Rate and Predictors of 30-day Readmission among Adolescents with Major Depressive Disorder: A Nationally Representative Study

Zeeshan Mansuri, MD, MPH

Boston Children’s Hospital

Scientific Abstract

Background: Major depressive disorder (MDD) related hospitalizations cause a substantial burden on the patient and the healthcare system. There is a paucity of studies reporting the rate and predictors associated with readmission among adolescents with MDD. We sought to evaluate the rate and predictors of 30-day readmission among MDD adolescents.

Methods: Patients with index admissions for MDD were obtained from the National Readmission Dataset (NRD) for the year 2016. The primary outcome was the rate and predictors of 30-day readmission (RA). The t-test, Chi-square test, and logistic regression analysis were performed. Odds ratio (OR) and 95% confidence interval (CI) were used to present the logistic regression analysis.

Results: Weighted Data of 54,586 patients was included (mean age: 15 years, female: 72%). The rate of 30-day readmission was 6.1 %, and the mean time to readmission was 13.5 days (12.9 days male vs. 12.8 days female, p: 0.20). MDD was the main cause of RA (75%), followed by other mental illnesses (20%). Among patients with RA vs. without RA, there was a high prevalence of anxiety disorders (58% vs. 52%, p: 0.003), personality disorders (6.6% vs. 4.8%, p: 0.003), personal history of self-harm (23% vs. 20%, p: 0.01), history of abuse (15% vs. 12%, p:0.002), and insomnia (11.9% vs. 9.3%, p: 0.006). There was no difference in RA by gender, discharge disposition, health insurance, and hospital type (by bed size, rural-urban location, and government-private). RA was low in patients living in the areas with the median income in the lower quartile (23% vs. 27%, p: 0.02). After controlling for relevant covariates, personality disorders (OR:1.49 (1.18-1.88)), history of abuse (OR: 1.26 (1.04-1.53)), personal history of self- harm (OR: 1.19 (1.01-1.39)), anxiety disorders (OR: 1.23 (1.05-1.44)), and substance use disorders (OR: 0.77 (0.65-0.90)) were predictors of RA.

Conclusions: There is a high rate of readmission among the MDD population mainly because of the higher prevalence of psychiatric comorbidities, history of abuse and self-harm. Addressing these factors may help reduce the readmission rate and the overall cost in this vulnerable population.

Live Zoom Session – April 21st

research Areas


Zeeshan Mansuri, MD, MPH, Chintan Trivedi, MD, MPH, Abhishek Reddy, MD