Study Design Clinical Measurement. Introduction Thumb carpometacarpal (CMC) osteoarthritis is a frequent cause of thumb pain. Purpose of the Study To evaluate the interrater reliability and the criterion validity of the grind test for thumb CMC osteoarthritis. Methods Radiological evaluation was used as the gold standard. Seventy thumbs of 54 persons with various hand disorders were included in the study. The grind test was performed by two physical therapists, and radiographs were evaluated by a certified hand surgeon, all blinded to the participants' diagnosis and other test results. Interrater reliability was compared with a kappa statistic, and the sensitivity, specificity, likelihood ratios, and predictive values were calculated from a 232 table. Results A kappa value of 0.48 was calculated as the interrater reliability. The sensitivity of the grind test was 42% and 53%, depending on tester. The specificity of the grind test was 80% and 93%, depending on tester. The positive likelihood ratio was 2.65 and 6.00, and the negative likelihood ratio was 0.59 and 0.62. The positive predictive value was 91% and 96%, depending on tester. The negative predictive value was 68% and 70%, depending on tester. Conclusion The grind test has moderate reliability and accurately confirms the diagnosis of CMC osteoarthritis and identifies those who do not have CMC osteoarthritis. However, a negative grind test does not necessarily reflect negative radiographic evidence of thumb CMC osteoarthritis. Level of Evidence n/a