A peer-led physical activity intervention in schools for adolescent girls: a feasibility RCT

Phase 1 objectives

1. Adapt and refine the ASSIST intervention to develop a PS training programme that focuses on promoting PA among Year 8 girls.

2. Develop an intervention logic model to refine in the feasibility trial.

Phase 2 objectives

1. Estimate the recruitment rate of PSs and non-PSs (NPSs), and monitor attendance at the PS training.

2. Qualitatively examine the acceptability of the intervention to students, trainers, schools and parents, and identify necessary refinements.

3. Report accelerometer and questionnaire data provision rates, examine data quality and explore the implications of missing accelerometer data in terms of how these data might be imputed in a definitive trial.

4. Estimate the potential effect of the intervention on daily accelerometer-derived MVPA and secondary PA-related and psychological variables immediately after the intervention (T1) and at 12 months after baseline (T2).

5. Estimate the school-related intraclass correlation (ICC) for daily MVPA.

6. Estimate the sample size for an adequately powered definitive trial evaluation.

7. Identify and test the feasibility of collecting the data needed to cost the intervention and conduct a cost-effectiveness analysis in a definitive trial.

8. Qualitatively examine parental views regarding allowing their child’s data to be used for data linkage with academic records kept by the National Pupil Database (NPD). Examine the completeness of data required to link participant data to educational attainment (via the NPD).

Recruitment and description of the pilot study school

One pilot school was identified by Wiltshire Council to be above the median of the Pupil Premium (i.e. relatively more deprived) and which would give a robust early test of the intervention. Recruitment materials were sent to the head teacher, who then delegated responsibility to a school contact (assistant leader in learning in PE). A school study agreement form was completed. The school (N = 1086, students; n = 70, Year 8 girls) was located in a medium-sized town. The school was given £500 as a thank you for participating.

Iterative focus groups: round 1

Iterative focus groups: round 2

Summary

Phase 1 resulted in the co-production of the draft PLAN-A, which was taken forward to pilot testing. The ASSIST intervention framework was successfully blended with the views and preferences of the user group. A description of how these findings were incorporated in to the final intervention is presented in Appendix 1.

Aims and objectives

The objective of the pilot study was to test the student recruitment approach and rehearse the peer nomination and PS selection processes, train the trainers and test the PS training (see Chapter 2, Description of the final intervention elements for details). In evaluating the delivery and receipt of the pilot intervention to identify refinements, the process evaluation methods were also piloted.

Methods

Results

Participant recruitment, peer nomination, peer supporter selection and recruitment

Participant recruitment flow is summarised in Figure 3. Of the 70 Year 8 girls, 54 (77%) attended the study briefing. Three (4%) study opt-out forms were returned. A total of 54 girls completed the peer nomination questionnaire, and the two researchers who led this process were satisfied that no amendments were required. The PS selection process was successful and 14 girls (20%) were invited to be a PS, all of whom attended the briefing meeting. The meeting was well received and 11 girls (16%) provided assent and parent consent to participate in the training.

FIGURE 3.

Flow diagram of PS recruitment in the pilot study. a, Number of Year 8 girls invited to PS meeting was > 18% as, in accordance with the ASSIST protocol, any student with an equal number of nominations as the final nominee was also invited to the training.

When PSs reflected on the peer nomination process they commented that the majority of PSs were from the same friendship group (many from the netball team) and believed that most of their friends were physically active:

. . . a lot of the people chosen are from the same friend group anyway, so there’s a lot of different people in the school who won’t have any friends that will have been voted for . . .

Post-training focus group 1

One participant suggested that rewording the peer nomination form to refer to ‘their closest friends’ could increase the diversity nominations across friendship groups:

I think maybe if you’d got everybody to write down, I don’t know, their five closest friends, then you could have a person from every friendship group.

Post-training focus group 1

Reasons PSs gave for wanting to be a PS included wanting ‘to help me help more people’ and personal skill development:

. . . you could benefit a lot of skills, not just for use in school but for when you get older for your jobs and stuff like that.

Post-training focus group 1

One participant took part because she thought the course would be focused on playing sports:

I thought it was to do with sports, to be honest.

Post-training focus group 2

During the PS meeting, participants suggested that they would prefer to wear PE kit, which the school contact agreed with. For the feasibility study, a decision was taken to recommend that the girls wear comfortable, casual clothing so as to (1) highlight that the training is not a sports event, (2) avoid the potential negative connotations of PE or PE kit for some PSs and (3) separate the PLAN-A training from school.

Peer supporter training

Attendance

Attendance was 91% (n = 10) at both training days; one girl did not attend because of a disagreement with another PS outside the PLAN-A training.

Logistics and resources

The PS training space and venue were rated ‘adequate’ by both trainers (Figure 4) and PSs (see Figure 7). This was largely because of its small size and lack of break-out space. This led to an environment in which it was difficult to concentrate. The observer also felt that the layout of the room (including a lack of tables) led to a very informal learning environment. Trainers and PSs rated the refreshments for the training highly (see Figures 4 and 7).

FIGURE 4.

Trainer ratings of PS training arrangements.

The training manual and resources were considered to be ‘very good’ and ‘good’, respectively, by the trainers (see Figure 4). Trainers agreed that the session plans were helpful and clear, but suggested condensing the quantity of information to make delivery easier:

There’s too much going on on the page . . . I don’t think you can take that in and use that as a thing to use in the classroom, for me.

Trainer 1

As a result, cue cards were developed that allowed the trainers to write their own notes about each activity to use in the PS training.

The girls found the PS booklets to be a helpful prompt of key information and good to show others what they did. They suggested that the booklet should be used more for reference by PSs rather than using it to write in during the training:

If you were to forget something, then you have a booklet to look back on. I feel maybe instead of copying it down you could have it written on the board and already written in the booklet.

Post-training focus group 2

The PSs felt that the PS diaries were a good idea; however, during the intervention the girls reported not using them because they forgot about them or did not want to record conversations:

I just don’t need it. I just can’t be bothered with it.

During-intervention focus group 2

Fidelity to the session plans

Given the time constraints, on day 1 two activities were not delivered: the ‘thought shower: why be active?’ (10 minutes in duration) and ‘post box’ (10 minutes in duration) activities. On day 2, the ‘relaxation activity’ (5 minutes in duration) was excluded.

The allocated time and observed delivery time for each activity are shown in Appendix 5. On day 1, three out of nine activities before lunch took more time than allocated to deliver (e.g. ‘what do we know already?’). Lunch was slightly longer than scheduled, which resulted in activities after lunch being excluded or shortened. On day 2, activities before and immediately after lunch were an average of 12 minutes longer than scheduled, again leading to activities in the afternoon being given less time. Qualitative findings supported the observations: trainers felt that there was too much content to deliver and a lack of time for discussion. As a result, both trainers suggested that the training should be extended and more time should be added for discussion:

It’s too much in 2 days . . . I think it would have been nice to spread it over 3 days if you keep the activities, have a bit longer for each of them.

Trainer 2

The observer recorded that trainers added in active games to energise girls when necessary. Despite intending to benefit the girls, these contributed to the drift from the schedule. In addition, the observer noted that trainers could have shortened some sessions that were longer than intended:

Too long was given for discussions amongst small groups/activities . . . The girls would then talk about ‘off task’ things . . . time could have been used better.

Observation note

The extent to which PS training objectives were met is reported in Table 3. For timetabled sessions that comprised more than one distinct activity, the mean objective fulfilment is reported. Seven activities (41.18%) across both days were not rated. The mean rating for activities was 3.00, suggesting that objectives were achieved. However, fulfilment of objectives was lower (1.00) for activities at the end of day 2.

TABLE 3 - Fulfilment of pilot PS training activity objectives

Rating based on 0 = not at all, 1 = a little, 2 = to some extent, and 3 = lots.

Activities are listed in the order in which they occurred.

Day
1		2
Activity	Fulfilment, mean (SD)	Activity	Fulfilment, mean (SD)
Introduction	3.00 (–)	Empowering girls’ confidence	3.00 (0.00)
Getting to know each other	3.00 (–)	Holding conversations	3.00 (0.00)
Knowledge is power	2.67 (0.58)	Communication skills	3.00 (–)
Recognising your skills	3.00 (–)	Gathering information	3.00 (–)
A day in the life of a Year 8	3.00 (0.00)	Tricky issues	1.00 (–)
Busting barriers	3.00 (–)	Finishing up	1.00 (–)
Day average	2.89 (0.33)	Day average	2.50 (0.93)
Average objective fulfilment	2.71 (0.69)

Both trainers felt they successfully enhanced PSs’ knowledge about PA, communication skills, interpersonal skills and understanding of their PS role (Figure 5). The objective to enhance PSs’ confidence to spread informal messages among their peers was not met well (1.5 ± 0.71); this supports PSs’ reported low confidence in this area (see Peer supporters’ experience of training).

FIGURE 5.

Trainer-rated achievement of pilot PS training objectives.

The qualitative data gave insight into why some objectives were not achieved fully, including losing focus because of activities over-running (and other activities being rushed), omitting key activities and trainers not being clear on general group rules:

When it came to the core message of it and becoming a PS, I think we did not do enough because we physically did not have enough time.

Trainer 2

Peer supporters’ experience of training

Peer supporters’ engagement with training was high on day 1 and slightly less so on day 2 (Table 4). A need to maintain engagement in the ‘communication skills’ and ‘finishing up’ activities was identified.

TABLE 4 - Observed PS engagement per pilot activity section

Rating based on 0 = not at all, 1 = a little, 2 = to some extent, and 3 = lots.

Day
1		2
Activity section	Engagement, mean (SD)	Activity section	Engagement, mean (SD)
Introduction	3.00 (–)	Empowering girls’ confidence	3.00 (0.00)
Getting to know each other	3.00 (–)	Holding conversations	3.00 (0.00)
Knowledge is power	3.00 (0.00)	Communication skills	2.00 (0.00)
Recognising your skills	3.00 (–)	Gathering information	3.00 (–)
A day in the life of a Year 8	3.00 (0.00)	Tricky issues	3.00 (–)
Busting barriers	3.00 (–)	Finishing up	2.00 (–)
Day average	3.00 (0.00)	Day average	2.67 (0.50)
Average engagement	2.83 ± 0.38

Figure 6 shows that the trainers felt the PSs were very involved throughout the training. They perceived the PSs to be somewhat engaged and interested. PSs wanted to discuss topics and ask questions throughout:

Every time we did something they wanted to talk about it. They were quite chatty and so they wanted to talk about things.

Trainer 1

That was quite good because you had to debate your point. That was quite good.

Post-training focus group 1

FIGURE 6.

Trainers’ ratings of PSs’ responses to the pilot training.

Girls were particularly engaged in activities that they could relate to and that involved being active:

It worked well because the video is so powerful. It was a ‘visual’ and some of them probably can relate to that . . . that’s why, because they can relate to those questions.

Trainer 2

However, although most activities were rated as highly engaging, there was evidence of some boredom, fatigue and distraction because of the ineffective timings of activities (i.e. rushing activities at the end of day 2), overuse of writing rather than more physical tasks and a disagreement between PSs on day 2:

Some silly[ness], some look bored [afternoon] day 2? Lethargic.

Observation notes

Some became distracted or I saw it in their face. OK, they might be looking but I could see they’re still not there.

Trainer 2

I just think there could have been more activities. . . . We probably learn more by doing stuff than writing because I think we get a bit bored writing.

Post-training focus group 1

Enjoyment over the 2-day training was moderate to high and was marginally higher on day 2 (3.80 ± 0.79) than on day 1 (3.30 ± 0.68). Open-ended responses revealed that all girls enjoyed the ‘listening train’ activity most and 60% felt they were good at role-play tasks. Activities that were more active and involved discussion and debate were also enjoyed, which supported trainers’ perceived drivers of engagement (see above):

I liked the big discussion we had about fitness between boys and girls and self-confidence. I liked that big debate.

Post-training focus group 2

Peer supporters believed that the training helped them understand their role; they enjoyed the games and learnt something new (Figure 7). Eighty per cent reported five different facts they learnt about PA in free text. The PSs felt prepared to support their friends by enhancing their knowledge, confidence and ability to empathise with others:

It does actually work, that you feel like you can go out and talk to people about it. It’s not just something that you do and then you forget about it.

Post-training focus group 1

Like, if one of your friends has an insecurity, you know how to understand them and how to talk to them about it.

Post-training focus group 1

FIGURE 7.

Peer supporter ratings of pilot training content.

Trainers expressed concerns about the PSs’ ability to be realistic about how to peer support. This was also reflected in observation notes that questioned how empowered the girls were to make changes at school. Similarly, 60% of PSs suggested that more time could be devoted to practising being a PS, which would further enhance their confidence to be a PS (see Figure 7). Some PSs were unsure of how to use the facts they had learnt or how to start a conversation with their peers:

It feels like if we have all these facts – like, you can’t just casually slip a fact into, like, a conversation that you’re having with someone.

During-intervention focus group 2

Peer supporters thought that including more activities that focus on building confidence would improve the training:

Like with the confidence, we didn’t go a lot over how to be confident.

Yeah, because that’s like a major thing. If you’re going to help someone else you need to be confident in yourself as well.

Post-training focus group 1

Need-supportiveness of trainers and perceptions of need-support

On average, PSs reported relatively high levels of trainer autonomy support (3.42 ± 0.47; range 2.80–4.00) and thought that trainers were helpful (3.80 ± 0.42) and friendly (4.00 ± 0.00). Similarly, the observation documented that the trainers circulated among the girls, interacted with them and helped them with their tasks. The girls respected the trainers, commenting that they were different from teachers, made the training fun, were trusted, not boring and ‘understood us more than actual teachers’:

If you said something to them, you really felt you could trust them. They were just very friendly and stuff.

Post-training focus group 2

Autonomy support

Trainers provided autonomy by giving PSs choice within a structure of clear instructions and the opportunity to lead discussions and express their opinions:

They gave us lots of choices. They kind of said what to do but then they were giving you choices of how to do it, which was better than everybody saying, ‘you’ve got to do it like this’.

Post-training focus group 1

Trainers empathised with the PSs, and when the training required concentration or when engagement was low, they responded by incorporating elements that they felt the girls enjoyed:

In the lunchtimes, they were desperate to move. We were getting the tennis balls out and [Trainer 2] got some tunes on and was getting them dancing.

Trainer 1

Competence support

Trainers frequently provided competence-based feedback, praising the girls’ efforts. Trainers circulated around the group asking questions and they provided help when needed. Both trainers reinforced the girls’ understanding of activities and provided them with the skills to become a PS, in particular ensuring that the girls were realistic about their role:

We talked a lot about being realistic. You’ve got to be realistic, so it’s not saying, ‘Right, let’s go for a 2-hour run.’ ‘Oh, but I don’t like running.’ Where do you go from there? ‘I want to watch TV.’ ‘Let’s go for a half-hour walk and then come and watch TV.’ So it’s that compromise . . .

Trainer 1

Trainers explicitly tried to help the PSs understand their role and spent time empowering the girls’ confidence to be a PS:

I think if they feel confident and empowered then they can go off and do anything. And I know that was quite a big part of the course, and I think we spent a lot of time doing that and I hope and think that went in.

Trainer 1

PSs, competence may have been undermined when the trainers rushed activities. Sometimes the PSs felt confused, and trainer 2 felt that some core messages were missed because of time constraints:

When it actually came to the core message of it and becoming a PS, I think we didn’t do enough of because we physically didn’t have enough time.

Trainer 2

Relatedness support

Both trainers endeavoured to, and were successful in, forming trusting relations with the PSs within the short training window. Trainers valued the PSs’ opinions, empathised with them and were interested in them as individuals. The PSs described the trainers as ‘friends’ and felt that the trainers understood them:

They [the trainers] related well to the girls and their delivery style supported SDT. Connections between the girls and with trainers [are] clear.

Observation notes

They were trusting. If you said something to them, you really felt you could trust them. They were just very friendly and stuff.

Post-training focus group 2

I could see they’re still not there [engaged], and when they sat down is when I suggested different ideas as a group, because I know when I was that age I was a really, really shy girl so I wouldn’t want to say anything.

Trainer 2

Template for intervention description and replication checklist

Table 7 summarises PLAN-A in accordance with the Template for Intervention Description and Replication (TIDieR) checklist. 57

Protocol amendments

The original study protocol was submitted to the National Institute for Health Research (NIHR) on 26 April 2013 and was published on 15 January 2016. 55 A small number of revisions were made to the protocol following NIHR approval during the study and are summarised in Table 8.

Ethics and governance

The study was granted ethics approval from the School for Policy Studies, University of Bristol (SPSREC14–15.A27), and was sponsored by the University of Bristol. Parents were given study information and parental informed opt-out consent was obtained covering both the quantitative and qualitative data collections. Parents of PSs provided written informed consent and students gave written informed assent. All adults involved in the research (e.g. trainers) gave written informed consent. Adverse events during data collection or when the intervention was being delivered were recorded and reported to the chairperson of the Trial Steering Committee (TSC) and the chairperson of the Ethics Committee.

Sampling and participants

To be eligible for inclusion in the study schools had to:

• be state-maintained mainstream secondary schools

• be located in Wiltshire and South Gloucestershire

• have girls in Year 8

• be above the median of the local Pupil Premium (i.e. more deprived: Wiltshire = £115,000 and South Gloucestershire = £145,000 for 2014/15)61

• not be currently implementing the ASSIST study intervention.

Eligible schools (N = 30: South Gloucestershire, n = 9, and Wiltshire, n = 21) were invited to participate in writing, followed by follow-up telephone calls and e-mails. In collaboration with Wiltshire Council, PLAN-A staff were advised to initially contact four schools that had previously proven difficult to engage in research and reach with health-based interventions. Consenting schools were selected on a first-come-first-served basis in line with the stratification by local authority (LA). A school contact teacher was designated to the study and a short, standardised briefing was delivered to all Year 8 girls in each school. This explained the study, participation, randomisation, data collection and PLAN-A. Students were able to ask questions and were provided with written student and parent information sheets and a parent opt-out consent form.

Trainers were recruited using snowball sampling and word of mouth, with the aim of reaching trainers with expertise in both PA promotion and youth work/drama. A recruitment advert was distributed to relevant organisations (e.g. Active Trowbridge, Nuffield Health, Travelling Light Theatre and Youth Action Wiltshire). Interested individuals contacted the study team and had an informal meeting to discuss the project and assess their fit with the recruitment criteria set out by Year 8 girls in the formative phase.

Sample size

As this was a feasibility study, a sample size calculation was not performed. We chose to recruit six schools to provide sufficient variety of schools and student numbers in which to test recruitment, retention and acceptability of the intervention, PLAN-A, and feasibility of the research to evaluate it. Participants were Year 8 girls and all girls in Year 8 were eligible to participate.

Randomisation

The unit of randomisation was school. After baseline data collection was complete, six schools were randomly allocated, stratified at an intervention-to-control ratio of 2 : 1 within the LA area (i.e. Wiltshire and South Gloucestershire). Two schools were allocated to the control arm and four schools were allocated to the PLAN-A arm by a member of the Bristol Randomised Trials Collaboration (BRTC), who was independent of the study and blind to the school identity. The lead and research associate statistician and all team members apart from the project manager, research assistants and fieldworkers were blind to allocation.

Intervention and control group provisions

The four intervention schools received PLAN-A, as described in Chapter 2, Description of final intervention elements. One PLAN-A school was not able to release a member of school staff to chaperone students to the off-site PS training and, as such, it was negotiated with the school that training would be held in the school. Peer nomination occurred in September 2015 and PS selection and recruitment took place in December 2015, with the initial 2-day PS training in February 2016 and the 1-day top-up training in April 2016. Control schools did not receive PLAN-A and continued with normal practice; all students completed identical measurements to the PLAN-A schools.

Measures

Accelerometer and questionnaire measures were collected at baseline [time 0 (T0): the beginning of Year 8, September–October 2015], immediately post intervention [time 1 (T1): the end of Year 8, May–June 2016] and follow-up [time 2 (T2): the beginning of Year 9, September–October 2016]. All questionnaire measures at each time point, apart from the EuroQol-5 Dimensions, child-friendly version (EQ-5D-Y) and KIDSCREEN,62 were completed using a Samsung Galaxy Tab 4 (Samsung Electronics Limited, Surrey, UK). Demographic information was collected at T0 only. Process evaluation measures were taken throughout the course of the project with key times being the training observations and evaluations in February and April 2016 and qualitative work in February, April and May 2016. Students received a retail voucher (£5 for T0 and £10 for each follow-up, i.e. T1 and T2) in recognition of the time given to each data collection. Participating schools (i.e. PLAN-A and control) received £500 and a summary of the findings in recognition of the time devoted to accommodating the study.

Health economics methods

Process evaluation methods

A mixed-methods process evaluation was conducted to document and assess training attendance, receipt of training, intervention fidelity and the experiences of being a PS and to identify further intervention refinements. The process evaluation also sought to understand the context in which the intervention was delivered. The majority of the methods used are identical to those used and described in the pilot study (see Chapter 2, Pilot study). As such, reference will be made to these descriptions with additional methods described in Process evaluation quantitative analysis and Process evaluation qualitative analysis. Initially, however, the quantitative measures used and the different qualitative data collected will be described.

Data analysis

School recruitment

There were 46 secondary schools in the two study areas; 30 schools were ineligible (26 were below the median Pupil Premium). The 16 remaining schools (eight in each LA area) were invited to participate. Eight (50%) schools accepted (six feasibility schools, one pilot school, one reserve), one (6.25%) declined after expressing initial interest and seven (43.75%) failed to respond to the study invitation.

Participant (student) recruitment

The recruitment briefing was delivered to 451 Year 8 girls. Four girls (0.89%) returned opt-out forms and 20 girls (4.43%), despite not handing in an opt-out form, chose not to take part resulting in a 94.68% participation rate across the six schools (Table 10). A total of 427 girls from six secondary schools were recruited (PLAN-A arm, n = 269; control arm, n = 158).

Peer supporter recruitment

Peer nomination resulted in 57 girls being invited to be a PS; 56 attended the PS meeting and 55 (96.49%) consented to the role. One PS withdrew before the training citing reasons including low confidence and PA motivation, resulting in 54 PSs. The number of PSs per school ranged from 11 (16.67% Year 8 girls) to 17 (17.71% Year 8 girls) (see Table 10).

Trainer recruitment

Expressions of interest were received from 10 potential trainers with a range of experience in youth work, PA or both. Five female trainers were recruited: two had experience in drama and working with young people and three had experience in PA and working with young people (Table 11).

Physical activity variables

Table 16 shows that at T1 (immediately after the 10-week intervention period, i.e. the end of Year 8) there was no evidence of a difference between PLAN-A and control groups for weekday MVPA or weekend day MVPA. There was some evidence that the girls in the PLAN-A arm were less sedentary than the control arm on weekdays at T1 (–31.8 minutes, 95% CI –57.44 to –6.18 minutes). In addition, there was some evidence that the control arm performed a higher volume of activity (i.e. CPM) at weekends, although the CI was wide (–124.24 CPM, 95% CI –213.48 CPM to –35.01 CPM).

At T2 there was evidence for a between-group difference in weekday MVPA in favour of the PLAN-A arm (6.09 minutes), with the 95% CI exceeding 10 minutes of MVPA per day (95% CI 1.43 to 10.76 minutes). The data suggested that the between-group difference was based on the PLAN-A arm girls maintaining their baseline level of MVPA, while the MVPA of the girls in the control group decreased over time. Similar to T1, there was some evidence that the girls in the PLAN-A arm were also less sedentary on weekdays at T2 (–23.26 minutes, 95% CI –43.73 to –2.79 minutes).

Sensitivity analysis

Psychosocial variables

Internal consistency statistics (Cronbach’s α) are reported in Appendix 9. With the exception of the peer PA norm (acceptance) variable, which exhibited low internal consistency at all time points, all psychosocial composite variables had good internal consistency at all time points. Table 19 shows the adjusted multivariable regression results for psychosocial variables. There was little evidence of any differences between the PLAN-A and control arms on any of the variables at T1 or T2. There was no evidence of harm or detriment to these variables because of the intervention.

Physical activity social support

Figure 10 shows that in the PLAN-A schools, immediately after the intervention period, there was an increase (9%) in the number of Year 8 girls believing that someone in their year group had spoken to them recently about PA, compared with a decrease (12%) at this time point in the control arm schools. At T2, scores in both schools had returned to approximately baseline values, albeit slightly higher relative to baseline in PLAN-A schools. Figure 11 shows that, at baseline, a similar proportion of girls in control and PLAN-A schools thought that talking to someone in their year had helped them to be more active (41–42%). However, at T1, although this level of belief was maintained in PLAN-A schools, in control arm schools there was a 5% decrease. At T2, 34.7% and 27.8% of girls agreed with this statement in PLAN-A and control arm schools, respectively.

FIGURE 10.

Responses (% yes, no, not sure) to peer support question 1 ‘has anyone in your year group talked with you recently about PA?’ at each time point.

FIGURE 11.

Responses (% yes, no, not sure, NA) to peer support question 1 ’did talking to anyone in your year help you to be more active?’ at each time point. NA, not applicable.

Outcome of peer nomination

The peer nomination process was largely successful. A small number of students copied others or felt pressure to nominate those that their friends had, but the majority of students nominated confident or active peers as they thought they would make legitimate PSs. Consistent across all schools and stakeholder groups was the perception that the girls selected to be a PS were ‘sporty’, ‘confident’, ‘outgoing’ and ‘self-motivated’, and this was the main reason for their nomination:

I think it’s because we’re all like quite sporty people.

School 2, PS focus group

If you have a non-active person telling you to be active, you wouldn’t listen to them as much would you? Because you think, well why should I, you’re not.

School 3, NPS focus group

Views on whether or not nomination had identified the influential PSs who were representative of the Year 8 girls varied between and within schools. For example, in schools 2 and 4, PSs thought that they represented different friendship groups, whereas NPSs thought that girls from the same ‘popular-type groups’ were selected. In school 2, trainers felt that the girls who were nominated could have been influential for the wrong reasons, describing them as ‘bullies’ (trainer B, top-up training day):

I think there was [sic] quite a few [girls] from lots of different ones [friendship groups].

School 2, PS focus group

There was evidence that some NPSs in school 3 viewed the PSs as potentially influential despite not being close friends:

I’m not close friends with them, but I can talk to them, but I don’t count them as friends.

School 3, NPS focus group

In one school, some PSs felt that their potential network of influence was reduced, as some of their friends were also PSs:

A lot of my friends went on the PLAN-A thing . . . like literally everybody that’s in our like group of friends went, and you kind of ran out of friends [to support].

School 6, PS focus group

Views on being nominated

There was consensus across participant groups that PSs were proud and happy to have been nominated and that they felt their nomination reflected their peers’ trust, support or confidence in their ability to be a PS. Parents suggested that their daughters were ‘shocked’, ‘proud’ and ‘really pleased’ to be nominated:

I felt quite privileged ‘cause like not everyone gets chosen, it’s nice that everyone like us gets chosen over everyone else really.

School 6, PS focus group

She was really pleased I think that people thought she could do it and like I said it just boosted her confidence.

School 2, parent

Consenting to be a peer supporter

Peer supporters were interested in becoming a PS because they liked the PLAN-A concept and wanted to help others be active:

I like the idea of helping others become more active . . . it’s the idea of helping my friends to do more stuff like inside or outside physically active [sic].

School 3, PS focus group

Peer supporters’ parents thought that the role aligned with their daughter’s characteristics, such as ‘wanting to help other people’ (school 4, parent), being ‘caring’ (school 2, parent) and having an interest in sports:

She’s just always looking for something sports wise and activity wise . . . not just for herself. So . . . I think that’s its appeal for her.

School 3, Parent

Non-peer supporters on not being nominated

Most NPSs did not mind or were relieved at not being a PS, providing some evidence that the nomination process was effective in identifying students who were sufficiently confident to influence the PA of their peers:

I didn’t want to do it because I didn’t want to have to meet new people. I don’t like talking to people who I don’t know and that made me feel really awkward about doing it.

School 4, NPS focus group

In contrast, a small number of NPSs were disappointed at not being nominated:

I like, didn’t want to do it, but when everyone else got chosen, I was a bit like, ‘well, does that mean that all of my other year group don’t think I’m good enough to do it?’.

School 4, NPS focus group

Attendance at peer supporter training

Attendance at all 3 days of the PS training was high, ranging between 90.91% and 100% (Table 25). In the majority of cases, absences were because of participant illness. In one case (school 2), a student had already been absent from school in the same week and her parents requested that she not attend. Attendance at training was not different in the school where training was delivered within the school setting (school 4) compared with the other schools where training was delivered off site.

View on the content of the peer supporter training

Table 26 shows that the PSs enjoyed all 3 days of the training. Enjoyment increased from day 1 to day 2 (mean ± SD = 3.60 ± 0.77 and 4.24 ± 0.58, respectively). Enjoyment levels were similar on the top-up day to day 1. Trainers’ perception of PS enjoyment was similar to the PS ratings of enjoyment (Table 27).

Qualitative evidence from all stakeholders suggested that PSs liked the novelty, participatory inclusiveness and autonomy of the training:

I think they enjoyed it. There was laughter, which is always a good sign.

Trainer C

I think it was about it being different; we hadn’t done anything like it before so it was just nice to try something new . . . It was a lot more we were involved, in classes we just get a teacher telling us things.

School 4, PS focus group

PSs found the training activities enjoyable and interesting (see Table 30). In particular, active games, competitions, working together and interactive activities (e.g. role-plays, discussions about videos) were enjoyable (see Appendix 17) and engaging because they differed from usual school-based learning:

I liked the alphabet game. It got us all together and it was like a competition and it made us think more as a group.

School 3, PS focus group

I think the things that are less like school are even better because they think then this is fun day, and then it’s less like they don’t even realising how much they’re learning at the moment, they’re just having fun.

Trainer B

. . . the discussions that come out of that ‘Like a Girl’ video exercise afterwards I think are brilliant in terms of what is going on in their groups . . .

Trainer A

They loved that video [Like a Girl] and they talked about that for ages.

School 2, school contact

Trainers and school contacts felt that the large majority of training activities were well conceived, flowed well, provided a clear message and were pitched at the right level. However, PSs from one school found some wording difficult to understand and struggled to give their opinion:

It probably meant something simple but it was a bigger word that no one really understood and that one didn’t have as many answers.

School 3, PS focus group

A lot of us struggled when we had to give our opinion on why picked stuff [sic] because you feel like you’re either going to be wrong [sic].

School 3, PS focus group

In contrast to the interactive tasks, PSs lost focus and struggled with behaviour when activities covered information that they already knew, and they did not enjoy activities that involved lots of writing, sitting and listening or repetition (see Appendix 17):

. . . the odd little bit that may have been a little bit wishy-washy in the sense that it is something that the kids were already aware of. And maybe that could have been why they lost concentration at times – or was less engaging.

School 2, school contact

It took ages and I think we remembered it more than they thought we would of [sic], so we didn’t really need to go through all of it again.

School 4, PS focus group

I think just sometimes sitting down for a bit longer than other times. . . . I think it is probably listening for a long time.

School 6, PS focus group

Many PSs believed that the 3-day training would be more active/sporty than it was:

I thought it was going to be more active. We talked a lot about being confident and stuff and I thought we were going to play some activities like, I don’t know, run round a field or something like that but we didn’t.

School 3, PS focus group

Although script writing and role-play was relatively unpopular at the 2-day training, it was the most enjoyed activity during the top-up day (see Appendix 17). This may reflect the PSs seeing the relevance of role-play at the top-up training, having had experienced being a PS for 5 weeks or their growing confidence:

I think some of the girls are kind of quite outgoing, confident, so the whole role-playing thing was great for them, they loved it. Whereas other girls, that’s their idea of hell and so for them that isn’t their thing so basically they’re a little bit out of their comfort zone.

School 3, school contact

Trainers reported that PSs also found it hard to understand the goal-setting task:

They found it [SMART (specific, measurable, achievable, realistic and timely) goals] quite hard to do it or they maybe misunderstood that it was in relation to . . . It’s more of a workplace, an adult thing . . . they did struggle a bit to get to grips with that one.

Trainer D

Most PSs would recommend the training to potential PSs, citing that it was fun and educational and motivated them to increase their PA:

You learn so much more and it really does change your mind about things.

School 2, PS focus group

I think it’s [sic] made me more confident.

School 3, PS focus group

The extent to which each training activity objective was achieved was assessed by the observer (see Appendix 18). For the majority of activities, objectives were achieved to at least some extent or lots on all training days and there was little variation between trainer pairs. When learning objectives were not achieved well [e.g. day 2 ‘communication skills’ (mean 1.75, SD 0.71)], the observations showed that delivery of the activity was suboptimal through either poor explanation, poor use of space or a lack of linking the activity to the role of a PS.

Peer supporter engagement with the training

On the whole, trainers and school contacts thought that the PSs had a positive attitude towards the training and their role as a PS. Trainers agreed that the PSs were excited and felt that they were ‘in a privileged position’ (trainer A) to attend the training:

I think the girls were really positive, they were positive about the whole event.

School 2, school contact

They seemed enthusiastic to be there. And they still seemed like something that was important to them.

Trainer E

In some cases, trainers thought PSs viewed the training as ‘a day out of school’ (trainer B, top-up), and in school 2 the trainers felt the PSs ‘were not really that fussed about it [being a PS]’ (trainer C, top-up). These findings are supported by the trainer survey findings (see Table 27).

Peer supporters engaged well in the activities on all 3 days. Engagement decreased throughout the afternoon of day 1, which corresponds with the lower enjoyment levels on day 1 (see Table 26). However, engagement was maintained throughout day 2 and the top-up day. Engagement levels did not differ between trainer pair/school and no particular activities elicited poor engagement. Notes from the observer’s and trainers’ evaluations support this; trainers rated the PSs as being engaged in the 2-day training (mean 2.50, SD 0.54) and top-up day (mean 2.13, SD 0.84) and were involved and interested in the activities:

You could tell they were really listening and not just pretending to listen, it was like they wanted to be there . . . when they were giving answers it was like they had the read the book! The way that they were genuinely understanding what we were talking about . . . what we were saying to them was getting through to them and they understood it.

Trainer B

However, the scores indicate some room for improvement and the qualitative evidence suggested that PSs were less engaged and/or lost concentration during longer and less varied or (inter)active tasks. At these times, the trainers found delivery difficult, particularly when PSs could not report any PS actions:

I would try and ask them . . . ‘Well, can you see yourself in a conversation?’ I wouldn’t get anything. Nobody would even shake their head yes or no. It was flat, nothing. . . . So then I felt like, ‘well now I’m putting things in their head and I’m assuming’ but they weren’t talking.

Trainer B

Peer supporter engagement was also influenced by the characteristics and behaviour of the group. School 6 in particular was challenging to engage and work with, with both trainers describing it as ‘like pulling teeth’ (trainer B). Behaviour problems in school 2 affected the level of engagement and interest. This is supported by the data in Table 27 that show lower interest and involvement for schools 2 and 6. In school 2, such behaviour made the trainers question how these PSs would carry out their role and how they would talk to others:

We were kind of thinking ‘oh my gosh, if they go and talk to their friends like this, it’s not going to be encouraging them, that’s really going to put them off’.

Trainer E

Interestingly, PSs recognised when their behaviour was poor and challenging for the trainers:

‘Cause we are still like kids, if we do something [wrong] and they like tell us off, but it won’t go into our heads straight away. If they tell us to stop talking then us being like kids we always try to finish the conversation . . .

School 2, PS focus group

Impact of training on peer supporters

Peer supporters felt that their confidence to promote PA among their peers increased as a result of completing the 2-day training (mean 3.33, SD 0.68) and top-up day (mean 3.38, SD 0.69) (see Table 30). Despite this, PSs commonly wanted further support to develop confidence to talk to people and start a conversation and with what to say (Table 28).

Trainers expressed some reservations (Table 29) about whether or not the training increased PS confidence to instigate conversations with their peers (mean 2.63, SD 0.74) and, in a minority of cases, questioned how realistic some of the PS-generated practice conversations were:

Sometimes when they were writing conversations I was thinking ‘is this really a conversation you know you would have?’.

Trainer A

The qualitative evidence supported the scores in Table 30 that show that the PSs understood their role. The girls demonstrated that their role was not to talk to students they did not know, that they knew when it was appropriate to offer support and that they understood that there are different forms of PA to suggest:

Knowing that a lot of things are exercise was really reassuring, you are doing exercise, it’s just not in some other ways [sic].

School 4, PS focus group

Peer supporters reported gaining new knowledge about PA among girls, PA guidelines and benefits and the distinction between PA and sport (see Table 30; see Appendix 19). In addition, PSs reported building their communication skills and self-confidence. This corresponds with the trainer-rated achievement of objectives (see Table 29; Table 31), including enhancing PS knowledge of PA and their interpersonal and communication skills. Focus group evidence supported the quantitative findings and demonstrated how improvements in having the confidence to talk to others, problem-solving and PA knowledge worked together:

When you have the right information, [it] sort of gives you confidence and they [the trainers] said to go round obstacles that might occur.

School 4, PS focus group

Trainers’ perceptions of achieving the training objectives were generally lower on the top-up day than on the 2-day training. This may be because there is a ceiling effect following the high achievement rates at the 2-day training and it suggests that different objectives should be assessed after the top-up day. This finding may also be because of the poor achievement of objectives in school 2. The qualitative work indicated that this was a result of the behaviour of the children in school 2; trainers felt that the time they spent trying to manage this may have had an impact on what they managed to achieve throughout the training. This was not experienced in any other schools.

As a result of improving PS knowledge and their confidence to talk to people, the PSs felt prepared to support their friends, and their parents agreed:

Like slipping it into a conversation was pretty easy after we learnt how to do it [agreement]. Like before you’d be like ‘urrrr’ . . . But, yeah, learning it I think it actually helped.

School 4, PS focus group

However, some PSs were apprehensive to peer support because they were unsure of how to start and have a conversation promoting PA:

They didn’t tell you so much about going up to people and like how to start off the conversation.

School 6, PS focus group

Peer supporters reported an increased personal awareness of their activity levels and those of girls in England, which motivated them to increase their PA:

It like changed how like active you were, because like I wasn’t very active but then I’ve heard like only 8% of girls are active, so I started being more active.

School 6, PS focus group

Furthermore, in the schools where the year group was structured in two halves, PLAN-A helped integrate the two sides of the year and build friendships:

I think it’s also good that we interact with the other side of the year and sort of made friends with them as well.

School 4, PS focus group

The trainers, PSs and school contacts believed that the top-up day was an important and beneficial element of the training. Key content involved revisiting important PA information, supporting confidence and exploring how to communicate, providing a reminder of the PS objective/role and encouragement to continue peer supporting:

I think the top-up day kind of enforced [reinforced] the fact that you are actually trying to do something important.

Trainer C, top-up

Just to remind us . . . because after you have been to a top-up day we always think ‘we need to do this’ and then I’ll forget about it.

School 2, PS focus group

Need-supportiveness of trainers and peer supporters’ perceptions of need support

Peer supporters rated the trainer autonomy support as high during the 2-day training (mean 3.31, SD 0.63; possible range 0–4) and slightly lower at the top-up day (mean 2.92, SD 0.80). These scores are reflected in the qualitative findings; however, there was no clear explanation for the lower scores at the top-up day.

Evidence of peer supporter diffusion

There was evidence from all stakeholders to suggest that the PSs made numerous attempts to peer support using different strategies:

I have overheard conversations because some of the mentors [PS] have been in my tutor group and even the girls saying to the boys ‘wow, you’re just always on the computer’.

School 2, school contact

What I quite often hear at the moment after that [PS training] is the girls will say to each other ‘for goodness sake woman up’.

School 2, school contact

A good few of them definitely had conversations.

Trainer B, top-up

Efforts to support peers included sharing knowledge with friends and offering to co-participate in clubs, sports, daily PA and active play. The PSs reported being empathetic to their friends and trying to encourage them to try new activities and persist with those in which they were already engaged:

There is a couple [of girls] [sic] actually, who I know got their friends joining [sic] a club doing a sport . . . and said that now a friend was doing a sports club or doing a different after-school club.

Trainer B, top-up

Whereas before they go [sic] to a youth club and I used to drop them off, now she’s walking. She gets her friends to call, they call for each other and they walk.

School 3, Parent

You know [NPS girl] because she is really good at athletics? She like . . .

Puts herself down a lot, doesn’t she?

Yeah, and then I have like let her join the athletics club with me. Because she is very good at high jumping.

So, you’ve tried to convince her to join the club with you? Have you succeeded yet?

Nearly there.

School 2, PS focus group

I do rounders, so me and [PS12] went up and spoke to her and said do you want to come and do rounders with us, you can come it’s all OK . . . she actually really enjoys it.

School 3, PS focus group

There was evidence that PSs carefully considered their diffusion attempts:

I tried to incorporate something that they liked. Erm, ‘cause some of my friends like to dance so it’s finding something that they like and encouraging them to do that more.

School 6, PS focus group

. . . at the top-up day we had a talk about when you’re in the car and talking about it and me and my friend were having a lift to go to someone’s house and we were talking about it and we ended up getting out the car early and then actually walking a bit because I was talking to her about all the benefits and we hadn’t done anything that day and we need to do stuff . . .

School 3, PS focus group

Generally, PSs felt that support given to close friends was more likely to be accepted and met with a positive attitude, and this was supported by NPSs who reported listening to PSs and accepting the support given if it was provided in an encouraging and polite way:

I feel [that] if I talk to my friend about it they won’t sort of be stubborn about it and when it’s coming from a friend or someone that they’re close with then they’re more sort of open about being active and what they would like to do, maybe rather than . . . So I feel that it’s sort of better coming from a friend.

School 4, PS focus group

If they say it [encouraging to be active] politely then yeah [they would try to be active], but if they get rude then probably not.

School 6, NPS focus group

The PSs in school 2 reported that it took time for some girls to accept support and that they persisted with their efforts. However, in school 6, NPSs were initially receptive ‘but they just sort of gave up after a while’ (school 6, PS focus group):

At first, yeah [NPS] was very negative about it, but over time she changed a lot – her mind . . . so we don’t want to give up on her because she does have potential, I think.

School 2, PS focus group

Trainers and PSs also reported that they had initially practised their role with family members as opposed to their peers; many found this easier:

She wasn’t actually talking to peers so she was trying to do it but she was doing it with her mum and her dad and her grandmother and her sister.

Trainer B, top-up

Despite the evidence of these PS activities, the trainers reported that many PSs seemed to have done little to peer support at the top-up training:

Some were like clearly covering up that they hadn’t done it and saying they had.

Trainer B, top-up

I just got the impression, particularly with the first school [school 6], that I’m not sure how much they were doing.

Trainer D, top-up

This was supported by many NPS:

No, no-one said anything to me.

School 4, NPS focus group

However, in line with their role, some PSs suggested that with effective peer support:

People might not know that you’re doing it.

School 2, PS focus group

They didn’t really know I was peer supporting because with some friends we did it quite subtly.

School 2, PS focus group

. . . you have just got to like every now and then just say something like ‘do you want to walk to school with me?’ or something or ‘why don’t you do PE today? Like have a go at it’.

School 2, PS focus group

The perceived effect of the Peer-Led physical Activity iNtervention for Adolescent girls on Year 8 girls (non-peer supporters)

The various stakeholders believed that PLAN-A had potential positive effects including increased participation in PA (e.g. extracurricular activities, daily PA or planned exercise), improved intrayear relations and knowledge of the importance of PA. One student reflected that a PS had helped her ‘realise that some sports are actually more fun than others’ (school 6, NPS focus group). In one school, students who were not nominated to be PSs appeared to take on the role of a PS by talking to others, this was perceived to be because these students did not want to ‘miss out’ and wanted to ‘be seen in the same way that PSs are’ (school 2, school contact):

It is good and it helps confidence and it makes people feel like people do trust them [PS], people do look up to them [PS] and people do actually think I would want them [PS] to do stuff with her; I want to be active with her [a PS].

School 3, PS focus group

They’ve grown in confidence again, I’m not sure whether I could say wholeheartedly that’s because of PLAN-A . . . Because it’s, they’ve matured because they feel more settled in the tutor group. But they are very happy to be themselves and not be a follower.

School 2, school contact

Some of them, like in my group a lot of them did actually start going to the gym now but I don’t know if it was ‘you’re allowed to go to the gym now let’s go for it’ or if it was ‘we should do this because of PLAN-A’.

School 2, NPS focus group

Non-peer supporters demonstrated some understanding of PLAN-A, but could not explain it fully. In their view, any impact that PLAN-A had on other girls within Year 8 was minimal and, if any, was short-lived. Reasons given for this were a lack of interfriendship group communication, ignoring advice or not being spoken to by a PS:

I’m that kind of person that will take it on for a day and then forget about it the next day . . .

School 2, NPS focus group

I don’t think that people should like encourage us because if we wanted to do it then we would do it but like them telling us, isn’t going to really change our opinion of if we want to do it or not.

School 3, NPS focus group

The impact of the Peer-Led physical Activity iNtervention for Adolescent girls on peer supporters

The perceived impact of the PS training on PS knowledge, motivation and PA is reported in Chapter 6, Peer supporter training, Impact of training on peer supporters. The most commonly reported effect of PLAN-A and actually being a PS was its positive effect on PSs’ self-confidence and self-perceptions. In addition to the pride and privilege associated with being nominated by one’s peers (see Chapter 6, Peer nomination), parents felt their daughters had ‘definitely become more confident’ (school 4, parent) and used terms such as ‘feel valued’, ‘empowered’ and ‘more assertive’:

I think it’s done their confidence a world of good and, like I say, they’ve taken quite a lot of pride in what they’re doing because they’ve been chosen to do it.

School 2, school contact

Although the NPSs identified an increased sense of confidence in the PSs, in school 3 they interpreted this negatively, describing PSs as ‘a bit more boastful’, with one girl suggesting that ‘it’s not good for the rest of us, because they are cocky’. Girls in school 4 suggested that the PSs were ‘confident anyway’ and that ‘they get a bit big headed, just ‘cause they did get chosen’.

There was also a perception that PSs were more empathetic in general with their peer group:

It’s just the everyday getting along. The supporting in a classroom and not if somebody gets something wrong – not laughing. You know, they are much more aware of people’s feelings themselves.

School 2, school contact

There was a clear understanding that being active was more than just participation in ‘sport’. Some girls claimed that their participation in PLAN-A had changed the way in which they think, and for many it led to an increase in their own PA levels:

It’s definitely helped me be a better person. It’s realised [sic] that I am active, I don’t have to just have to do loads of sports, I can just walk around the house, go up and down the stairs and I end up doing a lot of active things. It has helped me become a better person in how I talk to people as well, like making me think before I say to some people like I know that before I probably, if we didn’t have the PLAN-A thing, I probably would have just gone to someone, ‘oh you need to be active, come on a run’ and I think that would have then ended up not working.

School 3, PS focus group

I think it’s all the facts and it does make you realise how important it actually is having exercise.

I want to prove people wrong who say that girls aren’t as active as boys. Want to prove them wrong and say we are.

Like boys will still say, ‘ah, you run like a girl’ or something that’s like . . . you just want to prove them wrong. ‘Shut up’.

School 2, PS focus group

School policy and attitude towards physical activity

The control (mean 6.00, SD 1.41) and PLAN-A (mean 5.75, SD 1.71) schools scored similarly with regard to policies to encourage PA and PE (Table 32). One control school reported its annual PE budget and this was approximately £700 higher, on average, than PLAN-A schools. School 3 scored lowest (4.00) overall on policy and attitudes, reflecting few written PA policies and provisions, such as cycle training. School 4 was the highest scoring, reflecting a broad range of support for both staff and student PA and PE. All schools reported facilities to support staff PA (e.g. showers/changing); however, no schools required staff to receive training on PA promotion. The qualitative interviews revealed that schools 2 and 5 were involved in a variety of schemes and programmes (This Girl Can, Healthy Schools) to encourage student PA.

Physical activity within the curriculum was scored similarly in both PLAN-A and control schools (Table 33); however, both scores are relatively low, providing limited evidence for the promotion of PA across the curriculum, outside PE. Half of schools used PA in other subjects aside from PE, primarily in personal, social, health, economic (PSHE) education. Two PLAN-A schools encouraged staff members to promote PA to students. Mostly, the delivery of PE is not often compromised for other needs. However, in four out of the six schools, children are sometimes withheld from PE for other academic subjects, which is supported by the interview findings:

I think the schools at the moment are under so much pressure to get like the academic side, sometimes they forget the other [PA] side.

School 6, school contact

School physical activity provision

The audit of school-level PA provision (i.e. quality of cycling, walking and sports/play provisions) ranged from 13 to 25 out of 46 (Table 34), with higher scores indicating greater provision. The two lowest total scores (13 and 16) were from the PLAN-A schools; this is supported by the number of active after-school clubs offered in the same schools (20% and 37.5% of all clubs offered are physically active, respectively). School 4 scored highest (25 out of 46), commensurate with its high score for policies on PE and attitude towards PA. Although parents and students from this school thought that the sports facilities were excellent, space and equipment were not always accessible:

I heard that [the school has their own ‘sports facility’] the other day. I thought that was brilliant.

School 4, parent

One school contact recognised the need for appropriate equipment and space to encourage girls to be active, but the school may not have the capacity to do this:

[At the training] when they had the breaks and they had all the skipping ropes out and they were really happy just to do that. There was no boys around so they were not inhibited and I would love them to be able to do that somewhere but I’m not sure we’ve got the facility to do it but . . . they’re too embarrassed to do these activities so they don’t.

School 6, school contact

Five schools provided after-school activity club data. Out of the clubs provided for, and available to, Year 8s, approximately 50% were categorised as active in three schools; 37.50% of active clubs in school 4 were for girls only.

Physical activity facilities were rated as good at five out of six schools and adequate in school 2 (see Table 34). Of the three schools that had a written policy to encourage active travel (see Table 32), school 2 had relatively low scores for cycling and walking provision. When referring to PA in school, students commonly referred to formal sports or PE opportunities rather than broader opportunities such as cycling or walking:

And there’s clubs as well [sic].

Yeah, there are quite a few clubs [netball, rounders, badminton . . .

There’s a girls’ fitness club as well.

School 3, NPS focus group

School-level built environment

Table 35 provides a description of school grounds and the area within an 800-m buffer around each school. School field areas at PLAN-A schools were approximately 9000 m² larger than school field areas at control schools. The area of school playground and field do not appear to align with the data on level of sports and play precision within each school. The length (9952.67 m vs. 6381.04 m) and density (3.56 km/km² vs. 2.39 km/km²) of footpaths within the buffer was greater in control schools than in the PLAN-A schools. The data align with the scores in Table 34 in which school 2 scored lowest on walking provision; Table 35 shows that there are no footpaths within its buffer area. The majority of schools have no cycle paths within their buffer area; however, it is important to note that data are not available for cycle lanes that are part of roads or pavements, which are more common in England than separate cycle lanes. Therefore, these data do not agree with the cycling provision data in Table 34. The size of grassland areas is greater at control schools than at PLAN-A schools, and urban and suburban areas are larger at PLAN-A schools than at control schools.

Year group structure

All schools taking part in PLAN-A had a tutor system by which the year group was split in half. As such, girls in different halves of the year often were in different classes and did not interact. This influenced friendship groups, with girls often more likely to be friends with others in their half of the year:

Because we don’t see each other much, we’re never in any of the same lessons . . .

School 4, PS focus group

It also influenced peer nomination in some cases:

I’m on one side of the year so I chose the people who were on my side of the year that I knew were good at it but obviously I wouldn’t know the people on the other side of the year so I didn’t put [nominate] any of them.

School 3, NPS focus group

Awareness of and school-based support for the project

Awareness of PLAN-A was generally restricted to either the Year 8 students or the PE staff. In some schools, by focusing on PA promotion, PLAN-A was only considered important in the PE department. Among intervention schools, the level of priority depended on whether or not it fitted with the schools’ objectives and if it had the potential to be successful:

I don’t know about the school priorities, within the PE department it fits in very well with trying to promote PA . . .

School 4, school contact

The degree of support given by both school contacts and school leadership to the project was governed by contextual factors, such as changes in school senior management, work patterns of school contacts and teacher workload (data collection, accommodating the PS training and encouraging the PSs). In general, however, school contacts were helpful and facilitation of the study within schools was simple. In the school in which PS training was delivered in-school rather than off-site, there was a change in head teacher between initial school enrolment and intervention delivery:

The original head was really supportive towards it and I suppose the others have been – it’s not their thing – so it wasn’t a priority as such whereas [former head teacher] was very much keen to get involved.

School 4, school contact

School context summary

There were minimal variations between schools, as seen throughout the process evaluation. The majority of those that did exist (i.e. the behaviour issues in school 2 and poor PS engagement in school 6) could not be clearly linked to any school context factors. However, the year group structure can explain the barriers that some PSs experienced while trying to carry out their role as a PS. The split year group in most schools made it difficult for PSs to support girls outside the PS group and approach those in the other half of the year. Aside from this, evidence suggests that PLAN-A appears to be robust to a variety of contexts.

Parent understanding of and support for data linkage

The majority of parents understood the description of data linkage given; only a minority wanted additional information. All parents stated that they would be likely to consent to their daughter’s health and education data being used for data linkage, with some wanting to discuss it with their partner and/or seeking their daughter’s consent:

Yes [I would consent] if she wanted to do it.

School 4, parent

Parents supported data linkage as a contemporary and valuable scientific resource: ‘in this day and age it is something that needs to be done’ (school 4, parent), and commonly described their expectation that the process was secure and undertaken with appropriate governance so that data were ‘safe’ and ‘used positively’.

Most parents suggested that they would consent to data linkage studies until their daughter completed secondary or further education. Some parents would provide consent indefinitely, whereas others wanted to be able to review their consent in the future:

In principle, I would probably say until the end of school but it would be nice to be given the option on a yearly basis or whatever to change your mind, not that I probably would . . .

School 2, parent

Some parents suggested that the consent for data linkage should pass to their daughter either upon leaving school or at the age of 16 or 18 years:

I kind of think that then she’d be of an age, 18 [years], that, you know if you wanted to carry on it would be in her hands then. She’d be adult enough to say, ‘do you know what, now remove me’.

School 3, parent

Parents’ concerns

The majority of parents raised a range of concerns regarding anonymity, data security, potential negative effects of data linkage and who would have access to the data:

You know, where does that information go? Would it have her name, date of birth, etc.? So yes, I would want to know a lot more about it before I could be comfortable.

School 4, parent

I mean, because obviously of her age . . . it’s all secure isn’t it?

School 3, parent

I wouldn’t want to agree to something when she’s 12 years old that when she’s 30 years old might have a detrimental effect on her.

School 4, parent

Intervention refinement and piloting

The PLAN-A was based on the successful ASSIST28 intervention model, reworked to focus on influential Year 8 girls supporting their close friends’ PA. In line with recommendations for early intervention development and adaptation86,87 and a core focus of the NIHR through NIHR INVOLVE, phase 1 of the study involved extensive and iterative formative PPI. This included consultation with the DECIPHer ALPHA young persons’ advisory group followed by rounds of focus groups with Year 8 girls (using a ‘design–listen–refine’ cycle). The focus groups identified relevant barriers to PA, which the girls wanted help to overcome, how best to frame PS recruitment, PS training content, characteristics of PS trainers, design of the study logo and using social media in the intervention. The result was a co-designed intervention that, although remaining faithful to the ASSIST intervention model and structure, was adapted to include content focused on PA-based issues that were relevant for Year 8 girls, used appropriate terminology and equipped girls with the skills and confidence to support their peers.

The intervention was piloted in one school to identify key problems before the feasibility trial. The evaluation of the pilot identified a number of issues, including where insufficient or surplus time had been allocated to various PS training activities, the need to include more active tasks and more activities focused on the ‘how to’ of peer support and a clearer idea of the physical space requirements for PS training (i.e. all need a break-out space if possible). The pilot intervention test also provided an initial deployment of the peer nomination and selection process, as well as the majority of process evaluation methods (observations, interviews, focus groups) that were then used in the feasibility trial. In summary, the refinement and piloting stage of the current study was central to the development of an intervention with a higher likelihood of acceptance, which was further tested in the feasibility trial.

Peer nomination and peer supporter selection

Phase 2 comprised a feasibility trial of PLAN-A in six secondary schools, all above the median local Pupil Premium (i.e. schools serving more deprived communities). The peer nomination process was implemented as planned. The idea of being a PS was popular with the students: 97% (n = 55) of those invited gave consent to attend the PS training. This represented 18–22% of the Year 8 cohort of girls in each school. This proportion of PSs within each year group is stipulated in DOI theory34 and the ASSIST intervention28 as necessary for there to be sufficient opinion leaders within a social group to make peer support effective. Commensurate with findings from ASSIST,38 PSs valued and felt pride in being nominated and in their role. Some parents believed that the nomination alone had enhanced their daughter’s confidence. PSs were motivated largely by altruism in the form of helping others to be active.

The peer nomination process adopted in ASSIST28 and PLAN-A sets it apart from most PA interventions that involve a peer support component. It is common for peer mentoring to be hierarchical, in that older students (e.g. Year 10) are assigned an individual or group of younger students (e.g. Year 7) to mentor on a more formal basis. 33,88–90 The most robust study of this peer mentoring model,33 a cluster RCT in 1495 young people in 60 schools in the north of England, showed no effect of the intervention on accelerometer-assessed PA. In contrast, the peer nomination process adopted in PLAN-A allows the intervention to capitalise on the potential influence of close-knit peer groups. As PLAN-A is led by key influencers from within peer groups, it is reasoned that they are sufficiently similar (i.e. in age, interests, life and educational stage and context and priorities) to both understand and influence their close friends. There is also evidence to support the influence of peers’ PA modelling, co-participation and encouragement on the PA levels of adolescent girls,91 and that peer support can partly buffer the effect of low self-efficacy on PA levels among girls. 92 On average, the students selected to be a PS did approximately 12 minutes more MVPA on weekdays and weekends and were less sedentary on weekdays than NPS at baseline. They also had a more favourable profile on some psychosocial attitudes towards being active. These findings agree with qualitative reports in some focus groups that more active girls had been nominated and that some PSs experienced difficulty in supporting others to be active as their friends were also trained as PSs. However, equally, there was the view from focus groups that PSs represented a variety of friendship groups and that PSs would have more legitimacy if they were active versus less active. Several ways of controlling peer nomination to recruit PSs, who are as active as the NPSs and well distributed through all friendship groups, and its advantages and disadvantages, are shown in Table 37.

An unexpected influence on peer nomination was that the school year groups were structured as two relatively separate halves. The process evaluation revealed the existence of these structures in all schools and a tendency for students to nominate girls in their half of the year. This resulted in perceptions of unequal recruitment of PSs from each half or for PSs to feel that their influence was limited by having limited opportunities within school to interact with girls from the other half of the year. Future implementations of interventions using the peer nomination approach should determine how school year groups are split or structured and consider running peer nomination in ways that take account of the structure.

The process evaluation provided scant evidence that peer nomination had detrimental effects on the girls not selected to be a PS. No adverse events related to bullying were reported through either the adverse events reporting protocol or the school contact interviews. There was a small amount of disappointment voiced from some NPS girls in not being nominated, but this was not a commonly reported experience and there was no qualitative evidence that NPS girls’ PA was discouraged as a result of not being selected. Notwithstanding this, in a larger trial, with a greater number and diversity of participants, it is possible that some negative effects of not being selected may be seen. Therefore, it would be prudent to review the script with which the peer nomination process is delivered to ensure that terms such as ‘popular’ are not used (as was the case in this study).

Trainer recruitment and train the trainers

Five women were recruited and attended the train the trainers 3-day workshop. Trainers had a variety of backgrounds (e.g. youth work, theatre, yoga instruction, health promotion, coaching) and were paired as intended (balancing experience in PA promotion and working with young people). In this study, trainers were recruited using adverts and group contacts. Some trainers were self-employed, while others were employed by the LA (health promotion) or town council sports development teams. This profile is similar to the trainers recruited in ASSIST including youth workers, smoking cessation staff, teachers and NHS health improvement leaders. 93 Looking toward a larger trial and/or future wider dissemination, it will be important to identify a sustainable source of trainers while ensuring that the personal characteristics that were viewed as successful in this study (see below) are maintained.

The train the trainers course was effective and trainers found the materials and resources helpful. Trainers reported wanting a recap on theoretical principles and further advice on managing disruptive behaviour. Trainers also needed to use time (unpaid) outside the training to prepare for the delivery. Although it would be expensive to pay for all preparation time, future interventions could build in some personal preparation time. Alternatively, if trainers were identified through employers such as LAs, PLAN-A delivery would form part of their contracted tasks and some preparation time may be incorporated.

Trainers largely delivered the PS training with high fidelity to the training manual and in a manner that was consistent with SDT,46 supporting the PSs autonomy, competence and relatedness as planned. Challenges to this included disruptive behaviour, low PS engagement and long training activities. Previous school-based PA interventions based on SDT have identified similar challenges (i.e. delivering in an autonomy-supportive manner when facing disruptive behaviour). 94,95 Although the train the trainers’ workshops attempted to learn from these studies and provide advice on the use of structure and agreed ground rules, trainers requested greater practical guidance.

Focus groups with PSs reinforced the importance of the trainers’ ability to quickly build a strong and genuine rapport between themselves and the PSs, which was the foundation of much of the enjoyment and engagement in the PS training. PSs highly valued the autonomy-supportive approach of the trainers (with clear qualitative reports by PSs of trainer support for competence, autonomy and relatedness), identifying that their interpersonal style was different (i.e. more autonomy supportive) than their usual teachers. Many examples of autonomy support and enjoyment centred on when trainers delivered the training with flexible adaptation of the intervention materials and trainers being attuned to the PSs (e.g. recognising when it was appropriate to move on or spend more time on a task, and including their own experiences and skills, such as yoga in warm-up activities). As such, trainer innovations96 were in accordance with, and facilitated, the SDT underpinning of the intervention. Previous studies have identified drift (when adaptations are not in line with the spirit or theoretical underpinning of the intervention)94,95 in intervention delivery; however, this was not commonly seen in the delivery of PLAN-A. Consistent with previous research in which PA/educational intervention deliverers have been trained to deliver sessions based on SDT,94,97,98 this study suggests that trainers welcome the approach, and that it is possible to train deliverers to adopt this method and for them to deliver it in practice. Refinements to the train the trainers elements are shown in Appendix 20.

Peer supporter training and peer supporter experiences

Peer supporter training was held at non-school off-site venues for students from three out of the four schools. One school was unable to release a member of staff to chaperone the students at the training because of financial and staffing constraints. In this school, the PS training was conducted on the school site. Insight from the LAG suggested that this finding reflects the current challenging financial climate for schools. Delivery of the training in school had advantages (e.g. more time to conduct training) and disadvantages (e.g. small training space, disruption by other students using nearby rooms and a less marked sense of privilege among the PSs who were not taken off-site). Although off-site delivery is preferable, on-site PS training is feasible if carefully managed and could represent a pragmatic balance between intervention fidelity and acceptability to schools in the current challenging financial climate.

A key amendment to the ASSIST intervention approach was that, in addition to the initial 2 consecutive days of PS training, PLAN-A included a 1-day top-up training session (5 weeks after the initial 2 days) rather than four 1-hour in-school sessions (as implemented in ASSIST and the AHEAD study). Findings from the AHEAD study suggested that trainers found it more difficult to motivate PSs and deliver the in-school training. 40 In addition, feedback from the LAG (including representatives who had commissioned and delivered ASSIST) suggested that the four in-school visits yielded relatively poor engagement compared with the 2-day training. In PLAN-A, the top-up day built on the strong rapport between the PSs and the trainers and was valued as a good time to refresh key learning objectives, solve problems and re-energise the PSs. Trainers voiced some concerns about the length of the training days and suggested the option of adding another top-up day. However, it is felt that PS engagement could be improved through small changes to the training day content and structure that would be more financially viable than adding a further training day (this would cost approximately £1000 per school).

The PS training was successful with very high attendance (90–100%). PSs had fun, were engaged with most tasks and liked the trainers. Training helped the PSs prepare for their role; however, they would have liked some tasks to be more active (a finding replicated in the AHEAD study40) and greater help on how to start conversations about PA. Attending the training and being a PS had a number of perceived positive impacts on the girls: it increased their confidence to talk to others, improved their social connections, as well as increasing their drive to help their peers and maintain their activity levels. Similar to a recent process evaluation of the implementation of ASSIST in Scotland,93 PSs, NPSs and trainers raised questions about the level of diffusion of messages from PS to their friends. However, PSs reported employing numerous strategies to support their peers including sharing knowledge, giving encouragement and co-participation in sport and daily activity (e.g. active travel). Central to PS-reported efforts to peer support was their ability to empathise with their friends and be subtle in their support. This is suggestive of good theoretical fidelity, in that the PSs were faithful to the SDT-based foundations of the intervention (i.e. supporting and developing each other’s autonomy to be active in ways that suit them, rather than pressuring their peers, which was found to be an expectation that some students had pre training). Challenges to peer support included anticipating hostility, worries of offending others and difficulties in approaching peers; however, PSs reported that the anticipation of these factors was often worse than the reality and they were able to use an empathic and subtle approach to peer support to good effect.

The results of the feasibility study identified a number of refinements that could be made to the intervention if it was taken forward in a definitive trial; these are summarised below in Table 38 and further details are in Appendix 20. Refinements focus on the physical training space, manipulating the training structure and content to optimise engagement, simplifying some activities and terminology and adding more active training content, including practical guidance on the ‘how to’ of peer supporting. All of these changes are simple, would not change the core intervention design and are possible within current intervention cost estimates.

Evidence of promise to affect girls’ physical activity

The quantitative analysis for evidence of promise showed no evidence for a difference in minutes of weekday MVPA between control and PLAN-A arms at T1 (end of Year 8), but there was a 6.1-minute difference (95% CI 1.43 to 10.76 minutes) in weekday MVPA favouring the PLAN-A arm at T2 (i.e. beginning of Year 9). The CI exceeded 10 minutes, which is a level of daily MVPA considered to have important implications for health. 99 The between-intervention difference reflected a prevention of the decline in MVPA (seen in the control group) between the beginning of Years 8 and 9. The study also found that PLAN-A held promise to reduce the time that girls spend sedentary (i.e. sitting, lying down) as measured by accelerometer. At T1, girls in PLAN-A schools did 32 minutes less sedentary time per weekday (95% CI –57.44 to –6.18 minutes) than girls in control schools. This finding was largely replicated at T2, with girls in PLAN-A schools doing 23 minutes less sedentary time per weekday (95% CI –43.73 to –2.79 minutes). The findings were consistent when missing data were imputed and when the accelerometer validity criterion was reduced to 1 valid weekday. The results of the exploratory interaction analysis must be treated with caution, but suggested that PLAN-A might be more effective on PSs than NPSs at T1. However, there was no evidence for this at T2; when combined with the finding of evidence of promise at T2, this could indicate that the PSs had managed to diffuse the PA messages more successfully to their peers by this time point. In contrast to this, at T2, PLAN-A arm participants reported having been spoken to by a friend about PA less frequently than at T1. However, in process evaluation focus groups, the PSs believed that some of their peer support was subtle rather than overt; this is in line with the informal nature of PLAN-A and may suggest that PSs became more effective at peer support by T2 and explain these divergent findings.

Given the lack of interventions with evidence of effectiveness to increase (or maintain) adolescent girls’ PA, this finding is promising. Reviews of PA interventions for girls have shown very small positive effects,100 with an average effect of approximately 4 minutes of MVPA per day. 101 There was no effect on the accelerometer-assessed weekday MVPA of Year 7 girls in a robust RCT of an after-school dance intervention versus control. 83 A recent RCT of a multicomponent school-based intervention (Girls in Sport) in Australia,102 involving 1518 girls, aged 13 years at baseline, reported no effect on accelerometer-assessed MVPA. Among interventions that have shown effectiveness, the RCT of Physical Activity 4 Everyone, a multicomponent school-based intervention in Australia, showed that the effect in girls (4.0 minutes MVPA per day, 95% CI 0.01 to 8.0 minutes) was less than half that of the effect in boys (10.4 minutes MVPA per day, 95% CI 2.1 to 18.0 minutes). 23 Finally, there was no evidence of promise that the AHEAD intervention affected the MVPA of the boys and girls included. 40 Should the evidence of effectiveness of PLAN-A be replicated or increased (e.g. to 10 minutes MVPA per day) in a definitive trial, the intervention would be one of the most effective to date. In comparison to multicomponent school-based intervention approaches, such as Physical Activity 4 Everyone, PLAN-A is relatively simple, focused and a low burden for schools. As such, in addition to its potential effect on PA, PLAN-A also has the potential to be widely implemented in schools.

Reflections on the logic model

The logic model for the intervention was largely supported and it serves as a foundation for a definitive trial. Qualitative support was demonstrated for the proposed mediators of PLAN-A (i.e. SDT variables); however, this was largely based on the experiences of PSs, rather than NPSs, and there was no quantitative evidence that these variables changed as a result of the intervention. This may reflect a true lack of change, be as a result of the low power to detect changes or be as a result of measures with low sensitivity to change. To better understand mediation of the intervention effect, in a future trial there is a need to more fully explore what PSs did to support their peers and girls’ perceptions of this, in addition to more general PA social support, as these are likely key mediators (in addition to more abstract concepts, such as need satisfaction or motivation quality). Measuring these psychosocial factors with appropriate depth and precision is challenging, especially within the context of interventions, such as PLAN-A, which rely (at least partially) on subtle diffusion/support. The qualitative data in this study suggested that such support may not be in the conscious awareness of those receiving it, which further complicates its measurement. There is likely to be no single, simple solution (or questionnaire) to resolve this issue. Rather, further in-depth mixed-methods approaches involving both PSs and NPSs, in addition to more frequent monitoring (including the potential use of online resources to record peer support) of PS actions are needed.

Health economics: indications of affordability and potential cost-effectiveness

This study has established that it is feasible to both collect and cost resource-use data to estimate the cost-effectiveness of delivering PLAN-A. The study was able to collect all the costs associated with train the trainers (including estimates of how the cost would alter if more PSs were trained in a different location) and the information required to estimate the cost of the intervention per school and per Year 8 girl, including PS and Year 8 girl time attending PLAN-A meetings and training. As such, the study developed a robust and effective methodology for capturing cost data that could be applied in a definitive trial.

When compared with other studies, PLAN-A has the potential to be good value. PLAN-A cost an average of £2685 per school and £37 per Year 8 girl. Based on the identified 6.1-minute difference in weekday minutes of MVPA between trial arms, on average the cost per 10-minute increase in mean weekday MVPA at 12 months was £61.01 per Year 8 girl. For comparison, in the cost-effectiveness analysis conducted alongside the Physical Activity 4 Everyone in Australia,23 a societal perspective was taken and resource use measured included personnel costs, materials and printing. Opportunity costs for personnel time were estimated only for activities that took place outside usual PE and sport time. Participants in the PLAN-A arm were more active at 24 months (adjusted difference in mean daily MVPA 7.0 minutes, 95% CI 2.7 to 11.4 minutes). The intervention cost over 24 months was £240 per student and this equated to £34 (95% CI £21 to £90) per additional minute (or £340 per 10 minutes) of MVPA (conversion rate used, AUD$1 = £0.609317; date of exchange, 8 August 2017). In the UK, the Bristol Girls’ Dance Project83 provided after-school dance sessions for adolescent girls and effectiveness and cost-effectiveness were estimated. The intervention cost per girl was £73 (95% CI £71 to £75); however, there was no evidence that girls in the intervention arm were more active than girls in the control arm.

The PLAN-A was based on the ASSIST smoking preventing intervention model,28 which involved peer nomination, PS training and informal diffusion. A public-sector costing perspective was taken, similar to that used in the present study. The total cost of the ASSIST intervention per school was £5662 [including costs for a teacher chaperone for PS training (£633/school)]. The ASSIST intervention was estimated to cost £32 (95% CI £30 to £34) per student and, at 2 years post baseline, the incremental cost per student not smoking was £1500 (95% CI £669 to £9947). The AHEAD intervention,40 which used the PS model to target change in healthy eating and PA in secondary schools, cost £11,289 per school and was considered too expensive to replicate on a larger scale.

There was no evidence that the intervention could affect quality of life, as measured using the EQ-5D-Y. A limitation of the EQ-5D-Y is its focus on health outcomes; however, PLAN-A is a public health intervention that has the potential to impact quality of life more broadly. Given the focus of the EQ-5D-Y and given that we did not expect to see significant changes in the health of a largely healthy group, the limited variation in EQ-5D-Y scores and lack of association between EQ-5D-Y and MVPA was unsurprising.

Qualitative data

Directed thematic analysis was undertaken examining the following a priori issues: (1) views on the device prior to wearing them, (2) peer and family perception of the device, (3) participants’ experience of wearing the device and (4) which device participants would prefer to wear. The initial analysis was undertaken by an experienced qualitative researcher (ME), and was triangulated and refined by a second researcher (JM). Key quotations that represented the breadth of participants’ experiences were selected and checked by both analysts.

Quantitative data

ActiGraph data were downloaded via ActiLife 1.7 (ActiGraph, LLC, Pensacola, FL, USA) and analysed using Kinesoft (Kinesoft, SK, Canada). The GENEActiv data were downloaded using GENEActiv software. Given the difference in device placement, multiple analysis protocols were followed. ActiGraph waist data were analysed in line with the PLAN-A intervention protocol. Wrist-worn ActiGraph data were analysed in accordance with recommendations of Chandler et al. 107 (Axis 1). GENEActiv data were analysed using a Microsoft Excel^® (Microsoft Corporation, Redmond, WA, USA), macro analysis spreadsheet, supplied by GENEActiv. GENEActiv Open Platforms [https://open.geneactiv.org/geneactiv_macros.html (accessed 27 November 2018)] was used to interpret data.

Qualitative results

Quantitative results

The primary objective of this study was to gather participants’ experiences of wearing the different accelerometers. Although the accelerometers recorded data and the data were downloaded and analysed, the analysis is descriptive (as the study was not powered, nor was it intended, to use these data to identify representative differences in compliance and/or wear time).

Table 39 shows wear protocol compliance. Wear time exceeded the thresholds used in the PLAN-A feasibility study (i.e. 2 valid weekdays) for all devices, although there is greater wear time in the GENEActiv group.