Browse/search for people

Publication - Ms Sian Wells

    Long-term effects of the Active for Life Year 5 (AFLY5) school-based cluster randomised controlled trial


    Anderson, EL, Howe, L, Kipping, R, Campbell, R, Jago, R, Noble, S, Wells, S, Chittleborough, C, Peters, T & Lawlor, D, 2016, ‘Long-term effects of the Active for Life Year 5 (AFLY5) school-based cluster randomised controlled trial’. BMJ Open, vol 6.


    Objective To investigate the long-term effectiveness (12-months after the end of the intervention) of a school-based intervention to improve physical activity and diet in children. Assessment immediately at the end of this intervention found no effect on primary outcomes but three (out of nine) secondary outcomes showed evidence of benefit (reported time spent screen viewing at weekends, consumption of snacks and of high energy drinks).
    Design Cluster randomised controlled trial. Clustering was at the level of the schools, with eligibility for study entry being: (i) any state primary or junior schools that (ii) provided education to children aged 8 to 11 years and (iii) were within the Bristol City and North Somerset administrative areas (both areas in the South West of England).
    Setting 60 primary schools in the South West of England.Participants Primary school children who were aged 8-9 years at recruitment, 9-10 years during the intervention, and 10-11 years at the long-term follow-up assessment. We recruited and randomised 2221 children from 60 schools; 1064 (30 schools) randomised to the intervention and 1157 (30 schools) to standard educational practice control. The main analyses numbers differed by outcome type between 522 (30) vs 527 (30) for accelerometer based outcomes to 1062 (30) vs 990 (30) for all questionnaire based outcomes for pupils (schools) in interventions vs control schools, respectively.Methods to minimise bias Randomisation of schools to groups was undertaken remotely by computer to ensure concealment of allocation from anyone delivering the intervention and any of the research team. Fieldworkers who collected outcome data were unaware of which group the schools had been randomly allocated to.
    Intervention Teacher training, provision of lesson and child-parent interactive homework plans and teaching materials.
    Main outcome measures Primary outcomes were accelerometer assessed minutes of moderate to vigorous physical activity (MVPA) per day, accelerometer assessed minutes of sedentary behaviour per day, and reported daily consumption of servings of fruit and vegetables.
    Results As in the previously published assessment immediately after the end of the intervention, none of the three primary outcomes differed between children in schools allocated to the intervention, compared to those in control schools at the end of the long-term follow-up (1-year after the end of the intervention). Differences in secondary outcomes, based on point estimates, were consistent with those at the immediate follow-up, with no evidence that these had diminished over time. However, statistically there was little evidence for an effect on these outcomes at this longer term follow-up. Comparing intervention to control schools, the difference in mean child-reported screen viewing at the weekend was -16.03 minutes (95%CI: -32.82, 0.73, p = 0.06), for servings of snacks per day the difference was -0.11 (95%CI: -0.29, 0.06, p = 0.19) and in servings of high energy drinks per day -0.20 (95%CI: -0.39, -0.01; p = 0.04); servings of high fat foods per day also appeared lower in the intervention group -0.12 (95%CI: -0.25, 0.00; p = 0.05). None of these reached our predefined level of statistical significance, especially after accounting for multiple testing. We did not anticipate harmful outcomes and therefore did not assess any.
    Conclusion School based curriculum interventions are unlikely to have a major public health impact on children’s diet and physical activity.
    Trial registration Current Controlled Trials ISRCTN50133740.
    Funding: This trial was funded by UK National Institute for Health Research (NIHR) Public Health Research Programme (09/3005/04). The funders had no role in the study design, data collection, analysis or interpretation of results.

    Full details in the University publications repository