The research design used in this study is classified as descriptive (Babbie ER. The practice of social research. 6th ed. Belmont, Calif.: Wadsworth Pub. Co.; 1992:91), and the intent of the study was to describe the potential referred pain patterns of this population.
In a descriptive study, a second population (that is, a control or other group) is not utilized unless the researcher desires to compare the populations (Norman GR, Streiner DL. PDQ statistics. St. Louis: Mosby; 1986:159-62). If a research design utilizes an intervention, then I strongly recommend a control group be utilized so the intervention group can be compared with the control group in order to determine whether the intervention group had a significantly better effect than the control group.
Unfortunately, there is no objective method to identify referred pain (that is, cervicogenic headaches and the like) other than through patient reports. TMD pain (intensity, frequency and character) is similarly based on patient perception, with no objective technique available to measure the perceived pain. Probably the closest procedure we have is pressure algometer pain thresholds, which is also based on patient perception.
I concur with Dr. Marbach that researchers need to be cognizant of the tendency for subjects to distort their perceptions to please the examiner. During a study, subjects should observe an atmosphere in which the examiner is seeking the absolute truth and nothing less. I believe accurate study results are a product of the subject’s integrity as well as the integrity of the individuals running the study.
I disagree with Dr. Marbach that utilizing an examiner unfamiliar with the hypotheses makes the individual blinded nor does it necessarily enable him or her to obtain more honest subject perceptions.
In this article, I tried to convey that the results were collected with the greatest subject honesty possible. In the environment this study was conducted, I cannot believe a subject would have reported he or she felt pain in a distant location when he or she had not, and then fabricated a mythical location.
My belief is supported by the comparison of our results with the Branch et al. study (
Branch MA, Carlson CR, Okeson JP. Influence of biased clinical statements on patient report of referred pain. J Orofac Pain 2000;14[2]:120–7[Medline]
).
If only Dr. Marbach had the experience of working with me, he would know that my integrity is not an issue that needs to be challenged. My objective for donating my time to research is to provide myself and our colleagues with information that enables us to better diagnose and treat TMD patients, not to provide disingenuous information that would be refuted by future studies.
I concur with many of the points Dr. Marbach made. One central thrust is that readers need to critically evaluate the research they read to determine whether the results can be generalized back to their patients.
After reading Dr. Marbach’s comment, in hindsight, it would have been nice to evaluate subgroups (that is, more depressed subjects, subjects with greater pain chronicity and others) of this population for referred pain differences. I appreciate Dr. Marbach’s taking his time to review the important subject of research design.
