Impact and cost of the peer-led Students As LifeStyle Activists programme in high schools

Abstract

Objective:

To determine the impact of a peer-led, school-based programme (Students As LifeStyle Activists; SALSA) on energy balance–related behaviours (EBRBs) in Grade 8 students, and the cost of implementing the programme.

Design:

Pre–post test

Setting:

High schools in New South Wales, Australia.

Methods:

Grade 10 students (15- to 16-year-olds) were trained by university students to deliver four lessons on healthy eating and physical activity to Grade 8 students (13- to 14-year-olds). Students completed an online questionnaire pre–post lessons on EBRBs and intentions to change EBRBs over the next month. Items included fruit, vegetable and sugar-sweetened beverage (SSB) intake, frequency of eating breakfast, participation in moderate-to-vigorous physical activity (MVPA) and school-day recreational screen-time. Pre–post changes were analysed using generalised estimating equations, adjusted for clustering. Standard methods were used to estimate implementation costs.

Results:

In total, 2,056 Grade 8 students from 23 high schools participated. Significant changes included 5.2% increase in eating >2 fruit serves/day (p < .001), 2.5% increase in eating >4 vegetable serves/day (p < .05), a 4.3% increase in drinking <1 cup/day of SSB (p < .001) but limiting screen-time <2 hours/day decreased by 4.4% (p < .001). There were significant improvements in students’ intentions to change EBRBs over the next month, with the exception of MVPA. The average actual cost of implementing the SALSA programme in 2014–2015 was AU$1,388 (US$958) per school and AU$9.97 (US$6.88) per student.

Conclusion:

The SALSA peer education programme had a positive impact on most of the dietary EBRBs examined. The cost evaluation showed that it is a relatively low-cost programme to implement.

Keywords

Adolescents dietary intervention high schools peer education physical activity

Background

Overweight and obesity are major global problems due to high prevalence rates and significant consequences to human health across the lifespan (Global BMI Mortality Collaboration, 2016; Guh et al., 2009). In 2013, the prevalence of adolescent overweight and obesity was higher in 10- to 14-year-olds (26%–28% in developed countries; 15% in developing countries) than 15- to 19-year-olds (21%–23%, 10%–11%, respectively) (Ng et al., 2014). However, from this point, the prevalence of overweight and obesity rises sharply and continues to increase into adulthood (Ng et al., 2014). While the mean body mass index in young people appears to be plateauing in many high-income countries (NCD-RisC, 2017), this has not been observed in New South Wales (NSW), Australia, where the proportion of adolescents affected by overweight and obesity has increased from 22% to 28% between 2010 and 2015 (Hardy et al., 2016). The recent Lancet Commission on Adolescent Health and Wellbeing outlined a robust case for investing in the primary prevention of overweight and obesity among young people aged 10–24 years, including a recommendation that ‘more work is needed on interventions that capitalise on peer and social network influences’ for obesity prevention. (Patton et al., 2016: 2456).

Several systematic reviews (Ginis et al., 2013; Griebler et al., 2017; Yip et al., 2016) have identified studies engaging young people in delivering a nutrition and/or physical activity intervention to their peers, primarily in school settings. In some cases, students delivered classroom lessons together with teachers (Lytle et al., 2004; Siega-Riz et al., 2011) or graduate health professionals (Lo et al., 2008). In others, they delivered workshops (Wilson et al., 2012), promoted school-based activities (Aceves-Martins et al., 2017; Agron et al., 2002; Fulkerson et al., 2004; Siega-Riz et al., 2011; Thomas and Ward, 2006), informally diffused health promotion messages (Bell et al., 2017), mentored younger students about their health-related choices (Thomas and Ward, 2006), or modelled healthy behaviours through face-to-face discussions and handouts (Bogart et al., 2011). Notably, there is marked heterogeneity in both the specific targets and effects of these programmes. With the exception of a Spanish (Aceves-Martins et al., 2017) and a UK-based study (Bell et al., 2017), these studies were conducted in the USA and Canada, and only two studies appear to have published cost evaluations (Bell et al., 2017; Ladapo et al., 2016). Hence, the number of studies contributing to the evidence-base for the outcomes of peer-led interventions among young people is modest, with no studies from the southern hemisphere and little information available about the financial costs associated with implementing peer-led programmes in schools.

The Students As LifeStyle Activists (SALSA) programme was developed through a partnership between local family physicians, schools and community health staff in 2005. The rationale was to extend the limited evidence-base for adolescent-delivered nutrition and physical activity programmes in Australia that could be easily implemented in high schools. The SALSA programme aimed to improve the energy balance–related behaviours (EBRBs) of young people aged 13–14 years; specifically with respect to diet, physical activity and screen time behaviours, including students’ setting proximal goals (e.g. intentions) to change these behaviours. The theoretical framework of the SALSA programme was provided by Social Cognitive Theory (Bandura, 1998) which proposes a reciprocal determinism in the interaction between a person, a targeted behaviour and a particular social context. The intervention strategies were based on the empowerment educational approach (Wallerstein and Bernstein, 1988), wherein students gain knowledge through group sharing and understanding of the social influences that affect their lives, set their own priorities and decide on group action in order to change conditions or resolve issues (Shah et al., 2011).

The programme was implemented using a peer education model, in which university students were trained to coach Grade 10 high school students (15- to 16-year-olds) as SALSA peer leaders, who then delivered four healthy lifestyle lessons to their peers in Grade 8 (13- to 14-year-olds). This model was adapted from an effective and relatively low-cost, peer-led school-based asthma programme conducted in Australia and Jordan (Al-sheyab et al., 2012; Otim et al., 2015; Shah et al., 2001).

We have previously demonstrated the programme’s feasibility and acceptability in high schools in the Western Sydney region (Shah et al., 2011, 2017), an area of social disadvantage in Sydney. We have also previously shown that the EBRBs of adolescents in this region are sub-optimal (Shrewsbury et al., 2018), with most not meeting Australian dietary and physical activity recommendations (National Health and Medical Research Council (NHMRC), 2013; Department of Health and Ageing, 2014). For example, only around half of Grade 8 students ate breakfast daily, ate ⩾2 serves fruit/day, did not drink sugar-sweetened beverages (SSBs) daily, and limited recreational screen time to <2 hours/day. Only 1 in 5 (19.6%) participated in moderate-to-vigorous physical activity (MVPA) daily and 1 in 10 (11.4%) ate ⩾5 serves vegetable/daily (Shrewsbury et al., 2018).

The purpose of this study was to examine the effect of the SALSA programme on pre–post changes of Grade 8 students’ frequency of eating breakfast, fruit and vegetable intake, SSB intake, participation in MVPA, recreational screen-time, and intentions to change these behaviours over the next month. The secondary aim was to calculate the cost of implementing the SALSA programme in high schools using standard economic costing methods (Drummond et al., 2005).

Methods

Study design

A quasi-experimental, pre–post test design was employed to determine the impact of the SALSA programme on Grade 8 students’ EBRBs and their intentions to change their EBRBs. The study was registered as a clinical trial through the Australian New Zealand Clinical Trials Registry (registration number ACTRN12617000712303). Ethical approval was granted from the University of Sydney Human Research Ethics Committee (approval 2014/203), the NSW Department of Education and Training (State Education Research Application Process 2014096) and the local Catholic Education Diocese. The study was designed and conducted in line with the Declaration of Helsinki (World Medical Association, 2013).

The study was powered to detect an effect size of 0.1 for the within-subject change in frequency of eating breakfast. Based on a pilot study of the SALSA programme, we expected approximately 15 students per class to complete the study and at least five Grade 8 classes to be available for assessment in each school. We estimated the within-school cluster correlation coefficient (WCC) for the change in breakfast frequency as 0.011. Assuming an effect size of at least 0.1 and WCC of 0.011, a minimum of 19 schools were required to achieve 80% power to detect a statistically significant within subject change (5% significance level, two-sided test). Twenty-three schools were recruited to allow for expected attrition in school participation.

School and student recruitment

The SALSA programme was offered to government and Catholic high schools (n = 88) in the Western Sydney region through a bulk-email invitation and follow-up phone call. Schools were recruited until the target sample (n = 23) was achieved. Each school identified a teacher to coordinate the timetabling of the SALSA programme within the school. All Grade 8 students at participating schools were eligible to participate and were given an information sheet about the study to share with their parents. Students or parents and caregivers could decline the student’s participation through an opt-out process at any stage in the study.

Intervention – the SALSA programme

The SALSA programme is a peer-led, school-based educational programme designed to motivate high school students to improve their EBRBs in line with Australian recommendations (Foley et al., 2017; Shah et al., 2011, 2017). The programme involves trained SALSA Champions (project staff, doctors, teachers) training university students as educators to coach volunteer Grade 10 students as peer leaders, to deliver four 70-minute healthy lifestyle lessons to Grade 8 students at their school with classroom teachers providing supervision. Grade 10 students are provided with a manual and scripted notes so that the delivery of the lessons are standardised across schools. The lesson content aligns with learning objectives of the NSW Australia Personal Development, Health and Physical Education (PDHPE) curriculum (NSW Education Standards Authority, 2003).

For this study, 4 one-day workshops were conducted to train 96 university students as SALSA Educators for the programme. The trained university students visited participating schools (n = 23) in groups of four to coach volunteer Grade 10 students (n = 519) to be SALSA peer leaders during a 5-hour workshop. Guided by the university educators, the peer leaders learned about the programme, teaching and leadership skills to help them successfully deliver the four SALSA lessons to Grade 8 students (Foley et al., 2017). The delivery of the four lessons by Grade 10 peer leaders was recommended to be delivered over a school term (i.e. 10–13 weeks), but this was dependent on each school’s timetable. Peer leaders were provided with a lesson plan guide and activities, games and resources to help achieve each lesson’s learning objective, as outlined below (Table 1).

Table 1.

SALSA lesson outline.

Lesson	Objective	Activity	Purpose
Food choices	For students to understand the benefits of eating healthy foods and regular meals	Video	For students to discuss some of the challenges faced by young people to eat breakfast and a healthy diet, to be active and reduce screen-time
Food choices		Lunch box lotto	To inform students about different healthy options for lunch
Movement matters	For students to understand the importance of regular physical activity	Physical activity cards	For students to understand the differences between planned, incidental, and sedentary activities
		Mapping spaces	To identify a range of facilities where students can be active in their local area
		Ultimate frisbee	To give students the opportunity to participate in a form of physical activity that is simple, fun, and easy to do
Healthy lifestyle	For students to gain knowledge and skills to make more informed healthy food choices	Energy and food intake	To show students that for a healthy body you need to select from a variety of healthy foods
		Healthy snacks	To show students how to include healthy snacks in their diet and how to avoid snacks that are high in fat, salt and sugar
		Cooking with your family	To learn easy ways to make healthy foods and encourage students to share cooking tips with their family and friends
		Charades	To increase awareness about health services available to improve well-being
SALSA Action	For students to apply the knowledge learnt in previous lessons to maintain a healthy lifestyle	Quiz show	To revise key facts learnt about healthy eating and physical activity
SALSA Action		Create personal SMART goals	For students to set a SMART goal (i.e. Specific, Measurable, Achievable, Realistic, Timely)

SALSA: Students As LifeStyle Activists.

Data collection

The study was conducted during school terms two and three (autumn and winter) of 2014 and 2015. The evaluation involved students completing the same online questionnaire pre- and post-SALSA programme. Students were provided with a URL to access the questionnaire, which was administered via LimeSurvey (LimeSurvey Project Team and Schmitz, 2015), an open-source survey tool, hosted on a secure university network.

Each student was allocated a unique identifier and completed the questionnaire at school, supervised by their classroom teacher and assisted by a member of the research team. Students’ demographic characteristics included their sex and the primary language spoken at home by their parents (language background). The language background was used to broadly classify students to four groups (English-speaking, Middle Eastern, Asian and ‘other’). Socioeconomic status (SES) was determined by school-level SES. In Australia, every schools’ SES is determined from the Index of Community Socio-Educational Advantage (ICSEA) score which is located on the Australian MySchool website (Australian Curriculum Assessment and Reporting Authority [ACARA], n.d.). The ICSEA score is calculated using key student-level factors (parents’ occupation, school education, and non-school education) and school-level factors (geographical location and proportion of Indigenous students). ICSEA was used as a proxy measure of students’ SES and schools were dichotomised into low and high SES based on the mean (mean = 1000; SD = 100) (ACARA, 2015).

The EBRB questionnaire items were based on previously validated items (Flood et al., 2005; Hardy et al., 2016; Prochaska et al., 2001) and included reporting on the frequency of eating breakfast (response categories: 0–7 days), daily intake of fruit (response categories: ‘I don’t eat fruit every day, 1 serve, 2 serves, 3 serves or more), daily intake of vegetables (response categories: ‘I don’t eat vegetables every day, 1 serve, 2 serves, 3 serves, 4 serves, 5 or more serves), daily intake of SSB (i.e. soft drink, sports drink or cordial) (response categories: I don’t drink these drinks every day, 1 cup, 2 cups, 3 cups, 4 or more cups) (note: one cup was defined as 250 mL) and in the past week on how many days did you do a total of 60 minutes of MVPA (response categories: 0–7 days), and activities involving screens (i.e. mobile phone, iPad, tablet, computer, gaming console or watching TV/DVDs) during free time on a normal school day (response categories: 0–1 hours, 1–2 hours, 2–3 hours, more than 3 hours).

Students were also asked to report their intentions to change EBRBs in the following month, with the exception of SSB. These questions asked ‘Over the next month … “I plan to eat breakfast every day”’ (response categories: agree, neither agree or disagree, disagree); ‘… how many serves of fruit do you plan to eat every day?’ (response categories I don’t plan to eat fruit every day, 1 serve, 2 serves, 3 or more serves); ‘… how many serves of vegetables do you plan to eat every day?’ (response categories; I don’t plan to eat vegetables every day, 1 serve, 2 serves, 3 serves, 4 serves, 5 or more serves); ‘… I plan to be physically active on all or most days of the week’ (response categories: agree, neither agree or disagree, disagree); ‘… I plan to spend less free time sitting and doing activities involving screens’ (response categories: agree, neither agree or disagree, disagree).

The post-programme questionnaire included three additional questions to asses students’ perception of the SALSA programme: ‘Would you recommend the SALSA programme to your peers?’ (a measure of programme acceptability), ‘What was the most important message from the SALSA programme?’ and ‘Have you talked about the SALSA programme with your family? If yes, what did you talk about’.

Cost evaluation

Standard economic costing methods were used to estimate the cost of implementing the SALSA programme (Drummond et al., 2005). Programme implementation resource items were identified by the research team using a template from an earlier cost evaluation of a similar school-based asthma programme model (Otim et al., 2015). All programme activities were recorded and the measurement of resources and the valuation of the resources were collected in Australian dollars (AUD) using 2014 market rates. Prices were converted to US dollars using a purchasing power parity conversion (OECD Data, 2018).¹ Programme inputs included salaries, travels, hire of facilities, catering and resources (i.e. peer leader manuals). Two analyses were performed: (1) ‘actual costs’, representing all items for which there was reimbursement, including SALSA personnel salaries, catering for SALSA educator workshops, printing peer leader manuals and incidentals; and (2) ‘overall costs’, which included the actual costs plus valuation of in-kind contributions. In-kind contributions included relief teacher salaries (to enable teachers to attend to SALSA champion duties), SALSA educators’ time and travel expenses, peer leaders’ time, workshop venue hire, and catering for the champion and peer leader workshops. In both analyses, we calculated the mean cost of implementing the SALSA programme per participating school (n = 23) and per participating Grade 8 student (n = 3,203).

Statistical analysis

Data were analysed using IBM SPSS Statistics version 24. Generalised estimating equations (GEE) with a robust covariance estimator and exchangeable correlation structure were used to estimate the individual-level summary statistics and their 95% confidence intervals (CIs) adjusted for clustering (school) at baseline and post programme (Campbell et al., 2007; Donner and Klar, 2000; Murray, 1998). The within student change from baseline to post programme was calculated for each student for each of the outcomes of interest. GEEs were then used to estimate individual-level summary statistics and CIs for these within subject changes adjusted for clustering (school). School-level ICSEA scores were dichotomised as low (below the Australian mean of 1000) or high (at or above 1000). Students’ responses were dichotomised as ‘meeting’ or ‘not meeting’ Australian recommendations for adolescents’ daily fruit and vegetable intake (NHMRC, 2013), recreational screen-time, and MVPA (DOH, 2014). We considered drinking less than one cup of SSBs per day and eating breakfast every day to be levels of intake that aligned most closely with dietary guidance (NHMRC, 2013). The three additional post-programme questionnaire items were reviewed by two SALSA team members to identify main themes and sub-themes. These data were then coded independently by each team member. Themes were identified using a grounded theory approach, consistent with a framework analysis (Glaser et al., 1968; Ritchie et al., 2013).

Results

Twenty-three high schools (Government = 19; Catholic = 3 Independent = 1) agreed to participate in the programme and 22 high schools (96%) completed the programme. Over half (57.9%) of participating schools were categorised as of low SES. Of the 3,203 Grade 8 students eligible to participate, 2,056 (64.2%) completed the programme. The reported daily attendance rate for NSW government schools during the study period was 90.1%. This suggests an expected 2600 students attended on both the pre- and post-assessment day, and hence 79% (i.e. (2056/2600) x 100) of the expected number of available students completed both assessments. The demographic characteristics of the students are presented in Table 1. The majority students were girls (54.1%) and were from English-speaking backgrounds (67.3%). The median number of days between pre- and post-programme evaluation was 95 days (interquartile range 74–154 days; Table 2).

Table 2.

Demographic characteristics of participants (n = 2,056).

Characteristic	n (%)
Sex
Boys	944 (46.9%)
Girls	1,112 (54.1%)
Language background
English	1,383 (67.3%)
Asian	461 (22.4%)
Middle Eastern	104 (5.1%)
Other	108 (5.2%)
School-level SES^a
Low (<1000)	1,190 (57.9%)
High (⩾1000)	866 (42.1%)

SES: socioeconomic status.

Based on the Index of Community Socio-Educational Advantage for each student’s school and dichotomised as low if it was below the Australian mean of 1,000, and high if it was 1,000 or more.

Figure 1 shows the reported changes pre–post programme in the number of serves of fruit, vegetables and the number of SSB cups (250 mL) students had daily (unbroken lines) and student’s intentions to change these behaviours (broken lines). There were significant increases in the proportion of daily fruit intake for ⩾1 serve (4.1% [95% CI: 2.3, 5.9]; p < .001), ⩾2 serves (5.2% [95% CI: 2.3, 8.1]; p < .001), and ⩾ 3 serves (3.2% [95% CI: 1.2, 5.2]; p < .05). There were significant changes in students intentions to increase their daily consumption of fruit over the next month to ⩾1 serve (1.67% [95% CI: 0.1, 3.1]; p < .05), to ⩾2 serves (6.3% [95% CI: 3.5, 9.0]; p < .001), and to ⩾3 serves (6.0% [95% CI: 3.7, 8.4]; p < .001]. For vegetables, there were significant increases in the proportion of students who reported consuming ⩾2 serves/day (2.8% [95% CI: 0.7, 5.0]; p < .01) and ⩾4 serves/day (2.5% [95% CI: 0.3, 4.6]; p < .05). Students’ intentions to increase their daily vegetable intake over the next month changed significantly for ⩾2 serves (2.6% [95% CI: 0.6, 4.5]; p < .01), ⩾3 serves (4.0% [95% CI: 1.9, 6.2]; p < .001), ⩾4 serves (6.4% [95% CI: 3.3, 9.4]; p < .001), and ⩾5 serves (3.6% [95% CI: 0.9, 6.3]; p < .01). There was a significant decrease in the proportions of students drinking ⩾1 cup SSB daily (−4.3% [95% CI: −5.7, −2.9]; p < .001) and ⩾ 2 cups daily (−2.3% [95% CI: −4.5, −0.1]; p < .05).

Figure 1.

Grade 8 students’ reported the following: (a) daily fruit intake and intentions, (b) daily vegetable intake and intentions and (c) sugar-sweetened beverage (SSB) intake (SSB intentions not assessed).

Table 3 shows the pre–post changes in students reported EBRBs and their intention to change EBRBs over the next month according to EBRB recommendations. There was a 5.2% increase in the proportion of students who reported eating ⩾2 serves fruit daily (p < .001), a 4.3% increase in students drinking <1 cup SSB daily (p < .001) and a 4.4% decrease in students’ limiting screen-time on school days to ⩽2 hours/day (p = .01). There were significant changes in the proportion of students intending over the next month to eat breakfast daily (p = .05), to eat ⩾2 serves fruit (p < .001), to eat ⩾5 serves of vegetables daily (p < .01) and to limit recreational screen time on school days to ⩽2 hours/day (p < .001).

Table 3.

Cluster-adjusted percentages of students meeting lifestyle behaviour recommendations pre- and post-SALSA, changes and 95% CIs.

EBRBs	Behaviour, % (95% CI)			p	Intention, % (95% CI)			p
EBRBs	Pre-SALSA	Post-SALSA	Change	p	Pre-SALSA	Post-SALSA	Change	p
Daily breakfast (n = 2,056)	52.2 (47.7, 56.7)	51.7 (46.9, 56.5)	−0.4 (−2.5, 1.6)	.675	66.9 (62.8, 70.8)	69.6 (65.4, 73.5)	2.9 (0.8, 5.0)	<.05
Fruit ⩾2 serves/day (n = 2,033)	51.5 (47.6, 55.4)	56.6 (53.1, 60.0)	5.2 (2.3, 8.1)	<.001	66.1 (62.9, 69.2)	72.4 (69.3, 75.3)	6.3 (3.5, 9.0)	<.001
Vegetables ⩾5 serves/day (n = 2,031)	11.1 (9.8, 12.5)	12.6 (10.6, 14.8)	1.5 (−0.6, 3.6)	.156	18.9 (17.3, 20.6)	22.4 (19.8, 25.2)	3.6 (0.9, 6.3)	<.01
<1 cup/day SSB (n = 2,013)	47.2 (40.3, 54.1)	52.4 (45.0, 59.7)	4.3 (2.9, 5.7)	<.001	n/a	n/a	n/a	n/a
MVPA ⩾ 60 minutes/day (n = 2,004)	19.3 (16.5, 22.4)	20.5 (17.1, 24.4)	1.2 (−0.8, 3.2)	.249	72.9^a (70.1, 75.6)	73.9^a (70.9, 76.8)	0.9 (−1.5, 3.3)	.468
Screen-time ⩽2 hours/day (n = 1,990)	56.4 (54.0, 58.7)	52.0 (49.4, 54.5)	−4.4 (−7.4, −1.4)	<.01	37.3^b (34.3, 40.3)	41.6^b (39.1, 44.1)	4.2 (1.9, 6.6)	<.001

SALSA: Students As LifeStyle Activists; EBRB: energy balance-related behaviours; MVPA: moderate-to-vigorous physical activity; Screen-time: recreational screen-time on school days; SSB: sugar-sweetened beverage; n/a SSB intentions were not assessed.

Bold values: significant.

Intention to be physically active on all or most days.

Intention to reduce recreational screen-time on school days.

Table 4 presents the prevalence and change in EBRB by sex, language background and school SES (i.e. study covariates). The percentage change in students meeting EBRB recommendations was not significantly different (as indicated by the p-value for interaction) between any covariate sub-groups (e.g. boys and girls), but there were statistically significant differences within covariate sub-groups (e.g. girls increased their intake of fruit). Specifically, there were significant increases in eating ⩾2 fruit serves/day among girls, students from English-speaking backgrounds and in students from low- and high-SES schools. There were significant increases in the proportion of all student sub-groups drinking <1 cup SSB/day. There were significant decreases in limiting recreational screen time on school days among boys, students from English-speaking backgrounds and students in high SES schools.

Table 4.

Cluster-adjusted percentages and change for meeting EBRB recommendations pre- and post-programme, by sex, language background and school-level SES (% [95% CI]).

EBRB	Sex^a		Language background^a		School-level SES^a
EBRB	Boys	Girls	English	Other	Low	High
Daily breakfast
Pre-programme	57.5 [51.8, 63.0]	47.5 [42.7, 52.3]	49.5 [44.8, 54.3]	56.5 [51.3, 61.6]	46.0 [42.7, 49.4]	63.0 [57.3, 68.2]
Post-programme	56.3 [50.5, 61.8]	47.1 [41.7, 52.6]	49.3 [43.9, 54.7]	56.6 [51.2, 61.8]	45.0 [41.6, 48.4]	63.4 [57.1, 69.3]
Pre–post change	−1.3 [−4.4, 1.8]	0.0 [−3.2, 3.1]	0.0 [−2.4, 2.5]	−1.6 [−4.7, 1.5]	−1.0 [−3.1, 1.1]	0.5 [−3.1, 4.0]
p value	.417	.986	.992	.307	.350	.802
⩾2 fruit serves/day
Pre-programme	50.5 [45.5, 55.5]	52.2 [48.3, 56.1]	49.8 [45.8, 53.9]	53.5 [48.6, 58.3]	49.1 [44.8, 53.3]	55.6 [48.9, 62.0]
Post-programme	54.2 [49.8, 58.6]	58.9 [54.9, 62.7]	56.0 [52.6, 59.2]	57.3 [51.6, 62.8]	53.5 [50.6, 56.5]	62.0 [55.5, 68.0]
Pre-post change	3.7 [−0.2, 7.7]	6.7 [2.7, 10.7]	6.0 [2.3, 9.6]	3.9 [−0.8, 8.6]	4.5 [0.4, 8.5]	6.2 [2.1, 10.2]
p value	.066	<.01	<.01	.105	<.05	<.01
⩾5 vegetable serves/day
Pre-programme	13.6 [11.5,16.1]	8.9 [7.6, 10.4]	10.3 [8.9, 11.8]	13.1 [10.3, 16.4]	10.6 [9.0, 12.4]	11.7 [9.8, 14.0]
Post-programme	14.9 [11.9, 18.5]	10.2 [8.7, 12.0]	11.7 [10.1, 13.4]	13.7 [10.5, 17.8]	11.1 [9.1, 13.5]	14.6 [11.3, 18.6]
Pre-post change	1.3 [−1.9, 4.5]	1.3 [−0.5, 3.2]	1.4 [−0.4, 3.3]	0.2 [−3.0, 3.4]	0.5 [−1.7, 2.7]	2.8 [−1.0, 6.5]
p value	.436	.159	.123	.916	.679	.148
<1 cup SSB/day
Pre-programme	40.9 [33.2, 49.0]	53.9 [47.2, 60.4]	45.8 [39.5, 52.3]	48.0 [39.1, 57.0]	38.1 [34.0, 42.5]	65.0 [57.1, 72.2]
Post-programme	46.3 [38.4, 54.5]	58.4 [51.1, 65.5]	50.6 [43.3, 57.8]	55.4 [46.7, 63.8]	43.3 [37.5, 49.2]	69.3 [61.0, 76.5]
Pre-post change	4.1 [1.6, 6.5]	5.6 [4.1, 7.7]	5.2 [3.4, 6.9]	3.7 [0.4, 7.1]	5.1 [3.2, 6.8]	3.3 [1.4, 5.3]
p value	.001	<.001	<.001	.031	<.001	.001
>60 minutes MVPA/day
Pre-programme	25.8 [22.3, 29.7]	13.0 [11.0, 15.3]	20.8 [17.6, 24.5]	15.2 [12.3, 18.5]	19.2 [16.6, 22.1]	18.9 [13.3, 26.3]
Post-programme	28.7 [24.5, 33.4]	12.8 [10.0, 16.2]	21.9 [18.1, 26.3]	17.0 [13.5, 21.1]	20.8 [16.4, 25.9]	20.1 [15.1, 26.3]
Pre-post change	2.8 [−0.4, 6.0]	−0.1 [−2.1, 1.9]	1.1 [−1.3, 3.5]	1.4 [−1.9, 4.7]	1.1 [−2.0, 4.3]	1.4 [−0.8, 3.5]
p value	.082	.937	.365	.399	.483	.207
<2 hours screen-time/day
Pre-programme	59.7 [56.3, 63.0]	53.6 [51.0, 56.3]	57.2 [54.6, 59.8]	54.6 [51.1, 58.0]	56.1 [52.4, 59.8]	56.5 [54.6, 58.4]
Post-programme	54.0 [50.3, 57.6]	49.8 [45.8, 56.3]	52.2 [49.1, 55.3]	51.8 [46.5, 57.0]	51.3 [47.9, 54.7]	52.9 [49.2, 56.6]
Pre-post change	−5.7 [−9.3, −2.0]	−3.5 [−8.1, 1.1]	−4.9 [−8.1, −1.7]	−3.4 [−9.2, 2.5]	−4.7 [−9.4, 0.0]	−3.9 [−6.4, −1.3]
p value	<.01	.140	<.01	.261	.052	<.01

EBRB: energy balance-related behaviours; SSB: sugar-sweetened beverage; MVPA: moderate-to-vigorous physical activity. Screen-time: recreational screen time on school days.

Bold values: statistically significant.

p interaction between covariate sub-groups was non-significant for all behaviours.

Programme costs

A breakdown of the costs of implementing the SALSA programme in 23 schools is shown in Table 5. The average actual cost of implementing the SALSA programme in 2014–2015 was AU$1,388 (US$958) per high school and AUD9.97 (US$6.88) per student. The overall cost, which included valuation of in-kind contributions, was estimated to be AUD4,579 (US$3160) per high school and AUD32.89 (US$22.69) per student.

Table 5.

Cost of providing the SALSA programme to 3,203 Year 8 students at 23 high schools in 2014 and 2015.

Programme inputs	Quantity	Unit	Unit cost, AU$^a	Actual cost, AU$^a	Overall cost, AU$^a	Source/comment
SALSA personnel salaries over 2 years
1 Project Manager	0.05 FTE	Annual	82,750/p.a	10,758	10,758	Actual and overall cost includes 30% on-costs. https://intranet.sydney.edu.au/employment/enterprise-agreement/schedule-1/professional-35.html
1 Project Officer	0.1 FTE	Annual	65,485/p.a.	17,026	17,026
SALSA champion training
2 Relief teachers for participating teachers	2	Days	340/day	In-kind^b	680	NSW Department of Education
2 Family physicians	2	Days	−	In-kind^c	In-kind^c	n/a
Catering	1		200	In-kind	200	Receipts
SALSA educator training workshops (n = 4)
2 Relief teachers for participating teachers	2	Days each	340/day	In-kind^b	1,360	NSW Department of Education
2 Family physicians	2	Days each	−	In-kind^c	In-kind^c	n/a
96 university students–reimbursement for time	6	Hours	16.87/hours	In-kind^d	9,717	Minimum wage (adult): https://www.fwc.gov.au/documents/sites/wagereview2014/decisions/c20141_order.pdf
University venue hire	24	Hours	88/hour	In-kind^e	2,112	Hire rate
Catering	4		200	800	800	Receipts
SALSA peer leader training workshops (n = 23)
96 SALSA educator–reimbursement for time	5	Hours	16.87/hour	In-kind^d	8,098	Minimum wage (adult)
96 SALSA educators–reimbursement for travel	1	Day pass	11.5/day	In-kind^d	1,104	https://www.transport.nsw.gov.au/sites/default/files/media/documents/2017/Train%20Statistics%202014.pdf
519 peer leaders–reimbursement for time	5	Hours	7.97/hour	In-kind^d	20,682	Minimum wage (16 year old)
23 relief teachers	1	Day each	340/day	In-kind^b	7,820	NSW Department of Education
Use of school library/hall	n/a	n/a	free	n/a	n/a	n/a
23 catered lunches	1	Lunch	100	In-kind	2,300	School estimate
Peer leader manuals	519	Manuals	4.55/manual	2,361	2,361	Receipt
Year 8 lessons (4 x 70 minute)
519 peer leaders–reimbursement for time	4.67	Hours	7.97/hr	In-kind^d	19,317	Minimum wage (16 year old)
Incidentals	n/a	n/a	n/a	1,000	1,000	Receipts
Total				31,945	105,335
Mean cost per school				1,388	4,579
Mean cost per Year 8 student				9.97	32.89

SALSA: Students As LifeStyle Activists; NSW: New South Wales.

AU$ = Australian dollars.

Service provided by health or education professionals who volunteered their time.

Family physician time was not costed as it was considered to be a part of their regular duties.

Service provided by University students from health and education faculties and Year 10 students who volunteered their time.

As the SALSA programme was run by University of Sydney staff, these venues were provide free of charge.

The participant experience

The main messages students recalled about the programme were (1) the overall health benefits of having a healthy lifestyle (n = 44%), (2) healthy eating (eating fruit and vegetables, breakfast and reducing junk foods) (n = 43%) and (3) being physically active (being more active, doing more exercise and sports) (n = 38%). Fourteen percent of students reported that they had discussed the SALSA programme with their families. The most common themes discussed with families included participation in the programme, overall health benefits of living a healthy lifestyle and being more physical active. The majority of students (72%) reported that they would recommend participating in the SALSA programme to their peers.

Discussion

Study findings indicate that participation in the SALSA programme resulted in a higher proportion of students who met recommendations for two of the six EBRBs that we examined, and a higher proportion reported intentions that aligned with recommendations for four of five EBRBs intentions that we assessed. We also observed improvements in vegetable intake at levels of intake below the recommendation. The cost evaluation determined that the SALSA programme is relatively low-cost to implement which shows promise for programme sustainability. A key strength of the SALSA programme model is that, within the high school setting, it successfully engages older students to empower and motivate younger students to make healthy lifestyle changes.

These results are similar to the EBRB outcomes in a study of SALSA peer leaders where there were increases in the proportion of peer leaders meeting recommendations for fruit and vegetable intake and reductions in those consuming >1 cup of SSBs daily (Foley et al., 2017). Furthermore, in both of our studies language spoken at home had no influence on EBRB changes. However, differences are apparent in our Grade 8 student and peer leader samples regarding the influence of sex and SES on EBRBs (Foley et al., 2017). For example, in Grade 8 students, we saw no difference in the change in the proportion of students who met EBRB recommendations pre- to post-SALSA between gender or SES sub-groups. This is in contrast with our finding in peer leaders where changes in the proportion of peer leaders eating breakfast daily and meeting the MVPA recommendations was significantly different by sex, and the proportion meeting screen-time recommendations was significantly different by SES (Foley et al., 2017). This disparity between our two studies may be attributed to the different developmental stage of participants (Grade 8 students were 13- to 14-year-olds versus peer leaders who were 15- to 16-year-olds); receiving the content in different forms, that is, as a participant only (Grade 8 students) versus participating in training and delivering the lessons (peer leaders) and peer leaders being volunteers (i.e. a select group), whereas all Grade 8 students were invited to participate in the SALSA programme.

Similarly to the SALSA programme in Australia, only a few US-based studies to date have examined the effect of engaging older high school students or university students to deliver dietary intervention lessons to younger high school students (Lo et al., 2008; Wilson et al., 2012). Others have engaged same-age peers in delivering similar lessons (Lo et al., 2008; Lytle et al., 2004; Siega-Riz et al., 2011). In line with our results, these US-based studies also achieved success in improving short-term fruit (Lytle et al., 2004; Siega-Riz et al., 2011; Wilson et al., 2012), vegetable (Lytle et al., 2004; Wilson et al., 2012), and SSB intake (Lo et al., 2008; Siega-Riz et al., 2011) but longer-term impacts are less well studied and should be a priority outcome in future studies (Lytle et al., 2004).

A major strength of the study is that a large sample of schools participated with over half categorised as low SES, and over 2,000 students from a diverse range of cultural backgrounds included. Non-completion of the baseline questionnaire was higher among students from schools with low SES; this may reflect higher levels of student absenteeism from these schools, which was the main reason for missing data.

Another strength was including the cost evaluation of implementing the SALSA programme. To our knowledge, only two other studies (Bell et al., 2017; Ladapo et al., 2016) with adolescent peer-delivered nutrition or physical activity elements have included a cost evaluation. In comparison to our cost evaluation, the average cost of implementing the 10-week AHEAD intervention was GBP 11,289 per school (Bell et al., 2017), which was equivalent to AU$26,868 in 2014 (using the purchasing power parity [PPP] method) (OECD Data, 2018). The 5-week SNaX intervention cost US$8,637 per school or US$5.68 per student (Ladapo et al., 2016), which was equivalent to AU$12,524 per school and AU$8.24 per student in 2014 (using the PPP method). The authors of the AHEAD study concluded that while the intervention was acceptable to schools and students, it showed ‘no clear or consistent evidence of promise’ in relation to physical activity and eating behaviour change, and the implementation costs were one of the main barriers to implementation of the AHEAD intervention on a larger scale (Bell et al., 2017).

The SNaX intervention, which showed increases in the proportion of students served fruit and lunch, showed a significant decrease in the proportion of students buying snacks; cost less per school and per student than the SALSA programme (Ladapo et al., 2016). This difference is likely to have been due to the entire school population being considered as the denominator in the SNaX cost calculations (Ladapo et al., 2016), whereas, in our study, we included Grade 8 students only in the denominator. If we had been less conservative, it may have been reasonable to also include the number of peer leaders in our denominator. Taking into account the limited cost-evaluation data on peer-delivered diet and physical activity interventions in high schools, the SALSA programme stands out as having the lowest implementation cost per school, along with positive influences on EBRBs and intentions in high school students. Building on this success, the SALSA programme is currently testing the feasibility of a socioecological approach to create supportive school, home and neighbourhood environments to sustain healthy lifestyle changes (Shah et al., 2017).

Limitations of this study include the quasi-experimental design, as it was not logistically feasible to randomise either high schools or individuals, and the fact that we did not ask schools to postpone any whole-of-school healthy eating and physical activity initiatives for the duration of the programme. These circumstances limit the definitive attribution of the positive changes we observed to the intervention. However, at the time of the study, there were no competing interventions identified that specifically targeted Grade 8 students. Furthermore, we have shown reproducibility of the association between SALSA programme participation and the overall changes in most of the EBRBs and intentions that we examined, through both the analysis in students in this article, and the previous analysis in peer leaders (Foley et al., 2017). However, like most other school-based dietary and physical activity interventions with peer-involvement in delivering content via lessons, it is a limitation that longer-term follow-up data were not examined. Similar to most large-scale, school-based studies, the EBRBs also were self-reported, which can be seen as a limitation.

This study demonstrates preliminary efficacy of the SALSA programme in increasing fruit and vegetable intake, reducing SSB intake, and increasing intentions to meet recommendations for EBRBs. The cost of the SALSA programme compares favourably with costs reported for peer-delivered nutrition/physical activity interventions involving young people. Longer-term impacts of peer education programmes on high school students’ EBRBs, with accompanying cost-effectiveness analyses, are gaps in the literature that warrant further study.

Footnotes

Acknowledgements

We thank all the students and staff who participated in this study. We also acknowledge the contribution of members of the SALSA Advisory Committee.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Australian Government Department of Health and the Western Sydney Primary Health Network.

Notes

References

ACARA (2015) Guide to understanding ICSEA (Index of Community Socio-educational Advantage) values. Available at: http://docs.acara.edu.au/resources/Guide_to_understanding_ICSEA.pdf

Aceves-Martins

Llaurado

Tarro

, et al. (2017) A school-based, peer-led, social marketing intervention to engage Spanish adolescents in a healthy lifestyle ( ‘we are cool’-Som la Pera Study): A parallel-cluster randomized controlled study. Childhood Obesity 13: 300–313.

Agron

Takada

Purcell

(2002) California project LEAN’s food on the run program: An evaluation of a high school-based student advocacy nutrition and physical activity program. Journal of the American Dietetic Association 102: S103–S105.

Al-sheyab

Gallagher

Crisp

, et al. (2012) Peer-led education for adolescents with asthma in Jordan: A cluster-randomized controlled trial. Pediatrics 129: e106–e112.

Australian Curriculum Assessment and Reporting Authority (n.d.) My school. Available at: http://www.myschool.edu.au/

Bandura

(1998) Health promotion from the perspective of social cognitive theory. Psychology and Health 13: 623–649.

Bell

Audrey

Cooper

, et al. (2017) Lessons from a peer-led obesity prevention programme in English schools. Health Promotion International 32: 250–259.

Bogart

Elliott

Uyeda

, et al. (2011) Preliminary healthy eating outcomes of SNaX, a pilot community-based intervention for adolescents. Journal of Adolescent Health 48: 196–202.

Campbell

Donner

Klar

(2007) Developments in cluster randomized trials and statistics in medicine. Statistics in Medicine 26: 2–19.

10.

Department of Health and Ageing (DOH) (2014) Australia’s physical activity and sedentary behaviour guidelines for young people (13–17 years). Available at: http://www.health.gov.au/internet/main/publishing.nsf/Content/health-pubhlth-strateg-phys-act-guidelines/$File/FS-YPeople-13-17-Years.PDF

11.

Donner

Klar

(2000) Design and Analysis of Cluster Randomization Trials in Health Research.. London: Arnold Publishing Co.

12.

Drummond

Sculpher

Torrance

, et al. (2005) Methods for Economic Evaluation of Health Care Programmes. Oxford: Oxford University Press.

13.

Flood

Webb

Rangan

(2005) Recommendations for short questions to assess food consumption in children for the NSW Health Surveys. Available at: https://sydney.edu.au/science/molecular_bioscience/cphn/pdfs/2005_short_questions_children.pdf

14.

Foley

Shrewsbury

Hardy

, et al. (2017) Evaluation of a peer education program on student leaders’ energy balance-related behaviors. BMC Public Health 17: 695.

15.

Fulkerson

French

Story

, et al. (2004) Promotions to increase lower-fat food choices among students in secondary schools: description and outcomes of TACOS (Trying Alternative Cafeteria Options in Schools). Public Health Nutrition 7: 665–674.

16.

Ginis

Nigg

Smith

(2013) Peer-delivered physical activity interventions: An overlooked opportunity for physical activity promotion. Translational Behavioral Medicine 3: 434–443.

17.

Glaser

Strauss

Strutzel

(1968) The discovery of grounded theory: Strategies for qualitative research. Nursing Research 17: 364.

18.

Global BMI Mortality Collaboration (2016) Body-mass index and all-cause mortality: Individual-participant-data meta-analysis of 239 prospective studies in four continents. The Lancet 388: 776–786.

19.

Griebler

Rojatz

Simovska

, et al. (2017) Effects of student participation in school health promotion: A systematic review. Health Promotion International 32: 195–206.

20.

Guh

Zhang

Bansback

, et al. (2009) The incidence of co-morbidities related to obesity and overweight: A systematic review and meta-analysis. BMC Public Health 9: 88.

21.

Hardy

Mihrshahi

Drayton

, et al. (2016) NSW Schools Physical Activity and Nutrition Survey (SPANS) 2015: Full Report. Sydney, NSW, Australia: NSW Department of Health.

22.

Ladapo

Bogart

Klein

, et al. (2016) Cost and cost-effectiveness of Students for Nutrition and eXercise (SNaX). Academic Pediatrics 16: 247–253.

23.

LimeSurvey Project Team and Schmitz

(2015) Limesurvey: An open source survey tool. Available at: http://www.limesurvey.org

24.

Coles

Humbert

, et al. (2008) Beverage intake improvement by high school students in Saskatchewan, Canada. Nutrition Research 28: 144–150.

25.

Lytle

Murray

Perry

, et al. (2004) School-based approaches to affect adolescents’ diets: results from the TEENS study. Health Education and Behavior 31: 270–287.

26.

Murray

(1998) Design and Analysis of Group-Randomized Trials. New York: Oxford University Press.

27.

NCD-RisC (2017) Worldwide trends in body-mass index, underweight, overweight, and obesity from 1975 to 2016: a pooled analysis of 2416 population-based measurement studies in 128.9 million children, adolescents, and adults. The Lancet 390: 2627–2642.

28.

Fleming

Robinson

, et al. (2014) Global, regional, and national prevalence of overweight and obesity in children and adults during 1980–2013: A systematic analysis for the Global Burden of Disease Study 2013. The Lancet 384: 766–781.

29.

NHMRC (2013) Australian dietary guidelines. Publication references N55a and N55f. Available at: https://www.nhmrc.gov.au/guidelines-publications/n55

30.

NSW Education Standards Authority (NESA). (2003) Personal development health and physical education (PDHPE) 7-10 Syllabus. Available at: https://educationstandards.nsw.edu.au/wps/portal/nesa/k-10/learning-areas/pdhpe/pdhpe-7-10

31.

OECD Data (2018) Purchasing power parities (PPP). Available at: https://data.oecd.org/conversion/purchasing-power-parities-ppp.htm

32.

Otim

Jayasinha

Forbes

, et al. (2015) Building evidence for peer-led interventions: Assessing the cost of the Adolescent Asthma Action program in Australia. Australian Journal of Primary Health 21: 438–443.

33.

Patton

Sawyer

Santelli

, et al. (2016) Our future: A Lancet commission on adolescent health and wellbeing. The Lancet 387: 2423–2478.

34.

Prochaska

Sallis

Long

(2001) A physical activity screening measure for use with adolescents in primary care. Archives of Pediatrics and Adolescent Medicine 155: 554–559.

35.

Ritchie

Lewis

McNaughton Nicholls

, et al. (2013) Qualitative Research Practice: A Guide for Social Science Students and Researchers. London: SAGE.

36.

Shah

Foley

Molinari

, et al. (2017) The Students As LifeStyle Activists (SALSA) program. British Journal of Sports Medicine 51: 1445–1446.

37.

Shah

Patching Van Der Sluijs

Lagleva

, et al. (2011) A partnership for health – Working with schools to promote healthy lifestyle. Australian Family Physician 40: 1011–1013.

38.

Shah

Peat

Mazurski

, et al. (2001) Effect of peer led programme for asthma education in adolescents: Cluster randomised controlled trial. BMJ 322: 583–585.

39.

Shrewsbury

Foley

Flood

, et al. (2018) School-level socioeconomic status influences adolescents’ health-related lifestyle behaviors and intentions. Journal of School Health 88: 583–589.

40.

Siega-Riz

El Ghormli

Mobley

, et al. (2011) The effects of the HEALTHY study intervention on middle school student dietary intakes. International Journal of Behavioral Nutrition and Physical Activity 8: 7.

41.

Thomas

Ward

(2006) Peer power: How Dare County, North Carolina, is addressing chronic disease through innovative programming. Journal of Public Health Management and Practice 12: 462–467.

42.

Wallerstein

Bernstein

(1988) Empowerment education: Freire’s ideas adapted to health education. Health Education Quarterly 15: 379–394.

43.

Wilson

Jones

McClish

, et al. (2012) Fruit and vegetable intake among rural youth following a school-based randomized controlled trial. Preventive Medicine 54: 150–156.

44.

World Medical Association (2013) World Medical Association Declaration of Helsinki: Ethical principles for medical research involving human subjects. JAMA 310: 2191–2194.

45.

Yip

Gates

, et al. (2016) Peer-led nutrition education programs for school-aged youth: A systematic review of the literature. Health Education Research 31: 82–97.