An empirical model of therapeutic process for psychiatric emergency room clients with dual disorders

Abstract

Many individuals with dual disorders of mental illness and substance abuse enter the mental health system through psychiatric emergency rooms (PERs), but may resist treatment and not follow through with referrals to services. This study examined the impact of therapeutic process on referral outcome. Raters used the Vanderbilt Psychotherapeutic Process Scales and Working Alliance Inventory to evaluate audiotapes and transcripts of 39 PER sessions. Outcome was successful if clients attended all referral appointments. The authors found that therapist warmth and friendliness had a positive association with working alliance, which, in turn, was associated with successful referral. Paradoxically, there was a negative relationship between warmth and friendliness and success. Analysis demonstrated that, for a given level of warmth and friendliness, there must be a correspondingly higher level of working alliance to be associated with success. Key words: dual disorders; mental illness; psychiatric emergency rooms; substance abuse; therapeutic process ********** Individuals with dual disorders of mental illness and substance abuse exhibit a range of related difficulties, including suicidality, violence, homelessness, and arrest (Cornelius et al., 1995; Havassy & Arns, 1998; Pages, Russo, Roy-Byrne, Pies, & Cowley, 1997; Swartz et al., 1998). The literature indicates that substance use by those with a mental illness results in new difficulties such as depression, anxiety, disinhibition, and aggression (Brunette, Mueser, Xie, & Drake, 1997; Serper et al., 1995); exacerbates existing psychiatric symptoms; leads to noncompliance with medications or with therapy; and results in poorer treatment outcomes (Drake, Bartels, Teague, Noordsy, & Clark, 1993; Haywood et al., 1995; Mueser, Bellack, & Blanchard, 1992; Owen, Fischer, Booth, & Cuffel, 1996; Swofford, Kasckow, Scheller-Gilkey, & Inderbitzin, 1996). Even when medication compliance is controlled, there is evidence that those who abuse substances are more likely to relapse and require readmission than those who do not abuse substances (Gupta, Hendricks, Kenkel, Bhatia, & Haffke, 1996). However, clients with a dual diagnosis often deny their substance abuse problem, their mental health problem, or both (for example, Brady et al., 1996; Drake et al., 1990), making it difficult to engage them in treatment (Brady et al.; Drake et al., 1993; Lehman, Herron, Schwartz, & Myers, 1993). Because the substance use exacerbates their symptoms, clients with dual disorders tend to present in crisis (McCarrick, Manderscheid, & Bertolucci, 1985) and often enter the mental health system through psychiatric emergency rooms (PERs) (Elangovan et al., 1993). Unfortunately, although one of PERs' primary tasks is assessment and referral to services in the community, only about half of clients with dual disorders follow through with referrals (Solomon & Gordon, 1988). The purpose of our investigation was to develop an empirical model of therapeutic process for PER clients with dual disorders of mental illness and substance abuse. For the general population of psychiatric clients, a number of factors in the therapeutic process have been shown to have an effect on emergency room outcomes. Before discussing these factors it is necessary to present a framework from which to examine therapeutic process. Greenberg (1986) identified three major levels of therapeutic process. The first level, speech act, is the most basic element of the framework and is the type of statement made by the client (for example, resistance) or by the clinician (for example, question, confrontation). The second level, episode, consists of a set of speech acts made between the client and the clinician (the attitudes and behaviors of both the client and the clinician within a session). Episodes focus on a particular problem and, in the context of PERs, the focus is on identification of client needs and referral to appropriate services. …