Efficacy of a peer-led group program for unemployed people with mental health problems: Pilot randomized controlled trial

Abstract

Background:

People with long-term unemployment and mental health problems often find it difficult to take active steps toward help-seeking and job search and to navigate the complex system of available services. Likewise, job center staff would welcome interventions to improve the reintegration of long-term unemployed individuals with mental health problems into the labor market.

Aim:

To examine the efficacy of a peer-led group program that supports unemployed people with mental health problems in terms of help-seeking, job search and recovery.

Methods:

Based on participatory research, a four-session group program was designed and evaluated in a pilot randomized controlled trial (RCT) with 42 participants, randomized to the program (n = 23) or treatment as usual (n = 19). Outcomes were assessed at baseline (T0), 3 weeks (T1), 6 weeks (T2) and 6 months later (T3).

Results:

There were no significant intervention effects on primary outcomes (job search self-efficacy and help-seeking). But compared to the control group, intervention participants showed significant improvements in depressive symptoms (p = .02) and recovery (p = .04) at T2 with medium effect sizes. There were trend-level positive program effects on self-stigma, hopelessness and secrecy.

Conclusion:

This pilot RCT provides initial evidence for the efficacy of a peer-led group program to improve symptoms and recovery among unemployed participants with mental health problems.

Keywords

Unemployment mental health problems peer-led intervention group program recovery job search

Introduction

People with long-term unemployment and mental health problems often find themselves in a vicious circle defined by multiple social and medical problems (Brand, 2015). Qualitative research suggests that despite available services and sources of support, many find it difficult to take active steps toward help-seeking and job search (Audhoe, Nieuwenhuijsen, Hoving, Sluiter, & Frings-Dresen, 2018; Oschmiansky et al., 2017; Schubert et al., 2013). The efficacy of supported employment (individual placement and support) programs is well established, but they are not widely implemented in Germany, and many unemployed individuals with mental health problems find it difficult to navigate the complex system of available services. Likewise, job center staff would welcome new tools and interventions to improve the reintegration of long-term unemployed individuals with mental health problems into the labor market (Reissner et al., 2016).

At the time and location of this study in Southern Germany, unemployment rates were low at about 3%. Nevertheless, many unemployed individuals with mental health problems struggle to re-enter the labor market, resulting in long periods of unemployment. Taking into account an expert consensus that recommends to address mental health literacy and skills (Santos et al., 2018) as well as qualitative and quantitative work of our group among unemployed people with mental health problems (Rüsch et al., 2018, 2019; Staiger, Waldmann, Rüsch, & Krumm, 2017), we designed and evaluated a peer-led group program for this target population in a pilot randomized controlled trial (RCT).

It was the aim of our study to evaluate the efficacy of a group program as specific support in addition to any primary or secondary mental healthcare (treatment as usual (TAU)) as well as any employment agency support that participants received. Supported employment was not offered as part of the study intervention, by the employment agencies or other providers. Compared to the control group that received TAU alone, we expected that program participants would show improvements in terms of job search self-efficacy, help-seeking intentions as well as symptoms, stigma variables and indices of recovery and well-being.

Methods

Trial design and participants

In this parallel two-arm RCT, 42 participants were randomly assigned to the group program combined with TAU, if any (n = 23), or to a control group with TAU alone (n = 19), using a random number table sequence and closed envelopes. Participants were recruited in 2017 from job centers and community information centers for the unemployed and via newspaper advertisements in the Southern German regions of Ulm and Augsburg. Inclusion criteria were current unemployment (following German law defined as working <15 hours/week and earning ⩽450 Euros/month), age: 18–64 years, sufficient German language skills, current psychological distress as indicated by scores ⩾13 on Kessler’s K6 Psychological Distress Scale (M = 17.5, SD = 2.9; Kessler et al., 2010) and at least a moderate level of self-reported distress due to the current situation (score ⩾3 in response to ‘How mentally distressed do you feel with respect to your current situation?’, from 1/not at all to 5/very much; M = 4.0, SD = 0.8). Full disability pension was an exclusion criterion. The trial was approved by Ulm University’s ethics committee, and all participants provided written informed consent. Before including the first participant, the trial was registered on ISRCTN (38560825). Control group and intervention participants were reimbursed for their time during assessments, not for group participation.

Intervention

Based on focus groups with unemployed people with mental health problems and our previous qualitative (Staiger et al., 2017) and quantitative studies (Rüsch et al., 2018), an intervention was developed by our group that consists of researchers and people with experience of mental illness and/or unemployment. The intervention focused on peer support, values and acceptance, help-seeking, support for mental health problems and job search and disclosure decisions. The group program included elements of Hayes’ Acceptance and Commitment Therapy (Zhang et al., 2017) and of Corrigan’s peer-led Honest, Open, Proud program to support people with mental illness in their disclosure decisions (www.hopprogram.org). Groups were co-facilitated by one peer with own experience of unemployment and mental illness and by one clinical psychologist. Each group ran for 6 weeks and had four 2-hour sessions, one each in weeks 1, 2 and 3 plus a booster session in week 6. Lesson 1 introduced the mindfulness concept, followed by discussions of acceptance related to unemployment and mental health and help-seeking; it further discussed pros and cons of disclosing a mental health problem. Lesson 2 discussed participants’ values and their behavioral implications and levels of disclosure with risks and benefits. Lesson 3 repeated values and provided practical information about sources of help for job search, financial and related issues as well as for mental health. Lesson 4 covered acceptance of thoughts and feelings about one’s personal situation, and in terms of disclosure, Lesson 4 taught participants ways to tell their story that they could practice if they wished to do so. Groups were co-led by the two group facilitators; exercises, worksheets and homework were included in a workbook.

Measures

All outcomes were measured by self-report at four times: at baseline before randomization (T0), 3 weeks (T1), 6 weeks (T2) and 6 months (T3) after baseline (see Table 1 for ranges of possible scores). Higher mean or sum scores of each scale indicate higher levels of the measured constructs. Primary outcomes were assessed by the 6-item Job Search Self-Efficacy Scale (with Cronbach’s alphas between 0.76 and 0.89, for T0–T3; van Ryn & Vinokur, 1992) and the 13-item General Help-Seeking Questionnaire as a measure of help-seeking intentions for mental health problems (alphas 0.64–0.69, except for 0.39 at T3; Wilson, Deane, Ciarocchi, & Rickwood, 2005).

Table 1.

ANCOVAs for intervention and control group.

Scales (range of possible scores)		T0M (SD)	T1M (SD)	T2M (SD)	T3M (SD)	T1			T2			T3
Scales (range of possible scores)		T0M (SD)	T1M (SD)	T2M (SD)	T3M (SD)	F	η²	p	F	η²	P	F	η²	p
Job-search self-efficacy (1–5)	Intervention	3.1 (0.8)	3.2 (0.7)	3.4 (0.8)	3.2 (1.0)	2.1	0.06	.16	1.4	0.04	.24	<0.01	<0.01	.96
Job-search self-efficacy (1–5)	Control	3.3 (0.8)	3.2 (0.9)	3.3 (0.9)	3.2 (0.9)	2.1	0.06	.16	1.4	0.04	.24	<0.01	<0.01	.96
Help-seeking intentions (1–7)	Intervention	3.4 (1.0)	3.6 (1.0)	3.5 (1.0)	3.3 (0.6)	2.6	0.07	.12	0.2	0.007	.64	0.04	<0.01	.85
Help-seeking intentions (1–7)	Control	3.0 (1.0)	3.0 (0.9)	3.2 (1.0)	3.1 (0.9)	2.6	0.07	.12	0.2	0.007	.64	0.04	<0.01	.85
Recovery (SISR: 4–24)	Intervention	15.3 (4.1)	16.1 (4.2)	17.8 (3.4)	17.3 (3.5)	0.4	0.01	.53	4.6	0.13	.039	1.9	0.06	.17
Recovery (SISR: 4–24)	Control	16.3 (4.2)	15.9 (3.6)	16.1 (4.4)	15.7 (2.6)	0.4	0.01	.53	4.6	0.13	.039	1.9	0.06	.17
Self-stigma (SSMIS-SF: 5–45)	Intervention	17.3 (4.3)	15.3 (6.2)	14.6 (7.9)	15.4 (7.0)	0.8	0.02	.39	10.8	0.25	.08	1.5	0.05	.23
Self-stigma (SSMIS-SF: 5–45)	Control	17.3 (8.1)	16.9 (8.0)	19.1 (8.7)	18.3 (9.6)	0.8	0.02	.39	10.8	0.25	.08	1.5	0.05	.23
Depressive symptoms (CES-D: 15–60)	Intervention	39.2 (7.4)	35.2 (8.8)	31.5 (8.5)	32.0 (7.6)	0.6	0.02	.44	6.3	0.17	.017	0.9	0.03	.35
Depressive symptoms (CES-D: 15–60)	Control	41.4 (7.2)	39.1 (10.2)	40.1 (10.1)	37.5 (9.8)	0.6	0.02	.44	6.3	0.17	.017	0.9	0.03	.35
Hopelessness (BHS: 4–24)	Intervention	14.9 (3.6)	13.8 (3.9)	12.6 (3.6)	12.3 (4.2)	0.2	0.01	.65	3.4	0.10	.08	2.4	0.07	.14
Hopelessness (BHS: 4–24)	Control	14.1 (4.5)	13.7 (4.6)	14.2 (5.0)	14.4 (3.0)	0.2	0.01	.65	3.4	0.10	.08	2.4	0.07	.14
Secrecy (1–6)	Intervention	4.0 (1.2)	3.8 (1.0)	3.5 (1.3)	3.7 (0.9)	0.2	0.01	.64	2.8	0.08	.10	2.2	0.07	.15
Secrecy (1–6)	Control	3.7 (1.6)	3.7 (1.7)	3.9 (1.7)	4.3 (1.4)	0.2	0.01	.64	2.8	0.08	.10	2.2	0.07	.15

SD: standard deviation; SISR: Self-Identified Stage of Recovery Scale; CES-D: Center for Epidemiologic Studies-Depression Scale; BHS: Beck’s Hopelessness Scale; ANCOVA: analysis of covariance; SSMIS-SF: Self-Stigma of Mental Illness Scale-Short Form.

ANCOVAs with the baseline score of each outcome as covariate; baseline/T0; 3 weeks after baseline/T1; 6 weeks after baseline, that is after the booster session for intervention participants/T2; and follow-up 6 months after baseline/T3.

Secondary outcome measures were the 5-item self-concurrence subscale of the Self-Stigma of Mental Illness Scale-Short Form, indicating to what extent participants agreed that negative stereotypes applied to themselves (alphas from 0.60 to 0.76 from T1 to T3 and 0.40 at T0; Corrigan et al., 2012); a short 4-item version of Beck’s Hopelessness Scale (BHS, alphas from 0.74 to 0.78; Yip & Cheung, 2006); the 4-item part B of the Self-Identified Stage of Recovery Scale (SISR, alphas from 0.60 to 0.80; Andresen, Caputi, & Oades, 2010); the 15-item German version of the Center for Epidemiologic Studies-Depression Scale (CES-D, alphas from 0.83 to 0.93; Hautzinger & Bailer, 1993) and Link’s 5-item Secrecy Scale as an index of the preference to keep one’s mental health problem secret (alphas from 0.64 to 0.90; Link, Mirotznik, & Cullen, 1991).

Statistical analyses

Baseline characteristics of intervention versus control participants were compared using t-tests or Fisher’s exact tests as appropriate. Intervention effects were examined by analyses of covariance, with the baseline/T0 score of each outcome variable as covariate and partial η² values as effect size estimates. Analyses were done in SPSS 25 with a significance level of p < .05.

Results

Recruitment, baseline characteristics and attrition

Altogether, 92 individuals were assessed for eligibility of whom 50 were excluded (37 did not meet inclusion criteria, 3 were retired and 10 were not interested) and 42 were included in the study. Participants were Caucasian, on average in their mid forties (M = 46.1 years, SD = 9.9, median: 48), 22 (52%) were female, had on average 13.5 years of education (SD = 2.9), had been unemployed for nearly 4 years since their last job (M = 46.3 months, SD = 57.1, median: 27.0), most (n = 31, 78%) lived alone and two-thirds (n = 28) had debts. The average level of depressive symptoms about 40 in the CES-D (Table 1) was well above the recommended cutoff for clinically relevant depression of 37 (with possible scores from 15 to 60; Hautzinger & Bailer, 1993), and 29 participants (69%) scored above this cutoff at baseline. Recruitment was ended after including 42 participants due to continuing recruitment difficulties.

Three types of current mental health service use among the 42 included participants were assessed at baseline, with 50% (n = 21) reporting to see a psychiatrist or psychotherapist, 48% (n = 20) receiving care for mental health problems from a general practitioner and 43% (n = 18) taking psychiatric medication. Service use rates did not differ between groups (all three p values >.2 in two-tailed Fisher’s exact tests). Participants did not receive supported employment, neither as part of the study intervention nor as part of employment agency support or other services.

We compared baseline/T0 characteristics of program versus control group participants in terms of socio-demographic variables and outcome measures, and there were no significant differences. Starting with 23/19 (intervention/control group) participants at T0, data were available from 21 (91%)/17 (89%) at T1, 18 (78%)/17 (89%) at T2 and 20 (87%)/13 (68%) at T3, indicating increasing attrition rates over time with approximately 10% dropout at T1 and 20% at T3 across both groups.

Intervention effects

Regarding primary outcomes, there were no significant intervention effects on job-search self-efficacy or help-seeking intentions (Table 1). However, a number of secondary outcomes showed significant or trend-level positive effects at T2 (after the booster session for the intervention group): program participants reported significantly lower depressive symptoms and more recovery with medium effect sizes (Table 1). Small-to-medium trend-level positive intervention effects were also found for self-stigma, hopelessness and secrecy.

Participants’ views

Program participants gave feedback about the intervention in an open-ended questionnaire informally. Perceived strengths of the program included the positive and open atmosphere in the groups that provided the experience of not being alone with these kinds of problems; to have been taken seriously as a person; the peer group facilitator as a role model with own experience of unemployment and ill mental health; experiences with acceptance and mindfulness as a new way to deal with stressors that cannot be changed easily; and the fact that participants could, but did not have to, express their opinions. As a downside, some felt there was too much content for the four sessions.

Discussion

There is substantial evidence for the efficacy of supported employment programs for unemployed people with mental illness (Modini et al., 2016; Suijkerbuijk et al., 2017). However, they are not widely implemented in Germany, and for long-term unemployed individuals with mental health problems and unclear motivation to seek employment and to use services (Hagen, Bänfer, Werkstetter, Hebebrand, & Reissner, 2018), there is a need for tailored interventions. Our pilot RCT for this population provides partial support for the feasibility of a peer-led group program with moderate attrition rates. The main feasibility concern in our trial was recruitment, which may have been particularly difficult because some potential participants might not have considered themselves as having mental health problems, reducing their interest in participation. Another explanation might be general disillusionment with any kind of intervention after many years of unemployment. Motivating members of the target population to participate in group programs of this kind will likely be a challenge for future studies. A low-threshold in-house service, located in job centers, independent and confidential, might help. Partnering with job centers, primary healthcare providers, mutual help or peer support groups or information centers for unemployed individuals will also be needed to improve the outreach and delivery of such interventions. Another avenue is to approach unemployed people with mental health problems in the somatic and mental health service systems.

Our hypothesis about the program’s efficacy was not supported for primary outcomes, possibly because job search and help-seeking were too distal outcomes for a four-session intervention that focused on acceptance, values and disclosure decisions in a study among long-term unemployed participants. This highlights the need for early intervention, soon after the onset of mental health problems and unemployment. However, and encouragingly, this very brief intervention yielded significant benefits in terms of symptom reduction and increased recovery. Trend-level findings further suggest small-to-medium size positive effects in terms of reactions to stigma and discrimination that are associated with poor long-term outcomes, such as self-stigma and secrecy. The intervention might therefore, although yet without measurable short-term impact on employment outcomes, reduce barriers to the labor market and increase employability of participants over time. Consistent with this assumption, in a recent Swedish study, acceptance and commitment therapy appeared to increase employability among people with mental illness or chronic pain (Berglund et al., 2018). Future studies should also investigate the efficacy of an intervention that combines group programs such as ours with supported employment (individual placement and support), equaling a type of complex intervention often referred to as augmented supported employment (Reissner et al., 2016; Suijkerbuijk et al., 2017).

Limitations of our study need to be considered. The sample size was small, therefore, small-to-medium effects remained at a trend-level significance and negative findings need to be interpreted cautiously. Due to selection bias, our sample is likely not representative. Some scales had low internal consistency. Not all participants may necessarily have thought about themselves as having a mental illness, and therefore, parts of the group program that dealt with disclosure and secrecy may not have been relevant to them.

Conclusion

Despite these limitations, this pilot RCT provides preliminary evidence for the efficacy of a brief peer-led group program to improve symptoms and recovery among long-term unemployed individuals with mental health problems. Any future study will need to consider recruitment issues in this target population. Future interventions might need to use longer and more comprehensive interventions to achieve effects on help-seeking and employment.

Footnotes

Acknowledgements

The authors are grateful to all participants and to the German Federal Employment Agency and its Medical Service as well as to joint local agencies and approved local providers for their support with participant recruitment.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The study was funded by the German Research Foundation (DFG: RU 1200/3-1).

ORCID iD

Nicolas Rüsch

References

Andresen

Caputi

Oades

L. G.

(2010). Do clinical outcome measures assess consumer-defined recovery? Psychiatry Research, 177, 309–317.

Audhoe

S. S.

Nieuwenhuijsen

Hoving

J. L.

Sluiter

J. K.

Frings-Dresen

M. H. W.

(2018). Perspectives of unemployed workers with mental health problems: Barriers to and solutions for return to work. Disability and Rehabilitation, 40, 28–34.

Berglund

Anderzén

Andersén

Å.

Carlsson

Gustavsson

Wallman

Lytsy

(2018). Multidisciplinary intervention and Acceptance and Commitment Therapy for return-to-work and increased employability among patients with mental illness and/or chronic pain: A randomized controlled trial. International Journal of Environmental Research and Public Health, 15, Article 2424.

Brand

J. E.

(2015). The far-reaching impact of job loss and unemployment. Annual Review of Sociology, 41, 359–375.

Corrigan

P. W.

Michaels

P. J.

Vega

Gause

Watson

A. C.

Rüsch

(2012). Self-Stigma of Mental Illness Scale-Short Form: Reliability and validity. Psychiatry Research, 199, 65–69.

Hagen

Bänfer

Werkstetter

Hebebrand

Reissner

(2018). Inanspruchnahme des psychosozialen Hilfesystems bei jungen, psychisch erkrankten Arbeitslosengeld-II-Empfängern vor und nach Nutzung eines psychiatrisch-psychologischen Beratungsangebotes [Use of mental health service among young adults on unemployment benefit before and after receiving counseling at a psychiatric liaison department]. Der Nervenarzt, 80, 598–605.

Hautzinger

Bailer

(1993). Allgemeine Depressionsskala. Weinheim, Germany: Beltz.

Kessler

R. C.

Green

J. G.

Gruber

M. J.

Sampson

N. A.

Bromet

Cuitan

… Zaslavsky

A. M.

(2010). Screening for serious mental illness in the general population with the K6 screening scale: Results from the WHO World Mental Health (WMH) Survey initiative. International Journal of Methods in Psychiatric Research, 19(Suppl. 1), 4–22.

Link

B. G.

Mirotznik

Cullen

F. T.

(1991). The effectiveness of stigma coping orientations: Can negative consequences of mental illness labeling be avoided? Journal of Health and Social Behavior, 32, 302–320.

10.

Modini

Tan

Brinchmann

Wang

M. J.

Killackey

Glozier

… Harvey

S. B.

(2016). Supported employment for people with severe mental illness: Systematic review and meta-analysis of the international evidence. The British Journal of Psychiatry, 209, 14–22.

11.

Oschmiansky

Popp

Riedel-Heller

Schwarzbach

Gühne

Kupka

(2017). Psychisch Kranke im SGB II: Situation und Betreuung (IAB-Forschungsbericht, 14/2017) [IAB research report – Mentally ill people in the German social code II: Situation and care]. Nürnberg, Germany: Institut für Arbeitsmarkt- und Berufsforschung.

12.

Reissner

Scherbaum

Wiltfang

Kis

Meiler

Lieb

… Hebebrand

(2016). Psychiatrische Interventionen für Arbeitslose [Psychiatric interventions for the unemployed]. Der Nervenarzt, 87, 74–81.

13.

Rüsch

Corrigan

P. W.

Waldmann

Staiger

Bahemann

Oexle

… Becker

(2018). Attitudes toward disclosing a mental health problem and reemployment: A longitudinal study. The Journal of Nervous and Mental Disease, 206, 383–385.

14.

Rüsch

Malzer

Oexle

Waldmann

Staiger

Bahemann

… Corrigan

P. W.

(2019). Disclosure and quality of life among unemployed individuals with mental health problems: A longitudinal study. The Journal of Nervous and Mental Disease, 207, 137–139.

15.

Santos

Lopes

Virgolino

Stefanovska-Petkovska

Dinis

Ambrósio

Heitor

M. J.

(2018). Defining a brief intervention for the promotion of psychological well-being among unemployed individuals through expert consensus. Frontiers in Psychiatry, 9, Article 13.

16.

Schubert

Parthier

Kupka

Ulrich

Jörg

Philipp

(2013). Menschen mit psychischen Störungen im SGB II (IAB-Forschungsbericht, 12/2013) [IAB research report – Mentally ill people in the German social code II]. Nürnberg, Germany: Institut für Arbeitsmarkt- und Berufsforschung.

17.

Staiger

Waldmann

Rüsch

Krumm

(2017). Barriers and facilitators of help-seeking among unemployed persons with mental health problems: A qualitative study. BMC Health Services Research, 17(1), Article 39.

18.

Suijkerbuijk

Y. B.

Schaafsma

F. G.

van Mechelen

J. C.

Ojajärvi

Corbière

Anema

J. R.

(2017). Interventions for obtaining and maintaining employment in adults with severe mental illness, a network meta-analysis. Cochrane Database of Systematic Reviews, 9, CD011867.

19.

van Ryn

Vinokur

A. D

. (1992). How did it work? An examination of the mechanisms through which an intervention for the unemployed promoted job-search behavior. American Journal of Community Psychology, 20, 577–597.

20.

Wilson

C. J.

Deane

F. P.

Ciarocchi

Rickwood

(2005). Measuring help-seeking intentions: Properties of the general help-seeking questionnaire. Canadian Journal of Counselling, 39, 15–28.

21.

Yip

P. S. F.

Cheung

Y. B.

(2006). Quick assessment of hopelessness: A cross-sectional study. Health and Quality of Life Outcomes, 4, Article 13.

22.

Zhang

C.-Q.

Leeming

Smith

Chung

P. K.

Hagger

M. S.

Hayes

S. C.

(2017). Acceptance and Commitment Therapy for health behavior change: A contextually-driven approach. Frontiers in Psychology, 8, Article 2350.