed sustained reductions from baseline, and no between-group differences.

Self-reported psychological health and quality-of-life outcomes

Table 12 shows that improvements in self-esteem, positive and negative affect, and physical and mental HRQoL were sustained to 3.5 years in both groups, and that there were no between-group differences.

Randomised controlled trial secondary outcome trajectories

Table 13 shows significant differences in the trajectories of many RCT secondary outcomes from 12 months to 3.5 years between the FFIT-FU-I and the FFIT-FU-C groups. These are likely to reflect the fact that the FFIT-FU-C group had the opportunity to take part in the FFIT programme during this time (from month 13). The results for the FFIT-FU-C group, therefore, reflect both the changes men made as a result of taking part in the FFIT programme and the maintenance, or lack of maintenance, of these changes over the 2.5 years following the completion of the programme.

Taking each group separately, the FFIT-FU-I group showed some regain in waist circumference and BMI, and increases in systolic and diastolic BP between 12 months and 3.5 years (although these were all significantly improved from baseline), whereas the FFIT-FU-C group showed tendencies towards improvement in these outcomes over this time period.

In relation to self-reported PA, although the FFIT-FU-I group showed some reduction in total and vigorous PA between 12 months and 3.5 years, levels of moderate PA and walking appeared to be sustained. The FFIT-FU-C group showed improvements in total and moderate PA, and in walking. Between-group differences in moderate PA and in walking were not significant.

The FFIT-FU-I group also demonstrated sustained reductions in intake of fatty and sugary foods, portion sizes of cheese and meat, and total alcohol consumption from 12 months to 3.5 years, but consumption of fruit and vegetables decreased. The FFIT-FU-C group showed improvements in all dietary outcomes (apart from fruit and vegetables consumption) between 12 months and 3.5 years.

Finally, improvements in all psychological outcomes, apart from mental HRQoL, were sustained between 12 months and 3.5 years in the FFIT-FU-I group. The FFIT-FU-C group demonstrated improvements in self-esteem, positive affect and mental HRQoL during this period.

Baseline predictors of long-term weight loss

None of the prespecified baseline predictors (age, BMI, education level, socioeconomic status, marital status, orientation to masculine norms or number of long-standing illnesses) was significantly associated with weight loss at 3.5 years. For specific estimates and p-values, refer to tables 7.1–7.7 in Report Supplementary Material 8.

Mediators of long-term weight loss

Unlike the predictors, which are measured at baseline only, mediators of long-term weight loss were measured at multiple time points (12 weeks, 12 months and 3.5 years). Therefore, using these results to interpret the effect of each mediator on weight change, especially when the groups took part in the FFIT programme at different times, is not straightforward. Consequently, we performed multiple analyses at single time points. Mediators that were measured at 12 weeks and 12 months were analysed only in the FFIT-FU-I group, as the FFIT-FU-C group had yet to receive the intervention. The groups were combined for mediators measured at 3.5 years. The full results of the mediator analyses are presented in Report Supplementary Material 9. No 12-week variables (apart from objectively measured weight in the FFIT-FU-I group) showed any associations with long-term weight outcomes.

Table 14 shows that increases in self-reported total and vigorous PA, positive affect and physical HRQoL at 12 months all predict a lower weight at 3.5 years. In addition, increased total, vigorous and moderate PA, reduced intake of fatty and sugary foods, increased intake of fruit and vegetables, smaller portions of cheese, meat, pasta and chips, and increased positive affect, self-esteem and physical HRQoL at 3.5 years are associated with a lower weight at 3.5 years. Finally, increased time spent sitting at 3.5 years is associated with a higher weight at 3.5 years.

Table 15 shows that higher scores in relation to the self-determination theory constructs of autonomous regulation for diet and PA, perceived autonomy for diet and PA, perceived competence for diet and PA, and relatedness with peers (intimacy) and family members (acceptance), as well as higher satisfaction with current diet and PA, were associated with a lower weight at 3.5 years. In contrast, greater amotivation in relation to diet and PA was associated with a higher weight at 3.5 years.

Table 16 shows that self-reported routinisation of PA (walking, gym attendance and taking other forms of exercise), and healthy eating behaviours (eating regular meals, reducing portion sizes and reducing intake of fatty and sugary foods, sugary drinks and calories), self-monitoring of weight, reading the labels on food packaging, and ongoing contact with other participants and coaches from the FFIT programme were associated with a lower weight at 3.5 years.

Table 17 shows that greater weight loss at the end of the initial 12-week active phase of the FFIT programme (from objectively measured weight at 12-week RCT measurements in the FFIT-FU-I group, and self-reported retrospectively at 3.5 years by the FFIT-FU-C group) was associated with a lower weight at 3.5 years. Weight gain in the 12 months before starting the FFIT programme (from the RCT 12-month measurements in the FFIT-FU-C group only) was associated with a higher weight at 3.5 years.

Table 18 demonstrates that self-reported injuries that limit walking at 12 months predicted higher weight at 3.5 years, and that self-reported lower limb joint pain frequency, upper body and lower limb joint pain that limit activities, injuries that limit walking and using stairs, and total types of limitations due to injury were all positively associated with weight at 3.5 years.

Retention to the Football Fans in Training follow-up study

The use of the strategies that had been successful in achieving a retention rate of 92% at the 12-month RCT measurements44 resulted in the retention of 65% of the original RCT cohort at the 3.5-year follow-up study. There were some differences between men who took part in the 3.5-year measurements and those who did not. Men who were followed up were slightly older, more likely to be in paid work and more likely to own their home than those who were not followed up, perhaps reflecting a more stable lifestyle among those who were followed up. Men who took part in the 3.5-year measurements were also less overweight and had lower BP at baseline than those who did not take part in the follow-up study. Nevertheless, there were no differences in the RCT 12-month weight outcomes between men who were followed up and men who were not followed up, suggesting that success at maintaining weight loss to 12 months did not influence participation in the 3.5-year measures.

Long-term weight loss

Long-term weight trajectories

Between 12 months and 3.5 years, as expected, there were marked differences in weight trajectories between the groups. Men in the FFIT-FU-I group regained almost half of the weight they had lost (2.59 kg, 95% CI 1.61 to 3.58 kg) compared with men in the FFIT-FU-C group who lost 2.03 kg (95% CI 1.08 kg to 2.98 kg) over this period. This difference probably reflects the fact that men in the comparison group had the opportunity to take part in the FFIT programme at the start of this period, whereas those in the intervention group had had this opportunity 12 months earlier. As a result, the 12-month to 3.5-year weight trajectories of men in the FFIT-FU-C group reflect both their initial weight loss during the FFIT programme and its subsequent maintenance or lack of maintenance, whereas the weight trajectories of the FFIT-FU-I group simply reflect maintenance or lack of maintenance post programme.

Long-term secondary outcomes in the intervention group

Men in the FFIT-FU-I group showed sustained improvements in physical, behavioural and psychological outcomes at 3.5 years. Improvements in physical outcomes (in addition to weight) included sustained reductions in BMI, waist circumference, percentage body fat, and systolic and diastolic BP. Self-reported PA was higher at 3.5 years than at baseline and, on average, men sat less, continued to eat fewer fatty and sugary foods, continued to eat more fruit and vegetables, had smaller portion sizes and showed a sustained reduction in the amount of alcohol that they reported drinking. This group also showed sustained improvements in self-esteem, positive and negative affect, and physical and mental HRQoL at 3.5 years.

Long-term secondary outcomes in the comparison group

Men in the FFIT-FU-C group showed improvements from baseline in all secondary outcomes, and to levels similar to those seen in the FFIT-FU-I group. Thus, at 3.5 years, BMI, waist circumference, and systolic and diastolic BP were all reduced in the FFIT-FU-C group. Men reported increased PA and decreased sitting time, and also showed improvements in consumption of fatty and sugary foods, fruit and vegetables, alcohol intake and portion sizes. Self-esteem, positive and negative affect, and physical and mental HRQoL were also improved at 3.5 years.

Long-term secondary outcome trajectories

As with weight, men in the FFIT-FU-I group showed some attenuation of the impact that the FFIT programme had across other physical, behavioural and psychological outcomes over the long term, although all were still significantly better at 3.5 years than they had been at baseline. There was some regain of waist circumference and BMI, and increases in systolic and diastolic BP, as well as reductions in self-reported total and vigorous PA, fruit and vegetable consumption and mental HRQoL between 12 months and 3.5 years. Nevertheless, there was no significant reduction during this period in levels of moderate PA and walking. In relation to diet, men were still managing to limit their intake of sugary foods and alcohol and their portion sizes of cheese and meat. Finally, improvements in all psychological outcomes, apart from mental HRQoL, were sustained from 12 months to 3.5 years.

Again, as with weight, the long-term secondary outcome trajectories of the FFIT-FU-C group reflect both the initial impact of the FFIT programme and the subsequent maintenance, or lack of maintenance, of the changes made. This group showed significant improvements from 12 months to 3.5 years in BMI but not in waist circumference or BP (the lack of improvement in BP may reflect our protocol at the RCT baseline measures, whereby any man with an elevated BP recording was advised to consult his GP). Men in this group also demonstrated increases in self-reported total and moderate PA, and in walking. Dietary improvements between 12 months and 3.5 years included reductions in the consumption of fatty and sugary foods (but no significant changes in fruit and vegetables intake), in portion sizes of all key foods and in alcohol intake. Finally, men also showed improvements in self-esteem, positive affect and mental HRQoL between 12 months and 3.5 years.

Significant between-group differences in the trajectories of most of these secondary outcomes reflect the fact that many FFIT-FU-C men took part in the FFIT programme immediately after the 12-month measures (i.e. in autumn 2012).

Baseline predictors of successful long-term weight loss

None of the prespecified baseline predictors showed a significant relationship with weight loss at 3.5 years in either group.

Mediators of successful long-term weight loss

Self-monitoring, goal-setting and review

Many men spoke about continuing to monitor their weight. For some, this involved weighing themselves on a regular (often weekly), or semiregular, basis. There were some differences between the groups. Although few pre-FFIT achievers specifically described weighing themselves, a number of long-term maintainers said that self-weighing was a satisfying or enjoyable experience:

I love it. I absolutely love it. I still weigh myself every Thursday morning.

Interview 50, football club 02, long-term maintainer

A number of men, across all groups apart from the pre-FFIT achievers, spoke about monitoring their weight in relation to weight targets, although sometimes these targets were rather loosely defined:

I don’t really make goals noo [now]. One o’ [of] my aims is tae [to] lose the stone again, an’ [and] that’ll happen.

Interview 21, football club 12, long-term achiever

Other men were more specific, both in relation to weight loss targets and in relation to an upper threshold of weight that they did not want to breach:

If I got to the 15-stone marker, I’ll quite likely start to panic. But it’s actually, it’s getting a . . . It’s getting a sorta [sort of] target, a maximum target and saying ‘No further’. But quite honestly, I would like to do it in terms o’ getting down to about 13 and a half stone.

Interview 7, football club 05, delayed responder

By contrast, a few men (long-term regainers and delayed responders) viewed weighing themselves as a negative experience. Some long-term regainers said that it could trigger low moods, with one describing how the thought of finding out his weight filled him with trepidation:

An’ I don’t weigh myself or anything like that because I’m very – I feel bad that, you know, that, you know, I’ve not kept it up an’, you know, I’ve not got down tae my ideal weight an’ all the rest of it. [. . .] What I am now, I haven’t got the faintest idea ‘cause, as I say, I’m actually frightened tae weigh myself. I think it would throw me intae [into] a spiral o’ depression, you know, if I actually knew what my weight was, you know?

Interview 55, football club 02, long-term regainer

Men also described more indirect methods of monitoring their weight, in particular in relation to how their clothes fitted and changes in the sizes of clothing they had to buy. Some men used both weighing themselves and the fit of their clothes to monitor their weight, but others had come to rely solely on how their clothes felt:

Do you monitor your weight?

No’ [Not] as such but, you know, it’s just how I feel, you know in my clothes and stuff like that.

Right, yeah. So whether or not you still fit those jeans and things like that? Yep.

Yep, definitely.

OK. Which rung of the belt you’re . . .?

Exactly. Stuff like that. Whatever the weight checking something like that, that’s mair o’ [more of] a sensible way o’ judging it for me.

Interview 22, football club 07, long-term achiever

Many men also continued to monitor their PA, some by tracking their step counts, others by being aware of distances walked or time spent walking. A couple of men described monitoring the intensity of their PA through time taken to cover specific distances. Although many men had embraced using the pedometer during and immediately after the FFIT programme, only a few had continued to use it long term. Some had stopped using their pedometer when they lost it or when the battery ran out, and others described how they simply found that they no longer needed it:

For the first . . . I can’t remember exact, say the first 18 months or so, we were given a pedometer and I wore that faithfully, all the time and I did keep a close track on my steps. As time went on I realised I didn’t need the pedometer, I knew by the amount of, length of time I was spending walking how many steps I was taking and, and what sort of exercise it was. So I know that if I walk for a hundred minutes, I do 10,000 steps, you know that kinda [kind of] thing? Just at the pace that I go and what it does.

Interview 12, football club 05, long-term maintainer

Some of the men who spoke about monitoring their PA set themselves daily or weekly targets to aim for in relation to steps or distance. However, the review of these targets was sometimes rather infrequent:

I just do keep a record of . . . well, it’s on the phone actually, but I keep a record on this app [application] that I’ve got on the phone and just look at it every 2 or 3 weeks, see how many steps I’ve done, see if I’ve reached my goals an’ all that sort o’ stuff.

Interview 34, football club 12, long-term regainer

It was notable that only one man in the Pre-FFIT achievers group mentioned continuing to monitor his PA, and then only on occasion:

Well, I’ve got an app on my phone, so I can occasionally just kinda jump onto that and hae [have] a look and see how far I’ve walked in a day, or how many steps I’ve done, or whatever.

Sure. And is that something you use quite often? Or not so often?

Och [Oh], just now and again.

Interview 23, football club 13, pre-FFIT achiever

Men in all groups mentioned self-monitoring less frequently in relation to diet than to PA, perhaps reflecting the lack of emphasis on self-monitoring of food intake in the FFIT programme (men were asked to keep a food diary only at the start and the end of the programme). Nevertheless, some men did describe continuing to keep food diaries, monitoring their calorie intake or weighing their food:

I used to get up in the morning and just pour a bowl of cereal and, you know, it would just be a case of – it would be overflowing, if you like. But the information that we were given about the portion size, it’s usually on the product, that really helped me in the fact that I then got a set of scales and I would say a full one portion equals 30 grams, and I would get a, you know, I’d put my plate on the scales and then pour the 30 grams in there, and that would do me, you know?

[. . .]

Do you still go to that extent or . . .?

I do, yes. Yes.

Interview 14, football club 06, long-term maintainer

Some men (especially delayed responders) had adopted new technologies to monitor their weight, PA and/or diet. In addition to pedometer apps, technologies mentioned included the Fitbit (Fitbit, San Francisco, CA, USA), Apple Watch (Apple Inc., Cupertino, CA, USA) and Garmin watches (Garmin Ltd, Olathe, KS, USA), and fitness and running/cycling apps, such as MyFitnessPal [MyFitnessPal, Inc., San Francisco, CA, USA]. Often technologies were used to try to achieve targets in relation to steps, diet or weight:

I use that MyFitnessPal . . . to record what calories I’m taking on a day-to-day basis. However, I give myself a 1400 calorie target Monday through Thursday. Weekends, I’m a bit more relaxed aboot [about] that, but if I sustain the 1400 Monday through Thursday, in conjunction with the kind o’ exercise programme, that seems tae work for me an’ I can still lose a pound or a couple o’ pounds, you know, every couple o’ weeks.

Interview 41, football club 04, delayed responder

One long-term achiever described how he and his wife used monitoring technologies to support each other in achieving their step targets:

I bought one [a Fitbit] and my wife bought one, so, we have healthy competition there. And it makes it fun with the app on the phone and whatnot, you know? And we can get a bit of fun banter going on there, you know? But, yeah, I’m not consciously checking the step count really, you know? When it buzzes on my wrist I know that that’s 10,000 steps for the day kind of thing, you know? And then, yeah [yes], I kind of – you know, at the end of the week we, kind of, compare our results and look at the trend in them and whatnot, you know?

Interview 9, football club 13, long-term achiever

Strategies used to remain physically active

The majority of men reported continuing to walk to stay more active after the FFIT programme, often describing how they were still managing to incorporate walking into their daily routine. Some talked about walking for leisure, including when listening to music or football commentaries on the radio, or with their wife, or simply because it gave them something to do:

I do much, much more walking, without a doubt. And think about exercise. I’m not saying I do exercise – I still do, evidently – but, yeah, there’s more thought in ‘Right, what am I doing? I’m sitting here doing nothing. Och, let’s go for a walk’.

Interview 31, football club 11, long-term achiever

Nevertheless, many spoke about purposeful walking, including walking to get the paper, walking instead of taking the car and walking their dogs (a couple had even bought dogs to encourage them to walk more):

. . . certainly, the exercise side o’ things, I mean, because I have a dog now, you know . . . well, I bought a dog when I retired because I always wanted a dog but we were working full-time it wasnae [wasn’t] very practical. But getting a dog was also part o’ the reason tae give me a good reason for going out an’ getting out every day.

Interview 64, football club 11, long-term achiever

Fewer pre-FFIT achievers mentioned walking compared with men in other groups, and although long-term regainers did report walking, their descriptions of what they did when walking often seemed vaguer and appeared to demonstrate less commitment to walking regularly than the groups that had been successful in losing weight long term (long-term maintainers, long-term achievers and delayed responders):

I try to go out walking every day with the dog for a good . . . for a decent hike, although I’ve been a bit slack recently.

Interview 65, football club 10, long-term regainer

Many men talked about how they had taken up other activities after the FFIT programme, sometimes instead of, but often in addition to, walking. Popular choices included cycling, running and going to the gym:

I never cycled during the programme really. I toyed with the idea of cycling tae work an’ got as far as applying for cycle tae work scheme but that was about the time o’ my redundancy, so that fell through. But the one treat I bought myself wi’ [with] my redundancy cheque was a cycle, you know? An’ I’ve actually got two now, so . . . an’ that’s my tool if I overeat now, I go out on the bike for half an hour tae get my calorie count back down.

Sure. And is that something you do quite regularly now then?

Yeah. Minimum of twice a week I’m out on the bike. Yeah.

Interview 4, football club 02, delayed responder

One long-term maintainer, a taxi driver, described how, for him, running was a more convenient way than walking of fitting PA around his work commitments:

So, whereas before you were going out walking, let’s be honest, walking’s quite time-consuming, and to do it . . . You know, for me anyway, because I wasn’t doing much in the taxi, it would maybe take an hour and a half each night to get up to that sort of level of exercise in terms of walking and that’s –

OK, because you’d be getting all your steps going in one go?

Exactly. Exactly. So now I’m maybe going running three times a week. You’ll still go for a walk now and again, but it’s not that sort of specific measuring it, whereas I go for a run three times a week.

Interview 13, football club 07, long-term maintainer

Another long-term maintainer described how, after a period of inactivity caused by stress at work, taking up one form of PA (going to the gym) had acted as a catalyst for him to do other things as well:

So, basically I’ve been stuck in the office, it was all, kind of, it was quite stressful working so long, and after the year, you know, you kind of get stuck in a sort of rut, if you like. And I just suddenly realised that the weight was slowly going back on, but not in a great extent, you know? I just thought ‘Right, I’m gonna [going to] have to do something else.’ And, so, we, kind of, joined a different gym and, yeah, that’s the way it’s, kind of, coming off again.

So, again you’re, sort of, going to the gym rather than walking. Is that . . .?

I am walking as well now.

Interview 14, football club 06, long-term maintainer

Some men also spoke about doing exercise programmes at home to help them maintain their PA levels. One long-term maintainer described how he turned to the exercises that he had been shown during the FFIT programme when the weather prevented him from going outside:

I think they used sort of bean bags, you know for stretching exercises that you can do like in the living room. You know in your own house. So sometimes, yeah, sometimes I do that. Like say if it’s inclement weather and I think, ‘Oh God, I’m no’ going out in that’. You know, so just throw a cushion on the floor and do some exercises inside.

Interview 29, football club 09, long-term maintainer

Another man (a delayed responder), who had struggled with his weight after the FFIT programme before joining a slimming club to help him overcome his addiction to food, explained in the interview how he had bought some gym equipment for home so he could exercise without feeling embarrassed in front of other people:

I also in my spare room got some . . . got an exercise . . . piece of exercise equipment. That meant I didn’t have the self-consciousness of going to the gym. I could exercise in private so to speak.

Interview 5, football club 01, delayed responder

Men from all groups described playing football. Sometimes (but not always) this was with other men who had taken part in the FFIT programme (in some football clubs ‘follow-on’ groups were formed at the end of the intervention so men could meet up to play football together). Nevertheless, some men were reluctant to play football, especially when it became too competitive. One long-term regainer described how, when play had become too serious, he had stopped attending the ‘follow-on’ group at his football club:

There were people coming intae [into] the group that didnae [didn’t] know what we were all about. They were just coming for the football. And that, we came off the cinder pitch which I loved, you know, and I couldnae [couldn’t] care if it was snowing, it didnae make any difference. If you’re playing on a cinder pitch nobody’s gonnae [going to] be throwing themselves intae challenges. [. . .] But when we got to the [professional sports facilities], the people were coming in, were coming flying off their feet and there was a lotta [lot of] people, there’s a lotta people admitted, if they sat down and admitted it, they got injuries. Injuries that they shouldn’t have had, and it wasnae caused by them, it was caused by others, you know? No, that’s a thing and then they started playing against other FFIT clubs, it got a bit too serious as well. But I suppose maybe I was looking for an excuse to end it, and I didnae get a choice [laughs], you know.

Interview 58, football club 13, long-term regainer

There was some evidence that long-term regainers may have been doing fewer forms of PA (instead of, or in addition to, walking) than men in the groups that had been more successful at achieving long-term weight loss (long-term maintainers, long-term achievers and delayed responders).

Finally, a number of men described taking part in events or challenges to give themselves a goal to train for. These included 5-km and 10-km runs (e.g. a weekly Park Run), cycling races, triathlons, charity walks and climbing mountains. In general, long-term maintainers appeared to push themselves further in the types of challenges they undertook than men in the other groups:

I like to have one event or two events on the calendar that I know I’ve got to step up to or I’ll fail it, and I don’t want to fail it, so . . .

So having targets helps. So having events and then I suppose working out the . . . I was going to say the steps, but yeah, the steps it takes to kind of be prepared for that?

Yeah, exactly, if you’ve got a 10-km run to do you’re gonnae make sure that you can do 10 km, do you know what I mean? You’re not gonnae turn up only being able to do 5 km, you’re gonnae do it all sort of thing.

Interview 13, football club 07, long-term maintainer

Strategies used for healthy eating and drinking

Almost all men across all groups described how they were continuing to make healthier choices in relation to food and drink. Strategies used included eating more fruit and vegetables; eating fewer processed foods and takeaways; reducing fat, sugar and salt consumption; switching from sugary, fizzy drinks to diet drinks, diluting juice or water; and taking food to work instead of buying from a canteen, vending machine or fast food outlet:

I have considerably cut down on a lot of things I shouldn’t have been eating and have been eating since I was a boy, notably fry-ups, very much reduced that at home. Significantly reduced takeaways. And eating a bit more fruit and veg [vegetables]. Maybe not quite as good at that over the last little period but yes, I have taken, I take, like, grapes and apples and oranges now that I just never bothered with before.

Interview 47, football club 08, long-term achiever

The changes I kinda [kind of] maist [most] stopped doing was I basically stopped eating . . . at my work, with . . . were using the facilities on site, basically because it was . . . You know, you’re having the roll and sausage for your breakfast, and having the macaroni cheese and chips at lunchtime and stuff. And I stopped that and . . . To this day I still go in there and have, you know, I have, like, fruit for breakfast and take in a . . . a salad that I make up for my lunch. So I’ve kind of got into that habit. So I’ve changed, definitely changed the, you know, at least a couple of meals when I’m in a working routine [. . .] And I kinda kept going wi’ that from . . . from day one of the course to still doing it now, you know, almost 4 years later.

Interview 54, football club 11, long-term regainer

The majority of the men (including most of the pre-FFIT achievers) still spoke about the importance of being aware of portion sizes. This mirrors the impact that the discussion of the Eatwell Plate and portion sizes in week 2 of the FFIT programme had on many men in the RCT 12-week and 12-month focus groups, who described it as powerful. 44 In the 3.5-year interviews, some men described strategies that they still used, such as weighing scales or smaller plates, to help them control portion size:

One of the changes, believe it or not, that I made was, I had bought these plates from Ikea, and they were actually huge plates, to the extent I had to actually cut a little bit out of the back of my cupboards so they’d fit in. When I met my partner, we had both talked about the portion size thing, and she’s obviously quite conscious of her weight and stuff, and we ended up giving the plates away, going to Ikea, and buying a smaller set [. . .] that was a big change, you know, it definitely made a difference because you have a smaller plate, you don’t put as much food on it, you know?

Interview 2, football club 12, pre-FFIT achiever

Across all groups, men talked about changing their pattern or intake of (unhealthy) snacks, reflecting another key message from the Eatwell Plate discussion in week 2 of the FFIT programme. Strategies used included eating fewer, or healthier, snacks between meals, and eating unhealthy snacks only from time to time and in smaller quantities. Some of the men, including long-term regainers and pre-FFIT achievers, described not allowing themselves to eat snacks after a certain time in the evening:

I don’t stop myself from eating what I want to eat but I do it more in a regulated way. If I was sitting and I say, ‘Right, I’m gonna have a packet o’ crisps,’ I have a packet of crisps, where instead of opening four packets an’ putting them in a bowl (because it doesn’t look like four packets when it’s in a bowl, it’s easier tae eat then) and then eat that whole bowl, I’ll get one packet o’ crisps and have that packet o’ crisps.

Interview 46, football club 12, delayed responder

There’s very few times now that my wife and I will eat after 9 o’clock tae half past 9 p.m. So, you know, I found that out in the programme and that’s been pretty easy, to be honest, because the odd time we would’ve been eating later, but not now. An’ even when we go away on holiday it’s sort of unwritten law that we just don’t eat or drink anything after half past nine.

Interview 34, football club 12, long-term regainer

Reflecting the food labelling activity in week 8 of the FFIT programme, a few men in each group spoke about continuing to read food labels, especially to monitor fat and sugar content:

I think obviously the eating part of it made us more aware of the fat content in food and stuff, so obviously that’s always a thing when I’m buying stuff now. You tend to look twice, whereas before you’d just be ‘Och!’, pick up any old sandwich and things, and maybe get a wee [little] bit too obsessive about it at times. But you’re obviously looking at the labels now and trying maybe get one that’s a wee bit less fattening, or [less] saturated fat or whatever.

Interview 53, football club 13, long-term regainer

Men also spoke about continuing to limit their alcohol intake, reflecting what they had discussed in week 5 of the FFIT programme. Some simply described cutting back on the amount of alcohol they consumed; others said that they had virtually eliminated alcoholic drinks. A couple of long-term achievers said that they had reduced the amount of alcohol they drank by going out less. However, the most commonly reported strategy was to limit drinking alcohol at home:

I used to have a drink every Saturday in life whether I was in the house or whether I was out. Now I don’t even drink in the house at all now so . . . I have to be going out now if I’m having a drink, but, as I say, I just don’t, I don’t touch alcohol at all now really . . . or I very seldom do. So that’s been a change for the good as well.

Interview 54, football club 11, long-term regainer

The 12-month focus groups had suggested that a flexible approach to healthy eating (e.g. men allowing themselves to eat unhealthy foods on occasion) was associated with successful weight loss. 44 In the 3.5-year interviews, men in all groups described continuing to adopt a flexible approach, in particular having a more relaxed approach to healthy eating at the weekend or when on holiday. Nevertheless, there was a notable difference in the way that this flexible approach was described between men who were successful in maintaining their weight loss (long-term maintainers, long-term achievers and some delayed responders) and those who were not (e.g. the long-term regainers). Those who were successful tended to use positive terminology (e.g. enjoying themselves while on holidays and at weekends) and often appeared to have clear strategies for ‘damage limitation’, as illustrated by this man’s account of his approach to holidays:

. . . even when I’m on holiday, I’m watching what I’m eating, you know? We went on our first ever all-inclusive holiday back in March, just for 7 days (and the things that are there to tempt you, you know, like the food and the alcohol and stuff like that). But we, we went walking every day and [. . .] went to the gym a couple of times when I was there. Yeah, and I did have a couple o’ things, you know, that maybe I shouldnae [shouldn’t] have eaten, but when I came back and went on the scales, I’d only put on a pound, so you know, whereas if I hadn’t, you know, been aware that I need to go walking, I need to go to the gym, I need to get a bit of exercise in here, and for lunch, let’s just have a little bit of seafood and salad and some fruit, you know? So . . . and that’s the way you think, you know?

Interview 49, football club 10, long-term achiever

By contrast, men who were less successful in maintaining weight loss spoke more negatively and often demonstrated a less tempered attitude to maintaining a healthy lifestyle while on holiday:

I’m just back from a sort of city break abroad an’, well, most o’ the diet went out the window then . . .

Interview 34, football club 12, long-term regainer

Barriers to and facilitators of sustained weight loss, physical activity and healthy eating

In the 12-month focus groups, common barriers to maintaining weight loss included injury and health problems, changes at work, holidays, festivities, poor weather, life events, and the costs associated with leading a healthy lifestyle. 44 The 3.5-year interviews revealed that many of these barriers continued to be issues over the longer term.

Routinisation of physical activity and healthy eating

Throughout the FFIT programme, men were encouraged to make changes that suited their own eating and exercise preferences, and to try to incorporate them into everyday life. Men from all groups, apart from the pre-FFIT achievers, described how PA had become part of their daily routine. A number of long-term regainers, despite being less successful in long-term weight loss, still valued the fact that the FFIT programme had introduced regular PA into their lives:

So I’m sitting at a desk all day, you know? Eight, nine, ten hours. So that was the kind o’ scenario before I went on the course. After that, I’ve obviously I’m still doing the same kind o’ work, still working from home, but every lunch-time I now do a mile-round walk to go an’ buy my paper, as advertised on the television programme. Still doing that every day, primarily because it gets me out the house, get some fresh air, and forces me to take a longer break at lunch-time.

Interview 37, football club 02, long-term regainer

Men also spoke about how they continued to embrace the importance of eating regular meals (including breakfast), which was discussed in week 8 of the FFIT programme. Some described the importance of planning and organisation to help them achieve this:

It might sound a bit boring, but I now plan what I eat, what I will eat for the next day. You know, for example, I know exactly what I will eat tomorrow. I try and plan a menu. I find this helps me. I don’t know if it helps everybody, but I think it helps me. I try tae plan my eating for the week, so I’ll know what I’m gonna have Monday, Tuesday, Wednesday, Thursday, Friday. I have a bit of a blow out on a Saturday and a Sunday, but not to a great extent. But I do try tae, Monday tae Friday, I do try to plan what I’m gonna be eating, rather than just, you know, making decisions on the day. I feel for me personally it helps to plan the, to plan my programme for myself.

Interview 39, football club 13, delayed responder

Men across all groups spoke about the danger (and often the experience) of slipping back into old habits. Nevertheless, those who had been successful in long-term weight loss (long-term maintainers, long-term achievers and delayed responders) often appeared to recognise when old habits were re-emerging, and described the strategies they used to prevent these becoming established in their daily lives. For example, one long-term achiever described drawing on the knowledge and skills that he had learned during the FFIT programme to prevent him adopting old (unhealthy) eating habits when injury stopped him from following his normal PA routine:

I’ve slipped back into a couple of wee bad eating habits since I’ve been injured [. . .] but I know, I’ve identified them, you know, I know that ‘Right, that needs to get knocked on the head’. So, just a couple of bad habits [. . .] I’ve got a bad habit of eating a sandwich maybe about half 8, 9 o’clock [. . .] I used to have an 8 o’clock cut-off time for eating.

Interview 59, football club 04, long-term achiever

Men who were less successful at sustaining weight loss (long-term regainers and pre-FFIT achievers) seemed to take a more laissez-faire approach to the re-emergence of old habits:

I didn’t have any cheese whatsoever through the whole period [FFIT programme], you know? I mean, I’m not saying cheese is a big thing for me, it’s more chocolate and sweet things actually. So I hadn’t had them and I was sort of the belief that I’m OK as long as I don’t try . . . you know, it’s a bit like an addict with a drink or something like that, an alcoholic. Once I’ve tried some, that’s that, you know it’s . . . I can’t sort of rationalise that and say ‘Well, OK, that’s fine’. I say that, ‘Well, that’s it’. You know, I’ve . . . an’ then I’m a slippery slope again you see . . .

Interview 60, football club 11, long-term regainer

This section shows that men in all groups had retained some of the key messages around PA and healthy eating that they had learned during the FFIT programme over the long term. However, men in the more successful groups (in particular the long-term maintainers and long-term achievers) spoke in ways that suggested that they were more committed and disciplined about continuing to put these messages into practice than groups with less success in achieving long-term weight loss. For example, men in the pre-FFIT achievers group spoke less often about enacting the key healthy lifestyle advice from the FFIT programme, and men in the long-term regainers group were sometimes vague in their descriptions of how they put the tools and strategies from the FFIT programme into practice. Indeed, men in all groups spoke about facing continued challenges in maintaining their weight loss long term. The next section describes, in more detail, the reasons men gave for feeling motivated to continue to implement the changes they had made during the FFIT programme over the long term.

Externalised regulation (doing things because of an external demand or reward)

Men in all groups mentioned the role of external factors in influencing their efforts to lead a healthy lifestyle. Some told how family members reminded them of what they should and should not do, particularly in relation to healthy eating. One pre-FFIT achiever said that his partner had played a pivotal role in encouraging him to eat more healthily:

. . . being with my partner definitely helped me keep those changes up ’cause she’s there to keep me in check, you know? Whereas before I’d be, as I say, because you’re single and, when you work a long day, you come home, you just sometimes can’t be bothered cooking for yourself, and it’s certainly a lot easier when you’ve got somebody tae motivate you.

Interview 2, football club 12, pre-FFIT achiever

Other men described how discussions with health professionals about managing long-term health conditions, such as type 2 diabetes, hypertension and rheumatoid arthritis, had motivated them to try to keep their weight down:

I was certainly told by the doctor I really needed to lose about 5 stone. And, well, probably since I done the Football Fans in Training, it’s came down just over 2 stone, so . . . But they says ‘Don’t worry about how long it takes, just get there’, basically.

Interview 32, football club 08, delayed responder (diagnosed with type 2 diabetes since taking part in the FFIT programme)

Some men also described how compliments from other people acted as a reward that served to motivate them:

Oh, absolutely makes you feel a whole lot better about yourself and, like I say, if you’re getting compliments from people and stuff, you know, that’s gonnae help your self-esteem and make you feel better. There’s no doubt about that. Absolutely, you know, and it also keeps you motivated, you know, and . . . everybody’s the same, you know? If people say nice things about you, you’re gonna – it’s nice to have that, you know? But I think the challenge is to, rather than let that go over your head, is to keep it inside, and that keeps you motivated as well, you know? So it’s about learning to accept compliments. Before, you maybe werenae very good at that because it was about the way you looked and you knew that you were overweight and, you know, so it’s about keeping those inside you and using that as a continual motivation.

Interview 49, football club 10, long-term achiever

Introjected regulation (doing things because you feel you ‘should’)

Men from all groups described how they tried to manage their weight and lead a healthy lifestyle because they knew that they ‘should’ (e.g. they knew that it was good for their health), or to avoid negative emotions (e.g. embarrassment, guilt) or attain positive emotions (e.g. pride). Many said that they were aware of the health benefits of sustaining weight loss and leading a healthy lifestyle. Some reported that experiencing tangible benefits to their health and well-being, such as being able to reduce their medication, and improvements in back pain, quality of sleep and stamina, had spurred them on:

I suffer from high blood pressure, I had this at the start, and after the first 12 weeks of working there [on the FFIT programme], my – I went from hypertension just to high blood pressure. Now that’s fantastic as well, so my medication for that got reduced, and I really do believe that that’s been a direct impact or a combination of the exercise that we were doing there, and the eating habits that I had as well.

Interview 68, football club 03, long-term maintainer

Men described how the experience of negative emotions in relation to being overweight motivated them to keep the changes going. Some men recalled occasions, before commencing the FFIT programme, when they had felt embarrassed about their bodies, and they used these experiences to measure their progress:

[I]t’s lots of little things that make, you know, that you plan ahead, because you know you’re going to be embarrassed if such and such happens because you’re, you know, if, I always take my grandson to the [football] game or because if some other adult sat next to me [before he took part in the FFIT programme] we’d be struggling for space, ‘cause you know, my bum’s too big for the seat sort of thing, you know? And it’s these little things that . . . that make you want to . . . You know, lose weight.

Yeah, and do you find that that has changed for you?

Yes, aye, I can now, I can now sit on a seat myself [laugh].

Interview 59, football club 04, long-term achiever

Men also reported feeling disappointed and guilty when the new healthier PA and dietary routines that they had established were thwarted. For example, this long-term maintainer spoke about his negative emotional reaction to disruption to his normal PA routine:

I’ve kept it off, yeah, so, it’s a battle all the time just to keep it off, eh? It’s just – but that’s it, you just keep going. In fact, I feel – if I miss [an exercise] class now, I feel guilty.

Interview 51, football club 11, long-term maintainer

Other men were clear that negative emotions were unhelpful, and described how they sought to avoid them. One delayed responder told how the principles of the energy balance (discussed in week 11 of the FFIT programme) helped him keep a positive perspective on leading a healthy lifestyle:

I don’t treat the training as a chore and I don’t treat the eating as a sin. ‘Oh if I eat this, oh I’ve gottae [got to] do an extra mile, if I eat that.’ And I don’t hold back, if I want a packet o’ crisps, I’ll eat a packet o’ crisps, but I know that’s it. I don’t treat it where I feel to be guilty in doing any of it, which is good because I don’t feel guilty, I get on wi’ what I’m doing. I’ve been on these diets before, like Slimming World where you’ve gottae count this, and it’s a red day and a green day, and you cannae [can’t] eat this this day and you cannae eat that that day, and you can’t do this and you can’t do that, and you sit and you say tae yourself, ‘Oh I could really go this, I could really go that’. Where I just do it and work it off.

Interview 46, football club 12, delayed responder

In relation to positive emotions, pride in the changes they had succeeded in making was reported by men from all groups. One long-term maintainer described how losing weight helped him to feel good about the way he looked:

Well, I never used to bother aboot myself. I wouldnae [wouldn’t] get done up, if you know what I mean, to go oot. I wouldnae bother aboot shaving or anything, ken [you know]? Because you didnae think much o’ yourself before you started this course. And it makes you think a bit more about yourself when you’re . . . after you’ve done the course. And, sort of, didnae like going oot withoot [without] shaving or that, but as I say, I dinnae like looking as if I’m fat. That’s half the battle, if I dinnae look fat, I’ll feel better.

Interview 61, football club 08, long-term maintainer

Men across all groups described experiencing ups and downs in relation to sustaining their weight loss long term. For example, one delayed responder said ‘since the programme there’s been peaks an’ troughs’ (interview 41, football club 04, delayed responder). Weight regain was often accompanied by a range of negative emotions, including embarrassment, disappointment and guilt:

When I started to put weight back on, I felt guilty. I felt guilty that people had put me forward for this, or accepted me for it, and almost to a degree I’d let them down, I’d let myself down. And then you feel good when the weight starts to come back off. So . . .

Interview 17, football club 12, long-term regainer

Identified regulation (doing things because they are personally important)

Men from all groups spoke about how improvements in weight and lifestyle had enabled them to do things they valued and enjoyed, and that were personally important to them. These included playing or coaching football, doing more with their children, being able to do activities they enjoyed (e.g. hillwalking, deerstalking) more easily, or simply having more of an appetite for life:

I mentioned earlier about playing five-a-side, my son’s playing five-a-sides with me now, he – up until about 2 years ago he would be in company and, my friends would be saying aboot, ‘Oh your dad used to play football’. And he’s probably looking tae me, gaun [looking at me, going], ‘Nah [no] this wee dumpy guy here? Played football? At my level?’. And now we actually play five a-sides together. And I think he appreciates that actually I can kick a ball and play reasonable football. So that’s quite a good thing, it’s quite a good wee bonding thing as well.

Interview 62, football club 12, long-term maintainer

Although many men spoke about feelings of enjoyment in relation to PA, some also described enjoyment from eating a healthier diet and smaller portion sizes:

I enjoy cooking as it is, so the foods that I’m making for myself, it’s stuff that I enjoy eating but it’s also healthy stuff. Fish, pasta, rice, stuff like that that I like to cook. And obviously again portion control, I’m now enjoying what I’ve got, like I’m no’ wanting more, if you know what I mean. Or I’m no’ cooking too much that I’ve ate so much o’ it, an’ I cannae eat the rest. So I’m no’ wasting it, you know. So, but that’s a good thing that way.

Interview 40, football club 01, delayed responder

The preceding sections suggest that men in all groups (including those who had not sustained weight loss long term, i.e. long-term regainers and pre-FFIT achievers) had succeeded to some extent in internalising regulation for maintaining weight loss and positive behaviour changes. One long-term maintainer illustrated his shift from external regulation towards a more internalised form of regulation by describing how his weight loss was initially motivated by a desire not to be obese, a term that he saw as a ‘wake-up call’ when applied to him (at the RCT baseline measures). However, after successfully losing weight during the FFIT programme, he was now motivated to maintain the changes simply because he enjoyed his new lifestyle:

. . . now I hardly feel an incentive’s needed because I’m so much enjoying what we’re doing, it’s not a chore at all.

Interview 12, football club 05, long-term maintainer

According to self-determination theory, feelings of competence and relatedness to other people support this process of internalisation of regulation. 21 The following sections examine how competence and relatedness were portrayed by men in the 3.5-year interviews.

Competence

Men in all groups described feelings of competence and achievement in relation to weight control, PA, eating healthily and/or managing their health. Some men spoke confidently about their ability to use PA and diet to lose weight when they wanted to:

I feel as if I’m a lot more control over what I do, ’cause my wife was saying to me ‘Right we’ve got somebody coming up in a couple of weeks’. And I know I can, without being in some sort of like starvation, I know, if I think, I can just change my ways any time I like. And before I wouldnae really know what to do, to get that kind of weight loss or that fitness up, and just being really stupid and no’ eating for 2 days or something. So I now know that there’s nutritionist [nutritional] food oot there that I can make which isn’t bad for me. And I can cut oot a few things and do a wee bit of exercise, and I’m back on track again. So from that point of view it’s gave me, it’s gave me that wee bit of ownership of what I can do and what I cannae do.

Interview 62, football club 12, long-term maintainer

For most men, success in losing weight was accompanied by a real sense of achievement. Some illustrated this through accounts of being able to wear smaller-sized clothes:

Throwing oot the 40-plus waist – the 40-inch – sometimes it was almost a 42-inch waist, and throwing out those trousers and you know, gradually coming down. That’s the sort of thing that encouraged me, and also you know, when you go and buy the new trousers and everything, you don’t want to throw them out and go back to big stuff. So yeah, that’s encouraged me quite a lot. I’ve came down from like a two-XL to a large. And I’m still 36-inch waist, I’ve no managed to get below that, but it’s that . . . that’s encouraged me.

Interview 29, football club 09, long-term maintainer

Some men from all groups, apart from the pre-FFIT achievers, spoke about feeling competent and experiencing a sense of achievement in relation to PA, in particular when seeing improvements in their fitness and ability to run or play sports. One delayed responder described his satisfaction at now being able to play football with men who were much younger than him:

A minimum of twice a week I’m playing football with different groups. This is all through guys I’ve met through Football Fans in Training. I’m able to play football with guys that weren’t in Football Fans in Training because I’ve got healthier and I’m able tae keep up wi’ the younger kids, as it were.

Interview 4, football club 02, delayed responder

In relation to healthy eating, some men (again from all groups apart from the pre-FFIT achievers) described their skill in using PA to control their eating between meals, their ability to exert self-control over their eating habits or their competence in cooking healthy meals:

And we still do a lottae [lot of] the sortae [sort of] the cooking now, like I’m just standing here cooking my dinner for this evening which is made like bolognaise fae [from] scratch, ken. It’s no’ done wi’ like jars o’ sauces and things like that, ken; so a lot o’ our cooking techniques have totally changed.

Interview 38, football club 08, delayed responder

Finally, one delayed responder described how losing weight had allowed him to self-manage a long-standing back condition:

It turned out I had sciatica on the back o’ that, but because I’ve lost weight now it’s not bothering me any more. So it’s that kind o’ thing, you know? It’s made me think you can look after yourself without having tae go tae the doctor. I mean, the doctor tells you you’re overweight but you just think to yourself ‘Yeah, yeah, you’ve got tae say that’. But, no, it’s, the proof’s in the pudding.

Interview 4, football club 02, delayed responder

Relationships with other people

Subobjective 1: behaviour change techniques and practical strategies for controlling weight long term

The FFIT programme was designed to provide men with a toolbox of BCTs and practical strategies that they could choose from to help them incorporate PA and healthy eating (and drinking) into their normal daily routine. 46 Table 20 provides a summary of the BCTs and strategies that men across all 3.5-year interview groups said that they continued to use. However, there were some differences between groups, particularly in relation to PA. Men from the pre-FFIT achievers group were less likely to use self-monitoring, and did not appear to walk as much as men in other groups. Long-term regainers also seemed less committed to walking, and to engaging in fewer alternative forms of PA than some of the other groups.

Similar to the accounts of behavioural maintenance in the RCT 12-month focus groups, 3.5-year interviewees described how PA (often walking) and eating regular meals had become part of their everyday routine, although the pre-FFIT achievers spoke about this less in relation to walking than those in the other groups. However, the men’s accounts suggested that work was sometimes required to maintain their new healthy habits. It also appeared that those who were succeeding in controlling their weight long term were better at avoiding slipping back into old (bad) habits and tended to take a more positive attitude in relation to facing up to (and often overcoming) challenges.

Again, as in the 12-month focus groups, some men described how health problems, injury, work and family commitments had prevented them from pursuing the new PA and dietary behaviours that they had adopted as a result of taking part in the FFIT programme. Similarly, changes in jobs or working practices were described, particularly by long-term regainers, as being disruptive to newly adopted PA and healthy eating routines. Seasonal factors, such as Christmas and holidays, were also described as disrupting normal routine, as well as providing temptation and a licence to consume unhealthy and/or more food and drink. Poor weather continued to be a challenge and a negative influence on men’s intentions to be physically active, and the outdoor physical environment (e.g. traffic on roads) was also cited by a few men as a reason for being unable to engage in some forms of PA.

Subobjective 2: participant experiences of motivation

According to self-determination theory,20,21 internalised regulation is associated with behavioural maintenance. External factors (e.g. health professionals, family members or friends) continued to be important influences on behaviour for some interviewees. However, men from all groups also described more internalised forms of regulation, including pursuing weight loss, PA and healthy eating behaviours because of the health benefits of doing so, or to avoid embarrassment, guilt or disappointment. Many men across the groups spoke about their pleasure in being able to do things they valued or enjoyed or that were personally important to them, including being able to do more things with their children or grandchildren. Some men also described moving from an external form of regulation (e.g. being told that they were obese at the RCT baseline measures) to now leading a healthier lifestyle because they were enjoying the changes that they had made.

Relationships with other people featured widely in the men’s accounts. Many described feeling supported by their family and doing more things together, but some (including those who were less successful in long-term weight loss) complained that lack of support and involvement from family members threatened to undermine their efforts. Friends (including FFIT participants) also played an important role in the men’s efforts to maintain the changes they had made. However, some men, often those who had been unsuccessful in controlling their weight in the long term, had been unable to attend ‘follow-on groups’ that were set up after the FFIT programme, for different reasons. A few men described how the changes they had made had allowed them to become a positive role model for others. This is consistent with practice theory, which describes how ‘novices’ learn experientially and assimilate new behaviours into their routinised practice to become ‘full practitioners’, and predicts that some people will become ‘carriers of practice’ and take pleasure in influencing the behaviour (e.g. PA) of their family and friends. 75

Subobjective 3: acquisition of changed identities

Self-determination theory20,21 suggests that people who achieve fully integrated regulation in relation to weight loss, PA and/or healthy eating will be more successful in maintaining change. This was borne out in the 3.5-year interviews, in which accounts of viewing themselves as different from the person they had been before the FFIT programme featured most clearly in the accounts of men who had succeeded in long-term weight loss. Some of these men appeared to have distanced themselves from the (unhealthy, unfit or overweight) person they had previously been, seeing themselves now by contrast as a ‘new’ man who had succeeded in fully embracing a new, healthy lifestyle.

Resource use and costs

The cost of providing the FFIT programme in the 13 SPFL clubs in the RCT was estimated to be £61,700, which is equivalent to £164 per FFIT participant. 44 In addition to the intervention costs, self-reported data on the number and type of any NHS resources used in the preceding 12-week period were collected at all time points (baseline, 12 weeks, 12 months and 3.5 years) from each participant. This included visits to the GP, practice nurse or physiotherapist, and any attendances at accident and emergency departments. Unit costs for each of these visits were taken from the Personal Social Services Research Unit, using 2011/12 costs in the RCT79 and 2014/15 costs in the follow-up study at 3.5 years80 (see table i, Report Supplementary Material 12). We used 12 weeks as the time frame over which participants were asked to recall their health resource use, as longer periods have been found to be potentially subject to greater recall bias. 81 Self-reported data on inpatient stays and outpatient appointments over the preceding 12 weeks were also collected at all time points. The unit costs, which were taken from Information Services Division Scotland tariffs82 for 2012 for RCT data and 2015 for 3.5-year follow-up data, and, when necessary, NHS Reference Costs for 2011–12 (RCT)83 and 2014–1584 (3.5-year follow-up), are shown in table iii in Report Supplementary Material 12.

In the RCT, the only between-group difference in costs at baseline was related to inpatient costs, with the comparison group having higher self-reported inpatient costs than the intervention group. 44 However, there was no pattern in the data [i.e. participants with higher baseline inpatient costs did not continue to have higher costs during the RCT or follow-up study periods (confirmed through data linkage; see Chapter 6, Long-term clinical health outcomes)].

Finally, self-reported data were collected at each time point on the numbers of GP prescriptions of antidepressants, painkillers, asthma, pain gels or creams, anti-inflammatories and sleeping tablets in the previous 12 weeks. These medications were identified as being most likely to be affected by the intervention, and were costed using unit costs for a typical prescription from the British National Formulary (BNF)85 (see table ii, Report Supplementary Material 12).

As we have no self-reported health resource use data between 12 months and 3.5 years, to estimate the total health resource costs associated with participation in the FFIT programme over the entire 3.5-year period, we imputed costs at £16 per year per BMI unit increase, as estimated in the UK Counterweight programme,86 between 1 year and 3 years 3 months (i.e. the period of time for which we have no self-reported follow-up resource use data), assuming that there was no inflation over the period. This value was added to (or, for overall weight lost, deducted from) the observed annual resource use costs at 12 months multiplied by 2.25 years. The costs based on 12-week recall at the 3.5-year follow-up (see table iii, Report Supplementary Material 12) were then added to this figure. For full details of the baseline and 12-month health resource use costs, see Wyke et al. 42

For the hypothetical control scenarios, costs were similarly imputed at £16 per year per unit of BMI increase between baseline or 12 months, and 3.5 years. The annual values were added to the observed health resource costs at baseline or 12 months, and multiplied by 3.5 or 2.5 years, respectively.

Outcome measurements

For the medium-term analysis, weight is assumed to be a modifiable risk factor that has an impact on costs and utilities. Our ‘imputations’ were informed by a review of the literature for weight and costs77,86 or by regressing self-reported SF-12 data collected at all time points (baseline, 12 weeks, 12 months and 3.5 years) against BMI in order to predict SF-12 utilities (HRQoL). How each variable was imputed for the six hypothetical control scenarios is described in detail in the following sections.

Analysis

The individual-level data on achieving a ≥ 5% weight reduction at 3.5 years were summed to provide the number of men attaining this outcome in each group (i.e. the FFIT-FU-I group and the six hypothetical control scenarios). The individual-level QALY data were averaged within each group. Differences in the average utility change between the FFIT-FU-I group and hypothetical control scenarios give an estimate of the QALYs gained from participation in the FFIT programme, assuming no differences beyond 3.5 years.

The cost associated with each individual is the sum of the per-participant intervention cost (for the FFIT-FU-I group) and the cost of NHS resource use and medications. Individual costs were averaged within each group to give an estimate of the average cost associated with the FFIT-FU-I group and the hypothetical control scenarios. The incremental cost associated with the FFIT programme is the difference between the average cost of the FFIT-FU-I group and the average cost of each hypothetical control scenario.

The incremental cost-effectiveness associated with the FFIT programme is presented as the incremental cost per additional individual who achieved ≥ 5% weight reduction over 3.5 years, and the incremental cost per QALY gained. Costs and utilities were discounted at 3.5% in accordance with NICE guidance. 76

Utilities

Utility scores were fitted for each of the six hypothetical control scenarios using observed BMI at baseline, 12 weeks and 12 months, and the estimated BMI values at 3.5 years. The area under the curve method, using the trapezoid method,89 was used to determine the overall utilities. All available data were used at each time point (baseline, 12 weeks, 12 months and 3.5 years) to calculate the area under the curve (< 1% of utility scores were missing at baseline, and 8–11% of scores were missing at the various follow-up time points). Change from baseline was used to account for differences in the average baseline values between the FFIT-FU-I group and each of the hypothetical control scenarios in order to give an estimate of the utility change over the 3.5 years, as shown in Table 22.

Costs

The total cost associated with the FFIT programme was estimated to be £571,000 (95% CI £401,000 to £740,000), with an average cost of £2450 per participant. The minimum cost necessarily incurred in the FFIT-FU-I group is the cost of the FFIT programme itself, at £164. As shown in Table 23, the total cost associated with the hypothetical control scenarios was estimated to be between £521,000 (95% CI £410,000 to £632,000) and £697,000 (95% CI £480,000 to £914,000), with an average cost per participant of between £1640 and £1870 for the different hypothetical control scenarios.

At the 12-month follow-up, 130 men in the RCT intervention group had achieved and maintained ≥ 5% weight reduction, and 40 men in the comparison group had done so. At the 3.5-year follow-up, 53 men in the FFIT-FU-I group who had achieved ≥ 5% weight reduction at 12 months had maintained this to 3.5 years. An additional 22 men who had not achieved ≥ 5% weight reduction at 12 months had achieved this by 3.5 years. After modelling weight gain trajectories in the six hypothetical control scenarios, we examined how many of the 40 men in the comparison group who achieved ≥ 5% weight loss at 12 months were predicted to have maintained this to 3.5 years. In base case and scenario 1, as weight gain is modelled from baseline, no participant achieved ≥ 5% weight loss at 12 months. In the remaining scenarios (in which weight gain is modelled from 12 months) when a population trajectory is assumed, 34 men in the comparison group were predicted to have maintained ≥ 5% weight loss at 3.5 years, and when a FFIT-FU-I group trajectory is assumed, 24 men in the comparison group were predicted to have maintained ≥ 5% weight loss at 3.5 years.

Incremental cost-effectiveness ratios

As shown in Table 24, over 3.5 years the FFIT programme is more expensive than no active intervention, with an additional cost of £532–740 per individual. The results also indicate that the FFIT programme is more effective in terms of QALYs, with a gain of 0.046–0.051 QALYs over 3.5 years. As a result, the FFIT programme is associated with an incremental cost-effectiveness of £10,700–15,300 per QALY gained.

The Cardiovascular Disease (CVD) Policy model

The lifetime analysis involved linking the shorter-term impacts identified within the follow-up period to longer-term impacts through the use of a model. Following a review of published models, the Cardiovascular Disease (CVD) Policy model90,91 was selected for use in the lifetime cost-effective analysis during the FFIT RCT. 44 The model uses data from the Scottish Heart Health Extended Cohort (SHHEC),92,93 which measured ASsessing cardiovascular risk using Scottish Intercollegiate Guidelines Network guidelines (ASSIGN) risk factors [including age, sex, family history of CHD and stroke, socioeconomic status, diabetes, systolic BP, number of cigarettes smoked per day, and total and high-density lipoprotein (HDL) cholesterol] in random samples of the Scottish population recruited between 1984 and 1987. For a full description of the CVD Policy model, see Report Supplementary Material 13.

Employing the Cardiovascular Disease (CVD) Policy model in the Football Fans in Training follow-up study

Within the lifetime analysis in the FFIT follow-up study, the CVD Policy model was used simply as a means to extrapolate the medium-term results. The model uses individual sampling, which enabled us to generate estimates of long-term cost and outcomes for each individual within the FFIT-FU-I group with which to supplement the within-study data. This allowed us to estimate the longer-term costs and outcomes associated with the FFIT programme compared with no active intervention.

Weight and systolic BP were assumed to be the modifiable risk factors that have an impact on costs, life expectancy and QALYs. Weight and utilities were imputed for each of the six hypothetical control scenarios as described in Outcome measurements. Systolic BP was imputed for each hypothetical control scenario, informed by systematic review findings that 10% weight loss equates to a 6.1-mmHg drop in systolic BP on average7 (see table vii, Report Supplementary Material 12).

Updating the Cardiovascular Disease (CVD) Policy model to replace cholesterol with body mass index

We populated the CVD Policy model with data collected during the follow-up study on the risk factors for individuals in the FFIT-FU-I group. In the original longer-term modelling for the FFIT RCT, it was necessary to make a number of assumptions and adjustments in order to use the CVD Policy model. As shown in Table 25, three out of the nine ASSIGN risk factors (family history of CHD/stroke, and total and HDL cholesterol) were not collected during the RCT, and had to be imputed for each participant using information from other Scottish sources. 44

To avoid imputation in the current lifetime analysis, a question on family history of CHD/stroke67 was included in the follow-up study questionnaire (see Q41a, Report Supplementary Material 4), and the original model was updated using the original SHHEC data to replace the two cholesterol variables with a single BMI variable (see Table 25). Summary statistics showed that 50% of participants in the SHHEC data set had a BMI of > 25 kg/m², and 10% had BMI of > 30 kg/m². Even though the population prevalence of overweight and obesity may have increased since the SHHEC data were collected, we assume that the risk they pose has not and, thus, that the changing prevalence is not an issue. Table 26 describes the other variables used in the longer-term modelling, the data sources, and any assumptions made.

The FFIT follow-up economic evaluation supplementary tables provide the risk equations for each first event for men in the CVD Policy model with a BMI as a covariate (see tables viii–xi, Report Supplementary Material 12) and the original model that used cholesterol (see tables xii–xv, Report Supplementary Material 12). When BMI was included in the non-CVD death equation, it had a fairly strong (unintuitive) protective effect, as did cholesterol in the original model. In both models, it was assumed that this protective effect was an association, not causal, and therefore, it was disabled, and not included as part of the model. 90,91

Analysis

Comparing the Cardiovascular Disease (Body Mass Index) Policy model with the original Cardiovascular Disease (CVD) Policy model

The comparison of the CVD (BMI) Policy model with the original CVD Policy model indicated that the incremental costs and effects were approximately of the same order in both models. However, as shown in Table 27, costs were higher, and absolute QALYs were lower, in the CVD (BMI) Policy model.

Cost-effectiveness acceptability curves and expected value of perfect information (EVPI) analyses

Figure 5 shows the uncertainty surrounding the estimates of the lifetime incremental costs and effects for the different hypothetical control scenarios. There is no uncertainty surrounding the existence of a cost difference; the FFIT programme is more expensive than no active intervention (i.e. all of the incremental costs are positive), although there is some uncertainty about the extent of the cost difference in each analysis. In addition, there is no uncertainty surrounding the existence of a difference in effect (i.e. all the incremental QALYs are positive), although again there is uncertainty about the extent of the effect difference. Furthermore, costs and health outcomes (effect) are highly correlated.

FIGURE 5.

Incremental cost-effectiveness planes for the six hypothetical control scenarios. (a) Base case and scenario 1; (b) scenarios 2 and 3; and (c) scenarios 4 and 5.

The cost-effectiveness acceptability curves for the lifetime analysis (Figure 6) illustrate the probability that the FFIT programme is cost-effective for any given value of the cost-effectiveness threshold. Figure 6 shows that in the base case scenario (which is also representative of scenarios 1–3) the FFIT programme becomes the cost-effective option at a willingness-to-pay threshold of just over £2000, and in scenario 4 (which is also representative of scenario 5) the FFIT programme becomes the cost-effective option at a willingness-to-pay threshold of just under £2000. This demonstrates that, when the decision-maker is willing to pay £20,000 or £30,000 per QALY (the willingness-to-pay threshold generally accepted by NICE76), there is no uncertainty that the FFIT programme is cost-effective compared with a no active intervention alternative. Indeed, if a decision-maker is willing to pay more than approximately £2500 per QALY, there is no uncertainty that the FFIT programme is cost-effective.

FIGURE 6.

Cost-effectiveness acceptability curves.

The EVPI analysis shown in Figure 7 demonstrates that there is very little value associated with further research if the decision-maker is willing to pay more than £2500 per QALY. For cost-effectiveness thresholds below this level, there is considerable value in undertaking further research. For the range of thresholds generally employed in the UK (£20,000–30,000 per QALY), there is no uncertainty about implementing the FFIT programme and, consequently, no potential value in undertaking further cost-effectiveness research.

FIGURE 7.

Expected value of perfect information (EVPI) for each eligible man: base case scenario.

Scenario analysis

The scenario analysis (as shown in Table 29 and Figure 6) indicates that the FFIT programme remains cost-effective when the impact of the beneficial behaviour change on the intervention is limited to 5.5 years. Specifically, although the FFIT programme remains more expensive than no active intervention, the average additional cost is reduced to £1025 (95% CI –£85 to £1220) per individual (from £1680, 95% CI –£1480 to £1870, per individual in the lifetime base case analysis). The results also indicate that the FFIT programme remains more effective, but to a lesser extent, with an average increase of 0.639 (95% CI –0.595 to 0.693) QALYs (compared with 0.781, 95% CI –0.732 to 0.831 QALYs in the base case lifetime analysis). The overall result is to improve the cost-effectiveness of the FFIT programme, with an incremental cost-effectiveness ratio of £1600 per QALY gained (compared with £1790–2200 per QALY gained in the lifetime analysis).

Subobjective 1: the impact that sustained changes have had as a result of taking part in Football Fans in Training on the cost-effectiveness of the programme at 3.5 years

The medium-term (3.5 year) analysis found similar cost-effective results to those of the FFIT RCT,44 and these results were well below the willingness-to-pay threshold range of £20,000–30,000 per QALY used by NICE. 76 In the follow-up study, the FFIT programme was associated an additional cost of £532–740 per individual compared with no active intervention. The results also indicate that the FFIT programme is more effective than no active intervention, with a gain of 0.046–0.051 QALYs over 3.5 years. As a result, the FFIT programme is associated with an incremental cost-effectiveness of between £10,700 and £15,300 per QALY gained.

In the scenarios that model a population weight gain trajectory (base case and scenarios 2 and 4; i.e. 0.46 kg per year), the participants in the hypothetical control scenarios gain as much weight as the participants in the scenarios that model the FFIT-FU-I group weight gain trajectory (scenarios 1, 3 and 5; i.e. 1.04 kg per year). The FFIT programme is therefore found to be slightly less cost-effective (but negligibly so in terms of decision-making) in the scenarios modelling the population trajectories.

Subobjective 2: updating the modelling of the longer-term health outcomes and resource use assessing the potential for cost-effectiveness

In the follow-up study lifetime analysis, the FFIT programme was associated with an incremental cost-effectiveness of around £2000 per QALY gained, with no uncertainty that the FFIT programme is cost-effective compared with a no active intervention for any value of the cost-effectiveness threshold beyond approximately £2500 per QALY. Therefore, if the decision-maker is willing to pay more than this value, there is no uncertainty that the FFIT programme is cost-effective.

Assumptions about the duration of the effect of the intervention are a key driver of cost-effectiveness, and we undertook a scenario analysis limiting the treatment effect of the FFIT programme to 5.5 years. The FFIT programme remained cost-effective, although there was a fall in the additional cost per QALY associated with the programme. This result reflects the fact that restricting the period over which behaviour change is sustained has a greater impact on the QALYs associated with the FFIT programme than on the QALYs associated with no active intervention (reducing the incremental outcomes associated with the FFIT programme). In addition, the reduction in QALYs has a greater impact on the lifetime costs associated with the FFIT programme than on the cost of no active intervention (reducing the incremental costs associated with the FFIT programme).

Subobjective 3: heterogeneity of the cost-effectiveness of the Football Fans in Training programme

As the predictor analyses conducted as part of the outcome analyses (see Chapter 3, Baseline predictors of long-term weight loss) did not show any baseline predictors of weight loss, we took the practical decision of basing the cost-effectiveness heterogeneity analyses on BMI, one of the main eligibility criteria for taking part in the FFIT programme. This analysis suggests that participation in the FFIT programme has a greater impact on health outcomes for men with a BMI of > 35 kg/m², and that, although the FFIT programme is highly cost-effective overall, it becomes cost saving for men with a BMI of > 40 kg/m².

Definition of time periods for data linkage comparisons

In autumn 2015, we requested data to be extracted from the beginning of January 2009 to the most recent data available from each database. The period covered by the data linkage study was divided into different time periods to allow comparisons between the RCT intervention and comparison groups at various points in time, and across different periods of time for both groups combined. The four time periods (periods A–D) are shown in Figure 8 and defined as follows.

• Period A: the 12 months before the intervention group started the FFIT programme (i.e. 12 months before the RCT baseline measurements). The start date of period A was 365 days prior to the randomisation date for each man.

• Period B: the 12 months during which participants were involved in the RCT. For the intervention group, this included the period during which they took part in the FFIT programme; for the comparison group, this was the 12-month period prior to being offered the opportunity to take part in the FFIT programme. The start date of period B was the randomisation date for each participant, with the end date being 365 days later.

• Period C: the 2.5-year period between the end of the RCT and the 3.5-year follow-up measures. At the beginning of this period, participants in the comparison group were all offered a place on the FFIT programme. The start date for period C for each participant corresponds to the end date of period B.

• Period D: the 12 months during which participants in the comparison group were given the opportunity to take part in the FFIT programme (i.e. during the first 4 months of period C). Period D allowed comparison with period B in the intervention group. The start date of period D for each man was the autumn 2012 programme start date at the football club where he attended the FFIT programme.

FIGURE 8.

Data linkage time periods. a, FFIT programme delivery period.

Frequencies of events and percentages of participants admitted to hospital and using medications were then reported according to the following comparisons.

1. Pre-FFIT programme delivery: compared the intervention group during period A with the comparison group during period B

2. During FFIT programme delivery: compared the intervention group during period B with the comparison group during period D

3. Long-term follow-up: compared the intervention group with the comparison group during period C

4. During FFIT RCT: compared the intervention group with the comparison group during period B

5. Pre-FFIT programme versus FFIT programme: the clinical health outcomes for the intervention group during period A and the comparison group during period B were combined and compared with the combined clinical health outcomes for the intervention group during period B and the comparison group during period D.

Data management

The RCT database was archived at the Robertson Centre for Biostatistics (RCB) at the University of Glasgow. RCB information security (IS) managers submitted participant names, gender, dates of birth, addresses and RCT ID numbers securely to a named contact at the electronic Data Research and Innovation Service (eDRIS: www.isdscotland.org/Products-and-Services/EDRIS) using a previously established secure file transfer process. The identifiable data were used at eDRIS to index each participant to their Community Health Index (CHI) number to allow the extraction of records on hospital episodes [Scottish Morbidity Records (SMR): SMR01 and SMR04], deaths and recorded cause of death (National Records of Scotland, Vital Events – Deaths),101 cancers (SMR06) and prescribing [Prescribing Information System (PIS)]. The extracted data had all identifiers replaced by ID numbers and were then transferred back to a named IS manager at RCB, as before. RCB IS managers linked the data to the RCT database using participants’ RCT ID numbers. The new linked data set was made available to the project statistician for analysis within a secured network.

NHS research and development (R&D) approval for the data linkage utility study was obtained from Greater Glasgow and Clyde Health Board (reference number GN15DI154). Further approval for linkage and use of routinely collected clinical data was obtained from the Public Benefit and Privacy Panel for Health and Social Care (reference number 1516-0215).

Preparation of linked data variables for analysis

Records from the SMR01, SMR04, SMR06 and recorded cause of death were classified using International Classification of Diseases, Tenth Edition (ICD-10) codes102 and Office of Population, Censuses and Surveys Classification of Surgical Operations and Procedures version 4 (OPCS-4) codes. 103 Prescribing data were classified using BNF categorisations,85 and other data sets used local coding schemes.

Analysis

Clinical outcomes were reported only for the participants who had consented to data linkage at the RCT 12-month measurements, with percentages of those who could be followed up through data linkage reported overall and by group (intervention and comparison groups) also summarised. Numbers of hospital admissions, deaths, cancer incidence and prescriptions were summarised for all participants who consented to data linkage, and also by group, as frequencies of events and percentages of participants. Continuous variables were summarised by the number of observations, number of missing values, mean and SD. Categorical variables were summarised by the number of observations, the number of missing values and the number and percentage of individuals in each category.

Formal group and time period comparisons were made using two-sample or paired t-tests or non-parametric equivalents, as appropriate, for:

1. number of hospital admissions

   1. overall

   2. by ICD-10 group (i.e. the first letter of the ICD-10 code) of primary diagnosis

2. number of prescriptions

   1. overall

   2. by BNF chapter (first-level classification) to reflect broad disease categories (e.g. CVD)

   3. total number of BNF paragraphs (third-level classification) per participant to provide an indication of number of different conditions being treated

   4. for specific medications, including antidepressants (BNF section 4.3), pain relief (BNF subsections 4.7.1, 4.7.2, 4.7.3), statins and other cholesterol-lowering drugs (BNF section 2.12), diabetes medications (BNF section 6.1), non-steroidal anti-inflammatory drugs (BNF subsection 4.7.1), aspirin (BNF approved name aspirin), hypertensives (BNF subsections 2.5.1, 2.5.2, 2.5.3, 2.5.4) and hypnotics and anxiolytics (BNF section 4.1).

The following outcomes were summarised overall and by RCT group, but no formal comparisons were performed owing to small numbers of:

• deaths

• cancer incidences

• psychiatric hospital admissions

• strokes or myocardial infarctions.

Mixed-effects regression models, with a random effect for football club and a fixed effect for RCT group, assessed the relationship between hospitalisations and prescriptions (yes vs. no) as the clinical outcome, and the following explanatory variables measured from the start date of time period B to the end date of time period C:

• weight at RCT baseline

• change in weight from baseline to 3.5 years

• self-reported number of long-standing illnesses at 3.5 years.

All mixed-effects regression models were reported as relative risks or odds ratios, as appropriate, with corresponding 95% CIs and p-values.

Proportion of consented men followed up through data linkage

As shown in Table 31, indexing was successfully completed for 645 men: 86.3% of the original RCT population and 99.5% of the data linkage consented group. Indexing was successful for more men in the comparison group (89.8%) than in the intervention group (82.9%), but this was mainly a result of the greater loss to follow-up at 12 months in the FFIT RCT intervention group.

Long-term clinical health outcomes

As shown in Table 32, linkage was performed for the 645 indexed men to hospital episode statistics, cancer registry, the PIS and National Records of Scotland death records. We had hoped to also link to diabetes care records and blood test laboratory results, but these were still unavailable at the time of the follow-up study. Owing to the small numbers of recorded deaths, cancer registrations and admissions to a psychiatric unit, statistical comparisons between the groups and across different time periods were restricted to general hospital admissions (SMR01) and prescribing (PIS) records.

Associations between clinical health outcomes, weight and long-standing illnesses

The relationships between hospitalisations and prescriptions in the 3.5 years after the start of the FFIT RCT, and baseline weight, change in weight and number of long-standing illnesses at 3.5 years are shown in Table 36. Only the number of long-standing illnesses self-reported at 3.5 years was related to admittance to hospital and receiving a prescription.

Phase 1: spring 2015

The SPFL Trust identified 25 clubs as being due to start new deliveries of the FFIT programme between January and March 2015. Of these, 20 clubs were selected (on grounds of proximity to Central Scotland) for visits by members of the research team to ask participants for permission for data transfer and data linkage.

The research team aimed to visit the clubs during the early weeks (sessions 1–4) of the FFIT programme to maximise the number of participants in attendance. Once the session began, the researcher provided a brief overview of the rationale of the data linkage study and what this meant for the individual, and gave reassurances about anonymity in relation to men’s NHS records. All men were then asked to give written permission or refusal for future data linkage via the completion of a data linkage permission form. The protocol for phase 1 researcher data linkage visits to football clubs in spring 2015 is provided in Report Supplementary Material 16. Field notes, focusing on the men’s responses and questions asked, were written up electronically as soon as possible after each visit.

The number of men who were approached for data transfer and linkage permissions, as well as the number and percentage who gave permission, were reported overall and by club. A descriptive summary of the field notes was used to inform the development of a data linkage feasibility study coach protocol for coaches to pilot during routine deliveries of the FFIT programme in autumn 2015.

Phases 2a and 2b: autumn 2015

The aim of phase 2 was to test the feasibility of an implementation model (i.e. with the coaches who were delivering the FFIT programme requesting permissions for data transfer and linkage) that could be used in post-research deliveries of the FFIT programme. The SPFL Trust identified 22 clubs that were due to start new deliveries of the FFIT programme between September and December 2015.

Coaches delivering these FFIT programmes were trained by a member of the research team in the new data linkage feasibility study coach protocol (see Report Supplementary Material 16). The training was conducted either in person during the routine 2-day training of new coaches to deliver the FFIT programme (n = 4) or individually by telephone (n = 19). Coaches were asked to return all data (including the session during which permissions were collected and the total number of men registered on the current delivery of the FFIT programme) to the research team by Royal Mail tracked delivery, or courier.

Phase 1

The research team visited 18 out of 20 clubs between January and May 2015 to ask participants in new deliveries of the FFIT programme if they would be willing to provide permission for data transfer and linkage. The remaining two clubs did not deliver the FFIT programme in spring 2015. The visits took place across programme sessions 1–4 as follows: two clubs were visited in session 1, six in session 2, five in session 3 and four in session 4, and one club was visited during the enrolment session (the coaches had informed the research team that this was when session 1 was scheduled). The total number of men in attendance was 297 (range 9–35 men at each club). Of these, 264 (88.9%) agreed to data transfer and linkage. The percentage of men giving permissions ranged from 70.0% to 100.0% across the clubs. (For detail, see table I, Report Supplementary Material 18. )

Common questions noted during early phase 1 visits (e.g. around anonymity and how the NHS records would be identified, what data would be made available to the research team and how the results of any data linkage would be reported) were incorporated into the explanation of the rationale for, and process of, data transfer and linkage for later phase 1 club visits. Within sessions 1–4 of the programme, session 2 appeared to be the optimal time for asking for data transfer and linkage permissions: the men had ‘settled’ into the programme and attendance was high.

Phase 2a

Coaches at 11 clubs asked participants for permission for data transfer and linkage between October and December 2015. Despite being fully trained to ask for permissions in session 2, fewer than half (5/11, 45.5%) did so. Others asked for permissions in session 1 (n = 1), session 3 (n = 1), session 4 (n = 1), session 5 (n = 2) and session 6 (n = 1). The total number of men who completed data linkage permission forms across the 11 clubs was 128 (range 5–19 men at each club), with 117 (91.4%, range 68.4–100.0% men at each club) agreeing to data transfer and linkage. (For detail, see table ii, Report Supplementary Material 18. )

Although coaches had been trained to report the total number of men who were registered on the current delivery of the FFIT programme, only 7 out of 11 (63.6%) did so. This meant that we were unable to gain an accurate picture of the proportion of men registered on the programme who agreed to data linkage and transfer. When coaches did provide information about the number of men registered on the programme, comparison with the number of data linkage permission forms returned suggested that attendance at later sessions was similar to attendance at the earlier sessions.

There was slightly more variation across the clubs in terms of the percentage of men returning completed data linkage permission forms who agreed to data transfer and linkage in phase 2a (range 68.4–100.0% men at each club) in comparison with phase 1 (range 70.0–100.0% men at each club). This variation may indicate some inconsistencies in the way the protocol was delivered. In the four clubs observed, it was clear that, despite having been trained in the data linkage delivery protocol, most of the coaches still lacked confidence in explaining the rationale and process of data linkage, and in answering questions. Often the process was rushed, with a lack of discussion, and it was not clear that the men fully understood what they were being asked to do.

Given the poor compliance with the delivery of the data linkage feasibility study coach protocol in session 2, and the apparently high attendance at some of the later sessions, the decision was taken to delay the introduction of information about data transfer and linkage until session 6 [when it was felt that the men’s relationships with the coach(es) should be established]. In addition, to encourage the men to fully consider the implications of data transfer and linkage before deciding whether or not to give their permission, and to reduce the time pressure on coaches having to deliver the whole process in a single session, men were not asked to complete a data linkage permissions form until session 7. Finally, instead of being a stand-alone document, the data linkage feasibility study coach protocol was incorporated into sessions 6 and 7 of the main FFIT coach delivery notes, to try to increase compliance with the timing of delivery.

Phase 2b

Coaches at 10 clubs asked participants for permission for data transfer and linkage between February and April 2016. Data on the timing of the delivery of the protocol were available for 6 out of 10 clubs, either coach-reported or from researcher observation field notes, and five out of six clubs delivered the information on data transfer and linkage during session 6, as per protocol. The total number of men who completed data linkage permission forms across the 10 clubs was 87 (range 5–17), and 78 of these (89.7%, range 76.9–100% men at each club) agreed to data transfer and linkage. (For detail, see table iii, Report Supplementary Material 18.)

The number of completed data linkage permission forms returned in phase 2b (range 4–17) was smaller than in phase 1 (range 9–35) and phase 2a (range 5–19), which may reflect the fact that the data linkage permission process was conducted over two sessions. This conclusion is supported by variation in the dates on the completed data linkage permission forms in some clubs, suggesting that coaches may have had some problems in getting men to complete their forms after session 6 (i.e. not all did it in session 7, as per protocol).

Observations were timed to coincide with the session in which the coaches informed the men about the data linkage process. In four out of five clubs, this activity took place in session 6. Coaches at the fifth club delivered it in session 3. In one club, session 6 was led by a coach who had not received training in the data linkage protocol, because the main coach (who had been trained) was unavailable. This led to confusion about the materials that were to be handed out, and the men being largely left to learn about the process by reading an information sheet. In three of the clubs, the activity was well delivered, with the coaches raising most (if not all) of the key points, including the fact that agreeing to data transfer and linkage was completely voluntary and would not affect men’s involvement in the FFIT programme. In the club where information on the data linkage process was delivered in session 3, the coach appeared rushed; nevertheless, the men did appear well informed, and the voluntary nature of the process was clearly emphasised. Therefore, it appears that with adequate training, and when the data linkage permissions activity is included in session 6, the coaches are able to deliver it confidently to allow men to be fully informed before deciding whether or not to agree to data linkage. However, asking men to complete permission forms the following week may have contributed to a lower rate of completion than in previous phases.

Following up consenting men through data linkage

The data linkage utility study demonstrated that it was highly feasible to follow up participants through data linkage within the Scottish system, with a 99.5% success rate in linking data to routine NHS records for men who provided permission to do so.

Long-term clinical health outcomes

The linkage data showed that high proportions of participants had been admitted to hospital and/or received prescribed medications from 2009 onwards. This finding suggests that the FFIT RCT cohort (or at least those who consented to data linkage) had a number of existing conditions before, during and after their participation in the RCT, confirming (as we reported previously47) that this was a high-risk group of men. We did not detect any significant differences in hospital admissions; however, it is important to note that the RCT power calculations were based on weight loss at 12 months (our primary outcome) and not on hospitalisations or prescriptions.

Nevertheless, there were some differences in prescriptions, with the RCT comparison group (or at least those who consented to data linkage) tending to receive more prescriptions than those in the intervention group. Differences were seen in both the number of men receiving at least one prescription and the number of prescriptions per man at various time points. Men across both groups also received more prescriptions in the 12-month period during which they took part in the FFIT programme than in the 12 months prior to the FFIT programme. Finally, there was some suggestion that, in the 12 months during which men took part in the FFIT programme, they received more preventative cardiovascular (e.g. statins and aspirin), urinary tract, and musculoskeletal and joint medication.

Associations between clinical health outcomes, change in weight and self-reported long-standing illnesses at 3.5 years

Small numbers of reported deaths, cancers, psychiatric hospital admissions, and strokes or myocardial infarctions meant that we were unable to explore associations between these outcomes and weight and long-standing illnesses. For hospital admissions and prescriptions, we did not find any associations with weight. Associations between hospital admissions and prescriptions and self-reported long-standing illnesses confirmed the validity of our self-reported long-standing illnesses data.

The role of physical activity and diet

We did not find any subgroups of men who were more (or less) likely to achieve good long-term weight outcomes on the basis of our prespecified potential baseline predictors (age, BMI, education level, socioeconomic status, marital status, orientation to masculine norms and number of long-standing illnesses) Our mediator analyses suggested that regular PA (including walking and vigorous PA), and conscious dietary regulation (including reductions in fatty and sugar foods, portion sizes and regular mealtimes) and increased intake of fruit and vegetables were all associated with improved long-term weight outcomes. The findings are in line with existing evidence pointing to the role of both PA and dietary restriction in weight loss maintenance,16–18 and confirm the utility of many of the strategies that men were introduced to during the FFIT programme and still talked about using to control their weight in the 3.5-year qualitative interviews. Routinisation and the formation of new (PA and dietary) habits are important for behaviour maintenance;23 however, the men’s accounts suggested that work was often required to maintain new habits. They talked about the importance of planning and organisation to allow them to pursue PA and healthy eating in the face of competing work and family commitments.

Many men also spoke about the role of PA in weight loss and weight regain. For some, PA and weight appeared to be almost inextricably linked, whereas others spoke about a more indirect association (i.e. being more active encouraging them to eat better). Previous studies32,112,113 have highlighted that men are more likely to use PA as a strategy to manage their weight than women. Given the important role of PA in long-term weight control,16,17,19 the promotion of PA (e.g. encouraging the incorporation of walking into daily routine, as in the FFIT programme) may be a particularly useful strategy to improve weight loss maintenance in men.

The mediator analyses also revealed that increased self-reported sitting time was associated with poorer long-term weight outcomes, which is consistent with previous evidence of an association between increased sedentary time and risk of obesity. 114 The provision of information and advice around decreasing sedentary time, alongside increasing PA, may therefore be useful to improve the long-term outcomes of weight management programmes (see Further research).

Finally, given the finding of a positive association between the consumption of fruit and vegetables and long-term weight loss, and the fact that men in the FFIT-FU-I group had reduced their consumption of fruit and vegetables during the follow-up period, programmes aimed at men, who may be more reluctant than women to eat fruit and vegetables,115 should place more emphasis on the importance of finding sustainable ways to increase the intake of fruit and vegetables.

The role of behaviour change techniques

As theoretical accounts of behaviour change maintenance predict,23 regular self-monitoring of weight and a supportive social environment (e.g. through ongoing contact with other men and coaches from the FFIT programme) were associated with lower weight at 3.5 years. Most men had continued some form of monitoring of their weight (either directly, through weighing themselves, or indirectly, often through the fit of their clothes) and many also described continuing to monitor their PA (some, but not many, continuing to use pedometers or alternative strategies, which included being aware of the distance or time spent walking). Some men had adopted new technologies that allowed them to monitor steps taken and other forms of PA (e.g. running and cycling), and some also spoke about using these to monitor diet and calories consumed. A number of men continued to set themselves goals and targets in relation to weight and PA, although sometimes these goals were rather loosely defined. Nevertheless, some men found it useful to set an upper weight threshold that they did not want to cross to help them maintain their weight at a level that was comfortable for them.

The coaches who deliver the FFIT programme are trained to build on the men’s common interest in football and to encourage the formation of a positive group dynamic to allow the men to support each other in making changes to their PA and diet. 116 During the programme (in session 12) the coaches encourage the men to consider how they can meet up once the FFIT programme is finished. In the 3.5-year interviews, men described the importance of continuing to meet up with other FFIT participants in their ongoing efforts to control their weight. However, not all men were able to attend ‘follow-on group’ meetings. Some men complained that there was no such group at their football club, that the meeting times or locations were not convenient or that the activity that the group did (often playing football matches, which some men found too competitive or strenuous) did not suit them. Some of these men attributed their weight regain to their lack of ongoing contact with fellow FFIT participants, but others described finding sources of social support elsewhere, for example through sports club membership, other friends and work colleagues, and family members.

The role of innate psychological needs

Self-determination theory suggests that maintenance of behaviour change requires satisfaction of innate psychological needs for autonomy (associated with the internalisation of regulation), competence to perform behaviours and relatedness to others. 20,21 Our mediator analysis results were consistent with these hypothesised relationships: associations with improved long-term weight outcomes were found for autonomous regulation, internal locus of control, perceived competence for PA and dietary behaviours, and relatedness to other people. In the 3.5-year interviews, men described a range of factors that continued to regulate their behaviours. These included interactions with health professionals, family members and friends (external factors), and more internal factors, such as personal health benefits, avoidance of guilt or embarrassment, and doing things because they valued them (e.g. being able to do more things with their children or grandchildren) or simply because they found them to be enjoyable. A few men articulated their journey from initially wanting to make changes because of external factors, to continuing to manage their weight because they were enjoying the changes they had made. However, accounts of changed identities (or of viewing themselves differently) as a result of the changes that they had made were evident in interviews only with men who had been successful in long-term weight control. This is consistent with self-determination theory, which suggests that this fully integrated regulation is the most stable form of regulation and, therefore, most supportive of maintenance of change.

Our mediator analyses suggest that the type of relatedness with other FFIT participants that is supportive of positive long-term weight outcomes (intimacy) may be qualitatively different from that required from family members (acceptance). Consistent with this, the extent to which they were able to continue meeting up with other FFIT participants seemed important in many men’s accounts of their motivation to maintain their weight loss and lifestyle changes. However, men’s descriptions of maintenance of change and relationships with family members appeared more complex. This echoed findings from observations of non-RCT deliveries of the FFIT programme and post-programme participant focus groups, which suggested that although some men experienced wholehearted support from spouses/partners (characterised in our analysis as ‘facilitative allies’), for their efforts to improve their eating practices, others experienced more muted support (‘detached allies’) and in relatively rare cases felt that their efforts to change were undermined by spouses or partners. 117 Examples of supportive ‘facilitative or detached allies’ and family members ‘undermining change’ were also evident in the 3.5-year interviews, but these extended beyond spouses or partners and eating practices to include other family members (i.e. children) and PA.