Abstract: AIM To investigate the value of long RR interval in clinical diagnosis. METHODS Study subjects were randomly selected from patients with the longest RR interval of no less than 1.5 s from a 24-hour ambulatory electrocardiogram examination and their diagnosis information was recorded. The data were analyzed by Chi-square test, Fisher exact test and variance analysis. RESULTS The varying length of the longest RR interval corresponded to different electrocardiogram (ECG) diagnosis (Fisher test, P<0.0001). Multiple comparison analysis showed that the ECG diagnosis between the three groups were not the same: mainly arrhythmia and nonconducted atrial premature in 1.5-2.0 s group, mainly atrial fibrillation in 2.0-4.0 s group and mainly sinus arrest in 4 s and above group. Different frequencies of the long RR interval correlated to different ECG diagnoses. Multiple comparison analysis showed that the ECG diagnoses between the three groups were not the same: the frequency in 1-12 were mainly arrhythmia and nonconducted atrial premature, the frequency in 13-96 were mainly arrhythmia and atrial fibrillation and the frequency in 97 and above were mainly atrial fibrillation and sinus bradycardia. The proportions of nonconducted atrial premature and arrhythmia decreased gradually with increase of the long RR interval frequency, but those of atrial fibrillation, sinus bradycardia and sinus arrest were on the contrary. Fisher test showed that larger proportions of sleep unrelated diseases were atrial fibrillation, arrhythmia and nonconducted atrial premature, while the proportions of the last two diagnoses in sleep related group were also large. The distribution of the disease in the study was mainly high blood pressure, arteriosclerosis, coronary heart disease, atrial fibrillation, diabetes mellitus and hyperlipidemia. The results of variance analysis showed that the longest RR interval was found in patients with three or more kinds of the above diseases (1.98±0.56), followed by patients with two kinds of the above diseases (1.87±0.48) and by patients with one of the above diseases (1.75±0.26). The RR interval frequency had the same trend. CONCLUSION Longer length and higher frequency of long RR interval are associated with the complexity of the disease. Patients with two or more kinds of diseases often have a longer and a higher frequency of the long RR interval, suggesting that the diseases exert an additive or synergetic effect on the heart.