Volume 88, Issue 11 , Pages 1535-1541, November 2007
Postacute Rehabilitation Research and Policy Recommendations
Article Outline
- Abstract
- Track 1: Measurement and Methodology Recommendations
- Track 2: Access to Postacute Rehabilitation Services
- Track 3: Organization of PAC Rehabilitation Services
- Track 4: Effectiveness of Postacute Rehabilitation
- Conclusions
- References
- Copyright
Abstract
Clohan DB, Durkin EM, Hammel J, Murray P, Whyte J, Dijkers M, Gans BM, Graves DE, Heinemann AW, Worsowicz G. Postacute rehabilitation research and policy recommendations.
This article summarizes the research and health policy recommendations developed by participants at the symposium “State-of-the-Science on Post-Acute Rehabilitation: Setting a Research Agenda and Developing an Evidence Base for Practice and Public Policy,” held in February 2007. A diverse group of participants represented federal government agencies, private insurers, professional organizations, providers of rehabilitation services, patients and their advocates, and health researchers. During roundtable discussions and theme-specific break-out sessions, participants were asked to consider 5 major topics regarding postacute rehabilitation care: (1) the extent of our knowledge, (2) the limitations of our knowledge, (3) the barriers that limit research, (4) research priorities to reduce these obstacles and assemble needed evidence, and (5) the major policy implications. Several key themes emerged: the need for improved measures, particularly of case-mix factors and treatment ingredients; the need for a more uniform and coherent system of postacute rehabilitation care to facilitate gathering of common data; the need to attend to underutilization as well as overutilization of rehabilitation services; the need for cooperation among payers, providers, and researchers to advance a rehabilitation effectiveness research agenda; and the desire to develop payment policies that are based on research evidence.
Key Words: Health policy, Outcome and process assessment (health care), Outcomes research, Rehabilitation
THIS ARTICLE SUMMARIZES the state of our knowledge about a general assessment of postacute rehabilitation (settings of which include inpatient rehabilitation facilities [IRFs], skilled nursing facilities [SNFs], long-term care hospitals [LTCHs], and home health agencies [HHAs]) as well as research and health policy recommendations that were developed by participants of the symposium, “State-of-the-Science on Post-Acute Rehabilitation: Setting a Research Agenda and Developing an Evidence Base for Practice and Public Policy” held in February 2007.
After attending a series of plenary presentations, 4 groups met separately to pursue the themes of the symposium. Track 1 focused on the need for improved measurement of key rehabilitation variables and methods to collect and analyze this information; track 2, on factors that influence access to postacute rehabilitation care; track 3, on the similarities and differences in quality and quantity of services across postacute care (PAC) settings; and track 4, on the effectiveness of postacute rehabilitation services. The working groups started their deliberations with state-of-the science presentations (published in Archives of Physical Medicine and Rehabilitation), which were followed by 6 sets of work-in-progress presentations (an agenda and list of symposium presentations are available at http://www.foundationforpmr.org/programs/postacuterehab.html).
Working sessions in each track continued on the second day of the meeting. During these sessions, participants focused on 5 major issues: (1) the extent of our knowledge about the questions and topics discussed; (2) the gaps in the evidence that bears on the questions and topics; (3) the obstacles that limit our ability to assemble the evidence needed to address the questions and topics; (4) the research priorities that would help solve these obstacles and assemble the needed evidence; and (5) the major policy implications of these observations.
Discussion leaders (Daniel E. Graves, Bruce M. Gans, Greg Worsowicz, John Whyte) promoted discussion and assured that diverse perspectives were represented. Reporters (Patrick Murray, Dexanne B. Clohan, Elizabeth M. Durkin, Joy Hammel) developed a report summarizing the themes, problems, solutions, research questions, and actions related to their track. Summaries from each of the track sessions follow.
Track 1: Measurement and Methodology Recommendations
The measurement track participants developed recommendations regarding a wide range of questions, a subset of which included:
What Do We Know About Measurement and Methods?
The participants agreed that the body function and structure domains of the World Health Organization’s International Classification of Functioning, Disability and Health (ICF)1 are well defined and characterize many of the short-term outcomes that require measurement. Activity and participation domains are less well defined, but they are the domains in which many of the long-term outcomes should be measured. Physical outcomes (activities of daily living, mobility) generally have well-developed measures, especially for persons with neuromuscular disease and injury. The major problems in this area are the nonuniformity in scaling across postacute instruments and the targeting of measurement to patients’ ability, so to avoid floor and ceiling effects.
Efforts to improve the quality of care through the development and utilization of clinical treatment standards can be supported by inclusion of process items taken from evidence-based rehabilitation assessment and treatment guidelines. Extant guidelines related to pressure ulcer prevention in high-risk populations and for the rehabilitation care of persons with stroke provide a basis for quality improvement.2, 3, 4
Measurement of environmental factors that affect outcomes can be informed by the environmental factors taxonomy within the ICF. However, perceptions of environmental supports and obstacles may be affected by people’s activity and participation levels, suggesting that direct measures of the environment, though sometimes cumbersome, may be important for validating environment-outcome relationships. Little is known about factors affecting case-mix measurement, yet this is a critical issue that affects access to PAC rehabilitation.
Whether measuring outcomes, treatment ingredients, case-mix factors, or environmental influences, instruments have unknown validity when they are used for purposes for which they were not developed. Validation studies should be undertaken to assess the utility of instruments when used in new settings.
Knowledge Limitations
Symposium participants perceived that there are fewer concise and valid measures available for cognitive and psychosocial outcomes than for medical and physical outcomes. Even in the physical domain, many of the available measures that are used for neuromuscular conditions lack adequate validation for application in conditions such as cancer, organ transplantation, and cardiopulmonary conditions, in which fatigue, endurance, and discomfort are more salient issues than capacity to perform specific tasks. With respect to the ICF activity and participation domains, measures are inadequately developed, especially in the arenas of long-term living in nonrestrictive settings, emotional well-being, vocational return, and maintenance of family systems.
Case-mix issues remain a measurement challenge. Needed is an improved understanding of the relative contribution of disease severity, functional status, motivational, coping, and learning factors to functional improvement across a range of health conditions and outcome domains. In addition, comparable case-mix factors must be available across settings to allow for correction for selection bias and different prognoses. Premorbid status is thought to be an important aspect of case mix, yet its measurement is inadequately developed across the continuum of types of health issues for which people use postacute rehabilitation. Premorbid mental health and coping skills may be particularly relevant to subsequent rehabilitation outcome. It should be a long-term goal to provide financial incentives for optimizing treatment outcomes (eg, pay for performance). However, current prospective payment system (PPS) assessment instruments and Centers for Medicare & Medicaid Services (CMS) administrative data are insufficient to serve as case-mix adjusters in such a system, and to avoid patient “cherry picking.” We should focus on process measures of quality until better case-mix variables are available.
Uniformity in outcome instruments and the ability to crosswalk between similar instruments will allow for comparison among conditions and treatment approaches. However, uniform timing of measurement across PAC settings is required to support comparisons. Issues of timing must also be addressed for conditions, such as arthritis, that lack a clear onset date.
With respect to rehabilitation treatment and services, little is known about how evidence-based guidelines should be implemented in conditions with large variations in severity, when the original treatment studies focused on a limited severity spectrum. In addition, we know little about the effects of licensed and nonlicensed rehabilitation nursing care on PAC and its relationship to outcomes. Although researchers have taken initial steps to measure rehabilitation processes, there is much to learn about the processes of care that are critical to outcomes.
Barriers to Progress
Although participants envisioned a program of research that would provide greater linkage between appropriately adjusted outcome data and rational service policies, the timetable mandated by the U.S. Congress for the development of a tool to measure postacute rehabilitation needs, inputs, and outcomes is too short for a thorough assessment of validity and reliability across settings, diagnostic groups, and the intended uses of the instrument.5 There is evidence of inconsistency in ratings between raters of patient status, across settings, and within raters at different times during a PAC episode. Rater bias and a common measurement period must be resolved in patient measurement systems. Reaching agreement will be complicated by the variable recovery rates and treatment approaches across settings and for different diagnostic groups.
There is a relatively poor understanding of the selection process for PAC settings, aside from those that are driven by economic factors unrelated to patients’ status and needs. There is a lack of evidence for the utility of process variables6 in conditions other than stroke, and the few guidelines that exist support the use of process items that are poorly defined and hard to measure. Thus, meaningful implementation of pay-for-performance policies for many conditions must await demonstration of effective process factors and identification of case-mix factors sufficiently powerful to prevent abuse of such a system.
Long-term outcomes and premorbid status have not been measured systematically, though both are keys to judging the effectiveness of what is often a sequence of postacute services. Collection of data on these domains requires new resources and incentives. The instruments must be suitable for telephone administration, and the costs of their collection must be minimized.
Research Priority Recommendations
The participants recommended several research priorities.
Policy Implications
Participants made the following policy recommendations.
Track 2: Access to Postacute Rehabilitation Services
Making access to postacute rehabilitation a priority topic in a research symposium signals the need for answers to several questions, including:
What Do We Know About Access to Postacute Rehabilitation?
The participants concluded that knowledge about access to rehabilitation is based on limited evidence. One of the few examples is the symposium presentation by M. Elizabeth Sandel, MD,7 on an integrated health system in California. She reported that women were less likely than men to receive rehabilitation hospital care. Research is needed to determine the reasons for these variations.
Participants observed that insurance status and payment issues were among the factors influencing rehabilitation access, but not always in expected ways. For example, indigent patients who are able to access comprehensive rehabilitation through a county-supported system might have a longer hospital stay than their insured counterparts because their discharge options are more limited. Although public funding may benefit individual patients, hospitals could find that public funding is inadequate to cover their care.
Participants expressed concerns regarding the access implications of the 75 percent rule for IRF inpatients. Patients can be denied access to inpatient rehabilitation by virtue of having a diagnosis that limits a hospital’s opportunity to comply with the rule. Yet at a different hospital, or even the same hospital in a different time period, the patient would be welcomed as an appropriate candidate for rehabilitation. The rule’s effect on access was viewed by participants as reflecting the absence of research on benefits and costs of rehabilitation in various settings.
The absence of compelling evidence in this area was a concern to participants, given the barriers they observed in conducting research. The group welcomed presentations by researchers that begin to illuminate issues that affect access, for example, the severity of comorbidities that might predict the level of service a patient requires for optimal outcomes.
Knowledge Limitations
Rather than list unanswered access questions, which are documented by Ottenbacher and Graham,8 the participants focused on identifying ways to overcome knowledge gaps.
Barriers to Progress
The participants in the work group expressed a need for:
Research Priorities
Two research priorities were identified to quantify the availability of and need for services. Research is required to determine the:
Having solid information in these 2 areas would promote policy that aligns patient needs with service delivery capacity, and would help identify and address gaps or redundancies. Such research provides a population-based view of access, but additional research would be required to learn more about access at the level of individual patients in their communities. Participants expressed concern that patients may be denied access to rehabilitation programs because of uninformed beliefs of well-intentioned physicians, discharge planners, and case managers. Research is needed to learn if access is affected by:
Not only may the attitudes of health professionals reduce rehabilitation access, but attitudes and assumptions of patients and families also play a role. Limited consumer education in how to judge rehabilitation quality and family members’ fear of taking on an unfamiliar caregiver role can inhibit access.
Attitudes and beliefs that are incongruent with existing evidence that determine patients’ access to rehabilitation were described by participants as the “ethics of admission decisions.” Like other biomedical ethics topics, this topic is ripe for research.
Participants recommended more research into the effects of specialized rehabilitation nursing on patient outcomes. They wondered if variations in the intensity and focus of nursing services in different postacute settings might be an important factor in predicting patient outcomes, and therefore should be a more active consideration in determining the optimal care setting.
Policy Implications
Participants concluded that determining the right level of service requires understanding each person’s medical and rehabilitation needs. Public policy should encourage the delivery of the right level of service, not undermine the capacity to meet patients’ needs.
Track 3: Organization of PAC Rehabilitation Services
In the symposium’s third track, participants examined the state of knowledge regarding the organization and delivery of rehabilitation services across multiple settings. Discussion focused on 2 areas of inquiry. The first was variation in the organization of rehabilitation services across the 4 PAC settings (HHAa, SNFs, IRFs, LTCHs), primarily in terms of process. Questions included:
Following Kaplan’s presentation from the perspective of the Medicare Payment Advisory Commission,9 participants pursued a second area of inquiry: the implementation and evolution of reimbursement mechanisms for postacute rehabilitation providers. Of substantial interest were the consequences attributable to CMS’s implementation of a unique PPS in each of the PAC settings. The issue that unites these 2 lines of inquiry is that advancing the state of knowledge about financing effects on postacute rehabilitation should include: (1) a clearer understanding of which elements of care change in response to new reimbursement mechanisms; and (2) a better understanding of how these care elements contribute to effective rehabilitation, and for whom.
What Do We Know About Service Organization?
Several presentations illustrated how PAC providers, like other health care organizations, change their behavior in response to financial incentives such as those found in PPS or capitation. Earlier studies of the 4 Medicare PAC PPSs show that providers respond to incentives, but that the strategies chosen reflect, in part, the design of the particular PPS. The fact that providers change their behavior, sometimes unexpectedly, may explain why the 4 PAC PPSs have not achieved the intended reductions in expenditures. Other consequences of PPS on quality, access, and outcome, are mixed, based on the review by Chan.10 His review suggests that PPS effects might vary for impairment groups (eg, stroke, traumatic brain injury) or even subgroups of providers (small vs large, free-standing facility vs hospital unit). The group also considered research on organizations such as the Veterans Health Administration and large health maintenance organizations to draw lessons from alternative payment systems.
Resource utilization varies substantially across the PAC settings, raising questions about the cost effectiveness of providing rehabilitation in each setting. These questions have yet to be answered because, with the exception of a small number of diagnoses (eg, stroke), effectiveness studies are rare.11, 12 Track participants agreed that even if it were possible to match patients to the PAC setting that is most effective for their needs, nonclinical factors such as proximity to family, physician and discharge planner familiarity with providers, local market factors, and funding source, usually determine where a patient is treated.
Knowledge Limitations
In addition to problems assessing the effectiveness of PAC settings, participants identified several gaps in knowledge concerning the organization of rehabilitation services. Each PAC setting is like a black box, and how treatment is provided within these “boxes” is poorly understood. Although some studies have evaluated the contributions made by commonly available therapy services, less is known about other services such as recreation therapy and mental health services. Participants noted that in inpatient PAC settings, patients spend most of their hours outside of therapy, and that the contributions made by nurses, physicians, and family and patient educators are studied infrequently. Finally, participants noted that questions about preferable approaches for organizing and managing services and the consequences of these approaches for resource utilization are rarely studied.
There is a paucity of information on how specific patient groups fare under each PPS and whether different PPSs create access disparities. Beyond anecdotal evidence, little is known about the effects of PPS on the “culture of care” as an emphasis on costs becomes more prominent.
Barriers to Progress
In addition to discussing commonly cited barriers in PAC research (eg, lack of funding, the existence of multiple payment systems, assessment tools, and regulatory policies), participants focused on a less often mentioned, but critical, aspect of the research process. Given the division of services across 4 organizational silos, conducting research that opens the black boxes of PAC rehabilitation relies greatly on the development of trust between researchers and the providers they are studying. Creating trust requires a significant investment of time and energy to counteract suspicion that the research may favor one sector of the PAC industry. Another issue was the increasing pace of change within health care organizations. Policy-makers often request evidence on which to base health care policy at a faster pace than scientists can produce it.
Research Priorities
Research on PAC rehabilitation should be “person-centered”: less attention should be paid to comparing sites of care defined by licensure and regulations and a greater focus on understanding the elements of rehabilitation treatment (the “active ingredients”) that are critical to patient success. Some participants argued for examining trajectories of care that begin in the acute care setting. That said, researchers cannot ignore that services are delivered within organizations and that organizational factors affect service delivery. To identify the critical elements of treatments and of organizations that contribute to effective rehabilitation, data sets with uniform outcome and organizational level measures are required.
Policy Implications
Policy changes should be instituted that, at least as a demonstration, allow for innovative approaches for organizing services beyond the current PAC silos.
Track 4: Effectiveness of Postacute Rehabilitation
Discussions in this track highlighted the distinction between efficacy (eg, effect of an intervention in a controlled trial) and effectiveness studies (eg, effect of an intervention in clinical settings). Participants addressed the following questions.
What Do We Know About Rehabilitation Effectiveness?
The underlying severity of the disabling condition, as well as premorbid and demographic factors, are powerful predictors of long-term functional outcomes. The effects of rehabilitation services on these outcomes, although clinically significant, may be smaller in magnitude. Thus, it is important to prioritize research to examine areas in which rehabilitation is likely to have the greatest effect, and to choose outcomes that are likely to be strongly influenced by rehabilitation.
We must move beyond site comparisons (eg, IRF vs SNF) to focus on rehabilitation content. Treatment content and process vary considerably within PAC sites, and there may be considerable overlap in treatment across PAC sites. The need for substantial research infrastructure may result in participating sites being unrepresentative of the population of sites with the same label. Rather, the focus should be on opening the black box of rehabilitation to assess the effects of treatment ingredients. Required is a taxonomy of treatments that is site neutral.
Although randomized controlled trials (RCTs) are valued as the criterion standard design, there are situations in which they are infeasible or inappropriate. Even where they are possible, their findings need to be evaluated in subsequent effectiveness studies. RCTs that address health policy questions are constrained by multiple factors. Because treatment comparisons may involve different lengths of institutional stays or different intensities of therapy, research grants can rarely support the cost of the experimental treatments. The financial incentive, staff training, and research infrastructure necessary to participate in RCTs are absent in many rehabilitation sites. RCTs are appropriate and feasible when studying the effects of focused rehabilitation treatments within a broader system of care.
Efficacy and effectiveness research should alternate in an iterative process in which the value of specific, focused treatments is suggested by quasi-experimental research and evaluated by RCTs, and in which treatment shown to be effective by RCTs can be incorporated into treatment systems and quasi-experimental studies used to confirm their value in practice. Effectiveness research should make use of short-term outcome indicators that are tied specifically to the anticipated effects of rehabilitation and can be used to build a body of evidence to inform long-term outcome research.
Knowledge Limitations
Case mix and selection bias are central issues within PAC rehabilitation effectiveness research. An unknown factor can result in one group of patients receiving a specific type of care, and that factor might also affect outcomes. There are several ways to address selection bias, including propensity score adjustments and instrumental variable techniques. The type of selection bias adjustment used can affect results.13 Multiple regression is a more powerful adjustment technique but is less intuitively accessible. Instrumental variables are related to the likelihood of receiving the treatment but not to the outcome of care. These predictors may be difficult to identify.
We know little about how to organize research related to treatment effectiveness. Two possibilities are focusing on disease states or functional deficits. A multilevel analysis approach was proposed by Prvu Bettger and Stineman14 based on their meta-analysis of PAC rehabilitation research, which is supported by Whyte’s macro-to-micro conceptualization of research.15, 16 These approaches include:
Effectiveness research should include iterative movement among these levels. Studies that begin with participation or functional level selection may need to subdivide by etiology to explain variations in treatment response. Studies that begin with narrowly defined physiologic subgroups may be collapsed into functionally defined groups as commonalities in treatment response across groups are identified.
A critical topic is how to address the multiple services, interventions, intervening variables and their influence on outcomes. Different stakeholders value different outcomes (eg, consumers value instrumental activities of daily living, whereas therapists value activities of daily living). Different values may require outcome assessment in peoples’ homes and communities.
Participants emphasized discrete outcomes such as long-term functional status, living status, resource utilization, and supports. They called for sensitive and valid measures of community participation, integration, quality of life, employment and learning, social support and networking, and environmental barriers and supports9, 17, 18 that would inform level I and II research questions. Policy-makers and providers stressed the importance of cost or expenditure-effectiveness analyses, but emphasized long-term effects, such as costs of institutionalization versus community living. Payment based on outcome may have deleterious access effects unless accompanied by case mix adjustment.
Barriers to Progress
Participants identified several barriers to the measurement of specific effects of rehabilitation interventions on long-term outcomes. Several approaches should be considered:
Research Priorities
Policy Implications
Conclusions
Several key themes emerged from the discussions and recommendations of all groups.
Measurement
Despite much progress in measurement of functional outcomes, measurement of case-mix variables is inadequate for research or payment purposes. Measurement of the “active ingredients” of the rehabilitation process is even more limited, but will be crucial to allow research to go beyond comparisons of sites of care with unclear characteristics. We must develop new and improve existing measures of patient characteristics, treatment contents, and long-term outcomes. Advancing this research agenda will require focusing attention on differences among individual patients and how these differences affect the long-term outcome of PAC rehabilitation.
Rehabilitation Systems of Care
PAC rehabilitation is fragmented into silos that are often parts of different industries, reimbursed differently, and whose outcome measures and data systems are incompatible. The lack of an integrated health care system is an obstacle to gathering common data efficiently across different sites of care. Lack of integration provides incentives for cost-shifting and disincentives for delivering the most cost-effective sequence of postacute services. The development of a common outcome instrument across sites will facilitate research, but attention must be given to the timing of data collection, because treatment phases intersect at varying points with patients’ recovery trajectories.
Partnerships Among Payers, Providers, and Researchers
Current payment policies are not well supported by research on the effectiveness of specific postacute rehabilitation services, yet those same payment policies constrain efforts to study the effectiveness of alternative arrangements of services. Because the “independent variables” in effectiveness research are generally services delivered by clinicians over considerable time periods, providing them experimentally is generally beyond the resources of research budgets. Thus, progress in studying rehabilitation effectiveness will come only through flexible partnerships between payers, providers, and researchers, which allow short-term flexibility in the models of care and reimbursement. This partnership would benefit from a shared strategic plan to guide investigation and policy. To this end, the collaboration of CMS with the National Institute of Child Health and Human Development and clinical investigators on research projects should help accelerate progress. To quote:
CMS recognizes the need to enhance the evidence base for clinical practice, and will work with providers and research groups to encourage the design of clinical studies that meet NIH [National Institutes of Health] standards. CMS will also work with researchers conducting NIH-approved studies so that they can meet their study objectives within the overall framework of the Medicare program benefit.20
Optimization of Access
Evidence to indicate what constitutes over- and underuse of rehabilitation services is sparse. However, as indicators are developed, it is critical that as much attention be paid to ensuring universal access to services that are effective as to reduce delivery of services that are ineffective.
Educational Needs
Finally, participants made 2 recommendations related to educational initiatives. PAC researchers, providers, and policy-makers would benefit from a better understanding of CONSORT21 guidelines for observational and quasi-experimental research. An improved understanding would help improve the quality of varied research models, and assure the availability of additional meta-reviews. Technical support or expert group consultation could help disseminate rigorous research guidelines and assist with research designs and funding initiatives.
References
- . International classification of functioning, disability and health: ICF. Geneva: WHO; 2001;
- Management of adult stroke rehabilitation care: a clinical practice guideline. Stroke. 2005;36:100–143
- . Pressure ulcer: the management of pressure ulcers in primary and secondary care. London: NICE; 2005;
- . NICE pressure ulcer guideline: summary and implications for practice. Br J Community Nurs. 2006;11(6):S18, S20-4.
- Deficit Reduction Act of 2005, Pub L No. 109-171 (2006).
- . Evaluating the quality of medical care. 1966. Milbank Q. 2005;83:691–729
- . Disparities in post-acute stroke rehabilitation: results of an integrated health system study in Northern California. 2007;Paper presented to: State of the Science Symposium on Post-Acute Rehabilitation: Setting a Research Agenda and Developing an Evidence Base for Practice and Public Policy; Feb 12; Arlington (VA).
- . The state-of-the-science: access to postacute care rehabilitation services. A review. Arch Phys Med Rehabil. 2007;88:1513–1521
- . Growth and payment adequacy of Medicare postacute care rehabilitation. Arch Phys Med Rehabil. 2007;88:1494–1499
- . The state-of-the-science: challenges in designing postacute care payment policy. Arch Phys Med Rehabil. 2007;88:1522–1525
- Outcomes and reimbursement of inpatient rehabilitation facilities and subacute rehabilitation programs for Medicare beneficiaries with hip fracture. Med Care. 2005;43:892–901
- Poststroke rehabilitation: outcomes and reimbursement of inpatient rehabilitation facilities and subacute rehabilitation programs. Stroke. 2006;37:1477–1482
- . Analysis of observational studies in the presence of treatment selection bias: effects of invasive cardiac management on AMI survival using propensity score and instrumental variable methods. JAMA. 2007;297:278–285
- . Effectiveness of multidisciplinary rehabilitation services in postacute care: state-of-the-science. A review. Arch Phys Med Rehabil. 2007;88:1526–1534
- . Assessing medical rehabilitation practices: distinctive methodologic challenges. In: Fuhrer MJ editors. The promise of outcomes research. Vol 2:Baltimore: Brookes; 1997;p. 43–59
- . Using treatment theory to refine the designs of brain injury rehabilitation treatment effectiveness studies. J Head Trauma Rehabil. 2006;21:99–106
- . The Community Participation Index: a measure based on constituency input and evaluated in rehabilitation client and general population surveys. 2007;Paper presented to: State of the Science Symposium on Post-Acute Rehabilitation: Setting a Research Agenda and Developing an Evidence Base for Practice and Public Policy; Feb 12; Arlington (VA).
- . Keynote presentation. 2007;Paper presented to: State of the Science Symposium on Post-Acute Rehabilitation: Setting a Research Agenda and Developing an Evidence Base for Practice and Public Policy; Feb 12; Arlington (VA).
- . How comorbid conditions contribute to rehabilitation admittance decisions—SNF or IRF?. 2007;Paper presented to: State of the Science Symposium on Post-Acute Rehabilitation: Setting a Research Agenda and Developing an Evidence Base for Practice and Public Policy; Feb 12; Arlington (VA).
- Information for investigators engaged in medical rehabilitation research (Notice NOT-HD-07-005). Available at: http://grants.nih.gov/grants/guide/notice-files/NOT-HD-07-005.html. Accessed March 13, 2007.
- CONSORT: transparent reporting of trials. Available at: http://www.consort-statement.org. Accessed March 13, 2007.
Supported in part by the National Institute on Disability and Rehabilitation Research through the Rehabilitation Research and Training Center on Measuring Rehabilitation Outcomes and Effectiveness (grant no. H133B040032).
No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit upon the authors or upon any organization with which the authors are associated.
PII: S0003-9993(07)01449-9
doi:10.1016/j.apmr.2007.07.022
© 2007 American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation. Published by Elsevier Inc. All rights reserved.
Volume 88, Issue 11 , Pages 1535-1541, November 2007
