Reliability of the visual assessment of cervical and lumbar lordosis: how good are we?

STUDY DESIGN Blinded test-retest design. OBJECTIVE To measure the intrarater and interrater reliability of the visual assessment of cervical and lumbar lordosis. SUMMARY OF BACKGROUND DATA Cervical and lumbar lordoses are frequently evaluated using visual assessment, but little attempt has previously been made to measure the reliability of visual assessment. METHODS Twenty-eight chiropractors, physical therapists, physiatrists, rheumatologists, and orthopedic surgeons were recruited to evaluate the posture of photographed subjects (with and without back pain). Each clinician rated the lordosis of the cervical and lumbar spines as normal, increased, or decreased. Kappa coefficients (kappa) were calculated to determine intrarater and interrater reliability. RESULTS Twenty-eight clinicians evaluated photographs of 36 individuals (17 with back pain, 19 without). Mean intrarater reliability was kappa = 0.50 (95% confidence interval 0.02-0.98) and mean interrater reliability was kappa = 0.16 (95% confidence interval 0.00-0.48). No statistically significant difference existed among the five groups of clinicians or between the evaluation of the subjects with and without back pain. CONCLUSION Intrarater reliability of the visual assessment of cervical and lumbar lordosis was statistically fair, whereas interrater reliability was poor.