Interprofessional teamwork in decentralized child welfare in The Netherlands: A comparison between the cities of Amsterdam and Utrecht

ABSTRACT

The recent transformation of child welfare in the Netherlands has improved opportunities for interprofessional working. We compared two models of teamworking within newly established interprofessional teams in the cities of Amsterdam and Utrecht, conducting a secondary analysis of semi-structured interviews collected through three broader research projects. Respondents include seventeen interprofessional team members (six from Utrecht, eleven from Amsterdam), representing a variety of teams across city, as well as two policymakers from Utrecht and one from Amsterdam. Team members were approached using convenience sampling, policymakers were purposively recruited. In different rounds of open and focused coding, we found that differences in team organization between the two cities have led to differences in the quality of interprofessional teamworking. Teamworking is best developed in Utrecht partly because team members are recruited and employed by a single organization. This has enabled a more careful process of selection and team composition than in Amsterdam, where a delegation approach entailed fragmentation as well as the risk of divided loyalty between team and mother organization. In addition, while the development of interprofessional teamwork in Utrecht is served by certain structures, teams in Amsterdam have suffered from an imbalance between freedom and structure, causing insecurity amongst staff and reduced chances of interprofessional integration. Despite the apparent success of the Utrecht model of interprofessional teamworking, interprofessional collaboration across team boundaries might suffer from the fact that teams in Utrecht, unlike in Amsterdam, do not comprise representatives of relevant partner organizations.

KEYWORDS:

• Child protection
• child welfare
• decentralization
• Interprofessional collaboration
• interprofessional learning
• teamwork

Introduction

Poor interprofessional and inter-organizational coordination in child welfare and protection has been a recurrent finding in reviews of child death cases and child welfare systems in different countries (e.g. Kuijvenhoven & Kortleven, 2010; Munro, 2011; Sidebotham, 2012). In the Netherlands, this was one of the reasons for a major transformation of the child welfare system. The Youth Act, which came into force on 1st January 2015, removed bureaucratic and sectoral barriers to interprofessionalism by loosening legal constraints, merging budgets and making local governments responsible for all types of child welfare, ranging from parenting support to child mental health care and child protection. Local governments have developed new child welfare structures that are meant to advance interprofessional working in ways tailored to the local situation. In most if not all municipalities, a key role has been given to newly established interprofessional teams for initial assessment of children’s and families’ problems, support, and coordination of care provision.

In order to understand how these local interprofessional teams have functioned thus far and to what extent they have fostered effective interprofessional teamworking, we compared two models of teamworking in the largest and fourth-largest cities of the Netherlands, Amsterdam and Utrecht. These cities, which have built a reputation as pioneers in the development of interprofessional teams, have adopted distinct approaches to team organization and composition, scope of tasks and modes of operation. We investigated whether the differences in approach have consequences for the quality of teamworking.

Background

Interprofessional teams in the cities of Amsterdam and Utrecht consist mainly (in Amsterdam) or exclusively (in Utrecht) of so-called generalist professionals. The generalist role, whose development is considered critical to the transformation of child welfare, is carried out by professionals with different professional and disciplinary backgrounds who have also gained (or are still gaining) basic knowledge of other professions and disciplines. Such broad expertise should enable them to make an integral assessment of needs, provide support, and continue to coordinate welfare provision thereafter.

Generalists may need to consult with specialized professionals when they lack certain expertise among themselves. To this end, interprofessional teams in Amsterdam also comprise specialists (such as a child psychologist) and representatives of specific agencies (e.g. child health care and a child protection agency). In Utrecht, specialist expertise must be sought outside the teams. In addition, team organization differs between these cities in the sense that teams in Amsterdam are divided into two tiers, with first-tier teams responsible for initial assessment and basic support and second-tier teams for complex, multi-problem cases, whereas Utrecht has a single-tier system.

A final, and crucial, difference concerns the way teams are staffed. Teams in Utrecht are embedded in a single organization, which recruits and employs the team members, in order to prevent professionals from experiencing divided loyalty between their team and a mother organization. Teams in Amsterdam are staffed according to a more traditional delegation model, in which team members remain formally employed by their mother organizations.

Methods

This study undertook a secondary analysis of part of the data from three research projects. One project, carried out by the first author, focused on improvisation practices associated with the decentralization of child welfare in the Netherlands. The other two projects concerned the master’s thesis research of the second and third authors, supervised by the first author. The second author investigated in what way the decentralization of child welfare changed interprofessional collaboration in Amsterdam and Utrecht, both within and across the boundaries of interprofessional teams. The third author studied how the generalist role within interprofessional teams in Amsterdam was being developed.

Data collection

The secondary analysis was conducted on nineteen semi-structured interviews, in which three policymakers and seventeen members of interprofessional teams partook. The small group of policymakers consisted of the municipal program manager responsible for the transition of child welfare in Amsterdam as well as the municipal cabinet’s portfolio holder for child welfare in Utrecht, who was interviewed together with a municipal policy officer. They were purposively recruited because of their leading role in the organization of decentralized child welfare and were interviewed by the first author in September 2013, during the preparation of the decentralization. An appointment to also interview the Amsterdam portfolio holder for child welfare was twice rescheduled, then cancelled.

The group of seventeen professionals comprised six generalists from Utrecht, ten generalists from Amsterdam (two of them based in first-tier teams, eight based in second-tier teams), and one team leader (first-tier team) from Amsterdam. All professionals were recruited by convenience sampling and interviewed by the second and third authors in May–July 2016. Professionals were approached using the authors’ networks and snowballing techniques, and by e-mailing interprofessional teams with an invitation to participate. Those who agreed to participate represent a variety of teams across city. All interviews were audio-recorded and transcribed verbatim. Table 1 provides an overview of the respondents.

Table 1. Overview of respondents.

City	Code	Function	Professional and organizational background
Amsterdam	R1	Municipal program manager transition child welfare	N/A
	R2	Team leader 1st-tier team	Clinical child psychologist at education consultancy firm; now employed by municipality
	R3	Generalist 1st-tier team	Investigator & behavioral expert at (statutory) Child Protection Board; now delegated by local welfare and social work organization (A)
	R4	Generalist 1st-tier team	Parent counselor; delegated by local child welfare organization (B)
	R5	Generalist 2nd-tier team	Family worker; delegated by (and working part time at) supraregional organization treating children & youth with complex behavioral problems (C)
	R6	Generalist 2nd-tier team	Family worker; delegated by regional child welfare & parenting support organization (D)
	R7	Generalist 2nd-tier team	In-home counselor specializing in mental health & addiction problems; delegated by (and working part time at) Protestant Christian social work organization (E)
	R8	Generalist 2nd-tier team	Social worker & autism specialist; delegated by national welfare organization supporting people with disabilities (F)
	R9	Generalist 2nd-tier team	General social worker; delegated by (and working part time at) local welfare and social work organization (G)
	R10	Generalist 2nd-tier team	Adult social worker; delegated by organization F
	R11	Generalist 2nd-tier team	Social worker; delegated by organization F
	R12	Generalist 2nd-tier team	Social worker & trainer; delegated by organization F
Utrecht	R13	Municipal cabinet’s portfolio holder	N/A
	R14	Municipal policy officer	N/A
	R15	Generalist	Youth/school social worker; formerly employed by organization F, a local child welfare & parenting support organization (H) and other welfare organizations
	R16	Generalist	Clinical child psychologist & foster care employee; formerly employed by regional child welfare & parenting support organization (I)
	R17	Generalist	Youth probation officer & child protection social worker; formerly employed by regional child protection agency (J)
	R18	Generalist	Child/school social worker; formerly employed by organization H and other welfare organizations
	R19	Generalist	Expertise in child protection; organizational background unknown
	R20	Generalist	(Forensic) social worker & systems therapist; formerly employed by supraregional child welfare & parenting support organization (K) and forensic institutions

The selection of these interviews from the wider data set we had available was based on two decisions. First, we decided to focus on the only two cities in which we interviewed both policymakers and professionals, so as to be able to relate policy choices to professional experiences. Consequently, we excluded interviews with policymakers from other municipalities obtained in the first research project, since no professionals were interviewed in those municipalities. Second, given the word limit imposed on short reports, we decided to limit the scope of this report to interprofessional teamworking and to not consider interprofessional collaboration across team boundaries. We therefore left out from the analysis several interviews with other stakeholders and child welfare professionals outside the interprofessional teams.

Data analysis

The secondary analysis of the selected interviews, conducted by the first author, started with a round of open coding of the interviews with professionals. This brought to the attention the possible relationship between differences in team organization and the quality of teamworking. In subsequent rounds of focused coding, this relationship was further investigated, disentangled, and found sufficiently confirmed. While coding the interviews with professionals, we also encountered apparently non-organizational factors influencing the quality of teamworking, most notably personal characteristics of team members. However, on second thought, it turned out that the role personal characteristics play could be traced back to aspects of team organization as well, since the way of selecting team members determines in large part which personalities a team is composed of. The interviews with policymakers were predominantly used to outline the organization of interprofessional teams in both cities, putting the experiences of professionals into context.

Ethical considerations

All respondents agreed to be interviewed for scientific purposes and consented to audio-recording the interviews. During one interview, audio-recording was temporarily suspended due to the sensitivity of details discussed. Data have been sufficiently anonymized to prevent identity disclosure. As identification could not be entirely excluded in the case of policymakers, sensitive quotations have been avoided. The Ethical Review Board of the Faculty of Social Sciences, Vrije Universiteit Amsterdam (Reference: ERB/17-08-01, declared that this study complies with the ethical guidelines of the faculty.

Results

Table 2 (see online supplementary file) provides a selection of quotes that illustrate and support the findings presented in the results section. All quotes have been translated from Dutch into English by the first author. The results suggest that differences in team organization between Amsterdam and Utrecht have led to differences in the quality of interprofessional teamworking. Whereas in Utrecht five of the six interviewed generalist professionals indicated they felt safe and happy with interprofessional relations within their team (R15, R16, R17, R18, R20 Theme 1), respondents from Amsterdam expressed more negativity towards team dynamics (four out of ten generalists: R7, R8, R11, R12 Theme 1), and those expressing more positivity were less unequivocal than in Utrecht. Amsterdam respondents who said team collaboration was fine often added this was despite some adverse circumstance like high staff turnover, or that things were worse in the near history, or in other teams (R4, R5 Theme 1). In explaining these differences, two factors seem especially relevant.

First, a single team organization recruiting and employing team members, the Utrecht model, apparently provides stronger incentives and safeguards for teamworking than the Amsterdam model, with team members delegated by different organizations. The Utrecht model enabled a careful process of selection and team composition, increasing the likelihood of capable and motivated team members as well as adequate teamworking (R15, R16, R18 Theme 2).

Due to the involvement of different organizations, the selection process in Amsterdam has been fragmented and less careful (R2, R4 Theme 2). As a consequence, team composition and dynamics have been largely left to chance (R8, R11 Theme 2) and professionals have been allowed to become team members without thorough consideration or even with some reluctance, as the decision to delegate them to a team has not always been purely their own (R6 Theme 2 & Theme 5). Together with the possibility of returning to one’s mother organization (R7 Theme 3), this seems to explain part of the high staff turnover that stands out as an issue in the interviews with Amsterdam team members (R5, R7, R11 Theme 1).

The delegation model also complicates identification with a team. Whereas in Utrecht team spirit turned out to be generally well-developed (R15, R16, R17, R18, R20 Theme 1; R17, R18 Theme 3) and differences in background and perspective were predominantly seen as complementary (R15, R16, R18 Theme 2; R16, R17, R18, R20 Theme 4), many respondents in Amsterdam defined themselves and others as representatives of an organization and profession rather than close colleagues in the same team (R2, R5, R7, R8, R12 Theme 3), and perceived different perspectives more often as conflicting (R11, R12 Theme 4). Typically, one’s own mother organization (some professionals even continued to work there part-time, cf. R5 Theme 3) was contrasted positively with other organizations and was valued by various respondents as a place where one may temporarily retreat from interprofessional team dynamics (R5, R7, R8 Theme 3; R12 Theme 4). Nonetheless, some Amsterdam respondents reported they primarily identified with their team, showing that a continued connection with a mother organization, if plainly formal, need not be an obstacle (R4, R10 Theme 3).

Second, the quality of interprofessional teamworking relates to the balance between freedom and structure in the organization of the teams. Several respondents from Amsterdam said a lack of guidelines made them feel quite insecure about how to deal with team collaboration and generalist working (R4, R6 Theme 5). This was mentioned as another cause of the high staff turnover. In Utrecht, professional freedom has been structured more strongly right from the beginning, reducing the uncertainty associated with interprofessionalism to a manageable level (R16, R17 Theme 5).

Illustrative is the practice of working in pairs. In Utrecht, cases are allocated to varying pairs of generalists, depending on which expertise is required. This practice appears to be a crucial mechanism for interprofessional integration and the development of the generalist approach. It facilitates approaching cases from more than one professional perspective and enables interprofessional learning (R15, R16, R17 Theme 6). Working in pairs is sometimes also practised in the Amsterdam teams, but not in a consistent way. Respondents referred to it either as an option or as a rule, which is often deviated from under time pressure (R6, R8, R9 Theme 6). Those regularly working in pairs seemed to have the freedom to work often or always with the same preferred colleague(s) (R9 Theme 6). Thus, chances of interprofessional integration have been missed, leading some Amsterdam respondents to doubt the possibility of becoming a real generalist. They felt one should expect a generalist approach only from the team as a collective, with team members invoking each other’s expertise rather than learning from each other (R6 Theme 6).

Discussion

This study lends support to several of Hudson’s (2002) optimistic hypotheses on interprofessionality, particularly his third hypothesis: “socialisation to an immediate work group can override professional or hierarchical differences amongst staff” (p. 16). Nearly all respondents from the city of Utrecht and some respondents from Amsterdam reported they identified with their interprofessional team, considering professional differences an asset rather than an obstacle. Our findings suggest that the chances of such socialization, and thus the quality of interprofessional teamworking, may be significantly increased by embedding interprofessional teams in a single organization. Such a radical way of eliminating inter-organizational barriers to interprofessional collaboration, evading classification in terms of inter-organizational integration (Willumsen, 2008), was shown in the Utrecht case to support a high level of interpersonal integration (Willumsen, 2008), with professionals experiencing team psychological safety (O’Leary, 2016), having positive perceptions of each other (Widmark, Sandahl, Piuva, & Bergman, 2016), and being able to align different perspectives (Rowland, 2017).

This approach, however, may not be feasible or desirable in most care settings. Even in the city of Utrecht, there could be a trade-off, as interprofessional teams still have to collaborate with other organizations and professionals, like care providers, child protection agencies, and physicians. Such inter-organizational boundary-spanning might suffer from the fact that teams in Utrecht, unlike in Amsterdam, do not comprise representatives of relevant partner organizations. That could be a reason to hesitate copying the Utrecht model and first explore the possibilities for improvement within a delegation model. Our findings indicate that such possibilities lie for instance in a consistent practice of working in (varying) pairs.

Since this report is written relatively shortly after the decentralization of child welfare in the Netherlands and the underlying research is rather limited, both in terms of sample size and the number of municipalities included, the findings reported have a tentative character. Moreover, it must be noted that our account of interprofessional teamworking is solely based on the stories of the team members themselves. These stories were collected during two research projects with partly different foci, using different interview guides, which may have led to more than usual variation in the level of attention paid to certain topics. Further research would be necessary to trace the direction of ongoing developments and to gain more robust knowledge of interprofessional teamworking in a broader set of municipalities, which should also take into account the perspectives of stakeholders outside interprofessional teams. In addition, attention should be given to the question to what extent the introduction of interprofessional teamworking translates into better service delivery to children and families.

Declaration of interest

The authors report no conflicts of interest. The authors alone are responsible for the content and writing of this article.

Supplemental data

Supplemental data for this article can be access on the publisher’s website.

Additional information

Notes on contributors

Willem J. Kortleven

Willem J. Kortleven is an assistant professor of Public Administration at the Vrije Universiteit Amsterdam. He received a PhD in Socio-Legal Studies from Erasmus School of Law, Erasmus Universiteit Rotterdam.

Shelita Lala

Shelita Lala is a legal counsel with Dutch trade union FNV. She graduated in Human Resources Management (BSc) from the University of Applied Sciences Utrecht and Public Administration (MSc) from the Vrije Universiteit Amsterdam.

Youssra Lotfi

Youssra Lotfi is a trainee with the Amsterdam Metropolitan Area. She graduated in Political Science (BSc) and Public Administration (MSc) from the Vrije Universiteit Amsterdam.