To diagnose any condition, a doctor needs to obtain an accurate history, perform a clinical examination that includes appropriate diagnostic tests, and differentially diagnose between or among various pathological conditions.
Numerous extremity-specific exams are available to practitioners. But using lots of tests indiscriminately does not necessarily isolate the offending lesion. Moreover, more tests can result in more false positives and false negatives, thereby leading to an inaccurate diagnosis.
We need to exercise great care when administering tests and interpreting their results. The more accurate the "ortho-neuro" tests, the better the chances of arriving at the correct diagnosis. Although errors associated with individual tests may be small, the (incorrect) use of multiple tests might compound errors, leading to an incorrect diagnosis, and the cascade of subsequent management could quickly drift off target.
Practitioners should determine whether the extremity tests they use are, in fact, providing accurate information. The key question is: Are the reference materials based on good evidence or simply the habits and beliefs of the test authors? In addition, many chiropractic colleges and department heads have favorite ortho-neuro tests, irrespective of the tests' information value as clinical tools.
Critically assess the sources of your reference materials. Be careful if you are still using your college notes or textbooks, because information becomes quickly outdated. More information has been produced in the past decade than in the previous 5,000 years, and our knowledge base is doubling every two years.
Even if you use a newer source, like a recent book or seminar, be aware that those sources might lack scientific rigor. Some authors and speakers have hijacked the term evidence-based or evidence-informed to describe their materials, when in fact, their materials are not supported by valid research.
How can practitioners decide which tests are worth using and which ones should be discarded? The decision to use any test should follow a three-step process:
1. Review the Literature
Conduct a search of the scientific literature in databases such as MEDLINE, EMBASE and CINAHL. It is important to consider all tests that have published research, because this helps reduce the bias of tradition (using a test because you have always used that test). The Institute of Evidence-Based Chiropractic recently did a comprehensive search for special extremity tests using multiple databases and search terms. The search identified numerous tests for each extremity: shoulder – 114 tests; elbow – 29 tests; wrist / hand – 40 tests; hip – 40 tests; knee – 39 tests; and ankle / foot – 43 tests.1
2. Assess Evidence Quality
Appraise the quality of evidence for each test. Health care literature suffers from inconsistent quality, which complicates assessment and distorts conclusions. It is erroneous to assume that professional journals publish only "sound studies" that are properly designed and implemented.
In an analysis of 53 published studies, Sonis, et al., found the mean quality of articles to be 35 percent out of 100 percent, and Rubinstein, et al., found that only 30 percent of the studies had a low risk of bias.4-5 In addition, Evidence-Based Medicine, a journal that appraises studies from over 100 peer-reviewed journals, found that on average, less than 0.1 percent of the studies it reviews are of high quality.
Moreover, you cannot rely on abstracts or the prestige of a journal to establish validity.3 If reference materials do not explicitly state that an assessment of each article's scientific rigor was performed, you should assume it was not performed.
3. Determine Accuracy and Value
Assuming that a study was designed and implemented with reasonable scientific rigor, the third step is to interpret the evidence and determine whether a diagnostic test is accurate and useful. The three key attributes of a diagnostic test are reliability, validity and clinical utility.
Reliability in diagnostic testing is the degree to which observations of the same phenomenon agree among different observers or by the same observer at different times. If a test is unreliable, test results will change without the patient's condition really changing. However, a reliable test is worthless if it is not also valid. Cohen's kappa coefficient and intra-class correlation coefficient are measures commonly used to describe reliability. Cohen's kappa coefficient greater than 0.60 or intra-class correlation coefficient greater than 0.80 are sensible standards.2
Validity is the degree to which a test truly measures what it purports to measure. A valid test is able to detect a true positive and discriminate positive from negative; that is to say, it finds almost all patients who have the condition and hardly any who do not.
Sensitivity and specificity are measures commonly used to describe diagnostic accuracy. Sensitivity refers to the proportion of individuals with a particular disease who are correctly classified as diseased by the test (true positives). Specificity refers to the proportion of individuals without a particular disease who are correctly classified as disease-free by the test (true negatives). Sensitivity and specificity greater than 0.80 are reasonable benchmarks.2
The clinical utility or worthiness of a test is determined by its effects on patient outcomes and patient management. A test should direct the practitioner to deliver a specific therapy or use a therapy in a certain way that improves patient results.
Use this three-step protocol to evaluate the reference materials for the tests you use. At the very least, compare the benchmark statistics we have provided with the individual statistics provided in your reference materials. Do the tests you use meet or exceed these standards?
Assessment tools that demonstrate good validity, reliability and clinical utility can help practitioners deliver management strategies that yield optimal patient outcomes. Tools that do not exhibit those qualities can lead to less-favorable results.
References
- Feise RJ. Doctor's Clinical Protocols: Evidence-Based Extremity Exams. Scottsdale, AZ: Institute of Evidence-Based Chiropractic, 2013.
- Feise RJ. Evidence-Based Chiropractic: Research Methodology, Biostatistics & Critical Thinking. Scottsdale, AZ: Institute of Evidence-Based Chiropractic, 2013.
- Pitkin RM, Branagan MA, Burmeister LF. Accuracy of data in abstracts of published research articles. JAMA, 1999;281:1110-1.
- Rubinstein SM, Terwee CB, Assendelft WJ, de Boer MR, van Tulder MW. Spinal manipulative therapy for acute low back pain: an update of the Cochrane review. Spine, 2013;38:E158-77.
- Sonis J, Joines J. The quality of clinical trials published in the Journal of Family Practice, 1974-1991. J Fam Pract, 1994;39:225-35.
Click here for more information about Ronald Feise, DC.
Click here for previous articles by J. Michael Menke, MA, DC, PhD.
Dr. Edward Crowther is an associate professor in the Faculty of Health and Medicine at the International Medical University in Kuala Lumpur, Malaysia.