Partnering for change (P4C) in Sweden- a study protocol of a collaborative school-based service delivery model to create inclusive learning environments

A reference group, comprised of representatives from organisations at different levels of government in Sweden, such as the Swedish Association of Local Authorities and Regions, and the National Agency for Special Needs Education and interest organisations will be constituted. The reference group will be involved in discussions concerning methodological issues, interpretation, and disseminating the results. The reference group will meet on a regular basis throughout the study period.

The primary outcome is children’s engagement with learning, measured by ‘Engagement Versus Disaffection with Learning: Teacher Report’ (EEL), focusing on children’s engagement and disaffection with learning in school [33]. The EEL consists of four subscales across 25 items, rated on a four-point Likert scale, ranging from 1 (happens almost never) to 4 (happens very often), and is rated by teachers. The subscale behavioural engagement (5 items) focuses on effort, attention, and persistence in initiating and participating in learning activities. The sub-scale of behavioural disaffection (5 items) addresses a lack of effort and withdrawal from learning. The third sub-scale—emotional engagement (5 items) —focuses on emotions indicating motivated involvement during learning activities; and emotional disaffection (9 items) addresses emotions indicating withdrawal during learning. The EEL has been shown to have good convergent validity and good test–retest reliability, and to differentiate between behavioural and emotional school engagement [4, 34]. Timeframe: baseline, four months, eleven months (both intervention and control classes).

Secondary outcomes include data on children’s health-related quality of life (HRQOL) and wellbeing, children’s occupational performance, attainment of school-based occupational performance goals, school attendance, and student-environment fit. Also, teacher and OT knowledge and skills, as well as documentation of services provided including data on special educational needs will be collected. Additionally, focus group interviews will be performed with school staff and children after P4C.

Children’s HRQOL will be measured using Swedish Child Health Utility 9D (CHU9D), a generic preference-based HRQOL measure constructed for use in children and youth. CHU9D has nine questions representing nine dimensions of HRQOL: worried, sad, pain, tired, annoyed, schoolwork/homework, sleep, daily routine, and activities [35]. Each question has five response levels representing increasing degrees of severity within each dimension and is self-completed by the child. The CHU9D has shown to be practical to use, and has acceptable face and convergent validity and internal consistency [36‐39]. The CHU9D has been validated in a Swedish context and has been shown to be a feasible, reliable, and valid measures of preference based HRQOL [40]. Timeframe: baseline, four months, eleven months (both intervention and control classes).

Children’s well-being in school-related situations will be measured using How I Feel About My School (HIFAMS) [41]. HIFAMS addresses well-being in seven school-related situations: on the way to school, in the classroom, when doing work, at the playground, thinking about other children, thinking about their teacher, and about school in general. The HIFAMS uses a three-point Likert scale, where emotions are used to convey the responses sad (0), OK (1), and happy (2), and is self-completed by the child. The total score is calculated as the sum of the individual items (maximum total score:14), with higher scores indicating greater happiness. HIFAMS is suitable for children aged 4 to 12 years. It has shown moderate internal consistency and test–retest reliability [41, 42]. In a Swedish sample, the HIFAMS showed moderate and satisfactory internal consistency [42]. Timeframe: baseline, four months, eleven months (for both intervention and control classes).

Occupational performance in school-based activities will be observed by OTs and documented in a digital questionnaire created especially for the study and distributed using Survey&Report. The digital questionnaire is based on the Dynamic Performance Analysis (DPA) [31] and uses a rating scale (1–10), being 1 = no performance 10 = excellent performance. Attainment of occupational performance goals will be rated as attained, partly attained, or not attained before and after implemented interventions for each identified performance breakdown using the digital questionnaire. Time frame: continuously during the four-month intervention period (intervention classes).

Student-environment fit in school setting will be examined only before and after individual-level interventions using the School Setting Interview (SSI) [43]. The SSI includes 16 items concerning everyday school activities, and the children’s perceived need for adjustments before and after implemented interventions. Each item is scored using a four-step rating scale; 4 = perfect fit, 3 = good fit, 2 ≤ partial fit, and 1 = unfit. The SSI has shown usability and construct validity [44]. Time frame: applied continuously during the four-month intervention period (intervention classes).

All delivered interventions will be documented by OTs and teachers in a digital questionnaire created and distributed using Survey&Report. The type and level of interventions (class, group, individual), as well as time needed to deliver P4C each week will be documented. The questionnaire is developed for the project and has been previously used by OTs and teachers in Swedish schools. Time frame: weekly during the four-month intervention period (intervention classes).

Teachers’ knowledge, skills, and experience with P4C will be measured using the Educator Knowledge, Skill and Experience Questionnaire [27]. The questionnaire uses a five-point Likert scale from strongly disagree to strongly agree. The questionnaire has been translated into Swedish following a forward–backward translation process and previously used in Swedish elementary schools by teachers. Time frame: baseline, four months, and eleven months (both intervention and control classes).

OTs knowledge, skills, and experience of P4C will be measured using the Partnering for Change Occupational Therapist Questionnaire: Exploring Occupational Therapist Skills, Knowledge and Beliefs [27]. The questionnaire uses a five-point Likertscale from strongly disagree to strongly agree. The questionnaire has been translated into Swedish following a forward–backward translation process and has been complemented with questions relevant for the Swedish school system including questions on children with neurodevelopmental disorders. The questionnaire has been previously used in Swedish elementary schools by OTs. Time frame: baseline, four months, eleven months (for both intervention and control classes).

Additionally, after the intervention study, focus group interviews will be conducted with school staff (OT, teacher, special education teacher and/or principal) (one focus group/school) to explore the experience of working according to P4C, and if any facilitators and barriers for implementation were identified. The interviews will be digitally recorded and transcribed verbatim.

To describe children’s experiences of inclusive learning environments, focus group interviews will be conducted after P4C with six to eight children from intervention classes. The interviews will be digitally recorded and transcribed verbatim.

Descriptive statistics will be presented for demographic characteristics and outcome variables. To evaluate any potential changes following the P4C, the difference-in-differences method using regression modelling will be applied [45]. This method compares the changes in outcomes over time between the intervention and control group and allows for adjustment for confounders. Furthermore, the multilevel structure of the design of the study will be considered. If the proportion of missing data is between 5% and 40%, multiple imputation will be performed [46]. Estimates will be reported with 95% confidence intervals and associated p-values. A two-tailed p-value of < 0.05 will be considered statistically significant.

Cost-effectiveness analyses will be conducted as a within-trial economic evaluation comparing the health outcomes, school attendance, and societal costs between the intervention and control classes. We plan to conduct a cost-utility analysis (CUA) and cost-effectiveness analysis (CEA), which focus on the health outcomes produced, while the total society costs are related to health gains and the result is expressed as an incremental cost effectiveness ratio (ICER). The primary health outcome in economic evaluation is defined as quality adjusted life years (QALYs) gained and will be estimated using any change in utilities, measured by CHU9D instrument at baseline and respective data collection points. The change in QALYs will be estimated by multiplying the change in estimated utilities with the intervention period expressed in years. The reduction in absenteeism rate will be used as a secondary health outcome and collected during the follow-up period as well as the consumption of additional support and healthcare at school. Costs to deliver P4C will be estimated based on professionals’ time needed to deliver the intervention at school, costs of materials, and other related costs. Other societal costs such as additional support materials/equipment at school will be estimated on the group level. The total societal costs will be estimated for intervention and control classes. The cost–utility ratio will estimate the price for one additional QALY—i.e., one life year with full health gained—while the cost–effectiveness ratio will calculate the price for an avoided case of absenteeism comparing intervention and control classes. The total QALY-gain for the children during the follow-up period will be estimated using the ‘area under the curve’ method [47]. QALYs will be analysed with general regression model (GLM) [48], which allows for the control of confounders and adjustments to baseline. The costs and potential savings will be related to health gains and reduction on absenteeism, and then presented as cost–effectiveness ratios on a cost–effectiveness plane, together with uncertainty for cost- and effect data. Probability for P4C to be cost-effective compared with TAU will be presented as a range of ‘willingness-to-pay’, representing the intentions of decision-makers to pay for an additional QALY or for reduction of certain percent in absenteeism, using a ‘cost–effectiveness acceptability curve’ (CEA) [49], and bootstrap regression analysis.

The focus groups will be digitally recorded, transcribed verbatim, and analysed using content analysis [50]. The transcripts will be read through several times to get a sense of the data. To capture key thoughts or concepts, codes will be derived from the transcripts, comprising the initial coding scheme. The codes will then be organised into categories and themes highlighting the respondent’s experience.