Summary Workers and the other stakeholders who make decisions regarding return-to-work interventions often have diverging interests and goals. A shared decision-making (SDM) process is a person-centred approach that helps close gaps between workers and other stakeholders regarding the rehabilitation aim and the action plan to be implemented. To the best of our knowledge, no SDM program has ever been specifically adapted to the occupational rehabilitation context. The ultimate aim of this study is to promote attainment of a common goal by health practitioners and workers in the context of occupational rehabilitation programs, while taking into account employers’, unions’ and insurers’ interests and constraints. The following objectives were defined to this end: Document the acceptability and feasibility of a shared decision-making program applicable to existing rehabilitation programs, with four groups of key rehabilitation stakeholders (workers, employers, unions and insurers). Evaluate the efforts required to implement the shared decision-making, with four groups of key rehabilitation stakeholders. Explore the impact of the implementation context on satisfaction and the return to work, with four groups of key rehabilitation stakeholders. This study falls into the category of evaluation research. For objective 1, a mixed sequential design combining quantitative and qualitative data was chosen. An adapted version of the Technique for Research of Information by Animation of a Group of Experts (TRIAGE) was used. It included two phases: (1) individual consultations by means of a self-administered questionnaire to establish level of agreement on the acceptability and feasibility of the various components of the program (i.e. objectives, indicators, resources and activities); and (2) a series of focus groups with each stakeholder group. Descriptive analysis of the agreement scores and content analysis of the focus groups were performed. A sample of ten participants per group (employers, unions, workers and insurers) was targeted, for a total of 40 participants. The selection criterion for insurers was at least two years’ experience in occupational rehabilitation. The employers and unions had to have participated, within the past two years, in the rehabilitation process of at least one worker who had been on sick leave for more than three months for persistent pain caused by a musculoskeletal disorder (MSD). The workers had to have participated (successfully or not), within the past two years, in an occupational rehabilitation program for persistent pain caused by an MSD resulting in a work absence of more than three months. For objectives 2 and 3, a multiple case study approach was used. A case was defined as the dyad comprising an occupational therapist and a worker with a persistent MSD causing a work disability. The context was the rehabilitation centres with which the various stakeholders (insurers, employers and unions) have to collaborate. Different sources of information were identified in each case: with the worker/occupational therapist dyad as well as the other stakeholders working with the dyad worker, i.e. the employer, union representative and rehabilitation counsellor. Regarding the information sources, a variety of methods (questionnaires, interviews and observation) were used to measure the degree of attainment of each objective of the SDM program (one longitudinal objective, 11 specific objectives), using indicators predetermined in the SDM program. Prior to program implementation, the occupational therapists did training on the SDM program. Intra-case analyses were performed first. An analytical matrix was developed containing the objectives, indicators, their measurements and the way the score was interpreted to establish a level of implementation: complete implementation (score of 1), nearly complete (score of 0.75), partial (score of 0.50), minimal (0.25) and no implementation (score of 0). Inter-case analyses were then performed to identify patterns between cases, or more specifically, arrangements of actions or results that were grouped together and were visible in the raw data. For objective 1 of the study, a total of 39 key stakeholders completed a questionnaire during the individual consultation phase, and 38 of them attended focus groups. The individual consultations generated 37 proposals for changes to the objectives of the SDM program, 17 proposals regarding the activities and 39 comments on the feasibility of implementing the SDM model in an occupational rehabilitation context. The focus group results highlighted the fact that the SDM program was well accepted by the stakeholders. It was mainly nuances in the operationalization of the program that were suggested to increase implementation feasibility. A physical resource (an interview guide) was added to objective 2 of the SDM program to comply with the guidelines set forth in the insurance contract and the employer’s constraints. Three indicators were added to the longitudinal objective of the SDM program. For objectives 2 and 3 of the study, 39 cases were analyzed and, for 25 implementation scores (45%) out of 56, an implementation rate of more than 75% was observed. Conversely, six scores (11%) had an implementation rate of less than 25%. The least effectively implemented indicators were found first in relation to objective 1 of the program concerning the building of a working alliance between the worker and the occupational therapist, but particularly in relation to the indicators measured using an interview observation checklist. These indicators evaluated whether the occupational therapist allowed the worker to ask questions or express his or her opinion, or if the therapist checked whether the worker had accurately understood the information provided. The longitudinal objective of the SDM program, i.e. maintaining the alliance with the other stakeholders, was also less effectively implemented. Three out of four indicators revealed that there was rarely stakeholder consensus on decisions regarding the option, the objective and the action plan. The inter-case analyses generated three main case types. Case type 1 (n=2) was dubbed “return to work with positive progression and joint action in which there was consensus among all the stakeholders with regard to understanding, the objective, the option and the plan of action.” Case type 2 (n=14) involved a “return to work with various minor or a few moderate gaps with regard to understanding or to the implementation of the option or the chosen plan of action.” However, these obstacles were resolvable. Case type 3, where there was no return to work, involved major obstacles for which no solutions could be implemented, and revealed two sub-types: with decision-making regret (n=3) and without decision-making regret (n=11) regarding major changes made in the initially chosen option and the implementation of the action plan. All of the results emerging from the second and third objectives of the study supported a mixed quantitative and qualitative approach to developing the intervention. The SDM program in the occupational rehabilitation context offers clinicians a systematic means of helping workers make decisions regarding their return to work. The findings of this study also support one of the foundations of the SDM approach, namely, that when a worker is involved in a truly shared decision-making process, even if the intervention outcome is not favourable, there is no regret about the decision made.