Recovery of People With Psychiatric Disabilities Living in the Communityand Associated Factors

Yen-Ching ChangI-Shou University

Tamar Heller and Susan PickettUniversity of Illinois at Chicago

Ming-De ChenKaohsiung Medical University

Objective: Consumer-oriented recovery has been discussed for more than two decades in the mentalhealth field. Although there some qualitative recovery studies have shown important findings, fewquantitative studies of this concept currently exist. This study examined the relationship betweenrecovery and associated socialenvironmental and individual factors. Method: A total of 159 peoplewith psychiatric disabilities receiving services from a large community mental health agency participatedin the study. Participants completed a self-report survey that assessed individual recovery status, socialsupport, perceived recovery-oriented service quality, psychiatric symptoms, and demographics. Onehundred twenty-four surveys were analyzed. Hierarchical multiple regression analysis was conducted toexamine the relationship between recovery and associated factors. Results: Social support and perceivedrecovery-oriented service quality had significant positive relationships with recovery; psychiatric symp-toms had a significant negative relationship with recovery. The final regression model accounted for 58%of the variance in recovery, F(9, 114) 17.72, p .001. Conclusion and Implications for Practice:Socialenvironmental factors play an important role in peoples recovery, even after taking into accountpsychiatric symptoms. Namely, people with psychiatric disabilities can pursue recovery with symptomsas long as they receive appropriate support and services. Mental health professionals should provideservices adhering to recovery principles in order to help their clients achieve personal recovery.

Keywords: recovery, psychiatric disability, recovery-oriented care, social support

Consumer-oriented recovery principles were used as the con-ceptual framework for this study. Unlike the traditional scientificview on recovery, which focuses on cure and symptom reduction,this consumer-oriented recovery model was developed by peoplewith psychiatric disabilities and emphasizes personal goals andpotential (Bellack, 2006; Young & Ensing, 1999). It endeavors toimprove the lives of people with psychiatric disabilities and toredesign service delivery methods. Under this concept, people withpsychiatric disabilities are treated as independent individualsrather than as dependent patients; mental health professionals arehelpers rather than controllers. Common consumer-oriented recov-

ery components include hope, empowerment, taking personal re-sponsibility, self-redefinition, and participating in meaningful ac-tivities (Davidson, OConnell, Tondora, Lawless, & Evans, 2005;Ridgway, 2001; Young & Ensing, 1999). This consumer-orientedrecovery concept has been regarded as a guiding vision of thefuture of mental health services (Anthony, 1993). More and moreagencies provide recovery-oriented services.

Social support and mental health services, which belong tosocialenvironmental factors, have been discussed in much of therecovery literature, and are regarded as important facilitators of therecovery process (Jacobson & Greenley, 2001; Spaniol, Wewior-ski, Gagne, & Anthony, 2002). Recovery advocates believe that,with appropriate support and services, people with psychiatricdisabilities can experience a better recovery process and improvetheir quality of life (Davidson, OConnell, Tondora, Styron, &Kangas, 2006; Spaniol et al., 2002).

Furthermore, advocates have also stated that people with psy-chiatric disabilities can pursue recovery even though symptomsexist (Anthony, 1993; Davidson et al., 2005). Psychiatric symp-toms, which have commonly been regarded as individual prob-lems, are simply viewed as one attribute of psychiatric disabilitiesin the consumer-oriented recovery perspective. Although somestudies have found an inverse relationship between psychiatricsymptoms and recovery (Brown, Rempfer, & Hamera, 2008;Resnick, Rosenheck, & Lehman, 2004), these advocates believethat symptoms do not necessarily prevent recovery. While people

This article was published Online First May 6, 2013.Yen-Ching Chang, PhD, OT, Department of Healthcare Administration

and Department of Occupational Therapy, I-Shou University, KaohsiungCity, Taiwan; Tamar Heller, PhD, Department of Disability and HumanDevelopment, University of Illinois at Chicago; Susan Pickett, PhD, De-partment of Psychiatry, University of Illinois at Chicago; Ming-De Chen,PhD, OT, Department of Occupational Therapy, Kaohsiung Medical Uni-versity, Kaohsiung City, Taiwan.

Supported by the Provosts Award of University of Illinois at Chicago.Correspondence concerning this article should be addressed to Yen-

Ching Chang, PhD, OT, Department of Healthcare Administration andDepartment of Occupational Therapy, I-Shou University, 8 Yida Road,Yanchao District, Kaohsiung City 82445, Taiwan. E-mail: ychang@isu.edu.tw

Psychiatric Rehabilitation Journal 2013 American Psychological Association2013, Vol. 36, No. 2, 8085 1095-158X/13/$12.00 DOI: 10.1037/h0094975

80

with physical disabilities are not expected to regain their mobilityin order to live successfully in the community, similarly, peoplewith psychiatric disabilities are not expected to eliminate theirsymptoms in order to pursue their recovery (Davidson et al., 2006;Davidson et al., 2005).

Many studies have examined factors related to traditional sci-entific definition of recovery, but few studies have investigatedfactors associated with consumer-oriented recovery (Resnick et al.,2004). Some qualitative recovery studies have found commonrecovery components and statements (Ridgway, 2001; Smith,2000; Spaniol et al., 2002). However, these results have beenlimited by small sample sizes (i.e., n 20). The above statementssupported by advocates have not been examined by quantitativestudies with a large sample size.

Therefore, this study aimed to investigate the relationship be-tween recovery and socialenvironmental and individual factors(i.e., social support, perceived recovery-oriented service quality,and psychiatric symptoms) through a self-report survey. We usedhierarchical multiple regression to examine whether socialenvi-ronmental factors have a significant relationship with recoveryafter taking into account psychiatric symptoms, and controlling fordemographic characteristics (i.e., age, illness length, sex, race, andeducation). Although few consumer-oriented recovery studieshave discussed the influence of demographic characteristics, it ispossible that these factors impact consumer-oriented recovery. Forexample, people of different ages tend to have various personalgoals, and may exhibit different recovery perspectives. Sex differ-ence may also influence recovery expectations. Because demo-graphic factors were not the focus of this study, they were con-trolled to examine accurately the relationship between recoveryand associated factors.

Method

Participants and Data Collection

Study participants were recruited from a large community men-tal health agency located in metropolitan Chicago, Illinois. Theagency provides a wide range of services, including case manage-ment, housing, vocational rehabilitation, and social skills training,to people with psychiatric disabilities, regardless of their diagnosison the Diagnostic and Statistical Manual of Mental Disorders.Because the inpatient population is relatively unstable and theconsumer-oriented recovery model may be inapplicable (Frese,Stanley, Kress, & Vogel-Scibilia, 2001), only the communitysample was considered for inclusion in this study.

Study participants had to meet the following inclusion criteria:having a diagnosis of mental illness, being age 18 years or older,living in the community, receiving services from the study site(i.e., the collaborating agency), and being able to fill out the studysurvey independently. People who were actively symptomatic orcould not understand the survey content were not enrolled.

With assistance of program staff, the first author convenedmeetings in several community programs of the agency and ex-plained the studys purpose and procedures to potential partici-pants. During these recruitment meetings, program staff helped toidentify individuals who were actively symptomatic or had limitedliteracy. These individuals were not allowed to complete surveysand were excluded from the study. After informed consent was

obtained, participants completed the self-report survey. They re-ceived a $5 gift card as a reimbursement for their time andparticipation. This study was approved by the institutional reviewboards of the University of Illinois at Chicago and the study site.Data collection occurred from June 2010 through August 2010.

A total of 159 participants filled out the survey. After removing35 surveys with significant missing data (i.e., the individual an-swered 70% of scale items; n 32) and inattentive responsesets (i.e., the individual responded to the whole survey with aspecific answer or a pattern; n 3), 124 surveys were included inthe regression analysis. A summary of characteristics of bothanalyzed and excluded participants is shown in Table 1. Except forrace and ethnicity, there were no significant differences betweenthe analyzed sample and the excluded sample. Sixty-seven percentof participants in the analyzed sample were men. Most participantswere single (73%). Nearly equal percentages of Blacks (40%) andWhites (42%) completed surveys. Regarding education level, 53%of participants had a high school degree or lower while 47% ofparticipants reported some college or higher. Most participantswere unemployed or not in the workforce (83%). The majority ofparticipants (71%) lived in a private residence or household. Fourdiagnoses were reported most often: bipolar disorder (37%),schizophrenia (24%), major depression (19%), and schizoaffectivedisorder (15%). The average age of the analyzed sample was 47.10years (range: 2068, Mdn 47.96) and the average illness lengthwas 23.82 years (range: 056, Mdn 24.41).

Instruments

The self-report survey included two parts. The first part col-lected personal information, such as age, sex, and education. Thesecond part included a battery of self-report scales. First, therevised Mental Health Recovery Measure (MHRM-R) was used tomeasure the individual recovery status. The original MHRM (Bull-ock, 2005; Young & Bullock, 2003) was developed according tothe recovery process model of Young and Ensing (1999). Chang,Ailey, Heller, and Chen (in press) evaluated the MHRM usingRasch analysis. Four items inappropriate for the measured recov-ery concept were removed from the scale to improve its validity.The revised scale (MHRM-R) has 26 items and uses a 4-pointLikert scale, ranging from 0 (strongly disagree) to 3 (stronglyagree). It assesses comprehensive recovery content, includingovercoming stuckness, self-empowerment, learning and self-redefinition, basic functioning, overall well-being, new potentials,and advocacy/enrichment. It showed high internal consistency inthe present study (Cronbachs alpha .95). Higher total scoresrepresent a better recovery status.

The 19-item Social Support Survey (SSS; Sherbourne & Stew-art, 1991) measures five types of social support: emotional sup-port, informational support, tangible support, positive social inter-action, and affectionate support. Respondents were asked howoften the support is available if they need it. Response choicesinclude: none of the time, a little of the time, some of the time, mostof the time, and all of the time. The SSS showed high internalconsistency (Cronbachs alpha .97). Higher total scores repre-sent greater receipt of social support.

The revised version of the Recovery Self-Assessment(OConnell, Tondora, Croog, Evans, & Davidson, 2005) was used

81RECOVERY OF PEOPLE WITH PSYCHIATRIC DISABILITIES

to assess perceived recovery-oriented service quality, defined inthis study as participants perceptions of the degree to which theservices they received follow recovery principles. This 32-itemscale assesses life goals, involvement, diversity of treatment op-tions, choice, individually tailored services, and inviting space. Ituses a 5-point Likert scale, ranging from 1 (strongly disagree) to5 (strongly agree), and includes an N/A (not applicable) option; itshowed high internal consistency in this study (Cronbachsalpha .97). A higher item average indicates better perceivedrecovery-oriented service quality.

Finally, psychiatric symptoms were measured by the BriefSymptom Inventory (BSI; Derogatis & Melisaratos, 1983). TheBSI is a 53-item self-report symptom scale, and has ninedimensions: somatization, obsessive compulsive, interpersonalsensitivity, depression, anxiety, hostility, phobic anxiety, para-noid ideation, and psychoticism. Respondents were asked torate the presence and severity of their symptoms in the past 7days. Each item is rated on a 5-point scale, ranging from not atall, a little bit, moderately, quite a bit, to extremely. The BSIuses three global indices of distress to describe the individualscondition, including the General Severity Index (GSI), thePositive Symptom Distress Index, and the Positive SymptomTotal. This study used the GSI to represent the severity ofpsychiatric symptoms. The BSI showed high internal consis-tency (Cronbachs alpha .98). Higher GSI scores indicategreater symptom severity.

Data AnalysisIn addition to descriptive statistics, which explored data distri-

butions and characteristics of participants, a hierarchical multipleregression analysis was conducted to further examine the relation-ship between recovery and associated factors. Demographics (i.e.,age, illness length, sex, race, and education) were entered firstbecause they were regarded as control variables. Then, the psy-chiatric symptom variable (i.e., GSI score) was entered in thesecond block, and socialenvironmental factors (i.e., social sup-port and perceived recovery-oriented service quality), which werethe focus of this study, were entered in the third block. Thisregression model explored whether the socialenvironmental fac-tors had a significant relationship with the individual recoverystatus, after taking into account other variables.

Missing data are common in self-report surveys, and they oc-curred in this study. Hawthorne and Elliott (2005) found that if atleast half the items of the scale are present, person mean substi-tution is a better choice because it has simpler computation and itsefficiency is as good as hot deck imputation. Hence, this studyused person mean imputation to handle missing data in each scale.To maintain each survey scale, person mean imputation was usedif the individual answered 7099% of the scale items. As notedabove, the 32 surveys that had significant missing data (i.e.,individuals failed to answer at least 70% of scale items) wereremoved from the analysis. SPSS, Version 17.0, for Windows wasused for data analysis.

Table 1Characteristics of Participants

Characteristics Total sampleaAnalyzed samplea

(n 124)Excluded samplea

(n 35) Test statistic dfSex 2 0.68 1

Female 50 (31%) 41 (33%) 9 (26%)Male 109 (69%) 83 (67%) 26 (74%)

Marital status 2 4.84 2Single 114 (72%) 91 (73%) 23 (66%)Married/partner 13 (8%) 7 (6%) 6 (17%)Other 32 (20%) 26 (21%) 6 (17%)

Race and ethnicityb 2 10.17 2Black 70 (45%) 49 (40%) 21 (68%)White 56 (36%) 52 (42%) 4 (13%)Other 29 (19%) 23 (19%) 6 (19%)

Education 2 0.57 1High school or lower 86 (55%) 66 (53%) 20 (61%)College or higher 71 (45%) 58 (47%) 13 (39%)

Employment 2 0.32 1Work 25 (16%) 21 (17%) 4 (13%)Not work 129 (84%) 102 (83%) 27 (87%)

Living environment 2 0.02 2Private Residence/household 110 (71%) 87 (71%) 23 (72%)Supportive/transitional housing 40 (26%) 32 (26%) 8 (25%)Other 5 (3%) 4 (3%) 1 (3%)

Diagnosis 2 1.61 3Schizophrenia/schizoaffective disorder 61 (40%) 47 (38%) 14 (50%)Major depression 27 (20%) 23 (19%) 4 (14%)Bipolar disorder 54 (36%) 46 (37%) 8 (29%)Other 9 (6%) 7 (6%) 2 (7%)

Age (years) 47.04 (11.12) 47.10 (10.81) 46.77 (12.73) t 0.14 148Illness length (years) 24.07 (12.18) 23.82 (12.17) 27.93 (12.41) t 0.92 130a Total sample values are n (%) or M (SD). b p .006.

82 CHANG, HELLER, PICKETT, AND CHEN

Results

The hierarchical multiple regression results are presented inTable 2. The first regression model was not significant,F(6, 117) 0.63, p .71. After entering psychiatric symptoms,the second model, F(7, 116) 4.85, p .001, accounted for 23%of variance in recovery, and R2 change was significant, F(1,116) 29.29, p .001. When socialenvironmental factorswere entered, the third model, F(9, 114) 17.72, p .001,accounted for 58% of variance in recovery, and R2 change wassignificant as well, F(2, 114) 48.79, p .001.

Social support, perceived recovery-oriented service quality, psy-chiatric symptoms, and illness length were significantly associatedwith recovery status in the third model. Participants with greatersocial support, better perceived recovery-oriented service quality,lower symptom severity, and longer illness lengths tended to havehigher recovery scores.

DiscussionThis study provided preliminary findings on the factors signif-

icantly associated with consumer-oriented recovery. Results of thehierarchical multiple regression analysis found that social support,perceived recovery-oriented service quality, psychiatric symp-toms, and illness length had significant relationships with individ-ual recovery status and accounted for a significant amount ofvariance in recovery. These findings have several implications formental health providers who seek to enhance clients recovery.

Social support had the positive and highest standardized coef-ficient in the final regression model (see Table 2), which indicatesthat it had the most impact in the model. People with more socialsupport tend to have a better recovery status. This result is similarto that of previous research findings (Corrigan & Phelan, 2004;Hendryx, Green, & Perrin, 2009; Pernice-Duca & Onaga, 2009),and indicates the importance of social support for people in recov-ery. It also suggests that programs that facilitate connectionsamong people with psychiatric disabilities may enhance theirrecovery. For example, because peer support has been regarded as

an important support for people with psychiatric disabilities (Mead& Copeland, 2000; Substance Abuse and Mental Health ServicesAdministration, 2005), adding peer support groups to existingprograms may facilitate peoples recovery process.

Perceived recovery-oriented service quality also showed a sig-nificant positive relationship with recovery. In this study, per-ceived recovery-oriented service quality was assessed by partici-pants perceptions of whether the services they received wererecovery-oriented. Although recovery-oriented services have notbeen clearly identified, they have several characteristics in com-mon, including offering services that are consumer-centered andthat assist individuals in achieving personal goals (Anthony, 2000;Noordsy et al., 2002). In addition, attitudinal changes in mentalhealth professionals are the key of recovery-oriented services.Mental health professionals need to believe that recovery is pos-sible, to respect clients decisions, and to provide different sug-gestions and options for people in different recovery levels (An-thony, 1993; Mead & Copeland, 2000; Smith, 2000). Although allstudy participants were from the same mental health agency, theymay have had different experiences in receiving services due tovarious attitudes or behaviors of designated service providers andvariability in the quality of the specific programs provided. Theresearch finding indirectly confirms the effectiveness of servicesthat are perceived as recovery oriented. Namely, people whoreceive services that are viewed as adhering more to recoveryprinciples tend to have better recovery statuses. Therefore, it isrecommended that mental health professionals, administrators, andpolicy-makers implement recovery principles in their work.Greater adaptation of recovery-oriented services is likely to lead togreater recovery among people with psychiatric disabilities.

Moreover, psychiatric symptoms had a significant negative re-lationship with recovery. Using the MHRM-R, this study hadfindings similar to other previous studies, which assessed recoverywith recovery-related measures, such as hope and empowermentscales (Brown et al., 2008; Resnick et al., 2004). The result is notsurprising. Symptom reduction has been the main focus of thetraditional scientific recovery paradigm and has been discussed by

Table 2The Hierarchical Regression Model for Recovery (n 124)

Block

First model Second model Third model

t t t

Block 1Age 0.01 0.05 0.09 0.88 0.11 1.61Illness length 0.06 0.58 0.11 1.12 0.16 2.24Female 0.13 1.39 0.13 1.57 0.02 0.37Black 0.07 0.57 0.01 0.10 0.07 0.80White 0.03 0.20 0.05 0.48 0.12 1.36High school or lower 0.12 1.28 0.13 1.55 0.10 1.48

Block 2Psychiatric symptoms 0.46 5.41 0.27 4.15

Block 3Social support 0.49 6.23Perceived recovery-oriented service quality 0.22 2.85

R2a 0.03 0.23 0.58F for change in R2 0.63 29.29 48.79

a Unadjusted R2. p .05. p .01. p .001.

83RECOVERY OF PEOPLE WITH PSYCHIATRIC DISABILITIES

numerous researchers and professionals (Harrow, Grossman, Jobe,& Herbener, 2005; Whitehorn, Brown, Richard, Rui, & Kopala,2002). However, besides investigating the effectiveness of certainmedication and interventions, researchers often overlook that it ishard for people with psychiatric disabilities to cope with symptomswithout a good relationship with their service providers. In fact,many of them have unpleasant experiences with mental healthprofessionals (Mead & Copeland, 2000). To better assist thispopulation, mental health professionals should learn how to workwith their clients and to help them find better medication or copingstrategies (Sowers & Quality Management Committee of theAmerican Association of Community Psychiatrists, 2005). Treat-ing clients with respect and support is one of the critical principlesof recovery-oriented services (Jacobson & Greenley, 2001). Whenpeople with psychiatric disabilities are treated appropriately, it ismore likely that they can handle their symptoms and make goodprogress in their recovery.

In terms of demographics, only illness length showed a signif-icant positive relationship with recovery in the final model. Peoplewith longer illness lengths tend to have a better recovery status.This result corresponds with previous qualitative findings (Dee-gan, 1988; Smith, 2000; Spaniol et al., 2002). Recovery does nothappen suddenly, and it is not an easy process. It takes time forpeople with psychiatric disabilities to accept their illness, have adesire to change their lives, and to seek help. People with psychi-atric disabilities are likely to enter their recovery journey whenthey get to know themselves and the illness better.

Overall, the research findings are encouraging. Socialenviron-mental factors had a significant contribution to the final regressionmodel, which supports the assertion of recovery advocates: peoplewith psychiatric disabilities can experience a better recovery pro-cess and pursue better quality of life if they receive appropriatesupport and services (Davidson et al., 2006; Spaniol et al., 2002).Although psychiatric symptoms seem to have a negative impacton peoples recovery, the positive relationships between recov-ery and social support as well as perceived recovery-orientedservice quality after taking into account psychiatric symptomswere evident. This finding advances another advocates asser-tion: people with psychiatric disabilities can pursue recoveryeven though symptoms persist (Anthony, 1993; Davidson et al.,2005). Namely, even though they have symptoms, as long as thesocialenvironmental support is available, people with psychi-atric disabilities can gradually improve their lives and pursuetheir recovery.

Study LimitationsSeveral limitations exist in this study. First, cross-sectional data

cannot determine the causal direction of these relationships be-tween recovery and associated factors. It is unclear whether theimprovement in social support, perceived recovery-oriented ser-vice quality, and psychiatric symptoms influences recovery and/orwhether the enhancement of recovery helps people with psychiat-ric disabilities to receive social support, better services, and controltheir symptoms.

Second, several factors limit the generalization of this study.The present study sample is limited to people with psychiatricdisabilities served in one recovery-oriented mental health agency.The results may not reflect experiences of people without services

or receiving no recovery-oriented services. From their demograph-ics and responses to the survey, this sample tended to represent apopulation that had relatively stable conditions, both in regard tosymptoms and to the environmental support system. Moreover,because participants were required to fill out the survey indepen-dently, the experiences of people with limited literacy were ex-cluded from this study.

Finally, the use of a self-report survey also resulted in severalstudy limitations. Social desirability (Huang, Liao, & Chang,1998) and missing data are common in self-report surveys. Par-ticipants might answer items in a way that matches social desir-ability and could skip items that they did not want to answer. Thesemay cause an overestimation or underestimation of the results.However, the self-report survey is valuable for appropriately re-flecting the respondents perceptions. This study used anonymousparticipation to decrease the influence of social desirability, andused person mean substitution to better estimate participantsresponses, hence increasing the reliability of the findings.

Although the study participants may not represent all peoplewith psychiatric disabilities living in the community, our studyresults contribute important quantitative evidence for theconsumer-oriented recovery concept. Future research should col-lect data from various mental health agencies and attempt to reachpeople with limited services. People with low literacy may beincluded by face-to-face interviews. A larger and diverse samplecan expand the generalization of study findings. Moreover, al-though these associated factors may truly have significant contri-butions, it will be beneficial to have more empirical studies toreconfirm these results.

Conclusion and Implications for PracticeThe study explored the relationship between recovery and so-

cialenvironmental and individual factors. We found that socialsupport and perceived recovery-oriented service quality had sig-nificant positive relationships with recovery; psychiatric symp-toms had a significant negative relationship with recovery. Thefinal regression model accounted for 58% of variance in recovery.These findings support the statements of recovery advocates. So-cialenvironmental factors do play an important role in peoplesrecovery, even after taking into account psychiatric symptoms. Itindicates that people with psychiatric disabilities can pursue re-covery with symptoms as long as they receive appropriate supportand services. In addition to symptom control, people with psychi-atric disabilities who live in the community also need adequatesupport and services to improve their lives and achieve theirpersonal goals.

The results of this study are useful for mental health servicedesigns and mental health policy-making. Mental health profes-sionals can have more confidence to follow the consumer-oriented recovery paradigm, and they are encouraged to adoptand provide recovery-oriented services to help their clientsachieve personal recovery. These research results expand theknowledge base of the consumer-oriented recovery concept, andthey are beneficial for further follow-up or randomized controlledstudies, which can provide stronger evidence to verify the rela-tionship between recovery and associated factors.

84 CHANG, HELLER, PICKETT, AND CHEN

ReferencesAnthony, W. A. (1993). Recovery from mental illness: The guiding vision

of the mental health service system in the 1990s. Psychosocial Reha-bilitation Journal, 16(4), 1123.

Anthony, W. A. (2000). A recovery-oriented service system: Setting somesystem level standards. Psychiatric Rehabilitation Journal, 24(2), 159168.

Bellack, A. S. (2006). Scientific and consumer models of recovery inschizophrenia: Concordance, contrasts, and implications. SchizophreniaBulletin, 32(3), 432442.

Brown, C., Rempfer, M., & Hamera, E. (2008). Correlates of insider andoutsider conceptualizations of recovery. Psychiatric RehabilitationJournal, 32, 2331. doi:10.2975/32.1.2008.23.31

Bullock, W. A. (2005). Mental Health Recovery Measure (MHRM). In T.Campbell-Orde, J. Chamberlin, J. Carpenter, & H. S. Leff (Eds.), Mea-suring the promise: A compendium of recovery measures (Vol. 2, pp.3641). Cambridge, MA: The Evaluation Center@HSRI.

Chang, Y. C., Ailey, S. H., Heller, T., & Chen, M. D. (in press). Raschanalysis of the Mental Health Recovery Measure. American Journal ofOccupational Therapy.

Corrigan, P. W., & Phelan, S. M. (2004). Social support and recovery inpeople with serious mental illnesses. Community Mental Health Journal,40(6), 513523.

Davidson, L., OConnell, M. J., Tondora, J., Lawless, M., & Evans, A. C.(2005). Recovery in serious mental illness: A new wine or just a newbottle? Professional Psychology: Research and Practice, 36, 480487.doi:10.1037/0735-7028.36.5.480

Davidson, L., OConnell, M., Tondora, J., Styron, T., & Kangas, K. (2006).The top ten concerns about recovery encountered in mental healthsystem transformation. Psychiatric Services, 57(5), 640645.

Deegan, P. E. (1988). Recovery: The lived experience of rehabilitation.Psychosocial Rehabilitation Journal, 11(4), 1119.

Derogatis, L. R., & Melisaratos, N. (1983). The Brief Symptom Inventory:An introductory report. Psychological Medicine, 13(3), 595605.

Frese, F. J., 3rd, Stanley, J., Kress, K., & Vogel-Scibilia, S. (2001).Integrating evidence-based practices and the recovery model. Psychiat-ric Services, 52(11), 14621468.

Harrow, M., Grossman, L. S., Jobe, T. H., & Herbener, E. S. (2005). Dopatients with schizophrenia ever show periods of recovery? A 15-yearmulti-follow-up study. Schizophrenia Bulletin, 31(3), 723734.

Hawthorne, G., & Elliott, P. (2005). Imputing cross-sectional missing data:Comparison of common techniques. Australian and New Zealand Jour-nal of Psychiatry, 39(7), 583590.

Hendryx, M., Green, C. A., & Perrin, N. A. (2009). Social support,activities, and recovery from serious mental illness: STARS study find-ings. Journal of Behavioral Health Services & Research, 36, 320329.doi:10.1007/s11414-008-9151-1

Huang, C. Y., Liao, H. Y., & Chang, S. H. (1998). Social desirability andthe clinical self-report inventory: Methodological reconsideration. Jour-nal of Clinical Psychology, 54(4), 517528.

Jacobson, N., & Greenley, D. (2001). What is recovery? A conceptualmodel and explication. Psychiatric Services, 52(4), 482485.

Mead, S., & Copeland, M. E. (2000). What recovery means to us: Con-sumers perspectives. Community Mental Health Journal, 36(3), 315328.

Noordsy, D. L., Torrey, W. C., Mueser, K. T., Mead, S., OKeefe, C. O.,& Fox, L. (2002). Recovery from severe mental illness: An interpersonaland functional outcome definition. International Review of Psychiatry,14(4), 318326.

OConnell, M., Tondora, J., Croog, G., Evans, A., & Davidson, L. (2005).From rhetoric to routine: Assessing perceptions of recovery-orientedpractices in a state mental health and addiction system. PsychiatricRehabilitation Journal, 28(4), 378386.

Pernice-Duca, F., & Onaga, E. (2009). Examining the contribution ofsocial network support to the recovery process among clubhouse mem-bers. American Journal of Psychiatric Rehabilitation, 12, 130. doi:http://dx.doi.org/10.1080/15487760802615566

Resnick, S. G., Rosenheck, R. A., & Lehman, A. F. (2004). An exploratoryanalysis of correlates of recovery. Psychiatric Services, 55(5), 540547.

Ridgway, P. (2001). ReStorying psychiatric disability: Learning from firstperson recovery narratives. Psychiatric Rehabilitation Journal, 24(4),335343.

Sherbourne, C. D., & Stewart, A. L. (1991). The MOS social supportsurvey. Social Science & Medicine, 32, 705714. doi:10.1016/0277-9536(91)90150-B

Smith, M. K. (2000). Recovery from a severe psychiatric disability: Find-ings of a qualitative study. Psychiatric Rehabilitation Journal, 24(2),149158.

Sowers, W., & Quality Management Committee of the American Associ-ation of Community Psychiatrists. (2005). Transforming systems ofcare: The American Association of Community Psychiatrists guidelinesfor recovery oriented services. Community Mental Health Journal,41(6), 757774.

Spaniol, L., Wewiorski, N. J., Gagne, C., & Anthony, W. A. (2002). Theprocess of recovery from schizophrenia. International Review of Psy-chiatry, 14(4), 327336.

Substance Abuse and Mental Health Services Administration. (2005).National consensus conference on mental health recovery and systemstransformation. Rockville, MD: Department of Health and Human Ser-vices.

Whitehorn, D., Brown, J., Richard, J., Rui, Q., & Kopala, L. (2002).Multiple dimensions of recovery in early psychosis. International Re-view of Psychiatry, 14(4), 273283.

Young, S. L., & Bullock, W. A. (2003). The Mental Health RecoveryMeasure. Toledo, OH: University of Toledo, Department of Psychology.

Young, S. L., & Ensing, D. S. (1999). Exploring recovery from theperspective of people with psychiatric disabilities. Psychiatric Rehabil-itation Journal, 22(3), 219231.

85RECOVERY OF PEOPLE WITH PSYCHIATRIC DISABILITIES