The authors are affiliated with the School of Psychology, University of Wollongong, Australia. Send correspondence to Dr. Crowe at Illawarra Institute for Mental Health, Building 22, University of Wollongong, New South Wales, 2522, Australia (e-mail: email@example.com).
Anthony (1) defined recovery as "a deeply personal, unique process of changing one's attitudes, values, feelings, goals, skills, and/or roles. It is a way of living a satisfying, hopeful, and contributing life even with limitations caused by illness. Recovery involves the development of new meaning and purpose in one's life as one grows beyond the catastrophic effects of mental illness." Allott and colleagues (2) suggested that individuals should be supported in their own personal development by placing the "emphasis on building self-esteem, discerning identity, and finding a meaningful role in society. In this view, recovery does not necessarily mean restoration of full functioning without supports (including medication); it does mean building on personal strengths and resources to develop supports and coping mechanisms which enable individuals to be active participants in—as opposed to passive recipients of—their mental health care."
These perspectives imply that the concept of recovery should no longer be restricted to medical model definitions (symptom management or amelioration) or rehabilitation model definitions (improved functioning) but should expand to emphasize psychological recovery processes (3).
Providers of mental health services represent a very important environmental factor that can either help or hinder recovery (4). Tarrier and Barrowclough (5) demonstrated that people with psychiatric and psychological disorders are significantly affected by interpersonal interactions, including those with mental health professionals.
The degree of adoption of recovery-oriented principles and practices by mental health professionals may be influenced by their attitudes and hopefulness regarding the possibility of recovery. Hugo (6) found that mental health professionals were less optimistic than the general public about prognosis and longer-term outcomes for people with schizophrenia or depression. Others have suggested that the more negative attitudes of professionals may be more realistic and in line with greater knowledge of mental disorders, but they could also be biased as a result of the proportion of contacts they have with people with chronic and recurring disorders at times when significant interventions are required (7).
Rickwood (8) stated, "implementing a recovery orientation requires an attitude shift for many service providers in order to support consumer rights and provide the types of services that maximize well-being for people with mental illness." She also suggested that an understanding of the factors that affect recovery, rehabilitation, and relapse is essential. Attitudes are thought to reflect the "mental readiness" or learned "disposition" that influence actions and reactions (9).
Modification of attitudes of mental health professionals and individuals with mental illness is a key component of most general mental health competencies as well as specific "recovery-oriented" competencies (10). However, it is unclear whether a recovery-based staff training program can enhance staff attitudes and level of hopefulness regarding recovery of the consumers with whom they work.
The Collaborative Recovery Training Program (11) aims to train mental health professionals in the collaborative recovery model (12). The model integrates principles and skills—for example, motivational enhancement and collaborative goal setting—supporting the unique recovery processes of individuals with chronic and recurring mental illness and collaboration between mental health workers and these individuals. The program emphasizes recovery as a process that belongs to the individual with mental illness, a process that is more comprehensively defined by subjective experience than the presence of illness symptoms. The program also emphasizes issues of autonomy, hope, and individual experience that are central to the recovery movement of mental health consumers. The rationale for the program and a more detailed description has been published elsewhere (12). An appropriate recovery-oriented "attitude" from the perspective of the model involves being hopeful regarding the consumer's ability to set, pursue, and attain personal goals that facilitate recovery.
It was hypothesized that after completing the program, mental health workers would show significant improvements in knowledge of recovery processes, attitudes supporting recovery, and hopefulness related to the recovery process. This hypothesis is based on the assumption that exploring recovery concepts and engaging in specific skills training might raise trainees' awareness of potential attitudinal barriers and also empower them by providing specific skills to support recovery processes of people who have mental illness.
Pre- and posttraining measures were collected from 248 mental health workers from nongovernment organizations (101 workers, or 41 percent) and the government health sector (147 workers, or 59 percent) who were participating in the two-day Collaborative Recovery Training Program. Trainees were employed at organizations in eastern Australia. Trainees had a mean±SD age of 40.91±9.92 years, ranging from 22 to 60 years. The mean number of years in the mental health profession was 12.01±9.87 years, ranging from one month to 38 years. A majority of the sample was female (174 persons, or 70 percent).
Staff members were given the following measures immediately before and after the session. Measures were completed in ten to 15 minutes.
The Recovery Attitudes Questionnaire (RAQ-7) was designed to assess attitudes reflecting beliefs that people can recover from mental illness (13). Validation of RAQ-7 has been conducted with 249 consumers, 125 mental health professionals, 58 family members, and 159 students (13).
Factor analysis identified two factors comprising seven items on RAQ-7. The first factor consists of four items suggesting that "recovery is possible and needs faith" (for example, "Recovery can occur even if symptoms of mental illness are present"). The second factor consists of three items suggesting that "recovery is difficult and differs among people" (for example, "People in recovery sometimes have setbacks"). Each item is measured on a 5-point scale ranging from 1, strongly disagree, to 5, strongly agree. The test-retest reliability coefficient over a 19-day period was .67 for the full scale, .61 for factor 1, and .62 for factor 2. Cronbach's alpha coefficients for factors 1 and 2 and for the full scale were .66, .64, and .70, respectively (13). Concurrent validity was also found in that consumers who identified themselves as being in recovery, and who were in recovery for longer periods, had the most favorable attitudes (13).
The collaborative recovery knowledge scale was developed for this study and comprises 13 multiple-choice items related to knowledge of the key principles and intervention characteristics that represent components of the collaborative recovery model. Sample questions follow: "Research evidence demonstrates that well-being is related to: a) achieving as many goals as possible, b) achieving autonomous goals, c) not having goals, or d) having only one goal" and "Resistance is: a) a treatment opportunity, b) always an obstacle, c) the client's fault, d) proof the client is not motivated, or e) evidence that treatment is failing." Each item that was answered correctly was scored as 1, incorrect items were scored as 0, and a sum of correct responses was used in analyses. (The correct answer for the first sample question is b, and the correct answer for the second sample question is a.) Possible scores range from 0 to 13, with higher scores indicating better knowledge.
The staff attitudes to recovery scale (STARS) was also developed for this study. The scale assesses attitudes and hopefulness related to the goal striving and recovery possibilities for the consumers with whom they work. Principles and constructs that influenced item construction included the interrelatedness of hope, goal setting, and recovery. Items reflect Snyder and colleagues' (14) views of hope: "an enduring disposition that is subjectively defined as people assess their determination and strategies in relation to goals."
Three items on the STARS address general hopefulness (for example, "All of these clients are capable of positive change"). Eight items were adapted from the Adult Dispositional Hope Scale (14) (for example, "There are lots of ways around any problem" became "There are lots of ways to deal with any problems that these clients have"). Each item was rated on a 5-point scale ranging from 1, strongly disagree, to 5, strongly agree. The mean of all 19 items was used in analyses. Possible scores range from 19 to 95, with higher scores reflecting more positive and hopeful attitudes. Internal consistency of the scale in the study presented here was moderate to high (Cronbach's α=.81).
Independent-samples t tests used the pretraining measures to determine whether there were any significant differences in scores between government and nongovernment employees. There were no significant differences in pretraining scores between the two groups on the knowledge scale, STARS, and the first factor of RAQ-7. However, on the second factor of RAQ-7, ratings were significantly higher for nongovernment employees than for government employees (t=-3.05, df=246, p<.01).
Consequently, a 2×2 (sector, nongovernment versus government, by time, pre- and posttraining) mixed-design multivariate analysis of variance was conducted to determine whether improvements in knowledge (as measured by the collaborative recovery knowledge scale) and attitudes toward recovery (as measured by RAQ-7 and STARS) occurred over training and whether these improvements were different for the two employment groups. A significant multivariate time-by-sector interaction was found (Wilks' λ=.953, F=3.00, df=4 and 243, p<.05). Univariate tests revealed significant interactions for the attitudes as measured by STARS (F=4.85, df=1 and 246, p<.05) and by the second factor of RAQ-7 (F=5.79, df=1 and 246, p<.05). There were no significant interactions for knowledge or the first factor of RAQ-7.
Table 1 provides the means and standard deviations by time and sector. The interaction on STARS was such that both government and nongovernment groups had more positive attitudes over time, but the government group showed greater improvements. The interaction for the second factor on RAQ-7 suggested that the government group tended to have slight improvements in attitudes, whereas the nongovernment group had slight reductions. Univariate tests showed that the government group showed statistically significant improvements in attitudes on the second factor of RAQ-7 over time (F=4.48, df=1 and 146, p<.05). However, there was no significant change on this measure for the nongovernment group. For the knowledge scale and the first factor of RAQ-7, there was a significant main effect for time, such that both knowledge and attitudes improved for both employment groups (F=147.37, df=1 and 246, p<.01 for the knowledge scale; F=18.91, df=1 and 246, p<.01 for the first factor of RAQ-7). Table 1 also displays the effect sizes for each group by using partial eta squared values. By using Cohen's convention (15) for specifying effect size values as small (.2), medium (.5), or large (.8), it can be seen that effects on knowledge and attitudes for those working in government organizations were closer to medium, whereas for nongovernment organizations effect sizes were small.
Correlational analysis of posttraining ratings of the collaborative recovery knowledge scale with attitudes as reflected by STARS and the two factors of RAQ-7 indicated that the collaborative recovery knowledge scale is positively associated with recovery attitudes (first factor of RAQ-7, r=.41; second factor of RAQ-7, r=.39; and STARS, r=.13; p<.05 for all).
The two-day Collaborative Recovery Training Program (11,12) provides information about recovery principles and processes as well as specific skills related to motivational enhancement, identification of consumer needs, negotiation of goals, systematic specification of tasks to achieve the goals, and monitoring of progress over time. The study presented here found that over the course of the training session, desired changes were found in trainee knowledge, beliefs, and attitudes related to recovery. Specific attitude domains included increased beliefs that recovery is possible for people with serious mental illness and can occur even if symptoms are present. Overall, our study found that after completing the training, persons demonstrated higher levels of hopefulness and belief regarding the ability of individuals with serious mental illness to set and achieve their goals. A limitation of the study was the lack of a control group to be able to rule out other potential confounds, such as social desirability factors. However, the knowledge measure was significantly related to all recovery attitudes measures, which might suggest that increased knowledge regarding recovery principles and skills is associated with an increase in recovery orientation of mental health workers.
There was an interaction between improvements in knowledge and attitudes over the course of the training program and the employment sector of participants. This was a weak interaction, but it suggested that mental health workers from the government sector had more of a change in attitude toward recovery than those from the nongovernment sector. This finding might be explained in terms of different role functions and professional identities in the two employment groups. Although some participants from the nongovernment sector provided rehabilitation services, most of them were involved with providing supported housing and associated activities, for example, shopping, cooking, and recreation. Consequently, they may not have viewed themselves as taking as active role in facilitating consumers' recovery processes, thus tempering the strength of improvement in recovery attitudes. Some nongovernment workers may have felt challenged by the training to redefine their traditional role identities at work to include recognition of the potential impact that they could have on the recovery processes of those they work with.
However, despite these possibilities, it is worth noting that mental health workers from both sectors had relatively positive attitudes toward recovery. For example, in this study the mean rating of the scores on the two factors of RAQ-7 were higher than mean ratings by professionals (mental health workers) reported in Borkin and colleagues' study (13). Specifically, the pretraining mean score for the first factor for the entire sample in the study presented here was 16.19±2.22 compared with a mean score of 12.60±2.55 in the study by Borkin and colleagues. Similarly, for the second factor, this study found a pretraining score of 13.25±1.39 compared with a score of 10.44±1.44 reported in the study by Borkin and colleagues. It is possible that recovery attitudes have changed since Borkin and colleagues' study was conducted in 2000 because there is currently greater awareness and acceptance of the concept of recovery. However, it may also be that RAQ-7 is limited as a pre-post measure of attitudinal change. Furthermore, because the maximum possible rating is 20 for the first factor and 15 for the second factor, the ratings in this study may have been subject to a ceiling effect.
This study found preliminary evidence that staff attitudes and hopefulness regarding assisting consumers with their individual recovery processes can improve with training. Along with improved recovery attitudes, mental health workers who completed the program significantly increased their knowledge regarding principles of recovery and collaboration and consumer autonomy support, motivation enhancement, needs assessment, goal striving, and homework use. For people with enduring mental health disorders to become more hopeful about their own recovery prospects, they need to have hopeful people around to encourage and support the recovery process. Sustaining positive recovery attitudes in clinical practice may be influenced by the level of recovery orientation of the service setting. Future studies could examine the extent to which positive recovery attitudes are supported at an organizational level and whether they result in changes in clinical practice and consumer outcomes.
This study was supported by the National Mental Health and Medical Research Council Strategic Partnership Grant in Mental Health (219327). Contributing organizations to the Australian Integrated Mental Health Initiative High Support Stream of this project, in alphabetical order, include Aftercare, Illawarra Health Mental Health Service, Latrobe Regional Hospital Mental Health Service, Prince Charles Hospital Health District Mental Health Service, Neami, Psychiatric Rehabilitation Association, Richmond Fellowship Queensland and New South Wales, University of Wollongong, and Wentworth Area Health Service.