Psychiatr Q (2011) 82:69–84 DOI 10.1007/s11126-010-9148-9 ORIGINAL PAPER
Social Value of Supported Employment for Psychosocial Program Participants Paul J. Barreira • Miriam Cohen Tepper • Paul B. Gold Dana Holley • Cathaleene Macias
•
Published online: 3 September 2010 Ó Springer Science+Business Media, LLC 2010
Abstract Members of a psychiatric psychosocial program designed to provide both supported employment and peer support were surveyed about their current social activities, sources of social support, and social life improvement since joining the program. Survey respondents who worked a mainstream job (n = 17) reported greater peer contact in community locations, and correspondingly greater social life improvement, than those who remained unemployed or worked volunteer jobs (n = 45). Results of a hierarchical regression analysis (N = 62) that explored this positive correlation between mainstream work, community-based peer contact, and social life satisfaction suggest that working a job in an integrated setting that paid at least minimum wage encouraged program participants
P. J. Barreira (&) Behavioral Health and Academic Counseling, Harvard University Health Services, 5 Linden Street, Cambridge, MA 02138, USA e-mail:
[email protected] P. J. Barreira M. C. Tepper C. Macias Department of Psychiatry, Harvard Medical School, Boston, MA, USA M. C. Tepper Cambridge Health Alliance, Cambridge, MA, USA e-mail:
[email protected] P. J. Barreira M. C. Tepper D. Holley Waverley Place, McLean Hospital, Belmont, MA, USA e-mail:
[email protected] P. B. Gold Department of Counseling and Personnel Services, University of Maryland at College Park, College Park, MD, USA e-mail:
[email protected] C. Macias Department of Community Intervention Research, McLean Hospital, Belmont, MA, USA e-mail:
[email protected]
123
70
Psychiatr Q (2011) 82:69–84
to meet and interact in community locations, thereby strengthening peer mutual support while furthering social integration. This unique pattern of findings requires replication, and we recommend that other psychosocial programs conduct similar quality improvement studies to provide further insights into the relationship between peer support and community integration. Keywords Psychiatric rehabilitation Social integration Peer support Supported employment Participatory research Continuous quality improvement
Social rejection intensifies depression, anxiety, and suicidal ideation [1, 2], especially for adults who are middle-aged or older [3], and loneliness weakens the immune system, depletes the body’s ability to cope with chronic illness and aging, and can be as physically damaging as smoking or obesity [4]. Isolation also engenders distrust and paranoia [5, 6] and, for all these reasons, loneliness and social rejection are particularly adverse for anyone with a severe psychiatric illness [7, 8]. Conversely, positive social support is very therapeutic [9, 10]. Nevertheless, most adults with a severe mental illness are socially isolated [11–16], and many are so lonely that even negative relationships seem preferable to being alone [17, 18].
Definitions of Social Support and Social Integration for Adults with Psychiatric Illness The term ‘social support’ connotes the simple value of companionship for reducing loneliness, but the concept itself encompasses a complex set of subjective experiences, e.g., confidence-sharing, reciprocity, mutual trust [19, 20]. ‘Social integration’ is also a complex concept, but most human rights advocates use this term simply to refer to the right of disabled individuals to participate freely in organizations and activities open to non-disabled adults [21–23]. Recently, this basic definition of social integration has been extended to include the concept of social support, so that ‘social integration’ is now more often defined by mental health researchers as a disabled individual’s experience of connectedness to her or his own neighborhood and local community [24–27]. This more complex definition has become a catalyst for the development of supported socialization interventions parallel to supported housing, supported education, and supported employment. For instance, a number of new programs recruit non-disabled volunteers to accompany disabled adults in community activities [28–30]. It is not yet clear whether the primary value of such intentional friendships is their inherent ‘integration,’ or the companionship they provide for wider community participation [31]. To the extent that companionship makes community involvement easier, peer friendships may also promote social integration. Many peer support programs for adults with psychiatric illness were founded on the premise that friendships between disabled individuals generate momentum toward self-sufficiency, instilling the self-confidence needed to face challenging social situations [32–34]. It is less intimidating to shop for clothes, try a new restaurant, join a book club, or work out in a gym if you have a trusted friend as a companion. Friendship with another disabled individual is often characterized as more stigmatizing than friendship with a non-disabled individual, or more socially restricting, but a peer friendship may sometimes be more ‘normal’ in the sense of being more egalitarian and reciprocal [35], and peer friendships are not necessarily restrictive. For instance, in a study of British day centres for adults with psychiatric illness, program participants
123
Psychiatr Q (2011) 82:69–84
71
listed neighbors, shopkeepers, and workmates as their confidants nearly as often as they listed co-participants [36].
Supported Employment’s Promotion of Social Integration Supported employment (SE) was designed to increase disabled individuals’ access to competitive jobs that pay at least minimum wage and are located in mainstream, integrated settings. Social integration has historically been a viable secondary goal for supported employment programs because competitive jobs place disabled individuals on an equal footing with non-disabled co-workers, and often require disabled and non-disabled employees to work side-by-side [37, 38]. However, there is sparse evidence for SE effectiveness in encouraging social integration because few SE studies have measured integration in the workplace, or even social life satisfaction. Many non-experimental studies report a positive correlation between mainstream employment and social life quality [e.g., 37, 39–41], but it is not clear whether this correlation reflects the benefits of working in mainstream settings, success of SE programs in facilitating coworker relationships, or simply the superior social skills of people able to work mainstream jobs [18]. Moreover, no randomized trial has yet demonstrated that supported employment services effectively increase either social integration or social life satisfaction. In one SE randomized trial [42], only participants assigned to a comparison condition showed social life improvement following employment, apparently because this psychosocial program’s work-related activities enhanced peer friendships [43]. A second SE trial reported a trend toward better social outcomes for a similar psychosocial program in comparison to the focal SE intervention [44]. Two other SE trials reported that any type of employment had beneficial effects, by either decreasing ‘social disability’ [45] or increasing satisfaction with finances and leisure activities [46, 47]. The latter finding suggests mainstream employment may be socially beneficial primarily because work connotes normalcy [48, 49] and provides the disposable income needed for recreational pursuits [50]. None of the comparison programs in these trials provided supported employment services or prioritized mainstream work as a vocational goal, so there was no opportunity to evaluate the role of peer friendships in encouraging mainstream work, or to test the alternative assumption that peer friendships discourage job-site friendships with co-workers [51]. Nor was it clear in any psychosocial program description whether peer contact took place only within these service settings versus in the local community where peer mutual support might have fostered social integration. In the present naturalistic study, we conducted a confidential survey of 62 active participants in a typical psychosocial program that provided an array of community-support services, inclusive of supported employment. We asked each individual to rate his or her social life improvement since joining the program, and to describe sources of social support and participation in social activities inside and outside the program. We then compared the frequency, quality, and source of social support received by individuals employed in mainstream jobs versus their unemployed counterparts. Two research hypotheses were derived from the program’s aims: Hypothesis 1 An interactive psychosocial rehabilitation program can improve participants’ social lives by encouraging peer social contact in natural community settings. Hypothesis 2 Supported employment improves the social lives of participants in an interactive psychosocial rehabilitation program primarily because mainstream work encourages the transfer of program-based peer friendships to natural community settings.
123
72
Psychiatr Q (2011) 82:69–84
Method Program Description The program under study was Waverley Place in Belmont, MA, a community-based rehabilitation program for adults with severe mental illness. The program offers all its members (consumers) the opportunity to spend time in casual conversation, group discussions, and other interactive social, educational, and recreational activities in the same building where they receive professional and peer counseling and practical help with daily living [52]. However, attendance is not mandatory, and members are also welcome to sit quietly by themselves, reading, knitting, or listening to music. All group activities, including support groups, are planned and/or co-led by members based on current personal interests. To familiarize members with the surrounding neighborhood, Waverley Place staff and peer staff routinely accompany members on walks or drives to cafe´’s, shopping centers, exercise gyms, swimming pools, ball games, libraries, movies, and museums. The program is also open as a drop-in center on weekends. Job-hunting assistance and on-job support for Waverley Place members are provided primarily through one-on-one interactions with Waverley Place staff [53] One-on-one supported employment services include newspaper and internet searches, social networking for job leads and references, resume preparation, interview coaching and practice, job-related skill and travel training, and work-focused supportive counseling, including discussions of social security benefits, income budgeting, family expectations, time management, and job pressures. All program members are invited to join a weekly employment support group where they can share their own accomplishments and experiences, and gain insights and job information from other members. A weekly ‘soft skills’ vocational group focuses on how to finesse job interviews, get to work on time, and get along with bosses and coworkers. Members can also choose to work volunteer jobs in local community agencies (e.g., hospitals, libraries, nursing homes) to gain work experience and exposure to new situations. Sample Characteristics The survey sample represented three-quarters of the target population of currently active participants (74%, n = 53 of 73), as well as nine participants who had recently joined (n = 4) or returned (n = 5) to the program. At the time of the survey, 45% (n = 28) of the sample had been attending the program for 5 or more years, and 70% (n = 43) had been active members for at least 3 years. One-half of the survey sample were women (52%; n = 32), and, in keeping with the ethnicity of the surrounding community, 89% (n = 55) were Caucasian. Every participant had a disabling psychiatric disorder, with one-half (52%, n = 32) having been diagnosed with a schizophrenia spectrum disorder. The majority of participants (82%; n = 51) had never been married, and two-thirds (68%, n = 42) had a monthly income of less than $1200. First psychiatric hospitalization occurred at a median age of 25 years, and most participants had been psychiatrically ill for two or more decades (median age = 46.8; M = 46.9, SD = 10.2). Nevertheless, 40% (n = 25) had been able to complete some postsecondary schooling/training, and another 40% had achieved a college degree. Survey respondents did not differ significantly, or substantially, from other program participants on any of these background variables.
123
Psychiatr Q (2011) 82:69–84
73
Procedures All research procedures were part of the program’s continuous quality improvement efforts approved by the McLean Hospital institutional review board. The CQI survey described in this article was conducted in February and March of 2009 and targeted members who had more than 3 days of program attendance during either this first quarter (January–March) of 2009, or the preceding fourth quarter (October–December) of 2008. This selection criterion ensured that most participants had fresh memories of the same program events, as well as the same seasonal celebrations and news-worthy occurrences in the surrounding community. Every program participant in attendance during February and March of 2009 was invited by a staff worker to participate in the survey at his or her convenience, and each completed survey was placed by the participant into an envelope and then signed across the seal to ensure it would not be opened by anyone except the external program evaluator, who stored the data on a password-protected computer. A follow-up survey of employed members was conducted a few weeks later in the same manner. Members were considered to be working in mainstream employment if that job met the U.S. Department of Labor definition of ‘competitive employment:’ (a) any individually-held job, (b) located in a mainstream, integrated setting, which (c) paid minimum wage or higher [54]. This definition included both permanent and temporary positions in local businesses, as well as self-employment (e.g., yard work, snow removal) if the work exceeded 10 h per week. Program attendance records for the first quarter of 2009 were used to measure frequency of exposure to other members at the program site. Total days of attendance were also calculated for each survey respondent for each quarterly period in 2008 and 2009. Duration of membership was calculated as calendar days from date of program enrollment to March 31, 2009. Survey Measures (1) Social Life Improvement Scale (4 Items) Has your social life changed since you became a member? Have your friendships changed since you became a member? Have your family relationships changed as a member? Are you less lonely since becoming a member? Change was self-rated using a 7-point scale with anchors ranging from 1 = ‘Worse/Terrible’ to 7 = ‘Better/Great.’. Cronbach’s alpha (.66) was adequate for a four-item scale. (2)
Quality of Staff Social Support (4 Items)
Staff support was measured as the sum of ratings (1 = ‘Never’ to 4 = ‘Always’) across four items designed by program members and staff: Are staff genuinely concerned about you? Do staff treat you with dignity and respect? Can you locate a staff worker when you need to talk? Do you enjoy socializing with staff? Cronbach’s alpha was a = .75. (3)
Quality of Peer Social Support (4 Items)
Peer support was measured as the sum of ratings (1 = ‘Never’ to 4 = ‘Always’) across the four items of Brekke & Aisley’s ‘Client Interaction Scale’ [55]: Do other members help you reach your goals? When you have a problem, do other members help you out? When confused, do other members help you sort things out? Do you feel better when you spend time with other members? Cronbach’s alpha was a = .75.
123
74
Psychiatr Q (2011) 82:69–84
(4)
Frequency of Social Contact Outside the Program
Usual Contact with Family and Friends (2 Items) We measured usual social contact by summing the two ‘objective’ items that comprise the ‘Social Life’ subscale of Lehman’s Quality of Life instrument [56]: How often do you talk on the telephone with family or friends? How often do you get together with family or friends? Both item scales ranged from 1 = ‘Not at all’ to 5 = ‘Once a day,’ but the two items correlated only r = ?.39 (P \ .01) because while 85% (n = 53) of the survey respondents talked by telephone at least once a week, 61% (n = 38) visited this often. However, the sum of item ratings were normally distributed, with only one person reporting no contact with a significant other person by phone or in person, so the sum score appears to be a meaningful continuous measure of usual social contact. Contact with Other Program Members (2 Items) ‘Peer contact outside the program’ was measured using an adaptation of Lehman’s ‘Social Life’ items, described above, with the same 1–5 scaling and similar wording: How often do you talk on the telephone with another member? How often do you do something outside the program with another member? The correlation between these two items was r = ?.74. In addition, the 17 employed members who took part in the follow-up survey were asked to record the names of the friends from the program who they saw regularly in one another’s homes or local public places, and to describe these activities using checklists and personal narratives. Employed survey respondents also reported on friendships at work, and were asked to compare these friends to the peer friendships they had formed in the program. Data Analysis Plan Our research hypotheses were tested in two separate hierarchical regression analyses. Each analysis had the sum score for the 4-item Social Life Improvement Scale as the dependent variable. The first regression analysis (Table 2) tested Hypothesis 1 which posits that members of this psychosocial program who report more frequent contact with peers outside the program will also report greater social life improvement since becoming a member. In Block 1, we statistically controlled for opportunities to form peer friendships (program attendance & duration of membership) and usual frequency of social contact (Lehman’s QOL subscale). In Block 2, we entered members’ ratings of the quality of peer social support along with their ratings of staff social support. In Block 3, we added ‘peer contact outside the program.’ If this final variable is a significant predictor of social life improvement, and if ‘quality of peer social support’ also remains statistically significant and the full model is significant, we will infer that the location of peer contact is a salient predictor of social life quality over and above relationship quality. That is, we will infer that peer friendships that take place outside a rehabilitation program can enhance members’ social lives in ways uniquely different from social interactions that take place during program-sponsored activities. The second regression analysis (Table 3) tested Hypothesis 2 which posits that supported employment provided by a psychosocial program improves the social lives of employed individuals by increasing ‘peer contact outside the program,’ which in turn increases ‘quality of peer social support.’ Block 1 included the main predictor variable, ‘mainstream employment’ and controlled for frequency of program attendance to rule out
123
Psychiatr Q (2011) 82:69–84
75
the possibility that employed individuals may have had greater exposure to peers during their receipt of supported employment services. ‘Peer contact outside the program’ was then entered in Block 2, followed by self-rated ‘quality of peer social support’ in Block 3. To assume that ‘mainstream employment’ mediates the positive relationship between ‘peer contact outside the program’ and ‘social life improvement’ predicted for the first regression analysis, ‘mainstream employment’ would need to be a significant predictor of social life improvement over and above program attendance, but then become non-significant, with a reduced beta value, when ‘peer contact outside the program’ is added to the regression model. Likewise, ‘peer contact outside the program’ would need to be a significant predictor of social life improvement when first entered in Block 2, but no longer statistically significant after ‘quality of peer social support’ is added. If these changes in beta coefficients are observed, and if the regression model as a whole is statistically significant, we will infer that program members who were employed in mainstream jobs reported greater social life improvement than non-employed members because they spent more time with other members outside the rehabilitation program, and this peer contact outside the program increased the quality of their peer friendships. Confidence in this interpretation would derive from our a prior prediction of a complex pattern of findings, but the interpretation would remain tentative because we could not randomly assign study participants to employed versus unemployed conditions, nor measure peer contact and peer support at intervals before versus after periods of employment.
Results Association Between Peer Social Support and Social Life Improvement Survey respondents’ average rating of social life improvement after joining the psychosocial program was M = 20.56 (SD = 2.86), which corresponds to an average item rating of 5.1 (‘‘More Satisfied’’). The difference in mean scores across the four items that comprise the Social Life Improvement Scale (F = 10.43; df = 3, 61; P \ .001) varied from a mean of 4.75 (SD = 0.98) for change in family relationships to a mean of 5.51 (SD = 0.85) for change in loneliness. Most social support measures correlated moderately with one another and with social life improvement (r = ?.36 to ?.52), except that social support received from staff had minimal association with peer contact outside the program (r = ?.13, ns). However, none of these survey measures of program-related social support, or the measure of social life improvement, correlated significantly with Lehman’s quality of life measure of usual contact with family or close friends, or with any participant background characteristic. Hypothesis Test 1 An interactive psychosocial program can improve participants’ social lives by encouraging peer social contact in natural community settings. The results of the first hierarchical regression analysis (Table 2) support this hypothesis. In Block 1, we statistically controlled for participants’ routine exposure to the program (duration of membership, attendance) and usual rate of contact with family and close friends. When entered together in Block 1, none of these three variables predicted social life improvement. In Block 2, we entered our first predictor variable, ‘quality of peer social support,’ along with ‘social support from staff’ to control for the possibility that affiliation with staff may have enhanced a participant’s social standing among peers. Quality of peer social support significantly predicted social life improvement even when controlling for
123
76
Psychiatr Q (2011) 82:69–84
quality of relationships with staff, level of exposure to peers at the program, and usual contact with family and other friends. In Block 3, we entered ‘contact with peers outside the program,’ which was a statistically significant predictor of social life improvement independently of ‘quality of peer social support,’ which also retained a significant beta in the statistically significant full regression model (P \ .001). Association Between Employment, Peer Social Support, and Social Life Improvement The program under study appears to have provided high quality supported employment services. At the time of the survey in the first quarter of 2009, 27% (n = 17 of 62) of the survey respondents were employed in mainstream jobs. This employment rate was fairly comparable to the program’s overall 25% mean quarterly employment rate for 2008, and previous 25% mean quarterly employment rate for 2007, and only slightly lower than the employment rate (34%, n = 25 of 73) for the target population of all currently active members over a longer 6-month period (last quarter of 2007 through first quarter of 2008). These quarterly employment rates compare favorably to quarterly rates reported for similar types of supported employment programs in other locations within the United States [57]. Nearly one-half (46%; n = 12 of 26) of the employed participants in the target population worked 20 h or more a week, and a third (35%, n = 9) worked full-time. Likewise, one-half of the employed respondents in the survey sample (53%; n = 9 of 17) worked 20 h or more per week. All except two employed participants in the target population held onto their jobs for at least 6 months, and only one survey respondent worked less than 6 months. Median hourly wage for both the target population and the survey sample was $11. Job descriptions were diverse, ranging from landscaping work to nursing home aide to motel desk clerk. Overall, study participants who worked a mainstream job reported greater social life improvement than unemployed participants (Table 1). Employed participants also reported significantly higher quality social support from peers in the program, and were more likely to socialize with peers outside the program (82%, n = 14 vs. 56%, n = 25). Among those survey respondents who reported seeing a fellow participant outside the program at least monthly (n = 24), 69% of unemployed participants (n = 9 of 13) socialized about once a month, while 82% (n = 9 of 11) of employed participants socialized with another program member at least once a week (Chi square = 8.28, df = 2, P \ .05). On the other hand, employed and unemployed participants had similar rates of program attendance, similar durations of program membership, and comparably good relationships with program staff. Hypothesis Test 2 Supported employment improves the social lives of participants in an interactive psychosocial program primarily because mainstream work encourages the transfer of program-based peer friendships to natural community settings. The results of our second hierarchical regression analysis (Table 3) support this second study hypothesis. In Block 1, mainstream employment was a significant predictor of social life improvement, even when statistically controlling for exposure to peers during program attendance. However, when ‘contact with peers outside the program’ was added in Block 2, it also significantly predicted social life improvement, while substantially reducing the significant relationship between mainstream employment and social life improvement that had been evident in Block 1. That is, as predicted, time spent socializing with other participants outside the program appears to account for the positive relationship between mainstream employment and social life improvement. When ‘quality of peer social support’ is then entered in Block 3, it also predicts social life improvement, and its addition to
123
Psychiatr Q (2011) 82:69–84
77
Table 1 Employed versus unemployed members in the 2009 survey (N = 62) Variable
Mainstream work (N = 17)
No mainstream work (N = 45)
Characteristic
N
N
Male gender
10
59
17
38
ns
3
18
14
31
ns
Chronic health condition Over age 50
%
P
%
0
0
16
36
.004
15
88
36
82
ns
Schizophrenia diagnosis
8
47
24
53
ns
College degree
7
41
18
40
ns
Never married
Service receipt in 2008 Membership duration (years) Program attendance (total days)a Survey scale ratings
M ± SD
M ± SD
P
4.1 ± 2.0
4.2 ± 2.3
ns
99.6 ± 58.6
127.0 ± 93.1
ns
M ± SD
M ± SD
P
Social life improvement
22.3 ± 2.6
19.9 ± 2.7
.003
Social support from staff
14.5 ± 1.8
13.8 ± 2.0
ns
Quality of peer social support
11.6 ± 2.5
9.7 ± 2.5
.008
Phone contact with peersb
3.0 ± 1.6
2.0 ± 1.2
.017
Got together with peersb
3.0 ± 1.3
2.0 ± 1.1
.004
a
Average (mean) program attendances from January 1, 2008 to March 31, 2009
b
Two measures of peer contact outside the program: ‘How often do you talk on the telephone with another member?’ & ‘How often do you do something outside the program with another member?’. Scale anchors: 2 = Less than once a month; 3 = At least once a month; Scores on both items ranged from 1 to 5
the model substantially reduces the positive association between ‘contact with peers outside the program’ and social life improvement, suggesting that the reason peer contact outside the program improves the social lives of employed members is that it strengthens the quality of peer friendships. Because the predicted changes in variable betas were observed, and the full regression model is statistically significant, these ‘mediational’ inferences receive empirical support from our retrospective survey data. However, our interpretation of change will remain tentative until this same pattern of findings is replicated by prospective studies. It should be kept in mind that this positive impact of employment on peer social support holds only for individuals who are younger than age 50. As can be seen in Table 1, the significant age difference between employed versus unemployed participants in the survey sample (M = 49.5; SD = 9.5 vs. M = 40.0; SD = 8.8; t = 3.57; P \ .01) was due entirely to the fact that no employed participant was older than age 50. Within the subsample of individuals under age 50 (n = 46), employed and unemployed participants were more comparable in age (M = 41.0; SD = 8.4 vs. M = 43.7; SD = 7.9; t = 1.07; P = ns). Individuals under age 50 did not differ significantly from those older than age 50 on any study variable other than employment, and only two individuals in the survey sample had reached age 65, so the complete absence of employed individuals over age 50 may have more to do with age discrimination than age differences in functioning or retirement eligibility. Fortunately, age does not correlate with any measure of social support (r = -.02 to
123
78
Psychiatr Q (2011) 82:69–84
Table 2 Hierarchical regression analysis of the relationship between psychosocial program peer friendships and social life improvement (N = 62) Predictor variables
Step 1 Beta b
Step 2 b 95% CI
Step 3
Beta b
b 95% CI
Beta
b
b 95% CI
Block 1:a (control variables) Years of program membership
.12
.15 .7 [-.2, ?.5]
.07
.10
.6 [-.2, ?.4] .03
.04
.5 [-.2, ?.3]
Attendance in current quarter
.02
.00 .1 [.0, ?.1]
.11
.02
.1 [.0, ?.1]
.01
.1 [.0, ?.1]
Usual contact with family/friends
.12
.21 1.0 [-.3, ?.7] .01
.01
.8 [-.4, ?.4] -.03 -.05 .9 [-.5, ?.4]
Support from program staff
.13
.18
.8 [-.2, ?.6] .13
Quality of peer social support
.47
.51** .6 [?.2, ?.8] .39
.04
Block 2:b .19
.8 [-.2, ?.6]
.42** .6 [?.1, ?.7]
Block 3:c Peer contact outside program
.26
Intercept
18.23***
11.88***
R2, DR2
.03
.30***
.31*
.6 [.0, ?.6]
12.33*** .27***
.36***
.06*
Beta standardized coefficient, b unstandardized coefficient, CI confidence interval * P \ .05, ** P \ .01, *** P \ .001 a
Attendance = program contact days during quarter when survey was conducted: Jan 1, 2009–March 31, 2009; Usual contact with family/friends = Lehman’s [56] Quality of Life social contact scale; score range: 2–10 b
Support from Program Staff = Staff Satisfaction Scale; score range: 9–16; Quality of Peer Social Support = Sum score on Brekke and Aisley’s [55] Client Interaction Scale. Score range: 4–16
c
Frequency of non-program contact with other program participants was measured using similar wording and the same ratings as Lehman’s [56]; Social Life subscale: 1 = Not at all, 2 = Less than once a month, 3 = Once a month, 4 = Once a week, 5 = Once a day
?.08; P = ns), or social life improvement (r = ?.04; P = ns), so our inability to control for the fact that employed participants were older than unemployed participants does not compromise the validity of these regression analyses. Follow-Up Survey with Employed Members The community-based social activities listed by employed respondents ranged from simply hanging out together in a coffee shop to playing basketball, and the locations of these activities were common-place, e.g., homes, cafe´s, restaurants, movie theatres, parks, shopping malls, gyms, and walks through local neighborhoods. Almost all of the currently employed members who were in the follow-up survey (89%, n = 16) reported getting together regularly with another program member to have a meal in a restaurant (n = 10), see a movie in a theatre (n = 5), or ‘hang out’ in one another’s homes to listen to music or watch DVDs (n = 9). On average, these peer friendships had lasted about 3 years (M = 3.2, Median = 3.0, range = 1–7 years) when the survey was conducted in early
123
.07
.14*
R2, DR2
0.00
2.44**
b
0.1 [0.0, 0.1]
3.1 [0.9, 4.0]
95% CI
.22**
18.57***
.31
-.01
.26
Beta
Step 2
.08*
.36*
-.00
1.63
b
0.6 [0.1, 0.7]
0.0 [0.0, 0.0]
3.3 [0.0, 3.3]
95% CI
.37***
14.29***
.42
.21
.06
.16
Beta
Step 3
.15**
.45**
.24
.00
1.0
b
0.5 [0.2, 0.7]
0.5 [0.0, 0.5]
0.1 [0.0, 0.1]
3.1 [-0.5, 2.6]
95% CI
c
Sum score for the four-item Client Interaction Scale [55]. Self-ratings ranged from 4 to 16
Frequency of non-program contact with other program participants was measured using similar wording and the same ratings as Lehman’s [56]; Social Life subscale: 1 = Not at all, 2 = Less than once a month, 3 = Once a month, 4 = Once a week, 5 = Once a day
b
a Mainstream employment was defined as any (1 = Yes, 0 = No) socially-integrated job paying at least minimum wage that lasted at least 1 month; Current Attendance = program contact days during quarter when survey was conducted: Jan 1, 2009–March 31, 2009
* P \ .05, ** P \ .01, *** P \ .001
Beta standardized coefficient, b unstandardized coefficient, CI b confidence interval
19.60***
Intercept
Quality of peer social supportc
Block 3
Peer contact outside programb
Block 2
.38
Attendance in current quarter
Beta
Step 1
Any mainstream employmenta
Block 1
Predictor variables
Table 3 Hierarchical regression analysis comparing mainstream work and psychosocial program friendships as predictors of social life improvement (N = 62)
Psychiatr Q (2011) 82:69–84 79
123
80
Psychiatr Q (2011) 82:69–84
2009. Except for one respondent who enrolled in the program with a friend she had just met, every peer friendship had developed in the psychosocial program under study. In general, there was substantial diversity and reciprocity in friendships. Most respondents listed one or two friends, and there were 22 unique friendship dyads listed by the 16 respondents. Three individuals formed a triad friendship, each listing the other two as close friends they saw together. Only one participant listed another respondent as a friend who did not also list her as a friend. More than half the respondents (n = 10) also listed unemployed program participants as friends they saw often outside the program. However, only 6 of these 16 employed individuals said their relationships with co-workers at their place of employment were as close as their relationships with other program members, and only three individuals saw their co-workers socially outside the workplace. Several survey respondents added a hand-written note that described their relationships at work as positive and supportive, but less intimate than their peer friendships.
Discussion Findings from this study of 62 adults with serious mental illness suggest that a psychosocial program can improve the social lives of its members by first encouraging peer friendships and mutual support, and then by encouraging the transfer of these peer friendships to local community settings. Hierarchical regression analysis findings further suggest that psychosocial programs can improve the social lives of members by providing supported employment services that help them find mainstream jobs, because mainstream work appears to encourage the transfer of peer friendships to community settings, thereby strengthening the quality of these friendships. We do not yet know why people who worked a mainstream job reported greater peer contact in community locations, and correspondingly greater social life improvement, but it seems reasonable to assume that mainstream jobs that required routine contact with non-disabled individuals strengthened social selfconfidence and provided the discretionary funds needed to eat out, see movies, shop, or join sports and hobby groups. These survey findings suggest that the primary contribution of supported employment to social integration may not be the formation of new workplace friendships, but rather the transfer of existing peer friendships into community settings where mutual-support provides a normative sense of connectedness to society [58]. Friends listed by employed individuals as their partners in community activities were predominantly co-members of the same psychosocial program. By comparison, most employed respondents reported that they spent relatively little time with co-workers from their place of work during off-job hours. Since the peer friendships that transferred from the psychosocial program to the community were almost all unique dyads, friends appear to have chosen one another deliberately based on personal preferences and shared interests. And, because frequency of attendance and duration of membership in the psychosocial program did not correlate with peer contact in the community or ratings of peer support quality, these friendships appear to have gained autonomy from the program that fostered them. We do not know at what point program members began seeing one another in the community, but it is likely that program co-participation provided a fertile environment for the social confidence gained from mainstream employment because community interaction was concurrent with program attendance when the survey was conducted. Our survey did not ask to what extent co-participation in community activities spurred new friendships with non-disabled adults,
123
Psychiatr Q (2011) 82:69–84
81
or helped these individuals get to know their neighbors and other people they saw on a daily basis, but it appears that these out-of-program friendships do prevent the ‘reinstitutionalization’ that can occur whenever a disabled individual restricts his or her social life to service settings [32]. Study Limitations An alternative explanation for our findings is that only those individuals who were confident and socially skilled enough to work a mainstream job were capable of sustaining peer relationships and interacting with friends in community settings. This alternative explanation applies to any study that contrasts employed versus unemployed samples because, while research participants can be randomized to services, it is impossible to randomly assign ‘employed vs. unemployed’ status. Although we did not measure social skills or social anxiety, the general similarity of employed and unemployed participants in regard to education, marital status, psychiatric diagnosis, program attendance, and social support received from staff (Table 1) suggest that reasons for not working were diverse, and difficulty socializing was only one of several obstacles to mainstream work. Moreover, there was no evidence that competitive work was more or less socially-demanding than the occupations of many unemployed participants. For instance, while about a third (n = 5) of competitive jobs required minimal social contact (e.g., construction, lawn care, dog walking, library book scanning), almost all unemployed participants who worked a volunteer job (n = 11 of 45; 24%) had routine social interaction with non-disabled adults in a hospital, library, church, nursing home, or museum, and others (n = 7 of 45; 16%) earned pocket-money selling Mary Kay Cosmetics or their knitting, artwork, or jewelry at local craft fairs. Our survey data provide only retrospective reports of social life improvement, so our cause-effect inference that mainstream employment increased social interaction in the community requires replication of this same pattern of findings in longitudinal studies that schedule data collection to correspond to the timing of employment. We would also have greater confidence that the social life improvement reported by employed participants were attributable to the program under study if we had been able to randomize participants to either this psychosocial program or a SE program that offered no opportunities for peer social interaction. If employed individuals participated more frequently in community social activities than unemployed participants only in the psychosocial program condition, we could then infer that providing peer support in tandem with SE is an effective way to improve the social lives of disabled adults over and above job-site social integration. Based on the present study’s correlational findings, we can say that our hypotheses are worth testing in a randomized trial. Implications for Service Design Our findings should generalize to most psychosocial programs for adults with severe mental illness that encourage peer friendships and adhere to a philosophy of consumer empowerment and self-determination. The program under study intentionally encouraged autonomy by allowing its members to choose how and when they wanted to participate, and what services they wanted to receive. It also appears advantageous to allow peer friendships to develop naturally in keeping with individual preferences, as they did in this program, since researchers report that psychiatrically disabled adults resist, and probably resent, being matched to one another as companions in community activities [28, 29]. On the other hand, it may have been helpful for this psychosocial program to introduce its
123
82
Psychiatr Q (2011) 82:69–84
members to the community activities that interested them, so they could become familiar with local neighborhoods and bus routes, and feel more comfortable venturing out on their own to local cafe´s, parks, and gyms. Conclusion This naturalistic study of a psychosocial program demonstrated a three-way positive association between mainstream employment, peer companionship in local community activities, and social life improvement that suggests the attainment of mainstream work improved the social lives of employed individuals by encouraging the transfer of peer friendships from a service program to community settings where these friendships continued to thrive. A prior study of this same program demonstrated that SE services can be effectively provided by psychosocial program professionals practicing as generalists [53], and this program’s 2008 and 2009 quarterly employment rates were similar to the quarterly rates reported for other SE programs [57], so it seems reasonable to recommend that psychosocial programs designed to promote both peer support and social integration add supported employment to their array of rehabilitation services.
References 1. Slavich GM, Thornton T, Torres LD, et al.: Targeted rejection predicts hastened onset of major depression. Journal of Social and Clinical Psychology 28(2):223–243, 2009 2. Morgan C, Burns T, Fitzpatrick R, et al.: Social exclusion and mental health. British Journal of Psychiatry 191:477–483, 2007 3. Cacioppo JT, Hughes ME, Waite LJ, et al.: Loneliness as a specific risk factor for depressive symptoms: Cross-sectional and longitudinal analyses. Psychology and Aging 21(1):140–151, 2006 4. Cacioppo JT, Patrick W: Loneliness: Human Nature and the Need for Social Connection. New York, NY, W W Norton, 2008 5. Birchwood M: Pathways to emotional dysfunction in first episode psychosis. Journal of Psychosis 182:373–375, 2003 6. Gilbert P, Boxall M, Cheung M, et al.: The relation of paranoid ideation and social anxiety in a mixed clinical population. Clinical Psychology and Psychotherapy 12:124–133, 2005 7. Avison WR, Mcleod JD, Pescosolido BA: Mental Health, Social Mirror. New York, NY, Springer, 2007 8. Nilsson B, Naden D, Lindstrom UA: The tune of want in the loneliness melody—Loneliness experienced by people with serious mental suffering. Scandinavian Journal of Caring Science 22:161–169, 2008 9. Gearing RE, Mian I, Sholonsky A, et al.: Developing a risk-model of time to first-relapse for children and adolescents with a psychotic disorder. Journal of Nervous and Mental Disease 197(1):6–14, 2009 10. Corrigan PW, Phelan SM: Social support and recovery in people with serious mental illnesses. Community Mental Health Journal 40(6):513–523, 2004 11. Brunt D, Hansson L: The social networks of persons with severe mental illness in in-patient settings and supported community settings. Journal of Mental Health UK 11(6):611–621, 2002 12. Meeks S, Hammond CT: Social network characteristics among older outpatients with long-term mental illness. Journal of Mental Health & Aging 7(4):445–464, 2001 13. Goldberg RW, Rollins AL, Lehman AF: Social network correlates among people with psychiatric disabilities. Psychatric Rehabilitation Journal 26(4):393–402, 2003 14. Kilbourne AM, McCarthy JF, Post EP, et al.: Social support among veterans with serious mental illness. Social Psychiatry & Psychiatric Epidemiology 42:639–646, 2007 15. Pescosolido BA, Wright ER: The view from two worlds: The convergence of social network reports between mental health clients and their ties. Social Science & Medicine 58:1795–1806, 2004 16. Macdonald EM, Hayes RL, Baglioni AJ: The quantity and quality of the social networks of young people with early psychosis compared with closely matched controls. Schizophrenia Research 46(1):25–30, 2000
123
Psychiatr Q (2011) 82:69–84
83
17. Green G, Hayes C, Dickinson D, et al.: The role and impact of social relationships upon wellbeing reported by mental health service users: A qualitative study. Journal of Mental Health 11(5):565–579, 2002 18. Padgett DK, Henwood B, Abrams C, et al.: Social relationships among persons who have experienced serious mental illness, substance abuse, and homelessness: Implications for recovery. American Journal of Orthopsychiatry 78(3):333–339, 2008 19. Rogers ES, Anthony W, Lyass A: The nature and dimensions of social support among individuals with severe mental illnesses. Community Mental Health Journal 40(5):437–450, 2004 20. Brekke JS, Long JD, Kay DD: The structure and invariance of a model of social functioning in schizophrenia. Journal of Nervous & Mental Disease 190(2):63–72, 2002 21. Boski P: Five meanings of integration in acculturation research. International Journal of Intercultural Relations 32(2):142–153, 2008 22. van de Ven L, Post M, de Witte L, et al.: It takes two to tango: The integration of people with disabilities into society. Disability & Society 20(3):311–329, 2005 23. Scheff TJ: A concept of social integration. Philosophical Psychology 20(5):579–593, 2007 24. Ware NC, Hopper K, Tugenberg T, et al.: Connectedness and citizenship: Redefining social integration. Psychiatric Services 58(4):469–474, 2007 25. Eklund M, Hansson L: Social network among people with persistent mental illness: Associations with sociodemographic, clinical, and health-related factors. International Journal of Social Psychiatry 53(4):293–305, 2007 26. Townley G, Kloos B: Development of a measure of sense of community for individuals with serious mental illness residing in community settings. Journal of Community Psychology 37(3):362–380, 2009 27. Mezzina R, Borg M, Marin I, et al.: From participation to citizenship: How to regain a role, a status, and a life in the process of recovery. American Journal of Psychiatric Rehabilitation 9:39–61, 2006 28. Staeheli M, Stayner D, Davidson L: Pathways to friendship in the lives of people with psychosis: Incorporating narrative into experimental research. Journal of Phenomenological Psychology 35(2):233–252, 2004 29. Davidson L, Shahar G, Stayner D, et al.: Supported socialization for people with psychiatric disabilities: Lessons from a randomized trial. Journal of Community Psychology 32(4):453–477, 2004 30. McCorkle BH, Rogers ES, Dunn EC, et al.: Increasing social support for individuals with serious mental illness: Evaluating the Compeer Model of Intentional Friendship. Community Mental Health Journal 44:359–366, 2008 31. Perese EF, Wolf M: Combating loneliness among persons with severe mental illness: Social network interventions’ characteristics, effectiveness, and applicability. Issues in Mental Health Nursing 26:591–609, 2005 32. Schutt RK, Rogers ES: Empowerment and peer support: Structure and process of self-help in a consumer-run center for individuals with mental illness. Journal of Community Psychology 37(6):697–710, 2009 33. Davidson L, Chinman MJ, Kloos B, et al.: Peer support among individuals with severe mental illness: A review of the evidence. Clinical Psychology: Science and Practice 9(2):165–187, 1999 34. Whitley R, Harris M, Fallot RD, et al.: The active ingredients of intentional recovery communities: Focus group evaluation. Journal of Mental Health 17(2):173–182, 2008 35. Forrester-Jones R, Barnes A: On being a girlfriend not a patient: The quest for an acceptable identity amongst people diagnosed with a severe mental illness. Journal of Mental Health 17(2):153–172, 2008 36. Catty J, Goddard K, White S, et al.: Social networks among users of mental health day care: Predictors of social contacts and confiding relationships. Social Psychiatry & Psychiatric Epidemiology 40:467–474, 2005 37. Forrester-Jones R, Jones S, Heason S, et al.: Supported employment: a route to social networks. Journal of Applied Research in Intellectual Disabilities 17:199–208, 2004 38. Riches VC, Green VA: Social integration in the workplace for people with disabilities: An Australian perspective. Journal of Vocational Rehabilitation 19:127–142, 2003 39. Banks B, Charleston S, Grossi T, et al.: Workplace supports, job performance, and integration outcomes for people with psychiatric disabilities. Psychiatric Rehabilitation Journal 24(4):389–396, 2001 40. Ru¨esch P, Graf J, Meyer PC, et al.: Occupation, social support and quality of life in persons with schizophrenic or affective disorders. Social Psychiatry & Psychiatric Epidemiology 39:686–694, 2004 41. Salyers MP, Becker DR, Drake RE, et al.: A ten-year follow-up of a supported employment program. Psychiatric Services 55(3):302–308, 2004 42. Rollins AL, Mueser KT, Bond GR, et al.: Social relationships at work: Does the employment model make a difference? Psychatric Rehabilitation Journal 26(1):51–61, 2002 43. Petryshen PM, Hawkins JD, Fronchak TA: An evaluation of the social recreation component of a mental health program. Psychatric Rehabilitation Journal 24(3):293–298, 2001
123
84
Psychiatr Q (2011) 82:69–84
44. Mueser KT, Clark RE, Haines M, et al.: The Hartford study of supported employment for persons with severe mental illness. Journal of Consulting & Clinical Psychology 72(3):479–490, 2004 45. Burns T, Catty J, White S, et al.: The impact of supported employment and working on clinical and social functioning: Results of an international study of Individual Placement and Support. Schizophrenia Bulletin 35(5):949–958, 2009 46. Drake RE, McHugo GJ, Bebout RR, et al.: A randomized clinical trial of supported employment for inner-city patients with severe mental disorders. Archives of General Psychiatry 56:627–633, 1999 47. Bond GR, Resnick SG, Drake RE, et al.: Does competitive employment improve nonvocational outcomes for people with severe mental illness? Journal of Consulting & Clinical Psychology 69(3):489–501, 2001 48. Perkins DV, Raines JA, Tschopp MK, et al.: Gainful employment reduces stigma toward people recovering from schizophrenia. Community Mental Health Journal 45:158–162, 2009 49. Rogers JA: Work is key to recovery. Psychosocial Rehabilitation Journal 18(4):5–10, 1995 50. Heider D, Angermeyer MC, Winkler I, et al.: A prospective study of Quality of Life in schizophrenia in three European countries. Schizophrenia Research 93:194–202, 2007 51. Catty J, Lissouba P, White S, et al.: Predictors of employment for people with severe mental illness: Results of an international six-centre randomised controlled trial. British Journal of Psychiatry 192:224–231, 2008 52. Barriera PJ, Macias C, Rodican CF, et al.: Choice of service provider: How consumer self-determination shaped a psychiatric rehabilitation program. Psychatric Rehabilitation Journal 31(3):202–210, 2008 53. Barriera PJ, Tepper MC, Gold PB, et al.: Adapting evidence-based interventions to fit usual practice: Staff roles and consumer choice in psychiatric rehabilitation. Psychiatric Quarterly 81(2):139–155, 2009 54. Department of Labor: Job Training Program Act, Disability Grant Program Funded Under Title III, Section 323, and Title IV, Part D, Section 452, vol 63, Federal Register, 1998 55. Brekke JS, Aisley R: The client interaction scale: A method for assessing community support programs for persons with chronic mental illness. Evaluation & the Health Professions 13(2):215–226, 1990 56. Lehman AF, Kernan E, Postrado L: Toolkit for Evaluating Quality of Life for Persons with Severe Mental Illness. Cambridge, MA, HSRI, 1997 57. Gold PB, Macias C, Barreira PJ, et al.: Viability of using employment rates from randomized trials as benchmarks for supported employment program performance. Administration & Policy in Mental Health and Mental Health Services Research, 2009 58. Ware NC, Hopper K, Tugenberg T, et al.: A theory of social integration as quality of life. Psychiatric Services 59(1):27–33, 2008
Author Biographies Paul J. Barreira, MD is Director of Behavioral Health and Academic Counseling at Harvard University and Associate Professor of Psychiatry at Harvard Medical School. He is the founding director of the Waverley Place outpatient program at McLean Hospital in Belmont, MA. Miriam Cohen Tepper, MD is a psychiatrist at the Cambridge Health Alliance in Somerville, MA, and Associate Director of Waverley Place at McLean Hospital in Belmont, MA. Paul B. Gold, PhD is an assistant professor in the Department of Counseling and Personnel Services at the University of Maryland in College Park, MD. Dana Holley, MSW, LCSW was the program manager of Waverley Place at McLean Hospital in Belmont, MA at the time the study was conducted. Cathaleene Macias, PhD is Director of Community Intervention Research at McLean Hospital in Belmont, MA.
123
