Incremental hospital utilization and mortality associated with co-morbid depression in pediatric hospitalizations

Abstract Objective The objective of this study was to evaluate the association between hospital utilization and mortality and the presence of co-morbid depression in pediatric hospitalizations. Methods Using the Kids' Inpatient Database (KID) for 2012, a nationally representative database of all inpatient admissions in the United States of America for patients younger than 21 years of age compiled by the Health Care Utilization Project (HCUP), we obtained hospital records for children aged 6–20 years with any one of the 10 most common diagnoses in this population excluding affective disorders. After using one to one propensity score matching on age, race, gender, obesity, insurance type, hospital location and hospital size to ensure exchangeability between the groups, we compared the Length of Stay (LOS), Total Hospital Costs (THC) and mortality in children with and without reported comorbidity depression. We employed descriptive statistics and linear regression methods in our analyses Results A total of 667,968 discharges were extracted estimating a total of 937,971 discharges for children aged 6 - 20 with a primary diagnosis of any one of the 10 most common non-affective diagnoses in 2012. The prevalence of comorbid depression was about 2.9%. Propensity score matching produced 17,071 pairs. The mean LOS among patients with comorbid depression (4.63 days) was 0.89 days (95% CI: 0.74–1.05 days) higher than among patients without comorbid depression (3.74 days). The mean THC among patients with comorbid depression ($10,643) was $2,961 (95% CI: $2,401–$3,512) higher than among children without comorbid depression ($7,682). The odds of death as an outcome among the depressed was 1.77 (95% CI 1.13–2.77) times the odds among non-depressed patients. Conclusions These findings show that comorbid depression significantly increases hospital utilization and mortality in childhood hospitalizations.