Length of Hospital Stay and Ventilator Days Among Trauma Patients With Major Psychiatric Illness

Abstract

Background:

Little research has been done on individuals with major psychiatric illness and longer mechanical ventilation and hospital stays. The purpose of this study is to examine the mechanisms of association between major psychiatric illness and mechanical ventilation days, length of hospital stay, and total ICU days.

Methods:

Data for this study consisted of subjects admitted to the intensive care unit in a level-1 trauma center after a motor vehicle accident from January 2011- September 2015 (n=2,802). Wilcoxon rank sum test and chi-square tests were used to determine differences between those with major psychiatric illness and without psychiatric illness for demographics, injury severity score (ISS), substance use, and clinical outcomes. Mediation models were computed for associations between psychiatric illness and clinical outcomes through alcohol and drug use. R 3.5.3 and Mplus (version 8) were used in this analysis. The university institutional review board approved this study.

Results:

Among all participants (n=2,802), 5.5% (n=155) were classified as having a major psychiatric illness. Mechanical ventilation days were higher among those with psychiatric illness (4.90) compared to those without (4.86). A higher percentage of those with major psychiatric illness had an alcohol use disorder (14.7%) compared to those without psychiatric illness (5.1%) (P<0.001). Additionally, a higher percentage of those with major psychiatric illness had a drug use disorder (16.1%) compared to those without (4.4%) (P<0.001). Individuals with a major psychiatric illness had a higher length of hospital stay (16.41) compared to those without (13.61) (P<0.001); additionally, major psychiatric illness was associated with a higher number of ICU days compared to those without a major psychiatric illness (8.49 vs. 8.21, respectively, P=0.004). The association between psychiatric illness and length of stay was partially explained by drug use and alcohol use in the mediation models (45.4% mediated through drug use and 22.5% mediated through alcohol use). The association between psychiatric illness and ICU days was partially mediated through alcohol use (15.2% mediated); however, psychiatric illness and ICU days was not statistically significantly mediated through drug use.

Conclusions:

Appropriate measures should be used to reduce length of stay, ICU stay, and mechanical ventilation days among those with major psychiatric illnesses.