Process evaluation

Text from this chapter has been reproduced from Guest et al.123 This is an Open Access article distributed in accordance with the terms of the Creative Commons Attribution (CC BY 4.0) license, which permits others to distribute, remix, adapt and build upon this work, for commercial use, provided the original work is properly cited. See: https://creativecommons.org/licenses/by/4.0/. The text below includes minor additions and formatting changes to the original text.

Overview

Process evaluations provide a contextual understanding of how a particular intervention or programme was delivered, how participants reacted to it and why it was successful/unsuccessful in influencing behaviour change. In the context of complex and multicomponent interventions, such as the SHIFT intervention, which was delivered across multiple heterogeneous working environments throughout the UK, process evaluations provide useful information relating to all aspects of programme delivery and potential success, helping to inform which intervention components contributed the most and least to overall effectiveness. Therefore, process evaluations allow potential modification of the intervention, if required, and can inform the future implementation of the intervention (e.g. the SHIFT intervention as a training resource to HGV drivers across the logistics sector).94

The purpose of this process evaluation, therefore, was to investigate the application of the SHIFT intervention to understand the context within which the intervention was applied and the key elements of its implementation, including fidelity (i.e. components of the intervention), adaptations, contamination, sustainability, barriers and facilitators. To determine the implementation fidelity (which is the extent to which the programme adhered to the protocol model initially developed124), we aimed to understand whether or not components were used as intended (e.g. dosage, attrition rates) and if any adaptations were made to the intervention. A further aim was to describe and understand the contextual factors that may have influenced the intervention’s implementation and/or effectiveness. In addition, the process evaluation will recommend refinement of the intervention for future sustainability to ensure that the intervention can be optimally embedded into the stakeholders’ routine policies. Finally, this process evaluation will also help to support the development of further effective RCTs that are evaluating lifestyle health-related behaviours in HGV drivers (and employees with similar enforced sedentary occupations).

Methods

The MRC guidance for process evaluation of complex interventions was the most suitable framework for this process evaluation.94 The MRC guidance offers key comprehensive direction to describe the intervention, implementation and mechanisms of change, while understanding the contextual factors throughout. This integrated process evaluation proceeded in a series of steps and took place alongside the RCT. Mixed methods were used to deepen analytic understanding of a specific issue and, in turn, triangulate results. Fidelity and dose were measured both quantitatively and qualitatively. Qualitative research techniques were used to identify how context affects implementation, barriers, mechanisms of impact and future sustainability. Using a feedback questionnaire, we aimed to collect data on key aspects from both control and intervention participants, and combined these data with information gained from in-depth interviews and focus groups with drivers and managers who were purposively sampled from each depot. Feedback questionnaires were given to every participant 1 month after baseline measures and 1 month after 6-month follow-up measures. Interviews were conducted with one participant and one manager from each site after the 16- to 18-month follow-up measures. Table 37 summarises each of the evaluation outcomes examined, how the evaluation outcomes were defined and how the evaluation outcomes were measured. Table 38 provides a detailed overview of the process evaluation data collected through each method. The process evaluation results are presented according to the key outcomes examined (see Table 37).

Quantitative data collection

To inform attrition rates, records were gathered from all participants, including participation uptake per site (i.e. the number of participants who completed the baseline health assessment per site), dropout rate per site (i.e. the number of participants who failed to complete each follow-up), engagement with text messaging service (i.e. the number of responses) and compliance with activity monitor wear.

Feedback questionnaires

All participants were asked to fill out feedback questionnaires 1 month after the baseline health assessment and 1 month after the 6-month follow-up health assessment. The questionnaires included a mix of multiple-choice, open-ended and Likert scale questions. At baseline, the questionnaires sought feedback about the usefulness of the baseline health assessment for awareness of health and what the most useful measurements were. The intervention follow-up questionnaire (given to intervention participants 1 month after the 6-month follow-up health assessment) sought feedback about the quality and usefulness of different elements of the health intervention, including the education sessions, booklets, cab workout equipment, Fitbit, step challenges, text message feedback and any other lifestyle or work changes that may have affected the results. In the questionnaire that was completed after their 6-month follow-up assessment, control participants were asked about both health assessments and any other lifestyle or work changes that may have affected the results. Open-ended questions were analysed by listing key responses into themes and calculating frequencies in each theme.

Qualitative data collection

Managers in intervention sites were asked to schedule three to six participants for an on-site focus group, during the drivers’ working hours. A semistructured focus group schedule was developed and piloted with participants from six pilot sites. Questions related to each SHIFT intervention component, barriers to achieving health goals and recommendations for improvements to the SHIFT intervention. A semistructured interview schedule was developed and piloted for the managers in the pilot sites. Managers were also asked to schedule time for an interview on the same day as participants. These interview schedules were then analysed and revised, as needed, before commencing data collection for the main trial sites (n = 19).

Main trial interviews

Owing to the outbreak of COVID-19 and a national lockdown (March to July 2020), when face-to-face data collection (e.g. in-person interviews) was no longer feasible, mobile telephone interviews were completed for the main trial sites. As these interviews occurred during non-work hours, all participants were sent a text message to ask if they would be interested in participating, with the offer of a £5 high street voucher and a chance to win a Fitbit. The main liaison manager from each site was asked to participate via e-mail.

Data analysis

All focus groups and interviews were carried out by one or two researchers (AG and YLC). The focus groups were recorded on a digital audio device and transcribed verbatim (by AG) into Microsoft Word^® (Microsoft Corporation, Redmond, WA, USA), before inputting into NVivo 12 software (QSR International, Warrington, UK) for analysis. Transcripts were then re-read, coded and developed into themes using the deductive method of thematic analysis, where themes were already preconceived based on the interview schedule and existing knowledge. This was followed by an inductive method where themes were identified entirely based on the data (by AG).125 Each stage of the analysis was critically analysed from an informed external perspective. (Dr Anna Chalkley, an expert in qualitative research methods and programme evaluation, was independent to the research team and acted as a ‘critical friend’ throughout the analysis and reporting stage of this process evaluation.) Quantitative data, including dose, attrition and questionnaire data, such as multiple-choice and Likert scale answers, were analysed descriptively in IBM SPSS v25 (IBM Corporation, Armonk, NY, USA).

Results

Questionnaires

A total of 71.2% (n = 272) and 57.8% (n = 167) of the participants invited to complete a feedback questionnaire following the baseline and 6-month follow-up health assessments, respectively, responded.

Focus groups and interviews

During the internal pilot, on completion of the final follow-up measures, four focus groups and three interviews with 13 participants (i.e. an 83.3% site response rate) and eight individual interviews with managers (i.e. an 83.3% site response rate) were carried out to capture their views on and experiences of implementation. In the main trial phase, 15 telephone interviews were conducted with participants from the remaining 19 sites (i.e. an 82.4% site response rate), and 13 telephone interviews were conducted with managers (i.e. a 76.5% site response rate) after the 16- to 18-month follow-up assessments.

Through synthesising the data, we found there to be no substantial differences between pilot testing and main trial sites in both the questionnaire responses and the content of the interviews and focus groups. Therefore, the combined findings from both the pilot and main trial sites are presented below.

Dose

It was intended that participants in the SHIFT arm would receive a total of 12 hours of face-to-face contact with the research team over 12 months, broken down into three 2-hour health assessments and one 6-hour education session. Control participants would receive 6 hours of face-to-face contact, consisting of three 2-hour health assessments. Owing to the COVID-19 pandemic, a national lockdown prevented all main trial participants from receiving a face-to-face final follow-up health assessment, hence a reduction of 2 hours of face-to-face contact with participants from the main trial.

Text messages

Throughout the duration of the study, the average number of text messages sent to each driver was 20.2 and the average number of text messages received from each driver was 3.8, which resulted in a 18.8% response rate. Although drivers did not engage much with the text messaging, it appeared that the overall use of text messages was relatively positive, and a number of drivers used this service at the end of the trial to relay their appreciation to the research team:

I haven’t really text backwards and forwards, but when you sent me the, the little challenges like I thought good, yeah, I enjoyed that.

. . . it’s been interesting and enlightening and hopefully you and your team have added a few years to my life and others with the result of your research.

Thank you to yourselves [students] your university, staff, teachers, etc., for all your endeavours helping me to take my health more seriously. Wishing you all the very best in the future.

Cab workout equipment

Fifty-nine per cent of intervention participants said that they had used their cab workout equipment in the last 6 months, with few planning to use it in the future. For participants who did use it, the most common piece of equipment was the hand gripper, followed by the resistance bands. There was minimal reported use of the fitness ball. Of the participants who used the cab workout equipment, only 16.3% agreed/strongly agreed that it increased their physical activity levels. Reasons for low adherence levels were explained in more depth during the interviews. Some participants suggested that it was impractical to use the workout equipment in their cabs:

I mean to start doing all the . . . [demonstrates exercising], you know, just completely impractical. Really is . . . but . . . It’s all right if you’re office based.

Participants also reported that they prioritised sleeping and eating on their breaks:

. . . the only thing I do on my break is sleep.

I don’t, I don’t think it’s practical. You have 50 minutes’ break and by the time you’ve been to the toilet, you’ve had your dinner, you need a nap, if you’ve done 3.5, 4 hours of driving, you need shut eye for 10, 15 minutes. There’s just not enough time . . . To do those things.

Participants also suggested that, although not embarrassing for them, cab workouts would be embarrassing for some of the older drivers:

So the thing is . . . I think with a lot of like the old boys, like the drivers. Maybe they wouldn’t want to do something like that because they’d feel like, stupid doing it to be honest. Do you know what you mean? Whereas I don’t care. But like older drivers, they’d have just been like ‘oh I’m not doing that’.

However, a small number of participants did carry out cab workouts regularly and seemed to benefit from this:

Yeah, the hand gripper, I use that, the stretchy bands, I used to use them more, because I could tie it to the bottom of the handrail in the cab and I could exercise my arms, both arms. And you could do it whilst you were driving. Right hand only. I mean obviously you’ve got to keep holding the steering wheel all the time, but when I were parked up, I used the gripper, and the ball. And you know, I kept up with that. I’ve still got them, and I still use them. So you know, they are, they are quite, you know. They’re all right, they do what they’re meant to do.

Fitbit

The majority (92.4%) of intervention participants were still using their Fitbit 6 months later, and 6.1% of participants had previously used their Fitbit but no longer use it. Only 1.5% of participants had never used their Fitbit and did not intend to in the future.

Participants who did not use the Fitbit mainly said that it was due to the Fitbit causing skin irritation, already having an activity monitor or not knowing how to use the Fitbit:

I haven’t, I have been wearing it, but I had a rash on my wrist so I’ve had to take it off.

Umm. Fitbit no, because to be honest, I’ve got one of these . . . I’ve got my own.

If I remember, we were given the Fitbit at the end of the lesson weren’t we? I have, still no idea. I have no idea how to use the watch. I have no idea how to . . . Yes. I have no idea how to use it. The only thing I do is, press the button, and it tells me the time and that’s it. I mean, seriously. I mean, I mean, I’m hoping that this Fitbit thing is going to show me just how bone idle I am.

Fidelity

Health assessment

Eighty-three per cent of main trial respondents agreed/strongly agreed that the baseline health assessment met their expectations, whereas 86.0% of respondents agreed/strongly agreed that the baseline health assessment made them more aware of their current health status. Most (90.3%) respondents agreed/strongly agreed that the health assessment was thorough and understandable. Participants were asked to rank the measurements in terms of most interesting and useful, and cholesterol was considerably the most valued measurement, with waist circumference the least valued measurement (Table 39).

Most (87.8%) intervention participants agreed/strongly agreed that the first health assessment motivated them to want to change aspects of their lifestyle, including increasing physical activity (73.2%) or improving diet (70.7%). Fewer control participants agreed/strongly agreed (63.6%) that the first health assessment motivated them to want to change aspects of their lifestyle, including increasing physical activity (49.5%) or improving diet (52.5%). Participants mentioned that it was ‘eye-opening’ to see their results within the health assessment:

But when you see it wrote down in front of you, it sort of clicked something in the back of your mind, thinking ‘oh God’. You know what I mean. Numbers, numbers look a lot worse than looking at yourself because you’re used to looking at yourself . . . you look in the mirror and you just tap your belly, and you say ‘blimey I’ve got to lose some weight’. You know what I mean. But, when you’re sitting there, and someone writes down, like . . . I can’t think what it was, it was bleeding high, I know that! Yeah they said your BMI is this and that, then you start to think . . . You know what I mean. Because as I say, you look at the numbers and then it triggers something in you and you think ‘oh blimey, I’ve gotta sort this out’. You know what I mean.

Participants (including control participants) also mentioned that it was motivational to improve their health:

I didn’t realise I was overweight or, I’ve got a bit of belly but it’s quite literally, yeah it’s not too big but when they turned around and said you’re obese and your BMI is like a little bit not where it should be and then you’ve got the graph and they point to you in a certain section . . . Mine wasn’t too bad it was just a little bit above where I should have been so it gave me confidence that I could get there.

Six-hour structured education session

Most (90.9%) participants agreed/strongly agreed that the structured education session was the correct length of time, and participants agreed/strongly agreed that it motivated them to increase their physical activity (77.3%), reduce their sitting time while not at work (72.7%) and improve their diet (71.2%). When asked in the questionnaire about the key messages participants took away from the education session, 41% of 97 responses mentioned dietary knowledge changes and 31% mentioned exercise knowledge changes.

Based on the interviews and focus groups, participants were happy overall with the information they received. Most participants recalled learning about diet, more so than other elements (e.g. physical activity, smoking, alcohol, sleep and mental health):

Yeah, I enjoyed the workshop with the, with all the information we received about the food, I thought it was really enjoyable. And the understanding, gave me a better understanding of food. And gave me a better idea of how I could balance my diet. So I thought that was one of the best things of the study, was actually that, that part of it and the information that was available to us.

The way it was put across was great, the people we had in there was, was brilliant, and everyone partook in it brilliantly. But it was the information, the information that was there and available to us. And it was given to us in a way, it was given to us because it gives us a lasting knowledge of what was good for you and how easy it is to have the things you like, alongside the things you maybe no like to make a balanced diet, so you don’t have to be daft and cut everything out and just have fruit and veg, you can have a bit of everything and still be healthy. So I mean it was really good.

Fitbit

Participants agreed/strongly agreed that the Fitbit increased their awareness of their physical activity (90.5%), sitting time (73.8%) and sleep pattern (88.1%). In addition, the Fitbit motivated participants to make changes to their physical activity (81.0%) and their sitting time (69.0%), but less so their sleep pattern (40.5%). Three-quarters (75.4%) of participants engaged in the step challenges, 59.4% agreed/strongly agreed that step challenges were motivating and 54.7% agreed/strongly agreed that participating in the step challenges helped to increase their step count. Some participants reported that the free Fitbit was the main motivation for taking part:

The Fitbit was also mentioned as a useful tool for goal-setting and for providing a consistent sense of achievement:

. . . with having the tracker thing [points at Fitbit]. It’s like a reminder like, in better weather, I’d look at it, and if I hadn’t got enough steps in I’d go and take the dogs around the block again, you know what I mean, just to get my counts up.

Because of course, if you haven’t done your 250 steps, it buzzes, so I’d actually get up and start walking up down the living room to get those 250 steps in! If it’s nearly to midnight, I haven’t done me 10,000 steps, I’d be up and walking wherever I could.

I think the Fitbit, because I think that motivated me more than anything else because it challenged us all the time, you know it buzzes and said come on let’s go, 250 steps. That’s . . . where I live on my boat we’ve got washrooms we can use, that’s walking to the washrooms and coming back, I get my 250 steps so I maybe do that four times a day, whether I want to or not, I still like the challenge. You know, and I’m still getting that challenge now, which I try my best to carry out and get done. But I would say the Fitbit is the . . . it gets into your head, you get addicted to it. I am addicted, I always have it on, when the armband broke I actually used an elastic band and I wrapped it around so it stayed on my arm, ha ha ha! But that really got me addicted, and I’m still addicted to it now like you know.

Text messages

Participants agreed/strongly agreed that the frequency (84.5%) and content (91.1%) of the text messages was appropriate and informative. Based on the focus groups and interviews, the consensus of participants was that the text messages were good at providing logistical information for the upcoming tasks and that it was beneficial to know that the SHIFT team were readily contactable:

Well, they were sent as they were needed, you know it wasn’t an overload for the brain or anything but they were you know, when something had to be done, and yeah they were fine, unless you wanted . . . because I had the number I knew if I had any problems, any time I actually messaged you I did get a reply back . . . I did. Yeah every time I always got my reply back, and that’s what I was happy about. At least I, I knew I was doing something at my end, and you’s were taking it serious at your end . . . You were always there, because you said if you need us, we’re there. You kept your word, like I said, I messaged you, you got back, and you had kept your word.

However, participants suggested that if the messages were to be used in a motivational capacity, then more frequent messages would be beneficial:

The only thing the only thing I could think of personally, would be to just ramp them up a little bit more. You know what I mean . . . Because they were quite well spaced out. You know what I mean. If I’m right, there were only about two of them, weren’t there?

Adaptations

All baseline health assessments were conducted at the beginning of the drivers’ shift. Carrying out baseline health assessments for all drivers was logistically challenging for both the researchers and transport managers because of the 24 hours a day, 7 days a week (24/7) nature of the job (i.e. drivers started their shifts at different times of the day and night). As a result, post-intervention measures were carried out at a time of day that was more feasible and occasionally this was at the end of the drivers’ shifts. Drivers were still asked and reminded to fast for 4 hours prior to the assessment; however, this change in protocol may have affected body weight and blood pressure measurements.

The initial target was to include six participants per education session; however, it was logistically challenging for the transport managers to take six drivers off the road at any one time, particularly for the smaller sites, and, as a result, most education sessions consisted of fewer drivers per workshop (mean: 3.4 drivers; range: 1–7 drivers).

Owing to the COVID-19 pandemic, a national lockdown prevented all main trial participants from receiving a face-to-face final follow-up health assessment. Instead, in the 16- to 18-month follow-up health assessment, participants were asked to fill in the questionnaire, self-report their weight and wear the activPAL device (to monitor physical activity and sitting) while filling in the wear log continuously for 8 days. The pandemic also caused a 4- to 6-month delay in the final follow-up assessments for sites in the main trial phase, resulting in these taking place 16–18 months after randomisation.

Attrition

A total of 382 participants (SHIFT intervention, n = 183; control group, n = 199) received the baseline health assessment. Two hundred and sixty-two participants (68.6%) took part in the 6-month follow-up health assessments. The retention rate at 6 months was higher in the control group (73.9%) than in the intervention group (62.8%), and this may be because of intervention participants being ineligible to continue with the study if they failed to take part in the 6-hour structured education session. Of the 183 intervention participants, 145 (79.2%) took part in the education session. At 6 months, the retention rate for intervention participants who were eligible to continue after the education session (n = 145) was 79.3%. Participants who did not attend the session were removed from the trial. The dropout rate in the intervention group may have been a result of their expectations for greater health improvements among them:

I think that for them that dropped out they just . . . I think they’d realised that they’d not really made that much of a change, or they made them, but they were short-lived if you know what I mean?

Table 40 shows reasons for discontinuation of the study.

Retention rate between the 6- and 16- to 18-month follow-ups was higher (78.2%). In this instance, retention was better in the intervention group (87.8%) than in the control group (70.7%). It may be that overall retention rate was higher after 6 months because participants who were disinterested in the programme had already dropped out. Participants in the intervention group were possibly more invested in the intervention, and yet little time spent engaged in the project was required of them during this time. From start to finish, 46.3% (n = 177) of participants dropped out of the RCT. Mangers were asked in interviews if they knew the reasons for participants dropping out. One of the primary reasons mentioned was that truck drivers are notoriously transient workers, with a high staff turnover rate. It was evident that a lot of drivers had left their company before the cessation of the programme:

Like I say, the drivers that had to drop out, they left, obviously that’s one of those things, they left and went on to different contracts.

Other reasons for participants dropping out included long-term illness, and a few participants were isolating or on furlough:

Yeah, a couple on long term sick, and I think a couple that were shielding as well.

Another commonly mentioned reason for participants dropping out was that some saw the timings of the health assessments as too inconvenient:

They just want to come in and go home. Yeah, they don’t want to do anything else that adds on to their day. Which I totally get.

Barriers to behaviour change

In the 6-month questionnaires, all participants were open-endedly asked to identify their main barriers to a healthy lifestyle. Among the 145 responses, 46% suggested that the biggest barriers were work related, predominantly the length, irregularity and start times of their shifts. The second biggest barrier was family (e.g. child care) commitments (12%) and this was followed by self-motivation (10%), with drivers referring to themselves as being ‘lazy’ or ‘need[ing] more discipline’. Table 41 shows a summary of the mentioned health barriers.

Using an inductive approach in the thematic analysis, we discovered that participants discussed barriers to living a healthy lifestyle both at work and at home. We deemed it important to mention these barriers, as this may influence guidance for future health interventions in this demographic. Participants mentioned that there was little time in the day to fulfil a healthy lifestyle:

The trouble is, is when do you get the time, isn’t it? . . . You know, you finished at 10 o’clock in the morning, from 8 o’clock at night. All you want to do is go back and have a shower go to bed.

You’ve got all the things to do, you know I don’t have any chance to do any exercise, I don’t.

Linked to time pressure are the long, irregular shift patterns and early start times that almost all drivers from both control and intervention sites mentioned at least once in their interview:

. . . well, yeah, you know and the start times as well, I start between midnight and 4 a.m. in the morning and that’s no, no good for your body rhythm if you like. It’s not good for your body. By 2 o’clock in the afternoon you want to go to sleep.

But I’ve done this for 35, 36 years now, umm, and it’s just, it’s just part of being a lorry driver, it is how it is, it’s the territory that you’re in. You know, the uncertainty of where you’re going to work, whether you’re going to go home that night, or it might be 5 days later.

I think there should be, once you have worked a 12-hour shift, you should have a minimum of 12 hours off. Whereas you can actually work 15 hours and have 9 off. That’s the legal requirement, 9, 9 minimum . . . because it’s 15 and 9, that’s your 24 hours in a day isn’t it. Alright, so you could start at 6 in the morning, work till 9 at night and realistically you could start at 6 o’clock in the morning the next day.

And, I mean, to say that the trouble with the HGV world is uh, it’s poor wages long hours, you know. You can make a good life out of it, but you’ve got to put a lot of hours in, and when you put the hours in, it is detrimental to your health at the end of the day. You know, unfortunately that’s the aim of the game, you know.

Therefore, there’s often unrealistic, unrealistic expectations required of drivers, and long hours. And short breaks. And that doesn’t help at all with this side of things, which is looking at trying to keep healthy. Try to have a diet, a decent diet. When you’ve got guys going out for 13, 14 hours a day, when you get home . . . when I’m on 12 hours a day, I get home, I have a snack and go to bed practically. A snack, a shower and bed and up in the morning. And it’s a snack, and into work, when I’m on my 12-hour shift. So the guys that are getting a couple of 15 hour shifts on the bounce, they get home, go straight to bed. And they’re actually back the next day driving, munching as they go down the road because that’s their time for having something to eat. So, the fact that the pressures that are put on us in the workplace, and often or not the uhhh . . . the mismanagement of that is, is a factor towards the healthiness and well-being of the driver himself.

A lack of managerial support and excessive expectations also affected drivers:

It’s operational requirements, like. It’s, it’s operational requirements, because you can’t always eat when you’re supposed to eat, or sleep when you’re supposed to sleep or anything like that. It all revolves around your working day. And that’s how it is.

It’s not doing one 12-hour shift or one 15-hour shift, it’s one after another, after another, after another. You know, they think you’re machines like these trucks, you can drive them for 24 hours a day, 7 days a week, they’re just trucks, put some more diesel in, put some more oil in, they’ll go forever, we can’t do that. They think we can.

I’ll be honest, well, me and my partner have now gone part time, because it had just got ridiculous. And they just expect so much of you. And like, especially . . . for some reason, I don’t know why. In our place, the night drivers are just the dirt on their feet . . . They’ve just, no respect for them . . . but you try working nights. They don’t seem to understand that, you know, your natural body clock, even no matter how many years you work nights, your natural body clock is, you go sleep at night.

Another key barrier included unavailability of healthy foods while at work:

Well, umm, basically being lorry driver is . . . the diet is shocking, you can’t really get healthy food it’s all fast food. Basically working the shift that I do I finish my day at the end of the night, around 11 o’clock at night. And if I’m parked up somewhere, the only thing that’s available is McDonald’s [McDonald’s Corporation, San Bernardino, CA, USA], or Burger King [Jacksonville, FL, USA] or sandwiches from WHSmith [WHSmith, Swindon, UK].

. . . you’re feeling knackered, you’re feeling drowsy, you’re feeling, you know, you need to be doing something, you eat. And unfortunately that’s the nature of the beast. And it’s not necessarily good for you, but you can also have the mindset, well, it’s better that I eat than crash. Because I could kill somebody, or myself, you know. And it’s one of those, umm, it’s not ideal, and I do try and eat more healthily when I’m snacking, but sometimes you are thinking an apple ain’t doing it, a banana ain’t doing it, grape ain’t doing it, dried fruit’s not doing it. **** it, I’m having a choccy bar.

But yeah, the biggest problem is, when you go to a service station, what’s the first thing you see there? It’s fast food. They are trying to change it but it is fast food, burger bars, so . . . They are popping up aren’t they. You don’t see enough healthy food options, certainly in the UK, and whether that’s because of the climate we have I don’t know.

Less mentioned, and more debated, was physical activity, with some drivers suggesting that they do a lot of physical activity at work and others saying that they do not:

Well, the thing is my exercise regime did go out the window a little bit. Because I was starting an hour earlier . . . I was starting at . . . getting up at 2 and starting at 3 in the morning. And of course you know, that knocks you right up, you know, you come in and all you want to do is have a sit down an hour, half an hour, you have your food and then you’re back in the bed, you know.

It’s a shame, we haven’t sort of got a place, a room or something with gym equipment in it. Because I mean, you know, for example, like drivers, they sometimes sat down there for like 2 or 3 hours waiting for a job. I mean, if you have like gym facilities on site, rather than just sitting there, you could just come in and do a bit of something, you know what I mean?

Yeah, so, and the pumping weights, well I do enough of that at work, loading and unloading trailers.

My job is quite physical anyway. Umm and you know you unload like 300, 400 tyres, umm, so yeah you know that’s quite a good workout.

Self-motivation was also a key barrier that was mentioned, and this appeared to be at both home and work:

Just being lazy . . . That’s it. Me just telling myself ‘oi get your big fat ass out of there and go for a walk!’ or do this, or do that!

I just think it’s that lazy mindset really, like I said I’ve been preparing, like when I get home now like, my fruit or anything that I’m going to bring with me today, umm, it was a lot easier before to maybe just get to shop to buy sandwich. It’s pretty packed. It’s just lazy that mindset. Isn’t it?

Suggested improvements to the SHIFT intervention

By a significant margin, the most frequently suggested improvement from both drivers and managers was that the SHIFT team could have had more regular contact and engagement with the participants:

I suppose, how could I say it. I think if I was to say anything, and this is not being negative or nothing, but you just asked to think of something so . . . Umm, maybe just the odd phone call just to cheer you up and keep you going. You know what I mean . . . Barring that, I couldn’t imagine anything else. You know what I mean. But yeah, a bit of a phone call every now and again, how are you keeping, how are you doing? We’re keeping an eye on you like. If you know like, someone is watching you, it makes you, it pushes you along a bit more. You know what I mean.

I haven’t really text backwards and forwards, but when you sent me the, the little challenges like I thought good, yeah, I enjoyed that. And that’s one thing I would say. There could have been a lot more of that. Yeah, a lot more challenges, like, you know, you know, let’s say, look, everybody’s gotta hit 15 . . . But if there was more goals in there . . . And sort of push us along because I would rise to the occasion with the goals. I like that sort of thing. If I’ve got a target, that’s my goal that, you know me. But when there’s no targets, you tend to take a backward step.

I think for them they were given all the information and then there was quite a long gap, so that was the only negative thing I would say, that gap was too long, they were saying what’s happening and they were starting to lose interest.

Lots of drivers and managers would have liked to have seen more feedback about the results:

Yeah, just better . . . more publication of results for, for me, personally, I would have got something out for that. And I would have probably been able to get more of the guys talking about it if they have that data as well.

I’d have liked to have been able to have access to the results of the monitors that we were wearing to see over the period of time that we were wearing them how things fluctuated as well.

From the perspective of the managers, more clarity at the beginning would have been helpful in organising and managing expectations:

Umm, I think the only thing for me would be, at the very, very beginning, having a more understanding of what was involved . . . [exhales] it was a little bit, um, it was a little bit, ‘it won’t affect your business, it won’t affect your everyday’ bla bla bla, bla bla bla. But it does affect it when it’s 2 or 3 hours and it, and you’ve got to umm, plan that around the drivers’ times, rest days, and all that kind of stuff. Umm you know that alone, the admin side from myself or one of my managers, to prep it, umm, probably, it’s a few hours in a day, each time, to try and prep that. And it’s not about not doing that part of the role, I don’t mind doing it, but it’s about understanding that’s what is involved. Because at first it was sort of addressed, that ‘yeah, there’s no problems, there’s no impact to yourselves’, etc., etc., etc., but that is 2 or 3 hours of the drivers days that then have to rearrange his job for the day so that he can complete his job within legal time.

Future sustainability

All managers were asked if they could see the SHIFT intervention becoming sustainable in the future, and if they would support it. All managers said that they would support the SHIFT intervention:

Absolutely. 100%. I can say that without even speaking to my, my higher tiers . . . But I know they’ll be on board, they will, yes . . . Again, it’s all about the well-being of drivers isn’t it?

I think to have it as part of a CPC [Certificate of Professional Competence] I think certainly, would be, be a much easier route and you’d obtain a lot more numbers to be involved.

However, some managers caveated this support with comments about issues with practicalities:

Umm, I suppose within DHL to get everybody done for the assessments and stuff would take some major planning. You know what I mean, just on my contract alone you’ve got 650 drivers. Umm, within DHL you’ve got 7500 drivers. Umm, so you imagine trying to have consultations and assessments with their health and things on site with 7500 people . . . Umm, and that’s just like 7500 driving colleagues, staff at the moment. Within DHL . . . It’s quite big . . . If you want to start planning your days around, you can plan 7500 in if you want to haha . . . Keep you busy until 2029 I think! Haha.

Yes, it could. Ummm, you would have to work around sort of different areas or different drivers, because not all drivers would want to participate. Not everybody would want to be part of it. Like I said, in the planning of the drivers and getting the drivers back, it’s sometimes not that easy to be able to manage small contracts and get drivers back on site, the time that you need them.

Intervention participants were asked about their thoughts on turning the education session into a Certificate of Professional Competence (CPC) module, and all drivers said that it would be a good idea:

Now that would be interesting that would. That would be interesting . . . yeah, I think that would be a good thing actually. It would be well educational actually for a lot of the drivers.

However, almost all participants said that the education session would need a medical professional rather than a driver trainer to lead the workshop:

Well I think I’d prefer it to be a health professional, uh, because it just seems more appropriate that it comes from a health professional. You’d think that you’d take it more seriously.

But no with that thing I think a health professional could present it better than a driver trainer because driver trainers understand driving but they don’t understand . . . other things . . . it’s always better to have an expert talking about something they know, then somebody talking about something they’ve been told to talk about.

No . . . because they haven’t got the depth of knowledge that you guys have, I don’t think it would work. I really don’t think it would work. All they’d do is just read it. And then ask you five questions at the end of it and say right, you sign that, then sign the form to say that you understood. And people will just be saying, god is it time to go home?

One driver mentioned that cost would prevent DHL Supply Chain from doing the CPC module:

They have a programme that the whole of DHL use and that’s all you’re doing, not doing nothing else. Because we may have to register for something else, or we may have to pay out for something else. So we’re not going to do that. So . . . companies would then have to pay for it, they’d have less interest . . . Is it going to help them get a truck down the road?

All drivers said that moving the education session to an app-based workshop would be detrimental:

I mean if you’re in the classroom and that, with other people and that, then you get involved more, I think if you’re on your own, doing an app, I think a lot of people probably won’t bother.

I’d rather have face to face because you can ask questions and that can’t you really. And if you’re there with a group of people, which you can’t do at the moment, everybody would be coming up with different questions . . . I might be asking one question and the guy next to me asks another question that I never even thought about asking.

Being honest. I wouldn’t have paid much attention to it at all. I’d have flicked it on and flicked it back off just to say I’d been on it! I’m being totally honest. I’ve always worked with face to face, and I work well that way. I’m more of your audio kinetic sort of learner and everything. Umm, on a web page or on an app, you’re going to have loads of writing and all that sort of thing. I’m not going to read it and that’s being honest.

Control site participants

There was a lack of awareness from participants and managers in the control group with regard to whether they were in the control or intervention arm, suggesting that the SHIFT team could have provided more communication and explanation to both the managers and participants of their allocated arms and what this means for participants:

Contamination

Randomisation occurred at the site level to ensure that there was a minimal risk of contamination between intervention and control participants. All participants and managers were asked in the interviews if they had spoken about the SHIFT intervention to anybody from any other depot, and all participants and managers confirmed that they had not.

The COVID-19 pandemic

A large, unforeseen confounding variable to the SHIFT intervention was the COVID-19 pandemic, which caused three major lockdowns in the UK from March 2020 to June 2021. Fortunately, almost all (254/262, 96.9%) participants who were still enrolled in the trial at the time had completed the 6-month follow-up health assessment before the national lockdown. However, the final follow-up was greatly affected by the pandemic and, as such, all main trial sites required remote data collection, where participants were asked to fill out the questionnaires, self-report their weight, wear the activPAL and complete the sleep/work diary by themselves. Although this method has clear shortcomings, this was deemed the most appropriate, pragmatic and safe form of data collection at this time to assess the key main outcomes. When the COVID-19 pandemic was discussed in the interviews, it appeared that all sites were affected by the pandemic in different ways. For example, participants who delivered essential items were busier and worked longer hours:

Some drivers mentioned that it was business as usual:

It was the same run on the same start times, so not many changes if you like.

The same, whatever happens. Our job is the same. It never ever changes. We go out, two runs a day. Fourteen pallets, it might be a bit less. But it all depends how much the shop’s selling. But yeah, that’s it, it never changes. It’s just the same every day.

Some drivers mentioned being on furlough for most of the year. However, there was ambiguity between the furlough status of all sites. Several sites furloughed staff on a rotational basis, some sites furloughed all staff, other sites made it optional for drivers to choose furlough and the remainder of sites, which provided essential goods, did not make furlough an option:

I was put onto furlough, I was put on that until August . . . then we were locked down again in November, so we were put on furlough then . . . Since April I have probably done less than . . . 10 days’ work?

Some managers also mentioned that the pandemic halted the enthusiasm for the study, with priorities moving elsewhere:

It’s just after that the, the communication and the engagement because of the COVID-19, where that stopped, then obviously, this . . . the SHIFT stopped as well, well in the drivers’ minds. So they just said ‘it’s not continuing anymore obviously’. Nobody’s ringing us, and everybody else had obviously their personal interests in their minds during this pandemic as well. So they just lost focus on it.

Behavioural changes

The primary outcome measure was the difference in steps per day at 6 months between the SHIFT and control arms, measured using the activPAL3 accelerometer. Although a significant difference in daily steps was observed at 6 months, in favour of the SHIFT arm, drivers (from both trial arms) commented that the biggest modifications they made were to their diet, with 57 comments/references to dietary changes recorded, whereas only 27 comments/references were made with regard to changes in physical activity. There were no significant differences in fruit and vegetable intake or dietary quality observed between trial arms at either follow-up period:

Well, the barriers are still there, but you’ve just got to, it just puts thoughts in your mind because of the training you’ve had, I consider this to be like a training exercise to me, you know, I’m thinking about things before you just go in. So I used to like, when I got meals at McDonalds, I used to get the large meal. Now, I never do that. Always just your standard meal. I know there might not be many calories difference between a large and a medium meal, but there is some difference. So, that’s, that’s what I do now, and I manage with that. You know, I used I used to think I’ve got to have a large meal to be sort of, satisfied but now I don’t. I just have a standard meal, I have that, and that’s it.

I don’t eat as much chocolate as I used to do now. I take an apple or something like that for work, whilst I’m travelling down the motorway.

The small snack thing, before I just, I don’t know. I’d eat . . . in the truck for example, I’d always have a . . . I don’t know, say humbugs or something. And I’m eating a packet of biscuits and you’d just happily munch on them. Whereas now, I deliberately won’t have them in the vehicle, because if they’re not in the vehicle, I can’t . . . I can’t . . . do you know what I mean? I can’t eat them . . . I’d rather take like a bag of like roasted monkey nuts or something like that instead.

I’ve started eating more fruit . . . Rather than when I get hungry going for a chocolate bar, I go for some grapes, a banana or an orange or something like that.

My diet changed very much straight away . . . from not eating any fruit because I wasn’t interested, because I was a biscuit sweet person, I take fruit with me at work, I got down from nearly 14 stone to 13 [stone], 3 [pounds].

At 6 months, significant differences in steps, sitting and standing time were evident on non-workdays between trial arms, with no differences in these variables seen between groups on workdays, and this is reflected by the drivers’ and managers’ opinions about the inherent characteristics of the job: