Specialty substance use disorder services following brief alcohol intervention: a meta-analysis of randomized controlled trials May 27, 2016 Joseph E. Glass, PhD, MSW Assistant Professor School of Social Work University of Wisconsin-Madison
REVIEW doi:10.1111/add.12950 Specialty substance use disorder services following brief alcohol intervention: a meta-analysis of randomized controlled trials Joseph E. Glass 1, Ashley M. Hamilton 2, Byron J. Powell 3, Brian E. Perron 4, Randall T. Brown 5 & Mark A. Ilgen 6 School of Social Work, University of Wisconsin-Madison, Madison, WI, USA, 1 Chrysalis, Inc., Madison, WI, USA, 2 Department of Psychiatry, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA, 3 School of Social Work, University of Michigan, Ann Arbor, MI, USA, 4 Department of Family Medicine, School of Medicine and Public Health, University of Wisconsin-Madison, Madison, WI, USA 5 and VA Center for Clinical Management Research (CCMR), VA Ann Arbor Healthcare System and the Department of Psychiatry, University of Michigan Medical School, Ann Arbor, MI, USA 6 ABSTRACT Background and aims Brief alcohol interventions in medical settings are efficacious in improving self-reported alcohol consumption among those with low-severity alcohol problems. Screening, Brief Intervention and Referral to Treatment initiatives presume that brief interventions are efficacious in linking patients to higher levels of care, but pertinent evidence has not been evaluated. We estimated main and subgroup effects of brief alcohol interventions, regardless of their inclusion of a referral-specific component, in increasing the utilization of alcohol-related care. Methods A systematic review of English language papers published in electronic databases to 2013. We included randomized controlled trials (RCTs) of brief alcohol interventions in general health-care settings with adult and adolescent samples. We excluded studies that lacked alcohol services utilization data. Extractions of study characteristics and outcomes were standardized and conducted independently. The primary outcome was post-treatment alcohol services utilization assessed by self-report or administrative data, which we compared across intervention and control groups. Results Thirteen RCTs met inclusion criteria and nine were meta-analyzed (n = 993 and n = 937 intervention and control group participants, respectively).
Primary Rationale and Aims Estimate the main and subgroup effects of brief alcohol interventions in increasing the utilization of alcohol-related services Secondary To estimate the association of post-sbi treatment utilization with alcohol-related outcomes
Methods A systematic review and meta-analysis of RCTs published in electronic databases through 2013 Inclusion criteria Intervened with alcohol use in a medical setting Reported alcohol services utilization as an outcome Not a treatment-seeking sample English language Exclusion criteria Drug/alcohol samples without alcohol-specific analysis Alcohol services outcome delivered as part of the RCT Primary care-behavioral health integration
Measures Post-intervention alcohol services utilization: Formal treatment and/or mutual help Type of medical setting: Primary care (PC), inpatient, emergency department (ED), other Age: Adolescent vs. adult sample Sample Severity: High Strictly alcohol dependent, alcohol detox, alcohol-induced medical problems Low Excluded dependent drinkers Mixed Dependent and non-dependent drinkers Intervention intensity: Low: no in-person contact, Medium: single session, High: multiple sessions
Analyses Meta-analysis compared alcohol services utilization across study arms using STATA metan package Calculated risk ratios and 95% CIs I 2 statistic to measure heterogeneity across studies Subgroup analyses of studies with similar characteristics Qualitative review of interventions that attempted to evaluate the association between post-intervention alcohol services utilization and study outcomes
Results Database search (n = 637) Other sources (n = 35 hand search, n = 4 expert query) Abstracts screened (n = 676) Full-text articles assessed for eligibility (n = 111) Included studies (n = 13) Excluded (n = 565) Excluded (n = 98)
Table 1. RCTs examining the association between brief alcohol intervention and post-intervention utilization of alcohol-related care (n = 13) Study Setting Effect? Group n, % who received services Intervention Control Batel 1995 ED Yes 188 11.2% 181 1.1% Bernstein 2010 ED No 207 1.9% 209 2.4% Bischof 2008 PC No 37 18.9% 36 11.1% Cherpitel 2010 ED No 80 2.5% 97 1.0% Copeland 2003 PC No 100 N/A 105 N/A Crawford 2004 ED No 131 32.8% 159 30.8% Field 2010 ED No N/A N/A N/A N/A Gentilello 1999 ED No 194 4.6% 215 4.7% Kuchipudi 1990 Inpatient No 59 20.3% 55 16.4% Monti 2007 ED No 75 29.3% 80 20.0% Monti 1999 ED No 47 23.0% 37 18.0% Saitz 2007 Inpatient No 107 56.1% 105 56.2% Wutzke 2002 Various No N/A N/A N/A N/A
n = 1,930 patients in 9 studies Study % ID RR (95% CI) Weight Bernstein (2010) 0.81 (0.22, 2.97) 1.67 Bischof (2008) 1.70 (0.54, 5.32) 2.17 Cherpitel (2009) 2.42 (0.22, 26.26) 0.50 Crawford (2004) 1.07 (0.76, 1.49) 24.69 Gentilello (1999) 1.00 (0.41, 2.40) 3.65 Kuchipudi (1990) 1.24 (0.57, 2.72) 4.61 Monti (1999) 1.24 (0.53, 2.88) 3.96 Monti (2007) 1.47 (0.84, 2.57) 8.94 Saitz (2007) 1.00 (0.79, 1.27) 49.81 Overall (I-squared = 0.0%, p = 0.924) 1.08 (0.92, 1.28) 100.00 NOTE: Weights are from random effects analysis.0381 1 26.3 Comparison condition is better Intervention condition is better
Subgroup analyses No subgroup analyses of studies (stratified by age, setting, severity, intervention intensity) yielded statistically significant results Five studies compared referral-specific interventions to a control group that did not employ similar referral interventions (pooled RR=1.08, 95% CI=0.81-1.43) Originally published: 1,930 participants Re-analysis 2,380 participants Letter to the Editor Revised results: RR = 1.16 (0.96-1.40) (NS) I 2 6 months by injured patients from two emergency departments compared with 6.8% an inactive control (p condition > 0.05) REVISITING OUR REVIEW OF SCREENING, BRIEF INTERVENTION AND (NS) REFERRAL TO TREAT- [19.2% compared with 4.5%; odds ratio (OR) = 5.1, 95% MENT (SBIRT): META-ANALYTICAL RESULTS Liu et al, 2011, SBI Medical/Surgical wards in Taiwan: confidence interval (CI) =2.1 12.2]. The RCT by Apodaca STILL POINT TO NO EFFICACY IN INCREASING et al. [2] reported an increase in further treatment-seeking THE USE OF SUBSTANCE USE DISORDER at 5 months among a sample of in-patients from a trauma SERVICES centre with a mean Alcohol Use Disorders Identification Test (AUDIT) score of 20 for brief advice, compared with We appreciate the response to our meta-analysis [1] by 8.3% (n = 19) vs. Simioni 2.1% and colleagues [2]. They (n published = two systematic 4) Letter to the Editor IS THERE REALLY NO EVIDENCE OF THE EFFI- CACY OF BRIEF ALCOHOL INTERVENTIONS FOR INCREASING SUBSEQUENT UTILIZATION OF ALCOHOL-RELATED SERVICES? COMMEN- TARY ON THE PAPER BY GLASS ET AL. (2015) Recently, a meta-analysis of randomized controlled trials (RCTs) on utilization of substance abuse services following brief alcohol interventions in general health-care settings by Glass et al. [1] has been published in Addiction. The authors aimed to estimate the main and subgroup effects of brief alcohol interventions, regardless of their inclusion of a referral-specific component, in increasing the utilization of alcohol-related care. To identify RCTs, the control group [6 of 15 (40%) versus two of 15 (13%)], but this increase was not found statistically significant given the small sample size (n = 40). The study by Liu et al. [3] (n=616) also reported that brief intervention with post-discharge sessions was associated significantly with treatment utilization at 12 months in in-patients reviews this year in other peer-reviewed addiction journals on this topic [3,4]. The major differences between our review and theirs is that our study meta-analyzed the data, whereas they were focused on providing an overview of the literature, and each research team chose different eligibility The end-point for the third, unpublished study was an assessment to see if treatment was needed [7], whereas we were interested in treatment utilization. There were also issues of bias in how the outcome was assessed that would have led to its exclusion had we identified it. Simioni and colleagues posed the question: Is there really no evidence of the efficacy of brief alcohol interventions for increasing subsequent utilization of alcoholrelated services?. We believe the answer is still yes, but there is much room for innovation. One of the published RCTs identified by their team was a pilot study of a bibliotherapy intervention among emergency department patients, which was not designed to detect statistically significant effects and did not detect them [6]. However, it
Limitations Assessment of alcohol-related care varied across studies Studies had limited descriptions of RT-related interventions Many RCTs of brief interventions have been conducted, but most do not assess treatment utilization Did patients receive SBI, decrease consumption, then no longer need treatment?
Relevant studies not meeting metaanalytic criteria Controlled clinical trials Studies of other potentially relevant outcomes Initiation of substance use disorder evaluations Research intervention appointments Blow et al., JSAD, 2010 Chafetz et al., J Nerv Ment Dis, 1962 Elvy et al., Br J Addict, 1988 Goldberg et al., Medical Care, 1991 Runge et al., Unpublished, 2002 Siomioni et al., JSAT, 2015
Implications Calling RT of SBIRT evidence-based may mislead us What do we do? Identify and implement AUD interventions within the medical setting RT in novel care models (shared decision making, collaborative care, chronic care management) Bradley et al., JAMA, 2014 Oslin et al., JGIM, 2014 Saitz et al., JAMA, 2014
COMMENTARIES Commentaries on Glass et al. (2015) SBIRT IS THE ANSWER? PROBABLY NOT Screening, brief intervention and referral to treatment (SBIRT) addresses the full spectrum of unhealthy substance use [1]. It sounds like the answer to the question: how can we reduce substance use and disorders? by addressing everything except the delivery of specialized treatment itself. The best evidence suggests that brief intervention among people who drink risky amounts identified by screening (SBI) can reduce self-reported consumption very modestly in primary care patients (among those advised to drink less who are then asked if they have done so). The limitation of efficacy to a narrow band of those who drink too much but not too much is demonstrated by the disparate results of two emergency department SBI studies at the same site that differed largely in drinking eligibility/exclusion criteria [2,3]. The evidence for effects of alcohol SBI on any clinically important outcomes is limited, despite half a century of randomized trials and meta-analyses [4,5]. Studies have largely ignored patients with more severe unhealthy use by excluding such participants [6]. Glass et al. [7] assessed the evidence for the main way in which SBIRT addresses more severe unhealthy use referral to treatment. They find randomized-controlled trial evidence that RT in BRIDGING THE GAP BETWEEN MEDICAL SETTINGS AND SPECIALTY ADDICTION TREATMENT In their study, Glass et al. [1] examined whether Screening, Brief Intervention and Referral to Treatment (SBIRT) is efficacious for linking medical-setting patients with alcohol misuse, but not seeking alcohol-related care, to specialty addiction treatment. The results of their meta-analysis indicate that the referral to treatment component of SBIRT is not effective for promoting this care linkage. We commend the authors for highlighting the need to specify and evaluate interventions that can bridge this care gap. This commentary briefly summarizes some of the challenges to bridging the gap between medical settings and specialty addiction treatment and promising approaches to promoting this care linkage. The authors bring to our attention that substantial barriers exist to linking patients in medical settings with alcohol misuse to specialty addiction treatment. Barriers to linking patients to addiction treatment are well documented and can be categorized as those pertaining to the patient, provider and care system [2]. Patient-level barriers include not perceiving oneself as in need of services, difficulties accessing treatment, stigma associated with receiving treatment and limited nancial resources [3,4]. CHALLENGES AHEAD IN DEVELOPING AND TESTING REFERRAL TO TREATMENT INTERVENTIONS We are grateful for these thoughtful commentaries on our meta-analysis [1]. Saitz [2] and Cucciare & Timko [3] raise important issues about the numerous challenges related to linking individuals with alcohol use disorders who are identified in medical care to appropriate alcohol treatment services. As described by Saitz [2], a 10 15-minute brief intervention is probably too brief to help most people develop sufficient motivation to seek addiction treatment. Cucciare & Timko [3] have described additional barriers to seeking treatment and potential solutions to providing treatment linkages [4]. Medical patients in primary care settings with more severe unhealthy alcohol use exhibit higher levels of readiness to change [5], and thus may be further along on the pathway of being ready to seek treatment. However, our meta-analysis [1] supports the assertion that Screening, Brief Intervention and Referral to Treatment (SBIRT) may be insufficient to lead to help-seeking, regardless of patient severity, even when supplemented by a booster session. There is consensus in these commentaries that repeated contact and monitoring over time seems to be a more sensible solution. Cucciare & Timko [2] described case man- Comprehensiveness and feasibility are they at odds? Saitz, Addiction, 2015; Cucciare & Timko, Addiction, 2015; Glass, Addiction, 2015
Enhancing RT Addressing limitations of prior studies Treatment Utilization RT Drinking Outcomes Barnett et al., Addiction, 2010 Bertholet et al., JGIM, 2010 Elvy, et al., Br J Addict, 1988
Conclusions Brief alcohol interventions as currently tested and implemented are not efficacious in linking patients to specialty treatment services More dialogue is needed on RT development
Thank you! 17