Shoulder pain is common in wheelchair users living with spinal cord injury (SCI) [1,2]. A lesion to the spinal cord can result in loss of innervation to muscles of the trunk and lower limbs. Consequently, many people living with SCI (pwSCI) rely on their upper extremities not only for performance of daily activities but also for locomotion. Shoulder pain can have a significant impact on their activity, reducing mobility, independence, and quality of life [1-5]. Digital and web-based interventions have increasingly been offered to pwSCI to promote exercise and physical activity [6-9]. These interventions minimize barriers to rehabilitation to address many health concerns, including managing their shoulder pain. Previous authors have found that in the general population, digital or web-based interventions can produce positive effects in various outcomes, such as physical activity [10,11].

Technology-supported exercise interventions for pwSCI with persistent shoulder pain are currently available, but they have some limitations. They either require ongoing input and monitoring from a clinician [12-15] or provide general self-management advice [13] but without enough guidance to allow for clear and structured exercise progression specifically for shoulder pain. Self-guided digital exercise interventions have been successfully implemented for people with knee osteoarthritis [16,17], dizziness [18,19], and breast cancer [20] and may be a viable option for pwSCI. Our recent systematic review and meta-analysis of self-guided digital physical activity and exercise interventions demonstrated positive effects on physical activity at both short- and longer-term follow-up, in people living with chronic conditions [21]. We also found that interventions that used behavioral strategies and were underpinned by a theoretical framework were more effective. This suggests that self-guided digital interventions have the potential to support pwSCI to manage their shoulder pain, but that the intervention would need to be designed systematically and intentionally.

We have designed Shoulder Pain Intervention delivered over the interNet (SPIN) as a self-guided digital intervention to give pwSCI who experience shoulder pain the ability to access and progress evidence-based exercises. The intervention guides pwSCI to monitor symptoms and improvement [22] to promote autonomy in the management of their condition. The aim of SPIN is to be an engaging program that is responsive to the needs of pwSCI who have shoulder pain.

To achieve this, we were guided by the person-based approach (PBA) in the design of SPIN [23]. The PBA follows 4 iterative phases of intervention development that include (1) planning which seeks a deep understanding of the perspectives and psychosocial context of potential users through iterative qualitative research, (2) design based on guiding principles that have been created from insights from the first phase, (3) development and refinements which are made through iterative user feedback, and (4) trialing to evaluate the effectiveness on outcomes and impact on behavior change to make any necessary adjustments. Due to its focus on the development of digital behavior change interventions, the intent and purpose of PBA align well with adjacent behavior change theory and tools such as the COM-B [24], “Behaviour Change Wheel” [25], and behavioral analysis [26]. Furthermore, the PBA process is sufficiently flexible to enable the use of these (and other) tools to achieve the aims and purpose of a given phase. Integrating behavioral science theory and evidence while keeping users’ needs and contexts in focus has been found to maximize engagement and effectiveness of interventions [18,25,27-31]. This tutorial focuses on the first 2 PBA phases of planning and design. See Table 1 for an example of how our study was mapped onto the PBA.

We drew heavily on the existing PBA literature during the planning and designing stages of SPIN. However, the lack of access to detailed examples of how the PBA has been operationalized in practice made it challenging to translate the principles of this approach into reality. This is not a unique problem. Duncan and colleagues [32] noted that published work on the development of an intervention is frequently sparse because it is often included in the same publication as the reporting of a pilot or feasibility study.

The aim of this paper is to make clear how the principles of the PBA were operationalized in intervention design and the development of SPIN. We have illustrated our use of the PBA framework by outlining the detailed and explicit steps involved in the translation of the evidence, theory, and person-based recommendations into intervention design. In doing so, we have built on the existing methodological framework and enabled others to draw on this approach in future intervention design and development.

The planning and design phases of the PBA are described below, along with an overview of how they were operationalized in the design of SPIN. Figure 1 provides an overview of the SPIN design process and the components involved. Each step and its subsequent outcome have been described in detail in the sections that follow.

In the PBA, the guiding principles inform the intervention features by providing a framework for selecting and shaping features that directly support the specific design objectives, and to improve resonance, engagement, and acceptability of an intervention [23]. A range of evidence informed the selection of behavioral intervention features: (1) in our review, we identified a range of features used in digital interventions that have been associated with better health-related outcomes [27,34-36]; (2) we identified possible behavioral intervention design features from our Interpretive Descriptive qualitative study [33]; and (3) we identified behavioral “intervention functions” we were trying to achieve using a behavioral analysis as per Michie and colleagues’ framework [26]. We then mapped these to the most relevant intervention features. Figure 2 represents the layers of evidence that informed SPIN’s intervention features. We will describe each of these in detail below.

We believe that by explicating how we used the PBA in the development of SPIN, we can support others to use the PBA in the design of interventions. Table 10 provides a summary view of our process and includes some questions that we hope will prompt other researchers to consider how they might operationalize the use of PBA in their work. The table provides an overview of key phases of PBA and possible timelines (column 1) and examples from SPIN (column 2), including tools and methods we drew on as complementary to PBA and which we found useful in operationalizing the approach. In column 3, we have included our reflections on the benefits of our approach. The final column has questions that we hope will serve as prompts for researchers and designers when using this approach.

Ethics approval (Auckland University of Technology Ethics Committee-AUTEC 18/263) and participant consent were received for the earlier work [33] that informed this work.

This paper has described how we applied evidence, theory, and person-based approaches in the design of a self-guided digital intervention to help pwSCI manage their shoulder pain. We have detailed the processes of applying the PBA to the design of SPIN.

This builds on Yardley and colleagues’ [23] collection of work. The PBA emphasizes a detailed, qualitative understanding of users’ psychosocial contexts to inform intervention design. It adds value to user-centered design by addressing factors that influence behavior change, beyond just usability. The PBA complements theory- and evidence-based frameworks, such as the “Behaviour Change Wheel” [25] by tailoring interventions to the needs and preferences of specific populations. Despite growing evidence for the use of the PBA framework in intervention design [29-31,62,63], there is little available on its operationalization. To our knowledge, the detailed reporting of each step has not been available before.

In a recent systematic review on the effectiveness of self-guided digital exercise interventions, Stavric et al [21] found that interventions with theoretical underpinning had increased congruence with the intervention features leading to significant positive results. This is supported by findings from McEwan [64] who found that theory-based interventions resulted in more consistent significant improvements in physical activity. The pwSCI and shoulder pain who will use SPIN are likely to have minimal contact with a health care professional. Therefore, successful design required an understanding of how SPIN would meet their needs and how pwSCI would use it in daily life in a self-directed way. Engaging people and evidence in intervention design is supported by a range of researchers and designers [23,34,65]. Using a person-based approach, drawing on evidence from the people who will use the intervention, to derive the behavioral strategies has been shown to be effective in a variety of settings and methods of delivery [18,29,30,66,67].

Despite acknowledgment that interventions supported by theory and evidence maximize outcomes [68,69], there remains a paucity of full intervention description or design disclosure [70-73], making it challenging to explicate the link between theory and evidence and intervention features. A key tension we encountered was the limited availability of detailed examples of how the PBA had been operationalized in practice. This required us to make interpretive decisions when translating PBA principles into design elements, often without clear guidance. Additionally, balancing adherence to the PBA’s iterative, user-focused process with practical constraints such as time, resources, and access to participants posed challenges. These limitations were compounded by the fact that we were largely self-taught in the application of both the PBA and behavioral analysis frameworks.

Michie and colleagues [74] recognized the challenges and lack of clarity around the purported mechanisms by which digital interventions work during an international workshop on developing and evaluating digital interventions to promote behavior change in health. DiLiberto and colleagues [75] support the importance of “insider accounts” of intervention implementation and argue that the same transparent reporting practice should apply to intervention design. Of the 16 self-guided interventions included in our systematic review and meta-analysis conducted in the planning phase [21], only 6 provided any reference to methods used to plan, design, and develop them [19,76-81]. Of these, there was little supporting detail on how the design was carried out and none of the included studies reported exploring the behavioral needs of the users before designing the intervention. Future researchers might benefit from greater transparency and reporting of the design phase, including more practical examples of operationalizing person-based and behavioral approaches. Considerations for mitigating these challenges include allocating sufficient time and resources for user involvement beyond the planning stage, documenting key design decisions, and seeking opportunities for peer collaboration to support methodological alignment and confidence.

We have shown commitment to providing a robust and transparent process in the operationalization of the design phase of SPIN drawing on the PBA approach. This process included explicitly addressing the identified behavioral needs of the users and kept these central throughout the entire design process. The design of SPIN has demonstrated how we used evidence (from existing literature and from a previous Interpretive Descriptive qualitative study) and theory (from behavioral analysis, “Behaviour Change Wheel,” and COM-B) to enhance the person-based process. This explicit and thorough process of planning and designing SPIN has provided a blueprint for intervention development when using PBA. It also addresses many of the limitations in the reporting on the development processes for existing self-guided digital exercise and physical activity-related interventions.

Our operationalization of the PBA design phase reflects our interpretation of the PBA steps through available readings. We acknowledge there may be other perspectives and understandings. However, we believe that it is important to make our experiences visible to build on previous work and support future intervention design. Similarly, we relied on literature and online course instruction for support when we conducted the behavioral analysis using the COM-B. Being self-taught in both the PBA and behavioral analyses may mean that some aspects of our approach are not consistent with the original intent of these approaches. However, this is perhaps an artifact of the knowledge mobilization process, where the application of knowledge can change as knowledge changes hands. By offering transparency in our process, we hope that people can draw their own conclusions regarding the robustness of our approach. The design of SPIN did not include a logic model. Logic models typically include the main intervention components, how they relate to one another, which are meant to produce which effect, and include processes and expected outcomes. However, we did not believe a logic model would have been pragmatically useful as they assume causal relationships which may have restricted our thinking about solutions [82]. Our development process was underpinned by relationship building and community interaction, both of which are complex and require flexibility [83,84].

With the proposed intervention features selected, SPIN wireframes have been constructed. Wireframes are images or screenshots that show how screens of a website or app are structured and how content is arranged. These have provided a visual representation of the product and an opportunity to comment on content, features, and organization without getting distracted by aesthetics. Further participant consultation and design refinement have occurred. Frontend and backend software programming will occur at a later phase. Reporting of these stages will follow in a subsequent publication.

The design of SPIN has incorporated a deep understanding of the users’ needs and best available evidence by drawing on the PBA design process to maximize chances of engagement and outcomes. This paper has made visible the operationalization of each of the phases and can act as a blueprint to provide guidance to future researchers when using this approach.