Parts of this chapter have been reproduced with permission from Willis et al.1 and Jago et al.2 These are Open Access articles distributed in accordance with the terms of the Creative Commons Attribution (CC BY 4.0) license, which permits others to distribute, remix, adapt and build upon this work, for commercial use, provided the original work is properly cited. See: https://creativecommons.org/licenses/by/4.0/. The text below includes minor additions and formatting changes to the original text.
Girls’ physical activity during adolescence
Physical activity during adolescence is associated favourably with body composition, blood pressure, cholesterol and blood lipid levels,3 as well as with higher well-being, self-esteem4 and academic performance.5 There is moderate evidence that physical activity tracks through adolescence and into adulthood,6–8 and physical activity in adulthood is strongly associated with reduced risk of heart disease, stroke, type 2 diabetes and all-cause mortality.9
On average, children and adolescents do not meet the current international guidance of 1 hour of moderate to vigorous physical activity (MVPA) per day.10 Girls are less physically active than boys at all ages, but especially in adolescence [difference in daily MVPA 17 minutes, 95% confidence interval (CI) 16 to 18 minutes], and are more sedentary (difference in sedentary time 22 minutes, 95% CI 19 to 25 minutes).10 Thus, there is a need to support girls to be more physically active.
Effectiveness of physical activity interventions for girls
Increasing physical activity among girls has been identified as a public health priority,11 and a number of studies have attempted to help girls to be more active. However, systematic review evidence suggests that interventions have been only minimally effective at increasing physical activity among young people12 and, specifically, among girls.13–15
In their meta-analysis of the effect of school-based interventions on objectively measured physical activity, Borde et al.12 pooled data from 12 studies (two with boys only and two with girls only) and found that the effect on both overall physical activity (standardised mean difference 0.02, 95% CI −0.13 to 0.18) and MVPA (standardised mean difference 0.24, 95% CI −0.08 to 0.56) was small and non-significant. Common components of the reviewed interventions included traditional top-down strategies, such as increasing active breaks, providing health education and information, running extra lessons and distributing pedometers for self-monitoring. The authors reported that studies with higher compliance with accelerometer protocols yielded stronger positive effects.12 A similar review and meta-analysis looking at school-based interventions targeting only girls concluded that, to date, interventions have recorded a very small effect on adolescent girls’ physical activity levels (k = 16, g = 0.07; p = 0.05); nonetheless, it also concluded that multicomponent interventions showed more promise, and interventions underpinned by theory may be more effective.14 This is aligned with findings from an earlier review that found that multicomponent interventions were more effective at increasing physical activity; however, the findings also differed in that atheoretical interventions of higher quality yielded more positive outcomes than those of lower quality that were underpinned by theory,13 identifying the importance of trial quality on results.
A narrative review of 15 interventions (10 of which were school based) aimed at changing objectively measured health outcomes (i.e. physical activity, body mass index, body fat percentage) in adolescent girls15 identified little evidence that interventions increased objectively assessed physical activity. The authors also highlighted that the impact of the studies may have been limited by attendance rates to the interventions, as well as compliance with accelerometer measurement of physical activity.15 Collectively, this evidence suggests that interventions to increase physical activity levels in adolescent girls, whether school based or not, are minimally effective, at best, and are potentially limited by methodological issues, such as the measurement of physical activity, intervention design and dose, or a combination of these factors.
Correlates of girls’ physical activity
Factors known to be associated with physical activity in adolescent girls are increased enjoyment, increased perceived competence, higher self-efficacy and higher socioeconomic position (SEP).16 There is also some evidence that adolescents from lower socioeconomic groups engage in lower levels of physical activity,17 which may, in part, be explained by less favourable psychosocial determinants of physical activity.18 Furthermore, a scoping review exploring gender norms in relation to physical activity and nutrition concluded that lifestyle components are greatly affected by feminine ideals through complex negotiations, perceptions, body-centred discourse and societal influences.19 Similarly, changes to friendship groups, social/peer support, perceived competence, competing priorities, body-centred issues (e.g. weight status and menstruation) and ‘sporty’ gender stereotypes encountered during adolescence are factors that may contribute to the decline in girls’ physical activity.20–23
Peers and physical activity
Peers play a central role in adolescents’ physical activity through peer support, co-participation, peer norms, friendship quality, peer affiliation and peer victimisation.24,25 Qualitative studies among adolescent girls have shown that co-participation in physical activity with peers is an important motivating factor,23 and that a shared interest in physical activity with friends is associated with higher levels of motivation to be active, as well as greater enjoyment of being active.26 Social network research has shown that adolescents have similar physical activity levels to their close friends, and they may adapt their physical activity behaviour over time to be more like that of their friends.27,28 Peer-led interventions, therefore, have the potential to increase adolescents’ physical activity. There is a need to develop effective physical activity interventions, especially among girls, that capitalise on existing peer processes in schools by promoting peer support and enhancing peer communication skills.24
Peer-led health interventions
Peer-led approaches have been used in attempts to change a range of health behaviours among young people, including tobacco, alcohol and drug use,29 nutrition education, water consumption30,31 and sexual health.32 A small number of studies have attempted to use peer influence to change physical activity and sedentary behaviours;33 however, more recently, studies using a peer-led approach to target physical activity in young people have increased. A recent scoping review of peer-led physical activity interventions highlighted several approaches to increasing physical activity; however, the results are mixed in terms of their effectiveness, with no clear indication of specific approaches or intervention factors affecting impact.34
The authors of this review note that many approaches use young people to deliver adult agendas. For instance, most peer-led health interventions to date have been delivered in secondary education settings; through information provision and skill development, these interventions also trained peer leaders to educate other pupils by teaching sections of the curriculum (that would otherwise be taught by adults) through buddy learning or by leading new physical activity initiatives and supporting teacher-led activities.34,35 However, using a different approach, young people in the role of peer leader could take more control over the intervention that they delivered, which may have enhanced their credibility among their peers, in contrast to delivering a clearly adult agenda.34,36
An alternative peer-led approach, based on diffusion of innovations theory (DOI),36 is to provide training to young people that empowers them with the information and skills to diffuse health promotion messages to their peers through informal means. DOI explains how ideas, beliefs or behaviours can be informally communicated through members of a social system in ways that are authentic to their peer group, an approach that was adopted in the effective and nationally disseminated ASSIST (A Stop Smoking in Schools Trial).37 The findings from ASSIST demonstrated that informal school-based peer-led interventions can be effective in changing smoking behaviour.37
Theoretical underpinning
Evidence of the influence of theory on the effectiveness of health behaviour interventions is, generally, mixed, although robust analysis of interventions, based on theories for which there are sufficient data to analyse (e.g. social cognitive theory and transtheoretical model), show that their use is unlikely to increase intervention effectiveness.38 Overall, the evidence is restricted in several ways; notably, there is a lack of studies with a robust theoretical basis (especially a limited use of theory during intervention development), limited use of a range of theories, inadequate mapping of behaviour change techniques to theory in interventions and scant use of theory to select participants or tailor the intervention.
Conversely, theory that is used with clarity and focus, and that is well matched to the intervention design and target group (including to select participants) and underpinned by a logic model highlighting mediating mechanisms, with clear mapping of behaviour change techniques to theoretical mediators, can provide an important element of intervention design, delivery and evaluation. Using theory effectively will also contribute higher-quality theory-based interventions that can advance research into the effectiveness of the use of theory.
Few peer-led physical activity interventions integrate theoretical principles in their design or delivery.33,34 The present trial combines two complementary theories: DOI and self-determination theory (SDT). DOI was used to inform the intervention design and the selection of key participants (i.e. peer supporters as agents of change). SDT was used to inform the content (i.e. the ‘what’ and the ‘how’) of the intervention, including behaviour change techniques to proposed mediators between the intervention (i.e. peer support) and the outcome (i.e. physical activity).
Diffusion of information uses pre-existing social systems, such as friendship networks, to diffuse information and influence behaviour.39 Therefore, it provides a framework for harnessing the influential capacities of change agents in social networks, who can informally diffuse positive health messages to their peers. The relevance of DOI for interventions targeting adolescent girls is clear because of the role that peers play in influencing drivers of physical activity.23,24,26
Self-determination theory40 is concerned with the personal and social conditions required to foster high-quality and sustainable motivation, and has been used extensively to understand young people’s motivation to be physically active,22,41,42 as well as to guide interventions.43–45 SDT argues that autonomous motivation (i.e. motivation based on choice and personal value or inherent satisfaction) is associated with more positive behavioural and psychological outcomes than controlled motivation (i.e. motivation driven by external factors, such as meeting others’ expectations or guilt). Evidence points to consistent positive associations between autonomous versus controlled motivation and physical activity in children, adolescents and adults alike.41,42,46–49 Autonomous motivation is associated with perceptions of autonomy, competence and social belonging;42,50 therefore, it is supported by social environments and interactions (e.g. interactions with peers) that satisfy these three psychological needs. SDT is, therefore, fitting for a peer-led intervention, as peers can create a social climate that either facilitates or undermines girls’ interest in,33 enjoyment of and motivation for physical activity;23,26 perceptions of competence; social support; and choice of how to be active.20,22
Summary of formative work
The Peer-Led physical Activity iNtervention for Adolescent girls (PLAN-A) intervention design was informed by extensive formative, pilot and feasibility research, including iterative qualitative research conducted to adapt the ASSIST37 peer-led smoking intervention model to target physical activity, as well as an intervention pilot51 and a cluster-randomised feasibility trial.52 The feasibility trial was conducted in six UK secondary schools (four intervention and two control) with girls in Year 8 of the UK education system (i.e. girls aged 12–13 years), and included a health economics assessment and a comprehensive mixed-methods process evaluation.51–53
The intervention consisted of three stages:
- Girls anonymously nominated the peers they felt were well suited to being a peer supporter (the peer nomination process).
- The 18% of girls who were most frequently nominated in each school were invited to attend 3 days of specialist peer supporter training to provide them with skills and information.
- The trained peer supporters then informally influenced the physical activity of their close peers using a range of supportive techniques over a 10-week diffusion period.
The intervention design was based on DOI, and SDT was layered into the delivery, resources and content of the peer supporter training. Pupils’ physical activity was measured at baseline, immediately after the 10-week diffusion period and then approximately 5 months after the intervention period.53
The results showed that the intervention was deliverable, acceptable and affordable. There was evidence of promise at the 5-month follow-up time point, at which time there was a 6.1-minute difference in weekday MVPA favouring the intervention arm, after adjustment (95% CI 1.4 to 10.8 minutes). In addition, mean sedentary time per day was 23 minutes lower (95% CI –43.7 to –2.8 minutes) in the intervention arm than in the control arm at the same time point.51 The economic evaluation demonstrated that the programme has potential for cost-effective delivery from a public-sector perspective. The mean cost of intervention delivery equated to £37 per Year 8 girl and £6 per additional minute of weekday MVPA,45 comparing favourably with other multicomponent physical activity interventions in adolescents.54 The mixed-methods process evaluation identified key refinements for the intervention delivery and content, including increasing participatory learning, focusing more on how to start conversations with peers and providing more support to overcome challenges to giving peer support.55
Rationale for the current trial
Physical activity during childhood is positively associated with physical and psychological health. Physical activity declines with age and, by adolescence, few girls are sufficiently active. School-based interventions have had limited success and novel interventions are needed to mitigate girls’ lack of participation in physical activity. The power of peer influence presents a natural and sustainable intervention opportunity that has received little attention in high-quality research to date. Having followed the Medical Research Council framework for the development of complex interventions56 and made revisions to intervention content informed by robust feasibility work, the aim of this definitive trial was to test whether or not the PLAN-A intervention can increase adolescent girls’ physical activity and be cost-effective.