J Occup Rehabil (2015) 25:141–152 DOI 10.1007/s10926-014-9532-7

Operationalizing a Shared Decision Making Model for Work Rehabilitation Programs: A Consensus Process Marie-France Coutu • France Le´gare´ • Marie-Jose´ Durand • Marc Corbie`re Dawn Stacey • Lesley Bainbridge • Marie-Elise Labrecque

•

Published online: 8 July 2014 Ó Springer Science+Business Media New York 2014

Abstract Purpose The objective of this study was to design and operationalize shared decision making (SDM) rehabilitation model for worker rehabilitation programs. SDM has previously been shown to improve decision outcomes in patient-health care professional relationships. To date, SDM has not yet been adapted to work rehabilitation, although it could be a valuable approach to better understand and agree on return-to-work decisions. Methods We designed a preliminary model for return-to-work decisions for workers suffering from pain due to musculoskeletal injuries. We submitted the preliminary model and a questionnaire to expert health care professionals. Using the Technique for Research of Information by Animation of a Group of Experts method, a group consensus M.-F. Coutu (&)  M.-J. Durand  M. Corbie`re  M.-E. Labrecque Research Center – Hoˆpital Charles-Le Moyne, Universite´ de Sherbrooke, 150, Place Charles LeMoyne, Longueuil, QC J4K 0A8, Canada e-mail: [email protected] M.-F. Coutu  M.-J. Durand  M. Corbie`re  M.-E. Labrecque School of Rehabilitation, Universite´ de Sherbrooke, Longueuil, QC, Canada F. Le´gare´ Research Center of Centre Hospitalier Universitaire de Que´bec, Hospital St-Franc¸ois d’Assise, Quebec, Canada D. Stacey Faculty of Health Sciences, School of Nursing, University of Ottawa, Ottawa, Canada D. Stacey Ottawa Hospital Research Institute, Ottawa, Canada L. Bainbridge University of British Colombia, Vancouver, Canada

process was used to discuss and refine the experts’ responses to operationalize a model adapted for rehabilitation. Results Eleven occupational therapists (three were clinical coordinators) and four psychologists participated in three group consensus sessions. The final version of the model included one general longitudinal objective (the maintenance of a working alliance and assuring mutual comprehension among all stakeholders), and 11 specific objectives: establishing a working alliance, seven in the deliberation phase of the SDM process, and three in the implementation of the decision. Participants also reached consensus on between 1 and 8 indicators per objective. Conclusion We developed and operationalized an SDM rehabilitation model intended for a return-to-work implementation plan. The next step will be to document its feasibility among the main stakeholders (employer, union, insurer and worker) taking part in decisions about return to work. Keywords Rehabilitation  Shared decision making  Work  Injuries  People with disabilities  Chronic pain  Consensus  Process theory

Background Work is the main activity of 66.2 % of adults in Canada [1]. It provides workers with financial autonomy, selfrespect, dignity, quality of life, and a sense of self-worth [2, 3]. A recent survey in 2007–2008, of the Quebec population in Canada (EQCOTESST) shows that nearly 238,000 workers reported being absent from work in the last year because of a musculoskeletal disorder (MSD) that was perceived as being mainly associated with their work [4]. In general most of the workers will return to their

123

142

regular work after 3–4 weeks [5]. In EQCOTESST, 7 % of workers reported being absent more than 12 weeks. After this period, work disability becomes more complex and requires specific interventions. The Organisation for Economic Co-operation and Development (OECD) states that disability benefits have a significant impact on the public finances of most OECD countries and currently limit economic growth and reduce the effective supply of labor [6]. Because MSDs are a significant public health problem [7], great effort has gone into providing effective interdisciplinary work rehabilitation interventions for patients with prolonged work disability due to an MSD. Many studies and systematic reviews have recognized the effectiveness and cost benefits of these interventions [8–12]. However, workers and health care professionals making decisions about returnto-work interventions often have different expectations. Workers tend to focus primarily on the reduction of pain, while health care professionals focus on the reduction of the disability [13]. The different expectations of workers and health care professionals result in the use of different criteria for evaluating treatment efficacy and even different action plans, which contribute to negative outcomes [13]. For pain secondary to an MSD, a panel of experts convened by the American Pain Society recommended that a shared decision making (SDM) approach is helpful in managing the complex trade-offs between potential harm, benefits, costs and burdens of available treatments [14]. SDM is commonly defined as a relationship between a patient and a health care professional where there is an exchange of information, a process of expressing and discussing preferences, and an agreement about the goals and action plan, followed by an assessment and readjustment of the plan [15]. Nine essential elements for SDM have been identified [16]: defining/explaining the problem; presenting the options; discussing pros and cons; values and preferences; self-efficacy; verifying comprehension; doctor’s recommendation; making or explicitly deferring the decision; follow-up. Consequently, if well implemented, SDM can reduce gaps between health care professionals’ and patients’ understanding of each other’s knowledge and values/goals regarding treatment options [17, 18]. A systematic review of barriers and facilitators to implementing SDM in clinical practice, as perceived by health professionals, revealed that among 28 studies using data collected from 15 countries, none addressed implementation of SDM in a work rehabilitation program for patients with pain secondary to an MSD [19]. Since then, one study exploring health professionals’ beliefs about chronic pain treatment and how these beliefs influence the process of care concluded that SDM should be implemented to improve patient care [20]. To the best of our knowledge, an SDM model has not been specifically designed for work rehabilitation. In a

123

J Occup Rehabil (2015) 25:141–152

knowledge-to-action process, the adaptation of the model to the context is crucial [21]. Consequently, our goal was not only to identify essential elements of SDM but also to operationalize the model, based on Rossi et al. [22], in order to relate objectives to specific activities, resources and indicators. Also, providing a detailed description of behaviour change interventions facilitates their implementation in clinical settings and scientific replication [23]. In studying a complex intervention, such as SDM, one of the first steps is to build a model that is plausible, as well as theoretically and empirically derived [24, 25]. According to Rossi et al. [22], the explicit conceptualization of a program’s objectives and how to attain them, are generally not sufficiently developed when planning new programs. A program theory ‘‘explains why the program does what it does and provides the rationale for expecting that doing things that way will achieve the desired results (p. 165)’’ [22]. This approach is part of evaluative research, which is a scientific procedure that allows the understanding of causal relationships between components of an intervention [26]. The objective of this study was to develop a replicable SDM rehabilitation model and to operationalize it for work rehabilitation programs/interventions. This objective is part of a broader study, whose general objective was to implement an SDM program adapted to the realities of work rehabilitation for workers suffering from persistent pain secondary to a musculoskeletal injury.

Methods Ethics Statement The study was approved by the Research Ethics Committee of the Research Centre of the Charles Le Moyne Hospital in Longueuil, Quebec; all participants gave written informed consent. Design of the Study This study is related to the field of evaluative research [27]. Our theoretical perspective is pragmatic, i.e. knowledge arises out of action, situations and consequences and occurs in a social and political context [28]. Therefore, we used a sequential mixed methods design [28]. To operationalize the model, which involved relating objectives to specific activities, resources and indicators (for an example see Fig. 1), we followed a method by Rossi et al. [29]. Within a program theory, we find a ‘‘program process theory’’. It is ‘‘a plan or blueprint for what the program is expected to do and how.’’ Program process theory is built on needs assessment, links program design with the social conditions the program is intended to improve, and should

J Occup Rehabil (2015) 25:141–152

include input or endorsement by major stakeholders [29]. We describe our program process theory as ‘‘operationalizing an SDM rehabilitation model.’’ We first gathered empirical data and developed a preliminary version of the SDM rehabilitation model. Second, according to Rossi et al. [29], the logic and plausibility of the model can be assessed with these two questions: ‘‘is it well defined?’’ and ‘‘is it reasonable?’’ To answer these two questions for the operationalized model, we consulted expert health care professionals, who are our major stakeholders, as they are the potential service providers for this intervention. Specifically, we consulted a group of experts, during working hours, to document if the model was entirely clear, precise, and practical given current practices [29]. If the model was found to have shortcomings, we needed a technique to adapt the model to the work rehabilitation context. We used an inductive and interactive technique called the ‘‘Technique for Research of Information by Animation of a Group of Experts’’ (TRIAGE) [30]. This technique is part of evaluative research and is useful when seeking indicators, objectives or improvements [30]. It is a twophase group consensus method by which data is compiled, studied, analyzed, and then discussed among experts to reach a consensus. Individual questionnaires were sent to expert health care professionals, and they were provided with the preliminary model and the data. Then we conducted group consensus sessions with the expert health care professionals in four rehabilitation centres to discuss and propose, if judged necessary by the experts, an adaptation of the preliminary SDM model. TRIAGE has been recommended for developing rehabilitation programs by expert committees and was also identified as a strategy for implementing partnerships with clinical settings, as it includes health care professionals, one of the major stakeholders, early in the research process [31]. Details of the study protocol have been published elsewhere [31, 32]. Participants and Setting We recruited a convenience sample of experts comprising occupational therapists (OTs) and psychologists from four public rehabilitation centres in Quebec, Canada. These centres were chosen because they all apply the same work disability evidence-based intervention principles [33]. They also provided a variety of expertise, given the culture and the context of the different referring agencies, including the Quebec workers’ compensation board, the Quebec automobile insurance agency, the public health insurance program, and private insurers. We recruited an initial group of 14–16 experts, taking into account the possibility of attrition. We were aiming for at least 10–12 participants in the consensus group to achieve data saturation and to maintain a good group dynamic [34]. The inclusion criteria were as

143

follows: having been working full-time for more than 2 years as psychologists or OTs in work rehabilitation for workers suffering from pain secondary to an MSD. We also wanted to include OTs or psychologists with 2 years of experience, but that also had experience in a clinical coordination position; this would provide the administrator’s point of view in terms of the barriers and facilitators for implementation. These inclusion criteria were based on our prior studies [35]; in the past, these criteria have facilitated recruitment and provided suitable data. Data Collection and Analysis TRIAGE comprises two steps: individual reflection by participants followed by a series of group consensus sessions [30]. Step 1: we sent our preliminary version of the operationalized SDM rehabilitation model (based on the literature review) by mail to each expert, accompanied by a document explaining the model and a paper questionnaire (for an example of the model see Fig. 1). The questionnaire had been pilot tested with other health care professionals. Seven items solicited responses on the level of agreement of the expert. The labelling of the items was based on the Rossi et al. [29] method for assessing the logic and plausibility. The items were: (1) the objectives are necessary and sufficient for SDM; (2) the objectives are clear and specific; (3) the objectives are feasible in the context of SDM; (4) the activities are necessary and sufficient to attain the objectives; (5) the indicators are necessary, sufficient and pertinent to measure the objectives; (6) the resources are necessary and sufficient to attain the objectives; and for plausibility; (7) the SDM process theory can be implemented in the current practice. The level of agreement could vary from 1 (totally disagree) to 4 (totally agree). When respondents ‘‘totally disagreed’’ or ‘‘disagreed’’ with a statement (rating of B2/4), they were asked to give a maximum of five suggestions of ways to improve the objective, activity, resource or indicator that was problematic. They had to complete and return the questionnaire within 2 weeks, prior to the series of group consensus sessions. Step 2: Each answer was anonymously transcribed on separate sheets of 8‘00 9 1100 paper and posted on the wall so participants could see all the answers before taking part in the group consensus. Consequently, no answers were modified or analysed prior to the group sessions. The TRIAGE method requires that participants be provided with a visual representation of all the answers [30]. To aid the process of reaching a consensus, six sections on the wall were created to help the experts to organize their answers. Group discussions started with the ‘‘memory’’ section, in which all the answers on objectives were gathered and pinned to the wall. After discussing and reaching consensus on an objective, it could either be

123

144

eliminated and placed in the ‘‘garbage’’ section or, if deemed relevant, transferred to the ‘‘grouping’’ section, which was for objectives with similar significance and which might need to be renamed. When consensus was reached on the name of a new objective, it was transferred to the ‘‘selection’’ section. Once consensus was reached on the selected objectives, they were placed in the SDM rehabilitation model that was being built by the experts on the wall. When consensus was difficult to obtain for an objective, it was temporarily placed in the ‘‘refrigerator’’ section. If consensus still could not be reached when revisited later, it was placed in the ‘‘veto’’ section, to be discussed by an external group of experts. By the end of the series of group consensus sessions, all answer sheets had been discarded, adopted for the final SDM rehabilitation model, or vetoed. The meetings were recorded to supplement the research assistant’s notes on the group’s decisions. The principal investigator (MFC) facilitated the groups. Data analysis with TRIAGE is simultaneously performed by all participants as they discuss and arrange the sheets. The process was repeated to operationalize the model (reach consensus on activities, resources and indicators). Thus a fully operationalized SDM rehabilitation model emerged at the end of all the meetings.

Results Participant Characteristics A total of 15 experts started the group consensus sessions; 11 occupational therapists (OTs) (including 3 coordinators), and 4 psychologists were recruited from 4 public rehabilitation centres in Montreal, the Montreal South Shore, and the Sherbrooke region. One OT, who was a clinical coordinator, could not attend the group because of previous engagements, but he completed the questionnaire. Two OTs dropped out after the first group, due to personal leaves of absence. Most of the participants were women, which is also representative of professionals in the rehabilitation field. The mean age was 42.46 years. Participants on average had 9.62 years of experience, with a range of 5–20 years of experience. We conducted three meetings with the same participants. Each lasted 3 h and the meetings were held over a 2-month period. Objectives were discussed in the first meeting, activities and resources in the second, and indicators in the third. Developing an Operationalized SDM Rehabilitation Model Table 1 shows the mean agreement scores obtained for the 7-item questionnaire sent to participants on the preliminary model that we derived from the literature review. The

123

J Occup Rehabil (2015) 25:141–152

average scores were C3 (‘‘agree’’ to ‘‘totally agree’’) for the objectives are necessary and sufficient for SDM (item 1), the objectives are clear and specific (item 2), and the objectives are feasible (item 3). However, individually, some participants provided ratings B2/4 (‘‘disagree’’ to ‘‘totally disagree’’) and provided modifications for objectives 1–7, which had to be discussed among participants in order to reach a consensus according to the TRIAGE method. Also, on average, participants provided agreement scores\3 for the activities (item 4), indicators (item 5), and the resources (item 6) being necessary and sufficient. Consequently, the individual consultation on the preliminary model generated a total of 22 statements on the objectives; 7 on the activities; 10 on the resources; and participants suggested 158 indicators. Many answers were similar and were therefore grouped or renamed by participants. Overall, the modifications tended to reformulate objectives to be more in line with rehabilitation practices. Two specific objectives were grouped into one longitudinal objective. Moreover, the experts added activities such as consulting files and exchanging information with relevant stakeholders. The human resources required to perform SDM were initially identifying OTs, but the experts added an interdisciplinary rehabilitation team and stakeholders such as the insurer, employer, union, family, etc. The material resources (e.g. questionnaires, decision support resources) indicated in the preliminary model were judged necessary and sufficient to attain the specific objectives, although participants unanimously required that the material resources needed to be brief in order to be implemented in practice. Participants reached consensus on a minimum of one indicator and a maximum of eight indicators per objective. By the end of the consensus sessions, there were three statements that the experts were still unable reach a consensus on, and were therefore vetoed. The group knew that these three would afterwards be discussed by the research team who would decide on the final elements. Two of the statements concerned indicators/outcomes: ‘‘the problem requiring an SDM process is solved’’ and ‘‘the common objective is reached.’’ There was no consensus on these indicators because, as mentioned by some participants, factors outside the SDM process might affect the outcome. For example, an employer might contest that a condition is work-related. If the claim is found to be non-work-related, the rehabilitation services could be terminated. Thus although a perfect SDM process was performed on the options offered for a safe and sustained return to work, these two indicators would not show positive results. The last vetoed element was the ranking of one objective (‘‘identify a common objective for returning to work’’): some participants wanted to place it before ‘‘exploration of the determinants associated with work disability,’’ and others wanted to place it after (see Fig. 1). The vetoed

J Occup Rehabil (2015) 25:141–152

145

Fig. 1 SDM process theory. 1 The number after each indicator identifies the instrument presented in Table 3

123

146

J Occup Rehabil (2015) 25:141–152

Table 1 Mean scores for the level of agreement with all the items of the questionnaire Rehabilitation centre

Item 1 Mean

Item 2 Mean

Item 3 Mean

Item 4 Mean

Item 5 Mean

Item 6 Mean

Item 7 Mean

Centre 1

4.00

3.00

3.50

2.00*

3.50

2.50*

3.50

Centre 2

2.33*

3.00

3.50

2.33*

2.50*

2.00*

3.00

Centre 3

3.00

3.00

2.50*

3.33

2.25*

2.00*

2.50*

Centre 4

3.50

3.25

3.00

3.25

3.25

2.67*

2.00*

All centres

3.27

3.07

3.07

2.71*

2.93*

2.33*

2.83*

Scores on the level of agreement varying from 1 (totally disagree) to 4 (totally agree); * Problematic items needing discussion (rating of B2/4); Item 1 the objectives are necessary and sufficient for SDM, Item 2 the objectives are clear and specific, Item 3 the objectives are feasible in a context of SDM, Item 4 the activities are necessary and sufficient to attain the objectives, Item 5 the indicators are necessary, sufficient and pertinent to measure the objectives, Item 6 the resources are necessary and sufficient to attain the objectives, Item 7 the SDM process theory can be implemented in the current practice

Table 2 The SDM essential elements regrouped into new categories for operationalizing our SDM model for work rehabilitation programs Nine essential elements most cited in the models found in SDM literature [16]

Shared decision making model for work rehabilitation programs

1.Defining/explaining the problem

Objective 2: Identify the problem needing an SDM process

2.Presenting the options

Objective 7: Identify and evaluating the options

3.Discussing pros and cons

Objective 7, Indicator 1: Advantages and disadvantages of the option

4.Values and preferences

Objective 6: Evaluate the determinants influencing the worker’s decision

5.Self-efficacy

Objective 7, Indicator 5: Worker’s self-efficacy toward the options

6.Verifying comprehension

Objective 1: Establishing a working alliance between the worker and health care professional

7.Doctor’s recommendation

Objective 7, Indicator 4: The health care professional makes a recommendation on the preferred option

8.Making or explicitly deferring the decision

Objective 5: Make the decision explicit

9.Follow-up

Objective 9, Indicator 6: The obstacles that could hinder the performance are identified and solutions are identified. Objective 10, Indicators 1: The plan of action is readjusted if needed and Indicator 3: A follow-up on the course of the action plan is made with the worker and stakeholders

elements were discussed by our research team who were experts in work rehabilitation and shared decision making, and a consensus was reached. We divided the item ‘‘the

123

problem requiring an SDM process is solved’’ into ‘‘Functional status’’ and ‘‘Return to work status.’’ We also replaced ‘‘the common objective is reached’’ with ‘‘the level of attainment of the common objective.’’ Although the outcome may be subject to factors outside the SDM process, it is important to document it. Finally, we studied the final version of the operationalized SDM rehabilitation model to make sure it still reflected an SDM approach. The essential elements of SDM found in the literature were regrouped into new categories for operationalizing our model. Some objectives were left as such; others were grouped together and given a new name, while still others were categorized as indicators. For example, ‘‘discussion of pros and cons’’ was classified as an indicator of the objective ‘‘presenting the options,’’ and ‘‘patient’s comprehension’’ was considered an indicator for the newly named objective ‘‘creating a working alliance’’ (Table 2). The final model included the long-term objective of achieving a safe and sustained return to work and the medium-term objective of a safe return to work, while the short-term objectives consisted of 1 longitudinal objective and 11 specific objectives (Fig. 1). The longitudinal objective was ‘‘to maintain a working alliance and assure mutual comprehension among all stakeholders.’’ Three activities for operationalizing this were identified: information exchange, consulting file, and workplace visit. Two indicators for assessing the attainment of the objective were the level of common understanding among the stakeholders about the work rehabilitation program, and level of common understanding about the work disability problem. The research team grouped specific objectives 2–8 under the general objective named ‘‘conduct a deliberation phase of the SDM process’’, which could be performed during a first decision making meeting. Specific objectives 9–11, grouped under the general objective named ‘‘implementation/act on the option agreed upon’’, can be addressed during the work rehabilitation program, which lasts on average 12 weeks.

J Occup Rehabil (2015) 25:141–152

147

Table 3 Instruments identified to measure the indicators Instruments

References

Concepts measured

Validation

1. OPTION observing patient involvement

[39]

Rate the 12 basic skills needed for SDM on a five-point scale from 0 (behaviour not observed) to 4 (skill observed and exhibited to a high standard)

Good construct validity [55] and has yielded very high inter-rater agreement, with intraclass correlation coefficient scores of 0.77 [39]. High internal consistency has also been observed [55]

2. Observational grid

Developed for this study

22 items rated on categorical scales (i.e.: yes, no, and minimal observation) document aspects of the working alliance (cooperation, health care professional’s self-regulation); problem needing an SDM process; establishing common objective and an action plan

Not available

3. Working alliance (worker and health care professional versions)

[56]

High Cronbach’s alphas have been observed [56] for each of the three constructs, and ranged from 0.83 to 0.98

4. Ways of mattering

[57, 58]

12 items rated on a Likert scale (1 = never to 7 = always) measuring the perception of the relationship established, of the task performed, and of the goals pursued 24 Items rated on a Likert scale (1 = very rare to 5 = very often) measuring the attention, importance given to the patient by the health care professional. The feeling that the health care professional is interested in the patient and the dependence (the patient’s contribution is seen as essential)

5. Return to Work Obstacles and Coping Efficacy Scale—musculoskeletal disorder

[36]

97 items rated on a Likert scale (1 = not at all to 7 = completely) measuring 8 dimensions that could be factors hindering return to work and the level of self-efficacy to overcome each obstacle (general health; motivation/selfefficacy; social support; workplace; workstation; interpersonal relationships with colleague and supervisor; relationship with insurer; personal life)

Under study

6. The Brief Illness Perception Questionnaire (Brief-IPQ)

[59]

8 items rated on a Likert scale varying from 0 to 10 with the anchor description being adapted to the item. It measures the components of illness representation (identity/symptoms; control, timeline, coherence, emotional representation, consequences, causes). One item is to rank-order the three most important factors that caused the illness

Moderate to strong correlations (r = .46–0.63) were found between the scores on the Brief IPQ and the long version of the IPQ-R, with the exception of the personal (r = .33) and treatment control (r = .32) subscales [59]. Test–retest reliability coefficient at 3 and 6-week intervals varies from .42 to .75. The predictive validity at 3-months is satisfactory on functional status and quality of life

7. Decisional Conflict Scale (DCS)

[60]

16 items rated on a Likert scale (1 = strongly agree to 5 = strongly disagree) measuring the uncertainty in choosing options, modifiable factors contributing to uncertainty (information, values and social support)

Test–retest reliability coefficient at 2-week interval is of 0.81. Internal consistency coefficients ranged from 0.78 to 0.92. Discriminant validity is satisfactory. A weak inverse correlation (r = -0.16, p \ 0.05) is observed between the DCS and knowledge test scores.

8. Ottawa personal decision guide

[41]

A structured interview based on four sections assessing the decisional needs, such as the knowledge, values, level of certainty, and social support or pressure

Good apparent construct validity [61]

High Cronbach’s alphas varying from 0.82 to 0.91

123

148

J Occup Rehabil (2015) 25:141–152

Table 3 continued Instruments

References

Concepts measured

Validation

9. Agreement Questionnaire (worker and health care professional versions)

Developed for this study

7 items rated on a Likert scale (1 = strongly disagree to 7 = strongly agree) measuring the perception of the feasibility and the meaningfulness of the objective, the option and the action plan. 3 open questions assess the objective, the option and the action plan as perceived by the worker or the health care professional are also documented. If health care professional made a recommendation (y/ n) is documented. The dyadic measure helps identify the agreement between the worker and the health care professional on the objective, the option and the action plan

Not available

10. Work Status Questionnaire

[62]

By means of a decision tree, three categories of work status are identified: (1) back at work (old job or another job; full-time or part-time); (2) absent from work due to a treated problem; (3) absent from work for a reason other than the treated problem.

Not available

11. Decisional Regret Scale

[63]

5 items on a Likert scale (1 = strongly agree to 5 = strongly disagree) measuring the level of regret regarding a decision

Cronbach’s alphas varying from 0.81 to 0.92

12. Patient’s Satisfaction Questionnaire

[64]

Cronbach’s alphas varying from 0.8 to 0.9

13. Practitioner’s Satisfaction Scale

[64]

Adaptation of the 6 items rated on a Likert scale (1 = strongly disagree to 5 = strongly agree) measuring the level of satisfaction with the SDM consultation An adaptation using 11 items on a Likert scale (1 = not at all to 5 = completely) measuring the level of satisfaction with the SDM consultation

14. Functional status

[65]

2 items on a 0–100 % scale (0 = not at all to 100 % as before the accident) assesses the percentage of work or activity of daily living currently being carried out by the worker and compared with before the accident or injury

Not available

15. Semi-structured individual interview

Developed for this study

Semi-structured individual interview guide documenting stakeholder’s perception of the factors hindering and facilitating SDM implementation

Not available

The instruments for measuring indicators (Table 3) were identified by the research team, since participants did not feel qualified to identify valid and reliable instruments. The Return to Work Obstacles and Coping Efficacy Scale identified in Table 3 currently has 97 items, but a shorter and validated version should be available shortly [36]. The Brief Illness Perception Questionnaire was retained since it has 8 items. However, a longer version, the Revised Illness Perception Questionnaire Adapted for Work Disability [37, 38], is available online with automatic scoring and interpretation (http://qrit.recherche.usherbrooke.ca/en/home) and was validated for workers having a musculoskeletal disorder. Finally, some tools had to be developed by the research team, since no tools were available in the literature. These included an observational grid of the SDM

123

Cronbach’s alphas varying from 0.8 to 0.9

meeting, semi-structured interview guides with stakeholders, and a dyadic questionnaire to assess the feasibility and the meaningfulness for the worker of the objective, the option and the action plan. We also wanted to go beyond the perception of having reached an agreement. Therefore, we posed questions for the worker and the health care professional on what was the objective, the option, and the action plan.

Discussion With a series of three group consensus sessions, we were able to develop an operationalized SDM rehabilitation model based on consensus among service providers that

J Occup Rehabil (2015) 25:141–152

could be applied to the realities of work rehabilitation for workers suffering from persistent pain secondary to a musculoskeletal injury. Also, we were able to incorporate the nine essential elements most cited in the models found in SDM literature [16] into our final model. Thus our SDM rehabilitation model not only builds on current evidencebased knowledge and essential elements of existing SDM models, but also translates theoretical approaches to SDM into a fully operationalized model appropriate for rehabilitation that is ready for feasibility studies. Our results led us to make four observations. A first observation is that we were able to incorporate elements of SDM-related skills development tools into the operationalization of our model. We included the OPTION instrument [39, 40] and the Ottawa Decision Support Framework (ODSF) [18] in our SDM rehabilitation model, as these are tools that rehabilitation health care professionals could use to implement the deliberation phase of SDM [41]. Health care professionals can consult a website (https://decisio naid.ohri.ca/decguide.html) and get a tutorial to learn how to use the ODSF. This framework stood out as especially helpful in the rehabilitation context because it identifies determinants of decisional conflict, such as perceptions of the decision and treatment options, ‘‘significant others’’ involved in the decision, and the personal and external resources required to make and implement a decision [18]. Also, we took into account the inter-professional approach to SDM [42, 43], based on a review of 15 conceptual models of SDM and models consistent with the definition of SDM [44]. This approach is also especially pertinent to rehabilitation, where several kinds of professionals must collaborate to reach a decision with the injured worker. A second observation is that our SDM rehabilitation model formalizes goal setting and the development of a working alliance as new objectives. The need to establish a partnership with the patient has been recognized in the SDM literature [17, 44] and identified in SDM models [44]. However, to our knowledge, it has not been operationalized with objectives and indicators in order to help health care professionals identify what to implement. The establishment of a working alliance is the foundation for successfully identifying a mutual goal [45]. We therefore included an objective related to the establishment (objective 1) and maintenance (longitudinal objective) of the working alliance in our operationalized SDM rehabilitation model. More recently, Matthias et al. [46] have identified four categories of behaviours critical for effective communication during an SDM encounter as elements that contribute to a working alliance, namely, investing in the beginning; eliciting the patient’s perspective; demonstrating empathy; and investing in the end. It was also important to formalize goal setting, which has to be the starting point from which options for treatment or management approaches will be

149

discussed. In the SDM literature, an association between goal setting and treatment adherence was found [47]. Goal setting is also considered a fundamental component of rehabilitation programs [48]. When goals are meaningful for the patient [49] and formulated in a specific manner, the outcomes in terms of physical performance are higher compared to non-specific goals [50]. A third observation relates to the ultimate goal of the work rehabilitation program. When patients are referred by a third party payer (private or public insurance), the insurers expect that the ultimate goal of rehabilitation will be to return to work, whereas workers often see the goal as a reduction of pain. In this context, the insurers’ goal cannot be changed, but intermediate goals that both parties can agree on can be identified through SDM. In fact, our previous studies show that other important decisions can be made, such as how to conduct the process of a gradual return to work [13, 51]. Thus SDM should focus on specific options related to a safe return to work. Moreover, it is important in its implementation to remind all parties that the option of keeping the status quo, i.e. not participating in the work rehabilitation program, should always be offered, even though this could mean a loss of wage replacement for the worker, since he or she will have to decline receiving treatment. The goal is that the worker makes an informed decision. We have seen in continuing education training activities that health care professionals sometimes misunderstand the working alliance and the client-centered approaches, and prematurely terminate treatment without informed consent when the worker mentions seeking another treatment or another objective. A fourth observation relates to the association of SDM with the patient-centered approach, which can lead to confusion and become an obstacle to full implementation of SDM in clinical settings. When we first approached the collaborating rehabilitation facilities for the purposes of this study, we were told that SDM already takes place in most of them. The evaluation report of the UK’s Making Good Decisions in Collaboration (MAGIC) program also mentions that health care professionals assumed they were already performing SDM [52]. In reality, they were not, strictly speaking, practising SDM, which is a structured and systematized process rather than an overall ‘‘patient-centered’’ approach. Our SDM rehabilitation model clearly presents the objectives, activities, resources and indicators needed to operationalize it. During implementations using our SDM rehabilitation model, for example, it will be possible to assess the level of attainment of each objective using its related indicators. Currently the evaluation of the MAGIC program assesses the perception of having the skills to perform SDM [52], which may be different from actually having the skills. Our SDM rehabilitation model, designed and operationalized by expert health care

123

150

professionals and based on evidence-based knowledge from both the SDM and work rehabilitation fields, should enable the implementation and accurate assessment of SDM skills and programs in work rehabilitation.

Strengths and Limitations We consulted the literature to design a preliminary SDM rehabilitation model to provide a strong basis for the consensus group. This constitutes a partial triangulation of the data with two different methods (literature and experts), which should increase the credibility of the results [53]. However, our literature search was not a systematic review of the literature. We reviewed many research fields (shared decision making, chronic pain, work rehabilitation, working alliance, and goal setting), and within each field a variety of key terms are used for similar concepts. This could compromise the feasibility of our operationalized model. The quality of the information gathered using the TRIAGE method depends on the representativeness and the credibility of the participants. One strength is that we included professionals who were clinical coordinators working in rehabilitation in our consensus group sessions for the development of the SDM rehabilitation model. Including administrators meant we could design and operationalize a model that takes administrative barriers into account, and which will help us plan the implementation of SDM rehabilitation programs. However, all our experts came from public rehabilitation centres, which mean that our model might need further adaptation for private rehabilitation settings.

Conclusion Using a series of three group consensus sessions, we were able to design and operationalize an SDM rehabilitation model applicable to the realities of work rehabilitation for workers suffering from persistent pain secondary to a musculoskeletal injury. The model includes the essential concepts of an SDM process but is now adapted for the rehabilitation field [54]. These adaptations include the building of a working alliance and formalized goal setting. Our detailed description of the development of our SDM rehabilitation model should facilitate behaviour change implementation and scientific replication. Prior to the implementation of SDM rehabilitation programs, the next step will be to document the feasibility of our operationalized model among other important stakeholders (employer, union, insurer and workers) taking part in return-to-work decisions.

123

J Occup Rehabil (2015) 25:141–152 Conflict of interest declare.

The authors have no conflicts of interest to

References 1. Statistics Canada. Labour force characteristics, seasonally adjusted, by province (monthly). Government of Canada, Canada. 2014. http://www.statcan.gc.ca/tables-tableaux/sum-som/l01/cst01/ lfss01c-eng.htm. Accessed April 2014. 2. Travail GM. In: Bonte P, Izard M, editors. Dictionnaire de l’ethnologie et de l’anthropologie (Dictionary of ethnology and anthropology). Paris: Quadrige/Presses Universitaires de France; 2000. p. 717–20. 3. Leufstadius C, Eklund M, Erlandsson LK. Meaningfulness in work—experiences among employed individuals with persistent mental illness. Work. 2009;34(1):21–32. doi:10.3233/WOR2009-0899. 4. Ve´zina M, Cloutier E, Stock S, Lippel K, Fortin E´, Delisle A, et al. Enqueˆte que´be´coise sur des conditions de travail d’emploi et de sante´ et de se´curite´ du travail (Quebec survey on working and employment conditions and occupational health and safety) (EQCOTESST). Montre´al, Que´bec, Canada: Institut de recherche Robert-Sauve´ en Sante´ et en Se´curite´ du Travail; 2011. 5. Wynne-Jones G, Cowen J, Jordan JL, Uthman O, Main CJ, Glozier N, et al. Absence from work and return to work in people with back pain: a systematic review and meta-analysis. Occup Environ Med. 2014;71(6):448–56. doi:10.1136/oemed-2013-101571. 6. OECD. Sickness, Disability and work: breaking the Barriers. A synthesis of findings across OECD countries. OECD; 2010. 7. Henderson M, Glozier N, Elliott KH. Long term sickness absence. BMJ. 2005;330:802–3. doi:10.1136/bmj.330.7495.802. 8. Hlobil H, Staal JB, Spoelstra M, Ariens GA, Smid T, van Mechelen W. Effectiveness of a return-to-work intervention for subacute lowback pain. Scand J Work Environ Health. 2005;31(4):249–57. 9. Karjalainen K, Malmivaara A, van Tulder M, Roine R, Jauhiainen M, Hurri H et al. Multidisciplinary biopsychosocial rehabilitation for neck and shoulder pain among working age adults. CLIB. 2006;4. doi:10.1002/14651858.CD002194. 10. Schonstein E, Kenny DT. The value of functional and work place assessments in achieving a timely return to work for workers with back pain. Work. 2001;16(1):31–8. 11. Franche R-L, Cullen K, Clarke J, Irvin E, Sinclair S, Frank J, et al. Workplace-based return-to-work interventions: a systematic review of the quantitative literature. J Occup Rehabil. 2005;15(4):607–31. 12. Loisel P, Lemaire J, Poitras S, Durand MJ, Champagne F, Stock S, et al. Cost-benefit and cost-effectiveness analysis of the Sherbrooke model of back pain management: a six-year followup study. Occup Environ Med. 2002;59:807–15. 13. Coutu MF, Baril R, Durand MJ, Coˆte´ D, Cadieux G. Clinicianpatient agreement about the work disability problem of patients having persistent pain: Why it matters. J Occup Rehabil. 2013;23(1):82–92. doi:10.1007/s10926-012-9387-8. 14. Chou R, Loeser JD, Owens DK, Rosenquist RW, Atlas SJ, Baisden J, et al. Interventional therapies, surgery, and interdisciplinary rehabilitation for low back pain. Spine. 2009;34(10): 1066–77. doi:10.1097/BRS.0b013e3181a1390d. 15. Charles C, Whelan T, Gafni A. What do we mean by partnership in making decisions about treatment? BMJ. 1999;319(7212):780–2. 16. Makoul G, Clayman ML. An integrative model of shared decision making in medical encounters. Patient Educ Couns. 2006;60(3): 301–12. doi:10.1016/j.pec.2005.06.010. 17. Towle A, Godolphin W. Framework for teaching and learning informed shared decision making. BMJ. 1999;319:766–71. 18. O’Connor AM, Tugwell P, Wells G, Elmslie T, Jolly E, Hollingsworth G, et al. A decision aid for women considering

J Occup Rehabil (2015) 25:141–152

19.

20.

21.

22.

23.

24. 25.

26.

27.

28. 29.

30.

31.

32.

33.

34.

hormone therapy after menopause: decision support framework and evaluation. Patient Educ Couns. 1998;33:267–79. doi:10. 1016/S0738-3991(98)00026-3. Le´gare´ F, O’Connor AM, Graham ID, Saucier D, Coˆte´ L, Blais J, et al. Primary health care professionals’ views on barriers and facilitators to the implementation of the Ottawa decision support framework in practice. Patient Educ Couns. 2006;63:380–90. doi:10.1016/j.pec.2006.04.011. Parsons A, Harding G, Breen A, Foster N, Pincu T, Vogel S, et al. Will shared decision making between patients with chronic musculoskeletal pain and physiotherapists, osteopaths and chiropractors improve patient care? Fam Pract. 2012;29(29):203–12. doi:10.1093/fampra/cmr083. Graham ID, Logan J, Harrison MB, Straus SE, Tetroe J, Caswell W, et al. Lost in knowledge translation: time for a map? J Contin Educ Health Prof. 2006;26(1):13–24. doi:10.1002/chp.47. Rossi PH, Lipsey MW, Freeman HE. Expressing and assessing program theory. Evaluation: a systematic approach. 7th ed. Sage Publications; 2004. p. 133–68. Michie S, Fixsen DL, Grimshaw J, Eccles M. Specifying and reporting complex behaviour change interventions: the need for a scientific method. Implement Sci. 2009;4(40):1–6. doi:10.1186/ 1748-5908-4-40. Chen HT. Issues in constructing program theory. New Dir Progr Eval. 1990;47:7–18. doi:10.1002/ev.1551. Champagne F, Brousselle A, Hartz Z, Contandriopoulos AP. Mode´liser les interventions (Modelling interventions). In: Brousselle A, Champagne F, Contandriopoulos A-P, Hartz Z, editors. L’e´valuation: concepts et me´thodes (Evaluation: concepts and methods). Montreal, Quebec, Canada: Les Presses de l’Universite´ de Montre´al; 2009. p. 57–70. Dubois CA, Champagne F, Bilodeau H. Historique de l’e´valuation (History of evaluation). In: Brousselle A, Champagne F, Contandriopoulos A-P, Hartz Z, editors. L’e´valuation: concepts et me´thodes (Evaluation: concepts and methods). Montreal, Quebec: Les Presses de l’Universite´ de Montre´al; 2009. p. 13–34. Contandriopoulos AP, Champagne F, Denis JL, Avargues MC. L’e´valuation dans le domaine de la sante´: concepts et me´thodes (Evaluation in the health field: concepts and methods). Rev Epidemiol Sante Publique. 2000;48(6):517–24. Creswell JW. Research design: qualitative, quantitative, and mixed methods approaches. 2nd ed. Thousand Oaks, CA: Sage; 2003. Rossi PH, Freeman HE, Lipsey MW. Expressing and assessing program theory. Evaluation: a systematic approach. 6th ed. Beverly Hill: Sage; 1999. p. 155–88. Gervais M, Pe´pin G, Carrie`re M. Triage ou comment adapter une technique de recherche a` l’intervention clinique en ergothe´rapie (Triage or how to adapt a research technique to clinical intervention in occupational therapy). Revue Que´be´coise d’Ergothe´rapie. 2000;9:11–5. Demers L, Poissant L. Connecting with clinicians: opportunities to strengthen rehabilitation research. Disabil Rehabil. 2009;31(2): 152–9. doi:10.1080/09638280802037936. Coutu MF, Le´gare´ F, Durand MJ, Corbie`re M, Stacey D, Loisel P et al. Fostering shared decision making by occupational therapists and workers involved in accidents resulting in persistent musculoskeletal disorders: a study protocol. Implement Sci. 2011; 6(22):1–8. doi:10.1186/1748-5908-6-22 Loisel P, Abenhaim L, Durand P, Esdaile JM, Suissa S, Gosselin L, et al. A population-based, randomized clinical trial on back pain management. Spine. 1997;22(24):2911–8. Geoffrion P. Le groupe de discussion (Group discussion). In: Gauthier B, editor. Recherche sociale de la proble´matique a` la collecte de donne´es (Social research: from the problem to data collection). 5th ed. Sainte-Foy, Quebec, Canada: Presses de l’Universite´ du Que´bec; 2009. p. 391–414.

151 35. Durand MJ, Vachon B, Loisel P, Berthelette D. Constructing the program impact theory for an evidence-based work rehabilitation program for workers with low back pain. Work. 2003;21(3): 233–42. 36. Corbie`re M, Durand MJ, St-Arnaud L, Briand C, Goulet C, Charpentier N et al. Return to work obstacles and coping efficacy scale—musculoskeletal disorders (ROCES-MSD). 2009. 37. Albert V, Coutu MF, Durand MJ. Adaptation d’un questionnaire visant a` mesurer les repre´sentations lie´es a` l’incapacite´ de travail (Adaptation of a questionnaire designed to measure representations related to work disability). L’Ence´phale. 2013;39(3): 174–82. doi:10.1016/j.encep.2012.05.002. 38. Albert V, Coutu MF, Durand MJ. Internal consistency and construct validity of the revised illness perception questionnaire adapted for work disability following a musculoskeletal disorder. Disabil Rehabil. 2013;35(7):557–65. doi:10.3109/09638288. 2012.702849. 39. Elwyn G, Hutchings H, Edwards A, Rapport F, Wensing M, Cheung W-Y, et al. The OPTION scale: measuring the extent that clinicians involve patients in decision-making tasks. Health Expect. 2005;8:34–42. doi:10.1111/j.1369-7625.2004.00311.x. 40. Elwyn G, Edwards A, Wensing M, Grol R. Shared decision making: measurement using the OPTION instrument. Wales, UK: Cardiff University; 2005. 41. O’Connor AM, Stacey D, Jacobsen MJ. The Ottawa personal decision guide for people facing tough health or social decisions. Ottawa, Ontario, Canada: Ottawa Hospital Research Institute and University of Ottawa; 2012. 42. Legare F, Stacey D, Pouliot S, Gauvin FP, Desroches S, Kryworuchko J, et al. Interprofessionalism and shared decisionmaking in primary care: a stepwise approach towards a new model. J Interprof Care. 2011;25(1):18–25. doi:10.3109/ 13561820.2010.490502. 43. Le´gare´ F, Stacey D, Gagnon S, Dumm S, Pluye P, Frosch D, et al. Validating a conceptual model for an interprofessional approach to shared decision-making: a mixed methods study. J Eval Clin Pract. 2011;17(4):554–64. 44. Stacey D, Le´gare´ F, Pouliot S, Kryworuchko J, Dunn S. Shared decision making models to inform an interprofessional perspective on decision making: a theory analysis. Patient Educ Couns. 2010;80(2):164–72. doi:10.1016/j.pec.2009.10.015. 45. St-Arnaud Y. L’interaction professionnelle: efficacite´ et coope´ration (professional interaction: effectiveness and cooperation). 2nd ed. Collection intervenir. Montreal: Les Presses de l’Universite´ de Montre´al; 2003. 46. Matthias MS, Salyers MP, Frankel RM. Re-thinking shared decision-making: context matters. Patient Educ Couns. 2013;91(2): 176–9. doi:10.1016/j.pec.2013.01.006. 47. Evans L, Hardy L. Injury rehabilitation: a goal-setting intervention study. Res Q Exerc Sport. 2002;73(3):310–9. doi:10.1080/ 02701367.2002.10609025. 48. Levack WMM, Dean SG, Siegert RJ, McPherson KM. Purposes and mechanisms of goal planning in rehabilitation: the need for a critical distinction. Disabil Rehabil. 2006;28(12):741–9. doi:10. 1080/09638280500265961. 49. Randell KE, McEwen IR. Writing patient-centered functional goals. Phys Ther. 2000;80(12):1197–203. 50. Theodorakis Y, Beneca A, Malliou P, Goudas M. Examining psychological factors during injury rehabilitation. J Sport Rehabil. 1997;6:355–63. 51. Loisel P, Durand MJ, Baril R, Langley A, Falardeau M. De´cider pour faciliter le retour au travail: E´tude exploratoire sur les dimensions de la prise de de´cision dans une e´quipe interdisciplinaire de re´adaptation au travail (decision-making to facilitate a return to work—exploratory study of the aspects of decisionmaking in an interdisciplinary work rehabilitation team).

123

152

52. 53.

54.

55.

56.

57.

58.

59.

J Occup Rehabil (2015) 25:141–152 Montreal, Quebec: Institut de recherche Robert-Sauve´ en sante´ et se´curite´ au travail; 2004. King E, Taylor J, Williams RM, Vanson T. The MAGIC programme; evaluation. London, UK: The Health Foundation; 2013. Maggs-Rapport F. Combining methodological approaches in research: ethnography and interpretive phenomenology. J Adv Nurs. 2000;31(1):219–25. doi:10.1046/j.1365-2648.2000.01243.x. O’Connor AM, Jacobsen MJ. Ottawa decision support tutorial: training practitioners in decision support. Ottawa Health Research Institute. 2007. https://decisionaid.ohri.ca/ODST/. Accessed July 2007. Elwyn G, Edwards A, Wensing M, Hood K, Atwell C, Grol R. Shared decision making: developing the OPTION scale for measuring patient involvement. QSHC. 2003;12:93–9. doi:10. 1136/qhc.12.2.93. Corbie`re M, Bisson J, Lauzon S, Ricard N. Factorial validation of a French short-form of the working alliance inventory. Int J Meth Psych Res. 2006;15(1):36–45. doi:10.1002/mpr.27. Amundson NE. Mattering: a foundation for employment counseling and training. J Employ Couns. 1993;30:146–52. doi:10. 1002/j.2161-1920.1993.tb00173.x. Corbie`re M, Amundson NE. Perceptions of the ways of mattering by people with mental illness registered in supported employment programs. Career Dev Q. 2007;56(2):141–9. doi:10.1002/j.21610045.2007.tb00026.x. Broadbent E, Petriea KJ, Maina J, Weinman J. The Brief Illness Perception Questionnaire. J Psychosom Res. 2006;60(6):631–7. doi:10.1016/j.jpsychores.2005.10.020.

123

60. O’Connor AM. Validation of a decisional conflict scale. Med Decis Making. 1995;15:25–30. doi:10.1177/0272989X95015 00105. 61. O’Connor AM, Jacobsen MJ, Stacey D. An evidence-based approach to managing women’s decisional conflict. J Obstet Gynecol Neonatal Nurs. 2002;31:570–81. doi:10.1111/j.15526909.2002.tb00083.x. 62. Durand MJ. E´tude des effets du Retour The´rapeutique au Travail chez des travailleurs absents du travail suite a` une le´sion professionnelle au dos (study of the effects of a therapeutic return to work in workers on sick leave following a work-related back injury). Sherbrooke, Quebec, Canada: Universite´ de Sherbrooke; 1996. 63. Brehaut JC, O’Connor A, Wood TJ, Hack TF, Siminoff L, Gordon E, et al. Validation of a decision regret scale. Med Decis Making. 2003;23(4):281–92. doi:10.1177/0272989X03256005. 64. Tercyak KP, Johnson SB, Roberts SF, Cruz AC. Psychological response to prenatal genetic counseling and amniocentesis. Patient Educ Couns. 2001;43(1):73–84. doi:10.1016/S0738-3991 (00)00146-4. 65. Durand MJ, Berthelette D, Loisel P, Beaudet J, Imbeau D. Travailleurs de la construction ayant une dorso-lombalgie: e´valuation de l’implantation de la the´orie d’un programme de collaboration pre´coce en re´adaptation au travail (construction workers suffering from lumbar back pain: evaluation of the implementation of the theory of an early collaboration occupational rehabilitation program). Montreal, Quebec, Canada: Institut de recherche Robert Sauve´ en sante´ se´curite´ du travail; 2007.