Date Published: October 30, 2012
Publisher: BioMed Central
Author(s): Antoinette Krupski, Jeanne M Sears, Jutta M Joesch, Sharon Estee, Lijian He, Alice Huber, Chris Dunn, Peter Roy-Byrne, Richard Ries.
Although brief intervention (BI) for alcohol and other drug problems has been associated with subsequent decreased levels of self-reported substance use, there is little information in the extant literature as to whether individuals with co-occurring hazardous substance use and mental illness would benefit from BI to the same extent as those without mental illness. This is an important question, as mental illness is estimated to co-occur in 37% of individuals with an alcohol use disorder and in more than 50% of individuals with a drug use disorder. The goal of this study was to explore differences in self-reported alcohol and/or drug use in patients with and without mental illness diagnoses six months after receiving BI in a hospital emergency department (ED).
This study took advantage of a naturalistic situation where a screening, brief intervention, and referral to treatment (SBIRT) program had been implemented in nine large EDs in the US state of Washington as part of a national SBIRT initiative. A subset of patients who received BI was interviewed six months later about current alcohol and drug use. Linear regression was used to assess whether change in substance use measures differed among patients with a mental illness diagnosis compared with those without. Data were analyzed for both a statewide (n = 828) and single-hospital (n = 536) sample.
No significant differences were found between mentally ill and non-mentally ill subgroups in either sample with regard to self-reported hazardous substance use at six-month follow-up.
These results suggest that BI may not have a differing impact based on the presence of a mental illness diagnosis. Given the high prevalence of mental illness among individuals with alcohol and other drug problems, this finding may have important public health implications.
Brief interventions (BIs) for alcohol and other drug problems have been consistently associated with decreases in self-reported drinking  and, to a lesser extent, decreases in illicit drug use [2,3]. However, alcohol and drug problems do not always occur in isolation. It has been estimated that 37% of individuals with an alcohol use disorder (AUD) and more than 50% of individuals with a drug use disorder (DUD) have a co-occurring mental health disorder [4-8]. Thus, in any screening, brief intervention, and referral to treatment (SBIRT) program, it is likely that a significant subset of persons screened will have mental illness. Because individuals with co-occurring mental illness and substance abuse are known to have poorer outcomes when treated in traditional, non-integrated substance abuse treatment settings relative to individuals without mental illness  and often present clinical challenges when faced with traditional interventions based on motivational interviewing (MI) , it raises the question of whether such individuals would benefit from traditional SBIRT programs. Yet, few studies have compared the effects of BI on individuals with mental illness and those without.
This was a secondary data analysis of survey and administrative data gathered as part of the Washington State SBIRT (WASBIRT) program. The WASBIRT study protocol was reviewed and approved by the Washington State Institutional Review Board and three institutional review boards that oversee research within specific participating hospitals.
Descriptive information about the two samples (statewide and HMC) is presented in Table 1. At baseline, approximately 49% of the statewide sample reported using illegal drugs in the previous 30 days (15% cocaine, 35% marijuana, 10% methamphetamine, 4% heroin, 6% prescription opiates, and 5% other illegal drugs). Approximately 58% of the HMC sample reported using illegal drugs in the previous 30 days (29% cocaine, 36% marijuana, 9% methamphetamine, 9% heroin, 5% prescription opiates, and 7% other illegal drugs).
Results reported here are consistent with much of the existing literature showing that self-reported hazardous alcohol/drug use decreases six months after individuals receive a BI through an SBIRT Program. They extend existing findings by suggesting that the BI may not have a differing impact based on the presence of a mental illness diagnosis. Given that mental illness is estimated to coexist in 37% of individuals with an alcohol disorder and more than 50% of individuals with a drug disorder, this finding has important public health implications.
All authors are currently employees of either the University of Washington or the Washington State Department of Social and Health Services and have no conflicts of interest or financial interests.
AK conceived, wrote, and edited the manuscript and made substantial contributions to analytic design and data/interpretation. JMS designed and executed the analyses and statistical approach, took full responsibility for data management for the HMC sample, and wrote the methods and results sections of the manuscript. JMJ provided methodological and statistical support; she also made substantial contributions to the design of the analysis, interpretation of the data, and review and editing of the manuscript. SE, Director of the Washington State SBIRT Evaluation Project, made substantial contributions to the conception and design of the analysis, acquisition of data, and drafting, review, and editing of the manuscript. LH contributed to the acquisition of data, consulted on the statistical approach, and executed data analyses for the statewide sample. AH made substantial contributions to the design of the analysis and participated in the review and editing of the manuscript. PR-B, CD, and RR participated in the review and editing of the manuscript. All authors read and approved the final manuscript.