RESUMEN
BACKGROUND & PURPOSE: The cervicothoracic differentiation test (CTDT), cervical and thoracic unloading are used clinically to guide treatment. This study sought to determine the reliability and diagnostic accuracy of these tests. METHODS: A prospective diagnostic accuracy study was performed at two outpatient clinics and one university research center. A convenience sample of 48 individuals with neck pain was recruited. Cervical and thoracic unloading tests and CTDT were performed with symptom relief considered a positive test. Pain was assessed using a visual analog pain scale (VAS) at rest and during provocative movements. The reference standard was pain relief following thoracic manipulation. Change in pain was used to identify improvement at the MCID (15â¯mm) and 50% improvement thresholds. RESULTS: All three tests demonstrated high levels of inter-rater reliability, K = 0.90[0.77-1.00]. Of 48 individuals who completed the study, 39 (81.3%) were improved ≥ MCID; compared to 34 (70.8%) at the 50% threshold. As a single test, the CTDT yielded the strongest diagnostic utility (at MCID threshold) based on ROC curve: AUC 0.791 s.e. 0.078; with high specificity (0.89[51.75-99.72]); LR+ 6.23 [0.97-40]; LR- 0.35 [0.20-0.58]; and PPV 96.43. Unloading tests demonstrated high sensitivity, but poor specificity and likelihood ratios. Composite tests improved specificity, but with lower accuracy and minimal changes in ROC area compared to the CTDT in isolation. CONCLUSIONS: The CTDT is a specific test with significant diagnostic utility to identify individuals who will experience immediate pain relief following thoracic manipulation. The CTDT should be considered during the clinical decision making process when treating individuals with neck pain.