به کارگیری مدل فرایند شمارشی Andersen- Gill در تعیین عوامل مؤثر بر عود اختلال دو قطبی نوع یک

Background and purpose: Bipolar disorder is a severe psychiatric disorder with a variety of natural courses and manifestations, and a recurrent nature. The aim of this study was to determine the effective factors on recurrence using Andersen-Gill counting process model. Materials and methods: In an ambidirectional cohort study (4 years retrospective and 1 year prospective from 2007 to 2012), 255 patients (141 men, and 114 women) with bipolar type I disorder and history of hospitalization in Zare psychiatric center (Sari, Iran) were studied. Demographic, psychiatric and other variables related to recurrence were gathered from patients' medical records. Andersen-Gill counting process model via Statistical Package of STATA was used to determine predictive variables of recurrence. Results: 693 recurrences occurred during the 5 years follow-up period all the patients experienced recurrence at least once, 29 patients (11.4%) only once, 90 (35.3%) twice, 87 (34.1%) thrice, 25 (9.8%) four times, 21 (8.2%) five times, and 3 (1.2%) six time. Counting process regression model showed that the number of previous episodes and psychiatric family history in first-degree relatives (hazard ratio > 1 P < 0.05) increased the probability of recurrence and presence of psychotic features in each recurrence decreased hazard of recurrence (hazard ratio < 1 P < 0.05). Conclusion: Bipolar disorder has a high rate of recurrence. Predictor variables include number of previous episodes, first-degree relatives’ psychiatric family history and presence of psychotic features in each recurrence.