Improving Work Climate in a Residential Setting for Juveniles with Mild Intellectual Disability through Training of Staff in Non-violent Resistance

ABSTRACT

In residential settings for juveniles with Mild Intellectual Disability (MID) employees have to face tough work conditions, because of frequent aggressive incidents Eventually, this might result in a negative work climate, high staff-turnover and absenteeism. In turn, it hampers a positive therapeutic environment for youngsters. A method that may endorse a better work climate is Non-violent Resistance (NVR). As earlier research suggests that the residential NVR, adjusted for juveniles with MID, is successful in diminishing aggressive incidents and improving group climate, we hypothesize that NVR-MID will be also effective in enhancing work climate. We investigated the perception of work climate by staff of three residential settings before, during and after implementation of NVR-MID. Work climate was assessed both quantitatively and qualitatively. Outcomes were subsequently integrated. Quantitative data was collected using the Living Group Working Climate Inventory (LGWCI). Multilevel analyses were carried out to analyse the development of work climate over time. Qualitative data were collected via semi-structured interviews with NVR-MID trained staff and analysed through qualitative coding. Quantitative results showed that Job Satisfaction decreased after NVR-MID implementation. However, within qualitative reports, staff members experienced a positive influence on work climate of NVR-MID.

KEYWORDS:

• Non-violent Resistance
• Work climate
• intellectual disbaility

Practical Implications

• NVR-MID may improve work climate in residential settings for juveniles with MID

• NVR-MID may be beneficial to enhance team spirit within residential units for juveniles with MID

• (One of) the first studies on the effects of work climate after implementation of NVR-MID

Employees who work in facilities for juveniles with mild intellectual disabilities (MID) are often confronted with aggressive behavior and therefore have to deal with tough work conditions (Embregts et al., 2009). Possible consequences for staff having to deal with a high workload and high amounts of aggression are (longtime) absenteeism or ultimately resignation and job change (Colton & Roberts, 2007; Knorth et al., 2010; Seti, 2008). High staff turnover is a major problem in institutions with MID youth (Hatton et al., 2001). In the Netherlands, as in many other countries, the outflow of staff outnumbers the inflow of new staff who start working in youth care settings (Welzijn, 2017). A recent study revealed that in 2018 almost half of the staff in residential institutions in the Netherlands indicate that perceived workload has increased, compared to earlier years (CAOP & Prismant, 2018). The most common reasons for the increased perceived workload were lack of staff (16%) and high rates of sick leave (14%) (CAOP & Prismant, 2018). Consequences of this high rate of staff turnover again are high proportions of relatively inexperienced and untrained staff working in the services, and the lack of continuity in the service provided to people with MID (Hatton et al., 2001). Evidently, it is important to identify the particular reasons for increased perceived workload and staff resigning in order to being able to influence them. A recent review of d’Ettorre and Pellicani (2017) has shown that aggressive behavior in psychiatric inpatient wards accounts for 30% of the overall costs of accidents and absenteeism among employees on the psychiatric wards. Therefore, it should be a common goal to decrease aggressive incidents and improve work climate, by implementing an intervention enabling that. Less aggression on the ward and improved work climate should ultimately lower the stress levels among employees who work in residential facilities for children and adolescents with MID.

Non-violent Resistance (NVR) (Weinblatt & Omer, 2008) is a method that, amongst others, aims to reduce aggressive incidents and enhance work satisfaction. Teams are taught how to prevent escalations with the admitted juveniles and how to strengthen their mutual relationship. NVR for the residential setting is based on the philosophy and methods used by Gandhi and Martin Luther King in their social political struggle and originates from the method “Non-violent Resistance: A New Approach to Violent and Self-Destructive Children,” developed by the Israeli psychologist Haim Omer (Omer, 2004a). NVR primarily focuses on changing the behavior of the caretaker instead of that of the juvenile and aims to decrease problematic behavior by restoring the troubled relationship between juvenile and caretaker. As NVR is a systemic approach, its core is about being a team and acting together. One of the key factors of NVR is the importance of the “we,” the feeling in a team of working colleagues, that they are not alone but that they rather work as a team.

As NVR has proven to reduce helplessness and stress for parents (Van Holen et al., 2016; Weinblatt & Omer, 2008), it is suggested that when translated into the residential setting, where parental tasks are partly taken over by staff, it might also have positive effects. This assumption was supported in a qualitative study of NVR trained staff, wherein staff is highly positive about the method itself (Van Gink et al., 2017). The same authors have also studied the possible influence of NVR on work climate in residential settings (not specified for MID) (Van Gink et al., 2018). They found that NVR enhances team functioning, job satisfaction, task-significance, and shared vision and commitment (Van Gink et al., 2018). These findings have to be interpreted with caution, due to only small-to-medium effect sizes, as well as that positive effects were only found in settings where staff turnover was relatively stable and NVR was implemented well (Van Gink et al., 2018). The authors argue that due to the large-scale reorganization at the other settings, NVR could not be implemented properly. As such, it remains unclear what the effects of NVR on working climate when there has been a successful implementation process.

Notably, the regular residential NVR method has recently been adapted for the use in residential settings for youngsters with MID; “NVR-MID” (Haspels et al., 2017; Visser et al., 2017). A short overview on the NVR-MID program approach is given in the Appendix A. A link between training of staff in NVR-MID and a decrease in aggressive incidents has been observed (deleted because of blinding reason). Furthermore, the same authors (on the same population) have found that NVR improved an open group climate as reported by the admitted juveniles (deleted because of blinding reason). Hence, it is hypothesized that NVR-MID would over time improve work climate and therefore ultimately decrease staff turnover and absenteeism.

Therefore, the main aim of this study is to investigate the influence of residential NVR-MID training on the perceived work climate of staff members in residential settings for juveniles with MID in the Netherlands.

Method

This study consists of two parts: both quantitative and qualitative data collection. To enhance a clear structure, we chose to report the two parts successively, starting with the quantitative part. The data used in this study stem from a longitudinal study collected from October 2015 until October 2018. The study has been approved by the Medical Ethics Committee of the VU medical Center of Amsterdam (2015.344).

Quantitative Part

Design

We use a quasi-experimental stepped wedge design as an advanced variant of a pre-post design, as random allocation to different conditions was not desirable. The stepped wedge design is a widely used design that is ideally suited for effectiveness studies in clinical practice (Brown & Lilford, 2006; Campbell & Cook, 1979). The advantage of this design is that each cluster functions as its own control group and that the phased implementation per cluster ensures that the measurements are independent of any other event in the implementation period. Potential influences of any outside events, such as Christmas or the holiday period, are more easily detected and can therefore be controlled for. In addition, it is also a pragmatically favorable design, because the departments do not have to be trained at the same time, which greatly enhances the feasibility of such as study. In each institution there were three stages: 1) before NVR-MID training, 2) during NVR-MID training and 3) after NVR-MID training. Each stage was introduced into the institutions at a different time-point, see Table 1. There were two pre-training measurements (T0, T1), three during-training measurements (T2, T3, T4) and two post-training measurements (T5, T6). The training of staff in the NVR-MID method covered a total of 9 months (T2, T3, T4).

Table 1. Training in NVR-MID at the three Residential Settings in a Quasi-Experimental Stepped Wedge Research Design

Site	Sep 16	Nov 16		Feb 17	May 17	Aug 17	Nov 17	Feb 18	May18	Aug 18
1	T0	T1		T2	T3	T4	T5	T6
2		T0		T1	T2	T3	T4	T5	T6
3				T0	T1	T2	T3	T4	T5	T6
			Pre-Training Measurement
			During-Training Measurement
			Post-Training Measurement
			No Measurement

Participants

Staff of the participating residential settings for juveniles with MID and co-morbid behavioral and/or psychological problems were invited to participate in this study (N = 264). The sample consisted of 64 males and 200 females with an age range between 19 and 64 years (M_age = 34.7, SD = 10.1). Thirty-two participants worked at setting A, 93 at setting B and 139 at setting C. Of the sample, 221 participants worked as group workers (i.e. residential care staff who works directly with a group of youth) and 43 had other professions, such as psychologists, psychiatrists, and/or systemic therapists. For an overview of participant characteristics, see Table 2.

Table 2. Participants per Youth Care Setting, Gender, Mean Age (SD) and Percentage Group Leader

	Site 1	Site 2	Site 3	Total	Statistical comparison
N	32	93	139	264
% Male	46.9	18.3	23	24.2	X^2= 10.8, p = .05 ^a
Mean Age (SD)	36.6 (10.4)	36.5 (11.1)	33.1 (9.1)	34.7 (10.1)	F = 6.32, p = .01 ^b
% Group Leader	87.5	87.1	80.6	83.7	X^2= 2.1, p > .05

^a= significant difference between site 1 and site 2 ^b = significant difference between site 2 and site 3 ^c = significant difference between site 1 and site 3

Participating Sites

Site 1 forms the first cluster, with two units, located in a rural area of central Netherlands and was intended for clients with mild-to-moderate IDs. Clients often stayed here for a long period of time or even lived there permanently.

Site 2 forms the second cluster with four units, located in the province Southern Netherlands in a more urban environment and was aimed for clients aged from 6 to 21 years with MID and other comorbid psychiatric problems, such as autism, ADHD, and/or emotional problems. Some of the clients also had physical impairments, such as being in a wheelchair.

Site 3 forms the third cluster with six units, located in a northern part of the Netherlands in a rural surrounding and was aimed for clients between 12 and 23 years old, having severe behavioral problems besides their MID. Most clients were admitted by civil law or if parents could no longer handle the situation at home.

All sites used therapeutic measures that mainly focused on the behavior of the clients, ensuring safety for staff and clients themselves, for example, “Competence-based Guidance” (Competentie gericht begeleiden (Movisie, 2017)) or the Triple C method (Triple-C (Weerd & Wouwe, 2015). Besides a basic method, some of them also used more specific methods, which were suitable for working with youngsters with autism such as “Give me 5” (Geef me de vijf; (de Bruin & Gelderland)). The most important difference between these methods and NVR-MID is that within NVR-MID we focus primarily on the behavior of the professional than on the behavior of the youngster. Staff members of all sites were asked to participate in the study through a presentation on the ward and were handed written information on the study. Participants had a few days’ time to consider whether they wanted to participate and if they agreed, they signed an informed consent form.

Procedure

NVR-MID was implemented during a nine-month training period and consisted of three full training days for staff members (two days at the start of the training and one day at follow-up) in which the principles and interventions of NVR-MID were explained (Haspels et al., 2017). Every 10 weeks, there were additional supervision meetings in which the use of NVR-MID was evaluated and deepened through case studies, examples and questions form the work floor.

Work climate was assessed at seven time points, every three months: two measurements before (T0, T1), three during (T2, T3, T4) and two after (T5, T6) the NVR-MID training. At each measurement point, research assistants (Master’s students in psychology or remedial education) visited the residential youth care settings and handed out the questionnaires (see materials). At the same time, links were sent via e-mail so that participants could also complete them via the computer. Most work climate questionnaires were completed via the online questionnaire, which took them on average 45 minutes. However, the second measurement was linked to the training day, meaning that all questionnaires were handed to the participants on paper and they were asked to complete it right away. The percentage of completed questionnaires varied in between the measurements, between 72.8% (before), 62.5% (during) and 40.4% (after). The drop-outs were on the one hand caused by normal variations of professionals being on maternity leave, sick leave or resigning from jobs and on the other hand there were some reorganizations at one of the institutions, where 4 units were closed toward the end of this study. Most of the staff who worked at those units had been divided over the remaining units.

Materials

Work climate was measured with the Living Group Working Climate Inventory (LGWCI, (Dekker et al., 2014)). The LGWCI is a composite questionnaire consisting of different validated questionnaires. Work climate focuses on the vision of staff of their own work and possible factors that could influence their work climate. The questionnaire is developed for staff in different youth care and treatment facilities who maintain professional contact with the admitted juveniles. The questionnaire consists of a total of eight subscales, of which only four were used in this study because other subscales (such as Leadership) were considered less relevant to NVR-MID implementation. The subscales used were 1) Team functioning, 2) Work motivation, 3) Perceived Workload and 4) Shared Vision and Commitment. Items range on a five-point Likert-scale from 1 = totally disagree to 5 = totally agree. The completion of the questionnaire takes approximately 25 minutes. Internal consistency across these scales range from α = 0.71 to 0.88 (Dekker et al., 2014).

1) Team Functioning was based on the PANTRIX (18 items) (Van Der Helm et al., in press) and consists of positive and negative team functioning.

a) Positive Team Functioning (eight items) consists of items like “Team members can handle unexpected situations at the living group.” Reliability for this subscale is α = 0.87. b) Negative Team Functioning (10 items) knows items like “The team is disintegrating, roles and positions are unclear.” Reliability for this subscale is α = 0.79.

2) Work Motivation (14 items) is based on the Dutch questionnaire The List of Task Perception and Organizational Climate [Uitsprakenljst Taakbeleving en Organisatieklimaat (TABOK)] (Jansen & Zaal, 1984) and the Questionnaire Experience and Assessment of Labor [Vragenlijst Beleving en Beoordeling van de Arbeid (VBBA)] (Van Veldhoven et al., 2002) and consist of two subscales Job-Satisfaction and Task-significance.

a) Job-Satisfaction (seven items) measures the extent on how satisfying staff experiences their work. An example item is “I am satisfied with my job” Reliability for this subscale is α = 0.81.

b) Task-Significance (seven items) measures the extent to whether staff experiences the things they do at work as significant. An example is “I enjoy making contact with the clients.” Reliability for this subscale is α = 0.77.

3) Perceived Workload (eight items) is based on the VBBA (Van Veldhoven et al., 2002) and is about pressure, work-related stress and perception of the workload. An example item is “At my unit there has been a shortage of staff.” Reliability for this subscale is α = 0.68.

4) Shared Vision and Commitment (13 items) is based on both, the TABOK (Jansen & Zaal, 1984) and VBBA (Van Veldhoven et al., 2002) and consists of two subscales shared vision and commitment.

a) Shared Vision (five items) provides insight in the way staff experiences agreement in one’s own vision and vision of others (colleagues and the organization). An example item is “My team members and I share our vision of the clients and organization.” Reliability for this subscale is α = 0.74.

b) Commitment (eight items) is about the affective commitment of staff, an example item is “I feel emotionally attached to this organization.” Reliability for this subscale is α = 0.78.

Data Analysis

Differences in staff characteristics between sites were determined using one-way ANOVAs with post-hoc Bonferroni test and chi-square tests using SPSS 22.0. Data of the seven measurement points (T0 – T6) were averaged into three blocks: before (T0 and T1), during (T2, T3, and T4) and after (T5, T6) NVR-MID training. To test the hypotheses, linear mixed models analyses were performed to examine the relation between NVR-MID and work climate over time. Each subscale functioned as dependent variable in a separate linear mixed models analysis. To test whether NVR-MID was related to a change in work climate over time, the main effect of the fixed factor time (within-subject: before, during, after) was studied. As it was expected that participants of the three youth care settings would differ from each other, youth care setting (between-subject: site 1, site 2, site 3) was added as a fixed variable to all analyses, so that differences between the residential youth care settings could be accounted for. The interaction between time and youth care setting was studied to test whether the relation between NVR-MID and the change in work climate over time was different for the three residential youth care settings.

Qualitative Part

Study Design

A qualitative exploratory design was used. The consolidated criteria for reporting qualitative research (COREQ) guideline (Tong et al., 2007) was followed during the conduct and reporting of the study.

Participants

Participants were 18 employees of the three residential settings described above. The mean age of the participants was 39,4 years (range 23– 61 years; SD = 11.5). Five of them were men (27.8%). 77.8% worked as group workers, the other 22.2% as staff. Working experience varied between less than a year and more than 10 years (6 of them worked longer than 10 years, 10 between 4 and 10 years, 1 between 2 and 4 years and 1 less than 1 year). All participants were trained in NVR-MID for the residential setting (Haspels et al., 2017).

Procedure

A total of 25 employees were invited to participate via an e-mail explaining the purpose of the study and the informed consent procedure. Employees could ask questions about the study by contacting the researchers. Purposive sampling was used to ensure a maximum in diversity of the perspectives of the participants (Boeije, 2005). Diversity was sought in gender, levels of working experience with bringing NVR-MID into practice and profession (e.g., group workers, psychologists, parent counselors and psychiatrists). Managers were asked to invite employees to participate in this study and were explicitly asked to not only include staff-members who had positive feelings toward the NVR-MID implementation. Eventually, 19 employees agreed to participate. The other 6 employees did not want to participate due to time issues or other reasons.

The interviews were conducted by a female final-year Master’s degree student in forensic psychology. To minimize possible interviewer influences or biases, the interviewer was trained by one of the authors (KV) via role-play and supervision by listening to the audiotapes and providing feedback. Furthermore, the interviewer had to follow a detailed manual. All interviews were individually conducted in a quiet environment at the institutions and only the interviewer and the participant were present. The interviews lasted at average 26 minutes (ranging from 17 to 42 minutes). All interviews were recorded with a voice recorder and notes were made during the interviews. The participants did not know the interviewer beforehand, she introduced herself briefly at the beginning of the interview and indicating her involvement was part of her Master’s degree work

Materials

Beforehand a semi-structured topic list was drafted by the authors and pilot tested, covering topics such as 1) what NVR-MID had done in work climate 2) in group climate for the admitted juveniles, 3) how work- and group climate could be maintained and finally 4) what they thought about the NVR-MID training trajectory. The questions asked participants about their impressions of NVR-MID; any changes they noticed since implementing NVR-MID in either their working environment or the group climate and which elements of the NVR-MID approach might have contributed to these changes; and finally what would be needed to sustain NVR-MID in the institution.

Data Analysis

After being transcribed verbatim, the interviews were uploaded into ATLAS.ti, a software program for qualitative analysis. Then, they were coded three times according to the qualitative coding model as described in Boeije (2005) by two researchers separately. The first step is called open coding wherein all interviews were read repeatedly by the researchers and all information was sorted and coded into different paragraphs and fragments (e.g., relationship NVR-MID and work climate, interventions NVR-MID). In the second step called axial coding, the fragments and their corresponding codes were again sorted into different categories. This resulted in the following categories: 1) Client-related NVR-MID, 2) Feedback on the Quantitative Study, 3) Implementation NVR-MID, 4) Interventions NVR-MID, 5) Maintenance NVR-MID, 6) Negative NVR-MID 7) Positive NVR-MID, 8) Training NVR-MID and 9) Work Climate NVR-MID. The third and last step of the analysis is called sensitive coding and consists of finding new categories, through defining and redefining, sorting the topics into new ones, and finally forming new categories/codes or merging existing categories. As for this study, only the part on work climate is applicable, we will describe the categories which are related on work climate. These final categories were: (1) Work Climate/Team enhancing features of NVR-MID, (2) Challenges implementing NVR-MID.

Results

Quantitative Part

Means and standard deviations for the scores on the seven subscales are given in Table 3 and the corresponding line graphs can be found in Figure 1. No changes over time were identified for either positive or negative team functioning, task significance, perceived workload, shared vision, or commitment. However, the subscale job satisfaction showed a significant decrease over time (F(2, 112.17) = 4.44, p = .014).

Table 3. Test of fixed Effects, Means and Standard Deviations on Work Climate for the Whole Sample Before, During and After NVR-MID Training

Measurement point	Before	During	After	Test of fixed Effects
	M (SD)	M (SD)	M (SD)	F, p
N completed questionnaires	115	100	64
Teamfunctioning Positive Teamfunctioning	3.77 (0.56)	3.74 (0.61)	3.72 (0.62)	F = 0.55, p = .57
Negative Teamfunctioning Work motivation Job Satisfaction	2.74 (0.61) 4.09 (0.52)	2.70 (0.61) 4.02 (0.49)	2.75 (0.69) 3.89 (0.55)	F = 1.29, p = .28 F = 4.44, p = .01
Task Significance Perceived Workload Shared Vison & Commitment Shared Vision Commitment	4.32 (0.45) 3.56 (0.59) 3.92 (0.62) 3.34 (0.59)	4.24 (0.51) 3.46 (0.58) 3.90 (0.60) 3.20 (0.54)	4.19 (0.43) 3.55 (0.53) 3.81 (0.65) 3.19 (0.57)	F = 1.37, p = .25 F = 0.52, p = .59 F = 0.74, p = .47 F = 2.06, p = .13

All analyses were adjusted for significant influences of inter-correlation of repeated measures and clustering on the unit.

Figure 1. Work climate scores per site before, during and after NVR-MID implementation.

Qualitative Part

Participants from the three residential settings were interviewed, with saturation of information after 15 interviews (Boeije, 2005); no new themes emerged from participants’ narratives in subsequent interviews. The aim of the qualitative part of the study was to explore the judgment of staff on work climate in a qualitative manner, extending insights. Two themes were identified from the qualitative comments. The first theme, Work and Team climate effects, included three subcategories: Stability in the teams, Stability on the wards and Increase in team spirit. The second theme, Implementation challenges of NVR-MID, had two subcategories: Mismatch between method and staff and Timing of Implementation. Categories were ranked into an higher order and will be presented as such, with the main goal answering the question on how work climate is possibly influenced through training of staff in NVR-MID. The two core themes will be discussed in more detail and illustrated by original quotes, below.

Work Climate Enhancing Features of NVR-MID

When discussing with staff how NVR-MID could have influenced their work climate, they mentioned three main subcategories that were most important, namely stability in the teams, stability on the wards and increasing of the team spirit. We will discuss these domains in more detail below.

Stability in the Teams

The staff mentioned that NVR-MID succeeded in stabilizing their teams of group workers, after they had been trained. Interestingly, this was seen as well in relatively recently renewed teams as in well-established teams. “It was an unstable team, with a lot of problem behavior and the coming and going of supervisors and temporary workers. They got a new treatment manager and with him, also NVR-MID was implemented. That together has led the ward to a very good and stable situation. Through that, job satisfaction has improved among employees.” [p.n. 527]

“ … on our ward there was already a lot of solidarity, that has grown even stronger because of NVR-MID. And I think it also has improved the job satisfaction.” [p.n. 527]

Stability on the Wards

Besides the role teams play in the training of NVR-MID, it was also mentioned that the fact that there were less escalations on the wards also influenced the work climate in a positive way. “Yeah, less escalation, of course, improves the work here (…) and sometimes the atmosphere on the group has become a bit more relaxed.” [p.n. 705]

Increase in Team Spirit

The most important aspect that was recalled by staff when dealing with the improvement of the work climate, they said that NVR-MID succeeded in enlarging the team spirit feeling on the ward. Colleagues were working side by side again, instead of pointing out the mistakes they made.

“Yeah, well, the ‘we’ feeling, the more team feeling. We have more things to talk about (.) We experience more that you support each other as a team during your work.” [p.n. 519]

“Instead of one person saying ‘we’re going to do it like this’, the whole team is involved, so you’ll have more capacity in your team to pursue.” [p.n. 596]

Furthermore, staff pointed out that they could handle more stressors on the ward, could handle more difficult situations due to the enhanced team feeling through NVR-MID implementation.

“ … that you’re stronger, that you can tolerate more. Yes, it (NVR-MID) does that with the team.” [p.n. 570]

“ … that in an escalation, it is okay to hand it over to your colleague. That he can handle it too.” [p.n. 589]

One of the interventions of NVR-MID is the belief “We, instead of I alone.” This means that staff members are asked to talk in the “We” when they address the juveniles. For example, they are asked to say “we do not accept that you are late for dinner,” instead of ”I want you to do this or that”. This results in that the team is feeling more as a team, and also the juveniles feel that there is more interpersonal connection.

“I feel more supported by the NVR-MID, because you are one with your team in certain situations. Also because you can talk in the ‘we’ form, we as a team, we from the (name ward). It has also changed, that we have all started to look more closely at what we find acceptable and unacceptable and how are we are going to react to behavior.” [p.n. 527]

Challenges When Implementing NVR-MID

When discussing the matter of implementing NVR-MID into the teams, staff also mentioned points of attention or difficulties within the implementation process, which should be kept in mind when implementing the method. They mentioned the themes mismatch between the method and staff and timing of implementation. We will illustrate these themes below.

Mismatch between Method and Staff

On behalf of work climate and its influence by NVR-MID implementation, staff mentioned that there should be a match between the method and staff. More specifically, they mentioned that younger or more inexperienced staff had problems with embedding the new method into their work. They forgot to continue doing what they had done before NVR-MID came into sight. That led to a lot frustration and agitation on the ward.

“What I do notice is that (.) it (NVR-MID) has brought good things, but it has also caused turmoil, and uncertainty. And, look, I do have quite a bit of work experience, but for people who are still learning, setting limits, learning how to behave on such a group (…) then there’s just nothing more to choose, then it is just the survival mode (.) it has also caused some commotion, so to speak.” [p.n. 701]

Timing of Implementation

Even as important as the match between method and staff, is the right timing of implementation of NVR-MID. It is not advisable to train teams when they are in the middle of a reorganization or when there is no basic method from where teams work. NVR-MID is not a replacement for a basic work method, it is seen more as an extra tool, which can be used when needed.

“Because we’re still messing around with our staff like crazy. Staff who are insufficiently trained in a basic work method. And then NVR-MID came on top of that. So I’m much more concerned with how we have left the staff. We don’t have a permanent team. Timing could have been better, we actually started applying NVR-MID when it became too much for us.” [p.n. 534]

“Because you know, most things from NVR-MID are so valuable. But we are not capable enough, or the teams are not capable enough to put these good facets on top of the good things we already did before NVR-MID. Instead, it is going to take the place of something else and that ultimately leads to confusion. So maybe we would have needed more intervision, but that’s just not feasible. [p.n. 573]

Also, it is important that staff realizes that NVR-MID implementation is not a quick fix, but that it takes time to implement the method decently in order to have positive influence on the work climate.

“Yes, NVR-MID might be a bit of a cultural change, too. That takes time … ” [p.n. 527]

Discussion

The main aim of this study was to assess the influence of residential NVR-MID training on the perceived work climate of staff members in residential settings for juveniles with MID in the Netherlands. Work climate was measured in three residential settings during a period of 18 months quantitatively with the Living Group Working Climate Inventory (LGCWI) and qualitatively with a semi-structured interview with a part of the trained staff.

Quantitative Part

Quantitative results showed that most subscales of work climate: positive and negative team functioning, task significance, perceived workload, shared vision, and commitment, did not change significantly over time. Only one subscale, job satisfaction, did change, and surprisingly decreased over time. As such, we could not confirm our hypothesis that NVR-MID improves work climate. Previous research by Van Gink et al. (2018) was also inconclusive. Only at one site, where situation in the team in terms of staff turnover was stable, work climate improved after implementation of the regular version of the residential NVR (thus not specifically adapted for MID). On two other sites however, results showed a worsening in work climate over time, and in one of those sites job satisfaction decreased after implementation of NVR. The authors argue that sites where there was more stability in teams, NVR could be implemented more properly and could therefore positively influence the work climate. In the present study, one site (site 1) was more stable in terms of experienced team members compared to the others. When studying the graphs (Figure 1) it is evident that only at site 1 all work climate scales tend to develop in the expected, more positive, direction, although not significantly. This also supports earlier findings of Van Gink et al. (2018) that stability in the teams and in the institution enhances a successful implementation of the method and may eventually have an influence on work climate.

Qualitative Part

The qualitative results revealed that staff members described the influence of the NVR-MID training on work climate as positive. Generally, staff members reported NVR-MID training increased stability in teams and on wards and lifted the team spirit. For example, they accounted the increasing “We-feeling” as a result from the implementation of NVR-MID. Research suggests that cohesion is greater in teams where there is such a We-feeling (Kinicki & Kreitner, 2006). This cohesion may lead to a more positive work climate. Nevertheless, the interviewees also identified some complex aspects of NVR-MID. Such as, when implementing NVR-MID it is crucial to pay attention to the match between the method and staff as it seemed that for example younger or more inexperienced staff has problems with embedding the new method into their work and furthermore, the right timing for implementation; teams should not be trained in in the middle of a reorganization or when there is no basic method which teams work with, as NVR-MID is not a replacement for a basic work method, it is seen more as an extra tool, which can be used when needed.

Integrated Findings

As such, quantitatively we observed that NVR-MID did not influence the work climate positively, while qualitatively we díd observe such an effect. We will discuss this contrasting findings in detail.

A first possible explanation is that the questionnaire LGWCI had a broader scope and was therefore not specific enough to measure the aspects of work climate that were influenced by NVR-MID, as many items may have been also influenced by other contextual factors such as reorganizations in management within the institutions or changed of staff within teams. When we study our data, it is striking that only at institution 1 (which was the most stable one in all aspects) an increase in job satisfaction was seen, this would support this explanation. Earlier research has shown that these contextual factors are thought to hinder proper successful implementation of new methods (Woltmann et al., 2008).

Additionally, the questionnaire is long and consists of 147 items, some of them seemed not always to be very straight-forward and easy understandable, as some interviewees staged. Furthermore, some interviewees indicated that due to the agitation about having to fill in the long questionnaire every three months, this could have led to low motivation and reporting bias. In retrospect, we could have used a shorter instrument, especially because our study setup consisted of multiple measurements every three months. Additionally, with the knowledge we have now that NVR-MID especially has an influence on the team climate, we would recommend an instrument with specific focus on team climate (f.i. the Team Climate Inventory – short version,(Strating & Nieboer, 2009)). On the other hand, the LGWCI is a widely used instrument in the residential settings in the Netherlands (also in the ones that participated in our study). The developers of the LGWCI are currently working on a renewed version of the questionnaire (personal communication with the developers).

A different explanation for these findings could be the following: NVR is generally speaking seen as a method, which can be very different from other commonly used methods. Research has shown that implementing a new method can arise concern, turmoil, and irritation in the beginning (Durlak & DuPre, 2008). As NVR is seen as a method that changes the culture as a whole on the ward, it is imaginable that the change initiated by NVR would occur gradually. That would mean that it possibly cannot be detected through an ordinary questionnaire, or only on the long run. However, other researchers did find positive changes using the same questionnaire (Van Gink et al., 2018). When we consult literature on the theory of how change works in residential settings, we conclude that it takes time and effort to implement a method properly (Durlak & DuPre, 2008). And that sometimes effects are contrary from what is expected, due to the fact that staff has to find its way in working again and is startled by the newly introduced method (Durlak & DuPre, 2008). That would mean that effects can only be visible after a certain time. Future research should investigate group climate after a longer period of follow-up.

Possibly it is too soon to draw solid conclusions on the effect of NVR-MID on the work climate, but as there is almost no literature on this issue, it was very important to make a first attempt to objectify the possible effects of NVR-MID on work climate. Other research that studied the effect of NVR-MID on group climate (deleted due to blinding reasons) and aggressive incidents (deleted due to blinding reasons) showed a positive relationship with the implementation of NVR-MID (e.g., group climate measured by juveniles increases, aggressive incidents decrease). The developers of the group- and work climate argue that group climate is intertwined with work climate, so it could be suggested that because group climate does increase and staff is positive on the change of work climate in the interviews, NVR-MID actually does have a positive influence work climate. Either way, it might be evident that there is more research needed in order to support these assumptions.

This study investigated the relationship of the implementation of NVR-MID and work climate in the residential setting in a quantitative and qualitative manner. Although the quantitative questionnaires showed no change, or even negative change, the qualitative data derived from interviews, showed a preliminary effect on team stability, stability on the wards and team spirit, all being elements connected with a more positively perceived work climate (Roy et al., 2020; Van Gink et al., 2018). However, staff also indicated some important points of attention; it was crucial to pay attention to a good match between staff and method (e.g., professionals should not be too young nor too inexperienced) and the timing of the implementation (e.g., not before nor during a reorganization). On the basis of these findings, we have preliminary evidence that NVR-MID may have positive influence on the work climate, provided that the aspects we have just described are kept in mind. As work climate is a key element in a healthy environment in terms of less staff-turnover, burn-outs and sick-leave, it is very important to find methods that increase work climate – NVR-MID could be one of those methods.

Acknowledgements

We would like to thank all juveniles, staff and group workers who participated in this study.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Appendix A

The NVR–MID method

NVR-MID focuses on improving the relation between-group workers and juveniles instead of only focusing on changing the unacceptable behavior of the juveniles, by training the professional how to change her or his own behavior. NVR-MID is a systematic approach to help caregivers, as parents, teachers, and staff of mental health-care facilities, cope with juveniles’ problematic behavior by restoring the troubled relationship between juvenile and caretaker. Hereby primarily focusing on changing the behavior of the caretaker instead of the behavior of the juvenile. Central tenets of NVR-MID are presence, de-escalation and self-control, (re)building relationship, support and active resistance. A more detailed explanation of the method is presented in the methods section of this article or can be found in (Omer, 2004b) or Omer and Lebowitz (2016). These key aspects translate into a few viewpoints, such as “the team must take a stand against unacceptable behaviour,” “it is an illusion to think that I can control the other,” “ we work together as a team, with juveniles and their parents.” These viewpoints together form one of the parts of the NVR-MIDa method, the thinking part. The other part of the method consists of the more active doing part, the NVR-MID interventions. Examples of the NVR-MID interventions, in order to resist unacceptable behavior, are:

The announcement

The announcement is a written statement to the juvenile about what kind of behavior is no longer accepted. Besides that, it also tells the juvenile which alternative behavior is accepted and that the group will support the juvenile in achieving that. The fundamental tone of the letter is unmistakable but friendly and warm. The announcement is usually read to the juvenile by one of the group workers in the presence of other caregivers, such as parents, teachers, and/or friends.

The sit-in

When the unacceptable behavior discussed in the announcement is not decreasing, the group workers can decide to do a sit-in. During the sit-in, the group worker and sometimes other caregivers will enter the room of the juvenile in a calm and peaceful manner. One of them will do the talking, the others will remain silent the entire time of the sit-in. The aim of the sit-in is to make it unmistakable for the juvenile to understand that the shown behavior is unacceptable and that it has to change. Furthermore, it is the purpose of the sit-in that the juvenile takes responsibility for his behavior and suggests a solution how to change this behavior in the future.

The reparation act

A reparation act is done to repair a situation that went wrong. For example: an adolescent breaks a window deliberately. This situation can be discussed afterward and the group workers ask the juvenile what he or she could do to show that he or she is sorry. Within the reparation act it is important to keep in mind that every solution made by the juvenile is accepted even when it is not in proportion to the value of the damage that has been done, for example, the juvenile could suggest to bake a cake for the group to say sorry for the broken window. The only exception is repeating a suggestion that did not work before