Clinical Prediction Rules Currently Fail to Predict

Many musculoskeletal conditions represent only symptoms related to a myriad of possible underlying pathologies.  For example, due to the many underlying causes of low back pain, no single treatment option including drugs, surgery, exercise or spinal manipulation successfully treats all patients.  Yet each of these treatment options can be successful in individual cases.  The challenge for clinicians is to be able to identify patient attributes that would help predict a positive outcome; matching the patient most likely to benefit from one of the treatment options.

Several research efforts have been made to identify patient characteristics and positive diagnostic findings in order to develop a clinical prediction rule (CPR).  The focus of much of the research in this area to date has been on variables like the patient’s level of pain, duration, location and distribution of pain, positive orthopedic and neurologic findings, socioeconomic variables and comorbidity issues.  For chiropractors the question becomes even more complex when trying to develop a prediction rule that would direct patient care to one of 100+ chiropractic techniques.  Chiropractors use a variety of markers that are often technique specific to identify therapy options.  These tests often include leg length, algometry, skin temperature differential, motion palpation and muscle testing along with the usual patient history, orthopedic and neurologic testing.  To date there are only a handful of studies that compare chiropractic techniques