Skip to main content


We’d like to understand how you use our websites in order to improve them. Register your interest.

Results of a feasibility cluster randomised controlled trial of a peer-led school-based intervention to increase the physical activity of adolescent girls (PLAN-A)



Most adolescent girls in the UK do not meet government physical activity recommendations and effective interventions are needed. This study reports the results of a feasibility trial of PLAN-A, a novel school-based peer-led physical activity intervention for adolescent girls.


A two-arm cluster randomised controlled feasibility study was conducted in six English secondary schools (4 intervention & 2 control). Year 8 (age 12-13) girls were eligible and randomisation was at school-level. The intervention involved training Year 8 girls (out of school for two consecutive days, plus one top-up day 5 weeks later), who were identified by their peers as influential, to provide informal support to their friends to increase their physical activity. Feasibility of the intervention and the research was examined, including: recruitment, training attendance and data provision rates, evidence of promise of the intervention to affect weekday moderate-to-vigorous physical activity (MVPA), intervention cost and estimation of the sample size for a definitive trial. Accelerometer and questionnaire data were collected at the beginning of Year 8 (Time 0), the end of Year 8 (10-weeks after peer-supporter training) and the beginning of Year 9 (Time 2).


Four hundred twenty-seven girls were recruited (95% recruitment rate). 55 girls consented to be a peer-supporter and 53 peer-supporters were trained (97% of those invited). Accelerometer return rates exceeded 85% at each time point and wear time criteria was met by 83%, 71% and 62% participants at Time 0, 1 and 2 respectively. Questionnaire data were provided by >91% of participants at each time point. Complete-case adjusted linear regression analysis showed evidence of a 6.09 minute (95% CI = 1.43, 10.76) between-arms difference in weekday MVPA at Time 2 in favour of the intervention arm. On average PLAN-A cost £2685 per school to deliver (£37 per Year 8 girl). There were no adverse events. A trial involving 20 schools would be adequately powered to detect a between-arms difference in weekday MVPA of at least six minutes.


The PLAN-A intervention adopts a novel peer-led approach, is feasible, and shows evidence of promise to positively affect girls’ physical activity levels. A definitive trial is warranted.

Trial registration

ISCTRN, ISRCTN12543546, Registered on 28/7/2015, URL of registry record:


Physical activity is associated with benefits to young people’s physical and psychological health [1,2,3]. The UK and US governments recommend that children and adolescents undertake 60 minutes of moderate-to-vigorous physical activity (MVPA) each day [4, 5], however, many adolescents do not meet this guideline [6]. A consistent finding is that girls are less active than boys and that their physical activity declines at a faster rate than their male peers [7, 8]. Late childhood to mid adolescence is a key period of physical activity decline [9]. As such, there is a specific need to develop strategies to prevent the decline in, or increase adolescent girls’ physical activity.

Recent evidence syntheses have identified that previous interventions have had small positive effects on the physical activity of adolescent girls [10, 11], with one meta-analysis equating the average intervention effect to approximately 14% more physical activity in the intervention versus control group. Larger effects were shown for interventions targeting girls only, were multi-component, based in schools, utilised theory and targeted sedentary behaviour as well as physical activity [10, 11]. A recent trial in Australia of a whole-school, 24 month, multi-component intervention (Physical Activity 4 Everyone) resulted in a 7-minute (95% CI = 2.7, 11.4, p=0.002) increase in adolescents’ accelerometer-assessed MVPA compared to controls [12]. However, the intervention was less effective for girls (+4 minutes) than boys (+10 minutes).

Correlates of girls’ physical activity include physical self-perceptions, self-efficacy, perceived competence and enjoyment [13]. Yet, girls’ relationship with physical activity is also described as complex, including being intertwined with gender roles, societal norms and self-presentation [14]. Peer-influence, which increases as girls transition from childhood to adolescence [15], plays a role in girls’ physical activity [16, 17] and can comprise social support, presence of peers during physical activity, peer norms, friendship quality, changes to friendship groups, activity preferences of peers, affiliation to peer groups, and peer victimisation [17]. Despite being a key part of the complexity of girls’ physical activity, until recently peers have been relatively overlooked in many physical activity interventions [18]. Tymms et al. [19] reported results of a cluster randomised controlled trial of “MOVE” in England. MOVE was a cross-age peer-mentoring intervention in which Year 9 (age 13-14) students were trained to mentor same sex Year 7 students (age 11-12). The intervention comprised of six weekly 1:1 meetings (20-30 mins) aimed at identifying physical activity barriers and setting/reviewing goals using self-monitoring but there was no evidence that it affected (accelerometer-assessed) physical activity of the Year 7 students.

Despite the potentially important role of peers in adolescent girls’ physical activity, there is a lack of robust trials (Excluding Tymms et al. [19] and two ongoing studies [20, 21]) of peer-led physical activity interventions [22]. Previous interventions are limited to cross-age approaches, where older pupils support younger pupils (i.e. they are “peers” but not necessarily influential friends) using formal methods (e.g., mentoring or leading classes). A potential limitation of cross-age interventions is that pairs of pupils are unlikely to be friends or peers which may hamper the development of a foundation of trust, mutual understanding, shared experiences and key peer-support strategies (e.g., an older mentor may not be comfortable co-participating in physical activity with a younger pupil out of school). In the ASSIST (A Stop Smoking in Schools Trial) intervention [23] an alternative peer-led approach was used which involved students in a secondary school year group nominating influential peers who were trained as peer-supporters to informally diffuse messages and provide support to help their friends stop or not start smoking. In a cluster randomised controlled trial of 10,730 12-13 year olds from 59 schools in England and Wales the intervention reduced the odds of students being a smoker compared to controls (OR = 0.78, 95% CI = 0.64, 0.96) when data from three follow-ups was considered together.

Based on the success of ASSIST and the centrality of peers in young people’s physical activity, we hypothesised that a peer-diffusion intervention (i.e., informal peer-peer verbal support, encouragement, co-participation, sharing knowledge and shifting norms) that targeted adolescent girls’ physical activity could hold promise as a means of promoting physical activity. We therefore developed the PLAN-A intervention (Peer-Led physical Activity iNtervention) and in line with the UK Medical Research Council (MRC) framework on the development and evaluation of complex interventions [24], tested it in a feasibility trial [25, 26]. Specifically we aimed to assess recruitment and retention rates, estimate data provision and quality, estimate the potential effect of the intervention on accelerometer-derived week-day physical activity (evidence of promise), identify and collect the data needed to cost the intervention and estimate the sample size required to conduct a definitive trial.


Study Design

The study was a two-arm cluster randomised controlled feasibility study in six secondary schools to compare the PLAN-A intervention against a usual-practice control with embedded process and health economics evaluations. The study protocol was published in 2016 [25, 26] and was registered with ISCTRN (ISRCTN12543546) prior to data collection. A completed CONSORT checklist is in Additional file 1.

Sampling and participants

Schools were eligible and invited to participate via letter and email to the Head Teacher if they were; state-maintained mainstream secondary schools in Wiltshire and South Gloucestershire (South West England), had girls in Year 8 (aged 12-13 years), were above the median of the local Pupil Premium Indicator (i.e., more deprived) [27], were not special educational needs schools and were not implementing ASSIST at the time. Since this was a feasibility study, a sample size calculation was not performed but we chose to recruit six schools as we expected this to provide sufficient variety of schools in which to test the feasibility of the research and the intervention. Of 46 secondary schools in the two areas, 30 were ineligible (26 were below median pupil premium) and the 16 remaining (eight in each area) were invited to participate and six were selected on a first-come-first served basis, in line with the stratification outlined below (Fig. 1). A face-to-face group recruitment pitch was made to Year 8 girls in each school and study information for students and parents was distributed. Ethical approval was obtained from the University of Bristol’s School for Policy Studies Research and Ethics committee (Ref: SPSREC14-15.A27). Parents could opt their child out of the study. Parents of peer-supporters provided written informed consent and peer-supporters gave written informed assent. All adults involved in the research (e.g., peer-supporter trainers) gave written informed consent. Pupils received a retail voucher (£5 for baseline & £10 for each follow up) in recognition of the time given to each data collection. Participating schools (intervention & control) received £500 and a summary of the findings in recognition of the time devoted to the study. An independent Trial Steering Committee provided study monitoring and oversight.

Fig. 1

Trial profile for the PLAN-A study (based on CONSORT 2010 flow diagram). a Returned at T2, b Took part in T2, c Present at T2


Following the completion of baseline data collection, all schools were randomly allocated (school being the unit of randomisation), at an intervention:control ratio of 2:1 stratified within Local Authority area (Wiltshire & South Gloucestershire). Four schools were allocated to the intervention arm and two schools to the control arm by a member of the Bristol Randomised Trials Collaboration who was blind to the school identity and independent of the study. The statisticians, and all team members except the Project Manager, Research Associate and Fieldworkers were blind to allocation.

Intervention description

The intervention was adapted from the ASSIST intervention through detailed formative research and its refinement and piloting is described in detail elsewhere [28]. Briefly, it comprised: (a) peer-supporter nomination, (b) a train-the-trainers programme, (c) peer-supporter training followed by a ten-week informal physical activity message diffusion period. Control schools did not receive the intervention but pupils completed identical measurements to the intervention schools. A TIDieR checklist is presented in Additional file 2.

Theoretical background

Commensurate with ASSIST, the PLAN-A intervention was primarily grounded in Diffusion of Innovations theory (DOI) [29] in which influential peers are identified as agents of change within a social system. In addition to DOI, the behaviour change principles and pedagogical approach within PLAN-A training were guided by self-determination theory (SDT) [30, 31] a framework which has been used previously to encourage the personal and social conditions needed to foster high quality (autonomous vs. controlled) motivation for physical activity amongst children and adolescents. Further explanation of SDT, and detailed examples of how it was incorporated into the train-the-trainers and peer-supporter training are reported elsewhere [25]. Briefly, trainers were educated about core SDT principles such as developing girls’ autonomous versus controlled motivation, and peer-supporter training tasks targeted developing participants’ autonomy, competence and/or relatedness and in turn using these principles to support their friends’ physical activity.

Train-the trainers programme

Peer-supporter training was delivered by female trainers who had attended a three-day (≈ 15 hours) education programme delivered by the study team (MJE, KB, JM & SS). Trainers were recruited as freelance individuals (e.g., youth workers, performing artists) or through physical activity/sport promotion teams to have a range of experiences in working with young people and physical activity promotion. Training covered gender inequalities in physical activity, the PLAN-A concept (informal diffusion) and theory in addition to teaching trainers about the sessions, activities and resources and giving time to practice delivery. Peer-supporter training was delivered by trainers working in pairs and trainers had time to work in their pairs and establish their delivery plan. Trainers were given a manual containing all resources needed.

Peer nomination

Prior to randomisation, Year 8 girls in all schools completed a four-item peer-nomination questionnaire to identify the influential girls in their school (i.e., Who do you respect? Who are good leaders in sport or other group activities? Who do you trust? and Who do you look up to?). Girls could list up to five girls for each question. The 18% of girls with the most nominations were invited by letter to be a peer-supporter and attended a peer-supporter briefing where the role and training were explained and written information and consent forms were distributed. 18% were invited with the aim of recruiting 15% of the year group as peer-supporters in line with DOI.

Peer-supporter training

Consenting peer-supporters attended an initial two-day course, followed by a one-day top-up course approximately five weeks later. Separate training was held for each intervention school and took place off the school site (e.g., at a local business campus or community centre). Peer-supporters were provided with return transport from school to the training by coach (unless it was possible / more convenient to walk) and were chaperoned by a member of school staff. Training for one school was held on the school site due to the school being unable to release staff to chaperone. The interactive training blended education about the importance of physical activity, inspiration and empowerment of the girls to be active and development of the interpersonal skills needed to be a peer-supporter (See Table 1 in [25]). Peer-supporters were presented with many ideas of what types of support they could give (e.g., co-participation, encouraging and co-participating in incidental activity such as active travel to and from school, suggesting walking and talking at break/lunch instead of sitting, knowledge sharing, offering verbal support / being empathetic) as well as how to give support (e.g., how to be persuasive, language and style, and identifying who, where and when to support and not support). The top-up training revisited core learning objectives from the initial training, and focussed on sharing successes and collective problem solving. Peer-supporters each received a training booklet to support the training activities and a diary to record their peer-supporting if they wished (diaries were an intervention tool and data were not collected in). Following training, peer-supporters were asked to informally diffuse messages and norms about leading physically active lifestyles to their friends to encourage and support them to maintain or increase their physical activity.

Table 1 Participant and peer supporter recruitment rates by study school

Data Collection

Data were collected at baseline (Time 0: the beginning of Year 8, September-October 2015), immediately post-intervention (Time 1: the end of Year 8, May-June 2016) and follow up (Time 2: the beginning of Year 9, September-October 2016).

Demographic information

At baseline (Time 0) participants reported their date of birth and ethnicity. Socio-economic position was estimated by: (1) an index of multiple deprivation (IMD) using home postcode and the English Indices of Deprivation [27] for each participant, (2) participant-reported receipt of free school meals and (3) the four-item family affluence scale [32].

Recruitment and retention to trial and intervention

Recruitment of schools and Year 8 girls (opt-out consent rate) was recorded at Time 0 and retention was recorded at Time 1 and 2. Recruitment of peer-supporters (consent rate) and peer-supporter trainers was recorded by the field team as well as attendance of peer-supporters at the briefing meeting and each peer-supporter training day.

Physical activity and sedentary time

Participants wore a waist-worn ActiGraph accelerometer (Model wGT3X-BT; ActiGraph LLC, FL, USA) during waking hours which recorded data at 30Hz for seven days. Participants were instructed to remove the accelerometer while sleeping, showering/bathing and participating in water sports. Scores derived from ActiGraph accelerometers have been shown to provide reproducible and valid estimates of physical activity amongst adolescents [33]. Periods of ≥ 60 minutes of zero counts were classified as non-wear and were removed. Accelerometer data provision was a primary outcome of the feasibility study, however for the purposes of assessing the secondary outcome of the potential effect of the intervention on girls’ MVPA, participants who provided at least two weekdays of at least 500 minutes of data between 05:00 and 23.59 were included in this analysis. The Evenson cut-point (≥2296 counts per minute)[34] was used to estimate mean daily minutes of MVPA on weekdays and weekend day(s) as this threshold is the most accurate for adolescents [35]. Participants’ weekday and weekend sedentary time was estimated using a cut point of ≤100 counts per minute [34].

Psychosocial Questionnaires

Full details of the psychosocial questionnaires will be published in a separate process evaluation paper. For the purposes of reporting data provision rates in this paper, pupils completed a survey (assessing physical activity motivation, psychological need satisfaction, self-esteem, physical activity-based self-efficacy, social support & peer norms) using Samsung Galaxy Table 4 tablet devices.

Economic analysis

The economic analysis aimed to assess the feasibility of collecting the data required to cost the intervention and conduct a cost-effectiveness analysis in a future definitive trial. The affordability and potential cost-effectiveness of the intervention was also explored. A public-sector perspective was taken assuming that if scaled up the intervention would be funded by schools or local government. Data were collected using expense claim and data collection forms, completed by the research team, students and school contacts. Students’ quality of life was assessed using the EQ-5D-Y [36]. Responses were mapped to a utility score (scores derived from adults were used in the absence of child specific scores [37]) at each time point and quality-adjusted life-years (QALYs) were estimated adjusting for baseline differences in utility scores [38].

Data Analysis

Descriptive summary statistics were calculated by trial arm at baseline. Continuous data were analysed using means and standard deviations or median and inter-quartile range, where the data were markedly skewed. School and student recruitment and retention were presented as a CONSORT flow chart for schools by trial arm. Recruitment, retention and data provision were also summarised descriptively.

Summary statistics for the (definitive trial) primary and secondary outcomes are presented, by intervention and control group according to the allocation of the student’s school (i.e., an intention to treat (ITT) analysis). Mixed effects linear regression was used to estimate the adjusted differences in means between intervention and control groups. School-level variance in the outcome was accommodated by inclusion of school as a normal distribution random effect, and any remaining differences in the baseline assessment of the outcome measure, and local authority area (by which the randomisation was stratified), was accommodated by their inclusion as covariates. Each difference in means is presented with its 95% confidence interval. As this study was a feasibility study and not powered to detect differences between arms, p-values are not presented.

Two pre-specified sensitivity analyses were conducted assessing the impact of missing data. Firstly, where there were 5% or more missing MVPA measures, missing MVPA data were imputed for Time 1 and Time 2, in turn using Multiple Imputation by Chained Equations [39]. Fifty imputed data sets were generated for MVPA at each time point, and the mixed effects linear regression analysis described above was applied to these data. The variables used in the imputation equations have been reported elsewhere [40]. Secondly, MVPA measurements based on ≥ 1 valid day of accelerometer data (rather than ≥ 2 valid days) were used to estimate the adjusted difference between intervention and control groups in mean MVPA. The Trial Steering Committee also recommended an exploratory interaction analysis to examine whether any potential intervention effect was moderated by peer-supporter status. A binary covariate identifying peer-supporters and non-peer-supporters was added to the regression model for the analysis of MVPA, and p-value calculated for the test of the null hypothesis, of equal effect of the intervention on both groups.

The school-level intra-class correlation (ICC) coefficient for weekday MVPA was estimated in a variance component random effects analysis. As this gives an imprecise estimate in a study of this size, the data were compared with ICC estimates from other studies [12, 41, 42]. Sample sizes for a future definitive trial were calculated for assumed true differences in mean weekday MVPA between intervention and control groups of 6, 8 and 10 minutes, and for 80% and 90% statistical power at the two-sided 5% significance level. Sample size estimates were multiplied by the design effect: [1+(k-1)ICC], k being the number of participants at each school (cluster size = 70 informed by this study) and ICC the intra-cluster correlation coefficient for weekday MVPA. Sample size estimates were inflated to account for 30% loss to follow up on the primary outcome. All analyses were conducted in Stata (Version 14.2).


School and participant recruitment

School and participant recruitment and retention are shown in Fig. 1 and Table 1. Within the six schools, 451 Year 8 girls attended the recruitment briefing. Four (0.89%) returned opt-out forms and 20 girls (4.43%) did not take part (no formal opt out) resulting in a 94.68% participation rate (N = 427 girls, intervention n =269; control n =158). Peer nomination resulted in 57 girls being invited to be a peer-supporter, 55 (96.49%) consented and 54 attended the training. The number of peer-supporters per school ranged from 11 (16.67% Year 8 girls) to 17 (17.71% Year 8 girls) (Table 1). Ten trainers with a range of experience in physical activity, youth work, or both, expressed an interest and five were recruited. All trainers were female (Mean age = 33.8, SD = 9.68, range = 21-45).

Data provision

At each time point, accelerometer return rates exceeded 85% (Table 2). At Time 1 accelerometer return was approximately 10% lower in the control versus intervention arm. Compliance with the wear protocol was not different between arms at baseline, and was 5% and 7% lower in control arm at Times 1 and 2 respectively. At Times 0, 1 and 2, the wear time criteria was met by 82.63%, 71.13% and 62.21% of participants respectively. The psychosocial questionnaire was completed by >91% of participants at each time point and was similar between trial arms. The completion of the EQ-5D-Y exceeded 92% at each time point.

Table 2 Accelerometer and questionnaire data provision rates by trial arm at baseline, Time 1 and Time 2.

Baseline data

The trial arms were well balanced at baseline (Table 3). IMD (i.e., deprivation) was slightly higher in the intervention versus control group although both medians were in the second quintile range (8.5 to 13.8) of least deprived households in England [27]. Weekday and weekend minutes of MVPA were similar between arms. Participants in the intervention group recorded more minutes of sedentary time on weekdays and overall. Peer-supporters had higher family affluence and similar IMD to non-peer-supporters and a greater proportion of peer-supporters than non-peer-supporters were of white ethnicity (Additional file 3). Peer-supporters recorded approximately 12 minutes more MVPA on weekdays and weekends and less sedentary time (mainly due to weekday activity) than non-peer-supporters. Physical activity guidelines were met by 59% of peer-supporters versus 35% of non-peer-supporters. Peer-supporters also reported higher EQ-5D-Y scores.

Table 3 Baseline (Time 0) descriptive data by trial arm.

Attendance at peer-supporter training

Almost all (94%; (53/55)) peer-supporters attended all of the training (Table 4). In the majority of cases, absences were due to illness. Attendance at training was not different in the school where training was delivered on the school site (School 4), compared to the off-site training for pupils in the other schools. There were no adverse events.

Table 4 Attendance at the peer-supporter training

Evidence of Promise

There was no evidence of a difference in weekday or weekend MVPA between intervention and control groups at Time 1 (Table 5). There was some evidence that the intervention group were less sedentary on weekdays than the control group at Time 1 (-31.8 minutes, 95% CI = -57.44, -6.18). At Time 2, there was evidence that pupils in the intervention arm performed more weekday MVPA than controls (6.09 minutes, 95% CI = 1.43, 10.76) which reflected the intervention group maintaining their baseline level of MVPA while the MVPA of the control group participants decreased over time. Similar to Time 1, there was some evidence that the intervention group were less sedentary on weekdays at Time 2 (-23.26 minutes, 95% CI = -43.73, -2.79). The confidence intervals around the estimated differences in weekend sedentary behaviour were too wide to indicate a potential difference between groups. In both arms EQ-5D-Y scores decreased between Time 0 and Time 1 and increased between Time 1 and Time 2 (Additional file 4). Differences between groups for unadjusted EQ-5D-Y scores and estimated QALYs were small with confidence intervals including (or close to) zero. Sensitivity analysis based on the imputed data set and when the criterion for participant inclusion based on their accelerometer data was relaxed to ≥1 valid day were very similar to the complete case analysis in both point estimate and 95% confidence intervals (Additional file 5). There was some evidence that peer-supporters benefitted more from the intervention than non-peer supporters at Time 1, but no evidence of a difference between the groups at Time 2 (Additional file 6).

Table 5 Adjusted between-group differences in physical activity variables at Time 1 and Time 2

Sample Size

The intra-class correlation (ICC) on weekday minutes of MVPA (i.e., the primary outcome for a definitive trial) from the PLAN-A sample at each time point was; Time 0: <0.00 (95% CI = 0.0, 0.09), Time 1: 0.02 (95% CI = 0.0, 0.06) and Time 2: <0.0001 (95% CI = 0.0, 0.03). Based on these findings, and ICCs from other studies involving adolescent girls, sample size calculations for a definitive trial were based on an ICC of 0.01 (and were also modelled for scenarios where the ICC was 0.02 & 0.03). To detect a 10-minute between arms difference in weekday MVPA with 90% power, an alpha of 5%, and with a design effect of 1.7 (based on 70 students per school), 560 students are required (280 per arm) (Table 6). This was inflated to 800 students in 12 schools to account for 30 per cent loss to follow-up. Using a smaller between-arms difference in MVPA of 6 minutes (i.e., the point estimate from this feasibility study) and keeping all other factors constant, 980 students would be required (490 per arm), inflated to 1400 students (700 per arm) in 20 schools to account for 30% loss to follow-up.

Table 6 Sample size calculation for a future definitive trial


The average cost per school of delivering the two-day and top-up day peer-supporter training were £1490 and £794 respectively, and the total cost of delivering the intervention in each school ranged from £2309 to £3235 (average £2685 per school and £37 per Year 8 girl) (Table 7). Trainer time and travel costs, whether the students and school contact incurred travel costs and whether the training was delivered in or outside of school were key cost drivers. Peer-supporters who attended the student briefing, peer nomination, peer-supporter meeting and all three days of training spent on average 16 hours and 34 minutes taking part in PLAN-A.

Table 7 Costs of intervention delivery


This study showed that the PLAN-A intervention can be delivered and that it is feasible to conduct the research required to evaluate its effectiveness and cost-effectiveness in a definitive trial. The study was acceptable to schools and students with approximately 95% of eligible students in the six study schools participating at baseline and 93% of these were retained at the final follow up. This level of involvement is important for the potential effectiveness of the intervention and evaluation of its effect given that PLAN-A operates at the whole school year level.

At baseline approximately 60% of girls did not meet the recommendations of 60 minutes of MVPA per day and average MVPA levels were approximately 50 and 40 minutes per week day and weekend day respectively. These findings are similar to a recent UK accelerometer study which showed that 68% of adolescent girls performed less than 60 minutes MVPA per day [43] and support the present and future research efforts to increase girls’ physical activity.

Recruitment of peer-supporters was successful as 97% of those nominated and invited consented to the role and 94% attended all the peer-supporter training, suggesting that the peer-supporter concept was acceptable to adolescent girls. Peer-supporters were more physically active on average than non-peer-supporters at baseline showing that more active girls had been nominated to be peer-supporters. However, 40% of peer-supporters did not meet the recommended 60 minutes of MVPA per day. Peer-supporter training was delivered as planned at a non-school venue for three of the four intervention schools and in one school was delivered on the school site as financial/staffing constraints prevented the school from releasing a member of staff to chaperone the peer-supporters off site. The purpose of off-site delivery is to avoid in-school distractions (e.g., other pupils, prior commitments for example at lunch time, space constraints), to allow peer-supporters to work together and focus on a shared goal with their (non-school teacher) trainers, and to inspire peer-supporters with a special or “grown up” experience. As such, it would be ideal to deliver peer-supporter training off the school site. In future implementations of PLAN-A it would therefore be helpful to develop a persuasive pitch to school head teachers explaining the relatively low burden placed on school staff by PLAN-A, compared to other interventions [12, 21, 44, 45] to garner their support for off-site peer-supporter training. The intervention delivery, receipt and fidelity (including analysis of the school-based peer-supporter training) will be reported in full in a separate process evaluation paper.

We showed that the research required to evaluate the effectiveness and cost-effectiveness of the intervention in a larger trial is feasible. Questionnaire data provision was very high (>90%) at all time points. Accelerometer return rates exceeded >85% at each time point, however compliance with the wear time criteria decreased to 62% at Time 2 and was 5-7% lower in the control versus intervention group. Despite this, the level of completely missing accelerometer data (i.e., not wearing the monitor at all on a given day) was low. This suggests that whilst participants were willing to wear the accelerometers, strategies are needed to ensure that they are worn for long enough on measurement days, especially amongst control group participants, in a future definitive trial. Such strategies could include incentivising sufficient wear time rather than, or in addition to, monitor return [46, 47], amending the study design to reduce participant fatigue (i.e., accelerometery at baseline and beginning of Year 9 in the first instance), or using alternative accelerometers (e.g., wrist worn units) which may be less obtrusive and more comfortable. Even using these strategies, some missing accelerometer data is inevitable, and we also showed that imputation of missing accelerometer data was feasible and had no discernible impact on the estimates of intervention effectiveness.

The study showed that the PLAN-A intervention has the potential to positively affect adolescent girls’ MVPA (≈ +6 minutes per weekday) and reduce their sedentary time (≈ -23 minutes per weekday) compared to controls 4-5 months post-intervention. The estimates in this study suggest that the between-arm difference represents a stemming of the decline in MVPA that was seen in the control group. Previous school-based physical activity interventions for adolescent girls show very small effects [10] and some multicomponent school-based interventions were more effective for boys than girls [12]. The findings of the present study suggest that PLAN-A has the potential to have a larger effect than previous interventions and provides initial support for the alternative peer-led physical activity intervention model used in PLAN-A.

The evidence presented above suggests that the conduct of a larger definitive trial to determine the effectiveness and cost-effectiveness of the PLAN-A intervention is warranted. Sample size calculations suggested that this study should include 20 schools (ten intervention & ten control) and 1400 Year 8 girls to have sufficient statistical power to detect a 10-minute between-arms difference in MVPA. The details of minor amendments to the intervention content that were identified in the process evaluation will be described in a separate paper. In comparison to other studies PLAN-A has the potential to be good value. PLAN-A cost £37 per Year 8 girl and £6 per additional minute of weekday MVPA (based on the between-arms comparison at Time 2) which compares favourably, for example, to the multicomponent Physical Activity 4 Everyone trial [48] which had an intervention cost over 24 months of £240 per student and a difference in mean daily MVPA of 7.0 minutes (95% CI: 2.7 to 11.4) which equated to £34 (95% CI: £21 to £90) per additional minute of MVPA.

Strengths and Limitations

A main strength of this study was its methodology. The PLAN-A intervention was carefully adapted from an effective adolescent smoking cessation intervention model through formative work with key stakeholders [40]. As such, the intervention used a novel means of peer-support for physical activity. The study was aligned with guidance on the development of complex interventions [49] and provided important testing of procedures, estimates of recruitment, indicators of intervention feasibility and promise to change behaviour and determination of the sample size required for a future effectiveness trial. Further strengths are that the study conduct adhered to the previously published study protocol [25], the use of an independent study oversight group, the use of accelerometers to measure physical activity, and the grounding of the intervention on complementary theories of peer-influence and motivation. A limitation of the study is the involvement of only two control schools which provides little information on recruiting and retaining schools in the control arm. Further, there was limited ethnic diversity of participants across all schools and additional research is needed to explore feasibility in schools with a more ethnically diverse pupil population. Finally, the relatively high level of missing accelerometer data at the second follow-up would need to be improved in a future trial.


It is feasible to implement a secondary school-based peer-led physical activity intervention for adolescent girls based on trained influential students providing informal support to their year group peers. It is also feasible to study this intervention using a robust cluster-randomised trial design. The intervention shows promise to stem the age-related decline in adolescent girls’ physical activity, and overall the evidence presented suggest that a definitive trial is warranted to investigate the effectiveness and cost-effectiveness of the intervention.



Diffusion of innovations


Intra-class correlation


index of multiple deprivation


moderate to vigorous physical activity


Peer-Led physical Activity iNtervention for Adolescent girls.


Self-determination theory


  1. 1.

    Janssen I, LeBlanc AG. Systematic review of the health benefits of physical activity and fitness in school-aged children and youth. Int J Behav Nutr Phys Act. 2010;7:40.

    Article  PubMed  PubMed Central  Google Scholar 

  2. 2.

    Strong WB, et al. Evidence based physical activity for school-age youth. J Pediatr. 2005;146(6):732–7.

    Article  PubMed  Google Scholar 

  3. 3.

    Ekelund U, et al. Moderate to vigorous physical activity and sedentary time and cardiometabolic risk factors in children and adolescents. JAMA. 2012;307(7):704–12.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  4. 4.

    Department of Health, Start Active, Stay Active: A report on physical activity from the four home countries’ Chief Medical Officers. 2011.

  5. 5.

    U.S. Department of Health and Human Services, 2008 Physical Activity Guidelines for Americans. 2008: Hyattsville, MD.

  6. 6.

    Hallal PC, et al. Global physical activity levels: surveillance progress, pitfalls, and prospects. Lancet. 2012;380(9838):247–57.

    Article  PubMed  Google Scholar 

  7. 7.

    Corder K, et al. Age-related patterns of vigorous-intensity physical activity in youth: The International Children's Accelerometry Database. Prev Med Rep. 2016;4:17–22.

    Article  PubMed  PubMed Central  Google Scholar 

  8. 8.

    Dumith SC, et al. Physical activity change during adolescence: a systematic review and a pooled analysis. Int J Epidemiol. 2011;40(3):685–98.

    Article  PubMed  Google Scholar 

  9. 9.

    Metcalf BS, et al. Exploring the Adolescent Fall in Physical Activity: A 10-yr Cohort Study (EarlyBird 41). Med Sci Sports Exerc. 2015;47(10):2084–92.

    Article  PubMed  Google Scholar 

  10. 10.

    Owen MB, et al. The effectiveness of school-based physical activity interventions for adolescent girls: A systematic review and meta-analysis. Prev Med. 2017;105:237–49.

    Article  PubMed  Google Scholar 

  11. 11.

    Pearson N, Braithwaite R, Biddle SJ. The effectiveness of interventions to increase physical activity among adolescent girls: a meta-analysis. Acad Pediatr. 2015;15(1):9–18.

    Article  PubMed  Google Scholar 

  12. 12.

    Sutherland RL, et al. The Physical Activity 4 Everyone Cluster Randomized Trial: 2-Year Outcomes of a School Physical Activity Intervention Among Adolescents. Am J Prev Med. 2016;51(2):195–205.

    Article  PubMed  Google Scholar 

  13. 13.

    Biddle SJH, et al. Correlates of participation in physical activity for adolescent girls: A Systematic review of recent literature. Journal of Physical Activity and Health. 2005;2:421–32.

    Article  Google Scholar 

  14. 14.

    Spencer RA, Rehman L, Kirk SF. Understanding gender norms, nutrition, and physical activity in adolescent girls: a scoping review. Int J Behav Nutr Phys Act. 2015;12:6.

    Article  PubMed  PubMed Central  Google Scholar 

  15. 15.

    Davison KK, Jago R. Change in parent and peer support across ages 9 to 15 yr and adolescent girls' physical activity. Medicine and Science in Sports and Exercise. 2009;41(9):1816–25.

    Article  PubMed  PubMed Central  Google Scholar 

  16. 16.

    Smith AL. Peer relationships in physical activity contexts: A road less traveled in youth sport and exercise psychology research. Psychology of Sport and Exercise. 2003;4:25–39.

    Article  Google Scholar 

  17. 17.

    Fitzgerald A, Fitzgerald N, Aherne C. Do peers matter? A review of peer and/or friends' influence on physical activity among American adolescents. J Adolesc. 2012;35(4):941–58.

    Article  PubMed  Google Scholar 

  18. 18.

    Martin Ginis KA, Nigg CR, Smith AL. Peer-delivered physical activity interventions: An overlooked opportunity fo physical activity promotion. Transl Behav Med. 2013;3:434–43.

    Article  Google Scholar 

  19. 19.

    Tymms PB, et al. Clustered randomised controlled trial of two education interventions designed to increase physical activity and well-being of secondary school students: the MOVE Project. BMJ Open. 2016;6(1):e009318.

    Article  PubMed  PubMed Central  Google Scholar 

  20. 20.

    Brown HE, et al. A cluster randomised controlled trial to evaluate the effectiveness and cost-effectiveness of the GoActive intervention to increase physical activity among adolescents aged 13-14 years. BMJ Open. 2017;7(9):e014419.

    Article  PubMed  PubMed Central  Google Scholar 

  21. 21.

    Edwardson CL, et al. A cluster randomised controlled trial to investigate the effectiveness and cost effectiveness of the 'Girls Active' intervention: a study protocol. BMC Public Health. 2015;15:526.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  22. 22.

    Harden A, Oakley A, Oliver S. Peer-delivered health promotion for young people: A systematic review of different study designs. Health Education Journal. 2001;60:339–53.

    Article  Google Scholar 

  23. 23.

    Campbell R, et al. An informal school-based peer-led intervention for smoking prevention in adolescence (ASSIST): A cluster randomised trial. Lancet. 2008;371:1595–602.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  24. 24.

    Craig P, et al. Developing and evaluating complex interventions: the new Medical Research Council guidance. Int J Nurs Stud. 2013;50(5):587–92.

    Article  PubMed  Google Scholar 

  25. 25.

    Sebire SJ, et al. Protocol for a feasibility cluster randomised controlled trial of a peer-led school-based intervention to increase the physical activity of adolescent girls (PLAN-A). Pilot and Feasibility Studies. 2016;2(2)

  26. 26.

    Sebire SJ, et al. Update to a protocol for a feasibility cluster randomised controlled trial of a peer-led school-based intervention to increase the physical activity of adolescent girls (PLAN-A). Pilot Feasibility Stud. 2016;2:68.

    Article  PubMed  PubMed Central  Google Scholar 

  27. 27.

    Department for Communities and Local Government. English indices of deprivation 2015. 2015 30 September 2015; Available from:

  28. 28.

    National Institute of Health Research - Journals Library. Development and feasibility cluster randomised control trial evaluation of a Peer-Led physical Activity iNtervention for Adolescent girls (PLAN-A). 2018; Available from:

  29. 29.

    Rogers, E.M., Diffusion of Innovations. 1983, NY: The Free Press.

  30. 30.

    Deci EL, Ryan RM. The "what" and "why" of goal pursuits: Human needs and the self-determination of behavior. Psychological Inquiry. 2000;11:227–68.

    Article  Google Scholar 

  31. 31.

    Teixeira PJ, et al. Exercise, physical activity, and self-determination theory: A systematic review. Int J Behav Nutr Phys Act. 2012;9(78)

  32. 32.

    Boyce W, et al. The family affluence scale as a measure of national wealth: Validation of an adolescent self-report measure. Social Indicators Research. 2006;78:473–87.

    Article  Google Scholar 

  33. 33.

    De Vries SI, et al. Validity and reproducibility of motion sensors in youth: a systematic update. Med Sci Sports Exerc. 2009;41(4):818–27.

    Article  PubMed  Google Scholar 

  34. 34.

    Evenson KR, et al. Calibration of two objective measures of physical activity for children. J Sports Sci. 2008;26(14):1557–65.

    Article  PubMed  Google Scholar 

  35. 35.

    Trost, S.G., et al., Comparison of accelerometer cut points for predicting activity intensity in youth. Med Sci Sports Exerc. , 2011. 43(7): p. 1360-1368.

  36. 36.

    Wille N, et al. Development of the EQ-5D-Y: a child-friendly version of the EQ-5D. Qual Life Res. 2010;19(6):875–86.

    Article  PubMed  PubMed Central  Google Scholar 

  37. 37.

    Dolan P, Gudex C, Kind P, Williams A. A social tariff for EuroQol: Results from a UK general population survey. 1995, University of York: York. England.

  38. 38.

    Manca A, Hawkins N, Sculpher MJ. Estimating mean QALYs in trial-based cost-effectiveness analysis: the importance of controlling for baseline utility. Health Econ. 2005;14(5):487–96.

    Article  PubMed  Google Scholar 

  39. 39.

    White IR, Royston P, Wood AM. Multiple imputation using chained equations: Issues and guidance for practice. Statistics in Medicine. 2011;30(4):377–99.

    Article  PubMed  Google Scholar 

  40. 40.

    National Institute for Health Research. Development and feasibility cluster randomised control trial evaluation of a Peer-Led physical Activity iNtervention for Adolescent girls (PLAN-A) - Project Webpage. 2018; Available from:

  41. 41.

    Jago R, et al. Effect and cost of an after-school dance programme on the physical activity of 11-12 year old girls: The Bristol Girls Dance Project, a school-based cluster randomised controlled trial. Int J Behav Nutr Phys Act. 2015;12:128.

    Article  PubMed  PubMed Central  Google Scholar 

  42. 42.

    Murray DM, et al. School-level intraclass correlation for physical activity in sixth grade girls. Med Sci Sports Exerc. 2006;38(5):926–36.

    Article  PubMed  PubMed Central  Google Scholar 

  43. 43.

    Collings PJ, et al. Levels and patterns of objectively-measured physical activity volume and intensity distribution in UK adolescents: the ROOTS study. Int J Behav Nutr Phys Act. 2014;11

  44. 44.

    Tibbitts B, et al. Action 3:30R: protocol for a cluster randomised feasibility study of a revised teaching assistant-led extracurricular physical activity intervention for 8- to 10-year-olds. Pilot Feasibility Stud. 2017;3:69.

    Article  PubMed  PubMed Central  Google Scholar 

  45. 45.

    Webber LS, et al. Promoting physical activity in middle school girls: Trial of Activity for Adolescent Girls. Am J Prev Med. 2008;34(3):173–84.

    Article  PubMed  PubMed Central  Google Scholar 

  46. 46.

    Bell S, et al. Lessons from a peer-led obesity prevention programme in English schools. Health Promotion International. 2014;

  47. 47.

    Sirard JR, Slater ME. Compliance with wearing physical activity accelerometers in high school students. J Phys Act Health. 2009;6(Suppl 1):S148–55.

    Article  PubMed  PubMed Central  Google Scholar 

  48. 48.

    Sutherland R, et al. Cost effectiveness of a multi-component school-based physical activity intervention targeting adolescents: the 'Physical Activity 4 Everyone' cluster randomized trial. Int J Behav Nutr Phys Act. 2016;13:94.

    Article  PubMed  PubMed Central  Google Scholar 

  49. 49.

    Craig, P., et al., Developing and evaluating complex interventions: New guidance 2008, Medical Research Council.

Download references


We are very grateful to the schools, pupils and peer-supporter trainers that took part. We would also like to thank the independent members of our Local Advisory Group and Trial Steering Committee. Thank you also to Lisa Campbell for her administrative support, Keeley Tomkinson for statistical analysis support in the early project phases and Dr Sarah Bell for her assistance in refining the PLAN-A intervention. The work was undertaken with the support of The Centre for the Development and Evaluation of Complex Interventions for Public Health Improvement (DECIPHer), a UKCRC Public Health Research Centre of Excellence. Joint funding (MR/KO232331/1) from the British Heart Foundation, Cancer Research UK, Economic and Social Research Council, Medical Research Council, the Welsh Government and the Wellcome Trust, under the auspices of the UK Clinical Research Collaboration, is gratefully acknowledged. This study was designed and delivered in collaboration with the Bristol Randomised Trials Collaboration (BRTC), a UKCRC Registered Clinical Trials Unit in receipt of National Institute for Health Research CTU support funding. None of the funders had involvement in the Trial Steering Committee, the data collection, analysis nor interpretation, nor writing of the paper. The views expressed in this publication are those of the authors and not necessarily any of the funding bodies listed here. RJ is partly funded by National Institute for Health Research (NIHR) Collaboration for Leadership in Applied Health Research and Care West (CLAHRC West) at University Hospitals Bristol NHS Foundation Trust.

RAL receives funding through the Farr Institute of Health Informatics Research. The Farr Institute is supported by a consortium of ten UK research organisations: Arthritis Research UK, the British Heart Foundation, Cancer Research UK, the Economic and Social Research Council, the Engineering and Physical Sciences Research Council, the Medical Research Council, the National Institute of Health Research, the National Institute for Social Care and Health Research (Welsh Government), Wellcome Trust and the Chief Scientist Office (Scottish Government Health Directorates). MRC Grant No: MR/K006525/1.

Ethical Approval and Consent to Participate

Ethical approval for the study was given by the University of Bristol’s School for Policy Studies Research Ethics Committee (SPSREC14-15.A27).

Availability of data and material

Anonymised study data will be made available via a University of Bristol repository


This study was funded by the NIHR Public Health Research Programme (13/90/16). The views expressed are those of the authors and not necessarily those of the NHS, the NIHR or the Department of Health. Intervention costs were funded by Wiltshire Council and South Gloucestershire Council.

Author information




The Study was conceived by SJS, RJ, RC, WH, RK, PSB and RAL. MJE was the Project Manager, and KB was the qualitative Research Associate. Data analysis was performed by BK (statistical data) and KG (health economics data). SS produced the first draft of this manuscript with all other authors providing critical review and intellectual content. All authors read and approved the final manuscript.

Corresponding author

Correspondence to Simon J. Sebire.

Ethics declarations

Consent for publication

Consent to include anonymised information in publications was given by all participants.

Competing interests

The authors declare that they have no competing interests.

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Additional files

Additional file 1:

Consort 2010 extension for pilot and feasibility trials checklist. Completed CONSORT checklist (version adapted for reporting pilot and feasibility studies). (DOC 225 kb)

Additional file 2:

Template for Intervention Description and Replication (TIDieR) checklist for PLAN-A.Completed TIDieR Checklist describing the PLAN-A intervention. (DOCX 16 kb)

Additional file 3:

Baseline descriptive statistics of peer-supporter and non-peer-supporter students. Descriptive data for peer-supporters and non-peer-supporters at baseline. (DOCX 15 kb)

Additional file 4:

EQ-5D-Y and QALYs results, by treatment arm. Baseline, Time 1 and Time 2 EQ-5D-Y results and corresponding QALY estimates, by treatment arm (DOCX 14 kb)

Additional file 5:

Sensitivity analysis of evidence of promise for intervention effects on weekday MVPA at Time 1 and Time 2 using the imputed data set and complete data where accelerometer wear was one or more valid day. Sensitivity analysis of evidence of promise for intervention effects on weekday MVPA at Time 1 and Time 2 using the imputed data set and complete data where accelerometer wear was one or more valid day (DOCX 13 kb)

Additional file 6:

Exploratory interaction of the intervention effect on weekday MVPA at Time 1 and Time 2 and peer supporter status. Exploratory interaction of the intervention effect on weekday MVPA at Time 1 and Time 2 and peer supporter status. (DOCX 13 kb)

Rights and permissions

Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (, which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver ( applies to the data made available in this article, unless otherwise stated.

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Sebire, S.J., Jago, R., Banfield, K. et al. Results of a feasibility cluster randomised controlled trial of a peer-led school-based intervention to increase the physical activity of adolescent girls (PLAN-A). Int J Behav Nutr Phys Act 15, 50 (2018).

Download citation


  • Physical activity
  • Peers
  • Adolescent girls
  • Intervention
  • School