Who, what, where? The influence of the Integral Care Agreement (IZA) on established regional networks in Dutch healthcare

For over a decade, government initiatives have encouraged organizations in healthcare and welfare to collaborate in cross-domain networks. In the Netherlands, nine experimental regions were designated in 2013 to monitor the development of such collaborative efforts [1], and in 2018, the movement ‘The Right Care at the Right Place’ (De Juiste Zorg op de Juiste Plek) was introduced, emphasizing collaboration as a prerequisite for quality care [2]. These networks involve various organizations such as health insurance companies, healthcare and welfare organizations, municipalities, and patient advocacy groups. They are often perceived as solutions to grand challenges, such as long waiting lists and high healthcare costs, to keep healthcare systems sustainable [3‐5]. Recently, the Integral Care Agreement (Integraal ZorgAkkoord, IZA) 2022 has emphasized such collaborative efforts in healthcare administrative regions [6]. While many networks predated the IZA, they often operated on a smaller scale than that of the healthcare administrative region, and questions regarding the existence of a definitive region to address these challenges have been raised previously [7]. Therefore, it is crucial to investigate the influence of the IZA, particularly the designation of collaboration in the healthcare administrative region, on existing regional networks.

The scientific literature has shown that numerous determinants are able to influence the effectiveness of networks [8]. Changes in the internal environment of networks can impact a network’s functioning, such as alterations in network composition or diminished commitment of network participants [9, 10]. Additionally, research has indicated that disparities in the catchment areas of network organizations can pose challenges for defining network boundaries and goals [11]. Furthermore, various differences between network organizations, such as the level of trust [12], perceived differences in influence within the network [13], and the degree of openness and transparency in decision-making [14], can affect network effectiveness. Changes in the external environment, such as policy changes (e.g., implementation of the IZA), can either facilitate or impede network goals, depending on how well they align with existing roles and responsibilities in practice [15]. However, the specific impact of the IZA as a policy change on existing networks and the dynamics between their members remains uncertain.

Therefore, the research question of this study is: ‘What influence does the IZA have on existing regional networks?’ Answering this question is important for several reasons. First, it can assist networks and participating organizations in coping with the changes brought about by the IZA. Conversely, it can contribute to preparing government policies, and specifically, ensuring that any further developments of the IZA align well with practical needs and support existing and developing networks in practice. Finally, it can help fill gaps in our knowledge regarding network effectiveness and the role of policy in creating effective networks, thereby stimulating targeted future research.

This study is part of a longitudinal qualitative case study focusing on a network in the Netherlands, spanning from April 2019 through September 2023. The region in which the network operates is characterized by a high degree of collaboration, and it was one of the nine national experimental regions in 2013 [16]. The network includes the major providers of primary and secondary healthcare, mental healthcare, support services, home care, youth care, and public health, as well as the dominant health insurance company, the largest municipality, and a patient advocacy group. The network is divided into four domains, each with its own focus area where numerous projects are implemented. The region in which the network operates is bound to several municipalities and is characterized by an aging population, high healthcare costs, low perceived health status, and low socioeconomic status [17, 18]. Figure 1 illustrates the intricate relationships between the organizations involved in the network’s initiatives. Data were collected during the project in which the four chairs of the four domains provided documents and were interviewed to obtain a comprehensive understanding of the network’s relationships. The data were then analyzed using UCINET.

Within the subregion, the network is seen as the overarching network in which the region’s direction is discussed and monitored. Given the network’s existence for over a decade and its active engagement in the region, this case is particularly suitable for studying the influence of the IZA because it allows us to compare the situation before and after its implementation. Moreover, the longitudinal research period allows us to observe the pre- and post-IZA situation instead of relying solely on retrospective assessments. In this study, ‘subregional’ refers to the region in which the existing network operates, including the four domains and initiatives. This subregion is one of two subregions within the healthcare administrative region. The study has been approved by the university’s ethics committee (anonymized).

We present the results in two main categories related to the influence of the IZA on the regional network: changes in the role, purpose, and composition of the network; and changes in the dynamics between organizations in the network.

The development of a new network introduced after the IZA at the healthcare administrative regional level adds an additional layer of governance. The literature indicates that networks are more challenging to be governed than other organizational forms. They typically involve multiple layers of governance [5], rely more on their members (i.e., organizations), and are less able to resolve conflicts [20]. Moreover, there needs to be a balance between broad representation of participants while simultaneously maintaining a network that is small and agile enough to be effective [21, 22]. There is limited understanding of how to govern such an additional network layer, often referred to as a ‘network of networks’ [20], but one can anticipate that these governance challenges would be exacerbated. This raises questions about the manageability of a network of networks in practice, both from the perspective of the network itself (e.g., what is the right balance between representation and effectiveness?) and that of organizations in the region (e.g., how do organizations manage their participation in these networks, both when operating in a single subregion and when operating across multiple healthcare administrative regions?). Since there is still limited literature on governing such networks of networks [5], it is important to continue monitoring and researching these developments. Policy should align more closely with existing network structures and initiatives in practice, especially considering a potential IZA 2.0 [23], to facilitate collaboration and avoid hindering further development [24].

Our finding that primarily organizations with a larger catchment area and/or higher market power are made central to the development of regional IZA plans may have implications for the way the IZA is implemented in practice. The literature suggests that larger organizations are often less innovative than smaller ones [25], while smaller organizations are often better at implementing change [26]. Consequently, given IZA’s focus on innovation and transformation [6], it appears to be crucial to also involve organizations with a smaller catchment area more prominently in the regional execution of the IZA. Additionally, the idea behind collaboration in networks suggests there must be a collectively recognized problem in a specific context and that participants must feel they need each other to solve it [3]. The healthcare administrative region is relatively large, and the catchment area of many organizations is much smaller. Therefore, it is uncertain to what extent they will perceive a shared problem in their daily practice with organizations that serve a completely different region and possibly a different population. Moreover, regional plans developed in relation to the IZA appear to vary little between healthcare administrative regions [27]. The bigger differences only become noticeable within healthcare administrative regions, for example, in neighborhoods [28]. This raises the question of whether the healthcare region is the appropriate level at which collaboration should be organized and what form it should take [7]. Networks at the healthcare administrative regional level should explore the appropriate scale for implementing IZA plans. Existing networks at the subregional level can be particularly valuable in this regard. The healthcare administrative region can play a crucial role in facilitating cross-pollination and the exchange of best practices between subregional networks and simultaneously be open to input from less involved organizations. The value of such a ‘meta-network’ has also been argued in research studies in other sectors [29]. It is important for the government and health insurance companies to encourage input from a wider range of organizations to drive this transformation. Additionally, it is crucial to monitor these transformations, to assess whether the desired transformation is being implemented correctly and whether scarce resources are not being utilized in these networks without achieving the intended results [30]. This will require collaboration between practitioners, policymakers, and researchers.

Finally, the results indicate that organizations with a smaller catchment area are not only less involved in a network at the healthcare administrative regional level but also have a need for transparency from organizations that are involved. The lack of transparency leads to feelings of uncertainty. Since the literature has demonstrated joint decision-making and transparency are crucial for the effectiveness of networks and participant commitment [14] and network members should not experience power imbalances [13], it is essential to ensure that the regional IZA network does not adversely affect these dynamics. This could result in existing networks in subregions losing participants or even their effectiveness or legitimacy, despite often having implemented successful initiatives. Government and policymakers should play an active role in ensuring diversity in regional network participation to ensure all interests in the region are represented. Additionally, dominant health insurance companies often participate in both healthcare administrative region networks and subregional networks. The literature has shown that organizations in such networks are highly susceptible to normative and coercive institutional pressures and expectations, particularly when health insurance companies are involved. For example, an organization may fear the consequences for negotiations with the health insurance company if they do not comply with the network [30]. Government and health insurance companies must ensure that organizations with less involvement and potential transparency issues do not suffer adverse consequences in future negotiations.

Collaboration in cross-domain networks is regarded as one of the solutions to sustain the Dutch healthcare system, and the government has been strongly encouraging this for several years. While implementation of the IZA focuses on innovation and transformation through regional collaboration, it impacts existing subregional networks. After implementation of the IZA, collaboration needs to be organized at a higher level (the healthcare administrative region) than before, making it increasingly complex and putting pressure on the legitimacy of existing networks as they need to redefine their role, purpose, and composition. A split arises among network members who participate in the new regional network and those who do not. When organizations participate at the regional level, they question their involvement in the subregional network because, given their catchment area, it is more efficient to operate solely in the network at the healthcare administrative regional level. When they are not participating at the regional level, organizations question their involvement in the subregional network because they are unsure whether the purpose of the subregional network still aligns with their organization and they feel a lack of transparency from organizations that do participate at the regional level. Government and policymakers should better align future policies with practice to facilitate collaboration more effectively. Additionally, strong emphasis should be placed on monitoring transformations to ensure that scarce resources contribute to the intended outcomes.