Cognitive-Behavioral Therapy for Schizophrenia: Applications to Social Work Practice

Abstract

Schizophrenia is a psychotic disorder that has been considered to be the epitome of a severe mental illness. The negative psychosocial consequences of schizophrenia are well documented. Despite the advent of antipsychotic medication, residual symptoms persist for many persons diagnosed with schizophrenia. Cognitive-behavioral therapy (CBT) has emerged as an adjunctive treatment to pharmacotherapy. Cognitive-behavioral theories of positive and negative symptoms are described, as are interventions. Meta-analyses have supported the use of CBT for schizophrenia. This article describes and synthesizes the literature to articulate precise applications to social work practice and the congruence between CBT for schizophrenia and the social work profession. Limitations of CBT are also discussed.

KEYWORDS:

• Cognitive-behavioral
• schizophrenia
• randomized controlled trial
• social work
• meta-analysis

Many consider schizophrenia to be the epitome of a severe mental illness (SMI) (Hofmann & Tompson, 2002). Characterized by hallucinations, delusions, and other psychotic symptoms (American Psychiatric Association, 2000), schizophrenia frequently has a deleterious influence on various aspects of idiographic functioning (Hofmann & Tompson, 2002). Schizophrenia is associated with lack of employment (Rosenheck et al., 2006), increased risk for homelessness (Folsom & Jeste, 2002), as well as marital discord (Hooley, Richters, Weintraub, & Neale, 1987). Symptoms of schizophrenia are linked to impairment in academic performance, daily living activities, parenting, and social relationships (Mueser & McGurk, 2004). In the United States, approximately 5 out of 1000 people are diagnosed with schizophrenia (Wu, Shi, Birnbaum, Hudson, & Kessler, 2006). After neuroleptics began to show an ability to decrease positive symptoms, treatment of schizophrenia switched from psychodynamic therapy to biologically based intervention (Pratt & Mueser, 2002). Currently, the United States views neuroleptic (antipsychotic) treatment as the fundamental aspect of intervention for schizophrenia (Turkington, Kingdon, & Weiden, 2006).

Although neuroleptics have had some success in addressing schizophrenic symptoms and reducing relapse, symptoms persist for numerous persons (Gould, Mueser, Bolton, Mays, & Goff, 2001). Despite the advantages of neuroleptics, individuals diagnosed with schizophrenia have high relapse rates (Ilott, 2005). One fourth to half of individuals who adhere to pharmacological treatment still have considerable difficulty (Rector, 2004). The limited symptomatic benefit of neuroleptics for schizophrenia is illustrated by the Food and Drug Administration's (FDA's) more liberal requirements for symptom improvement (Patterson, Albala, McCahill, & Edwards, 2006). To be approved by the FDA neuroleptics must demonstrate a 20% to 30% decrease in symptoms relative to the placebo. In contrast, the FDA requires that antidepressant medications show a 50% decrease in symptoms. With regard to the outcome of schizophrenia, “… approximately 20% make a full recovery, 20% have relapses with no intervening deterioration, 40% have relapses with some deterioration, and fewer than 20% remain chronically ill and show little recovery” (Kingdon & Turkington, 2005, p. 2). For therapy to have maximum effectiveness, pharmacological intervention for schizophrenia should be supplemented by psychotherapeutic treatment (Patterson et al., 2006).

Effective psychosocial treatment like cognitive-behavioral therapy (CBT) is needed to complement pharmacological interventions for schizophrenic symptoms (Bieling, McCabe, & Antony, 2006). Previously it was believed that clients with schizophrenia could not be treated with CBT, yet this is not true (Gould et al., 2001). Cognitive-behavioral therapy has become a recognized treatment for psychosis in the United Kingdom (Tarrier & Haddock, 2002). Despite its availability in the United Kingdom, there has been a lack of attention on CBT for schizophrenia in the United States (Turkington et al., 2006). Although studies (Bradshaw, 1997, 2003; Bradshaw & Roseborough, 2004) have provided insight regarding CBT for social work practice with persons who have schizophrenia, the applicability of CBT for schizophrenia in social work practice can be improved via reviewing the theory, practice, randomized controlled trials (RCTs), and effect sizes of CBT for schizophrenia. According to Soydan (2008), “… when it comes to measure the effects of social work interventions, experimental studies, especially when randomized, conducted very carefully, and large enough to generate statistical power, are the designs that best fit for the purpose” (p. 313). Although the implications of CBT for schizophrenia are widely available, explicit applications for social work are less prominent. In an effort to increase U.S. social workers' awareness, application, and evaluation of CBT for schizophrenia, this review has several purposes. This article seeks to explain cognitive-behavioral theories and interventions for schizophrenic symptoms, describe the experimental status of CBT for schizophrenia, synthesize the literature to articulate precise applications to social work practice and research, and explicate the congruence between CBT for schizophrenia and the social work profession.

COGNITIVE-BEHAVIORAL THEORIES OF SCHIZOPHRENIA

Vulnerability

Theories of schizophrenia have been used to explain causal factors, identify targets for intervention, and guide the psychosocial treatment process. Acknowledged in cognitive-behavioral treatment for schizophrenia (Fowler, Garety, & Kuipers, 1995; Kingdon & Turkington, 1994, 2005), the vulnerability model (Zubin & Spring, 1977) maintains that individuals have a certain level of susceptibility to schizophrenia and the individual's response to stressful situations will mediate the ultimate manifestation of schizophrenic symptoms. Intrinsic vulnerability to schizophrenia includes genetic risk factors. Extrinsic vulnerability pertains to life events such as traumatic experiences, stressful situations, disease, and familial and social experiences. Zubin and Spring consider positive (career advancement) and negative changes (divorce) as life events. Once schizophrenia has developed, the vulnerability model suggests that individuals try to adapt via “accommodation” or “assimilation.” Accommodation refers to intrapersonal changes the individual makes to adjust to environmental demands. Assimilation occurs when individuals seek to change their environmental circumstances to cope with stressors. These two processes can either enable or impede adaption (healthy responses to stressors).

The vulnerability model (Zubin & Spring, 1977) has cognitive-behavioral implications in that it states “… it must be remembered that the stress value of various life events depends on the perception of threat by the individual” (p. 114). Likewise, the cognitive-behavioral model of psychopathology maintains that affective and behavioral responses are determined by the individual's appraisal of events, rather than the events themselves (Beck, 1995). Since Beck first used CBT to treat schizophrenia in 1952, cognitive-behavioral theories of hallucinations (Birchwood & Chadwick, 1997; Chadwick & Birchwood, 1994) and delusions (Garety, Kuipers, Fowler, Freeman, & Bebbington, 2001) have advanced case formulation for and treatment of clients with schizophrenia.

Auditory Hallucinations

The cognitive model of auditory hallucinations maintains that clients' maladaptive beliefs about voices cause undesirable emotional and behavioral consequences (Chadwick & Birchwood, 1994). Believing that the voices intend to do harm (malevolent) causes the person to feel emotions such as sadness, angst, fright, and anger. With regard to behavior, malevolent beliefs about voices lead individuals to argue with voices, yell, and avoid stimuli that elicit voices. Chadwick and Birchwoods' model also implies that emotive and behavioral outcomes are influenced by beliefs the client has about whom the voice belongs to and how much power the voice possesses.

There is empirical support for the cognitive model of auditory hallucinations (Birchwood & Chadwick, 1997). Persons diagnosed with schizophrenia or schizoaffective disorder completed the beliefs-about-voices questionnaire (BAQV) (Chadwick & Birchwood, 1995) and were classified as having malevolent, benevolent (believing the voices have good intentions), or benign beliefs about voices. With regard to the level of distress for the three different belief systems, there were statistically significant differences. The malevolent and benevolent groups experienced the most and least amount of distress, respectively. The study demonstrated that persons who had malevolent beliefs about their voices had a statistically significant greater chance of having at least moderate depressive symptoms, in comparison to persons who had benevolent beliefs. Participants with malevolent beliefs about voices had significantly more positive symptoms than participants with benign beliefs about voices. The cognitive model of auditory hallucinations suggests that distress could me minimized via challenging and replacing core beliefs about the intention, authority, and identity of voices (Chadwick & Birchwood, 1994). If clients with schizophrenia believed that their auditory hallucinations were generated from themselves, the resulting affect and conduct may be substantially different (Chadwick & Birchwood, 1994).

Delusions

Maher's (1974) theory is frequently cited in cognitive-behavioral literature regarding the explanation for (Bentall, Corcoran, Howard, Blackwood, & Kinderman, 2001; Bentall, Kinderman, & Kaney, 1994; Birchwood & Jackson, 2001; Freeman, 2007; Garety et al., 2001) and treatment of delusional beliefs (Chadwick, Birchwood, & Trower, 1996; Chadwick & Lowe, 1990, 1994). Maher maintained that delusional beliefs develop from the individual's attempt to explain unusual perceptual stimuli (hallucinations). According to Maher's theory, delusions provide comprehensive explanations that account for the hallucination's occurrence, origin, and idiosyncratic nature. The explanatory function of delusions are said to provide a reduction in anxiety. Specific content of the delusion is believed to be a product of the individual's previous and current experiences. Maher's theory views delusions as an ordinary response to atypical stimuli. The theory does not claim to apply to all persons with delusions. Cognitive-behavioral therapy seeks to aid the individual in understanding that the delusion is the following: perceptual rather than actual, an attempt to explain their experience, accompanied by behavioral and emotive disadvantages, false and inferior to a more reasonable belief (Chadwick et al., 1996).

Negative Symptoms

The cognitive model acknowledges the role of biological vulnerability and suggests that negative symptoms in schizophrenia are partially influenced by beliefs pertaining to social interaction, pleasure, success, and resources (Rector, Beck, & Stolar, 2005). In describing and justifying their cognitive model, Rector and associates rely on published studies. Clients with negative symptoms are theorized as having beliefs that are antithetical to social interaction. Such clients are also viewed as having pessimistic beliefs about their intrapersonal and social adequacy. The cognitive model maintains that negative symptoms are in part facilitated by the client not expecting to obtain satisfaction or achievement from potentially gratifying activities. The authors stated that in reality clients do receive some gratification from activities when they participate in them. The model identifies the client's tendency to underestimate resources as a key part in the maintenance of negative symptoms. Cognitive-behavioral techniques used (Beck, Rush, Shaw, & Emery, 1979) for depression are also applied to the negative symptoms of schizophrenia (Rector, 2004). The aforementioned theories of vulnerability, auditory hallucinations, delusions, and negative symptoms are only four of the theoretical explanations that have been submitted to explain symptoms of schizophrenia. These theories were described here because of the frequency with which they are cited in cognitive-behavioral literature, their contribution to the symptomatic understanding of schizophrenia, and the practical implications they have for treatment. Cognitive-behavioral therapy for schizophrenia has been substantially influenced by these theories. Given the cognitive-behavioral theoretical underpinnings of schizophrenic symptoms, case formulation and CBT are logical extensions.

COGNITIVE-BEHAVIORAL INTERVENTIONS FOR SCHIZOPHRENIA

Cognitive-behavioral interventions for schizophrenia are currently available in several treatment manuals (Beck, Rector, Stolar, & Grant, 2009; Byrne, Birchwood, Trower, & Meaden, 2006; Chadwick et al., 1996; Fowler et al., 1995; Kingdon & Turkington, 1994, 2005; Morrison, Renton, Dunn, Williams, & Bentall, 2004). Social workers who use cognitive-behavioral techniques to treat other disorders will be familiar with many of the interventions that are described in these manuals. For example, common cognitive restructuring interventions like systematically evaluating the evidence for and against beliefs (Beck et al., 1979) are used in the treatment of schizophrenia. Conventional behavioral interventions like behavioral experiments and activity scheduling (Greensberger & Padesky, 1995) are frequently employed as well. Table 1 provides a list of cognitive-behavioral interventions and treatments manuals that discuss their application. Table 1 is not meant to be an exhaustive reiteration of the techniques described in the cited treatment manuals. Instead, the purpose of Table 1 is to provide social workers with an overview of cognitive-behavioral techniques that are frequently used in the treatment of schizophrenia and a list of treatment manuals where interventions are explicated. The cognitive-behavioral techniques are classified according to whether they are primarily used to identify and monitor beliefs, alter maladaptive beliefs, or cope with symptoms. The classification of the interventions are somewhat relative, as the techniques are not necessarily mutually exclusive with regard their purposes.

TABLE 1 Cognitive-Behavioral Interventions Used in the Treatment of Schizophrenia

Identifying thoughts and beliefs

ABC Technique

Downward Arrow Technique

Exposure

Thought records

Challenging maladaptive thoughts and beliefs

Alternative explanations for events

Behavioral experiments

Evidence for and against beliefs

Imagery

Pie charts

Pros and cons of beliefs

Psychoeducation

Relaxation

Coping skills

Activity scheduling

Distraction

Flashcards

Normalizing

Pie charts

Prodrome monitoring

Role-plays

CBT Treatment Manuals: Beck et al. (2009); Byrne et al. (2006); Chadwick et al. (1996); Fowler et al. (1995); Kingdon & Turkington (1994, 2005); Morrison et al. (2004).

In conjunction with other factors, the applicability of cognitive-behavioral interventions is determined by the extent of the client's engagement (Haddock & Siddle, 2003). Cognitive-behavioral therapy must include a strong therapeutic relationship between the clinician and client (Beck, 1976, 1995). Consequently, the development of therapeutic rapport is highly important (Kingdon & Turkington, 2002) and the initial goal of CBT (Tarrier, 2005). The therapeutic relationship is essential to case formulation and the application of cognitive-behavioral interventions to persons with schizophrenia (Hewitt & Coffey, 2005). Failure to develop a therapeutic relationship will minimize the effectiveness of cognitive-behavioral techniques (Beck, Wright, Newman, & Liese, 1993). Fortunately, although engaging a client with schizophrenia can seem to be a huge barrier, the process may not be as complicated as it initially looks (Kingdon & Turkington, 2005). Explicit details regarding how to build therapeutic rapport are beyond the scope of this article; however Kingdon and Turkington (2005) have provided a chapter length summary regarding ways to facilitate the therapeutic relationship with clients who have schizophrenia.

Cognitive-behavioral therapy is used in the treatment of schizophrenia to accomplish specific psychosocial outcomes. Objectives of CBT for schizophrenia include reductions in the frequency and intensity of positive symptoms, stigma, relapse, and co-occurring depression and anxiety (Rector, 2005). An associated objective of CBT is the development of the client's insight (Kingdon & Turkington, 1994). Insight refers to the client's acknowledgment that he or she has a psychiatric illness, attribution of hallucinations and delusions to psychiatric illness, and adherence to treatment (David, 1990). Cognitive-behavioral therapy for auditory hallucinations enhances client insight via assisting the client in recognizing that the voices are attributable to herself or himself (Chadwick & Birchwood, 1994).

Similar to some of the aforementioned goals of CBT for schizophrenia, psychotropic medication has been beneficial in decreasing the frequency and intensity of symptoms (Kingdon & Turkington, 1994). Psychosocial treatment of schizophrenia is maximized when it is used in conjunction with medical intervention (“Psychosocial Interventions,” 2005). When used simultaneously, the treatments collectively increase compliance and communal functioning (“Pharmacotherapy,” 2005). It is not currently known if CBT can be effective for schizophrenia in the absence of pharmacotherapy (Kingdon & Turkington, 2005). Evaluating the effectiveness of CBT, as a stand alone treatment for schizophrenia, raises ethical issues (Kingdon & Turkington, 2005). The absence of research examining CBT as a stand alone treatment is attributable to the severe nature of schizophrenia and the effectiveness of neuroleptics (Wright, 2004). To the contrary, numerous RCTs have examined the effectiveness of CBT with pharmacotherapy for schizophrenia in comparison to control groups.

METHODS

Given the many experimental trials that have examined CBT for schizophrenia over the past 30 years and the purpose of this review, this article focuses exclusively on meta-analytic studies. Meta-analyses are beneficial to summarizing research because they provide structure to synthesizing data coming from different studies, report effect sizes rather than probabilities regarding null hypothesis significance tests, and have the potential to determine relationships that may be masked by other summary methods (Miller & Salkind, 2002). The present study used Medline and PsychInfo databases (November 2008) to systematically obtain meta-analyses pertaining to CBT for schizophrenia. Keywords such as cognitive behavioral, cognitive behavior, CBT, cognitive therapy, meta-analysis, effect size, schizophrenia, randomized controlled trial, and RCT were used to identify pertinent meta-analytic studies. Reference lists of the included meta-analytic studies and seminal texts were reviewed to identify relevant meta-analyses that may not have been published in peer-reviewed journals.

Each of the meta-analytic studies was subject to inclusion and exclusion criteria. For a meta-analysis to be included in the current review the following inclusion criteria must have been met: (1) the study must be a meta-analysis, (2) CBT is identified as the independent variable in the analysis, (3) participants were classified as having a schizophrenic spectrum disorder, (4) the number of studies using a RCT is reported, (5) the average standardized mean difference effect size (see Henson, 2006; Rosnow & Rosenthal, 2003 for discussion) of positive (hallucinations, delusions) and/or negative symptoms (anhedonia, avolition, etc.) is explicitly reported, (6) the effect size represents a posttest or pretest to posttest comparison, (7) the statistic used to calculate the effect size is stated, and (8) a systematic literature search strategy is conducted and described. Meta-analytic studies were excluded if a comorbid DSM-IV axis I or II disorder was the specific focus of evaluation or if the study was a review of meta-analyses. Of the ten studies that were identified, four meta-analyses met inclusion criteria for this review. Excluded studies and their reason for exclusion can be found in the Appendix.

RESULTS

Research has shown CBT to be effective in challenging and altering maladaptive beliefs that contribute to hallucinations and delusions (Rector & Beck, 2002). As can be seen in Table 2, the average effect sizes for positive symptoms of schizophrenia show the CBT cohorts as having small to large treatment effects. The meta-analytic studies included in this review (Gould et al., 2001; Rector & Beck, 2001; Wykes, Steel, Everitt, & Tarrier, 2008; Zimmermann, Favrod, Trieu, & Pomini, 2005) provide insight and support regarding the efficacy of cognitive-behavioral interventions for positive and negative symptoms of schizophrenia. Additionally, the meta-analyses are overwhelmingly composed of experimental studies that encompass an interval of over 25 years. Meta-analyses which incorporate RCTs are optimal because such designs control threats to internal validity (Campbell & Stanley, 1963) and increase the likelihood of a cause-and-effect relationship (Kirk, 1999) between CBT and a reduction in schizophrenic symptoms.

TABLE 2 Cognitive-Behavioral Therapy for Schizophrenia—Meta-Analyses

Meta-analysis	Total number of studies	Number of RCTs (%)	ES statistic	Mean ES	Types of control groups	n	Attrition	Article timeframes
Gould et al. (2001)	7	7 (100%)	Glass's Delta	Positive symptoms = 0.65 ^a	ST, TAU, WL	340	12.4%	1978–2000
Rector & Beck (2001)	7	7 (100%)	Cohen's d	Positive symptoms = 1.31 ^b Negative symptoms = 1.08 ^b	ST, TAU	383	N/A	1993–2000
Wykes et al. (2008)	34	30 (88%)	Glass's Delta	Positive symptoms = 0.37 ^a ^d Negative symptoms = 0.43 ^a ^d	ST, TAU	58.2 ^c	Seven studies with an attrition rate above 25%	1978–2006
Zimmermann et al. (2005)	14	13 (92%)	Hedge's g	Positive symptoms = 0.37 ^a ^d	ST, TAU, WL	1484	N/A	1993–2004
^aPosttest effect size.
^bPretest to posttest effect size.
^cAverage.
^dRandom Effects Model (REM).
Acronyms: ES = Effect Size; n = Sample Size; N/A = Not Applicable, Limited or no information reported; RCT = Randomized Controlled Trials; ST = Supportive Therapy; TAU = Treatment as Usual; WL = Wait-List.

Effect sizes calculated by Rector and Beck (2001) demonstrate that on average CBT had a large effect size for positive and negative symptoms, while control groups had moderate effect sizes for the two sets of symptoms. All seven studies in Rector and Beck's analysis were RCTs and the experimental and control cohorts were all receiving pharmacotherapy. Gould and colleagues' meta-analysis included seven RCT's, yielded a large effect size at post-treatment, and supported CBT's ability to reduce the intensity of positive symptoms. Of the seven studies, one study involved 95% of the participants receiving pharmacotherapy, while all of the participants in the remaining six studies received pharmacological intervention.

The meta-analysis conducted by Zimmermann and associates showed CBT as significantly reducing positive symptoms of schizophrenia. The study found that at post-treatment, the average participant in CBT had greater reductions in positive symptoms than 64% of persons in control groups. Of the 14 studies included in the meta-analysis, 13 were RCTs. To be included in the Zimmermann and associates meta-analysis a supportive therapy, treatment as usual, or wait-list control was required. However, the meta-analysis lacked adequate discussion pertaining to the pharmacological intervention that control groups received.

The most extensive meta-analysis conducted thus far shows CBT as having favorable effects on positive, negative, and affective symptoms (Wykes et al., 2008). Functioning was also shown to be favorably influenced by CBT. Three of four studies in the meta-analysis did not support the effectiveness of CBT for hopelessness. Wykes and associates included 30 RCTs (34 studies in total) and used advantageous meta-analytic methods and practices that other studies did not. In the Wykes and colleagues meta-analysis, pharmacological intervention of all participants was required for studies to be included. The aforementioned meta-analyses have included some of the same studies.

CONCLUSIONS

The theoretical, practical, and empirical aspects of CBT for schizophrenia have many applications to social work practice. Given the current effect sizes of the CBT cohorts, there is considerable empirical support for the use of CBT in the treatment of schizophrenia. The meta-analyses are complemented by non-experimental (Bradshaw, 1997) and experimental studies (Bradshaw, 1996, 2003; Bradshaw & Roseborough, 2004), which demonstrates or implies the ability of CBT for schizophrenia to be successfully applied by social workers. Because the empirical evidence supports the effectiveness of CBT for schizophrenia, social workers will be operating within the ethics of the social work profession (see NASW, 1999) and adhering to the mandates of the educational policy and accreditation standards (see CSWE, 2008), when they competently use the treatment to improve the welfare of clients. Yet, social workers should be cognizant of the finding that studies with less scientific rigor are empirically associated with inflated effect sizes (Tarrier & Wykes, 2004).

Therapists must receive training in CBT for schizophrenia (Bradshaw & Roseboroughs, 2004). Such training is essential to social workers who wish to effectively integrate the intervention into social work practice. Clinicians must have a fundamental comprehension of CBT and experience with CBT in clients without psychosis (Turkington, Dudley, Warman, & Beck, 2004). In Bradshaw and Roseboroughs' study (2004), the therapists were licensed clinical social workers (LCSWs) who had master's degrees in social work (MSWs) and an average of five years in mental health experience. Over a duration of six months, the LCSWs were given 48 hours of training in CBT. Training also involved each of the social workers using CBT with three clients and clinical supervision. To effectively provide CBT for schizophrenia, psychiatric nurses were trained for ten days and participated in supervision that occurred on a weekly basis (Turkington, Kingdon, & Turner, 2002). Based on the study conducted by Turkington and others (2002), for counselors (who have previously treated schizophrenia) to learn the fundamentals of CBT for schizophrenia, it takes at least two weeks of rigorous training and continual supervision by a CBT expert (Turkington et al., 2006). For social workers who meet the prerequisite criteria, workshops are available (Turkington et al., 2004). Treatment guidelines (American Psychiatric Association, 2004; National Institute for Clinical Excellence, 2003) provide practical insight into the application and status of CBT for schizophrenia. The availability of CBT for clients with schizophrenia is dependent on the accessibility of supervision and administrative support (Turkington et al., 2004). All factors considered, given the empirical support of CBT for schizophrenia, social work administrators should seriously consider the practical and fiscal viability of integrating CBT into their agency's treatment protocol.

There are several aspects of CBT for schizophrenia that make its application among social work practitioners less complicated. To reiterate, many of the techniques described in the aforementioned treatment manuals are used to treat other axis I disorders in the Diagnostic and Statistical Manual–IV (DSM-IV). For example, cognitive-behavioral techniques used to treat schizophrenia are also used to treat other SMIs such as bipolar disorder and endogenous (biologically based) depression. Additionally, CBT is compatible with other treatments for schizophrenia, such as assertive community treatment (ACT), cognitive remediation, and family intervention (Turkington et al., 2004).

For social work practitioners who use CBT to treat persons with schizophrenia, perhaps the biggest clinical advantage to clients is the social worker's ability to substantively influence both intrapersonal and environmental factors. Social work is characterized by its emphasis on individual functioning in the social environment (NASW, 1999). It is this factor that sets social work apart from other helping professions (Sheafor, Horejsi, & Horejsi, 2000). The accommodation and assimilation processes that are articulated in the vulnerability model (Zubin & Spring, 1977) could be facilitated by micro and macro social work interventions, respectively. Adjunctive social work interventions to CBT (such as case management, class and case advocacy, policy formulation, etc.) could positively facilitate the assimilation and accommodation processes which can contribute to the restoration of adaption.

Clients with schizophrenia may benefit from social workers who simultaneously provide CBT and case management (see Kingdon & Turkington, 2005). Social workers have knowledge regarding ways in which the social environment contributes to the growth or stasis of client development (CSWE, 2008). Problems stemming from the social environment can potentially be mitigated via case-management services. Case management involves aiding the client with regard to the applicability, accessibility, advantage, and request of resources (Miley, O'Melia, & DuBois, 1998). Social work encompasses a number of professional roles, including clinician and case manager (Sheafor et al., 2000). Social workers have claimed to be the most suitable health profession to perform case-management tasks (Berger, 2002). When therapy and case-management services are provided by the same person the advantages include the possibility that: case-management services may aid the therapist in establishing a therapeutic relationship with the client, the client will relate better to one person instead of several, or after therapeutic services have ceased, case management may still be able to continue (Kingdon & Turkington, 2005).

Social workers can influence the availability of CBT for schizophrenia via advocating for national and organizational policies that increase the funding, dissemination, adoption, and evaluation of the treatment. The lack of attention for CBT of schizophrenia in the United States is partially attributable to “… the difference in health care research and delivery …” between the United States and United Kingdom (Turkington et al., 2006, p. 371). Social workers are assumed to have been trained to intervene at micro and macro levels (Haynes & Mickelson, 2000). Indeed, bachelors of social work (BSW) and MSW programs that are accredited by the CSWE (2008) are required to teach social work students to utilize research to enhance policy and the delivery of social services. Consequently, it is certainly within the professional scope of social workers to improve the availability of CBT for schizophrenia via advocating for policy change that is connected with the administration of social services. Such advocacy could ultimately improve the treatment options available to and well-being of clients.

Social workers should be cognizant of limitations that are associated with CBT for schizophrenia. Little is known about the effects of CBT for clients with schizophrenia from diverse cultures (Tarrier, 2005). In one study (Rathod, Kingdon, Smith, & Turkington, 2005), relative to Caucasians, persons who were of African descent had a significantly greater attrition rate at post-treatment and at the one-year follow-up. The authors stated that these results were similar to prior findings by other studies. The evident lack of research and potential lack of external validity that CBT could have for persons with schizophrenia who are not Caucasian, is noteworthy because social work places specific emphasis on vulnerable and marginalized groups (NASW, 1999) such as racial minorities (CSWE, 2008). Factors that may preclude or diminish the effectiveness of CBT for schizophrenia include cultural incongruence between the practitioner and client, extreme paranoia, absence of pharmacological intervention, and severe symptoms (Turkington et al., 2006). Currently it is premature to conclude that CBT lacks external validity for racial minorities.

The meta-analyses included in this review make a substantial contribution to the empirical status of CBT for schizophrenia; however there are still limitations and issues that require further investigation. Meta-analyses are needed which directly address the possible effects of moderating and mediating variables. Such variables are important because they could provide further explanation regarding the relationship between CBT and relevant outcomes of schizophrenia (see Frazier, Tix, & Barron, 2004 for discussion regarding moderator and mediator variables). Tests of heterogeneity allow the meta-analyst to determine if there is significant variability in the studies; if there is significant variability moderator variables can be evaluated to account for the variance (Huedo-Medina, Sáchez-Meca, Marín-Martínez, & Botella, 2006). Heterogeneity was examined in all of the included meta-analyses. Tests of heterogeneity were not significant for three of the included meta-analyses (Gould et al., 2001; Rector & Beck, 2001; Zimmermann et al., 2005). Even when heterogeneity tests are not significant, meta-regression is a viable option to examine variables that may possibly contribute to a lack of homogeneity (Thompson & Higgins, 2002). Gould and associates examined the role of gender on effect sizes. The effects of control group types and patients' status were explored by Zimmermann and colleagues. It would be advantageous if meta-regression was used to determine the potential moderating impact of the patients' age, marital status, employment status, or race. The meta-analyses provide scant attention to cultural variables that, at least theoretically, may influence treatment. The lack of attention to culture in the meta-analyses could be attributable to cultural variables not being reported in the studies included in the meta-analyses. The potential moderating effects of race still require further examination.

Meta-regression or other statistical analyses should be conducted that evaluate the possible influence of the clinicians' years of education, professional affiliation, and years of experience. These variables could substantially impact effect sizes. Current research provides little insight regarding the effectiveness of CBT administered by clinicians with less experience (Jones, Cormac, Silveira da Mota Neto, & Campbell, 2004). Meta-analyses included in this review do not substantively address this issue. Meta-analytic exploration of the aforementioned moderating and mediating variables may ultimately clarify and improve the delivery of CBT. It is worth reiterating that to a great extent, meta-analyses are restricted to the data that is reported in included studies. Before quantitative reviews can provide the most effective meta-analytic procedures, the relevant variables must first be collected and articulated.

This review has detailed the theoretical underpinnings, techniques, and empirical support for CBT of schizophrenia. The compatibility between CBT for schizophrenia and social work practice has been explicated. In summary, there are a number of key points for social work practitioners to consider regarding the application of CBT for schizophrenia: (1) Via meta-analyses and RCTs, which they consist of, there is empirical justification for using CBT in the treatment of schizophrenia; (2) While issues pertaining to external validity require further investigation, CBT for schizophrenia is consistent with the social work profession's mission and standards; (3) Cognitive-behavioral therapy for schizophrenia could be complemented by social work's person-in-environment perspective; (4) Given the aforementioned published outcomes, treatment manuals, and treatment guidelines regarding CBT for schizophrenia, social work administrators need to thoroughly consider the actual and potential implications of providing the treatment to clients served by their agency (see Bradshaw and Roseboroughs, 2004 for similar discussion); (5) Social workers have the prerequisite skills to use social policy to advance the availability of cognitive-behavioral interventions for clients with schizophrenia; (6) The current status of CBT for schizophrenia can be improved by the production and consumption of research that examines factors that are germane to social work's mission and practitioners. The social work profession has used treatments that have been shown to not help clients or has failed to use treatments that demonstrate an ability to help clients (Soydan, 2008). As it pertains to social work practice, cognitive-behavioral theory, interventions, and experimental research have been described and synthesized here to change the latter.

APPENDIX: EXCLUDED STUDIES

Jones, C., Cormac, I., Silveira da Monta Neto, J. K., & Campbell, C. (2004). Cognitive behaviour therapy for schizophrenia. Cochrane Database of Systemic Reviews 2004, Issue 4. Art. No.: CD000524. doi:10.1002/14651858.CD000524.pub2 (Study lacks an average standardized mean difference effect size for positive and/or negative symptoms)

Pfammatter, M., Junghan, U. M., & Brenner, H. D. (2006). Efficacy of psychological therapy in schizophrenia: Conclusions from meta-analyses. Schizophrenia Bulletin, 32(S1), S64–S80. doi:10.1093/schbul/sbl030 (Study provides a review of meta-analyses)

Pilling, S., Bebbington, P., Kuipers, E., Garety, P., Geddes, J., Orbach, G., & Morgan, C. (2002). Psychological treatments in schizophrenia: I. Meta‐analysis of family intervention and cognitive behaviour therapy [Electronic Version]. Psychological Medicine, 32(5), 763–782. doi:10.1017/S0033291702005895 (Study lacks an average standardized mean difference effect size for positive and/or negative symptoms)

Sensky, T. (2005). The effectiveness of cognitive therapy for schizophrenia: What can we learn from the meta-analyses? Psychotherapy and Psychosomatics, 74, 131–135. (Study is not a meta-analysis)

Tarrier, N. (2005). Cognitive-behaviour therapy for schizophrenia—A review of development, evidence, and implementation. Psychotherapy and Psychosomatics, 74, 136–144. (A systematic literature search strategy is not conducted or described)

Tarrier, N., & Wykes, T. (2004). Is there evidence that cognitive behaviour therapy is an effective treatment? A cautious or cautionary tale? Behaviour Research and Therapy, 42, 1377–1401. doi:10.1016/j.brat.2004.06.020 (A systematic literature search strategy is not conducted or described)