Individual differences, variability, and student research.

Amid the chaos and frustration that was my education in chemistry, there were some rare pleasant moments. The laboratory was my salvation, a place where my skills as a cook could be put to use in following specified procedures to identify unknowns and quantify mysterious aliquot parts. I particularly liked using those wonderful scales, the ones with which we calculated tare (the weight of containers and filter paper) and then measured ingredients after the materials were enclosed behind a glass door. I liked the idea that not even a breeze could affect the measurement. It was as though we were weighing things in a world separate from our own, a world in which error could not intrude. The precision was awe inspiring and something many of us still yearn for.

In the chemistry lab, we can wall out the world while we weigh minuscule quantities, but elsewhere such walls are not easily built, nor are they constructed without serious consequences. The journal receives many manuscripts in which authors "wall out" individual differences with great zeal. But physical therapists manage patients--that is, people--and patients do not come out of a cookie cutter. To conduct research that will relate to the real world, the passion for controls must be tempered by a realization of the effect that those controls can have on the inquiry.

Authors of research books too often treat the world as if it were one giant chemistry lab, a place from which we can eliminate all sources of error. Most of the texts bear little resemblance to the real world of scientific inquiry, and those in the biomedical sciences too often are rigid, when clinical research requires adaptation and flexibility. Novice researchers often depend too heavily on these texts instead of working with experienced scholars. From my perspective, the majority of physical therapists who teach research are novices themselves, and they use research books without tempering the content with knowledge borne from experience. In these books, relevance accidentally takes a back seat to a slavish desire to control things, and reality is left behind.

If you were developing an instrument to measure the motion available to patients with low back pain, you would need to study that instrument as it is applied to such patients. Yet we still hear authors suggest that reliability studies using clinically relevant populations introduce a source of error--the variability of the patient. But what is wrong with that? Won't this source of error be present in the real world in which the measurements will be used? Sometimes researchers eliminate the use of examiners to reduce sources of error; in other words, many people make judgments, but only one physical therapist interacts with the patient while the others watch live or via video-tape. When I last checked, however, our patients need to be examined by those who actually use the measurements.... Is it always wrong to reduce these sources of error? No. It all depends on the question and the need to generalize the results.

Clinical research has to be conducted in a real-world context if it is to be relevant, but that does not mean that variability in subjects is necessarily desirable or that we want to eliminate meaningful controls from clinical investigations. Research should be credible, and when there are interventions we need to isolate the effect; however, if research is to guide practice, our subjects should be similar to the patients we would see in the world of practice. Excessive heterogeneity in subjects may compromise our ability to understand experimental effects, but artificial homogeneity may make research irrelevant. How can you strike a balance? First, you have to know something about practice and about research. You learn this in part by making mistakes and through your interactions with patients, colleagues, peer reviewers, and others.

Poorly guided student research, which I believe is the dominant mode of student research in physical therapy, often is the best example of the excesses of which I speak. How can students have the clinical expertise--let alone the knowledge of how to make trade-offs--that is necessary to the research process? Faculty advisors should know how to balance those two necessities, but very few of those who supervise mandatory professional (entry-level) student research projects have adequate knowledge or experience. We simply do not have that many veteran researchers in our profession. If you doubt this claim, look up the publication records of faculty members who work with student researchers. If the faculty members do not have a sustained record of scholarly publication in peer-reviewed journals, they are inappropriate supervisors. The issue is peer-reviewed publication, which is the sine qua non of academic and scholarly credibility--not whether they are nice people, have lots of clinical experience, have won APTA awards, or can teach fantastic continuing education courses. Such individuals do have a role in guiding research questions and in dialogue about research, but they are ill equipped to help students design, implement, and publish research.

Even as we vigorously reaffirm the need to have licensed physical therapists supervise our students' clinical activities, we allow people without appropriate credentials to teach and guide student research. In the past I have called this academic hypocrisy "malpractice. " I suggest that students demand to see the credentials of their faculty and that academic leaders put an end to the inappropriate utilization of faculty. Not all faculty members, especially clinical faculty, need to have a research publication record, but if a faculty member presumes to guide others in the research process, he or she should have been there first--and publication is a mandatory part of the research process.

We talk more and more about evidence-based practice. Even if we did not need to make a case for reimbursement or for the provision of services, physical therapists still should crave evidence so that they can use it to guide patient/client management and provide better services. We can no longer afford the luxury of "playing" with research. We have to do research with dedication. We have to stop wasting resources on projects that lead nowhere and through which our new physical therapists are acculturated to accept mediocre data. For most students, their time and their faculty's time would be better spent elsewhere.

Most new physical therapists, including those who have done these research projects, graduate without knowing how to use science and evidence in practice. At a minimum, they should be able to write publishable case reports that are scientifically credible. In my opinion, they are not able to do so. We need physical therapist educators to start contributing to practice by guiding their students in relevant scholarly activities--not just in "projects" that often obscure the sad reality that few faculty have ongoing research or have contributed to the literature, not even through case reports.

The time has come to appreciate individual differences, and one of those differences between faculty and students is that the faculty should be doing the research. They should not be looking for students to "cover" them for what they are not doing. In addition, we should appreciate that we are training students to be scientific practitioners. More important than mandatory research projects is our students' ability to apply science every day to better serve their patients.