Abstract
Objective
To develop and validate an evidence-based framework guiding therapists in arm-hand rehabilitation in individuals with cervical spinal cord injury, to enhance performance and activities of daily living.
Design
The framework was developed using a mixed-method approach: (1) item generation based on the UK Medical Research Council's guidelines for complex interventions; (2) four-round e-Delphi study with Likert scales and qualitative input; and (3) expert panel discussion.
Participants
The e-Delphi panel comprised 24 international rehabilitation professionals (16 occupational therapists and eight physiotherapists) with mostly ≥5 years of clinical and/or scientific experience in spinal cord injury rehabilitation. A separate expert panel included seven individuals with cervical spinal cord injury.
Main measures
e-Delphi consensus for each item was predefined as ≥70% agreement or inclusion across four rounds, with content analysis for qualitative input. Items were further discussed and validated during the expert panel and analyzed by content analysis.
Results
65 initially generated items were included in the e-Delphi; 36 were added from qualitative input; consensus was reached on 100 of the 101 items. The expert panel confirmed these items, added five, and reprioritized personal load capacity and motivation. The validated framework comprises three interrelated phases: (a) knowledge gathering—assessing personal needs, load capacity, and environment; (b) goal setting—collaboratively formulating person-centered, realistic goals; and (c) acting—a task- and goal-oriented training targeting meaningful activities and daily life integration. Motivation plays a key role across all phases.
Conclusions
This evidence-based framework provides guidance for therapists to deliver person-centered arm-hand rehabilitation tailored to patients’ needs.
Keywords
Introduction
Arm-hand impairments (e.g. reduced hand and arm use and limitations in fine motor skills) in individuals with cervical spinal cord injury (SCI) often limit independence and participation, 1 making functional improvement of the arm-hand a key priority for enhancing activities of daily living and quality of life. 2 In this study, arm-hand skill training and skilled performance are defined at the activity level of the International Classification of Functioning, Disability, and Health. 3 Repetitive motor training remains the gold standard for arm-hand rehabilitation, promoting activity-dependent plasticity in the brain and spinal cord.4,5 Systematic reviews provide evidence that motor training improves muscle strength and activities of daily living performance.1,6,7 However, variability in therapy content and dose across studies limits comparability and hinders clinical translation, as no specific approach has demonstrated clear superiority. 7
A recent systematic review found that combining skill training with strength and endurance exercises, particularly when task-oriented training components are integrated, most effectively improves arm-hand skill performance. 8 Evidence on the optimal training dose of this type of intervention is limited: Mateo et al. 9 found no link between training duration or repetitions and improvements in functional outcomes, while Bertels et al. 8 suggested a minimum of eight weeks of training based on effect size analysis based on limited studies. In clinical practice, therapists emphasize person-centered rehabilitation, focusing on the individual and their environment, while also considering the individual's load capacity. 10 Moreover, motivation is widely recognized as a key factor in sustaining patient engagement.10,11
Although research has identified various important aspects of rehabilitation for individuals with cervical SCI, the translation of these findings into clinical practice remains limited. Clinical practice continues to prioritize stretching, range of motion, strengthening, and analytical training over task-oriented approaches.12–14 Therapists, therefore, emphasize the need for structured, evidence-based, and person-centered guidelines. 10 A conceptual framework may facilitate translating research into clinical practice to support the implementation of arm-hand skill training in practice. 15 The aim of this study is to develop and validate an evidence-based framework guiding therapists in arm-hand skill training in individuals with cervical SCI. The ultimate goal of the training is to enhance arm-hand skilled performance and the ability to perform activities of daily living.
Method
The framework was developed in three successive stages: (1) item generation and development, (2) an e-Delphi study with international experts; and (3) an expert panel discussion.
Stage 1: Item generation and development
Framework development followed the UK Medical Research Council's guidelines for complex interventions, focusing on the development phase. 16 A theory-driven, evidence-based, and target population-centered approach was integrated. 17 This was operationalized through the evidence-based practice model to synthesize research evidence, clinical expertise, and patient preferences. 18 In addition, practice-based evidence from an observational study was incorporated to capture real-world intervention details. 19 Items were generated based on the findings of four previously conducted studies. An “item” refers to a specific individual component within the framework, such as personal needs or realistic goals. An overview of the evidence used, including their research design, population, and objectives, is provided in Table 1.
Overview of the used evidence.
The framework was developed through an iterative process. First, drawing on findings from the systematic review, multicenter longitudinal observational study, focus group discussions with therapists, and semi-structured interviews with individuals with cervical SCI, the framework was structured into distinct phases and further divided into items through discussions among the authors (BN, JY, and SA). Task-oriented training components were included in the framework if supported by at least two studies. Specifically, for the systematic review, this meant demonstrating a mean effect size (Cohen's d) of ≥0.50, and for the observational study, being used in more than 50% of cases. In Supplemental Table 4, the included items and their supporting sources are described. Examples of items include personal needs, realistic goals, skill training, and positive feedback. The draft framework was reviewed and refined by an internal expert group (n = 7, occupational therapists, physiotherapists, sports scientists, and neuroscientists) and international experts at conferences (n = 6, occupational therapists and physiotherapists).
Stage 2: e-Delphi study with international experts
Ethical approval was obtained from the University of Hasselt's Medical Ethics Committee (CME2024/056) for the e-Delphi study and expert panel discussion. All participants provided written informed consent prior to participation. The Delphi study followed the DELPHISTAR reporting guidelines for Delphi studies in health and social sciences. 20
The Delphi panel was composed of international experts, occupational therapists, and physiotherapists, with at least five years of clinical and/or scientific experience in the rehabilitation of individuals with cervical SCI. To ensure this level of expertise, purposive sampling was used to recruit 20 experts, approximately 10 per profession, via professional networks, Special Interest Groups, and snowball sampling. Therapists who participated in the previously conducted studies were excluded.10,12
A modified e-Delphi technique was employed to reach expert consensus, with participants completing an electronic survey questionnaire in each round. The electronic survey included three types of questions: 5-point Likert-scale questions (ranging from “very unimportant” to “very important” and “strongly disagree” to “strongly agree”), open-ended questions for additional suggestions, and multiple-select questions (using a “check all that apply” format) to identify important items. The multiple-select questions were added to keep the first two rounds of the e-Delphi from becoming too extensive. From the third round onward, only Likert-scale questions were used. The Delphi questions were developed by the authors through internal group discussions, based on the findings from the previous studies conducted in Stage 1. Each item in the framework was presented as a question in the survey, for example, “How important is it to gather information on personal interests for goal setting in arm-hand skill training for individuals with cervical spinal cord injury?” The process continued until either ≥70% consensus was reached for each item or four rounds were completed.21,22 Consensus on an item was defined as ≥70% of participants rating an item at the same end of the Likert scale: “important” or “very important” (or “agree”/“strongly agree”) for inclusion, and “very unimportant” or “unimportant” (or “disagree”/“strongly disagree”) for removal. Each round incorporated iterative feedback, presenting group responses and allowing participants to revise their views, while anonymity was maintained to reduce bias. 23 Before launch, Round 1 was piloted with two experts to ensure clarity and feasibility who were not included in the main Delphi expert panel, to ensure clarity and feasibility. From January to April 2025, data were collected through multiple anonymous rounds using the eDelphi software (eDelphi.org).
The first round began with a 10-min framework introduction video and collecting participant demographics (age, gender, country, discipline, and experience). Round two incorporated detailed items and suggestions from round one. For rounds three and four, the response format was adjusted to a 4-point Likert to achieve consensus in the later stages of the e-Delphi. The mid-point option was removed to ensure that participants had to take a position. 22
Frequency tables, medians, and interquartile ranges were analyzed using IBM SPSS version 29.0. Unscored items were treated as missing data. 21 Based on the frequency tables, Likert-scale items that reached ≥70% were considered to have achieved consensus. In the multiple-select questions, participants indicated which of the summarized items they considered important to include. Items rated as important by 50% or more of the participants were carried forward to the next round, where they were evaluated using a Likert scale. Qualitative content analysis of the open-ended responses was conducted by two reviewers following the approach described by Elo and Kyngäs. 24 The first author (BN), a PhD student in rehabilitation sciences with a background in occupational science, and the second reviewer, a master's student in occupational science (VT), independently analyzed the data using Excel. Any disagreements were resolved through discussion between both reviewers and two additional authors (JY and SA). New framework items were derived from the content analysis and discussed with co-authors before being included in the questionnaire of the next round.
Stage 3: Expert panel discussion
An expert panel of Dutch-speaking individuals with cervical SCI was recruited through purposive sampling. Inclusion criteria comprised: (1) traumatic or non-traumatic SCI at neurological levels C1–T1, (2) American Spinal Injury Association Impairment Scale grades A–D, and (3) followed arm-hand skill training. Exclusion criteria were cognitive or communication impairments that would limit meaningful participation in group discussions. Recruitment was conducted in collaboration with rehabilitation centers and patient organizations in Belgium and the Netherlands.
As key stakeholders, individuals with cervical SCI were consulted in the final stage of framework development to refine its items. After reviewing an introductory video about the framework, they participated in an online expert panel discussion. Their feedback was based on their lived experiences and insights. Sessions were recorded via Google Meet (audio and video) while a master's student in occupational therapy took field notes to ensure accuracy in data collection. These complementary recording methods enhanced transcription accuracy by capturing spoken content clearly and enabling precise speaker identification. Verbatim data transcription was performed using OpenAI's Whisper (Whisper: OpenAI's Multi-Lingual Speech-to-Text Model) and reviewed for accuracy. To ensure data validity, the transcripts were returned to participants for member checking, allowing them to review and verify their statements.
Automated transcripts were verified against recordings and field notes before importing into NVivo (version 12). Content analysis was performed according to Elo and Kyngäs. 24 The analysis was conducted by two independent reviewers: the first author (BN) and the third author (DE), a PhD student in occupational science. Any discrepancies between the reviewers were resolved through discussion to ensure consistency and reliability in the coding process. An item was considered agreed upon and included in the framework only if all seven participants supported the element.
Results
Stage 1: Item generation and development
For the development of the framework, the e-Delphi panel agreed upon three key phases: (1) gathering knowledge, about the person by assessing personal needs, load capacity and the environment; (2) goal setting, where therapists and individuals collaboratively establish person-centered, realistic goals; and (3) acting phase, involving a goal- and task-oriented training program focused on meaningful activities and their integration into daily life. Motivation is a critical factor throughout the whole framework, promoting engagement and active participation. Sixty-five items were identified to be incorporated into the framework. The results of the item selection based on the existing evidence are presented in Supplemental Table 4.
Stage 2: e-Delphi study with international experts
The e-Delphi study included 24 participants, comprising 16 occupational therapists and eight physiotherapists. Their average professional experience was 16.7 years (SD = 9.9), and they represented 12 different countries. To maintain anonymity, the country of employment is only reported for the first round. An overview of participant demographics is provided in Table 2.
Demographics of the international experts.
The Delphi process began with the 65 items developed in Stage 1. To ensure feasibility within the intended 30-min timeframe, 40 items were included in the first round, followed by the remaining 25 items in the second round. A flowchart outlining the progression and inclusion of items across rounds is presented in Figure 1. All items presented to the international Delphi panel, together with their percentage agreement, median, and interquartile range for each round, are provided in Supplemental Table 5.

A flowchart outlining the progression and inclusion of items across rounds.
Round 1
In the first round of the Delphi study, the three rehabilitation phases of arm-hand skill training identified in the literature were confirmed by the Delphi expert panel.
Consensus was reached on 39 of 40 items. In the gathering knowledge phase, agreement was reached to include assessment of personal needs and load capacity, based on nine person-specific and two social–environmental items. Person-centered and realistic goals were considered important for phase 2. In the third phase agreement was reached on training meaningful activities to support their integration into daily life. Identification of activities based on goals was considered important. Early implementation of arm-hand skill training was supported, and 10 associated task-oriented training components were identified. Consensus was reached on four session-based dose dimensions: active arm-hand time, intensity (objective/subjective), and difficulty. The integration of relearned activities across multiple contexts, along with four motivation strategies to support engagement, was emphasized.
Round 2
Consensus was reached on 37 of 40 items. An agreement was reached on assessing personal needs through four specific items. Short-term and SMART-based goal setting was also supported. In the acting phase, it was agreed that strength and endurance training would be delivered through functional activities supported by nine and three task-oriented components, respectively. Consensus was reached on four program-level dose dimensions: training duration, number of training days, sessions per day, and session length. Training meaningful activities was considered essential for maintaining motivation.
Four session-based therapy dose dimensions were agreed upon for objective assessment: active arm-hand time, intensity (objective and subjective), and difficulty. The importance of a structured weekly therapy plan and clear communication of this plan to patients was also established.
Round 3
Consensus was reached on 21 of 25 items. Missing data, defined as unanswered questions in this round, were as follows: one participant omitted 12 items; two participants omitted 2 items each; and three participants omitted 1 item each. Cognitive function was included as a component in assessing personal load capacity. While round 2 supported strength training via functional activities, round 3 specified the need to combine isolated strength training with functional training. Two additional task-oriented components for strength training and three for endurance training were identified.
An agreement was reached on five reasons for integrating relearned activities into the patient's own environment and three additional strategies for sustaining motivation in arm-hand training. Consensus was also achieved on combining subjective and objective methods to assess personal needs, and on tailoring the timeframe for evaluating the framework according to time since injury. A therapist’s experience in SCI rehabilitation was identified as an important factor in goal setting.
Round 4
In round 4, consensus was reached on three of four items. Consensus was achieved on evaluating the entire process in the subacute phase between 4 and 6 weeks and including the institutional environment. It was also agreed that the task-oriented training component, “clear functional goal,” is inherent to “client-centered goal.” No consensus was reached on objectively measuring the environment.
The level of agreement, median, and interquartile ranges for the framework items are presented in the Supplemental material.
Stage 3: Expert panel discussion
Nine individuals with cervical SCI confirmed their willingness to participate in the expert panel; however, two were unable to attend the panel discussion due to health-related issues. The seven participants had a mean time since injury of 9.6 years, with lesion levels ranging from C2 to C6. One participant was unaware of the exact lesion level in the cervical spinal cord region; not all participants knew their abbreviated injury scale (AIS) score. All participants received arm-hand training within rehabilitation wards affiliated with the Dutch–Flemish Spinal Cord Society, which follows shared rehabilitation guidelines and procedures for individuals with cervical SCI. An overview of participant demographics is presented in Table 3.
Demographics of the experts with lived experience.
SD: standard deviation; AIS: abbreviated injury scale.
Gathering knowledge phase
Participants agreed with the items of the gathering knowledge phase and emphasized the importance of focusing on performance capacity within personal load capacity, particularly early post-injury. As a result, the prioritization of personal load capacity items was adjusted. As rehabilitation progressed, they highlighted the need for therapists to better understand them as individuals by assessing personal needs and load capacities. To support this, participants recommended early education, leading to the incorporation of educational components such as “What is a cervical SCI?,” “What can I expect?,” and “What does the rehabilitation process involve?” into the framework. Notably, participants also emphasized that the social environment should provide support rather than care, a distinction that was explicitly integrated into the environmental section of the framework.
Goal-setting process
Participants supported the goal-setting items, highlighting the motivational value of short-term goals. They noted that early education, as discussed in the previous phase, is also important in fostering collaborative and realistic goal-setting alongside the therapist's clinical expertise. Participants further indicated that being in their own environment enhanced their ability to define personalized goals, a point that has been incorporated into the framework.
Acting phase
Participants agreed with the acting phase items but stressed that therapists should explain how each exercise contributes to the predefined goals, as understanding the purpose of the exercise was essential for maintaining motivation. This recommendation was explicitly integrated into the framework. Participants consistently reported that working toward personalized goals, which need to be set in the second phase, helped them transfer relearned skills to meaningful daily activities, which fostered a sense of independence and increased therapy intensity in both subacute and chronic stages. In the chronic phase, individuals with cervical SCI described using daily life activities as their daily motor training.
Motivation
Participants agreed on the motivational strategies outlined in the framework but identified several items as particularly influential: showing progress, engaging in goal-directed and meaningful activities, and maintaining a positive therapeutic relationship. These items were therefore prioritized within the framework. Furthermore, the therapist's enthusiasm was described as contagious, with participants stating that the therapist's energy directly impacted their own engagement and perseverance. This insight was incorporated into the framework as a motivational factor.
Participants viewed the framework positively, recognizing its potential to enable more person-centered, individualized, and effective arm-hand skill training. Figure 2 presents a graphical representation of the validated evidence-based framework on arm-hand skill training.

(a) Validated evidence-based framework on arm-hand skill training. (b) Validated evidence-based framework on arm-hand skill training.
Discussion
This evidence-based framework, validated by experienced therapists and individuals with lived experience, offers structured guidance for therapists in delivering person-centered training to enhance arm-hand skilled performance and facilitate integrating meaningful activities into daily life. It comprises three phases—gathering knowledge, goal setting, and acting—with motivation embedded throughout.
Aligned with the World Health Organization’s (WHO’s) call for person-centered rehabilitation, the current framework places personal needs—encompassing individual interests, values, preferences, and pre-injury behaviors—at its core. 25 Person-centered care aims to tailor interventions to individual values and integrate the patient's perspective, which individuals with SCI consider essential for recovery. 26 However, implementation remains challenging due to evolving patient needs. 27 Our framework addresses this by incorporating regular reassessments. Beyond identifying needs, Constand et al. 28 highlight the importance of recognizing each patient's unique characteristics, though they do not specify which traits. Our framework addresses this by defining six personal load capacity categories, most notably cognitive impairments and preferred learning strategies, which are often overlooked despite their impact on recovery.29,30 Currently, therapists lack structured tools to assess these personal load capacity items, relying instead on intuition, which limits the potential for person-centered care.
Collaborative goal-setting is essential to person-centered rehabilitation. 31 Our framework supports personalized goal-setting aligned with individual needs and personal load capacity. While a Cochrane review found that focusing on meaningful goals improves quality of life and satisfaction, effects on physical and social outcomes are limited. 32 More recent studies in general and spinal cord rehabilitation show that active goal involvement boosts individuals' motivation, adherence, and confidence.33,34 However, patients in general and SCI rehabilitation indicate that early participation in goal-setting is often hindered by poor health or limited understanding of their condition.27,34 Our framework tries to overcome this through patient education in the gathering knowledge phase. As our expert panel of individuals with cervical SCI highlights, personal goals often become clearer after discharge, when real-world needs and priorities emerge. This underscores the importance of regularly reassessing personal needs, as incorporated in our framework, to ensure rehabilitation goals remain relevant and responsive. Structured tools such as the Canadian Occupational Performance Measure can support this process by identifying the skills individuals use most frequently, thereby facilitating meaningful and individualized goal-setting. 31
According to the WHO, 25 rehabilitation aims to enhance independence and participation in meaningful life roles. While our framework organizes existing evidence to guide therapists in arm-hand skill training, a major challenge remains: integrating relearned skills into real-life contexts. Both individuals with cervical SCI and therapists report difficulties in preparing for this transition 12 ; many individuals struggle to apply skills in unstructured, inaccessible home environments, further complicated by shifts in identity and life roles.35,36 Acceptance of the new life situation and social support facilitate more effective skill use, 36 yet the literature offers limited guidance on integrating meaningful activities into the home environment. The European Stroke Organization recommends a behavioral transfer package, based on expert consensus, that includes diaries, home practice, and follow-ups—to address this gap. 37 Though developed for stroke, such strategies may be adaptable to spinal cord rehabilitation.
Motivation is a key driver of behavioral change and a strong predictor of rehabilitation outcomes.38,39 The current framework includes nine therapist-led strategies to sustain motivation in arm-hand training, aligning with previous findings in individuals with cervical SCI, emphasizing optimism, positive coping, and therapist encouragement as essential for recovery.27,40 Current strategies included in the framework are consistent with those identified in stroke rehabilitation, including the importance of feedback, varied exercises, clarity of purpose, and experiences of success. 39 The framework highlights a person-centered approach that integrates goal-oriented training and meaningful activities to enhance engagement.
Future research should explore the implementation of this framework in rehabilitation contexts. Its effectiveness in improving arm-hand function and facilitating the integration of meaningful daily activities should be explored. Additionally, research should aim to establish consensus on appropriate assessments for measuring personal load capacity and to develop new tools where gaps exist. Finally, creating a reliable method to measure therapy dose objectively is essential to tailor interventions to personal load capacity and optimize outcomes.
The framework was developed by triangulating three peer-reviewed studies led by SCI experienced researchers. While physiotherapist participation in the Delphi study was lower, all therapists had substantial experience. Two participants—one occupational therapist and one physiotherapist—had slightly less than five years of experience but were included due to their relevant expertise in SCI rehabilitation. Of the initial 24 participants, 17 completed all four rounds, resulting in a 29% attrition rate. No standards for acceptable attrition rates have been established for Delphi studies. Minor missing data in round three arose from language clarity issues and one incomplete survey; unscored items were treated as missing. 21 Despite global recruitment, most experts were from Western countries, potentially limiting transferability. One item—patient-customized training load in strength training—was omitted from Delphi scoring due to a technical error, but included in the framework based on strong support from focus group discussion and the systematic review.
In conclusion, this framework was developed through an evidence-based process and validated by experienced occupational therapists, physiotherapists across multiple countries, and individuals with cervical SCI. The framework is specifically developed to guide therapists in delivering personalized arm-hand skill training for individuals with cervical SCI, aiming to enhance arm-hand skilled performance and the integration of meaningful daily activities into daily life. The framework structures training around three key phases—(1) gathering knowledge, (2) goal-setting, and (3) acting phase—with motivation recognized as a critical factor sustaining patient engagement throughout rehabilitation.
Clinical messages
Using the three phases—knowledge gathering, goal setting, and acting—while keeping patients motivated, can help therapists provide effective arm-hand rehabilitation for individuals with cervical spinal cord injury.
Arm-hand skill training should be highly person-centered, with realistic and personal goals set collaboratively.
Task-oriented training in skill, strength, and endurance training enhances arm-hand skilled performance and integration of activities of daily living.
Maintaining motivation is key to sustaining patient engagement and improving outcomes in rehabilitation.
Supplemental Material
sj-docx-1-cre-10.1177_02692155251413776 - Supplemental material for A validated framework to guide therapists in arm-hand rehabilitation for individuals with cervical spinal cord injury
Supplemental material, sj-docx-1-cre-10.1177_02692155251413776 for A validated framework to guide therapists in arm-hand rehabilitation for individuals with cervical spinal cord injury by Nele Bertels, Yvonne Janssen-Potten, Eva Delooz and Annemie Spooren in Clinical Rehabilitation
Footnotes
Acknowledgements
We would like to thank Trui Van Campenhout (VT), a master's student in the Occupational Science program, for helping with the data collection and analysis of the Delphi study. During the preparation of this work, the authors used ChatGPT 3.5 and DeepSeek-V3 in order to improve the readability and language of the manuscript. After using this tool/service, the authors reviewed and edited the content as needed and take full responsibility for the content of the publication.
Ethical approval
Ethical approval was obtained from the University of Hasselt's Medical Ethics Committee (CME2024/056) for the e-Delphi study and expert panel discussion. All participants provided written informed consent prior to participation.
Author contributions
NB, YJ, and AS were involved in the conceptualization and development of the methodology. NB, YJ, ED, and AS contributed to data curation and analysis. NB drafted the original manuscript. All authors participated in the writing, review, and editing of the manuscript. All authors have read and approved the final version of the manuscript.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by Bijzonder Onderzoeksfonds UHasselt (grant number 21OWB23).
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Data availability statement
The data supporting this study's findings are available from the corresponding author upon reasonable request.
Supplemental material
Supplemental material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
