Standardization of Outcome Measures

The Inuit people have many words for snow. No doubt, like me, you've heard this said for as long as you can remember. This precision of language captures the unique quality of each different class of snow and reflects an appreciation for its uniqueness. As a child growing up in Maine, I too appreciated that snow could be classified: there was the wet, sticky snow that was perfect for snowmen and snowballs; there was the hard crusted snow that could be stacked in sheets to make snow forts; there was the soft, powdery snow that was perfect for skiing on “snow days” off from school. It was clear that each class of snow was best suited to a particular use, and trying to compel the snow to a fill a role that did not suit its qualities was an exercise in frustration. Our patients share something in common with snow. They represent an amazing diversity, and not just by virtue of their singular humanity. Despite similar medical diagnoses each will manifest his pathology, impairment, and disability in a unique and individual way. Yet from this diversity, patterns emerge–patterns of signs and symptoms that form the basis for patient classification. In her 2005 McMillan address Dr. Rebecca Craik1 eloquently advocated for treatment based on patient classification and the use of standardized outcome measures, echoing an appeal that other luminaries of our profession2–4 have been making for 2 decades. Dr. Shirley Sahrmann2 has long advocated for classification in the form of physical therapy diagnoses, citing the benefits of directed intervention and improved communication among therapists. Dr. Steve Rose3 observed that the delineation of classification groups would facilitate research regarding the effectiveness of our interventions. In her March 2004 JNPT editorial, Dr. Judy Deutsch5 noted that classification based on prognosis would allow us to advocate for care based on what is known about rate of recovery and responses to intervention. While the Guide to Physical Therapist Practice6 is an essential first step in this direction and arguably represents the foundation of a classification system, it is not the system itself. So then, what is the next step in identifying which patients respond best to which approaches so that we can be sure that we are delivering the best possible care? There are now sophisticated techniques available to probe data for the purpose of identifying associations and interactions. Advances in data mining, a data analysis approach designed to discover patterns and to cluster data into meaningful subgroups, have allowed marketers to describe and predict consumer spending patterns. These techniques are increasingly being used in the health care arena as well.7,8 Large scale research endeavors such as multicenter trials are ideally suited to this form of analysis, and there is no doubt that we need to apply these strategies in our large-scale clinical trials. But does this mean that the quest for the development of classification lies solely within the purview of academe and its grant-funded researchers? I would argue that the opposite is true;all physical therapists have the opportunity to contribute to the development of a classification system. More than any other group of health care professionals, physical therapist clinicians are uniquely positioned to gather and evaluate information about our patients and their response to our interventions. However, in our covenantal relationship with our patients it is easy to lose sight of the forest for the trees. We need to look carefully at our patient records then take a step back, look at the forest, and become active observers of the patterns that emerge. We can, as individual therapists, identify characteristics of our patients who respond well to an approach versus those who do not. But there is a caveat, these relationships can only be discovered and compared within and among therapists if we use the same standardized outcome measures. Dr. Craik's address represents the most recent voice in a call to action that has been sounding for 20 years. Like a snowball rolling downhill, I believe this appeal has reached a critical mass and now is the time for our generation to take up this charge. Perhaps due to more straightforward pathokinesiology, classification of patient groups has advanced further in the orthopedic realm than it has in neurologic physical therapy. But even so, we in the neurologic realm of practice need not start from scratch. Scheets et al9 provide an excellent framework upon which to build. The next step is to find like-minded colleagues who share our practice interests and commitment to delivering the best possible care. I believe there already exists, within our Special Interest Groups, a structure that is ideal for undertaking this mission, a mission that would involve multiple phases. First, we must identify the best standardized tests for subjects within our realms of practice and agree to use these outcome measures. Second, we must develop subgroup classification of patients based on initial scores on these standardized tests. Third, we must identify the interventions associated with the greatest gains within each subgroup. Ultimately a database can be developed (HIPAA compliant, of course) to allow large-scale tracking of outcomes and facilitate refinement of both classification and intervention. While this effort would be a colossal undertaking, the payoff would be immeasurable–the ability to efficiently identify and effectively administer the best possible intervention. When this work is done, we will be able to appreciate our patients and best approach to each with the same clarity that Inuits–and children–appreciate snow. Could we have any greater aspiration? My heartfelt thanks to all the authors who contributed their time and expertise to this special issue. A sincere thanks also to D r. Carol Richards for your invaluable assistance; your keen insight and broad perspective continue to be a source of inspiration for so many of us.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.