Study design: A systematic review.
Objectives: To determine the quality of the research and assess the interexaminer and intraexaminer reliability of spinal palpatory diagnostic procedures.
Summary of background data: Conflicting data have been reported over the past 35 years regarding the reliability of spinal palpatory tests.
Methods: The authors used 13 electronic databases and manually searched the literature from January 1, 1966 to October 1, 2001. Forty-nine (6%) of 797 primary research articles met the inclusion criteria. Two blinded, independent reviewers scored each article. Consensus or a content expert reconciled discrepancies.
Results: The quality scores ranged from 25 to 79/100. Subject description, study design, and presentation of results were the weakest areas. The 12 highest quality articles found pain provocation, motion, and landmark location tests to have acceptable reliability (K = 0.40 or greater), but they were not always reproducible by other examiners under similar conditions. In those that used kappa statistics, a higher percentage of the pain provocation studies (64%) demonstrated acceptable reliability, followed by motion studies (58%), landmark (33%), and soft tissue studies (0%). Regional range of motion is more reliable than segmental range of motion, and intraexaminer reliability is better than interexaminer reliability. Overall, examiners' discipline, experience level, consensus on procedure used, training just before the study, or use of symptomatic subjects do not improve reliability.
Conclusion: The quality of the research on interreliability and intrareliability of spinal palpatory diagnostic procedures needs to be improved. Pain provocation tests are most reliable. Soft tissue paraspinal palpatory diagnostic tests are not reliable.