Overarching research design
The overarching study design is a parallel-group superiority RCT with integrated process and health economic evaluation [21] designed in line with MRC Guidance principles [20]. The RCT follows the Standard Protocol Items: Recommendations for Interventional Trials statement [27] (Additional file 1). The process evaluation will follow a concurrent design with data collection and analyses being completed during similar time frames [28] (Fig. 2). The trial participants will be 18–50 years old and will have radiographically verified acetabular dysplasia and have experienced hip pain for a minimum of 3 months [21]. The participants will be dichotomised into either eligible but unwilling to undergo a PAO or not eligible for the PAO [21]. Further information on the study setting, eligibility criteria, sample size, etc., is reported in the RCT protocol [21]. In brief, to obtain sufficient statistical power, the RCT will recruit a minimum of 85 participants to the intervention group and 85 to the control group (usual care). Usual care includes one oral consultation provided by the first author on self-management of hip symptoms, including advice about staying physically active and exercising and, if relevant, advice to lose weight [21]. All intervention group participants will be included in the generation of process data. The RCT aims to investigate the effectiveness and cost-effectiveness of exercise and patient education compared with usual care [21].
Flowchart of the MovetheHip trial proceedings and process evaluation. Abbreviations: SDT, self-determination theory; MI, motivational interviewing; HAGOS, Copenhagen Hip and Groin Outcome Score; iHOT-12, Short version of the International Hip Outcome Tool; PAO, periacetabular osteotomy
Implementation
We will conduct a multi-component examination of the implementation of the intervention, including process, fidelity, dose and reach [21]. We will examine the implementation process by describing the required structures and resources, and we will measure fidelity as the degree to which the intervention is delivered as intended. Dose will be measured as the number of completed supervised and home-based exercise sessions. High dose is defined as completing a minimum of 75% of scheduled training sessions, medium as completing 50–74% and low as completing less than 50% [21]. Additionally, we will measure the dose using the exercise adherence rating scale (EARS) [29, 30]. Reach will be examined by describing baseline participant characteristics in those receiving a high compared versus a low dose, and by comparing the age and sex of participants with those of non-participants. In an additional explorative analysis, we will examine if the dose is associated with changes in health outcomes.
Acceptability of the intervention
We will explore acceptability by examining the extent to which the participants, intervention providers and healthcare managers consider the intervention to be appropriate based on their anticipated (prospective acceptability) or experienced responses to the intervention and received training (concurrent and retrospective acceptability) [31]. In an additional explorative analysis, we will explore if the reason for deselecting surgery (i.e. dichotomised into surgeon’s decision (not a surgical candidate) and participant’s decision (unwilling to undergo surgery)) moderates dose.
Mechanisms of change
Mechanisms of change will be assessed by analysing how participants interact with the intervention activities to facilitate change in health outcomes. We will examine if autonomy, competence and relatedness relate to motivation and change in health outcomes and study the influence of contextual factors [22, 23].
Contextual factors
Contextual factors include events and a socioecological view of personal, interpersonal and organisational factors interacting with the implementation, acceptability and mechanisms of change. By adopting a socioecological perspective, we will consider the implication of individual attitudes and behaviour and the quality of bonds to the intervention providers and support from family and friends. Furthermore, we will consider the impact of organisational support to gain additional knowledge on how processes may be optimised to facilitate any up-scaling of the intervention.
Data sources
The following sections outline data collection methods and sources relevant to the process evaluation. The timing and relation to the research questions are described in Table 1. Trial procedures are described in Fig. 3. The trial protocol paper details all wider measures obtained as part of the effectiveness trial [21].
Schedule of procedures for the MovetheHip randomised controlled trial (SPIRIT figure)
Participant data measured
Orthopaedic surgeons and research assistants will measure and register baseline participant characteristics as reported in the trial protocol [21]. The baseline data will be used to describe intervention reach.
Online participant survey
Participants will enter baseline data using a survey option in a Research Electronic Data Capture (REDCap) database. Baseline process data include sex, age, duration of hip symptoms, educational level, employment status, cohabiting status and level of physical activity [21]. These data will be used to examine intervention reach as aforementioned.
Online participant-reported health outcome measures (health outcomes)
Self-reported pain, physical functioning in sports and recreation and quality of life will be measured using the Copenhagen Hip and Groin Outcome Score (HAGOS) [32]. Self-reported hip-related quality of life will be measured using the Short version of the International Hip Outcome Tool (iHOT-12) [33]. These health outcomes will be used to examine the underlying mechanisms of change by exploring how any changes in health outcomes are related to satisfaction of psychosocial needs and motivation.
Participant training records
Intervention dose is the number of completed training sessions (supervised and home-based) and the more comprehensive measure of doses received using the EARS [21]. The participants will prospectively register the number of completed training sessions using a weekly logbook and register with the EARS [29] the extent to which the four exercises described in the exercise programme are completed. The dose data will serve to examine intervention reach as aforementioned.
Online fidelity survey
Intervention providers will use a 100-mm visual analogue scale (VAS) in a REDCap survey form to measure their ability to deliver specific content of the intervention, ranging from not possible (0) to always possible (100). The intervention components describe the ability to deliver the following: (1) use the Borg CR10 to determine the difficulty level and repetitions of exercises; (2) use the participant’s expressions of exercise acceptability to determine difficulty level and repetition; (3) use the intervention manual to determine correct exercise performance; (4) council participants about pain mechanisms in acetabular dysplasia, give advice on physical activity, monitor weight loss (if relevant) and deliver support to increase exercise adherence. The data will be used to examine intervention tailoring as an integral aspect of the implementation process.
Semi-structured interviews with participants
Participants will be invited to participate in semi-structured one-to-one interviews by purposeful sampling at baseline and at the 6-month follow-up. Sampling will take into account factors such as gender, age, employment status and reason for not undergoing surgery. Data saturation will be checked prospectively, and further data collection will be undertaken if needed. The mode of the interviews will be either physical or via a video connection according to the participant’s preference. Baseline interviews will focus on previous experiences and expectations to changes and behaviour (degree of external, integrated or internal motivation) and the influence of personal, interpersonal and organisational contextual factors. The 6-month follow-up interviews will focus on experiences with the intervention, its acceptability and the participant’s behaviour (degree of external, integrated or internal motivation), as well as the influence of personal, interpersonal or organisational contextual factors. A semi-structured interview guide is developed and key questions are included in Table 2. As mentioned above, the interview data will be used to examine the underlying mechanisms of change.
Semi-structured focus group interviews with the intervention providers and the expert team
Regular evaluation meetings designed as online focus group interviews with the intervention providers and the expert team will be conducted in the study period [9, 21]. A semi-structured interview guide has been developed and key questions are included in Table 3. Focus group interviews will focus on the quality of delivery, comprising challenges, tailoring and experiences with the MovetheHip intervention. Data from these interviews will be used to examine how the intervention providers tailor the intervention to the individual participant, considering their different exercise behaviours and physical and mental functioning as part of the implementation examination. In addition, data will reflect a consensus about tailoring for the individual participant and will show how well the intervention providers feel prepared to deliver the intervention.
Semi-structured focus group interview with healthcare managers and practice consultants
We will conduct an online semi-structured focus group interview with 4–6 key healthcare managers and practice consultants within the field of physiotherapy. The managers were public and private physical therapy clinic managers and lead managers in local municipalities responsible for health and rehabilitation resources. Furthermore, we will invite practice consultants within physiotherapy from the Central Denmark Region. A semi-structured interview guide has been developed and key questions are included in Table 4.
This focus group interview will explore intervention acceptability regarding how healthcare managers and consultants decide if an intervention is relevant and beneficial in their organisation; specifically, what information or evidence they need, and how they prioritise interventions and provide the necessary financial resources. Furthermore, the interview will explore their experiences with inter-organisational behaviour change. The data from the interviews with the healthcare managers may add to our understanding of what type of information healthcare managers consider essential. This understanding is important should they consider supporting a forthcoming implementation of this intervention, provided it proves to be effective and cost-effective.
Quantitative data analysis
Descriptive statistics will be used on implementation data; normally distributed continuous data will be reported as means with standard deviations and categorical data will be reported as numbers and proportions. Mechanisms of change will be examined by linking intervention functioning (qualitative data) to mean changes in health outcomes. Specifically, mean changes in health outcomes will be calculated in normally distributed data using descriptive statistics. Furthermore, we will explore the possibility of dichotomising data on intervention dose into low versus high. Reach will be examined by describing baseline characteristics within each dose group using descriptive statistics.
Additional explorative analyses will be performed. To examine the implementation, we will explore if the dose is associated with changes in health outcomes, using simple linear regression models with doses as independent variables and changes in health outcomes as dependent variables. Furthermore, in a repeated measurement analysis using a mixed-effects model, an explorative analysis will be undertaken to determine if the reason for deselecting surgery (i.e. dichotomised into the surgeon’s or participant’s decision) moderates dose changes as part of the examination of acceptability. In the model, participants will be the random effects with a fixed factor for group and time and the corresponding interaction (group × time), adjusted for baseline values. Statistical significance is considered to have been achieved at p < 0.05, and the Stata 17 (StataCorp, College Station, TX, USA) software package is used for the data analyses.
Qualitative data analysis
All interviews will be recorded and transcribed verbatim. A theoretical thematic analysis will be conducted [34]. Blinded to the findings of the RCT, the last author (DS) will index a subset of the data and construct a coding framework for each dataset (i.e. individual interviews and focus group interviews). A priori codes covering the process evaluation domains and concepts from the self-determination theory and motivational interviewing will be included in the coding framework, and these domains and concepts will be supplemented with new codes emerging from the data. The remaining data will be analysed according to the analytic framework. The last author will code the first interviews of each dataset and refine each coding framework. Subsequently, the first author (JSJ) will code the remaining data according to the coding framework. The last author will then verify all coded data and refine them if needed. Once all data have been coded, the last author will identify relevant themes within each dataset. Themes from across all datasets will be compared and refined to agree on a final set of study-level themes. These themes will be accompanied by anonymised quotes, again collected by the last author.
Data integration
Data integration will be achieved through data merging once the qualitative and quantitative data have been collected and analysed separately [28]. The analysis focuses on the following three process evaluation domains: the degree to which the intervention is implemented, the acceptability, and how the intervention functions (mechanisms of change). Furthermore, the analysis will explore how contextual factors relate to these three process evaluation domains. Integration at the interpretation and reporting level will follow the weaving approach, involving writing the qualitative and quantitative findings together on a theme-by-theme basis according to the process evaluation domains and research questions [28]. The findings made during data integration will show if the quantitative and qualitative findings confirm, expand or are in discordance with each other [28].
Add Comment