The purpose of this paper is to investigate inter-organizational designs for care–cure conditions in which low-risk patients are cared for in specialized care organizations and high-risk patients are cared for in specialized cure organizations. Performance impacts of increasing levels of integration between these organizations are analyzed.
Mixed methods were used in Dutch perinatal care: analysis of archival data, clinical research and system dynamics simulation modeling.
Inter-organizational design has an effect on inter-organizational dynamics such as collaboration and trust, and also on the operational aspects such as patient flows through the system. Solutions are found in integrating care and cure organizations. However, not all levels of integrated designs perform better than a design based on organizational separation of care and cure.
A clear split between midwifery practices (care) and obstetric departments (cure) will not work since all pregnant women need both care and cure. Having midwifery practices only works well when there are high levels of collaboration and trust with obstetric departments in hospitals. Integrated care designs are likely to exhibit superior performance. However, these designs will have an adverse effect on organizations that are not part of this integration, since integrating only a subset of organizations will feed distrust, low collaboration and hence low performance.
The originality of this research is derived from its multi-method approach. Archival data and clinical research revealed the dynamic relations between organizations. The caveat of some integrated care models was found through simulation.
Pieters, A., van Oorschot, K., Akkermans, H. and Brailsford, S. (2018), "Improving inter-organizational care-cure designs: specialization versus integration", Journal of Integrated Care, Vol. 26 No. 4, pp. 328-341. https://doi.org/10.1108/JICA-02-2018-0018Download as .RIS
Emerald Publishing Limited
Copyright © 2018, Emerald Publishing Limited
Healthcare has become a highly specialized service sector where different professionals with different backgrounds, working in different departments or even in different organizations have to collaborate to deliver high-quality care (Mintzberg, 1997). This specialization is not always working well: it can result in fragmented, poorly coordinated care and low service quality (Herzlinger, 1997; Kenagy et al., 1999). One root cause on which there appears to be a broad consensus is that the design of the services provided is in urgent need of improvement (Institute of Medicine, 2001; Herzlinger, 2004; Porter, 2010). This especially holds for patients who have a variety of needs, as is the case with chronic conditions such as cardiovascular risk, diabetes, chronic obstructive pulmonary disease (COPD) and congestive heart failure, but also with some psychological conditions and pregnancy. These conditions require general and preventive monitoring, education, psycho-social and basic medical support (which we will call a need for care) and in the case of an acute episode, patients need specialized, medical intervention (which we will call a need for cure). These care and cure needs can be met by the same professional, although, in the highly specialized practice of healthcare, they are mostly delivered by different professionals. In Europe, with its strong distinction between primary and secondary care, these professionals work in different organizations. Healthcare in the USA traditionally consists mostly of cure organizations, i.e. hospitals and specialist care (Wagner et al., 2001), but this is changing: these cure organizations have developed some care services, and primary/community organizations that focus on delivering care are being developed.
The current research focuses on the performance of an organizational design where low-risk patients are cared for in care organizations and where high-risk patients are cared for in cure organizations. From an operations management perspective, such a design is based on the principles of the focused factory concept (Skinner, 1974), in which an organizational split is made based on risk level (low risk vs high risk) and/or needs (care vs cure needs). The focused factory concept is based on the intuitive notion that a plant can achieve superior performance by organizing its resources to perform one task instead of trying to meet all sorts of demands from internal and external sources. However, it is also known that specialization in healthcare can result in problems of coordination and fragmentation (Herzlinger, 1997; Kenagy et al., 1999). So, even though an organizational design based on the focused factory concept may work well in theory, it can have its limitations in practice. Based on this notion, this research focuses on the following research questions (RQs):
What are the problems of the focused factory concept applied in healthcare systems with specialized organizations for care and cure?
Why are these problems encountered?
How can these problems be solved?
Healthcare is a complicated domain; it involves patients, patient flows, different professionals working in different organizations and a variety of procedures and regulations. As such, this is best researched on different levels, and from a range of disciplines such as operations management, organizational science and health policy (Creswell and Clark, 2007). Meanwhile, research and theory regarding inter-organizational designs in healthcare are at their early, formative stages, and the variables are still unknown. As such, we will need exploratory research with a mixed methods approach that combines elements of qualitative and quantitative research approaches (Creswell and Clark, 2007; Johnson et al., 2007). We conduct our research in a care system that has been organized according to the principles of the focused factory concept for decades: Dutch perinatal care. To answer our three RQs, we will perform three studies. First, we will use archival data analyses to gain insight in patient flows between an independent midwifery practice and an obstetric department in a hospital. In our second study, we will perform clinical research (Lewin, 1948; Schein, 1987) to gain a deeper understanding of why the Dutch perinatal system is not always operating according to its design. Third, we will use simulation modeling (system dynamics (SD)) to investigate how Dutch perinatal care can be improved.
Overall, our mixed-method research in Dutch perinatal care teaches us that the inter-organizational design (i.e. a tiered system) has an effect on inter-organizational dynamics such as collaboration and trust, and on the operations such as patient flows through the system. Solutions are found in integrating care and cure organizations, however, not all levels of integrated care designs perform better than a design based on an organizational separation of care and cure.
Traditionally, healthcare has been delivered according to the “acute care” model (Wagner et al., 2001). It has been focused on cure, on medical intervention, delivered in hospitals, in secondary care. In the last decade awareness has increased of the value of care, of the psychological aspects and of the value of prevention in care–cure conditions (e.g. Rothman and Wagner, 2003). As a result of this shift, new designs for healthcare delivery have been developed. One notion from the OM literature that has exercised great appeal for the healthcare sector is the focused factory concept (Skinner, 1974). The concept is based on the intuitive notion that a plant can achieve superior performance by organizing its resources to perform one task instead of trying to meet all sorts of demands from internal and external sources. A focused factory with a narrow product mix for a particular market niche will outperform the conventional plant, which has a broader scope. This principle is also applied in healthcare (Carey and Mitchell, 2018; Herzlinger, 2004; McLaughlin et al., 1995). For care–cure conditions, two main designs can be discerned: a design based on organizational separation, where care and cure are provided by different organizations, as is the case with primary and secondary care (e.g. Chin et al., 2000), and a design based on organizational integration, in which both care and cure are provided by the same organization, as is the case with specialty hospitals and service lines in general hospitals (e.g. McDermott et al., 2011).
Neither one of the above designs stands out as clearly the best: all come with problems. As such, in a recent study, the effects of focus on healthcare performance are reported to be moderate, mixed or missing (Bredenhoff et al., 2010; Peltokorpi et al., 2016). Focus exploits division of labor, specialization, standardization and knowledge creation (Ding, 2014; Douma and Schreuder, 2008). But, when care and cure are delivered by different organizations, care processes often become fragmented and problems of coordination arise regarding the patient’s condition (e.g. Edgren and Barnard, 2015; Preston et al., 1999). Each organization tends to focus on its own tasks and resources and not on the system as a whole, that is, the system actually experienced by patients: the customer journey (Wener and Woodgate, 2016; Zomerdijk and Voss, 2010). As a result, the task of improving the quality of interaction, cooperation and communication across the interfaces is not seen as any individual group’s particular responsibility (Kvamme et al., 2001). In addition, when professionals and organizations specialize in a particular condition, this can cause problems for co-morbidity patients (i.e. patients with more than one, often chronic, condition), and when professionals do not specialize in a particular condition, as is the case with practice nurses and general practitioners, knowledge of a particular condition is often lacking.
There have been numerous attempts to overcome these problems by focusing on integrating the delivery of care and cure. An integrated service focuses on improving outcomes for a target population (Ovretveit, 1998), by improving coordination, communication and collaboration between providers and/or professionals and by integrating services. Many different definitions and concepts exist (Kodner and Spreeuwenberg, 2002; Memon and Kinder, 2016). They differ regarding the type of integration (functional, organizational, professional or clinical), the breadth of integration (horizontal or vertical), the degree of integration (linkage, coordination or integration) and the process of integration (Nolte and McKee, 2008). Integrated structures rarely integrate the actual delivery of care and cure at a patient level (Burns and Pauly, 2002): they focus more on administrative and organizational integration. If one is to achieve integrated care at the patient level, focus should be on the individual health professionals, on their collaboration, and on their behavior (Van Wijngaarden et al., 2006).
Collaboration between professionals and organizations is not only of importance to the healthcare sector, it is important in other sectors as well. Collaboration has been studied in various settings such as in buyer–supplier relations (e.g. Autry and Golicic, 2010) and in inter-organizational networks (e.g. Nooteboom, 2004). Inter-organizational relations are continually shaped and restructured by the organizations involved through a repetitive sequence of negotiation, commitment and execution (Ring and Van de Ven, 1994). In addition, relationship–performance spirals exist between collaborating organizations (Autry and Golicic, 2010) and there can be positive feedback loops between factors such as collaboration, travail (work/hardship), trust, information transparency and performance (Akkermans et al., 2004). In settings in which multiple providers are engaged in carrying out highly interdependent tasks under conditions of uncertainty and time constraints, as is often the case in healthcare, shared knowledge, shared goals and mutual respect are important for achieving desired outcomes (Gittell and Douglass, 2012).
Thus different designs exist for care–cure conditions. However, each of these designs has its problems and it is not known yet what design might work best. In this research, we will study the different inter-organizational designs by focusing first on what the problem is with a design where care and cure are delivered by separate organizations and why it is a problem, before turning to solutions of integrated care.
This research was conducted in the Dutch perinatal care system. This encompasses the care of pregnant women from the start of their pregnancy to giving birth. Different ideologies in perinatal care exist (e.g. Van Teijlingen, 2005), with the two extremes being the medical model (or illness-model) and the midwifery model (or wellness-model). In the former, pregnant women are primarily cared for by obstetricians within a medical, hospitalized setting, as is the case in the USA. In contrast, the midwifery model considers being pregnant and giving birth to be healthy and natural events, physiological processes involving no illness or disease, as is the case in the Netherlands. The current design of Dutch perinatal care is based on principles of the focused factory concept (Pieters et al., 2010; Van Stenus et al., 2017). In the Netherlands, all pregnant women are classified into two distinct groups: low risk or high risk. For these two sub-populations, two different services have been set up, staffed with different professionals (midwives and obstetricians) who have been educated and trained very differently, who belong to different professional communities and work in different organizations (independent midwifery practices and obstetric departments in hospitals). Moreover, policy development and financing structures are also different for midwives and obstetricians. The perinatal care system is organized in more or less the same way throughout the Netherlands.
In addition to the fact that Dutch perinatal care has been organized according to focused factory principles, it was chosen as the research setting for three further reasons. First, Dutch perinatal care has been known for many years for its midwifery model of care, which has a strong focus on care. In contrast, for many other care–cure conditions it is only recently that healthcare providers have become aware of these care aspects, and many providers are still figuring out how to organize the delivery of care. Second, Dutch perinatal care is widely regarded as an example of good practice by other Western countries who would like to move toward a system with stronger midwife involvement for low-risk pregnancies (e.g. Wagner, 2006). Third, the Netherlands has been investigating the effectiveness of its perinatal care system, since despite this international reputation it has problems regarding perinatal and maternal morbidity and mortality rates (Euro-Peristat, 2008), and the degree of user satisfaction with services expressed by pregnant women (Rijnders et al., 2008).
Research design and methods
This research applies a mixed-method approach, consisting of three parts, each with its own RQ, its own method and each with a focus on a certain part of Dutch perinatal care. A mixed methods approach is often used in management studies in general (Creswell and Clark, 2007), and in particular in studies that focus on the inter-organizational level such as supply chains (Singhal and Singhal, 2012). The first part focuses on RQ1 and is addressed by archival data analyses of the flow of patients between one hospital and one midwifery practice. The second part focuses on RQ2 and uses a variety of clinical research approaches applied to a region in the Netherlands. The third part focuses on RQ3, and is answered with SD modeling. Simulation is especially useful for theory development when the focal phenomena involve multiple and interacting processes, time delays or other non-linear effects such as feedback loops and thresholds, when the theoretical focus is longitudinal, non-linear or processual, or when empirical data are challenging to obtain (Davis et al., 2007).
Archival data analysis for insight into patient flows
The first phase focuses on the problems of the application of the focused factory concept in Dutch perinatal care, through an examination of operational performance. Performance is determined by internal and external fit (Pieters et al., 2010; Venkatraman and Camillus, 1984). The internal environment is composed of the business strategy, manufacturing/operations strategy and organizational design. The external environment consists of all those significant elements outside the organization, such as the industry, clients, competitors, governmental entities and professional groups.
Regarding the internal fit, since Dutch perinatal care is designed in line with focused factory principles, one would expect midwifery practices to focus on low-risk pregnant women only, obstetric departments in hospitals to focus on high-risk pregnant women only and that there should not be too many transfers between the two organizations. Regarding the external fit, one would expect internal processes to be aligned with the characteristics of the medical condition, i.e. pregnancy. The internal and external fit was investigated by archival data analysis of the flows of pregnant women between a midwifery practice and a hospital in Tilburg. Data from 737 pregnant women in one particular year (2007) were collected regarding the medical condition at the start of the pregnancy, the consultations during the pregnancy, the delivery and referrals between the midwifery practice and the hospital. The findings were discussed with representatives from both organizations.
Clinical research for insight into inter-organizational dynamics
In a system based on the focused factory concept and in which patients should be referred between organizations only when necessary, collaboration between the organizations becomes important. To try to understand why the focused factory concept does not work well in Dutch perinatal care, clinical research (Schein, 1987) was initiated, at the request of the system stakeholders — the 16 obstetricians of the obstetric departments of two hospitals and the 45 midwives of 12 midwifery practices — in an effort to improve system performance by increasing joint understanding of the underlying dynamics of their faltering collaboration.
Four activities (see also Akkermans, 2001) were undertaken. First, 13 obstetricians and 22 midwives (35 in total) completed a questionnaire, which was based on previous questionnaires that studied inter-firm relationships (Humpreys et al., 2003; Johnston et al., 2004). Second, semi-structured interviews were held with seven obstetricians in the hospitals and 19 midwifes working in the six midwifery practices. The following questions were used as a guideline for the interviews: what contributes to good collaboration? What contributes to bad collaboration? How do you notice the performance of the collaboration? Third, group model building sessions (Vennix, 1996) focused on one or two aspects of the collaboration were conducted. Causal loop diagrams (Sterman, 2000) were used to disentangle the problems. This resulted in ten different causal loop diagrams revealing the root causes of these problems. Fourth, in a plenary session the findings were presented to all participants. This approach allowed us to initially investigate a large number of variables and their relationships, find out which of these were the most significant and gain insight into people’s mental models. This is of importance since at the start of the project it was not even known what all the relevant variables were, let alone their precise inter-relationships.
Simulation modeling for insights into inter-organizational designs
The third phase focused on evaluating inter-organizational designs, taken into account the inter-organizational dynamics as studied in the second phase, and demonstrating why these inter-organizational designs might or might not work. This was achieved through the use of SD modeling. SD focuses on how causal relationships among system elements can influence the behavior of a system (Forrester, 1961; Sterman, 2000). It intensifies the perception of feedback, accumulation and delays (Größler et al., 2008). A further strength of SD is that models can be based on different sources of data (Forrester, 1980). In our case, mental, written and numerical data on patient flows and inter-organizational dynamics were elicited through archival data analysis, questionnaires, interviews and group model building sessions. Additional data were derived from the literature on the Dutch perinatal care system and collaboration.
Insight into patient flows through archival data analysis
The first phase focused on the problems of the application of the focused factory concept in Dutch perinatal care, investigating internal and external fit by analyzing patient flows. Our archival data analysis showed that there is not a good fit between the design of the system and the way it actually operates. Figure 1 represents four states and the inflows and outflows between states in Dutch perinatal care for pregnant women. In the diagramming notation, states, represented by rectangles, denote a level of a variable (in our case the number of pregnant women), flow variables fill or drain the state and are depicted as pipes with valves (Sterman, 2000). Here, the state in which a woman resides depends on her risk level (low risk or high risk) and on the organization that takes care of her (midwifery practice or obstetric department in a hospital). Note that, for reasons of readability, Figure 1 does not present the inflow and the outflow of pregnant women in each state by becoming pregnant and delivering. The other variables in this figure will be explained later.
Regarding the internal fit, according to focused factory principles one would expect only a small number of transfers between the two organizations. Instead, in our analyses, we found that there is a significant transfer of pregnant women between the organizations: 77 percent of the pregnant women who start in the midwifery practice also receive hospital care at some point. Moreover, a significant number of low-risk pregnant women are cared for in the hospital: 50 percent of the pregnant women in the hospital were categorized as low risk at the start of their pregnancy, but were not referred to the midwifery practice. Interestingly, the women who receive care from both the midwifery practice and the hospital have on average 17.5 consultations, and they see on average ten different healthcare professionals (midwives, obstetricians, residents).
Regarding the external fit, one would expect internal processes to be aligned with the characteristics of the medical condition, i.e. pregnancy. However, we found that it is hard to predict the risk level of pregnant women at the start of their pregnancy. In a focused factory system, one would want to decide at the start of the pregnancy which type of “factory” will take care of the pregnant women. Our archival data analysis showed that 77 percent of the women that were labeled as being low risk at the start of their pregnancy in the midwifery practice still had to see an obstetrician along the way. In addition, pregnant women do need both care and cure along the way. High-risk pregnant women also need psycho-social support, and low-risk pregnant women often do need some medical expertise, since the predictability of the risk level is low.
Summarizing, our RQ1 was aimed at finding out the problems of the focused factory concept applied in healthcare systems with specialized organizations for care and cure. Our analysis reveals that although the organizations may be specialized, the patients are not: the majority of patients need to move between care and cure organizations. The main reasons for this is the low predictability of the risk level at the start of the pregnancy and the need for both care and cure of most patients.
Insights into inter-organizational dynamics through qualitative research
In a system that is based on the focused factory concept and in which patients should be referred between organizations when necessary, collaboration between organizations is likely to be important. Our interviews, questionnaires, group model building and other clinical research revealed two key findings. First, midwives and obstetricians all believe there is value in collaborating together (exchanging information, streamlining care processes, knowing each other better), and that better collaboration will have a positive effect on performance (increased service to patients, increased quality of care, decreased costs, higher patient satisfaction, higher professional satisfaction). However, even though both professional groups expect the relationship to last a long time, the relationship is by no means symmetric: midwives trust obstetricians more than vice versa, midwives provide the obstetricians with more information and midwives feel they depend more on the obstetricians than vice versa.
Second, the root causes of why things go wrong include a lack of trust, feelings of competition, a sub-optimal exchange of information, sub-optimal communication between midwifery practices and obstetric departments in hospitals and not knowing each other. Although trust consists of several aspects (Nooteboom, 2004), our research shows that the trust midwifery practices have in hospitals seems to be determined by financial incentives, by fear of hospitals “stealing” their clients. On the other hand, the trust hospitals have in midwifery practices seems to be determined by the obstetricians’ perceptions of the competences of midwives in recognizing high-risk pregnancies. Trust and referral behavior are tied together in a self-reinforcing feedback loop (R1), the trust-loop (see Figure 1).
The more low-risk pregnant women are cared for in hospitals, the less trust midwifery practices have in hospitals, so midwifery practices refer fewer high-risk pregnant women to hospitals, and there are more high-risk pregnant women in midwifery practices, reducing the trust hospitals have in midwifery practices. As a result, hospitals refer fewer low-risk pregnant women to midwifery practices, which results in more low-risk pregnant women in hospitals, and so on.
Summarizing, in our first study we found that most patients need both care and cure which causes a need for specialized organizations for care and cure to refer patients from one organization to the other. Our second study was aimed at understanding why this referral may be a problem (our RQ2). Our findings reveal that when the specialized organizations in a focused factory concept do not trust each other, this results in fewer/slower referrals of patients between the two types of organizations, leading to more patients cared for in the “wrong” organization, which will further feed distrust in both types of organizations. In addition, the benefits of being a focused factory will be reduced.
Insights into inter-organizational designs through simulation modeling
In the third phase of this research a SD model of Dutch perinatal care was developed. A high level overview is presented in Figure 1. Whether or not pregnant women are referred between the organizations depends on the level of trust (as described above) and on the level of collaboration. Collaboration in the model is characterized by contractual obligations, rules and procedures; it is the level of formal collaboration (Smith et al., 1995; Wren, 1961). Within an organization, pregnant women “flow” from one state to another by developing and recovering from a high-risk pregnancy, which is affected by the quality of care the organization is delivering. The quality of care is determined by the competences of the professionals involved, the work pressure (Oliva and Sterman, 2001) and the collaboration. Broadly speaking, midwives lack cure competences (Amelink‐Verburg and Buitendijk, 2010; Reuwer et al., 2009) and obstetricians lack care competences (Franx, 2011; NVOG-HOOG, 2005). Note that in reality, there is inevitably an element of “care” in the hospital sector and an element of “cure” in the midwifery sector. However, these concepts are essentially archetypes, intended for gaining insights into the behavior of human actors in the maternity care system, rather than a rigid demarcation of clinical practice. Work pressure is determined by the number of consultations pregnant women require. Work pressure and quality of care are tied together in a self-reinforcing feedback loop: the higher the work pressure, the lower the quality of care, the greater the number of high-risk pregnant women, the more consultations and the higher the work pressure (in Figure 1: R2 for midwifery practices and R3 for hospitals).
One apparent solution to the problems with the current design is to make major steps toward a more integrated system. Besides the current focused factory based design of Dutch perinatal care, the SD model evaluates three inter-organizational models that all focus on integrating care, although in different ways:
Collaborative model: formal collaboration between all midwifery practices and hospitals is structurally improved by an inter-organizational improvement project.
Improved hospital model: hospitals employ midwives too. As a result hospitals are able to meet the care needs of pregnant women. There are no specific efforts to improve collaboration between hospitals and midwifery practices.
Partial merged care model: a percentage (in this case 20 percent) of the midwifery practices and the hospitals merge into one integrated care unit which is able to meet both the care and the cure needs of pregnant women. The remaining midwifery practices stay independent from the hospitals. There are no specific efforts to improve collaboration between hospitals and the remaining independent midwifery practices.
The main goal in a healthcare system is to improve the health of a population (Horvath, 1975), and thus it is not the performance of individual organizations that counts, but the performance of the system as a whole. Patients should be cared for in the right organization: high-risk pregnant women in hospitals (cure organizations), low-risk pregnant women in midwifery practices (care organizations). Having high-risk patients cared for in midwifery practices can lead to negative effects on their medical outcomes (Van Dillen et al., 2010), while low-risk pregnant women cared for in the hospital often results in unnecessary, dangerous and invasive obstetric interventions (Wagner, 2001).
The four models were run for ten simulated years (the model is available from the authors upon request). Table I presents the end values of the proportion of high-risk pregnant women in hospitals, including the improvement percentage compared to the current situation (base case).
There are several implications for Dutch perinatal care. First, for Dutch perinatal care the best thing to do is to implement the collaborative model, i.e. to improve collaboration between all independent midwifery practices and obstetric departments in hospitals, because of the virtuous cycles of trust and transparency that are nurtured in this setting. Second, the improved hospital model performs the worst. If hospitals opt for this model, they might be able to deliver a higher quality of care and increase their own performance, but the proportion of high-risk pregnant women cared for by the “wrong” organization (in the midwifery practices) will increase, and overall performance of the system will decrease. A third finding is that when a few midwifery practices decide to merge with the hospitals, while most practices remain independent and outside of the collaboration, overall system performance will get worse, due to vicious cycles of eroding trust between hospitals and independent midwifery practices, performance for the whole patient population will decrease.
Summarizing, our RQ3 was aimed and finding out how the problem of low trust between organizations involved in a focused factory design in healthcare can be solved. Our simulation results reveal that trust can be enhanced by improving collaboration between all organizations in a specific healthcare region. This will result in improved performance for the whole population of patients, since collaboration results in patients being referred between organizations when needed, which increases trust in the system and more clinically appropriate referrals between organizations. When only a subset of the organizations in a specific healthcare region is merged into an integrated care unit that focuses on both care and cure and on both low risk and high risk, the performance of this integrated care unit’s own patient population will increase. However, the performance for the whole population of patients will decrease, since the remaining independent organizations distrust the integrated care unit and are less likely to refer patients to it.
Implications and conclusions
This research has focused on the organizational design for care–cure conditions. Many different organizational designs exist, ranging from a complete separation of care and cure to the integration of care and cure in one organization. This research has investigated three RQs:
What are the problems of the focused factory concept applied in healthcare systems with specialized organizations for care and cure?
Why are these problems encountered?
How can these problems be solved?
This research has applied a mixed-method approach to investigate organizational design for care–cure conditions. The Dutch perinatal care system, traditionally based on the “focused factory” concept, was used as our research setting.
Implications for practice
Based on our research, we conclude that the current organizational design for Dutch perinatal care, which is based on organizational separation of low risk and high risk and of care and cure, is not optimal. Improved performance can be achieved through integrating care. However, there is a caveat: partially integrated designs actually result in worse overall system performance than the current segmented model. This is the case for the improved hospital model and the partially merged model that meets both care and cure needs and can serve both low-risk and high-risk pregnant women, respectively, by employing midwives or merging with some (but not all) midwifery practices. In these models, feelings of competition will dominate, the independent midwifery practices will distrust the improved hospital or the merged integrated care unit, and as a result they are less likely to refer high-risk pregnant women in a timely fashion, causing pregnant women to receive sub-optimal care. For Dutch perinatal care, the best thing to do is to implement the collaborative model, i.e. to improve collaboration between all independent midwifery practices and obstetric departments in hospitals in a specific region.
The implications of these findings for perinatal care systems in other countries depend on the existing system that is in place. For systems built on the medical model (e.g. the USA), where healthcare providers are thinking about moving toward a more midwifery-based model, this research advises caution. Choosing an inter-organizational design in which there is a clear split between midwifery practices focusing on low-risk pregnant women and obstetric departments in hospitals focusing on high-risk pregnant women will not work, since it is not known in advance whether a pregnancy will turn out to be a low or a high risk one, and since all pregnant women need both care and cure. Having independent midwifery practices only works well when there are high levels of collaboration and trust with cure providers (obstetric departments in hospitals).
Implications for theory, limitations and suggestions for further research
In general, for care–cure conditions, organizational designs that seek to integrate care and cure activities are likely to exhibit superior performance over designs that focus on specialization. One important caveat is that such integrated care designs will have an adverse effect on the overall population if they remain limited to only a subset of the overall population, since such an integrated subset will feed distrust, low collaboration and low performance from the rest of the healthcare system. This contribution to theory is inductively derived from our archival data analysis, clinical research and simulation. The dynamic relations between the organizations could not have been found with archival data analysis only: we needed clinical research in order to discover them. In addition, the insights we found regarding the caveat of some integrated care models could not have been found without simulation.
Although pregnancy is a care–cure condition, it differs from other care–cure conditions. First, it is time limited (i.e. nine months), whereas most other care–cure conditions are chronic, life-long conditions. Referral behavior might be different for chronic diseases. Second, most pregnant women do not have co-morbidities, whereas they are common in other care–cure conditions. Co-morbidities are likely to result in a more complicated care process, with more professionals and more organizations involved. Third, Dutch cure organizations are able to take care of low-risk pregnant women, which results in competition between care and cure providers. For other conditions, it might not be possible for the cure providers to also take care of the patients who only need care.
Despite the differences between pregnancy and other care–cure conditions, Dutch perinatal care is still an appropriate setting to study inter-organizational designs. Whereas healthcare providers have only relatively recently become aware of the psycho-social aspects of many care–cure conditions, Dutch perinatal care is widely known for its focus on care, since it has been operating according to the midwifery model for decades. Furthermore, regardless of the problems mentioned in this paper, the perinatal morbidity rate in The Netherlands is one of the lowest in Europe (Eurostat, 2015), pregnant women have high trust in the quality of care, and there is no increased risk of adverse perinatal outcomes for planned home births among low-risk women (De Jonge et al., 2015). In addition, Dutch perinatal care is a clearly defined system: only two types of organizations are involved, and it is set up the same throughout the whole country. The GP only plays a minor role in this system, since only 5 percent are involved in perinatal care (De Verloskundige, 2018). It seems to make sense to first understand the dynamics in a “simple” healthcare network (two types of providers) designed for a condition with hardly any co-morbidities (pregnancy). A suggestion for further research would be to conduct a similar study for a care–cure condition that involves chronic patients with a high risk of co-morbidities. Care–cure conditions such as diabetes, COPD and heart failure are commonly associated with co-morbidities and hence a larger number of different specialist professionals are involved, over a much longer time period.
Results of the scenarios
|High risk in hospital|
|% HR in H||Normalized value||Improvement %|
|Improved hospital model||62||91.2||−8.8|
|Partial merged care model||64||94.1||−5.9|
Akkermans, H. (2001), “Renga: a systems approach to facilitating inter‐organizational network development”, System Dynamic Review, Vol. 17 No. 3, pp. 179-193.
Akkermans, H., Bogerd, P. and Van Doremalen, J. (2004), “Travail, transparency and trust: a case study of computer-supported collaborative supply chain planning in high-tech electronics”, European Journal of Operations, Research, Vol. 153 No. 2, pp. 445-456.
Amelink‐Verburg, M.P. and Buitendijk, S.E. (2010), “Pregnancy and labour in the Dutch maternity care system: what is normal? The role division between midwives and obstetricians”, Journal of Midwifery & Women’s Health, Vol. 55 No. 3, pp. 216-225.
Autry, C.W. and Golicic, S.L. (2010), “Evaluating buyer–supplier relationship–performance spirals: a longitudinal study”, Journal of Operations Management, Vol. 28 No. 2, pp. 87-100.
Bredenhoff, E., van Lent, W.A.M. and van Harten, W.H. (2010), “Exploring types of focused factories in hospital care: a multiple case study”, BMC Health Services Research, Vol. 10 No. 154, pp. 1-16.
Burns, L.R. and Pauly, M.V. (2002), “Integrated delivery networks: a detour on the road to integrated health care?”, Business Health Care, Vol. 21 No. 4, pp. 126-143.
Carey, K. and Mitchell, J.M. (2018), “Specialization and production cost efficiency: evidence from ambulatory surgery centers”, International Journal of Health Economics and Management, Vol. 18 No. 1, pp. 83-98.
Chin, M.H., Auerbach, S.B., Cook, S., Harrison, J.F., Koppert, J., Jin, L., Thiel, F., Karrison, T.G., Harrand, A.G., Schaefer, C.T., Takashima, H.T., Egbert, N., Chiu, S.C. and McNabb, W.L. (2000), “Quality of diabetes care in community health centers”, American Journal of Public Health, Vol. 90 No. 3, pp. 431-434.
Creswell, J.W. and Clark, V.L.P. (2007), Designing and Conducting Mixed Methods Research, Sage Publications, Thousand Oaks, CA.
Davis, J.P., Eisenhardt, K.M. and Bingham, C.B. (2007), “Developing theory through simulation methods”, Academy of Management Review, Vol. 32 No. 2, pp. 480-499.
De Jonge, A., Geerts, C.C., Van der Goes, B.Y., Mol, B.W., Buitendijk, S.E. and Nijhuis, J.G. (2015), “Perinatal mortality and morbidity up to 28 days after birth among 743 070 low-risk planned home and hospital births: a cohort study based on three merged national perinatal databases”, BJOG: An International Journal of Obstetrics and Gynaecology, Vol. 122 No. 5, pp. 720-728.
De Verloskundige (2018), “De Huisarts”, available at: http://deverloskundige.nl/over-de-verloskundige/subtekstpagina/162/de-huisarts/ (accessed April 6, 2018).
Ding, D.X. (2014), “The effect of experience, ownership and focus on productive efficiency: a longitudinal study of US hospitals”, Journal of Operations Management, Vol. 32 Nos 1-2, pp. 1-14.
Douma, S. and Schreuder, H. (2008), Economic Approaches to Organizations, Prentice Hall, Essex.
Edgren, L. and Barnard, K. (2015), “Achieving integrated care through CAS thinking and a collaborative mindset”, Journal of Integrated Care, Vol. 23 No. 3, pp. 108-119.
Euro-Peristat (2008), “European perinatal health report”, available at: www.europeristat.com (accessed February 10, 2011).
Eurostat (2015) “Fetal, peri- and neonatal mortality rates by country of occurrence”, available at: http://ec.europa.eu/eurostat/web/products-datasets/-/hlth_cd_aperrto (accessed April 6, 2018).
Forrester, J. (1961), Industrial Dynamics, MIT Press, Cambridge, MA.
Forrester, J.W. (1980), “Information sources for modeling the national economy”, Journal of the American Statistical Association, Vol. 75 No. 371, pp. 555-566.
Franx, A. (2011), Lijn Der Verwachting. Inaugurale Rede, Universiteit Utrecht, Utrecht.
Gittell, J.H. and Douglass, A. (2012), “Relational bureaucracy: structuring reciprocal relationships into roles”, Academy of Management Review, Vol. 37 No. 4, pp. 709-733.
Größler, A., Thun, J.H. and Milling, P.M. (2008), “System dynamics as a structural theory in operations management”, Production and Operations Management, Vol. 17 No. 3, pp. 373-384.
Herzlinger, R. (1997), Market-Driven Health Care: Who Wins, Who Loses in the Transformation of America’s Largest Service Industry, Addison-Wesley, Reading, MA.
Herzlinger, R. (Ed.) (2004), Consumer-Driven Health Care, Implications For Providers, Payers, and Policymakers, Jossey-Bass, San Francisco, CA.
Horvath, W.J. (1975), “Obstacles to the application of operations research techniques in the health field”, in Schuman, L.J., Dixon Speas, R. and Young, J.P. (Eds), Operations Research in Health Care, A Critical Analysis, The Johns Hopkins University Press, Baltimore, MD and London, pp. 66-86.
Humpreys, P.K., Li, W.L. and Chan, L.Y. (2003), “The impact of supplier development on buyer-supplier performance”, Omega, Vol. 32 No. 2, pp. 131-143.
Institute of Medicine (2001), Crossing the Quality Chasm: A New Health System for the 21st Century, National Academy Press, Washington, DC.
Johnson, R.B., Onwuegbuzie, A.J. and Turner, L.A. (2007), “Toward a definition of mixed methods research”, Journal of Mixed Methods Research, Vol. 1 No. 2, pp. 112-133.
Johnston, D.A., McCutcheon, D.M., Stuart, F.I. and Kerwood, H. (2004), “Effects of supplier trust on performance of cooperative supplier relationships”, Journal of Operations Management, Vol. 22 No. 1, pp. 23-38.
Kenagy, J.W., Berwick, D.M. and Shore, M.F. (1999), “Service quality in health care”, JAMA: The Journal of the American Medical Association, Vol. 281 No. 7, pp. 661-665.
Kodner, D.L. and Spreeuwenberg, C. (2002), “Integrated care: meaning, logic, applications, and implications – a discussion paper”, International Journal of Integrated Care, Vol. 2 No. 4, pp. 1-6.
Kvamme, O.J., Olesen, F., Samuelson, M. and Samuelsson, M. (2001), “Improving the interface between primary and secondary care: a statement from the European Working Party on Quality in Family Practice (EQuiP)”, Quality in Health Care, Vol. 10 No. 1, pp. 33-39.
Lewin, K. (1948), Resolving Social Conflicts, Free Press, New York, NY.
McDermott, C.M., Stock, G.N. and Shah, R. (2011), “Relating focus to quality and cost in a healthcare setting”, Operations Management Research, Vol. 4 No. 3, pp. 127-137.
McLaughlin, C.P., Yang, S. and Van Dierdonck, R. (1995), “Professional service organizations and focus”, Management Science, Vol. 41 No. 7, pp. 1185-1193.
Memon, A.R. and Kinder, T. (2016), “Co-location as a catalyst for service innovation: a study of Scottish health and social care”, Public Management Review, Vol. 19 No. 4, pp. 381-405.
Mintzberg, H. (1997), “Towards healthier hospitals”, Health Care Management Review, Vol. 22 No. 1, pp. 9-18.
Nolte, E. and McKee, M. (2008), “Integration and chronic care: a review”, in Nolte, E. and McKee, M. (Eds), Caring for People with Chronic Conditions: A Health System Perspective, WHO/European Observatory on Health Systems and Policies, Open University Press, Berkshire, pp. 64-91.
Nooteboom, B. (2004), Inter-Firm Collaboration, Learning & Networks, An Integrated Approach, Routledge, London.
NVOG-HOOG (2005), “Curriculum opleiding tot gynaecoloog”, in Scheele, F. and Schutte, M.F. (Eds), Het Centrum Utrecht bv, Utrecht.
Oliva, R. and Sterman, J.D. (2001), “Cutting corners and working overtime: quality erosion in the service industry”, Management Science, Vol. 47 No. 7, pp. 894-914.
Ovretveit, J. (1998), Integrated Care: Models and Issues, A Nordic School of Public Health Briefing Paper, Nordic School of Public Health, Goteborg.
Peltokorpi, A., Linna, M., Malmström, T., Torkki, P. and Lillrank, P.M. (2016), “Five focus strategies to organize health care delivery”, International Journal of Health Care Quality Assurance, Vol. 29 No. 2, pp. 177-191.
Pieters, A., van Oirschot, C. and Akkermans, H. (2010), “No cure for all evils Dutch obstetric care and limits to the applicability of the focused factory concept in health care”, International Journal of Operations and Production Management, Vol. 30 No. 11, pp. 1112-1139.
Porter, M.E. (2010), “What is value in health care?”, The New England Journal of Medicine, Vol. 363 No. 26, pp. 2477-2481.
Preston, C., Cheater, F., Baker, R. and Hearnshaw, H. (1999), “Left in limbo: patients’ views on care across the primary/secondary interface”, Quality in Health Care, Vol. 8 No. 1, pp. 16-21.
Reuwer, P.J.H.M., Bruinse, H.W. and Franx, A. (2009), Pro-Active Support of Labour, Cambridge University Press, London.
Rijnders, M., Baston, H., Schönbeck, Y., Pal, K.V.D., Prins, M., Green, J. and Buitendijk, S. (2008), “Perinatal factors related to negative or positive recall of birth experience in women 3 years postpartum in the Netherlands”, Birth, Vol. 35 No. 2, pp. 107-116.
Ring, P.S. and Van de Ven, A.H. (1994), “Developmental processes of cooperative interorganizational relationships”, Academy of Management Review, Vol. 19 No. 1, pp. 90-118.
Rothman, A.A. and Wagner, E.H. (2003), “Chronic illness management: what is the role of primary care?”, Annals Internal Medicine, Vol. 138 No. 3, pp. 256-262.
Schein, E.H. (1987), The Clinical Perspective in Fieldwork, Sage Publications, Newbury Park, CA.
Singhal, K. and Singhal, J. (2012), “Imperatives of the science of operations and supply-chain management”, Journal of Operations Management, Vol. 30 No. 3, pp. 237-244.
Skinner, W. (1974), “The focused factory”, Harvard Business Review, Vol. 52 No. 3, pp. 113-121.
Smith, K.G., Carroll, S.J. and Ashford, S.J. (1995), “Intra-and interorganizational cooperation: toward a research agenda”, Academy of Management Journal, Vol. 38 No. 1, pp. 7-23.
Sterman, J.D. (2000), Business Dynamics, Systems Thinking and Modeling for a Complex World, Irwin McGraw-Hill, Boston, MA.
Van Dillen, J., Mesman, J.A.J.M., Zwart, J.J., Bloemenkamp, K.W.M. and Van Roosmalen, J. (2010), “Introducing maternal morbidity audit in the Netherlands”, BJOG: An International Journal of Obstetrics & Gynaecology, Vol. 117 No. 4, pp. 416-421.
Van Stenus, C.M., Boere-Boonekamp, M.M., Kerkhof, E.F. and Need, A. (2017), “Client satisfaction and transfers across care levels of women with uncomplicated pregnancies at the onset of labor”, Midwifery, Vol. 48, May, pp. 11-17.
Van Teijlingen, E. (2005), “A critical analysis of the medical model as used in the study of pregnancy and childbirth”, Sociological Research Online, Vol. 10 No. 2, pp. 1-15.
Van Wijngaarden, J.D.H., De Bont, A.A. and Huijsman, R. (2006), “Learning to cross boundaries: the integration of a health network to deliver seamless care”, Health Policy, Vol. 79 No. 2, pp. 203-213.
Venkatraman, N. and Camillus, J.C. (1984), “Exploring the concept of ‘fit’ in strategic management”, Academy of Management Review, Vol. 9 No. 3, pp. 513-525.
Vennix, J.A.M. (1996), Group Model-Building. Facilitating Team Learning Using System Dynamics, Wiley, Chichester.
Wagner, E.H., Austin, B.T., Davis, C., Hindmarsh, M., Schaefer, J. and Bonomi, A. (2001), “Improving chronic illness care: translating evidence into action”, Health Affairs, Vol. 20 No. 6, pp. 64-78.
Wagner, M. (2001), “Fish can’t see water: the need to humanize birth”, International Journal of Gynaecology & Obstetrics, Vol. 75 No. S1, pp. S25-S37.
Wagner, M. (2006), Born in the USA, How a Broken Maternity System Must be Fixed to Put Women and Children First, University of California Press, Berkeley, CA.
Wener, P. and Woodgate, R.L. (2016), “Collaborating in the context of co-location: a grounded theory study”, BMC Family Practice, Vol. 17 No. 30, pp. 17-30.
Wren, D.A. (1961), “Interface and interorganizational coordination”, Academy of Management Journal, Vol. 10 No. 1, pp. 69-81.
Zomerdijk, L.G. and Voss, C.A. (2010), “Service design for experience-centric services”, Journal of Service Research, Vol. 13 No. 1, pp. 67-82.