Residential care: the effects of Safe Path® intervention, substance (ab)use disorders, and
personality disorders on living group climate
Liana Davidian
University of Amsterdam
A master thesis
Faculty of Social and Behavioral Sciences CLINICAL MASTER FORENSIC PSYCHOLOGY
Words: 7523
Student number: 11407867 Supervisor: Evelyn Heynen
Table of Contents Abstract 3 Introduction 4 Method 7 Participants 7 Safe path® 8 Instruments/materials 8
Prison Group Climate Instrument (PGCI) 9
The Personality Inventory for DSM-5-Brief Form (PID-5-BF) 10
Substance (ab)use disorders 10
Procedure 11
Statistical analysis 12
Results 13
Correlation analysis 13
Independent samples t test 13
Factorial ANOVA 14
Factorial ANOVA (personality disorders) 16
Discussion 17
Abstract
Living group climate has been a predictor for both positive and negative factors within residential care settings, and an open living group climate has been linked to higher treatment motivation with better treatment outcome results. This study examined whether Safe Path® intervention was related to experiencing a more open and supportive living group climate, amongst 91 adult mental health patients with substance (ab)use disorders and personality disorders. Results of an independent samples t test showed no relation between Safe Path® intervention and the experience of a more open living group climate during residential stay. A factorial ANOVA indicated that Safe Path® intervention is positively associated with a decrease in levels of repression within the living group climate. This study can be seen as a basis for future experimental research into the effects of Safe Path® intervention on living group climate within residential care settings. However, longitudinal studies should be considered to explore causal relationships, improve living group climate and subsequently treatment outcomes and recovery in residential care.
Keywords: Safe Path® intervention, Prison Group Climate Instrument (PGCI), DSM-5
Personality Inventory Brief-Form (PID-5-BF), Residential care, substance abuse disorders, living group climate, personality disorders, comorbid disorders
Introduction
Living group climate has been a predictor for both positive and negative factors within residential care settings. A study by Van Der Helm, Klapwijk, Stams, and Van Der Laan (2009) has linked experiencing an open living group climate to higher internal locus of control and treatment motivation, within inmates of a youth prison. Underlying this experience of an open living group climate was the perception of the inmates of ‘being understood by the group workers’. Group workers who paid more attention to the (psychological) needs of inmates and gave them more room to grow, were perceived as being more positive. This had a positive effect on the inmates, in that it promoted feelings of being understood, which ultimately lead to higher treatment motivation and internal locus of control.
Additional research by Souverein, Van Der Helm, and Stams (2013) indicated that experiencing a repressive and closed living group climate is likely to lead to increased recidivism. In their research, living group climate is defined as the way a person is supposed to act, think, and feel in a certain environment or situation. This perception is often shared amongst patients of the residential care setting (Souverein et al., 2013). Living group climate is considered open when patients feel safe, supported, and have plenty of opportunities for personal growth. A closed living group climate on the other hand is experienced when patients feel repressed and often even humiliated (Van Der Helm, Stams, & Van Der Laan, 2011). According to Souverein and colleagues (2013) repression can be defined as punitive power of group workers, injustice, and too many and unfair rules within the residential care setting. Feelings of lack of safety and privacy of the individual also fall into this definition of repression.
Residential care settings are often home to patients with personality disorders and substance (ab)use disorders. According to research by De Graaf, Ten Have and Van Dorsselaer (2010) in the Netherlands, approximately 5.6% of the population was diagnosed with a substance abuse disorder. A study by Bernstein, Arntz, and Vos (2007) indicates that 60% to 90% of patients within Dutch forensic hospitals have been diagnosed with a personality disorder. Personality disorders are often classified as persistent and pervasive, which makes patients diagnosed with these disorders a challenging group to treat (Livesley, 2008).
Comorbidity amongst these disorders has proven to be high (Castel, Rush, Urbanoski, & Toneatto, 2006; Drake, Mueser, Brunette, & McHugo, 2004). Recent studies showed that rates of violent offences were higher in patients diagnosed with a substance abuse disorder and those with
comorbid personality disorders than in mentally ill patients without substance abuse disorders (Hernandez-Avila et al., 2000). These high rates of comorbidity are alarming at the least, since both personality and substance abuse disorders often lead to increased risks for criminal and violent behavior and higher rates of recidivism (Bernstein et al., 2007; Kraanen, Scholing, & Emmelkamp, 2012).
Treatments for individuals with comorbid substance abuse and personality disorders, mostly focus on single disorders at a time. This makes treatment harder for patients, because they are often forced to focus on several systems at once (Drake et al., 2004).
For personality disorders, schema (focused) therapy is one of the most used and effective treatments in forensic settings (Bernstein et al., 2007; Bernstein, Nijman, Karos, Keulen-de Vos, De Vogel, & Lucker, 2012; Masley, Gillanders, Simpson, & Taylor, 2012). Schema (focused) therapy combines different approaches, such as cognitive, social relations, and behavioral into one integrative form of psychotherapy. This form of treatment focusses on patients that showed insufficient results with other forms of treatment and are diagnosed with personality disorders, and/or other disorders which have been found difficult to treat (Young, Klosko, & Weishaar, 2003).
According to Moos (2007), there are several theories who promote the remission of substance (ab)use disorders. Social control theory is one of the theories that comes closest to the care provided in residential care settings. This theory states that strong bonds on several aspects in the individual’s social life are important for acceptable behavior. Lack of structure, monitoring, and support on the other hand, lead to undesirable behaviors such as alcohol and drug abuse. Thus, residential treatment programs often include high amounts of monitoring, structure and even coercion.
A meta-analysis into the effectiveness of coercion in treatment showed that high amounts of coercion lead to a decline in effectiveness of interventions (Parhar, Wormith, Derkzen, & Beauregard, 2008). Repression and coercion often go hand in hand and coercion is said to easily transform into repression, if there is a (extreme) power imbalance between individuals (Lammers & Stapel, 2011; Lammers, Stapel, & Galinsky, 2011; Souverein et al., 2013). Shearer and Ogan (2002) researched a similar topic; treatment resistance in residential substance abuse programs. Results showed that perception of treatment on a voluntary basis, leads to less treatment resistance
amongst patients within residential care settings. This in turn could have a positive effect on treatment motivation and overall treatment outcomes.
The above-mentioned studies show the need for residential treatment programs with an open and supportive living group climate, which at the least provide sufficient monitoring and structure, without coercion and can possibly thus be a benefactor for positive treatment outcomes and successful rehabilitation. Results for the effectiveness of these residential treatment programs are mixed, some showed improvement while others showed no difference in treatment outcomes when compared to non-residential treatment programs (Reif et al., 2014).
The current study, therefore, focusses on a team-based form of schema (focused) therapy, known as “Safe Path®” (Bernstein, 2014). Safe Path® intervention focusses on the use of schema modes, which are said to play a significant role in most personality disorders (Bernstein et al., 2012) and was specifically designed to support the need for an empathic and supportive living group climate within clinical settings.
Schema modes are also known as emotional states and are often referred to as “sides of oneself” that influence the individuals’ feelings, thoughts, and actions (Bernstein, 2014). Safe Path® intervention incorporates the central concept of schema modes into a unified treatment model, with an easy to understand non-judgmental language, and facilitates communication between therapists and patients. This approach makes it easier for patients to discuss their problematic (daily) interactions with others (e.g. therapists, other patients) and easier to resolve them. Safe Path® intervention encompasses several components, these include, but are not limited to: assessment, treatment planning, team-based schema therapy, and monitoring patient outcome. Furthermore, Safe Path® intervention makes incorporation of the elements most beneficial for the treatment of substance (ab)use disorders and personality disorders, like high structure and support and low levels of repression (Moos, 2007) easier (Bernstein, 2014).
Previous studies within a Forensic Psychiatric Centre, showed favorable treatment outcome and rehabilitation results for patients, after implementation of Safe Path® intervention, and Safe Path® training was also positively evaluated by staff members (Bernstein, 2014). The effectiveness of Safe Path® intervention in promoting an open and supportive living group climate and reducing the risk of recidivism is currently being tested in several residential care settings, specialized in substance abuse and personality disorders, within the Netherlands (De Kraker, Schaap-Jonker, & Scholte, n.d.).
In this study, it is expected that the implementation of Safe Path® intervention can have a positive effect on experiencing a more open living group climate for patients during residential stay. This expectation is based on previous literature studies by Bernstein (2014), Bernstein and colleagues (2012), De Kraker and colleagues (n.d.), and Van Der Helm and colleagues (2011).
Furthermore, personality and comorbid disorders are expected to play a significant role in the relationship between Safe Path® intervention and living group climate. Specifically, Safe Path® intervention is expected to have a more positive effect on living group climate, if the participant is suffering from a personality disorder or from a comorbid personality and substance (ab)use disorder. These expectations are based on literature studies done by Bernstein (2014), Bernstein and colleagues (2012), Masley and colleagues (2012), Moos (2007), Parhar and colleagues (2008), and Van Der Helm and colleagues (2011).
While it is uncertain whether personality or comorbid disorders would hinder rather than boost the effects of Safe Path® intervention, it is to be noted that participation in Safe Path® intervention is on voluntary basis for participants in this study. For this reason, it is safe to assume that participants are trying to accomplish something positive and will show more effort (treatment motivation) to gain from this form of therapy.
Method
Participants
111 patients participated in this study at the residential care setting for substance (ab)use disorders and personality disorders “De Hoop” in Dordrecht, the Netherlands. The criteria for participation were being at least 18 years of age and being a resident of “De Hoop” during the period in which this study took place. The sample which completed the study, consisted of 67 male participants with ages ranged between 20-63 (M = 37.13, SD = 10.78) and 24 female participants with ages ranged between 22-62 (M = 35.00, SD = 10.02). Participants were recruited from various living groups within the residential care setting “De Hoop” through verbal communication and flyers, specifically designed for this study. Patients were not compensated for participation. All participants were informed about this study beforehand and signed an informed consent form at the beginning of the study. Participants took part on a voluntary basis and could withdraw from the study at any given point, without having to provide a reason. Data from 20 participants was
excluded from data-analyses due to insufficient information about participation in Safe Path® intervention, possible personality disorders, and/or substance (ab)use disorders.
Safe path®
Participation in Safe Path® intervention was not measured using a questionnaire. Information regarding which groups of participants engaged in the Safe Path® intervention, was provided by the residential care setting.
Safe Path® training was provided to GZ-psychologists, schema therapists, group workers, and other staff members within the residential care setting in question, who in turn were expected to implement the Safe Path® intervention within their groups of patients.
Safe Path® training for all employees consists of a beginners training and an advanced training. The beginners training consists of a one-day course, in which the individual gets an introduction into the methods of Safe Path® intervention. In this beginners’ course, individuals also get to work towards limited reparenting attitudes and case-conceptualization, using schema modes. The advanced Safe Path® course focuses on practical exercises, in which setting boundaries and empathic confrontation are the key elements.
Patients were expected to receive the Safe Path® intervention within the different departments of residence. It is at this point unclear when and how exactly this was accomplished, since staff members were charged with the implementation of the Safe Path® intervention. There is a possibility that some groups of patients received Safe Path® intervention, while others did not.
Instruments/materials
Data was collected using self-report questionnaires filled out on pen and paper. Both were provided by the experimenter. The current study used one questionnaire, which was put together by combining the (original) Dutch version of the (Prison) Group Climate Instrument (PGCI; Van der Helm, et al., 2011) and the Dutch version of The Personality Inventory for DSM-5-Brief Form (PID-5-BF; Van Der Heijden, Ingenhoven, Berghuis, & Rossi, 2014). The questionnaire existed of four parts: quotes on living group climate, rating the four elements of living group climate, quotes on participants’ personality (PID-5-BF), and autobiographical items respectively. We chose not to disclose to the participants that part three of the questionnaire was in fact the PID-5-BF (Van
Der Heijden, et al., 2014). This was done to account for social desirability. A participant number was provided to each questionnaire and linked to the informed consent, to individualize the test results, but keep them as anonymous as possible.
Prison Group Climate Instrument (PGCI)
For measuring the living group climate within the residential care setting, the (original) Dutch version of the (Prison) Group Climate Instrument (Van Der Helm, et al., 2011) was used. This questionnaire is designed to measure the quality of the living group climate and related aspects within residential care settings. Results can be used to improve treatment outcomes, safety within the care setting and even rehabilitation.
The instrument consists of 36 items, measured on a 5-point scale ranging from 1 = ‘totally not applicable’ to 5 = ‘totally applicable’ and measures the four factors characterizing living group climate. These factors measure: growth, repression, support, and atmosphere within the living group climate. The scale growth has seven items and assesses whether individuals have developmental possibilities within the residential care setting and measures their feelings and thoughts about their stay and their future. An example item of this scale is “my stay here is focused on return to society”. Repression is measured using six items and assesses factors like repression and rules (strictness/fairness) within the residential care setting. An example item of this scale is “you always have to ask for permission”. Support is assessed with 11 items and measures the amount of support received by staff members. An example item of this scale is “supervisors always have time for my needs”. Atmosphere uses seven items to assess group atmosphere, which measures levels of feeling safe within the residential care setting and trust amongst individuals. An example item of this scale is “I can trust everybody here”. The PGCI was scored by totaling the scores on the 36 items, leading to a potential range of scores from 1-5. A total score closer to 5 indicates a more open and supportive living group climate, while a total score closer to 1 indicates a closed and repressive living group climate.
Psychometric properties of this instrument were tested in a study by Van Der Helm and colleagues (2011, as cited in Roest, Dekker, Van Miert, De Valk, & Van Der Helm, 2015). Results showed a Cronbach’s alpha of .88 for the PGCI, which indicates a high internal consistency. Scores for the factors support, growth, repression and atmosphere, were α = .90, α = .88, α = .76 and α =.76, respectively.
The Personality Inventory for DSM-5-Brief Form (PID-5-BF)
The Dutch version of The Personality Inventory for DSM-5-Brief Form (Van Der Heijden, et al., 2014) was used to measure personality disorders. This self-report questionnaire was developed to assess the severity of the individuals’ personality impairment within five personality trait domains. These trait domains measure: Negative Affect, Detachment, Antagonism, Disinhibition, and Psychoticism. Each domain consists of five items, with a total of 25 items for the whole questionnaire. An example item of Negative Affect is “I get emotional easily, often for very little reason”. An example item of Detachment is “I often feel like nothing I do really matters”. An example item of Antagonism is “It’s no big deal if I hurt other peoples’ feelings”. An example item of Disinhibition is “People would describe me as reckless”. An example item of Psychoticism is “Things around me often feel unreal, or more real than usual”.
These items are rated on a 4-point scale ranging from 0 = “very false or often false” to 3 = “very true or often true”. The total score ranges from 0 to 75 points for the overall measure, where greater personality dysfunction is indicated by higher scores. Trait domains range in scores from 0 to 15, with higher scores indicating greater dysfunction in the specific personality trait domain. A score of “2” indicates moderate personality dysfunction. In the current study, we used a cut-score of “1”, which indicates some form of personality dysfunction within the participant. This cut-score was chosen since this measure is more of a guidance tool, to help practiced professionals make better judgements on personality dysfunction, rather than a solid diagnosis tool.
Psychometric properties of this questionnaire where tested in a study by Quilty, Ayearst, Chmielewski, Pollock, and Bagby (2013). These properties were tested on a psychiatric patient sample consisting of 201 participants. Overall results provided support for the psychometric properties of this instrument, but showed mixed evidence for discriminant validity and internal consistency.
Substance (ab)use disorders
Substance (ab)use is assessed upon arrival of the patient at the residential care setting. All participants were assessed regarding different domains regarding substance (ab)use. Information was stored in casefiles of the participants, which have previously been filled out by health care professionals at the residential care setting “De Hoop”. The casefiles are based on results from the
Dutch version of the Measurements in the Addictions for Triage and Evaluation (MATE; Schippers, Broekman, & Buchholz, 2011).
This questionnaire was designed to make reliable and valid assessments of participants’ characteristics, including patients’ strengths and limitations and history of (treatment of) substance (ab)use disorders. Results give an indication of the severity of the drug abuse and are used to diagnose dependence on drugs and drug (ab)use disorders based on the Diagnostic and Statistical Manual of Mental Disorders. These results can also be used to provide the best possible care and treatment for patients with substance (ab)use disorders.
Schippers, Broekman, Buchholz, Koeter, and Van Den Brink (2010), tested the psychometric properties of this instrument. Results showed a satisfactory inter-rater reliability. Interviewer-reliability scores, however, indicated that some subscales of this questionnaire need improvement. Significant correlations between the different modules within this questionnaire, indicate concurrent validity.
Since collection of substance (ab)use data was already done prior to the start of the current study, no additional questionnaires were used to assess substance (ab)use disorders.
Procedure
Upon arrival at the appointed area for data collection, participants were given verbal instructions by the experimenter. It was made clear that participation in the study was completely anonymous, voluntary, and that participation could be stopped at any given moment without providing a reason. Following the verbal instructions, participants were asked if there were any remaining questions and instructed to take a seat at a table. Individuals who did not want to participate, could remain in the same room as the participants. Questionnaires were handed out to each participant. Prior to filling out the questionnaires, participants were required to read and sign an informed consent form. An experimenter was present in the room while participants were filling out the questionnaires, to answer any possible questions and to make sure that each participant individually filled out the questionnaires. In some living groups, a group worker or staff member was also present while participants were filling out the questionnaire. Questionnaires were filled out in the same order for each participant. Living group climate was measured first, followed by a personality trait questionnaire. demographic and personal questions were filled out last, including gender, age and duration of residency at “De Hoop”. Groups with a maximum of 15 participants
each simultaneously filled out the questionnaires. Completing the questionnaires took approximately 30 minutes. Questionnaires were collected by the experimenter immediately after completion and participants were thanked for their cooperation. After collection, questionnaires were stored to later be converted into a data-file.
Statistical analysis
Statistical analyses were chosen from the book: Discovering statistics using IBM SPSS statistics by Andy Field (2013) based upon their relevance to the research question and hypotheses in the current study. All data analyses were conducted using IBM SPSS statistics 23.
A bivariate-correlation analysis was used to determine if there was relationship between personality disorders, comorbid disorders, Safe Path® intervention, and living group climate.
For testing whether Safe Path® intervention is related to a more open living group climate during residential stay, an independent-samples t test was used.
Safe Path® intervention was expected to have a more positive effect on living group climate, if the participant is suffering from a personality disorder. Based on this hypothesis, a factorial ANOVA (two-way analysis of variance) was used to test whether there was a main and/or interaction effect between these variables. Specifically, we wished to examine whether engaging in Safe Path® intervention (while suffering from a personality disorder) is related to experiencing a more open living group climate during residential stay. In this analysis, living group climate was used as the dependent variable, while Safe Path® intervention and personality disorders were fixed factors.
A factorial ANOVA (two-way analysis of variance) was also used to test whether there was a main and/or interaction effect between Safe Path® intervention, living group climate, and comorbid disorders. We expected participants who engaged in Safe Path® intervention (while suffering from comorbid disorders) to benefit more from Safe Path® intervention and experience a more open living group climate during residential stay. In this analysis, living group climate was used as the dependent variable, while Safe Path® intervention and comorbid disorders were fixed factors.
Results
Correlation analysis
Means and standard deviations of the study variables and the results of the correlation analyses are presented in Table 1. No significant relationship was found between Safe Path® intervention, Comorbid disorders, and Living Group Climate.
Table 1
Means, standard deviations and bivariate correlations of study variables
Variable n M SD 1 2 3
1. Comorbid Disorders 15 1.64 .48 -.11 .03
2. Safe Path® intervention 48 1.47 .50 -.17
3. Living Group Climate 91 3.61 .55
Note. *p < .05, **p < .001. M and SD are used to represent mean and standard deviation, respectively.
Independent samples t test
An independent-samples t test was conducted to compare living group climate between the Safe Path® intervention (n = 48) and non-Safe Path® intervention (n = 43) groups. It was expected that participants who engaged in Safe Path® intervention would experience a more positive and open living group climate during residential stay. A series of independent samples t tests were also conducted to test the four factors of living group climate (support, growth, repression, atmosphere), individually.
Results showed a non-significant difference in experiencing a more open living group climate, between participants who engaged in Safe Path® intervention (M = 3.69, SD = .56) and participants who did not engage in Safe Path® intervention (M = 3.51, SD = .52), t(89) = -1.62, p =.107, two-tailed, d = .33, 95% CI [-.41, .04]. Thus, no difference in experience of living group climate was found between the Safe Path® intervention and non- Safe Path® intervention groups. Specifically, this indicates that living group climate was not experienced as more open in the group of participants who engaged in Safe Path® intervention during residential stay.
Results also did not show a significant difference in experiencing more support, between participants who engaged in Safe Path® intervention (M = 3.93, SD = .60) and participants who
did not engage in Safe Path® intervention (M = 3.79, SD = .75), t(89) = -.968, p =.336, two-tailed, d = -.20, 95% CI [-.43, .15]. Thus, no difference in experience of support was found between the Safe Path® intervention and non- Safe Path® intervention groups. Specifically, the group of participants who engaged in Safe Path® intervention did not experience more support from staff members during residential stay.
The factor growth had similar results and showed a non-significant difference in experiencing personal growth, between participants who engaged in Safe Path® intervention (M = 4.30, SD = .72) and participants who did not engage in Safe Path® intervention (M = 4.11, SD = .67), t(89) = -1.31, p =.194, two-tailed, d = .27, 95% CI [-.48, .09]. Thus, no difference in personal growth was found between the Safe Path® intervention and non- Safe Path® intervention groups. Specifically, the group of participants who engaged in Safe Path® intervention did not experience more personal growth during residential stay.
However, results showed a significant difference in levels of repression, participants who engaged in Safe Path® intervention (M = 2.53, SD = .66) scored 0.31 points lower, 95% CI [.05, .06] on the factor repression than participants who did not engage in Safe Path® intervention (M = 2.84, SD = .57), t(89) = 2.34, p =.02, two-tailed, d = .50. Specifically, this indicates that the group of participants who engaged in the Safe Path® intervention, experienced less repression by staff members during residential stay.
Results showed a non-significant difference in experience of atmosphere, between participants who engaged in Safe Path® intervention (M = 3.59, SD = .80) and participants who did not engage in Safe Path® intervention (M = 3.34, SD = .90), t(89) = -1.38, p =.172, two-tailed, d = .29, 95% CI [-.60, .10]. Thus, no difference in experience of atmosphere was found between the Safe Path® intervention and non- Safe Path® intervention groups. Specifically, the group of participants who engaged in Safe Path® intervention did not experience a better (more pleasant) atmosphere within the living group during residential stay.
Factorial ANOVA
A factorial ANOVA was conducted to compare the main effects of Safe Path® intervention and comorbid disorders on experiencing a more open living group climate. The interaction effect between engaging in Safe Path® intervention and suffering from comorbid disorders, on
experiencing a more positive and open living group climate, was also tested. Safe Path® intervention consisted of two levels (received Safe Path® intervention, did not receive Safe Path® intervention), comorbid disorders also consisted of two levels (absent, present). It was expected for the Safe Path® intervention to have a positive effect on experiencing a more open living group climate during residential stay, and if the participant is suffering from a comorbid personality disorder and substance (ab)use disorder.
Results of the factorial ANOVA showed a non-significant main effect of Safe Path® intervention on experiencing a more open living group climate during residential stay, F(1, 87) = 3.67, p =.058, partial 2= .041. There was a non-significant main effect of comorbid disorders on experiencing a more open living group climate during residential stay, F(1, 87) = .015, p =.904, partial 2= .000. There was a non-significant interaction between engaging in Safe Path® intervention and the presence of comorbid disorders, on experiencing a more open living group climate during residential stay, F(1, 87) = 1.93, p =.168, partial 2= .022.
When computing the ANOVA for the individual living group factors, results showed a non-significant main effect of Safe Path® intervention on experiencing more support during residential stay, F(1, 87) = 2.04, p =.156, partial 2= .023. There was a non-significant main effect of comorbid disorders on experiencing more support during residential stay, F(1, 87) = .498, p =.482, partial 2= .006. There was a non-significant interaction between engaging in Safe Path® intervention and the presence of comorbid disorders, on experiencing more support during residential stay, F(1, 87) = 2.47, p =.119, partial 2= .028.
Results of the factorial ANOVA showed a non-significant main effect of Safe Path® intervention on experiencing more growth during residential stay, F(1, 87) = 2.35, p =.128, partial
2
= .026. There was a non-significant main effect of comorbid disorders on experiencing more growth during residential stay, F(1, 87) = .569, p =.453, partial 2= .006. There was a non-significant interaction between engaging in Safe Path® intervention and the presence of comorbid disorders, on experiencing more growth during residential stay, F(1, 87) = .595, p =.443, partial
2
There was a significant main effect of Safe Path® intervention on repression during residential stay, F(1, 87) = 5.75, p =.019, partial 2= .062. Specifically, this indicates that individuals who engaged in Safe Path® intervention (M = 2.53, SD = .60), experienced less repression by staff members during residential stay than individuals who did not engage in Safe Path® intervention (M = 2.71, SD = .67). There was a non-significant main effect of comorbid disorders on experiencing less repression during residential stay, F(1, 87) = 1.42, p =.235, partial
2
= .016. No significant interaction was found between engaging in Safe Path® intervention and the presence of comorbid disorders, on experiencing less repression during residential stay, F(1, 87) = 1.04, p =.310, partial 2= .012.
Results of the factorial ANOVA showed a non-significant main effect of Safe Path® intervention on experiencing a better (more pleasant) atmosphere during residential stay, F(1, 87) = 2.53, p =.115, partial 2= .028. There was a non-significant main effect of comorbid disorders on experiencing a better atmosphere during residential stay, F(1, 87) = .564, p =.455, partial 2= .006. There was a non-significant interaction between engaging in Safe Path® intervention and the presence of comorbid disorders, on experiencing a better atmosphere during residential stay, F(1, 87) = 1.69, p =.196, partial 2= .019.
Factorial ANOVA (personality disorders)
A factorial ANOVA (two-way analysis of variance) was planned to be conducted, for comparing the main and interaction effects of Safe Path® intervention and personality disorders on experiencing a more open living group climate during residential stay. It was expected for Safe Path® intervention to have a more positive effect on experiencing a more open living group climate, if the participant is suffering from a personality disorder.
While examining the data used in this study, it was discovered that all participants had underlying substance (ab)use disorders. This finding indicated that there was no group of participants that only suffered from personality disorders. Hence no statistical analysis could be conducted to test this hypothesis.
Discussion
The goal of the current study was to examine the effects of Safe Path® intervention on living group climate in a group of mental health patients with substance (ab)use disorders and personality disorders within a residential care setting.
Firstly, it was hypothesized that participants engaging in Safe Path® intervention would experience a more open living group climate during residential stay. Results of the study showed no difference in experience of overall living group climate, between the groups of participants that engaged in Safe Path® intervention and the groups of participants that did not engage in Safe Path® intervention. Thus, Safe Path® intervention does not seem to be associated with an open and supportive living group climate. This finding is not in line with previous literature studies by Bernstein (2014), Bernstein and colleagues (2012), De Kraker and colleagues (n.d.), and Van Der Helm and colleagues (2011), where it was found that implementation of Safe Path® intervention and the use of the Prison Group Climate Instrument, could possibly lead to a more open and supportive living group climate, which is associated with higher treatment motivation and better treatment outcome results. Thus, no support was found for hypothesis 1.
A possible explanation for failing to find support for hypothesis 1 could be related to the “ceiling effect” (Cramer & Howitt, 2004). This phenomenon occurs when scores on one or more variables used in a study are quite high to start with. When a value is already approaching the upper level at a baseline measurement, it is quite plausible that no effects of an intervention will be found at a follow-up measurement. Since scores are practically at a maximum level, introducing a new variable probably won’t do a great deal of elevating the scores any further (Cramer & Howitt, 2004). In this study, high overall living group climate scores were discovered in both groups of participants. Both participants that had engaged in Safe Path® intervention and those that had not, showed high scores on the factors support, growth and atmosphere and low scores on the factor repression. The factors support and growth were most prominently present within the residential care setting. This appears to indicate that an open living group climate with high support and plenty of growth opportunities was already being experienced by patients within the residential care setting in question. Furthermore, it indicates towards sufficient structure, flexibility, and minimal repression from group workers. Since the living group climate was found to be quite open and supportive in both the Safe Path® intervention and non- Safe Path® intervention groups, the possibility exists that there was not much room for significant
improvement of the living group climate within the residential care setting in the first place. Hence, possible effects of the intervention could have been misinterpreted as not being present within the current study.
Another plausible explanation for failing to find support for hypothesis 1, but finding high overall living group climate scores, is social desirability bias. Grimm (2010) defines social desirability bias as a tendency of participants to give socially desirable answers to questions that reflect sensitive issues, such as drug abuse. These responses are often not reflective of the individuals’ true feelings or thoughts. Tourangeau and Yan (2007) state that fear of repercussions can be a source of social desirability bias and subsequently found that this can lead to substantial underreporting of problematic issues. In the current study, some staff members and group workers were present in the room during data collection. It is thus quite plausible that fear of repercussion by group workers and other staff members could have led to socially desirable answers within both groups of participants, which in turn could have led to higher overall living group climate scores in all participants. Thus, it is advised to account for social desirability bias in future studies. Tourangeau and Yan (2007) imply that it is important not to solely rely on one assessment method. In case of self-report questionnaires, one could use complementary methods, such as in-depth interviewing. Grimm (2010) states that it is highly important to use well trained interviewers, to help avoid this bias in some extent.
Results for hypothesis 1, however, did reveal significant lower levels of repression among the groups of participants who engaged in Safe Path® intervention. Patients within various living groups, that did engage in Safe Path® intervention, perceived rules to be less unfair and strict; and staff members were perceived as being more responsive and flexible. This indicates that Safe Path® intervention appears to at least positively contribute to one of the four factors that constitute living group climate, namely, repression. Repression is an important factor within the living group climate, since a high level of repression is associated with a closed living group climate and with more recidivism. A low level of repression on the other hand is associated with an open living group climate, leading to higher treatment motivation and better overall treatment results (Shearer & Ogan, 2002; Souverein et al., 2013; Van Der Helm et al., 2009). This association between Safe Path® intervention and the perception of less strict and unfair rules; and perhaps most important, an increase in responsivity and flexibility from staff members is a key finding in this study. Responsivity and flexibility from group workers influence important factors in the process of
rehabilitation. Specifically, responsivity and flexibility facilitate an open living group climate, which in turn helps to restore the patients’ damaged bond with society. Therefore, making it easier to reintegrate into society (Souverein et al., 2013). These findings indicate that implementation of Safe Path® intervention could lead to better treatment results of patients within the residential care setting in question and ultimately even to successful rehabilitation. However, further research on the effectiveness of Safe Path® intervention is needed for making inferences on causality.
Secondly, it was expected that participants suffering from personality disorders would benefit more from Safe Path® intervention; and experience a more open living group climate during residential stay. Due to unforeseen findings in the dataset, in which it was discovered that all participants had underlying substance (ab)use disorders, no statistical analysis could be conducted to test hypothesis 2. Since the hypothesis could not be subjected to further analysis, it can neither be confirmed nor rejected.
Lastly, Safe Path® intervention was expected to have a more positive effect on living group climate during residential stay if the participant is suffering from a comorbid disorder (personality disorder and substance (ab)use disorder). This effect was not found however. Thus, Safe Path® intervention does not seem to be related to a more open living group climate, for patients suffering from comorbid disorders. Studies done by Bernstein (2014), Bernstein and colleagues (2012), De Kraker and colleagues (n.d.), Moos (2007), and Van Der Helm and colleagues (2011) found that living group climate is considered as open if structure and support are prominently present. These elements are also incorporated in the Safe Path® intervention, supposedly making it both a beneficial treatment for patients with comorbid disorders and improving overall living group climate. Results in the present study are not in line with the afore mentioned studies. Hence, no support was found for hypothesis 3.
A plausible explanation for failing to find support for hypothesis 3, can be found in the use of the questionnaire for defining personality disorders; and the information provided regarding substance (ab)use disorders. The current study used the Dutch version of The Personality Inventory for DSM-5-Brief Form (Van Der Heijden, et al., 2014) to define whether the patient was suffering from personality disorders. An overall score of 2 on this questionnaire, indicates moderate personality dysfunction within the individual. However, at this point it is unclear how severity of the personality dysfunction, on the different personality traits within the PID-5-BF, can be determined. This in turn makes it very unclear to what end the PID-5-BF can be used to correctly
classify different personality disorder types (Al-Dajani, Gralnick, & Bagby, 2016). Furthermore, it is to be noted that this questionnaire cannot solely be used as a diagnostic instrument, rather it is to be used as a guidance tool to help practiced professionals make better judgements on personality dysfunction and suitable treatment methods (Van Der Heijden, et al., 2014). In the current study, no practiced professionals were called upon to assist in categorizing the participants into either the personality disorders, comorbid disorders, and no personality disorders or comorbid disorders groups. Since the PID-5-BF did not offer a clear cut-score to indicate personality dysfunction, an overall score of 1 was used to categorize individuals into the personality disorders group. Though the residential care setting did provide a document on the most common personality disorders within the different living groups, no diagnosis documents were available on individual participants’ personality dysfunction. The same is true regarding severity of substance (ab)use disorders within participants in this study, where no information was available regarding substance (ab)use on the individual level of the participants. Assigning participants to the different disorder groups was solely based on personal judgement of the PID-5-BF results. This makes it likely that misinterpretations were made, when assigning participants to the personality disorders, comorbid disorders, and no personality disorders or comorbid disorders groups. Future studies, should use several diagnosis tools who measure personality dysfunction, and severity of substance (ab)use in combination with judgments from practiced professionals, to provide a correct diagnosis for each participant.
An alternative explanation for failing to find support for both hypothesis 1 and hypothesis 3 can be found in literature studies done by Barth (2005), Parhar and colleagues (2008), Souverein and colleagues (2013), and Van Der Helm and colleagues (2009) regarding the overall effectiveness of treatment programs within residential settings. Residential treatment programs are often recommended for individuals suffering from comorbid disorders, though growing in numbers, uncertainty regarding overall effectiveness remains (Barth, 2005). Souverein and colleagues (2013) second this notion and state that there are no real evidence-based treatments within residential care settings, thus making it important to look for more promising treatment methods outside these residential settings. Moreover, it seems that residential treatment programs are high in cost while offering little benefit compared to non-residential care (Barth, 2005). Parhar and colleagues (2008) found close to no effects of treatment with incarcerated adults. However not all research findings are negative and some promising results with residential treatment programs
were found with youth residing in prisons. A possible explanation for this finding is that adolescents could be more susceptible to treatment regardless of setting. Another point of interest is that treatment effectiveness seems to only account for approximately 15% of outcome variance, indicating that external factors could be at play (Van Der Helm et al., 2009). Regarding the overall effectiveness of residential treatment programs, it is of importance to keep in mind that patients within residential settings frequently have ongoing problems due to a history of ineffective treatment and care (Souverein et al., 2013). As to residential treatment programs that are evidence-based, high levels of supervision seem to account for their effectiveness (Barth, 2005). These findings on the effectiveness of residential treatment programs could have, in part, accounted for the results of the current study. Though most finding on effectiveness were done in correctional settings (prisons), it is to be noted that the residential setting in this study shares a resemblance with most correctional settings (e.g. similar population, structure, and rules). Future research should focus more on the history of patients, possibly external factors counteracting the treatment effects, and possible most crucial; the quality and amount of supervision provided.
The present study is probably one of the first studies to examine the relationship between Safe Path® intervention and living group climate, within a residential care setting specialized in patients with personality disorders and substance (ab)use disorders. This makes findings prone to several limitations.
First, the sample size of 91 participants was selected from only one residential care setting. Furthermore, this number was not equally distributed amongst gender and age groups (67 male vs 24 female participants), and was also not based on statistical power analysis, making it likely that the sample is too small to find statistically significant results and make statements which can be generalized to different age groups and residential populations.
Second, personality disorders and substance (ab)use disorders were not categorized into specific disorder types, severity thereof, and treatment outcomes were also not measured. Thus, making it impossible to, at this time, make statements on the effectiveness of Safe Path® intervention for individuals suffering from personality disorders and substance (ab)use disorders at different degrees of severity.
Third, the goal of this study was to test, whether implementation of Safe Path® intervention was related to a more open and supportive living group climate during residential stay. This required the same groups of individuals to be tested over long periods of time. However, new
information received later in the study revealed that there were some shifts within the living groups at the residential care setting. Patients had transferred from one living group to another and it became clear that some patients might already have engaged in Safe Path® intervention, prior to the start of this study. It is thus probable that Safe Path® intervention could have played a role in contributing to a more open living group climate at the start of the study. This in turn could have accounted for the finding of high overall living group climate scores in both groups of participants, possibly leading to no effects of the intervention used. Though high overall living group scores are a positive finding, it is important to note that information on the individual participants who received Safe Path® intervention and those that did not, is not quite clear. This led to the inability to test the same group of participants over periods of time, which makes it impossible to make statements on causality between the study variables.
Last, participating staff members were instructed by practiced professionals in the use of Safe Path® intervention methods. Staff members in turn were expected to carry out the Safe Path® intervention within the various living groups at the residential care setting, making it unclear which methods were used to accomplish this. For making statements on the effects of Safe Path® intervention on living group climate, it is thus of great importance to consider the contents of the Safe Path® intervention and look into personal information regarding staff members in more detail. Factors like frequency and duration of the intervention, (psychological) state of the staff members who delivered the Safe Path® intervention, and other external factors that could infer with correct implementation of Safe Path® intervention, should be considered and ruled out one by one.
Because of these limitations, the findings of this study should be interpreted with great caution. Results should be replicated in a prospective, longitudinal study within several residential care settings, which would allow for the examination of participants on the individual level, and make statements on causality possible. Nevertheless, research findings in this study could be considered as a basis for future studies into the effectiveness of Safe Path® intervention for patients suffering from personality disorders and substance (ab)use disorders within residential care. Increased responsivity and flexibility by staff members, perception of less strict and unfair rules, and overall improvement of the living group climate, through implementation of Safe Path® intervention for instance, could lead to higher treatment motivation, and restoring damaged bonds with society. Conclusively, Safe Path® intervention, if implemented correctly and with the right
amount of supervision could elevate treatment results, and ultimately even lead to successful rehabilitation.
References
Al-Dajani, N., Gralnick, T. M., & Bagby, R. M. (2016) A Psychometric Review of the
Personality Inventory for DSM–5 (PID–5): Current Status and Future Directions, Journal of Personality Assessment, 98, 62-81. doi: 10.1080/00223891.2015.1107572
Barth, R. P. (2005). Residential care: From here to eternity. International Journal of Social Welfare, 14, 158-162. doi: 10.1111/j.1468-2397.2005.00355.x
Bernstein, D. P. (2014). “Safe Path”® for Patients with Addiction and Personality Disorders, Proposal for the Ommekeer. Unpublished manuscript.
Bernstein, D. P., Arntz, A., & Vos, M. D. (2007). Schema focused therapy in forensic settings: Theoretical model and recommendations for best clinical practice. International Journal of Forensic Mental Health, 6, 169-183. doi:
10.1080/14999013.2007.10471261
Bernstein, D. P., Nijman, H. L. I., Karos, K., Keulen-de Vos, M., De Vogel, V., & Lucker, T. P., (2012). Schema therapy for forensic patients with personality disorders: Design and preliminary findings of a multicenter randomized clinical trial in the Netherlands. International Journal of Forensic Mental Health, 11, 312-324. doi;
10.1080/14999013.2012.746757
Castel, S., Rush, B., Urbanoski, K., & Toneatto, T. (2006). Overlap of clusters of psychiatric symptoms among clients of a comprehensive addiction treatment service. Psychology of Addictive Behaviors, 20, 28. doi: 10.1037/0893-164X.20.1.28
Cramer, D., & Howitt, D. L. (2004). The Sage dictionary of statistics: a practical resource for students in the social sciences. Thousand Oaks: Sage.
De Graaf, R., Ten Have, M., & Van Dorsselaer, S. (2010). De psychische gezondheid van de Nederlandse bevolking. NEMESIS-2: Opzet en eerste resultaten (Research Report). Retrieved from Trimbos Instituut website: http://raymondgruijs.nl/wp-
content/upload_folders/raymondgruijs.nl/2015/09/x-cz-2995458b.pdf De Kraker, P. T., Schaap-Jonker, H., & Scholte, W. (n.d.). Schematherapie en een
stimulerende relationele omgeving bij complexe verslaving: verwachte veranderingen in neurocognitieve netwerken. Unpublished manuscript.
Drake, R. E., Mueser, K. T., Brunette, M. F., & McHugo, G. J. (2004). A review of treatments for people with severe mental illnesses and co-occurring substance use disorders. Psychiatric rehabilitation journal, 27, 360. doi: 10.2975/27.2004.360.374 Field, A. (2013). Discovering statistics using IBM SPSS statistics. Thousand Oaks, CA: Sage. Grimm, P. (2010). Social desirability bias [Wiley Online Library version].
doi: 10.1002/9781444316568.wiem02057
Hernandez-Avila, C. A., Burleson, J. A., Poling, J., Tennen, H., Rounsaville, B. J., & Kranzler, H. R. (2000). Personality and substance use disorders as predictors of criminality. Comprehensive Psychiatry, 41, 276-283. doi: 10.1053/comp.2000.7423 Heynen, E. J. E. (2016). No Guts, No Gains! The relation between living group climate and social development of juvenile delinquents in detention (Doctoral thesis, University of Amsterdam, Amsterdam, Netherlands). Retrieved
from http://dare.uva.nl/search?identifier=5849761c-012e-4813-b493-ebfd4aef5bfb Kraanen, F. L., Scholing, A., & Emmelkamp, P. M. (2012). Substance use disorders in forensic psychiatry: Differences among different types of offenders. International journal of offender therapy and comparative criminology, 56, 1201-1219. doi: 10.1177/0306624X11420252
Livesley, J. W. (2008). Integrated therapy for complex cases of personality disorder. Journal of Clinical Psychology, 64, 207- 221. doi:10.1002/jclp.20453
Masley, S. M., Gillanders, D. T., Simpson, S. G., & Taylor, M. A. (2012). A Systematic Review of the Evidence Base for Schema Therapy, Cognitive Behaviour Therapy, 41, 185-202, doi: 10.1080/16506073.2011.614274
Moos, R. H. (2007). Theory-based processes that promote the remission of substance use disorders. Clinical psychology review, 27, 537-551. doi: 10.1016/j.cpr.2006.12.006 Parhar, K. K., Wormith, J. S., Derkzen, D. M., & Beauregard, A. M. (2008). Offender coercion in treatment: A meta-analysis of effectiveness. Criminal Justice and Behavior, 35, 1109- 1135. doi: 10.1177/0093854808320169
Quilty, L. C., Ayearst, L., Chmielewski, M., Pollock, B. G., & Bagby, R. M. (2013). The psychometric properties of the Personality Inventory for DSM-5 in an APA DSM-5 field trial sample. Assessment, 20, 362-369. doi: 10.1177/1073191113486183
Reif, S., George, P., Braude, L., Dougherty, R. H., Daniels, A. S., Ghose, S. S., & Delphin- Rittmon, M. E. (2014). Residential treatment for individuals with substance use disorders: assessing the evidence. Psychiatric Services, 65, 301-312. doi: 10.1176/appi.ps.201300242
Roest, J. J., Dekker, A. L., Van Miert, V. S. L., De Valk, S., & Van Der Helm, G. H. P. (2015). GCI: Group Climate Instrument [Measurement instrument]. Retrieved from
https://www.hsleiden.nl/binaries/content/assets/hsl/lectoraten/residentiele- jeugdzorg/handleidingen/handleiding-gci-2015.pdf
Schippers, G. M., Broekman, T. G., & Buchholz, A. (2011). MATE 2.1. Handleiding en protocol. Nijmegen, Netherlands: Bêta Boeken.
Schippers, G. M., Broekman, T. G., Buchholz, A., Koeter, M. W., & Van Den Brink, W. (2010). Measurements in the Addictions for Triage and Evaluation (MATE): an instrument based on the World Health Organization family of international classifications. Addiction, 105, 862-871. doi: 10.1111/j.1360-0443.2009.02889.x
Shearer, R. A., & Ogan, G. D. (2002). Voluntary participation and treatment resistance in substance abuse treatment programs. Journal of Offender Rehabilitation, 34(3), 31-45. Souverein, F. A., Van Der Helm, G. H. P., & Stams, G. J. J. M. (2013). ‘Nothing works’ in secure residential youth care? Children and Youth Services Review, 35, 1941- 1945. doi: 10.1300/J076v34n03_03
Tourangeau, R., & Yan, T. (2007). Sensitive questions in surveys. Psychological bulletin, 133, 859. doi: 10.1037/0033-2909.133.5.859
Van Der Heijden, P., Ingenhoven, T., Berghuis, H., & Rossi, G. (2014). DSM-5 persoonlijkheidsvragenlijst [DSM-5 personality inventory]: PID-5-NL. Dutch
translation of The Personality Inventory for DSM-5® (PID-5) Adult, 2011 (American Psychiatric Association). Amsterdam: Boom.
Van Der Helm, P., Klapwijk, M., Stams, G.J.J.M., & Van Der Laan, P.H. (2009). ‘What Works’ for juvenile prisoners: The role of group climate in a youth prison. Journal of Children’s Services, 4, 36-48. doi: 10.1108/17466660200900011
Van Der Helm, G. H. P., Stams, G. J. J. M., & Van Der Laan, P. H. (2011). Measuring group climate in prison. The prison journal, 9, 158-176. doi: 10.1177/00322885511403595
Young, J. E., Klosko, J., & Weishaar, M. (2003). Schema Therapy: A Practitioner’s Guide.
