| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Letters and Responses |
 | To the editor |
|---|
From our perspective, for research to become meaningful, it needs to go beyond consideration of complex systems to include both design and process. We believe that our approach: (1) acknowledges and evaluates complexity, (2) is consistent with client- and family-centered care, (3) brings clinicians and scientists together in collaboration, (4) bridges the gap between conducting research and incorporating it into practice, and (5) promotes embodiment of evidence-based practice. We acknowledge that this approach has challenges, ones that we believe can be overcome with enthusiastic collaboration.
In 2004, we published an article in Physiotherapy Canada that described a framework for planning and conducting what we refer to as "comprehensive rehabilitation outcomes research."4 We have found the International Classification of Functioning, Disability and Health5 (ICF) to be useful because it articulates the complex interactions among contextual factors such as personal factors and aspects of the environment (including, but not limited to, physical therapy intervention) with the integrity of body structure and function, activity, and participation in determining outcomes.
In our 2004 article, we described a process of working with a team comprising, at a minimum, clients, clinicians, and researchers to identify relevant outcomes and determinants of those outcomes for the selected client population, functional limitation, and participation goal of interest to the group.4 Administrators and policy makers also can be team members, potentially generalizing the results even more widely. After team members identify important outcomes and determinants of those outcomes, they decide how those constructs are related to each other. Research evidence, theory, and personal perspectives inform this step. By way of example, we described 2 conceptual models: one describing determinants of motor change for young children with cerebral palsy,6 which was informed—in part—by dynamic systems theory as it applies to early motor development, and the other describing factors associated with health-related quality of life in adults with chronic obstructive pulmonary disease. Importantly, the configuration of the factors, or constructs, that make up these complex systems is different in every conceptual model.
Once the conceptual model has been developed to all team members’ satisfaction, the next step is to identify indicators of the constructs.4 In the spirit of collaboration, we again recommend working with team members at this point. In the "determinants of motor change" work, we (Bartlett and Palisano7) obtained consensus from Canadian physical therapists across the province of Ontario about the indicators of the constructs in the conceptual model that they believed were most important in contributing to the acquisition of early motor abilities. For example, for the construct involving family ecology, they identified "family’s support to the child," "family’s expectations of the child," and "support to the family" as the key influences. Interestingly, therapists’ perceptions of important indicators were consistent with available research literature. The objective at this stage is not to identify all possible indicators, but rather the key indicators. Subsequent to this, we obtained feedback from parents, who provided useful feedback for revision of the model.
Following the identification of the key indicators for each construct, it is primarily the researcher’s responsibility to identify tests that provide reliable and valid measurements and that are acceptable to clients and clinically feasible to administer.4 Here is one of the main challenges of this work: identification of psychometrically sound measurement tools for all aspects in the model, as guided by the ICF. One of the delays in testing the model of determinants of motor change for children with cerebral palsy has been the need to step back, develop some measures (again, in collaboration with clients, families, and clinicians), and test them for reliability and validity. Just this past fall, pilot testing was conducted with children, parents, and physical therapists to ascertain the acceptability and feasibility of collecting the data on measures of indicators for the entire model.
Prior to collecting data, ethical approval and funding must be obtained. Here is a second major challenge. Sample sizes to conduct this type of work aiming to understand complex systems are large, and consequently significant research funding is required. This is not the type of research that granting agencies have traditionally supported. The Levels of Evidence Table produced by the Oxford Centre for Evidence-based Medicine8 was designed to evaluate the quality of reductionist research and, as such, places "outcomes research" at level 2c evidence (ie, in the lower half of the table in terms of strength of evidence). To help educate granting agencies and reviewers that "level 2c evidence" might be precisely the design option that best fits important clinical questions in rehabilitation, Bartlett and colleagues developed a position statement9 and a manuscript.10
Once the data have been collected, the full model is tested using analytical approaches within Structural Equation Modeling, identified to be ideally suited to work in the rehabilitation disciplines.11 A third challenge of this approach is that this analysis is not simple to conduct; most rehabilitation researchers require a statistical analyst or consultant to assist with this work, adding to the funding costs. Once the data have been analyzed, we believe that team members should again assemble to discuss the key interpretations and plan the dissemination pieces (both peer-reviewed and non-peer-reviewed).4 Information sharing is most effective when it meets the needs and concerns of target audiences. Short, easy-to-read summaries should be prepared in collaboration with specific target audiences such as families or service providers. Based on experience at the CanChild Centre for Childhood Disability Research at McMaster University, these types of dissemination pieces have more impact on clinical practice than full-length, peer-reviewed articles.
We believe that this type of complex systems approach, conducted in collaboration with clients and clinicians, has the potential to inform practice in 2 important ways. First, determinants of the selected outcomes that are not modifiable are useful for realistic goalsetting, thus contributing to the efficiency of services delivered. Second, determinants of the selected outcomes that are modifiable are potential targets for intervention, thus contributing to the effectiveness of services delivered. These determinants could be at all levels of the ICF, including aspects of the client that are not related to health condition and the physical, social, or attitudinal aspects of the environment, thus recognizing potential therapeutic opportunities beyond traditional physical therapist practice. We believe that research conducted in this way has the greatest opportunity to affect practice because all stakeholders have been engaged at every step, ensuring that meaningful factors have been selected as a target for study. Phenomena can be investigated in a holistic, rather than reductionistic, fashion that more closely matches the real worlds in which physical therapists practice. Like Collins, we would be interested in learning from others about how they have dealt with the issues of bridging the research-practice gap.
Associate Professor
School of Physical Therapy
1588 Elborn College
Faculty of Health Sciences
The University of Western Ontario, London, Ontario, Canada N6G 1H1, djbartle{at}uwo.ca
Associate Memeber
CanChild Centre for Childhood Disability Research
McMaster University
Hamilton, Ontario, Canada
Associate Professor
School of Physical Therapy
Faculty of Health Sciences
The University of Western Ontario
deblucy{at}uwo.ca
References
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
