Challenges and possibilities when doing research on active school travel interventions in a school setting – a non-randomized pilot study assessing feasibility of an evaluation design | BMC Public Health
Study design
In accordance with the conceptual framework of feasibility studies [34], a non-randomized controlled pilot study was carried out, applying a combination of quantitative and qualitative methods. The study evaluated feasibility outcomes in relation to the evaluation design and methods for data collection in preparation for future evaluations of AST interventions. The Swedish Ethical Review Authority approved the study (no. 2021–03783).
Setting
The study was carried out during February–March 2022 in Falun, Sweden. Falun has a geographical area of 2275 km2 [35], and about 60,000 people live in the municipality, of whom about 39,500 live in the urban area [36]. The municipality has 25 elementary schools [37].
Falun is located in the province of Dalarna, in central Sweden, and experiences winter road conditions with snow and ice (December–April). During the beginning of the study period, the third wave of the COVID-19 pandemic swept over Sweden. This wave had implications for the study, which had to be conducted mostly through virtual interaction between the researchers and the children, parents, and teachers involved. However, focus group interviews succeeding the SICTA intervention were not affected by COVID-19-restrictions and therefore conducted onsite.
Participants and recruitment
Children in fifth grade (age 11–12 years) were the target population. In collaboration with Falun municipality (the Child and Education Administration as well as the Sustainability Strategist), invitations to participate in the study were distributed to principals in all 25 public schools in Falun. Three principals expressed interest for their schools to participate, two with explicit requests to participate as intervention schools and one with explicit request to participate as a control school. After a second request from the Child and Education Administration, a fourth principal agreed for their school to participate in the study, as a control school. The four principals were further informed about the study and contacted their personnel. Neither participating schools nor children received any payment for their participation.
Description of the SICTA intervention
The SICTA intervention was delivered in the classroom setting for 4 weeks, consisting of weekly assignments to promote AST. The assignments are related to the national curriculum, and teachers together with the children decided on relevant SICTA assignments for children to solve on the way to/from school, guided by an online platform (the teachers’ portal) [22].
For example, one assignment concerns parental fears and traffic safety when cycling or walking to school. The assignment is inspired by Astrid Lindgren’s book “Ronja Rövardotter” (Ronia, the Robber’s Daughter), which follows Ronja as she explores the forest and learns about its dangers. The book is a classic in Swedish literature, known for its themes of independence and bravery. The task addresses parents’ concerns about their children walking or cycling to school and provides an opportunity to discuss traffic safety.
In the first step, the children are asked to identify dangers on their way to school and document these dangers through photos or notes. Following this, the children read an excerpt from “Ronja Rövardotter” in the classroom where Ronja’s father warns her about the forest’s dangers. This is followed by discussions of the text in small groups.
Next, the children complete a writing assignment where they write a text mimicking Lindgren’s style about a child cycling or walking to school for the first time and their parents’ warnings. The children consider what dangers they might face and how to overcome them. They can also do sub-assignments to reflect on and discuss the dangers with their parents or interview an adult about their fears regarding the school route.
Lastly, the assignment is followed up by sharing and discussing the texts in small groups, focusing on common fears and traffic safety. The online platform includes several assignments for different subjects that the teacher could choose from, moreover, the teachers could also choose to develop their own assignments with inspiration from the existing ones.
The SICTA intervention consists of five steps: (1) Caregivers are informed during a parental meeting. (2) Children are involved and given information about the benefits of AST. (3) To promote AST, teachers and children decide on suitable assignments together. (4) Weekly assignments are performed during a 4-week period, while AST is measured and reported. (5) The SICTA intervention ends with an evaluation and a celebration.
Control schools
The control schools did not receive any AST intervention (passive control).
Outcome measures and data collection procedures
The outcome measures and data collection procedures are presented below in three parts. The first section describes the measures and procedures intended for describing the characteristics of the study sample. The second section describes the measures and procedures intended for evaluation of the SICTA intervention, and the third section details the targeted measures and procedures for the assessment of this study’s feasibility objectives.
Measures to describe characteristics of the study sample
Health-related quality of life (HRQoL)
To describe the population, HRQoL was self-reported using a child friendly EQ-5D version (EQ-5D-Y) [38, 39]. The EQ-5D-Y can be summarized using an index value (1–3), where index 1 indicates very good HRQoL [38, 39]. Data collection took place one week before the intervention/control phase.
Outcome measures for evaluation of the SICTA intervention
AST
In this study, AST was defined as actively traveling to school for the whole or part of the distance by walking, cycling, or school bus (children traveling by school bus also travel part of the distance actively [40, 41]). The wASTapp was used to report daily travel mode, commuting time, and distance traveled and was accessible from computers and smartphones. The children logged in via their digital school account using a personal “research ID.” As an option, a web link to the wASTapp was distributed daily via SMS to the child’s (or parents’) smartphone.
A paper report was also available for teachers to distribute to children if needed, as an analogue option for daily reports.
School travel was registered for each direction to and from school, with a maximum of 10 reported trips each week. Additionally, in two schools (one control and one intervention school) wASTapp reports were complemented with accelerometers (control school), and accelerometers together with GPS trackers (intervention school). Data collection included daily reports to and from school, one week before and one week after the intervention phase.
Mediators of AST
Three self-report questionnaires were used to collect data on social mediators of AST: Behavioural Regulation in Active Commuting to and from School questionnaire in Sweden (BRACS-SWE) [22, 42], Perceived Autonomy Support Scale for Active Commuting to and from School (PASS) [43], and Basic Psychological Need Satisfaction in Active Commuting to and from School (BPNS-ACS (SWE)) [44]. Furthermore, to collect data on parental attitudes about AST, caregivers completed the Parents Intentions to Let their Children use AST (PILCAST-52) questionnaire [45] via the web (data not reported in this study). Data were collected one week before and one week after the intervention phase.
Measures and procedures to assess feasibility of the evaluation design
Participation and response rates
Objective measures used as a quantitative indicator to describe the feasibility of the study design and methods for data collection were based on following data sources: recruitment of schools, number of schools and children accepting participation in the study, reporting rates of the daily school travels in the wASTapp, response rates of the questionnaires (BRACS-SWE, PASS, BPNS-ACS (SWE), and EQ-5D-Y), and adherence rates for wearing measuring devices (accelerometer and GPS tracker).
Feasibility questionnaire
The research group created a 10-item questionnaire (“the feasibility questionnaire”) to assess indicators of feasibility [34] based on children’s opinions of the research evaluation process and methods for data collection. A four-point Likert scale was used in which 1 indicated a very positive and 4 a very negative response. Free-text comments in the feasibility questionnaire for additional clarification were optional. The feasibility questionnaire was distributed together with the abovementioned other three questionnaires one week after the intervention phase.
Focus groups
Semi-structured focus group interviews [46] were performed to explore the children’s experiences with the research and methods for data collection and to gain a richer understanding of aspects of feasibility from their perspective. All children who participated in the study at one intervention school (where all methods for data collection were included) were invited to participate in the focus groups (Figs. 1 and 2). The focus groups were conducted according to a semi-structured interview guide based on the feasibility questionnaire. One of the authors (MA) moderated all focus groups with the children. To ensure dependability, the interview guide was followed carefully, and probes were used to facilitate discussions. The focus groups were carried out after school in the classroom, and the children could sign up to one of three different occasions, two or three weeks after the SICTA intervention.
Study procedure and data analysis
Study procedure
Information about the study was given to teachers, children, and caregivers using video conference. Teachers were oriented first, followed by the children. During the introduction, the children were informed about the study and asked to obtain signed informed consent from their caregivers to participate. Children were informed that they were part of a research project to test research methods related to assessing school travels. Additionally, children in ools were informed about the SICTA intervention.
Caregivers were informed in a parents’ evening presentation as well as by printed study information, with recorded information videos and text made available on the study website. In case of two caregivers, both were asked to sign informed consent for their child. For all families, it was emphasized that participation was voluntary and could be discontinued at any point without giving reasons. For ethical reasons, all fifth-grade children from the participating schools could fill in the questionnaires and use the daily wASTapp, but data were collected only for children with a signed informed consent obtained from a parent and/or legal guardian. For the same reason and to avoid non-equal situations, all children in the intervention schools participated in the previously described SICTA-intervention.
After the introduction to the study process, individual coded folders for each child were provided to the teachers, who distributed the materials to all children one week before the intervention. The folders included questionnaires and a “junior researcher business card” (with an individual ID number, the study website address, and password to the wASTapp). Via video conference, the researchers (MA and AP) gave instructions to all fifth-grade children and their teachers about how to use the password for login and how to register travel mode, commute time, and distance (using Google Maps™) in the wASTapp.
General information about the study, instructions related to wASTapp, and how to measure time and distance were also presented on the study website ( in the form of short instructional videos, as well as texts.
Instructions on use of accelerometers and GPS trackers were provided, where these were distributed. Following the SICTA intervention, the feasibility questionnaire and focus group interviews were added to assess aspects of feasibility. The timeline of the study procedure is further detailed in Fig. 1.
Timeline and content of the study procedure
Data analysis
Participation and response rates
Quantitative data were analyzed using the Statistical Package Software for the Social Sciences (SPSS version 28.0.1.0, Chicago, IL, USA). Descriptive data are presented as median and min–max or interquartile range (IQR) for ordinal and continuous variables or as number (n) and percentage (%) for categorical variables. For comparison between groups on ordinal variables, the Mann–Whitney U test [47] was used. In the analyses, a p-value < 0.05 was considered significant.
Feasibility questionnaire
The ratings from the feasibility questionnaire were dichotomized by answer choices that were positive (1: Very fun/ Very well/Very easy; 2: Fun/Well/Easy) and negative (3: Not fun/Not well/Not easy; 4: Boring/Not at all well/Not at all easy). Ratings were reported as proportions of positive and negative responses.
Focus groups
Qualitative data were analyzed using inductive manifest content analysis inspired by Graneheim and Lundman for content analysis [48]. MA transcribed the interviews. The transcriptions were then read separately by the two authors MA and AP, one being a junior researcher (MA) and the other a senior researcher with many years of experience in qualitative research (AP). The transcribed interviews were read several times to obtain a sense of the overall data, identified meaning units, and individually assigned codes to the meaning units, as suggested by Graneheim and Lundman [48–49]. The two authors then compared their coding for discrepancies, discussing them several times until consensus was reached [48]. The codes were then collated into preliminary categories by MA, with the aim of staying close to the text [49]. A third author (SR), a senior researcher with many years’ experiences in qualitative research, was involved to discuss and validate the preliminary categories, and after discussions among the three authors, the preliminary categories were collated into final categories. Quotes were used to strengthen the credibility of the analysis [48–49].
link
