STITCHING THE SOCIAL FABRIC OF SOCIETY The effectiveness of a social network intervention in forensic psychiatric care Lise Swinkels
STITCHING THE SOCIAL FABRIC OF SOCIETY The effectiveness of a social network intervention in forensic psychiatric care Lise Swinkels
The social fabric of society In the late 18th century, the metaphor “the social fabric” was first used in England in response to the industrial development of the machine loom, which raised concerns that citizens would become isolated as more people moved from villages to cities to work in factories (Lepore, 2021). To date, the metaphor has been used by various disciplines, including researchers (e.g., sociologists and psychologists), policy makers, and artists, to describe the strength of connectedness between groups in the society – social cohesion, individuals, institutions, systems, and in infrastructure (Bowen, 2015; Chang & Rubin, 2020; Kawachi & Berkman, 2000). In this dissertation, stitching the social fabric of society refers to community volunteers – “threads” – reaching out to lonely, vulnerable, or disadvantaged people – “loose threads” – with the intention of developing and strengthening reciprocal and supportive relationships in the society.
SPANNINGSVELD IK ZIT KLEM TUSSEN TWEE MACHTSSTRUCTUREN DE EEN OPEREERT VAN BOVEN NAAR BENEDEN DE ANDER VAN BENEDEN NAAR BOVEN IK VERZET MIJ MET HAND EN TAND Yahya Hassan Gedichten 2 (2020)
COLOPHON Stitching the social fabric of society: The effectiveness of a social network intervention in forensic psychiatric care Lise T. A. Swinkels ISBN: 978-94-6473-178-1 Cover and artwork by Gemma Haggenburg Layout and design by Wendy Bour Printed by Ipskamp Printing This research was funded by Stichting tot Steun VCVGZ and Inforsa Forensic Mental Healthcare. © 2023 L. T. A. Swinkels All rights reserved. No parts of this thesis may be reproduced, stored, or transmitted in any way or by any means without permission of the author.
VRIJE UNIVERSITEIT STITCHING THE SOCIAL FABRIC OF SOCIETY The effectiveness of a social network intervention in forensic psychiatric care ACADEMISCH PROEFSCHRIFT ter verkrijging van de graad Doctor aan de Vrije Universiteit Amsterdam, op gezag van de rector magnificus prof.dr. J.J.G. Geurts, in het openbaar te verdedigen ten overstaan van de promotiecommissie van de Faculteit der Geneeskunde op woensdag 15 november 2023 om 15.45 uur in een bijeenkomst van de universiteit, De Boelelaan 1105 door Lise Tilla Anne Swinkels geboren te Helmond
promotoren: prof.dr. A. Popma prof.dr. J.J.M. Dekker copromotor: dr. T.M. van der Pol promotiecommissie: prof.dr. T.G. van Tilburg prof.dr. E.D.M. Masthoff dr. M. Sentse dr. B.J. Verschuere prof.dr. B.G.M. Völker
CONTENTS 8 Preface Chapter 1 General introduction 11 25 Chapter 2 The effectiveness of social network interventions for psychiatric patients: A systematic review and meta-analysis 65 Chapter 3 Improving mental wellbeing of forensic psychiatric outpatients through the addition of an informal social network intervention to treatment as usual: A randomized controlled trial 93 Chapter 4 Patients’ and volunteer coaches’ experiences with an informal social network intervention in forensic psychiatric care: A qualitative analysis 127 Chapter 5 The effectiveness of an additive informal social network intervention for forensic psychiatric outpatients: Results of a randomized controlled trial 163 Chapter 6 The impact of COVID-19 restrictions on social relationships of forensic psychiatric outpatients with preexisting social network-related problems: A mixed methods study 181 Chapter 7 Summary and General discussion Appendices References 210 Supplementary material 236 Appendix A 237 Appendix B 295 Appendix C 299 Nederlandse samenvatting 308 Dankwoord 322 Publications 334 Curriculum vitae 338
8 PREFACE In 2012, the year of the Dutch national political elections, and also the year that I was about to finish my master’s degree in clinical forensic psychology, I remember being triggered by a slogan on a campaign poster of a large political party displayed alongside the highway that read: “Empathy for victims, not offenders”. Moreover, to quote the campaign agenda: “Police and the justice system focus on suspects and offenders and stand firm for the victims of crime. The [name party] does not accept that the good suffer from the bad.”. I perceived this as a polarizing message – victims are important and should be taken care of, whereas offenders, on the other hand, are evil who deserve harsh punishments and imprisonment? I wondered whether this slogan spoke to an underlying stigma in society towards people who committed criminal offenses, which I also encountered in my own educated and privileged social network. Why are you helping people who commit crimes? Is it safe for you to be alone with them? You must have heard all of these crazy stories? Although nobody could disagree with the first part of the slogan: “Empathy for victims”, I (obviously) highly disagree with the second part, as a practitioner in forensic psychiatric care. The distinction between people who are victims and people who commit offenses is often difficult to make. Naturally, not all criminal offenders are also victimized in the past and require care due to psychiatric problems related to their criminal behavior. Yet, research shows that a significant proportion of people who committed offenses have experiences of victimization and are vulnerable due to psychiatric problems (Appelman, Dirkzwager, & van der Laan, 2021; Dirkzwager, Verheij, Nieuwbeerta, & Groenewegen, 2021; Dienst Justitiële Inrichtingen, 2022b; Fazel, Hayes, Bartellas, Clerici, & Trestman, 2016; van Buitenen, van den Berg, Meijers, & Harte, 2020). Moreover, given the psychiatric problems as well as the complexity and diversity of problems (i.e., comorbidity and socioeconomic problems) that are often related to criminal behavior in people who committed offenses – forensic psychiatric patients, effective care for this group is highly warranted. Furthermore, there is an urgent need to further improve treatment effectiveness for forensic psychiatric patients to enhance mental health and reduce criminal recidivism (MacInnes & Masino, 2019; MacKenzie & Farrington, 2015; McIntosh, Janes, O’Rourke, & Thomson, 2021; Völlm et al., 2018). Therefore, practice-oriented research questions arose: How can we best help vulnerable people who are suffering from psychiatric problems not to commit offenses? How can we further improve forensic psychiatric care? Specifically, can empathy and support from people in the society help to stop reoffending?
9 In short, this research project was born out of personal experiences, a reaction to attitudes in media and people around me on forensic psychiatric care, the stigmatization of people with risky or criminal behavior in the society, and the difficulties that forensic psychiatric patients can encounter reintegrating in society. In the years that followed its inception in 2014, our research team explored a way to improve the connection between people who are actively participating in society and forensic psychiatric patients to enhance patients’ mental wellbeing. The start of a research project with an idealistic intention.
1 Chapter 1- General introduction 1 General introduction
12 Chapter 1 Forensic psychiatric care People who commit crimes and have experiences of victimization, psychiatric, or socioeconomic problems related to their risk of criminal behavior – forensic psychiatric patients – make up a large portion of the prison population (Dirkzwager et al., 2021; Fazel et al., 2016). A recent study found that 30% to 56% of the youth in prisons in the Netherlands reported emotional or physical neglect and 10% to 26% have experienced physical abuse in the past, which is substantially higher (38%) among youth in juvenile correctional institutions (Appelman et al., 2021). As for adult prisoners, most research on victimization has been conducted in females, showing that 30% of the females experienced physical abuse in childhood and 25% experienced sexual abuse. Furthermore, previous research demonstrated that a large part of the prisoners in the Netherlands were likely to have experienced socioeconomic and (mental) health disadvantages prior to imprisonment (Dirkzwager et al., 2021). For example, compared to a control group, prisoners were found to be twice as likely to have social problems (i.e., migration background, low income, no paid job, and no romantic relationship) and were 3.5 times as likely to suffer from psychological problems. In the Netherlands, about 60% of the prisoners are diagnosed with a psychiatric disorder (Dienst Justitiële Inrichtingen, 2022b). The mental health problems are also complex, as almost 60% is diagnosed with one or more comorbid psychiatric disorder (van Buitenen et al., 2020). Moreover, extensive research showed that prisoners worldwide are more likely to suffer from mental health problems (Fazel et al., 2016). Depending on the severity of the risk of recidivism and the complexity of mental healthcare needs, forensic psychiatric patients can be referred to either inpatient or outpatient forensic mental healthcare services. In the Netherlands, most referrals of offenders to forensic psychiatric treatment are imposed by judges, for example, as a part of their sentence or as a measure. Additionally, although less often, patients are being referred by general practitioners. In 2021, about 2500 individuals in the Netherlands received forensic inpatient care and 12.078 people received outpatient care (Dienst Justitiële Inrichtingen, 2022a). The overarching goal of forensic psychiatric treatment is twofold: (1) reducing the risk of criminal recidivism and (2) enhancing mental health and wellbeing of an individual. Forensic treatment is therefore personalized and multimodal, aimed at addressing both complex criminogenic and mental health needs of patients (Nicholls et al., 2022; Völlm et al., 2018). There are two dominant treatment models used to frame forensic mental healthcare. The risk need responsivity (RNR) model, applied since 1990, is the most widely
13 General introduction 1 used model in clinical practice for the assessment of risk factors related to criminal offending and the implementation of treatment (Andrews & Bonta, 2010; Andrews, Bonta, & Hoge, 1990a). Previous research demonstrated that treatment programs adhering to the RNRmodel aremost effective, which implies that (1) the treatment intensity is matched to the individuals’ risk of criminal recidivism (i.e., risk principle), (2) the criminogenic needs are assessed and addressed in treatment (i.e., need principle), and (3) the individuals’ ability to learn from treatment is maximized by tailoring the intervention to an individuals’ motivation, abilities, and strengths (i.e., responsivity principle) (Andrews & Bonta, 2010; Andrews et al., 1990a). The good lives model (GLM), a strength-based rehabilitation approach, is proposed as a complement to the RNR model (Barnao, 2022; Nicholls et al., 2022; Ward & Brown, 2004). The GLM model emphasizes the promotion of a meaningful life by providing patients with tools, both skills and resources, they need to attain personal goals related to wellbeing and criminal behavior – primary goods – in a socially acceptable way (Barnao, Ward, & Robertson, 2016). In clinical practice, patients’ personal goals should be explored to understand what motivates them to commit crimes and to determine what tools are needed to achieve these goals without harming others (Barnao, 2022). A focus on patients’ meaningful personal goals and strengths in addition to managing their risks in treatment, potentially enhances treatment motivation and adherence of forensic patients, resulting in enhanced wellbeing and a reduced risk of criminal recidivism (Barnao, 2022; Willis & Ward, 2013). The story of Johan presented below illustrates the multiple socioeconomic, criminogenic, and mental health needs that are often addressed in forensic psychiatric outpatient care. Furthermore, the story highlights several challenges that Johan experienced during treatment. Johan (41 years) identifies himself as a male from Amsterdam because he lived there his entire life, except for the months and years spent in institutions and prisons. Between his 12th and 18th year, he was transferred from home to a closed facility and stayed in juvenile detention. He has been going in and out prison since the age of 16. For the last 1.5 years, Johan received mandatory treatment at an outpatient care facility in Amsterdam. In the first 6 months, he only occasionally showed up for appointments. He had spoken with many psychologists and psychiatrists in the past and didn’t feel like any of the conversations had helped him. He remembers being told that he has an antisocial personality disorder, a traumatized childhood, and problems with substance use. He didn’t want to repeat his
14 Chapter 1 story again. After receiving a warning from his parole officer, Johan started attending appointments with the social worker who helped himwith the application for social benefits and found a place for him at a supported housing facility. Perhaps clinicians could be of some help after all, he thought. Johan then participated in cognitive behavioral therapy focused on limiting alcohol use and cannabis use. He learned to control his alcohol use and minimize triggers. For example, not spending so much time with the people he has been hanging out with all his life and not drinking more than three alcoholic beers at soccer games of Ajax, which has led to public urination, fines, and aggressive confrontations with police many times in the past. Furthermore, Johan is exploring pharmacotherapy with the psychiatrist, which could help him reduce the agitation he has experienced since childhood and help him arrive on time at appointments with important agencies and at vocational activities. Right now, it sometimes feels like he is drowning in responsibilities. Therefore, cannabis helps him to relax. In terms of criminal behavior, things are going well. However, Johan struggles to find his place in society and lately feels rather passive and down. He does not have much money to spend now that he is paying off his debts and is surrounded by people who are bothering him with their problems at the supported housing facility where he lives. This annoys him and sometimes he loses his temper towards the other residents. He isolates himself in his room in an effort to avoid conflicts and warnings. Being alone, his thoughts go in all directions. He does not want to bother family, as they have been through enough with him and all are busy. There are not much fun things for him to do now he has less money and is avoiding the people he used to hang out with. It makes him think more about the opportunities to earn a lot of money. He wonders if he could use that money to go out more and live more independently. Johan’s story shows an initial aversion towards forensic care. However, after ongoing efforts by clinicians to schedule appointments and the provision of practical care focused on important basic needs (i.e., income and housing), trust was built and a working alliance emerged. This illustrates that treatment was continuously tailored to the individual’s motivation and abilities (i.e., responsivity principle). Multiple (overlapping) criminogenic needs and mental health needs were addressed in treatment, such as substance use and agitation, which contributed to desistance in criminal behavior. However, at this point in treatment, Johan has difficulty adjusting to his new life (i.e., developing new social relationships and finding meaningful activities) and appears unhappy and lonely, ultimately increasing the risk of criminal recidivism. For Johan, self-agency, recreational activities and/or work, inner peace, social relationships, community and belonging, and happiness, are potential primary-good-domains that could be addressed in the treatment to further enhance treatment outcomes.
15 General introduction 1 Although some studies showed that forensic psychiatric care that complies with the RNR principles is most effective and is preferable to imprisonment in terms of treatment outcomes, evidence of effectiveness is still limited (Andrews et al., 1990b; Völlm et al., 2018). Furthermore, a meta-analysis found that significant treatment effects on criminal recidivism favoring the experimental conditions occurred in only a minority of forensic interventions (Wartna, Alberda, & Verweij, 2013a). Therefore, there is an urgent need to improve treatment effectiveness and develop evidence-based interventions for forensic psychiatric patients with complex needs (MacInnes & Masino, 2019; MacKenzie & Farrington, 2015; McIntosh et al., 2021; Völlm et al., 2018). In this dissertation, we focus on improving mental wellbeing of forensic psychiatric outpatients with an intervention aimed at strengthening social networks – one of the primary goods – in addition to forensic psychiatric care. Social networks of forensic psychiatric patients In general, the research on social networks is both extensive and complex, because it includes many aspects related to structural characteristics (e.g., size, roles or composition, density, quality, reciprocity) and functional characteristics (e.g., providing resources, positive and negative influences, companionship, and social support) (Heaney & Israel, 2008; House, 1987; Kawachi & Berkman, 2001). Regarding the social network size of forensic populations, a previous study found an average number of 5.9 social network members and 2.9 members in the core social network, consisting of people with whom probationers with mental and substance use problems spend most of their time (Skeem, Eno Louden, Manchak, Vidal, &Haddad, 2009). However, a study among forensic inpatients found an average number of 15 members with a large dispersion (SD = 7.6) (ter Haar-Pomp, Spreen, Bogaerts, & Völker, 2015). Therefore, the size of the network may vary across populations as well as on an individual level, and also depends on the type of measurement that was used. Despite this, previous studies demonstrated smaller informal social networks (i.e., personal network members, including family, friends, romantic partners, and peers) in forensic patients compared to general and other psychiatric populations (Skeem et al., 2009; ter Haar-Pomp et al., 2015; Wang et al., 2017). Furthermore, a decrease in the personal network size of forensic patients during inpatient treatment was shown (ter Haar-Pomp, Spreen, Völker, & Bogaerts, 2019). A study examining the core discussion network (i.e., the network consisting of people with whom a person discusses important issues) of prisoners revealed that although the network size remained stable throughout imprisonment, 60% of the network composition had changed after imprisonment (Völker et al., 2016).
16 Chapter 1 It was suggested that family relationships remained after a prison sentence and that friendships were more likely to disappear or to be replaced. Moreover, other studies found that the majority of the social networks of forensic populations consisted of family members who were likely to be victims in offenses by forensic patients (Estroff, Zimmer, Lachicotte, & Benoit, 1994; Ter Haar-Pomp et al., 2015). In addition, social networks predominantly consisted of formal network members, such as professionals, probation officers, and clinicians, as well as informal members with criminal behaviors and/or attitudes, psychiatric problems, and substance abuse (Skeem et al., 2009; ter Haar-Pomp et al., 2015). Johan used to hang out a lot with friends with whom he drank alcohol and committed crimes. After receiving forensic outpatient treatment, his social network (i.e., the people he trusts and enjoys spending time with) consisted of his younger sister (36 years), his sister's two children (10 and 8 years), his mother (62 years), an uncle (69 years), an old friend from elementary school (male, 40 years), and his social worker and psychologist. He only meets or talks to his mother who also lives in Amsterdam, social worker, and psychologist regularly. His friend he has not seen for over a year. Johan expressed an overall dissatisfaction with his personal social relationships. He would like to meet with people to do something fun. However, he also stressed that he prefers to be alone and that it is better for him to avoid people. The summary of Johan’s social network outlines several aspects of social network characteristics (i.e., network size, contact frequency, quality of relationships, and social network-related needs) (Heaney & Israel, 2008). However, many aspects of the social network remain unclear, such as whether social network members live nearby, the quality of the relationship between different social network members, and whether or not the social relationships provide positive social support (e.g., emotional, informational, instrumental, and appraisal support) (House, 1981; Langford, Bowsher, Maloney, & Lillis, 1997). Moreover, the negative social support (e.g., encouragement of drinking and criminal activities) remains unknown. Although Johan names eight people in his social network, including two children and two professionals who are important to him, he hardly meets anyone other than his mother and professional caregivers. He is dissatisfied, but his social network-related needs seem ambivalent. Stigma Developing and maintaining social relationships in society and actively participating in society can be challenging after imprisonment, long inpatient treatment, and being labeled
17 General introduction 1 as a forensic psychiatric patient (Mezey, Youngman, Kretzschmar, & White, 2016; ter Haar-Pomp et al., 2019; Völker et al., 2016; Willis, 2018). Moreover, portraying this group as insane and dangerous in the media contributes to stigmatizing thoughts and beliefs about people with psychiatric disorders and criminal behavior (West, Yanos, & Mulay, 2014). Bearing in mind the consequences of stigma, on the one hand, prejudice and negative beliefs of people with a migration background, psychiatric problems, and criminal behavior (i.e., triple stigma) might lead to avoidant attitudes of social networks in society, for example neighbors and employers, towards patients (West et al., 2014). On the other hand, patients themselves may feel unfairly treated, have feelings of distrust, fear rejection or harm by others, or lack the social skills to develop and maintain social relationships, thus avoiding connections with others and participation in society (Link, Struening, Neese-Todd, Asmussen, & Phelan, 2001; Mezey et al., 2016; Thornicroft, Brohan, Rose, Sartorius, & Leese, 2009). Social network and mental health For many decades, the protective effects of social networks on mental health and health behavior has been universally emphasized in various theoretical models (Cohen & Wills, 1985; Glanz, Rimer, & Viswanath, 2008). Two dominant models – the main effect model and stress buffering model – are mutually used to explain the relationship between supportive social networks and mental health (Cohen, 2004; Cohen &Wills, 1985; Kawachi & Berkman, 2001). In the main effect model, several pathways of social networks (i.e., social influence, supporting resources, positive psychological states) influencing a person’s health promoting behaviors (e.g., treatment compliance, self-care, and avoiding risky situations) are identified, which directly affect (mental) health (Cohen, 2004). The stress buffering model describes several pathways in which social support influences a person’s response to stressful events, preventing responses that could compromise (mental) health (Cohen, 2004). In short, first, the perceived availability of social support (i.e., the belief that others will provide support) following a stressful event could help to reframe a person’s perception of the situation as potentially harmful and could increase a person’s perceived coping abilities, preventing stress responses. Second, perceived or received social support could diminish and prevent cognitive, emotional, behavioral, and physiological responses following a stressful event, which in turn effects (mental) health. Furthermore, the social network is a multifaceted factor, including beforementioned structural and functional factors influencing mental health (Cohen & Wills, 1985). A more recent study therefore provided a comprehensive overview of seven theoretical mechanisms based on previous research through which social networks and social support
18 Chapter 1 could influence (mental) health: (1) social influence or comparison (i.e., a person’s health behaviors are influenced positively or negatively by their network members through observation and comparison), (2) social control (i.e., a person’s health behaviors are positively influenced by formal and informal network members through active encouragement and persuasion, but can be negatively influenced if social control is perceived as invasive), (3) role-based purpose and meaning (i.e., a person’s health behaviors are influenced by their commitment and obligations towards network members as well as by their perceived value or meaning to others), (4) self-esteem (i.e., a person’s positive evaluation of role fulfillment will increase self-esteem, which positively influences mental health), (5) sense of control (i.e., a person’s positive role fulfillment will increase a sense of control over one’s life, which positively influences mental health), (6) belonging and companionship (i.e., connections to and participation in activities with informal and formal network members lead to feelings of belonging and companionship, producing positive affect, which enhances mental health), and (7) perceived support availability (i.e., emotional, informational, and instrumental support from informal and formal network members directly influences mental health as well as indirectly influences mental health through self-esteem, and a sense of control and meaning to others) (Thoits, 2011). It was suggested that these mechanisms can be used to explain both positive and negative effects of social network factors on mental health outcomes. Furthermore, it was discussed that many of these mechanisms can be used to explain how social support impacts a person’s response to stressful events. Both positive and negative effects of social network factors on mental health outcomes have extensively been demonstrated in empirical research (Cohen & Wills, 1985; Harandi, Taghinasab, & Nayeri, 2017; Heaney & Israel, 2008; Kawachi & Berkman, 2001; Schön, Denhov, & Topor, 2009). However, the relationship between social network factors and mental health is considered complex, as the evidence varies across populations and life stages (Kawachi & Berkman, 2001). Moreover, effects vary across different types of social network factors (i.e., structural and functional characteristics). For example, a recent study showed that higher functional social support (i.e., the perceived availability of network members providing support), in particular, was related to decreased depressive symptoms in a general male population (Almquist, Landstedt, & Hammarström, 2017). In contrast, a lack of social support as well as smaller social networks could negatively influence mental health outcomes. Negative effects on mental wellbeing among males were found in those experiencing a lack of trustworthy members and demanding members in their social network (Antonucci,
19 General introduction 1 Akiyama, & Lansford, 1998). Furthermore, lacking support and smaller social networks were associated with increased mental health service use (Albert, Becker, Mccrone, & Thornicroft, 1998). Noteworthy, negative effects of social networks on mental wellbeing have been reported, particularly among females with a higher number of close social relationships, females experiencing pressure from network members, and females with obligations to provide social support (Antonucci et al., 1998; Kawachi & Berkman, 2001). Despite these various effects of social network factors, the positive relationship between supportive social networks and mental health is generally accepted (Kawachi & Berkman, 2001; Thoits, 2011). Social network and criminal recidivism As for mental health, the potential influence of social networks on the development and persistence of criminal behavior has been emphasized in several theoretical models from the field of criminology. The differential association theory considers criminal behavior as a social learning process (Sutherland, 1949). A person will develop criminal behavior in interaction with other criminal network members, depending on the frequency, duration, and intensity of the contact. The more exposure to certain behaviors, the more a person will copy and internalize those behaviors. The social control theory proposes that strong connections with social networks and society will discourage people from engaging in criminal behavior (Hirschi, 1969). The social network is understood as source of control. For example, network members can reinforce and disapprove of a person’s behavior, thereby constraining risk behaviors. Besides, a person with strong connections with network members and with the society risks losing these connections after engaging in criminal behavior, which could discourage this behavior. Following up on this, the social support theory assumes that higher social support levels decrease the risk of criminal behavior (Cullen, 1994). Moreover, the exchange of social support enables strong connections between network members, leading to social control. Social networks are both a risk and protective factors for criminal recidivism, as supported by empirical research. On the one hand, it was found that social networks encouraging criminal behavior, friends and family members with criminal behaviors, problematic family relationships, and a lack of social support predicted criminal recidivism in forensic (psychiatric) populations (Bonta, Blais, & Wilson, 2014; Eisenberg et al., 2019; Gendreau, Little, & Goggin, 1996). On the other hand, strong family relationships, strong relationships with prosocial adults, connections with friends in the community, a stable romantic relationship, and higher levels of social support predicted desistance from criminal behavior (Barrick et al., 2014; Berg & Huebner, 2011; Borowsky, Hogan, & Ireland, 1997;
20 Chapter 1 Laub, Nagin, & Sampson, 1998; Lodewijks, de Ruiter, & Doreleijers, 2010). One study among a forensic population with mental and substance use problems showed that large networks consisting of network members with prosocial attitudes predicted decreased probation violations and better treatment compliance (Skeem et al., 2009). Furthermore, good relationships with professionals and core network members, particularly with clinicians working with shared decision-making, decreased violations. Social network interventions Since the mid-nineties, various types of interventions have been developed to enhance informal supportive social networks (Kawachi & Berkman, 2001). A review distinguished three overarching approaches of social network interventions for psychiatric populations with different social network needs: (1) developing new social connections for patients with small networks, with few connections between different network members and network groups (e.g., family, friends, and colleagues), and with less access to various support groups (e.g., family, friends, and professionals), (2) maintaining and enhancing existing social connections for patients with sufficient network members but lacking social support or skills to develop and maintain supportive connections, and (3) enhancing family connections for patients with family members who are overburdened, family conflicts, or with unfortunate family interaction patterns (Biegel, Tracy, & Corvo, 1994). In line with these approaches, different types of interventions are described, such as social skills training, support groups, self-help groups, mutual-help groups, and volunteer-linking (Biegel et al., 1994; Perese & Wolf, 2005). Multiple reviews demonstrated modest effects of social network interventions on (mental) health and behavioral outcomes, including specific reviews of support group and volunteer-linking interventions (Anderson, Laxhman, & Priebe, 2015; Hunter et al., 2019; Latkin & Knowlton, 2015; Mead, Lester, Chew-Graham, Gask, & Bower, 2010; Siette, Cassidy, & Priebe, 2017). However, the need for additional empirical research regarding social network interventions for psychiatric patients has been repeatedly expressed to develop evidence-based interventions and determine which patients might benefit from these interventions. Moreover, studies examining the effectiveness of social network interventions in forensic psychiatric populations are warranted (Pettus-Davis, Howard, Roberts-Lewis, & Scheyett, 2011). As described in previous literature, volunteer-linking intervention aimed at developing new social connections could be indicated for patients with small networks and a lack of supportive network members or support groups (Biegel et al., 1994; Heaney & Israel, 2008). Given the evidence that informal social networks of forensic patients can become
21 General introduction 1 more fragile after institutionalization and imprisonment, and that the majority of the network consists of eithermemberswithapotential negative influenceand/or professionals (Skeem et al., 2009; ter Haar-Pomp et al., 2015), we hypothesized that volunteer-linking interventions would be promising for improving mental health outcomes in forensic patients. Therefore, the main aim of the current research project was to develop and examine the effectiveness of an informal social network intervention, based on volunteer-linking, to enhance mental wellbeing in forensic psychiatric patients. Forensic network coaching At Inforsa Forensic Mental Healthcare, a department of Arkin Mental Health Institute that provides forensic psychiatric care for youth and (young) adults in Amsterdam, the Netherlands, we sought collaboration with De Regenboog Groep [The Rainbow Group], an informal care institute providing volunteer services for people with social or mental challenges. We explored whether volunteers living in the community of Amsterdam could complement formal care provided by Inforsa and “weave threads” between reintegrating forensic patients or “loose threads” and people in the society. Hence the running title of this dissertation: Stitching the social fabric of society. With the academic supervision of researchers from the Department of Child and Adolescent Psychiatry & Psychosocial Care of the Amsterdam UMC and the Department of Research and Quality of Care of Arkin, both affiliated with the Vrije Universiteit Amsterdam, it was possible to conduct the research in clinical practice. In short, this research is a result of a collaboration between staff from a mental healthcare institute, an informal care institute, and research institutes. Furthermore, this research was financed by De Stichting tot Steun VCVGZ. The informal social network intervention, entitled forensic network coaching (FNC), was developed by De Regenboog Groep in collaboration with Inforsa. The intervention is provided by trained volunteer coaches recruited from the community, who are matched to participants based on mutual personal preferences. FNC is primarily focusing on the development of a supportive and nonprofessional relationship between volunteer coaches and participants – participants have the opportunity to develop a new connection with a person in the community. Secondary goals of FNC are: (1) enhancing the size of the informal social network and quality of relationships with informal network members, (2) enhancing social support, and (3) enhancing social participation in the community. Training of volunteer coaches focuses on knowledge of basic coaching skills, their expectations, attitudes, and commitment, as well as on how to provide or tailor the intervention to personal needs of forensic psychiatric patients. Furthermore, volunteer
22 Chapter 1 coaches received a brief training on the structured intervention Natuurlijk, een netwerkcoach! [Of course, a network coach!], which is based on the TO GROW (goal reality options will) model of coaching and on solution focused (brief) therapy (SFT) (De Shazer et al., 2021; Mezzo, 2015; Whitmore, 2017). This structured intervention is presented as a theoretical guideline for coaches to use if necessary. In addition, throughout the intervention period, supervision for volunteer coaches is organized by De Regenboog Groep. Aim and outline of this dissertation The aim of this research project was to examine whether an additive informal social network intervention could improve treatment outcomes among forensic psychiatric outpatients, to contribute to the development of evidence-based interventions warranted for this population with complex needs, and to guide clinical practice and policy. First, a systematic review and meta-analysis was conducted to gain insight in the existing literature on the effectiveness of social network interventions on treatment outcomes among psychiatric patients. Second, to examine effectiveness of the additive informal social network intervention for forensic psychiatric outpatients, we designed and conducted a randomized controlled trial in clinical practice. Both quantitative and qualitative analytic methods were used to explore effectiveness and to provide an in-depth understanding of the effects, respectively. Lastly, given the extraordinary situation due to the COVID-19 pandemic that crossed our research project, an additional mixed methods study was conducted to explore the effects of the restrictions on social relationships of forensic outpatients participating in the clinical trial. The studies that were conducted are presented in the following chapters of this dissertation. Chapter 2 presents the findings regarding the effectiveness of social network interventions on social network outcomes and other treatment outcomes in psychiatric populations. Furthermore, the potential moderators of treatment effectiveness, such as intervention and patient characteristics, were explored. We hypothesized that social network and mental health outcomes would improve more in psychiatric patients receiving social network interventions compared to treatment as usual in mental healthcare settings. Chapter 3 outlines the study protocol of the randomized controlled trial, detailing the methods, such as the additive informal social network intervention – forensic network coaching, patient population, and procedures. We hypothesized that the addition of an informal social network intervention to treatment as usual would improve effectiveness
23 General introduction 1 on mental wellbeing and other relevant treatment outcomes compared to treatment as usual alone in forensic psychiatric outpatients. Chapter 4 presents patients’ and volunteer coaches’ experiences with the additive informal social network intervention. Qualitative methods were used to analyze experiences and perceived barriers and facilitating factors that influenced compliance of patients and volunteer coaches who participated in the randomized controlled trial. Chapter 5 presents the quantitative findings of the randomized controlled trial – the effectiveness of the additive informal social network interventions on mental wellbeing and other treatment outcomes in forensic psychiatric outpatients. In addition, the effects among patients with different levels of compliance were explored, as well as potential moderating patient characteristics. Chapter 6 focuses on the impact of the COVID-19 restrictions on social relationships of forensic psychiatric outpatients participating in the randomized controlled trial. We hypothesized that social relationships would further deteriorate due to the COVID-19 restrictions. In addition, we hypothesized that loneliness and low levels of social support would predict deterioration in social relationships due to COVID-19 restrictions. Finally, in Chapter 7 the main findings of the studies in this dissertation are summarized and discussed. Implications for future research and clinical practice are considered.
2 Chapter 2 - The effectiveness of social network interventions for psychiatric patients: A systematic review and meta-analysis 2 The effectiveness of social network interventions for psychiatric patients: A systematic review and meta-analysis Lise T.A. Swinkels Machteld Hoeve Janna F. ter Harmsel Linda J. Schoonmade Jack J.M. Dekker Arne Popma Thimo M. van der Pol Published in Clinical Psychology Review
26 Chapter 2 ABSTRACT Strengthening social networks is an important goal in mental health treatment. This study aimed to determine the effectiveness of social network interventions for psychiatric patients. A systematic review and meta-analysis was conducted comparing these interventions with control groups on social and mental health-related outcomes in psychiatric patients. PubMed, EMBASE.com, PsycInfo, Scopus, and IBSS were searched for studies until December 21, 2022. Three-level random effects models were used to obtain Cohen’s d mean estimates on composite outcomes of social network and secondary mental health outcomes. Heterogeneity was examined with potential moderators. Thirty-three studies were included. Small-to-moderate effects of social network interventions were detected on positive social network (d = 0.115, p = 0.022) and support (d = 0.159; p = 0.007), general functioning (d = 0.127, p = 0.046), mental health treatment adherence (d = 0.332, p = 0.003), days substance use (d = 0.097, p = 0.004), and abstinence (d = 0.254, p = 0.004). Estimates of psychiatric functioning were higher in samples with more females. The quality of evidence was moderate-to-low. This evidence suggests that social network interventions can improve positive social networks, general functioning, mental health treatment adherence, and substance use in psychiatric patients receiving usual care. Keywords: social network interventions, effectiveness, psychiatric patients, mental health treatment, social support, meta-analysis
27 Review of social network intervention studies 2 INTRODUCTION Mental healthcare incorporates more than treatment of an individual with a mental disorder, especially ifmental problems are chronic and complex (WorldHealthOrganization, 2021). It has been extensively demonstrated that supportive social networks, with personal contacts and ties intended to positively contribute to an individual’s social live or identity, have a protective effect on physical and mental health in general populations (Almquist et al., 2017; Cohen & Wills, 1985; Harandi et al., 2017; Heaney & Israel, 2008; Kawachi & Berkman, 2001; Ozbay et al., 2007). In psychiatric populations, supportive social networks were found to promote mental wellbeing (Pinto, 2006), and mental health recovery (Schön et al., 2009), as well as to reduce depression (George, Blazer, Hughes, & Fowler, 1989), substance use, and criminal behaviors (Knight & Simpson, 1996; Wasserman, Stewart, & Delucchi, 2001). At the same time, social networks of psychiatric patients often consist of lower social support levels, and less reciprocal ties (Clifton, Pilkonis, & McCarty, 2007; Pinto, 2006; ter Haar-Pomp et al., 2015). Therefore, strengthening social environments during mental health treatment, utilizing collaborative approaches targeting supportive social networks and loneliness, could be seen as an important goal in mental healthcare (Giacco et al., 2017; World Health Organization, 2021). However, a more in-depth understanding of the additive effects of social network interventions for psychiatric patients is warranted to examine whether mental health treatments can be further improved. Various types of social network interventions have been developed and identified in previous research, such as clinical or formal interventions, social skills training, support groups, mutual help and self-help groups, community empowerment groups, and volunteer-linking, emphasizing different primary approaches to improve mental health (Andersson, 1998; Biegel et al., 1994; Hogan, Linden, & Najarian, 2002; Perese & Wolf, 2005). Clinical interventions are based on psychotherapeutic and systemic approaches used in professional mental healthcare in which a professional aims to enhance existing supportive social networks. A direct influence of the social network on mental problems of an individual is acknowledged, and therefore, network members are included in treatment to decrease these problems (Andersson, 1998; Cohen & Wills, 1985). Social skills training, often professional-led, is designed to enhance the necessary skills to develop and maintain social relationships by teaching these skills and providing an opportunity to practice skills (Biegel et al., 1994; Hogan et al., 2002). Social skills training is based on the social learning theory, assuming that individuals are able to develop and increase skills to enhance social networks by observing and copying behaviors of others (Bandura & Walters, 1977; Segrin
28 Chapter 2 & Givertz, 2003). In contrast, support groups as well as mutual help and self-help groups rely on the assumption that the provision of social support (leading to enhanced coping) results in improved mental health (Cohen & Wills, 1985; Lakey & Cohen, 2000). Social support can be provided in different ways: emotional (i.e., through the expression of care and empathy), informational (i.e., by providing of information and guidance), appraisal (i.e., by providing constructive feedback and affirmation), and instrumental (i.e., through the direct provision of resources and services) (Heaney & Israel, 2008; House, 1981). For example, support groups are aimed at improving coping in times of stressful events by providing the opportunity to share information, discuss advice and feedback (i.e., informational and appraisal support), and to practice social skills in a group (Perese &Wolf, 2005). Mutual help and self-help groups have a joint focus on the exchange of social support (i.e., emotional, informational, appraisal, and instrumental support) between individuals or peers in a group who share common problems (Andersson, 1998; Perese & Wolf, 2005). Community empowerment groups are aimed at the development of groups in the community characterized by collective strengths to facilitate access to resources and activities (i.e., instrumental support and companionship) in the community (Andersson, 1998; Biegel et al., 1994). Lastly, volunteer and/or peer linkage, including befriending, mentoring, and peer supported interventions, are interventions in which an individual is matched to a trained community volunteer, mentor, or peer respectively to develop an unidirectional social relationship, attain specific goals (i.e., particularly in mentoring), provide social support, engage in social activities, and facilitate access to resources and services in the community (Andersson, 1998; Biegel et al., 1994; Perese & Wolf, 2005; Siette et al., 2017). Although the abovementioned social network interventions are described separately, they are also combined. Thus, in general, the types of social network interventions can be distinguished on the basis of their main approach to improve mental health, such as a focus on maintaining and strengthening existing supportive social relationships (e.g., clinical interventions and social skills training), providing social support (e.g., support groups, mutual help and self-help groups), and building new supportive social relationships in the community (e.g., community groups and volunteer-linking) (Biegel et al., 1994; Heaney & Israel, 2008). Furthermore, social network interventions can differ in the way they are implemented, including the type of care provider used (e.g., natural support members, peers and community volunteers with or without a history of mental problems, and professionals), as well as the format (e.g., group and individual), location (e.g., the provision of social support in the community, neighborhood, or treatment setting), duration, and intensity of care (Andersson, 1998; Biegel et al., 1994; Hogan et al., 2002; Perese & Wolf, 2005).
29 Review of social network intervention studies 2 To date, the empirical evidence on the effectiveness of social network interventions for psychiatric patients is inconclusive. Previous systematic reviews and meta-analyses revealed modest positive results of volunteer/peer linkage interventions, such as befriending (Siette et al., 2017) and peer supported interventions (Barnett et al., 2022; Killaspy et al., 2022; Shorey & Chua, 2022; Smit et al., 2022), on treatment outcomes and recovery among psychiatric populations. Furthermore, multiple previous systematic reviews suggested that interventions directly focused on the development and maintenance of social relationships in the community and social or community participation may be most effective in enhancing social networks of psychiatric patients (Anderson et al., 2015; Barnett et al., 2022; Brooks et al., 2022; Webber & Fendt-Newlin, 2017). Although positive effects of social skills training on social network and treatment outcomes were demonstrated, previous research suggests that these interventions should include techniques to improve on social functioning and skills in real life (Barnett et al., 2022; Hoy, Roher, & Duncan, 2023; Killaspy et al., 2022; Turner et al., 2018). Despite beforementioned research suggesting promising results of several types of social network interventions in psychiatric populations, a recent meta-analysis, including eight studies, found only a small overall effect of social network interventions on perceived social support in psychiatric outpatients (Beckers, Maassen, Koekkoek, Tiemens, & Hutschemaekers, 2022). The previous reviews have offered valuable insights, yet several limitations exist. For example, some reviews included support group interventions providing enhancement of support between peers in the therapeutic environment (Anderson et al., 2015; Lloyd-Evans et al., 2014; Webber & Fendt-Newlin, 2017), thus the effects of enhancement of supportive social networks in real life, outside the psychiatric healthcare practice, remain unclear. Furthermore, most reviews included non-psychiatric populations (Andersson, 1998; Hogan et al., 2002; Hunter et al., 2019; Latkin & Knowlton, 2015; Mead et al., 2010; Siette et al., 2017). Reviews including psychiatric populations only studied specific types of interventions, a limited number of outcome measures, or a specific type of psychiatric population, which resulted in the inclusion of a limited number of independent studies (Beckers et al., 2022; Brooks et al., 2022; Lloyd-Evans et al., 2014; Webber & Fendt-Newlin, 2017). In short, evidence from previous reviews regarding the effectiveness of social network interventions is limited due to an absent focus on interventions aimed at social network enhancement in real life, psychiatric populations, and a broader range of mental health outcomes. The current meta-analysis aims to add to the literature by focusing on the effectiveness of social network interventions directly targeted at improving social networks in real life, in
www.proefschriften.netRkJQdWJsaXNoZXIy MjY0ODMw