Introduction

Changes during the past twenty years in the mental health service environment––especially the rise of the managed care industry have radically altered the provision of mental health services in the United States (Cohen 2003). Managed care was a response to an institutional demand to reduce health care costs and reorganize health care and mental health care systems with the rationale of efficiency (Scheid 2003). The overarching goal of managed care organizations is to provide efficient quality care at a lower cost than that offered in the fee-for-service professional community. Social workers as well as other mental health care providers have been required to learn new strategies and skills in order to cut costs of services by limiting access to services and limiting the utilization of more costly services (Cohen 2003; Scheid 2000). New skills involve computers and technology, documentation and paperwork, empirical validation of treatment methods, knowledge of brief treatment modalities, and a business orientation in managing services in a profitable way (Bolen and Hall 2007; Feldman 2001; Lu et al. 2002).

With the restructuring of mental health services to managed care practices, workers have been dealing with strict practice guidelines, increased accountability, reduced autonomy, and a requirement to become competent with new management skills (Cohen 2003; Feldman 1997; Hall and Keefe 2000; Koeske and Koeske 1993; Lu et al. 2002; Shera 1996). Participating in managed care is complicated, and those that are not apprised about the managed care world are likely to suffer of stress and anxiety concerning their ability to perform well and continue providing effective services to their clients (Hall and Keefe 2000; Spevack 2009). Perceptions that job activities are inappropriate and incongruent with workers’ training and expertise can result in feelings such as disliking the job, burnout, and turnover (Acker 1999; Arches 1997; Lu et al. 2002; Mechanic 2007; Feldman 2001; Shera 1996; Stone 1995; Tyler and Cushway 1998).

The purpose of the study reported in this article was to understand social workers’ perceived responses to professional activities and interactions when dealing with managed care organizations, and to examine the relationships between workers’ experiences when interfacing with managed care organizations and the psychological and somatic symptoms associated with burnout.

Managed Mental Health Care

Managed care has affected both the private and the public sectors of mental health care. Workers in the public sector have especially been subjected to increased administrative control, limited resources, and a new focus on short––term treatment modalities, and measurable outcomes. Those working with clients with severe and persistent mental illness (SPMI) face additional challenges. Managed care decisions in reimbursing mental health services are based on diagnostic criteria and treatments that aim at reducing symptoms associated with the client’s diagnosis. Treatments are expected to be efficient and short-term, with sufficient documentation of desirable outcomes. Those working with clients with SPMI know that these clients need an ongoing and long term support as well as their progress is slow and signs of improvement may not be noticeable (Acker 1999; Maslach 1978; Raquepaw and Miller 1986). The medical model adopted by managed care for treating clients with SPMI is perceived by providers, as incompatible with the social work approach which focuses on helping these clients in living successfully in their communities, by improving their social skills, and providing them with the full continuum of care including housing, vocational training, and ongoing social support (Scheid 2003). Social workers have been required to alter their role from serving as clients’ advocates to balancing clients’ needs against the need for cost control. These conflicted ideologies are a source of frustration for social workers who perform the largest portion of mental health work in the USA (Cohen 2003; Daniels 2001; Egan and Kadushin 2007; Scheid 2003).

The cost-effective approach of managed care organizations requires that providers become competent in new skills that enable them to work effectively in a managed care environment. Among these skills are problem-oriented, goal-focused, and short-term treatment modalities. Additionally, providers need to develop the appropriate capacities to meet the accountability imperatives associated with managed care. New administrative skills such as documentation and paperwork, computers and technology, empirical validation of treatment methods, and a business orientation in managing services in a profitable way are necessary to function effectively within the managed care environment (Cohen 2003; Hall and Keefe 2000; Keefe and Hall 1998). In their new roles as gatekeepers and treatment providers social workers must learn these new strategies and skills in order to reduce considerable cost savings expenditures when providing services to clients (Cohen 2003).

Competence in the Managed Care Context

Building on the concept of competence which describes feelings of confidence about one’s abilities in mastering organizational and work demands (Hall and Keefe 2000; Wagner and Morse 1975; White 1967), several theorists argue that persons’ belief that they are not able to professionally perform well increases their risk of becoming burned out (Bandura 1989; Cherniss 1993; Harrison 1980; Keefe and Hall 1998). The management literature claims that people have a drive to influence and master their work environment. The concept of “self-perceived competence” is a subjective evaluation of the person’s skills and abilities to perform well. When workers feel that they are competent about their skills and that they can perform well in spite of the interference of management, they are most likely to also manage their work tasks effectively. Management theorists argue that no matter what the problems and obstacles are, when workers feel that they are competent to solve challenges and problems at their workplace, they are also capable of solving them (Hall and Keefe 2000; Keefe and Hall 1998; Wagner and Morse 1975; White 1967).

Based on the management competence approach it is inevitable that social workers who feel that they cannot manage competently their practices under managed care guidelines, and that their professional roles are incongruent with the new requirements and regulations are more likely to get frustrated (Keefe and Hall 1998); thus at risk to suffer the negative consequences of burnout.

Social Support

Social support at the workplace, which includes supervision and other supportive mechanisms directed toward workers, has been diminishing with the dwindling resources and funding associated with social services (Acker 2003, 2004; Adams 2001; Pumariega et al. 2003). Supervision and other opportunities for professional development such as workshops, continuing education, and stress management programs cost too much; resulting in limited opportunities for workers to discuss job related problems and advance professionally (Acker 2003, 2004). It is not unlikely that workers who feel professionally unsupported are more likely to have negative attitudes toward their job (Acker 1999, 2003; Pines 1983; Um and Harrison 1998; Winnibust 1993).

Burnout

Burnout is defined as a negative psychological experience that is a reaction of workers to job-related stress. Burnout refers to a cluster of physical, emotional, and interactional symptoms, including emotional exhaustion, a sense of lacking personal accomplishment, and depersonalization of clients. Burnout symptoms can also include common colds, flu-like symptoms, gastroenteritis, headaches, and fatigue (Maslach 1982; Maslach et al. 1996; Mohren et al. 2005). It is critical to recognize that social workers who are confronted by the complex needs of clients with mental health problems, and organizational demands of cost containment, which often are incompatible with professional values and expectations are at risk to experience the negative symptoms associated with burnout.

Several hypotheses are suggested in order to better understand the relationships between the external and internal pressures associated with managed care requirements for efficiency and cost containment, and social workers’ psychological and somatic experiences as described by the concept of burnout.

  1. 1.

    Self-perceived competence in the context of managed care is negatively associated with the three burnout dimensions (e.g. emotional exhaustion, depersonalization, and a reduced feeling of personal accomplishment) and with somatic symptoms.

  2. 2.

    Professional involvement with clients with severe and persistent mental illness is negatively associated with self-perceived competence in the context of managed care; and positively associated with the burnout dimensions and with somatic symptoms.

  3. 3.

    Social support is positively associated with self-perceived perceived competence in the context of managed care; and negatively associated with the burnout dimensions, including somatic symptoms.

Procedure

The sample of this study, which was obtained from a professional list of social workers practicing in New York State, consisted of 591 social workers. Self-administered and anonymous questionnaire packets were mailed to 1,000 randomly selected individuals from this list. The overall response rate was 58%. This study was approved by the institutional review board of the university, where the author is employed.

Sample

Educational levels of respondents included 89% with master degrees in social work, and 5% with doctoral degrees. The respondents were primarily females (80%). The mean age was 51 ranging from 21 to 80. Seventy-one percent were married or involved in long term relationship with a partner, 28% were not married. The respondents were predominantly White (86%), African American or Black (5%), 7% were Latino, and 1% were Asian. The mean for years of experience in social work was 22 years, ranging from 2 to 50 years of employment. The median for client contact hr per week was 25. Forty-three percent were employed in outpatient mental health settings, 13% in community support systems, 23% in private practice, 7% in substance abuse rehabilitation settings, 8% in inpatient psychiatric settings, and 6% in schools.

Measures

Self-perceivedperceived competence in the context of managed care (CMC). These 16-item scale (Hall and Keefe 2000) measures workers’ self-perceived competence in meeting the demands of managed care (Table. 1). Example of items include: “Managed care allows me enough freedom to be effective in treating clients”, and “I feel like I am getting nothing done due to managed care requirements”. Respondents are asked to rate each item on a 4-point likert-type scale for their agreement with each statement (1 = strongly disagree; 2 = neither agree nor disagree; 3 = agree; 4 = strongly agree). The scale has both good reliability and validity (Hall and Keefe 2000). The Cronbach alpha coefficient for a sample of practitioners (social workers, psychologists and psychiatrists) in private practice was reported to be .90 (Hall and Keefe 2000). This measure was also reported to have adequate content validity based on factor interpretation; the predictive validity of the instrument was tested in terms of its hypothesized relationship with organizational task and performance and yielded verified and significant relationship (P < .05) (Hall and Keefe 2000). Cronbach’s alpha coefficient for this study sample was .86.

Table 1 Items from the self-perceived competence in the context of managed care instrument (Hall and Keefe 2000; Keefe and Hall 1998)
Full size table

Burnout was measured by using a slightly modified version of the Maslach Burnout Inventory (Maslach et al. 1996). It includes three subscales: Emotional exhaustion (EE), depersonalization (DP), and personal accomplishment (PA), all known to have good reliability and validity (Maslach et al. 1996). The EE subscale comprises nine items reflecting feeling of being emotionally overextended, exhausted, physical exhaustion, and emptiness. Examples of items include: “Working with people all day is really a strain for me”; and” I feel frustrated by my job”. The four items of the DP subscale describe an unfeeling and impersonal response towards clients of one’s service. It includes such items as “I feel I treat some as if they were impersonal objects” and “I don’t really care what happens to some clients”. The PA subscale consists of six items that describe feelings of competence and successful achievement in one’s work. Examples of items include: “Accomplishing worthwhile things at work”; and” Positively influencing my clients’ lives through work”. Respondents were asked to rate each statement on a 7-point likert-type scale for frequency of agreement (0 = never, 1 = a few times a year or less; 2 = once a month or less; 3 = a few times a month; 4 = once a week; 5 = a few times a week; 6 = every day). Cronbach’s alpha coefficient for this study’s sample included .92 for EE, .78 for DP, and .77 for PA.

Somatic symptoms which in this study are additional dimensions of the burnout syndrome were measured by two scales. One scale measured common colds and flu-like symptoms. It comprises 12 items describing symptoms such as colds, sore throat, cough, dizziness, and fever. The second scale measured symptoms of gastroenteritis (GA) and included three items. These scales are based on previous research done by Mohren et al. (2005) and Nakao et al. (2005). In this study respondents were asked to rate each item on a 7-point likert-type scale in terms of how often they have been experiencing each of those symptoms for the past six months (0 = never, 1 = rarely; 2 = sometimes; 3 = fairly often; 4 = often; 5 = very often; 6 = all or most of the time). Cronbach’s alpha coefficient for this study’s sample included .85 for common colds and flu-like symptoms, and .75 for symptoms of gastroenteritis.

The measure involvement with clients with severe and persistent mental illness (SPMI) is based on Hagen and Hutchison’s (1988) description of symptoms and behaviors related to having severe mental illness. This scale comprises eight items that reflect the workers’ perception about the extent to which they work with clients who suffer from symptoms related to severe mental illness, and engage in socially unaccepted behaviors which are also related to mental illness. Examples of items are”: Inappropriately angry or violent”, “Confused, disoriented or otherwise out of touch with reality”, “Having paranoid behavior”, and “severely anxious”. Each item is rated on a five point scale: 1 = not at all, 2 = sometimes, 3 = often, 4 = very often, 5 = most or all the time. The weighted scores for all the eight items are added together to measure this variable. The scale has both good reliability and face content validity. Cronbach’s alpha coefficient for this study sample was .88.

Social support at the workplace which reflects workers’ perceptions of how much social support and opportunities for professional development they receive from their workplace is measured by two scales. The social support scale is adopted from the social support from supervisor and social support from co-workers scales developed by Caplan et al. (1980). This scale comprises eight questions about the extent to which people around the worker (the worker’s supervisor and co-workers) provide support by listening and by being persons that the worker can rely on for help. Examples of questions are: “How much does your supervisor go out of the way to do things to make your life easier?” and “How much are other people at work willing to listen to your personal problems?” The scale has both good reliability and validity (Caplan et al. 1980). Cronbach’s alpha coefficient for this study sample was .80.

The measure of opportunities for professional development (OPD) comprises 7 items that reflect the extent to which the agency provides workers with opportunities for training and advancement including: Staff meetings, continuing education courses, in-service training, reimbursement of school tuition and others. Each item is rated on a seven point scale from 0 = not at all to 6 = most or all the time, for the extent to which each participant is provided with those opportunities. Face content validity was established by having experienced workers and social work faculty members from the researcher’s organization review the scale items in terms of the degree to which they reflect opportunities that social service workers receive at their jobs. Cronbach’s alpha coefficient for this study sample was .82.

Data Analysis

Pearson product- moment correlation coefficient was used to investigate the research hypotheses and other relationships among the study’s variables. A correlation matrix was computed for all the study’s variables. To further explore how self-perceived competence in the context of managed care (CMC) contributed to the burnout dimensions, and the somatic symptoms, above and beyond other work related variables (e.g. involvement with clients with SPMI, social support, opportunities for professional development, caseload size, satisfaction with salary, and years of social work experience), hierarchical regression analyses were used. Additional control variables including race, age, and gender were also included in the statistical analysis.

Nominal variables including gender, race (people of color or White), and type of setting in terms of private or public sectors were included in the statistical analysis by coding them as dummy variables. Data analyses utilized SPSS computer software.

Results

The analysis began with the investigation of the relationships among the study’s variables including the primary relationships as indicated in the study’s hypotheses. As shown in Table 2 workers’ self-perceived competence in the context of managed care (CMC) was found to have statistically significant low negative correlations with emotional exhaustion (EE) and with flu-like symptoms; and low positive correlation with opportunities for professional development (OPD). Positive correlation means that the two variables move in the same direction, and negative correlation means that the two variables move in opposite directions-that is, as one increases, the other one decreases. EE had medium positive correlations with depersonalization (DP), flu-like and gastroenteritis symptoms, involvement with SPMI; and low negative correlations with feeling of personal accomplishment (PA), and satisfaction with the salary. DP had medium positive correlations with involvement with SPMI, and with flu-like and gastroenteritis symptoms. DP also had low positive correlation with size of caseload, and low negative correlations with work experience and age. PA had low negative correlations with flu-like and gastroenteritis symptoms, and low positive correlations with work experience and age. Involvement with SPMI had medium positive correlation with OPD; low positive correlation with size of caseload; and low negative correlation with age. Social support had medium positive correlation with OPD.

Table 2 Intercorrelation among selected study variables (N = 591)
Full size table

White workers reported lower levels of CMC; and higher levels of DP and gastroenteritis symptoms than Non-Whites. Workers in public agencies reported higher levels of EE, DP, reduced PA, and involvement with SPMI. Workers in public agencies reported higher levels of social support and OPD than those in private practice. In relation to the correlations found among the burnout dimensions, the results indicate that the power of the correlations between the burnout variables including EE, DP, PA, and flu-like and gastroenteritis symptoms are sufficiently independent to be examined separately.

Multiple regression analyses provided information on the influence of self-perceived competence in the context of managed care (CMC) on the burnout symptoms above and beyond other work related variables (i.e., involvement with SPMI, social support, caseload size, work experience, and type of agency), and workers’ socio-demographic variables including age, gender, and race. CMC was entered first into the regression equation. Then, in the second step, the work variables were added to the regression in order to examine whether there was an increase in predictability above and beyond the information provided by the first variable. The third step included workers’ socio-demographic variables in order to examine if those variables also contributed significantly to the outcome variables. To reduce the problem of finding significance that’s actually produced by chance when multiple correlations are made with multiple regression analysis, the Bonferroni procedure calls for the researcher to divide the .05 probability level by the number of statistical tests to be conducted, which in the this case included ten tests. This resulted in a new p value which was = .005 (.05/10), (Montcalm and Royse 2002).

The first analysis for predicting EE which included the variable CMC in the first step resulted in non- significant relationship, R 2 = .00, F (1, 170) = .684, P < .401. Step two which included the work variables (involvement with SPMI, social support, OPD, caseload size, satisfaction with salary, years of SW experience, and type of agency) was significant, R 2 change = .22, F (7,163) = 6.671, P < .001. The third step with the socio-demographic variables (age, gender, and race) was not significant, R 2 change = .24, F (3,160) = .791, P < .501. The first step of the second analysis for predicting DP was not significant, R 2 = .00, F (1, 170) = .064, P < .800. The second equation which included the additional work variables was significant, R 2 change = .14, F (7, 163) = 3.860, P < .001. The third step with the socio-demographic variables was not significant, R 2 change = .02, F (3, 160) = 1.451, P < .230. The first step of the third analysis for predicting PA was not significant, R 2 = .01, F (1, 170) = .774, P < .380. The second equation which included the additional work variables was not significant, R 2 change = .10, F (7, 163) = 2.717, P < .011. The third step which included the socio-demographic variables was also not significant, R 2 change = .02, F (3, 160) = 1.283, P < .282. The analysis for predicting flu-like symptoms resulted in non-significant relationship for the first step, R 2 = .00, F (1, 170) = .056, P < .812). The second equation which included the additional work variables was not significant, R 2 change = .07, F (7, 163) = 1.606, P < .137. The third step for the socio-demographic variables was also not significant, R 2 change = .00, F (3, 160) = .186, P < .906. The analysis for predicting gastroenteritis symptoms resulted in non-significant relationship for the first step, R 2 = .00, F (1, 170) = .092, P < .762). The second equation which included the additional work variables was not significant, R 2 change = .06, F (7, 163) = 1.495, P < .172. The third step for the socio-demographic variables was also not significant, R 2 change = .01, F (3, 160) = .291, P < .832.

Based on the results of the multiple regression analyses, none of the outcome variables were predicted by CMC when controlling for all the other variables of the study. The second set which included the work variables (involvement with SPMI, social support, OPD, caseload size, satisfaction with salary, years of SW experience, and type of agency) had statistically significant influence on two outcome variables including EE, and DP. The set of the socio-demographic variables did not have statistical significant relationship with any of the outcome variables.

Discussion

The present study shows that workers who report higher levels of self-perceived competence in the context of managed care, report lower levels of emotional exhaustion (EE) and somatic––flu-like symptoms. These findings are consistent with previous research on management competence, which suggests that a sense of competence about one’s abilities and skills to function well in their work environment leads also to competent behavior such as the ability to solve and respond successfully to complex and changing tasks (Bandura 1989; Hall and Keefe 2000; Wagner and Morse 1975; White 1967). An important issue in the management literature is the point of view indicating that workers, who feel competent in specified skills areas such as those required under managed care guidelines, are more likely to have better partnership with management (Keefe and Hall 1998). On the other hand, when workers feel that they are not able to professionally perform well, they are more likely to feel stressed and anxious and suffer from the complex symptoms associated with the burnout phenomena (Cherniss 1993; Hall and Keefe 2000; Harrison 1980; Keefe and Hall 1998).

Another aspect of self-perceived competence in the context of managed care has to do with the incongruence of the traditional ideology of the community mental health movement with the managed care emphasis on cost effective treatment and measurable outcomes (Scheid 2003). Scheid (2003) and Minikoff (1994) call upon mental health care providers to try to resolve this incongruence with the assertion that the ideology of managed care is not so different from the original ideology of the community mental health movement. For example principles of community based care include: The need to address the entire community; making services accessible to the entire population, not just to those with SPMI; interventions should enhance clients’ empowerment rather than dependency on providers; and elimination of public versus private services, so private providers can care for public clients, and public providers can develop services that can compete for private dollars. There is no doubt that providers must be involved in the process of making decisions about service delivery and the assessment of treatment outcomes. The focus on performance accountability in terms of saving dollars is problematic unless there is an agreement with providers that clients do not receive inadequate services (Scheid 2003).

Self-perceived competence in the context of managed care is shown to be correlated with opportunities for professional development, which suggests that social workers who receive adequate professional development such as continuing education, and in-service training are better informed about organizational requirements such as those associated with managed care.

Involvement with clients who suffer from severe and persistent mental illness was correlated with higher levels of EE and DP; these results are consistent with the concern that social workers as well as other helping professionals (i.e., psychologists and psychiatrists) often have negative attitudes toward this type of population, because of the ongoing and significant amount of work needed so that these clients succeed to make a very small progress in treatment (Acker 1999; Finch and Krantz 1991). Working with clients with SPMI in a managed care environment contributes to additional challenges. The limited resources and access to more expensive programs for clients with SPMI can be another source of stress for social workers who want to provide these clients with the full continuum of care so they can function successfully in their communities (Scheid 2003).

Younger and less experienced social workers reported higher levels of DP and reduced levels of PA. These findings are consistent with the early burnout literature which claims that older workers with more life and job experience are at a lower risk to become burned out compared to those who are young and have less job experience (Maslach 1982). Although these findings confirm Maslach’s burnout theory (1982), they also pose the concern of how much schools of social work and agencies prepare younger and inexperienced workers to deal effectively with the pressures and requirements associated with the current mental health care environment.

White workers reported lower levels of CMC, and higher levels of EE, DP, and gastroenteritis symptoms than Non-Whites. Maslach (1982) claims that Non-White workers, compared to Whites, do not burn out as much; they have a more realistic perspective about life and an appreciation of their own difficult experiences; and thus are better prepared to cope with stressful job situations.

Social workers employed in public agencies reported higher levels of EE, DP and reduced PA than those in private practice. Although managed care has greatly affected those working in private practices, it appears that the job demands, the complexity of larger organizations, and the heavy caseloads associated with public agencies have resulted in producing higher levels of burnout among social workers employed in public agencies. Social workers in public agencies are also more involved with clients with SPMI than those who work in private practice. As discussed earlier, clients with SPMI have multiple needs, and require a large amount of resources that very often are not met or reimbursed by the organization and/or the insurer. The challenges and the frustration for those working in public agencies eventually result in higher levels of EE, DP and reduced PA, as shown in this study.

The hierarchical regression analyses provided further evidence about the relationships between work related stressors and burnout. The work related group of variables which consisted of SPMI, social support, opportunities for professional development, caseload size, satisfaction with salary, years of SW experience and type of agency had significant relationships with EE and DP, the two major dimensions of the burnout syndrome which are related to stress. CMC did not have statistically significant relationships with any of the burnout dimensions and somatic symptoms when controlling for the other work related variables. Although the work related variables as a set provided the most comprehensive influence on workers’ job attitudes, CMC can be viewed as an additional organizational factor that is going to contribute to the burnout phenomena among those practicing in public mental health care organizations.

Study Limitations

The convenience sample of social workers limits the generalizability of the study’s findings. This sample was obtained from a professional list of social workers which resulted in a relatively homogeneous sample of social workers, especially when it comes to variables such as age, gender, ethnicity, and level of education. In the community mental health settings many of the primary providers dealing with clients with serious mental illness are bachelor’s degree case managers who are young, men, and members of ethnic minority groups. Unfortunately, there are no lists available of this diverse population of case managers employed in mental health agencies. It is not uncommon that case managers who do not have graduate degrees, are not certified by the State, or have membership(s) in professional associations, thus making it very difficult to randomly recruit them from available lists.

The response set of subjects when responding to self-report measures is another limitation, which is inclusive for this type of research design. A similar response set to different scales which measure respondents’ feelings and perceptions about their job could be a result of a temporary mood of respondents at the time of responding to those measures, as well as it can be related to a social desirability factor. The items of the scale “self-perceived competence in the context of managed care” refer to the concept of managed care as a common term, which is used to describe a proxy for a number of specific organizational and management components associated with managed care. However, because this term can mean many things, the researcher in this study cannot be sure that respondents are making an assessment on the same variable, or very different variable. This poses a problem for the validity of this measure.

Although the findings included several statistically significant correlations, most of them would not be significant with a Bonferroni correction. Thus, with this type of research design, a cautious interpretation of the findings is suggested.

Implications for Research, Practice, and Education

Although this research provided some important answers about factors associated with managed care and burnout, it is clear that there are many more questions that future research may explore. Future studies should use more diverse samples with workers who are younger and who represent more diverse ethnic and racial backgrounds. It is also suggested that a new or a revised scale that assesses the perceived responses of workers to potential stressors associated with managed care is developed to increase the validity of the scale used in this study.

It is important that social work schools integrate the findings of this study in their curriculum in order to better equip their new graduates to face the real world working conditions. Social work programs must teach students the appropriate skills needed for working with managed care organizations. Emphasis should be on teaching market-driven therapies such as brief counseling, crisis intervention, and direct services and community interventions (Bolen and Hall 2007). Students must learn the necessary skills and attitudes required in order to balance the financial aspects and clinical aspects of care giving. Field education training experience should involve agencies that are impacted by managed care, and that enable students to develop skills which are relevant to working with managed care organizations (Keefe and Hall 1998). A focus on case management services will increase students’ realistic expectations about mental health work and about the important needs of clients with severe and persistent mental illness. In-service training and workshops are important resources for inexperienced social workers, as they can provide them with the knowledge and skills to interact effectively with managed care; thus increasing social workers’ competence and confidence and minimizing work-related stress.

This study’s findings also call for an organizational assessment and identification of potential strategies and plans by mental health professionals and managed care organizations. Providers should become more involved in professionals decisions about service delivery, at risk contracts, capitated case rates, and so on. It would be wise for mental health officials and policymakers, managed care executives, practitioners and consumers to work together in an effort to increase the effectiveness of services while maintaining professional quality and standards, and realistic mental health costs.