Abstract
Background and Purpose There is growing awareness that the poor uptake of evidence in health care is not a knowledge-transfer problem but rather one of knowledge production. This issue calls for re-examination of the evidence produced and assumptions that underpin existing knowledge-to-action (KTA) activities. Accordingly, it has been advocated that KTA studies should treat research knowledge and local practical knowledge with analytical impartiality. The purpose of this case report is to illustrate the complexities in an evidence-informed improvement process of organized stroke care in a local rehabilitation setting.
Case Description A participatory action approach was used to co-create knowledge and engage local therapists in a 2-way knowledge translation and multidirectional learning process. Evidence regarding rehabilitation stroke units was applied in a straightforward manner, as the setting met the criteria articulated in stroke unit reviews. Evidence on early supported discharge (ESD) could not be directly applied because of differences in target group and implementation environment between the local and reviewed settings. Early supported discharge was tailored to the needs of patients severely affected by stroke admitted to the local rehabilitation stroke unit by combining clinical and home rehabilitation (CCHR).
Outcomes Local therapists welcomed CCHR because it helped them make their task-specific training truly context specific. Key barriers to implementation were travel time, logistical problems, partitioning walls between financing streams, and legislative procedures.
Discussion Improving local settings with available evidence is not a straightforward application process but rather a matter of searching, logical reasoning, and creatively working with heterogeneous knowledge sources in partnership with different stakeholders. Multiple organizational levels need to be addressed rather than focusing on therapists as sole site of change.
Physical therapy subscribes to the ideal of evidence-based practice, but how do we achieve that ideal? The health care sector, including physical therapy, has difficulty in applying and achieving widespread uptake of evidence in local practices.1–4 Despite the ever greater technologies available for critically assembling, appraising, and synthesizing the relevant studies of interest on any given topic, therapists still experience difficulties when applying the results of systematic reviews and related guidelines to their routine daily practice.2 They complain about the poor quality in terms of relevance and fit of the evidence produced by researchers. Researchers, in turn, point to therapists who insist on doing it in their own experience-based way. There is growing recognition that the research-practice gap thus experienced is really a problem of knowledge production rather than one of knowledge transfer.5–7 Research goes unused because researchers fail to address the most important problems facing patients, therapists, and decision makers.8
The research-practice gap is increasingly articulated in terms of concern about the relative neglect of external validity when it comes to clinical epidemiological research. It is a sine qua non that researchers need to eliminate the possibility of bias in order to provide high-quality evidence. The focus of the research community is on the methodological quality of the clinical trials rather than on the usefulness of their results.1,8–11 A drawback that ensues from the stringent rating of research proposals and publications in terms of the rigor of their internal validity control is the creation of a rather “sterile” evidence base.1 Because clinical trials require such tight criteria when it comes to the interventions and selection of participants, they are likely to end up not having a very high external validity.1 Much more space, therefore, should be devoted to issues of external validity,1,8–11 engaged scholarship,5,6 and the conduct of knowledge translation work in a disinterested way.12
This case report illustrates the complexities involved in an evidence-informed improvement process in a local rehabilitation stroke unit. It is an integrated knowledge-to-action (KTA) process that draws on a sociologically informed way of considering what sciences are and do.12,13 In such a conception of science, research knowledge is not privileged over practical knowledge.12–14 Translation is conceived in our KTA process as a 2-way knowledge exchange between knowledge producers and users.6,8,12 Both forms of knowledge are thereby considered as being distinct but equally valued knowledge practices.
Case Description and Methods
Target Setting
The target setting was the stroke unit of a Dutch rehabilitation center to which patients with multifaceted problems are admitted for multidisciplinary rehabilitation after discharge from an acute stroke unit. The multidisciplinary team consists of physiatrists, physical therapists, occupational therapists, nutritionists, speech therapists, psychologists, social workers, and nursing staff specializing in stroke rehabilitation. A medical and team manager coordinate the multidisciplinary team's work.
Identified Problem
This KTA project started with a request from the head of the department to improve the local service with the best evidence available in the whole chain of care. The request stemmed from his concern that home- and community-based rehabilitation were less organized in the Netherlands compared with, for instance, Scandinavian countries. Despite (or because of) the wealth of systematic reviews and related guideline recommendations on the organization of stroke services, the local therapists experienced difficulties in critically appraising the research knowledge and guideline documents in terms of their usefulness for their local setting.15 Together with the multidisciplinary team, we performed an integrated KTA project on organized stroke care by co-creating knowledge that fitted the local circumstances.
Methods
A fieldwork approach including participatory action research was used to assist the local therapists in KTA activities regarding organized stroke care. In such an approach, active interaction between researchers and those researched influences the learning processes and the self-reflective capacity of both parties.16 An integrated KTA framework guided the study and involved researchers, therapists, and other stakeholders in a 2-way knowledge translation and multidirectional learning process (see Tab. 1 for an overview of phases, stakeholders, tools, and products).17 A mixture of research tools was used to bridge the research-practice gap, comprising literature study, individual and focus group interviews, a pilot study, and expert meetings. All individual and focus group interviews were digitally recorded, and files were verbatim transcribed. Data were subsequently coded and compared with regard to facilitators/barriers and advantages/disadvantages of the service to be put into action. The Atlas-ti software program (Atlas-ti GmbH, Berlin, Germany) was used for qualitative data analysis.18
Two-Way Knowledge Exchange and Multidirectional Learning Processa
A project team was appointed, consisting of the head of the department, the medical manager of the stroke unit, the participatory researcher, and a senior researcher. A human movement scientist, who also worked as a physical therapist in the stroke rehabilitation team, fulfilled both the role of participatory researcher and that of content and setting expert. A philosopher of science and medicine (senior researcher), who was not involved in local practice, fulfilled the role of reflective questioner and supported the participatory researcher in the conceptualization, analysis, triangulation, and interpretation of the data. Ongoing collaboration took place with the multidisciplinary team and their team manager in terms of consultation, information, and feedback meetings. The Board of Directors of University Medical Center Groningen and health care insurance experts were consulted to answer questions about financial, legislation, and regulatory matters.
Case Report
Phase 1: Specify Problem
In this phase, the participatory and senior researchers specified the problem of the local team by examining literature about the research-practice gap in general, and more specifically about organized stroke care. They identified the problem not just as a lack of knowledge of local therapists but also as a knowledge production problem. The evidence for well-organized stroke care was widely acknowledged.15,19 However, trials had been undertaken in several countries in various settings, using different names to distinguish well-organized from less-organized ones and selecting different target groups with respect to severity due to stroke.15 This variability in naming, setting, and target population complicated the translation of the body of evidence summarized in numerous reviews into local improvements. Thus, the poor uptake of evidence on organized stroke care in local practices was attributed to the great diversity in stroke services examined in clinical trials.15,20,21
Phase 2: Knowledge Creation
In this phase, the researchers searched for reviews summarizing the evidence produced in clinical trials on organized stroke care and stroke rehabilitation. Two guidelines on stroke management were used as additional knowledge sources.20,21
Review of reviews.
PubMed, CINAHL, and The Cochrane Library revealed a great variety of reviews on stroke rehabilitation and organized stroke care services. The researchers selected a total of 16 reviews between January 2003 and September 2013 that were relevant to the identified problem. These reviews had their focus on the organization of stroke rehabilitation in primary and secondary care.22–37 Reviews with a focus on diagnostic procedures and rehabilitation treatments were excluded.
Making research knowledge manageable.
The project team asked the researchers to make the diversity of stroke services and related evidence manageable in a user-friendly tool. Informed by distinctions made in the reviews, the researchers ordered the research knowledge in a matrix with 4 quadrants (Figure).22–38 They distinguished services in well- and less-organized services and inpatient and outpatient services (see Appendix for definitions of services).22,26–31 In addition, they differentiated stroke severity as mild (a), moderate (b), or severe disability (c) and evidence as firm (+), inconclusive (+/−), or no evidence (−).38 In so doing, they reflected the emerging hierarchy of service organizations and target group descriptions articulated by “stroke unit trialists.”22
Ordering of well- and less-organized and inpatient and outpatient services. a=mild disability, b=moderate disability, c=severe disability, +=evidence, +/−=inconclusive evidence, −=no evidence.
Phase 3: Tailor Knowledge to Address Identified Problem
In this phase, the project team applied the matrix as an ordering figure to: (1) determine the nature of the local rehabilitation service, (2) explicate the available evidence for the local service, and (3) pinpoint research knowledge that could be used locally for improving home- and community-based services. To address these issues, the researchers explored differences and similarities in target population, content, and implementation environment between the services reviewed and the local service under study.
A rehabilitation stroke unit: a firm evidence base to start with.
The project team determined their local service as an evidence-based rehabilitation stroke unit (quadrant I), as their service met the 4 criteria set out in stroke unit reviews.22–25 First, there was a multidisciplinary team that provided stroke care in a dedicated ward; second, the stroke team (including nursing staff) was specialized and trained in stroke rehabilitation; third, routine involvement of caregivers in the rehabilitation process was established in the form of partner groups; and, fourth, new insights and skills in stroke rehabilitation were taught in regular meetings and courses.22–25 Furthermore, the target population of the local service were adult patients with moderate to severe disability who had been admitted to a dedicated department after care in the acute stroke unit of a hospital. The multifaceted problems meant that rehabilitation treatment could not adequately and safely be managed at home. This approach was in line with stroke guidelines recommending that all patients not suitable for transfer home after completion of acute diagnosis and treatment should be treated in a specialist rehabilitation stroke unit.20,21 Research knowledge about rehabilitation stroke units thus served as a firm evidence base (level A) for the target group in the local setting.
Early supported discharge (ESD): a challenge to translate the evidence.
Evidence concerning home and community rehabilitation services was inconclusive because of variability across services (quadrant IV).28–30 The project team endorsed ESD as a well-organized and evidence-based cost-saving outpatient alternative to regularly supported discharge (quadrant III).31–37 Early supported discharge involves a multidisciplinary team specialized in stroke rehabilitation that plans and coordinates early discharge from the hospital, postdischarge care, and rehabilitation at home through weekly team meetings, usually followed by a visit from the case manager of the ESD team.31 Stroke guidelines recommend that patients should be discharged early from stroke unit care only if there is a specialist stroke rehabilitation team able to continue rehabilitation in the community from the day of transfer.20,21
The project team noted that well-organized, specialist home- and community-based stroke services were scarce in the region. Moreover, the greatest benefits from ESD are attributed to patients with mild to moderate disability in acute stroke units, whereas the target group in their local setting are patients with moderate to severe disability treated in a rehabilitation stroke unit.31–36 They concluded that the original evidence on ESD did not exactly fit the circumstances of the target setting. Even so, they argued that the local rehabilitation stroke unit could well profit from the ESD recommendation to shift more—and at an earlier stage—into home-based alternatives.31–33,35,36 In light of these findings, the idea of an alternative service tailored to patients with moderate and severe disability following stroke surfaced; that is, combined clinical home rehabilitation (CCHR), in which specialist multidisciplinary treatment in the rehabilitation unit could be combined with home treatment at an early stage. The participatory researcher used the matrix to inform the multidisciplinary team about: (1) how to build on the evidence available for stroke rehabilitation units and (2) how to learn from evidence for ESD.
Phase 4: Adapt Knowledge to Local Context
In this phase, the researchers traced and articulated ways of attending to clinical and home rehabilitation of the local team. Therefore, they conducted 2 focus group discussions, each with 8 participants from different disciplines of the multidisciplinary team. The focus groups addressed the question: “How could one learn the most from the advantages or disadvantages of combined clinical and home rehabilitation.” To elaborate on the practical contextual knowledge of the team, the researchers subsequently conducted a second literature review focused on items that the participants addressed as challenging.
Practical knowledge of the team.
The participants articulated the greatest advantage of CCHR as being an easier carryover of skills learned in the clinical setting to the home setting. They discussed the carryover problem in terms of “a gap, into which patients and their families threaten to fall after discharge from the rehabilitation unit.” Patients faced the problem of being treated in an institution and then having to master an environment that differed from this institution in various ways. This problem could not be prevented by means of weekend leaves and routine involvement of partners in the rehabilitation process, or even by home visits. Therapists critically reflected on the “artificial state” of their treatments by stressing that activities training in a clinical setting was always an imperfect simulation of the natural setting at home. Therefore, they welcomed CCHR because it made training more “tailor-made” and task- and context-specific. Patient goals could be better set in collaboration with patients and significant others at home, thereby taking problems in real life as a directive for treatment. The added value of CCHR was expressed in a better transition to the home setting rather than in cost savings such as in ESD.
One disadvantage discussed by the participants was the distance that needed to be bridged in order to deliver treatment at home. Travel time made them less employable in the clinical setting, and they doubted whether health insurance companies were willing to pay for such indirect costs. Asking community therapists to deliver home treatment was not an alternative because they lacked specialist expertise in stroke rehabilitation. Fatigue of patients with stroke was a concern that also needed to be considered with respect to travel time. Also, although family members could be actively involved in CCHR, they might not always be able to free themselves from work obligations during home treatment.
Research knowledge from the literature.
To build on the practical knowledge of the local team, the researchers conducted a literature review on task- and context-specific training and goal setting. This review confirmed and specified the practical knowledge of the team. Indeed, most patients with stroke had difficulty with generalizing what had been learned in the therapeutic setting to their own living environments.39 The context in which tasks are learned, therefore, should be as personalized and home-like as possible. Task-specific training enhanced functional outcomes in stroke rehabilitation, where there is increasing evidence available for neural plastic changes.40 Intensive, meaningful, repetitive task-specific and client-centered treatments that targeted the whole body were advocated in an enriched environment.40,41 Goal setting in close collaboration with patients and their families was considered a key element in rehabilitation treatment.42 Goals should be meaningful and challenging but achievable and should include both short-term (days/weeks) and long-term (weeks/months) targets and time-bound measurable outcomes.42 Qualitative studies on the actual use of goals in rehabilitation practice revealed, however, that goals are frequently unattained, modified, or contested.43,44 In rehabilitation, goals are set for the home setting that differ—in terms of spatial and social characteristics—from those in the clinical setting, where people are training for the accomplishment of goals.44 The research knowledge on task- and context-specific training and goal setting supported the practical knowledge and drives of the therapists and the uptake of CCHR. The therapists explicated these findings in a feedback meeting with the team.
Phase 5: Assess Barriers to Knowledge Use
In this phase, the project team decided in close cooperation with the team to conduct a pilot study with 4 patients with varying cognitive and motor problems poststroke (Tab. 2). Physical therapists and occupational therapists were indicated as best candidates for delivery of CCHR. The participatory researcher assisted the therapists in making CCHR practical. After the pilot study, the researchers assessed facilitators and barriers in the application of CCHR by conducting semistructured individual interviews with 9 therapists and the 4 patients and their caregivers involved in the pilot study. The head of the department and the senior researcher subsequently talked with members of the Board of Directors of University Medical Center Groningen and with regional health insurance experts about the potential of CCHR.
Patient Characteristicsa
Specification of CCHR.
Informed by ESD evidence, the participatory researcher decided in partnership with the therapists to begin with CCHR as early as possible, that is, within 4 weeks after admission to the rehabilitation stroke unit. The literature studies on task- and context-specific training and goal setting made them decide that goals needed to be set in the home setting in close collaboration with the patients and their caregivers. Whole tasks should subsequently be assessed and trained in the home setting and then problematic tasks repeatedly practiced in the clinical setting with a focus on the missing components. Subsequently, the whole tasks had to be trained in the home setting again.
Experiences of therapists and patients and their caregivers.
Therapists embraced the alternative: CCHR provided a great deal of insight into the home environment, including the problems their patients had to face. Although they experienced difficulties with restructuring their existing skills, they felt they could offer their patients a more customized treatment. According to one therapist, “CCHR was a lottery ticket for a patient with severe cognitive problems.” This patient showed astonishing improvements when training took place in his own living environment. Patients were satisfied, because they were taught to use, in their own home environment, what they had learned in the clinical setting. Home treatments made family members feel more secure in assisting patients in performing daily activities in the absence of the therapist during weekend leaves.
Most of the goals that were set at home, and subsequently trained for in the clinical setting, were attained. A frequency of two 45-minute home-treatment sessions a week appeared to be enough for goal setting and treatment. The total duration of CCHR ranged from 2 to 5 weeks, in which the total number of home treatments ranged from 2 to 10 sessions, depending on the complexity of problems poststroke. However, starting early (ie, within 4 weeks) became a barrier for some patients and their caregivers. For logistical reasons, the team could not select more than 2 patients for home treatment at the same time; otherwise, too few therapists would then be available for regular treatment in the clinic. Therapists also experienced time pressure in their schedules caused by additional travel time and time to fine-tune activities. As a solution, home treatments were connected to weekend leaves.
Expert consultation.
A barrier beyond the control of the therapists involved the organization and finance system of health care services in the Netherlands. The consulted board of directors and health insurance experts foresaw financial and legislative problems were it to be implemented structurally. Dutch inpatient and outpatient therapists work in separate organizational environments with different employment relationships and financing agencies. They suggested involving care financing agencies in the project because legislative procedures and partitioning walls between health-financing systems in primary and secondary care might hamper structural implementation of CCHR. This suggestion implied that the project team had to redefine the problem and integrate practical and contextual knowledge of regional managers, therapists, and financial and policy experts in the local knowledge co-creation process. Several therapists were eager to give content to the new idea.
Phase 6: Redefine Identified Problem
In this phase, the researchers placed the problematic integration of primary and secondary care services in a broader perspective by organizing an expert meeting with 1 local care manager as well as 4 care managers, 1 health insurance expert, 1 general practitioner, and 2 policy makers in the region. Another expert meeting was organized with 6 physical therapists, 5 occupational therapists, and 2 speech therapists treating survivors of stroke in the region. Subsequently, the researchers conducted a qualitative study to gain knowledge of the needs of survivors of stroke and family caregivers in the clinical, postdischarge, and reintegration phases of the rehabilitation process.
Expert meetings.
The region had identified active and healthy aging as a major societal challenge. All participants were supportive of a better integration between primary and secondary stroke care services. To meet the demands associated with demographic change in terms of low birthrates, increasing longevity, the related shortage of health care professionals, and increasing pressure on public budgets, the policy makers felt the need to rethink the way stroke services were organized in the regional chain of care. Delivery of stroke care with less fragmentation, close to home, and bridging distances with the help of innovative technology were shared challenges. The physical therapists, occupational therapists, and speech therapists expressed a shortage of knowledge in outpatient services concerning the treatment and multifaceted problems of survivors of stroke as a major concern. The qualitative study revealed that survivors of stroke foregrounded functional recovery in the clinical phase, were confronted with an identity confusion and related mourning process in the postdischarge phase, and longed for recognition and a sense of belonging in the reintegration phase.45
Reshape problem.
Informed by this practical, contextual, and research knowledge, the project team suggested that individual coaching of survivors of stroke and family caregivers rather than multidisciplinary treatment should be pivotal in home- and community-based rehabilitation. The project team decided to reshape the identified therapeutic problem by placing it in a regional policy and multi-stakeholder perspective and then go through the knowledge creation and action cycle again. The therapeutic question “How can therapists improve the transition from the clinical to the home setting with the available evidence on organized stroke care?” was reshaped into a decision-making question: “How do we organize rehabilitation service through to the home environment of survivors of stroke in a sustainable, efficient, and smart way, eventually with the help of e-health technology?” Combined coaching at home (face-to-face) and over distance (screen-to-screen) was articulated as an alternative that might contribute to regional challenges. How the project team addressed multiple organization levels in a second KTA study on organized stroke care and how they managed that therapists remained involved will be presented in the near future.
Discussion
This case report illustrates the iterative and dynamic process of integrated KTA research aimed at translating evidence, produced in controlled research settings, to complicated implementation environments such as those seen in clinics. To bridge the research-practice gap experienced, we used a participative action approach, thereby treating research and practical contextual knowledge with analytical impartiality.5,12 The lessons learned by doing are discussed below.
Different Translations of Evidence to Local Setting
By articulating similarities and differences in the target population and environments of local and research settings, we learned that evidence produced in clinical trials could help to advance local settings in different ways. Similarities indicate proof of actual practice in local settings, implying that local therapists already worked in an evidence-based manner, as was the case in the local rehabilitation stroke unit described in this case report. Differences indicate that there is a challenge to improve local reality in an evidence-informed way, as we did with evidence for ESD. This finding might imply that therapists need to deviate from original evidence produced in research settings, as was done in CCHR, in order to tailor it to the abilities of patients who are severely disabled poststroke treated in a rehabilitation stroke unit. In most reviewed trials, severity was measured with the Barthel Index. Whether the Barthel Index is a valid instrument for objectifying the multi-facedness of problems of survivors of stroke in their own living environments might be a topic for future research.
Generalizability of Case Studies
Case studies and participatory action research do not have generalizing power, in that we can now recommend how other stroke rehabilitation services can improve their service in a well-organized and evidence-informed manner. This conclusion, however, does not mean that the findings in this KTA project only have local value. We hope to have demonstrated that detailed analysis and description of a bidirectional evidence-informed translation process in one location might have value for stakeholders working in another location. Indeed, such a specification of a local improvement process renders it possible for therapists in other settings to examine what will remain the same and what will change, when they want to improve the organization of their service in an evidence-informed manner. The user-friendly matrix that we developed as an ordering figure for navigating through the different types of services on organized stroke care including available evidence also might facilitate others in improving their local setting in an evidence-informed way.
Lack of Attention to Multiple Organizational Levels
One limitation of this case report is that we did not take broader organizational levels into account from the very beginning of our KTA process. We learned that improving rehabilitation practice with available evidence on organized stroke care is not just a matter of mobilizing its key users but also one of augmenting its material and organizational environment.7,8 Partitioning walls between different financing systems along with stifling legislation and regulatory procedures were detected as main barriers for structural implementation of CCHR. Thus, multiple organizational levels need to be addressed in KTA studies rather than focusing on therapists as the sole site of change.46 That is why we reshaped our identified problem and started to go through the knowledge creation funnel and action cycle again. The philosophy underlying, and methodology used, in transition management research may prove to be useful in such a multi-actor process.47
Transformed Terminology and KTA Framework
Translation was conceived in the KTA process as a 2-way knowledge exchange and multidirectional learning process for knowledge “producers” and “users.”5,6,12 Conventional KTA frameworks suggest a one-way knowledge transfer by the KTA expert, as is illustrated by the arrows of the action cycle pointing in just one direction.2,5 We welcome the bidirectional arrows of present frameworks that visualize the iterative, dynamic, and multidirectional learning process in KTA work more adequately.17 Knowledge translation experts may have been too focused on applied and evaluation research, which privileges scientific knowledge over local practical knowledge.12 As such, they might overlook the multiplicity of KTA work, in particular the production of new knowledge, which arises from the interactions between scientific and practical contextual knowledge. This situation implies that we need to reconsider the terminology in which current KTA frameworks articulate their knowledge creation and KTA activities. Also, the contradiction between knowledge “producers” and knowledge “users” might no longer be appropriate terminology. We argue that evidence-informed improvements of local practices are achieved more easily if the terminology remains equally valued and yet is fluid enough to change shape, to make it adaptable to the requirements of the new surroundings.
Improving rehabilitation practice is quite a different enterprise from proving its effectiveness.14 Improving settings with the help of research findings from a rather “sterile” evidence base1 and translating these findings into the complex world of health care delivery is a matter of creatively working with heterogeneous information sources.3,13,14 It is an iterative, cyclical, and dynamic translation and transformation process that requires practical and contextually relevant wisdom on the part of all involved, such as researchers, therapists, patients, caregivers, managers, and policy makers.
Appendix.
Glossary of Stroke Services Ordered in a Matrix
Footnotes
All authors provided concept/idea/project design. Ms Nanninga, Dr Postema, Ms van Twillert, and Dr Lettinga provided writing. Ms Nanninga provided data collection. Ms Nanninga, Ms van Twillert, and Dr Lettinga provided data analysis. Ms Nanninga and Dr Lettinga provided project management. Ms Nanninga provided fund procurement. Ms Nanninga and Dr Schönherr provided patients and institutional liaisons. Dr Schönherr provided facilities/equipment. Dr Schönherr, Ms van Twillert, and Dr Lettinga provided consultation (including review of manuscript before submission). The authors gratefully acknowledge Stichting Beatrixoord NN for providing financial support for the present work. Special thanks go to the patients and therapists of the multidisciplinary rehabilitation team participating in this project.
Project approval was obtained from the Medical Ethical Committee of University Medical Center Groningen (UMCG).
- Received October 27, 2013.
- Accepted November 25, 2014.
- © 2015 American Physical Therapy Association
References
Issue highlights
