Objective To compare the difference in heart rate variability（ HRV） between patients with recurrent depression and long-term hospitalized schizophrenia. Methods From January 2014 to January 2019， a total of 120 patients with multiple episodes of depression or long-term hospitalization with schizophrenia who were hospitalized in the Department of Psychiatry， Shanghai Jiao Tong University School of Medicine Mental Health Center were enrolled retrospectively. The age and gender were matched between the two groups. HRV tests were carried out on the two groups of patients by means of 24-hour Holter recorder and the heart rate variability analysis software. The HRV related indicators RR interval standard deviation（ SDNN）， root mean square of the difference between adjacent RR intervals（ rMSSD）， the percentage of the difference between adjacent RR intervals ＞ 50 ms（ PNN50）， low frequency power（ LF）， high frequency power（ HF）， LF and HF The results of the ratio（ LF/HF） were compared between groups. Spearman correlation analysis was used to analyze the correlation between the number of hospitalizations and various indicators of HRV. Results The time domain index SDNN in the recurrent depression group was higher than that in the long-term hospitalized schizophrenia group［121.0（95.0，158.0）vs 111.0（87.0，144.0）ms，t=2.214］， and rMSSD and PNN50 were lower than those in the long-term hospitalized schizophrenia group［22.0（13.0，46.0）vs 20.0（12.0，28.0）ms， t=3.832；3.0（0，13.0）% vs 2.0（0，7.0）%，t=2.571］. The differences were statistically significant（ all P＜0.05）. There was no statistical significance in the difference of frequency domain indexes（ all P＞0.05）. The results of Spearman correlation analysis showed that the number of hospitalizations in patients with recurrent depression was related to rMSSD（ r=0.270，P=0.003）， PNN50（ r=0.263，P=0.004）， and HF（ r=0.246，P=0.015）. The number of hospitalizations of patients with schizophrenia was not correlated with the analysis indicators of HRV （all P＞ 0.05）. Conclusions The autonomic dysfunction of patients with recurrent depression is more serious than that of patients with long-term hospitalized schizophrenia. The number of hospitalizations of patients with recurrent depression affects the HRV time domain indexes.