Background/Aim: Obstetric brachial plexus injury (OBPI) is caused by traction to the brachial plexus during labor. Traction injury may vary from neurapraxia or axonotmesis to neurotmesis and can cause edema, avulsion, or rupture of the nerve. Improvement in the first two weeks after birth is a good indicator of outcome. The disability varies according to the location and severity of the effect in the plexus. However, most injuries are transient, with a total return of function in many cases. This study aimed to obtain clues for the prevention and follow-up of obstetric brachial plexus injuries by revealing the outcome and clinical features. Methods: In this retrospective cohort study, hospital records of patients with brachial plexus injury due to delivery were reviewed between January 2017 and September 2021. Injury levels, birth weights, other injuries at birth, maternal age, gravidity, gestation time, and treatment response were recorded. Brachial plexus injuries of the patients were classified per the NARAKAS classification. The Spearman correlation and Pearson correlation tests were used for correlation analyses. The variables were evaluated with the Chi-Square and Student's t-tests. The normality of the distribution was assessed with the Kolmogorov-Smirnov test. A value of P<0.05 was considered statistically significant. Results: Thirty-nine cases were included in the study (21 males, 18 females). The mean and median birth weights were 3857 (392) grams, and 3880 (3100-4600) grams, respectively. The median gestational week of birth was 39 weeks. Most mothers were primigravida. All patients were born by vaginal delivery. Per the NARAKAS classification, 29 patients (74.4%) were in group 1, 5 patients (12.8%) were in group 2, 4 patients (10.3%), in group 3, and 1 patient (2.5%) was in group 4. The mean follow-up period was 23.2 (14.1) months. Twenty-four patients recovered spontaneously; six had sequelae without functional impairment, five had sequelae with functional impairment, and two had contractures. The relationship between the NARAKAS groups and birth weight was insignificant (P=0.09). There was a significant correlation between the NARAKAS group and recovery (P<0.001). A correlation was found between sequelae and functional loss (P=0.01) and the NARAKAS group. Functional loss was not related to maternal age, week of birth, birth weight, baby gender (P=0.15, P=0.30, P=0.20, P=0.15 respectively). Conclusion: Permanent functional loss in brachial plexus injury is associated with the NARAKAS classification, and patients in groups 3 and 4 should undergo imaging as soon as possible. Electromyography (EMG), a complex invasive procedure for the newborn, should be preferred if there is no satisfactory recovery. We recommend performing brachial plexus magnetic resonance imaging before EMG to give the patient a chance for early surgery.