Patient- and practice-related determinants of emergency department length of stay for patients with psychiatric illness.

STUDY OBJECTIVE

To identify patient and clinical management factors related to emergency department (ED) length of stay for psychiatric patients.

METHODS

This was a prospective study of 1,092 adults treated at one of 5 EDs between June 2008 and May 2009. Regression analyses were used to identify factors associated with ED length of stay and its 4 subcomponents. Secondary analyses considered patients discharged to home and those who were admitted or transferred separately.

RESULTS

The overall mean ED length of stay was 11.5 hours (median 8.2 hours). ED length of stay varied by discharge disposition, with patients discharged to home staying 8.6 hours (95% confidence interval 7.7 to 9.5 hours) and patients transferred to a hospital outside the system of care staying 15 hours (95% confidence interval 12.7 to 17.6 hours) on average. Older age and being uninsured were associated with increased ED length of stay, whereas race, sex, and homelessness had no association. Patients with a positive toxicology screen result for alcohol stayed an average of 6.2 hours longer than patients without toxicology screens, an effect observed primarily in the periods before disposition decision. Diagnostic imaging was associated with an average 3.2-hour greater length of stay, prolonging both early and late components of the ED stay. Restraint use had a similar effect, leading to a length of stay 4.2 hours longer than that of patients not requiring restraints.

CONCLUSION

Psychiatric patients spent more than 11 hours in the ED on average when seeking care. The need for hospitalization, restraint use, and the completion of diagnostic imaging had the greatest effect on postassessment boarding time, whereas the presence of alcohol on toxicology screening led to delays earlier in the ED stay. Identification and sharing of best practices associated with each of these factors would provide an opportunity for improvement in ED care for this population.