Reliability of procedures used in the physical examination of non-specific low back pain: A systematic review

We present a treatment-based classification approach to the conservative management of low back syndrome. The approach has three levels of classification based on historical information, behavior of symptoms, and clinical signs. We first distinguish patients whose conservative care can be managed predominantly and independently by physical therapists versus patients who require consultation with other services (eg, psychology) or who require referral because of possible serious nonmusculoskeletal pathology. Once patients who can be managed by physical therapists are identified, the next level of classification is to stage their condition with regard to severity. We propose three stages: stage I for patients in the acute phase where the therapeutic goal is symptom relief, stage II for patients in a subacute phase where symptom relief and quick return to normal function are encouraged, and stage III for selected patients who must return to activities requiring high physical demands and who demonstrate a lack of physical conditioning necessary to perform the desired activities safely. The remainder of the article focuses on a third level of classification for stage I only in which patients are classified into distinct categories that are treatment-based and that specifically guide conservative management. The entire approach is diagnosis based, with specific algorithms and decision rules as well as examples presented.

A reliability study and case-control study were conducted. To determine the reliability and discriminative validity of the Biering-Sorensen test. A low Biering-Sorensen score has been found to predict who will have nonspecific low back pain. However, the reliability of the test remains controversial, implying that some studies may have produced results that underestimated the magnitude of the predictive validity of this test. Two raters measured the time holding a specific position (holding time) of 63 subjects (23 currently experiencing nonspecific low back pain, 20 who had had an episode, and 20 who were asymptomatic) while they performed the Biering-Sorensen test twice, 15 minutes apart. A standardized protocol was followed. Test-retest reliability was evaluated by calculating intra-class correlation coefficients (ICC 1,1), 95% confidence intervals (CI), and standard errors of the measurement (SEM) for the total group and for the subgroups. A three-way analysis of variance was used to determine whether test order, subject gender, or symptom status affected holding time. High reliability indices were obtained for the Biering-Sorensen test in subjects with current nonspecific low back pain (ICC [1,1], 0.88; 95% CI, 0.73-0.95; SEM, 11.6 seconds), in subjects who had had nonspecific low back pain (ICC [1,1], 0.77; 95% CI, 0.52-0.90; SEM, 17.5 seconds), and in asymptomatic subjects (ICC [1,1], 0.83; 95% CI, 0.62-0.93; SEM, 17.4 seconds). Results of an analysis of variance showed that subjects asymptomatic for low back pain had a significantly longer holding time than the other two groups (P < 0.05). The Biering-Sorensen test provides reliable measures of position-holding time and can discriminate between subjects with and without nonspecific low back pain.