Factors Associated with Rapid Readmission Among Nevada State Pyschiatric Hospital Patients

Rapid readmission (RR) of psychiatric patients within 30 days of discharge places a costly burden on state psychiatric facilities and may be an indicator of suboptimal service provision. Several studies have previously considered RR to inpatient facilities, but there is a lack of information available about the variables associated with RR of psychiatric patients to state-operated inpatient facilities in Nevada. This study attempts to identify factors associated with RR at a southern Nevada state psychiatric hospital. Participants included 7,177 patients admitted between May 1, 2012 and April 30, 2014. All 12,068 admissions, including 2,220 RR, were reviewed, and rapid readmits were compared to their counterparts who were not readmitted within 30 days in a multivariate model using logistic regression. Multinomial logistic regression was utilized to determine if risk factors varied based on the number of RR, and to ascertain whether or not previous length of stay (LOS) had an impact on time to readmission. Whenever possible, analyses were run separately including all admissions and each patient\u27s first admission only. Results from multiple logistic regression consistently demonstrated that those aged 35-44, never married, divorced, and living in a homeless shelter or other residential or institutional setting are at increased risk for RR; in contrast, females, persons not receiving social security disability, and those over 55 years of age were at reduced risk. Significant factors associated with RR also included a history of legal problems, medication noncompliance, lack of stable housing, and diagnosis with a psychotic or substance use disorder. Persons living in a homeless shelter and having a history of legal issues were more likely to be in the group most often rapidly readmitted (four or more times), and those that did not receive social security disability were less likely to be included in this high risk group. Previous LOS did not contribute significantly in multivariate modelling of time to readmission. Nevada currently suffers from budget cuts in mental health care spending, which were a result of the recent economic crisis, and a severe lack of bed space in southern Nevada. This study demonstrates that it may be possible to reduce rates of costly RR by focusing on those with a history of rapid readmission and modifiable factors including social and financial supports and housing