The Active for Life Year 5 (AFLY5) school based cluster randomised controlled trial: study protocol for a randomized controlled trial

The school years for all of our assessments are in primary/junior school and data from Bristol City Council shows very little (~5% of children) movement of children between schools or leaving school in the Bristol area over these school years; where movement does occur it tends to be to schools in neighbouring city council areas, such as North Somerset. We had very few problems during the pilot and feasibility studies, with just 2% of participants leaving their school and 8% missing one or other of the baseline or follow-up assessments because of school absenteeism. In this full scale trial fieldwork will be organised so that children who are out of school (for example, due to illness) on the day of data collection will have the opportunity to provide some data on an alternative day, which was not possible in the pilot/feasibility work. We will work closely with schools to ensure that visits for any of the assessments do not coincide with school trips or other activities that would make data collection difficult. Our sample size calculation anticipates a 15% loss to follow-up by the 12 month long term follow-up, which experience leads us to believe is likely to be an overestimate.

We will use ActiGraph accelerometers[21] and the same protocol that we used in our feasibility/pilot work. Accelerometers will be set to record at 10-second epochs on the day after the accelerometers are handed out. Where possible, accelerometers will be given out on a Wednesday and will be collected the following Tuesday, downloaded, recharged, reinitialised and taken to the next school the following day. The accelerometers will be shown to the children and verbal instructions provided in the class room with all children together. Each child will be given their accelerometer individually at the time of the anthropometric measurements and the child will be asked what they remembered about when the accelerometer should be worn and removed. They will be asked to wear them during the day (except when bathing or swimming or participating in contact sports such as karate) until the fieldworkers return to collect them. This will allow data collection on three weekdays (Thursday, Friday and Monday) and both weekend days. We will use established procedures for analysing the accelerometer data and calculate time spent in MVPA and sedentary behaviour.

All anthropometric measurements will be completed with children in a private room with both CRB checked trained fieldworkers present. Weight will be measured without shoes in light clothing to the nearest 0.1 kg using a Seca digital scale. Height will be measured, to the nearest 0.1 cm, without shoes using a portable Harpenden stadiometer. Fieldworkers will be trained to ensure correct position for height assessment. WC will be measured the nearest 1 mm at the mid-point between the lower ribs and the pelvic bone[22] with a flexible tape. BMI and WC will be used to define whether a child is generally overweight/obese (using BMI) or centrally overweight or obese (WC), using appropriate gender and age specific thresholds from BMI and WC growth charts [23‐25]. For BMI we will use both international obesity task force thresholds and those based on the UK 1990 growth charts[23] and will compare results using these two different definitions.

Fruit and vegetable consumption and other dietary outcomes will be assessed using the "A Day in the Life Questionnaire" (DILQ),[26] which was used in our feasibility/pilot studies [14]. The DILQ provides information about the children's entire food and drink intake the previous day. To improve recall the questionnaire is structured with sequential questions in a 24 hour segmented school day. Fruit and vegetable consumption and other dietary outcomes will be assessed using an established scoring scheme [14, 27, 28]. As in the pilot study, allocation of foods written in text to categories will be undertaken independently by two individuals with discrepancies checked by a third independent individual, and a random 5% further coded by another researcher to check for accuracy. In the pilot/feasibility work there were very high levels of agreement (> 97%) [14]. An abbreviated and updated version of a previously validated screen viewing questionnaire was used to assess self-report sedentary behaviour in our feasibility/pilot study and this will be used here [29]. The questionnaire asks about the length of time spent doing screen based activities on the previous weekday and Saturday. The changes that we made (with the author's permission) included adding new media, such as Xbox and Play Station, and removing detailed questions about the number and location of TVs in the house and eating whilst watching TV, in order to reduce the length of the questionnaire.

We hypothesise that change in physical activity and fruit and vegetable consumption will be mediated by change in the child's self-efficacy. As the study is designed to increase parental support for physical activity we also hypothesise that parental support will function as a mediator of behaviour change. Increasing our understanding of the mechanisms of behaviour change is essential for designing and delivering more effective interventions and establishing how to disseminate a complex intervention that has been shown to be effective in one setting to another setting [30]. Therefore, we will assess four hypothesized mediators using well-established scales that have been shown to have good reliability among primary school aged children. Specifically, physical activity and screen-viewing self-efficacy will be assessed using the 9-item scale of activity and sedentary behaviour self-efficacy[31] and fruit and vegetable self-efficacy will be assessed using an 18-item scale that assesses child's self efficacy for this behaviour [32]. Parental support for physical activity will be assessed using the 18-item, child completed Activity support questionnaire which provides information on parental logistic support (that is, transport to activity venues, facilitating participation in organised activities), parental modelling of physical activity behaviours and parental support for reducing screen-viewing [33, 34]. Perceived parental modelling for consuming fruit and vegetables will be assessed using a 15-item scale [35].

All questionnaires will be combined into one document and administered in the classroom (self-completed by the school children) with a fieldworker and teacher present to answer any queries and to assist the children with reading and writing as necessary. The fieldworkers and teachers will be instructed only to help with reading and spelling specific words and not to suggest answers to any questions.

Resource use related to the administration of the intervention in the form of time, travel and materials will be collected through trainer and teacher time and resource logs completed over the intervention period. A parental completed postal questionnaire will be administered at the end of the intervention. The parents will be asked about their time spent in helping children with relevant homework, and taking children to and attending out of school activities, as well as any cost involved. They will also be asked about health service use in relation to exercise related injury in their child. To facilitate the completion of this questionnaire resource use logs, in which they can record this information, will be given to parents at the beginning of the year in which the intervention takes place. In order to compare costs between intervention and control schools the same questionnaire will be sent to parents at all schools.

Audits and process evaluations will be undertaken throughout the study to ascertain whether the intervention was delivered as planned and to identify any issues that might impact dissemination.

Audits will be undertaken at all schools, in both the intervention and control arms, to assess current physical activity provision (number of scheduled physical education (PE) lessons per week and time allocation for unsupervised outdoor play), school physical activity and nutrition polices (active travel, break-time play, packed lunch policy, etc) and number and type of after school clubs provided. Audits will be conducted once per term in all schools during the intervention and follow-up periods of the study (September 2011-July 2013).

We will ask all Year 5 teachers in intervention schools to complete a log of session delivery. The log will ask teachers to report the date that the sessions were delivered, the dates that homeworks were set and returned, and any problems or positive comments about the lesson.

Process evaluation will be conducted using focus groups with children, face-to-face interviews with teachers and school administrators, and telephone interviews with parents. Focus groups will be conducted at the end of the intervention in six purposefully selected intervention schools and six purposefully selected control schools with 6-8 purposefully selected pupils in each school. Schools will be selected to ensure representation of schools from more and less deprived areas and those with few and many programmes or initiatives (in addition to the study intervention) to promote healthy behaviours in children. Pupils will be selected to ensure both girls and boys are represented, as well as those who express strong and weaker self-efficacy and parental support at baseline assessment. The topic guides will be largely identical for both children in intervention and control groups. The focus groups will first examine the children's knowledge, attitudes and behaviours in relation to physical activity, sedentary behaviour and diet. They will then examine whether the children remember any lessons or homeworks that they have received during year 5 that relate to physical activity, sedentary behaviour and diet. If the children in intervention schools spontaneously mention intervention lessons and homeworks then the interviewer will probe about whether they liked or disliked these lessons (and ask for reasons) and what they remember learning from these lessons. Similar probing will take place with children in control schools where they spontaneously mention specific lessons or homeworks related to diet, exercise or sedentary behaviour topics. Children will be encouraged to say what they liked and disliked about these activity/diet related lessons and ideas they have about improving how these subjects could be taught. In any focus group (intervention or control school) where no specific lessons or homeworks related to physical activity, sedentary behaviour and diet are mentioned the pupils will be asked about whether they can describe any specific lessons/homeworks. In intervention schools only, the interviewer will describe some of the intervention lessons and homeworks to see if that prompts the children's memory of these if none have been spontaneously mentioned as the group appears to be coming to an end. If the children acknowledge some memory of these lessons/homeworks, they will then be prompted further about what they remember and what they liked or disliked about them. Thus, the focus groups will examine children's attitudes towards being taught these subjects and the methods of teaching them and will also assess whether intervention school children remember anything specific about the intervention and whether control school children received similar lessons.

All classroom teachers who delivered the intervention and head teachers (or their deputies) from all of the 25 intervention schools will be invited to participate in face-to-face, in-depth, semi-structured interviews that will assess aspects of the intervention that worked well, or that could be improved, with a focus on any elements that would improve delivery within the school. Finally, phone interviews will be conducted with a sample of 30 parents from 12 different schools (6 intervention and 6 control) to assess the parents' perspective of teaching their children about healthy levels of physical activity, sedentary behaviour and diet in school. The attitudes of parents from intervention and control schools towards parent involvement with homeworks will be explored, initially with no prompts (so that as with children we can explore whether parents remember the specific intervention homeworks and whether parents from control schools report any homeworks that appear similar to those from the intervention school). If parents from the intervention schools do not spontaneously mention any of the intervention homeworks they will be asked about specific homeworks. If they report remembering these, they will be asked about their attitudes towards the parent-child active homeworks that were part of the intervention.