BackgroundArthrometry has been introduced to evaluate the laxity of ankle joint in recent years. However, its role in the diagnosis of chronic ankle instability is still debatable. Therefore, we assessed the diagnostic accuracy of a digital arthrometer in terms of sensitivity and specificity.MethodsAccording to the inclusion and exclusion criteria proposed by the International Ankle Consortium, we recruited 160 uninjured ankles (control group) and 153 ankles with chronic ankle instability (CAI group). Ankle laxity was quantitively measured by a validated digital arthrometer. Data of loading force and joint displacement were recorded in a continuous manner. Differences between the control and CAI groups were compared using 2-tailed independent t test. A receiver operating characteristic curve analysis was used to calculate area under a curve, sensitivity and specificity.ResultsLoad–displacement curves of the two groups were depicted. Differences of joint displacement between the control and CAI groups were compared at 30, 45, 60, 75, 90, 105 and 120 N, which were all of statistical significance (all p < 0.001) with the largest effect size at 90 N. Statistical significance was found in the differences between the two groups in load–displacement ratio at 10–120 N, 10–40 N, 40–80 N and 80–120 N (all p < 0.001), with the largest effect size at 10–40 N. Load–displacement ratio at the interval of 10–40 N had the highest area under a curve (0.9226), with sensitivity and specificity of 0.804 and 0.863, respectively, when the cutoff point was 0.1582 mm/N.ConclusionThe digital arthrometer measurement could quantitively analyze the ankle laxity with high diagnostic accuracy. The load–displacement ratio would be a reliable and promising approach for chronic ankle instability diagnosis.Level of evidence level II.