In last month's article, I discussed the underlying validity of approaching spinal analysis and treatment from a postural perspective. This month, I want to go one step further and discuss the concept of reliability. Let's take a few moments to think of "reliability" as it relates both to what we do as chiropractors in general, and to postural analysis in particular.
Reliability can be simply defined as the consistency of a method of measurement or analysis. In reading scientific papers, you will often encounter the terms "intra examiner" and "inter examiner" reliability. INTRA examiner reliability refers to the consistency of a single individual to get the same or similar results time after time (test-retest). INTER examiner reliability refers to the consistency of findings between different examiners (examiner A vs. examiner B). Obviously, it's a good thing if you can do a test three times in a row and get pretty much the same result. It's also a good thing if two doctors can do the same test and get the same findings, although no test is likely to be 100 percent reliable.
It's important to remember that, just because a test is "reliable," it still doesn't mean it's "valid." If we revisit last month's example of the good old Derifield Leg Check, just because I get a consistent (reliable) result with the leg check, doesn't necessarily tell me anything about what causes the phenomenon of a "short leg." One person may interpret a "short leg" finding to be indicative of sacral subluxation. Another may believe it due to pelvic misalignment. Or (my personal favorite) it might just be caused by shortness in the leg! My point here is that reliability has nothing to do with validity. In this case, all three hypotheses can't be correct (valid) at the same time and, yet, the method could still be highly consistent (reliable) in terms of producing the same result time after time. Got it?
So, why is scientific reliability important to you as a practicing chiropractor? Well, to start with, the reliability of all health care methods is constantly being called into question. Over the years, chiropractic's critics have railed against the subjective nature of various methods of analysis. To be honest, we might have avoided much trouble had the profession done a better job of being critical and introspective of our own science. This process of constantly looking inward to address shortcomings and figure out ways to better our processes is incumbent upon all professions. Understandably, it would be pretty hard to expect the world to take us seriously, if ten chiropractors performed a test and came up with ten different conclusions. Reliability, then, is a vital cornerstone in the foundation of any science.
The inclusion of postural analysis allows the doctor to access information about the patient's biomechanical status using methods of measurement that are among the most reliable available to chiropractors. Let me repeat that, because it is important. Postural analysis, particularly radiographic analysis, is documented to be among the very most reliable methods available to us as chiropractors. Here are just a few examples.
A 2003 study by Harrison, et al., looked at radiographs of subjects taken anywhere from a few days up to months, and sometimes years apart. In 48 of 50 measurements, the differences in measured posture was less than 1.5 degrees and under 2mm.
The conclusions here were two fold. 1) Human posture, itself, is a very consistent (reliable) phenomenon over time. Patients have a strong tendency to return to their same habitual standing posture day in and day out. 2) Proper use of radiographic analysis appears to be a highly reliable method of measuring spinal displacement (poor posture).1
Another study found X-ray line drawing analysis to have "high reliability with a majority of ICC's in the .8-.9 range."2 (Intraclass Correlation Coefficient is a statistical means of measuring consistency between multiple examiners. An ICC of 1.0 would indicate perfect agreement between examiners.) The authors point out that, in most studies looking at the inter examiner reliability of both chiropractic and medical radiologists in determining pathology from X rays, a widely accepted use of radiography will, generally, only show ICC's in the range of .40 to .75.
And finally, a 1999 study3 looked at, among other things, a simple method of analyzing the cervical lordosis using Jackson's angle. The authors found INTRAexaminer reliability in the range of .82 to .95, while INTERexaminer reliability ranged from .89 to .99. Remember, 1.0 would represent perfect agreement between examiners here! The inescapable conclusion here is that X-ray line drawing is a highly reliable method of analyzing displacements of spinal posture.
In short, posture matters because human posture appears to be a very consistent (reliable) phenomenon over time. Left uncorrected, patients will consistently return to the same patterns of global spinal misalignment (subluxation) time after time. Posture also matters, because the simple methods of radiographic analysis available to all chiropractors are the most reliable methods of spinal analysis we have. Combine valid and reliable methods of analysis with effective methods of postural rehabilitation, and our profession is slowly, but certainly, moving in a direction which is beneficial to the people we treat and care for.
Dr. Mark Payne is president of Matlin Mfg., a manufacturer and distributor of postural rehab products since 1988. For a FREE, unabridged copy of this article or other information on postural chiropractic, please contact Matlin Mfg. Inc. at 1-334 448 1210 or on the web at