Detailed summary of the research paper

SystemDynamicsModelingforIntellectualDisabilityServices-ACaseStudy.pdf

Home >Computer Science homework help >Detailed summary of the research paper

System Dynamics Modeling for Intellectual Disability Services: A Case Studyjppi_342 112..119 Meri Duryan*,†, Dragan Nikolik‡, Godefridus van Merode§, and Leopold Curfs*,§

*Gouverneur Kremers Centrum; †University of Maastricht; ‡Maastricht School of Management; and §Maastricht University Medical Center, Maastricht, the Netherlands

Abstract Organizations providing services to persons with intellectual disabilities (ID) are complex because of many interacting stakeholders with often different and competing interests. The combination of increased consumer demand and diminished resources makes organizational planning a challenge for the managers of such organizations. Such challenges are confounded by significant demands for the optimization of resources and the goal to reduce expenses and to more effectively and efficiently use existing resources while at the same time providing high quality services. The authors explore the possibilities of using “system dynamics modelling” in organizational decision-making processes related to resource allocations. System dynamics suggests the application of generic systems archetypes as a first step in interpreting complex situations in an organization. The authors illustrate the application of this method via a case study in one provider organization in the Netherlands. The authors contend that such a modeling approach can be used by the management of similar organizations serving people with ID as a tool to support decision making that can result in optimal resource allocation.

Keywords: allocation of resources, intellectual disabilities, system dynamics modeling, systems thinking, waiting lists

INTRODUCTION

Healthcare organizations are complex entities as they have multiple stakeholders with often conflicting objectives and goals (Drucker, 1993). Provider organizations specializing in intellec- tual disabilities (ID) are also complex because of the nature of the care and supports they provide and how they are organized. Some of the complexities relate to the difficulties that adults with ID might have in expressing themselves. Moreover, the specifics of the care often require a deeper involvement of carers with respect to their relationships with families and other sectors of society. Because of their complexity, ID provider organizations, com- pared with healthcare providers, often require a higher level of resource planning, collaboration, and cooperation among social, health, and education services, mental health services, and other sectors (WHO, 2010).

To manage the complexities and challenges ID provider orga- nizations face, managers need to analyze and understand complex interdependencies among the systems with which they are dealing. In order to achieve that, ID provider managers need to examine and shift their mental models regarding their role in managing the organization and in establishing relationships with all the stakeholders involved. However, as Forrester (1980) has noted,

traditional management generally does not have other ways to manipulate its mental models but by intuition. The question is, how to go beyond intuition? We posit that many challenges related to decision making in ID provider organizations may be effectively addressed with a “system dynamics modelling” (SD) approach. While SD modeling has been shown to be applicable in many industries (van Ackere, Larsen, & Morecroft, 1993; Braun, 2002; Davenport, 2009; Senge, 1990; Sterman, 2000), it also has been applied in healthcare (Dumas, 1985; Homer & Hirsch, 2006; Kommer, 2002; van Merode, Groothuis, Schoenmakers, & Boersma, 2002; Rohleder et al., 2007; Trochim, Cabrera, Milstein, Gallagher, & Leischow, 2006; Vissers, 1998; Wolstenholme, 2004).

The objective of this article is to explore the application of SD modeling when analyzing decision-making processes within an ID provider organization, with the intent of optimizing allocation of existing resources so as to improve efficiency and effectiveness of resource utilization. Specifically, the focus is on the use of systems archetypes, a class of systems thinking tools, which capture challenges common for any organization. One generic systems archetype, “shifting the burden,” can be used to gain insight into the nature of the dealing with the dilemma of waiting lists (i.e., the backup of admission to a service because of under- capacity and excess demand), a prevalent problem among many such organizations. To illustrate the application of SD modeling, we choose to use a case study conducted in one ID provider organization in the Netherlands. The case study demonstrates the need for a better resource allocation policy, especially considering increased consumer demand and diminishing resources in ID care in the Netherlands as well as in other nations.

Received September 1, 2011; accepted January 5, 2012 Correspondence: Meri Duryan, Gouverneur Kremers Centrum—University of Maastricht, Peter Debeyeplein 1, Room B2.076, Maastricht 6229 HA, the Netherlands. Tel: +31 624575559; Fax: +31 433618388; E-mail: [email protected]

bs_bs_banner

Journal of Policy and Practice in Intellectual Disabilities Volume 9 Number 2 pp 112–119 June 2012

Systems Thinking

Systems thinking is a theory addressing the interrelationships between parts and their connections to a whole system. As Rich- mond (1994) noted, systems thinking is the art and science of drawing conclusions about behavior as a result of deep under- standing of the underlying structure of the system. Within the constructs of this theory, goals and resources are established with a view toward the whole system, rather than artificially allocating them to parts of the system (Plsek & Wilson, 2001; Richmond, 1994; Senge, 1990). The originator of “general systems theory” is Ludwig von Bertalanffy, an Austrian biologist, who in 1954 formed the Society of General Systems Research along with four Nobel Prize winners from economics, physiology, physics, and mathematics (Haines, 2007).

There are a number of approaches that can apply systems thinking to a better understanding and improvement of systems. Those approaches have been characterized by Checkland (1981) as “hard systems” and “soft systems” approaches. The soft systems approach is defined mainly as a learning process designed to determine what needs to be done in not clearly or distinctly defined (i.e., ill defined) problem situations (Checkland, 1981). The hard systems approach is used to determine how to make improvements to a well-defined problem. Hard systems thinking helps us to analyze knowledge directly relevant to clients. It also allows for testing hypotheses related to system behavior via the development of models (Jackson, 2009). In this article, the hard systems thinking approach, termed SD, is applied.

SD is a method of studying complex systems (Sterman, 2000). A hallmark of good SD practice is the skill of seeing the big picture while not losing sight of an operating detail (Senge, 1990). The fundamentals of SD were developed in the late 1950s by Jay Forrester, of the MIT Sloan School of Management, with the establishment of the MIT System Dynamics Group (Forrester, 1961). A key emphasis within SD models is placed on the presence of feedback loops that have a critical impact on the behavior of complex systems and must be considered when designing policies to control those systems (Morecroft, 2007; Senge, 1990; Sterman, 2000).

SD modeling offers a unique opportunity for decision makers to understand the sources of, for instance, low performance in the organization. Many top companies, governments, consult- ing firms, and educational institutions use SD modeling while dealing with complex issues (Sterman, 2000). Forrester (1961) and Senge (1990) emphasized the effect of time delays on the dynamic behavior of the system. They noted that as often there is no explicit connection between cause and effect because of delays, this makes the management of the organization think that the changes made were ineffective. Consequently, such decisions may lead to more changes, which can cause side effects that eventually may throw the system out of balance. For this reason, when man- agers do not get the desired effect, SD practitioners recommend first seeing whether there is a delay involved.

SD involves mapping of system behavior with the help of causal loop diagrams to understand interdependencies between

parts of the system. To be able to see the dynamics of the complex systems over time, SD suggests the use of computer simulation of the problem situation using stocks and flows. Once the computer model is developed, available data are used to quantify the model so that it can be used to simulate various “what-if ” scenarios. The final step in model building is validation of the model via its initialization at past points in time and comparison of results they produce with historical data (Hirsch, 1979).

SD has a number of strengths that make this approach espe- cially useful in healthcare settings applications. First, the dynamic modeling approach involves decision makers with the goal to make the model more realistic. Moreover, decision makers’ involvement ensures that they will use the results produced with the model (Hirsch, 1979). Second, a key strength of SD is that it helps users understand the situation even when there is insuffi- cient data. With the help of causal description of a system, “a model can be developed around important variables regardless of whether or not good data on those variables exist” (Hirsch, 1979; p. 40). The mental models of managers are not always able to see the feedbacks that determine the dynamics of a system. Causal loop diagrams are useful in capturing the mental models of indi- viduals or teams and in communicating the important feedbacks that could be responsible for a problem (Sterman, 2000).

Causal loop diagrams Causal loop diagrams are analytical tools widely used in SD modeling. According to Reynolds and Holwell (2010; p. 32), “a causal loop diagram is a visual tool for the feedback systems thinker.” The diagrams provide a language for articulating our understanding of the dynamic nature of the system studied. Any organization can be viewed as made up of two kinds of system building blocks—reinforcing (positive) and balancing (negative) processes. Arrows can denote the direction of causality between a cause and its effect, and the symbols “s” and “o” at the arrowhead represent “same” and “opposite” directions of causality.

Thus, causal loop diagrams can describe the organization as a system via combination of reinforcing and balancing loops connected together with arrows. Reinforcing processes create exponential growth and collapse; balancing processes keep a situation at equilibrium. Particular combinations of reinforcing and balancing processes within the system can be the reason for system’s complexity (Senge, 1990). Thus, system thinking gives insight into the phenomena that certain patterns of systems behavior recur repeatedly (Senge, 1990). These causal loop struc- tures that capture challenges common for any organization are known as system archetypes.

System archetypes System archetypes were first identified in the 1930s by biologist Bertalanffy and expanded in the 1990s by Senge. They are fundamentally important to SD modeling as they capture the essence of “thinking” in systems thinking (Wolsten- holme, 2004). All archetypes are made up of causal loops with delays. The archetypes can serve as an effective tool for the orga- nization to analyze what causes the same problem to occur repeatedly over time. Difficulties in solving problems often stem from the fact that problems do not occur in isolation but in relation to each other. Archetypes do not describe any one problem specifically; they describe families of problems generi- cally. As Braun (2002; p. 1) has noted, “diagnostically, archetypes

Journal of Policy and Practice in Intellectual Disabilities Volume 9 Number 2 June 2012

M. Duryan et al. • System Dynamics Modeling

113

help managers recognize patterns of behavior that are already present in their organizations.”

Generic archetypes1 can help with the creation of dynamic hypotheses in the beginning of the modeling process. It is often beneficial to use the archetypes in parallel throughout the process of detailed modeling. When managers start thinking in terms of the systems archetypes, their perceptions will recon- dition, and so they will find it easier to identify leverages to solve complex management situations. An archetype modeling approach provides opportunity to experiment on strategic and tactical levels with different realistic scenarios and compare out- comes before making real life changes. Thus, this approach can help to model different resource allocation scenarios in order to find the one that is the most effective and efficient. A number of researchers have used generic systems archetypes as a first step in interpreting complex situations in healthcare (e.g., Dumas, 1985; Homer & Hirsch, 2006; Kommer, 2002; Rohleder et al., 2007; Trochim et al., 2006; van Merode et al., 2002; Vissers, 1998; Wolstenholme, 2004).

Modeling in Healthcare

Modeling helps decision makers better understand the complexity of an organization. Decision makers in healthcare organizations need a model that can help them allocate resources optimally and increase performance with existing resources. It is very important to develop an organization’s capacity to work with mental models as they are too often not made explicit (Senge, 1990). Having a language that everyone across the orga- nization can understand will definitely enhance performance improvement. For several decades, SD modeling has been used to more effectively understand the challenges that healthcare managers face (Harper, 2002; Harper & Gamlin, 2003; Hirsch & Miller, 1974; Homer & Hirsch, 2006; Lane, Monefeldt, & Rosenhead, 2000; Levin & Roberts, 1976; Sundaramoorthi, Chen, Rosenberger, Kim, & Buckley-Behan, 2010).

One of the success stories related to simulation modeling in healthcare settings is the redesign and implementation of patient service centers in Calgary, Canada (Rohleder, Bischak, & Baskin, 2007), in which a simulation model application was used to suggest ways to reduce demand variability and improve resource utilization in the city’s laboratory services. More directly relevant to ID, in the Netherlands, a waiting list model was developed for residential services for adults with ID (Kommer, 2002). The model helped find the critical factors influencing waiting lists for services. The model became an “eye-opener” for local policymak- ers as it revealed clearly that older and aged adults with ID also need to be considered in residential provision. It was found that there was a greater demand by older aged adults with ID for more intensive care in comparison with that evidenced by younger aged adults with ID. After the Dutch Ministry of Health, Welfare, and Sports used the model in developing strategies to address the waiting list problem, there was a 34% smaller waiting list (Kommer, 2002).

Another successful example of simulation modeling was nurse-to-patient assignment in a hospital in Texas, USA (Dumas,

1985). As the healthcare system in the United States has a shortage of nurses (noted to be about 12% in 2010), it was important to optimize the nurse-to-patient assignments to avoid nurse burn- outs. The research made a significant contribution to the scien- tific management of nurse-to-patient assignments. It was done by introducing a tool to evaluate different nurse-to-patient assign- ment policies with the goal to identify the one that was most optimal.

The World Health Organization (de Savigny & Adam, 2009) report on “Systems Thinking for Health Systems Strengthening” suggests that application of systems thinking does not necessarily mean that the challenges organizations come across will be easily solved without considerable changes in the nature of the organi- zations. However, it will help to identify where some serious challenges lie. SD modeling has been considered to be an appro- priate method for improving healthcare systems and has been used in healthcare environments to explore policies for ongoing operations (Sundaramoorthi et al., 2009). As such, the approaches suggested can be used by senior managers of even more complex forms of healthcare provider organizations, like those designed to provide for adults with ID, as a tool to support resource allocation decision making.

ID Services Provision

Complexity of ID provider organizations Healthcare providers are complex forms of provider organizations. However, there are specifics related to healthcare services for adults with ID that add to the complexity. It is well known that many adults with ID have certain limitations in communication and social skills and some have difficulties in caring for themselves. Thus, ID provider orga- nizations should utilize a higher level of planning for resources, should have better trained staff, and should collaborate more both with families and with other stakeholders. The involvement of family members of adults with ID in the process of care is crucial. Moreover, the family members and carers of the adult are themselves considered as a high-risk group with a significant need for support (WHO, 2010).

Some challenges that healthcare organizations face are common in nearly all developed countries (Akcali, Coté, & Lin, 2006; Carter, 2002; Harper & Gamlin, 2003; Polder, Meerding, Bonneux, & van der Maas, 2002; Smits & van der Pijl, 1999; Vissers, 1998). In the Netherlands specifically, ID service providers are under pressure because of planned reforms and changes in society, politics, and consumer expectations. The combination of increased consumer demand and diminished resources is pro- gressively a more urgent challenge for ID provider organizations that aim to improve the services they provide via delivering more person-focused care. One solution could be effective and efficient resource capacity planning. The approach can give managers more insight into their own decision-making processes and help them apply a new way of thinking through resource allocation problems.

ID services in the Netherlands A large share of total healthcare costs in the Netherlands is taken up by healthcare for people with ID. Funding for ID services in the Netherlands is provided on the basis of the Exceptional Medical Expenses Act (Algemene Wet Bijzondere Ziektekosten (AWBZ)). Over the last decade, the1Tools that reveal patterns of behavior in systems.

Journal of Policy and Practice in Intellectual Disabilities Volume 9 Number 2 June 2012

M. Duryan et al. • System Dynamics Modeling

114

expenditure on the AWBZ care has risen from just €14 billion in 2000 to over €23 billion in 2010. All parties participating in the AWBZ need to express the healthcare they supply in so-called indexed care packages (Zorg Zwaarte Pakket (ZZP) (severity of care packages)) that are combinations of different services and care functions. For long-term care, these ZZPs range from ZZP1 (sheltered living with some assistance) to ZZP8 (sheltered living with very intensive care) (ZZPs 9 and 10 concern rehabilitative care and palliative care, respectively) (Mot, 2010). To initiate funding, every request submitted to AWBZ funding must be assessed by an independent organization, the Care Assessment Centre (Centrum Indicatiestelling Zorg or CIZ). The Association for Care of the Disabled (Vereniging Gehandicaptenzorg Neder- land (VGN)) is the umbrella Dutch association representing individual provider organizations that provide professional care and support to persons with mental, physical, sensory, and/or communication disabilities. The VGN encompasses some 173 member organizations providing care to about 110,000 clients and accounts for a budget of €4.8 billion.

Community-based services in the Netherlands have evolved as alternatives to institutional residential care, although the exist- ing service structure remains dominated by institutional models (WHO, 2010). In 2010, more than 600,000 people in the Nether- lands made use of long-term care.2 However, it is projected that the prevalence of people with ID needing specialized long-term care in the Netherlands is going to increase because of the growth in the number of older and aging adults with ID (Kommer, 2002). Thus, ID services provision in the Netherlands, as in many other countries, is confronted today with pressures associated with a growing demand for medically oriented care and services (because of such factors as a “graying” population) and by a growing emphasis on standards for quality of life (Smits & van der Pijl, 1999).

Given that the costs of healthcare for people with ID as well as for those with mental disorders will inevitably increase (Jansen et al., 2004), there is currently a demand for the optimization of resource allocations—with the goal of reducing expenses and more effectively using the existing resources, while at the same time providing high-quality services. One of the solutions to this challenge can be a SD approach in analyzing healthcare systems, as suggested in the WHO report on health systems strengthening (de Savigny & Adam, 2009). The WHO promotes systems think- ing as a core approach to health systems analysis (de Savigny & Adam, 2009).

A Case Study on Underutilization of Resources

Background information According to the WHO European Ministerial Conference on Health Systems, particular attention should be given to improvement of the performance of health service delivery by making health systems more “patient-focused” (WHO, 2008). With this approach, it is important to keep the client in the center of focus especially in ID services provision and to respect his or her wishes, concerns, values, and priorities. It is equally important to give a prompt response to the needs of

adults with ID and to reduce waiting times for their seeing health practitioners. That is a challenging task for ID care providers facing diminished financial resources when they have to ensure optimal utilization of existing resources.

Given this, a study was conducted in one ID provider organi- zation in the Netherlands to determine how such outcomes could be best attained. This organization provides AWBZ-funded care for about 1,591 persons in a number of discrete residential facili- ties. The services are provided through nine divisions that are geographically positioned in different provinces of the Nether- lands. The complexity of the ID organization evolved not only from the critical nature of the organization’s mission to deliver a full package of services (i.e., “full-service concept”) but also by the diversity of services needed and geographic locales as well as the variety of stakeholders and employees. In order to manage the complexities and challenges arising from being a multilocational organization operating in a dynamic environment, managers needed to analyze and understand the complex interdependen- cies among the systems with which they were dealing. Resource capacity planning was thus essential for this kind of organization because ID services and supports are dynamic and not commodi- ties that can be stocked and stored.

The study was conducted within one of three divisions of the organization that provided similar kinds of services in the areas of living, working, day care, leisure, treatment, and support. Specifi- cally studied were the processes related to dealing with budget gaps when there was a shortage of clientele in one division of the organization and when there was a surfeit of persons on waiting lists in other divisions of the organization. We found that when a client applied to any division of the organization with appropriate indication from CIZ that he or she required 24-h residential supports in that division, there may not have been free space at that moment. When this occurred, the intake team of that divi- sion put the client on a waiting list, even if there was an empty bed in a residence in one of the other divisions. This happened because the waiting lists were not shared among the organiza- tion’s divisions. We determined that one reason they did not share the waiting lists was because of the funds that the client brought to the division. To be able to obtain and retain the funds, the division could keep the client by putting him or her on a waiting list while providing alternative care services (e.g., home visits). However, if the client disagreed with the suggested alter- native services and did not want to, or could not, wait some 3 months until there was a free space, then he or she would leave the division and, most of the time, the organization (thus resulting in a loss of funds for the organization). There are many competitors in the field and considering the fact that adults with ID are free to choose their provider organization, it is critically important for ID providers to improve their services and to avoid putting clients on a waiting list for an extended period of time—especially when there are residential spaces in other divisions. As there could be different scenarios for the organization to use to overcome this challenge and to utilize resources more effectively, SD tools could help to model different “what-if scenarios” to see the outcome of a strategic decision before its actual implementation.

Methodology To see how this organization might employ stra- tegic decision making in optimizing resource allocations, we used a case study approach. Such a case study approach has been

2Program letter concerning long-term care. Parliamentary document, Ministry of Health, Welfare and Sport, June 1, 2011.

Journal of Policy and Practice in Intellectual Disabilities Volume 9 Number 2 June 2012

M. Duryan et al. • System Dynamics Modeling

115

defined as an empirical inquiry that “investigates a contemporary phenomenon within its real-life context when the boundaries between phenomenon and context are not clearly evident and in which multiple sources of evidence are used” (Yin, 1984; p. 13). SD methodology, the same as a case study research, uses qualita- tive data to develop quantitative simulation models. Data- gathering techniques such as observations, interviews with clients and senior- and middle-level management of the organization, focus groups, and archival records were used. Following a case study research guidelines, the most important sources of the information were the interviews (Yin, 1984). We translated the interview scripts into the SD language of feedback loops.

As the organization’s senior management indicated that resource sharing issues (particularly waiting list sharing) were of primary importance due to the scarcity of resources and increas- ing consumer demand, we started our research with the challenge that concerns the managers the most. The logic of causal loops was revised in collaboration with the senior management of the organization to ensure that the processes were clearly understood by the modeler. Following the causal loop design, we found patterns related to one of the generic archetypes, “shifting the burden.” The archetype describes a situation when people take actions in response to the problem using quick fixes (Kim, 1999). The “shifting the burden” archetype helped to map the impli- cations of the decisions made to solve the problem related to underutilization of resources. This approach helped decision makers identify additional data needed to provide a complete picture of the forces affecting the performance of the system they managed.

Description of “shifting the burden” system archetype “Shifting the burden” is one of the generic SD’s archetypes that demon- strate the tension between short-term (symptomatic) solutions to visible problems and the long-term impact of fundamental solu- tions that consider the pattern of behavior (Kim, 1999; Senge, 1990). The archetype is composed of two balancing loops (B1 and B2) and a reinforcing loop (R) (see Figure 1). As Figure 1

demonstrates, there are two options to solve the problem. The first and the quick one is application of a short-term fix to the symptom. The second option, application of a long-term fix to the fundamental issue, is more time consuming and is harder to implement (Kim, 1999).

Short-term solutions relieve the symptom only and do not fix the problem. Moreover, quick solutions lead to unintended consequences that make the situation even worse in the long run, because the underlying problem persists and so the problem will reappear over time. Decision makers do not anticipate solutions- induced side effects as they act as if cause and effect are always closely connected in time and space (Sterman, 2000). Side effects worsen the situation in the long run, because symptomatic solu- tions reduce the tendency toward the fundamental solution and also they reinforce the perceived need for the symptomatic solu- tion. As Braun (2002; p. 4) noted, it “hypothesizes that once a symptomatic solution is used, it alleviates the problem symptom and reduces pressure to implement a fundamental solution, a side effect that undermines fundamental solutions.” The long-term fix (fundamental solution) requires building new capacity. It takes time because of delay before its effects will be visible and requires deeper understanding of the underlying problem as well as more commitment and patience (Braun, 2002). Considering the pres- sure managers usually face, they prefer fixing problems as soon as possible and move on. So, they do not see the need for exten- sive efforts and resources to identify and solve the fundamental, systemic problem (Braun, 2002).

Illustration of “underutilization of resources” problem via systems archetype The “shifting the burden” archetype in Figure 2 dem- onstrates the tension between fundamental and symptomatic solutions in relation to the issues that arise from not sharing the waiting lists between divisions. The problem symptom in this particular case is periodic shortage of clients, which means that

problem symptom

apply symptomatic

solution

(B1)

apply fundamental

solution

(B2)

solution-induced side effect(R)

FIGURE 1

“Shifting the Burden” system archetype.

resources underutilized

cut costs

(B1)

share waiting lists

(B2)

o attractiveness to

new clients

new clients admitted

consumer/employee satisfaction

(R)

employees’ schedule shifts

FIGURE 2

Causal loop diagram of the case study.

Journal of Policy and Practice in Intellectual Disabilities Volume 9 Number 2 June 2012

M. Duryan et al. • System Dynamics Modeling

116

the organization has resources in terms of beds and workforce that are not used efficiently. The shortage of clients leads to budget gaps, and the division must react as soon as possible in order not to incur significant losses of funds.

The easiest and fastest (symptomatic) solution in this case is to cut costs (loop B1) via reduction of the workload of employ- ees that have flexible contracts or via their temporal reallocation. Reduction of the workload leads to reduction of income, and employees have to look for part-time jobs in other places. The employee reallocation leads to changes in their schedules, which can cause client and employee dissatisfaction. Employees feel uncomfortable as they have to organize their lives around new schedules. In the long run, shifts in schedule and salary reduc- tion may bring tension and may even cause employees to ques- tion their loyalty to organization, which can cause increases in employee turnover. Clients may not be satisfied with the shifts as they get stressed because of the discontinuity in relation with care workers and frequent changes of people with whom they interact.

Adults with ID, especially those residing in congregate set- tings, “may develop emotional insecurity as a result of disconti- nuities in the care, as well as competition for limited amounts of personal attention” (Schuengel et al., 2010; p. 39). The research on personal attachment has shown that “an attachment figure may be especially important for people with intellectual disabil- ity, because they are less adept in dealing with stressful situations on their own” (De Schipper & Schuengel, 2010; p. 585). Attach- ment behavior may be “part of young persons’ adaptation to the stresses and challenges of group care” (De Schipper & Schuengel, 2010; p. 584).

In ID care situations, the care worker also serves as the point of contact between parents and the service system and plays an important role in the effectiveness of the delivery of services (WHO, 2010). Thus, the established relations between a care worker and the adult with ID are very important, and the negative impact of frequent shifts cannot be underestimated. Relations of trust between the care worker and the client become a very impor- tant indirect resource, especially in ID support situations. That intangible resource that reflects clients’ feelings and expectations takes time to accumulate; however, it can be destroyed rapidly. Warren (2002; p. 120) noted that “damage to intangible resources has powerful effects on tangible factors.” In this particular situa- tion, it may affect the number of clients who apply for admission. The organization may no longer seem attractive to new clients, which in turn can lead to a lower level of people seeking its services (i.e., admission). Reduction in client admissions reinforces the need for a fundamental solution while at the same time making it less feasible, thus less favorable and hence overdue.

If consideration is not given to the side effects of budget cuts and employee shift changes, any attempts to stabilize the system may destabilize it even more. The fundamental solution to this problem (loop B2) might be the sharing of information among the divisions about individuals on the waiting lists. The solution can be considered fundamental because it influences the problem’s cause rather than the problem’s symptom. If information about the clients who are on the waiting lists is shared among the divisions, the organization as a whole will not lose the clients. Moreover, the division will be able to provide a client with quick information on potential accommodations in other divisions. By doing this, the

organization will provide better customer services, especially in cases where there is an urgent need for residential care.

This fundamental solution can also bring about the need to share information on schedules of employees with flexible con- tracts, which will improve the organization’s flexibility to adapt better to changes caused by reduction in funding. As any other fundamental solution, this one will have time delays and will require changes in attitudes of the management. The decision makers have to see the problem not only from the perspectives of their divisions but also from the perspective of the whole orga- nization. In order to apply a fundamental solution, the heads of divisions should develop a vision of the organization they want to build up because problem-solving attitudes will always push them to apply quick fixes.

There is a common outcome from the application of symp- tomatic solutions over time, which is “addiction” to the symptom- atic solution and desire to see immediate results, even though they are short term and do not contribute to the vision of the organi- zation. There are clients who leave the divisions of the organiza- tion as they do not feel comfortable with getting alternative services (home visits) instead of permanent residential care. By suggesting alternative services to patients, the division attracts additional funds to the organization. Moreover, it helps to fill the budget gaps when there are empty beds. Thus the division tries to keep the clients by providing alternative services to avoid budget gaps in the future, and so it has no incentives in suggesting their clients apply to any other division of the organization. As the division does not have empty beds at that moment, it does not incur losses but the organization as a whole does.

DISCUSSION

We all tend to look for causes near the events we want to understand. Moreover, we pay more attention to symptoms rather than the underlying causes of the problem (Sterman, 2000). However, in complex systems like organizations providing services to people with ID, cause and effect are distant in time and space. SD modeling helps determine the dynamics of a system through learning about feedback processes, time delays, and nonlinearities in the system.

SD modeling also takes into account the multiple pers- pectives of all the stakeholders involved. Thus, considering the fact that one of the major reasons for the ID provider organi- zations’ complexity is having multiple stakeholders with con- flicting objectives and goals, the SD approach can be quite feasible. The generic system archetypes with causal loop dia- grams provided the language that helped explore thinking around complex situations in ID services, for instance, phenom- ena such as sharing of waiting lists. SD approaches can help develop the capacity to get a better understanding of the situa- tion to an extent where, in collaboration with the relevant stake- holders (e.g., organization’s management, care workers, and persons with ID and their families), a strategy may be developed that will address the problem in a way to minimize the likeli- hood of unintended consequences.

SD modeling is considered to be appropriate for the case example described in this article as it also allowed consider- ing the intangible resources that reflected clients’ feelings and

Journal of Policy and Practice in Intellectual Disabilities Volume 9 Number 2 June 2012

M. Duryan et al. • System Dynamics Modeling

117

expectations. Those resources are of significant importance in ID services provision as they have powerful effects on tangible factors. The process of mapping the problem with the help of systems archetypes facilitated sharing of mental models of the management of the organization. It also demonstrated that systems approaches may help ID care managers to simplify the process of their thinking about complex realities and to overcome the feeling of helplessness when confronted with complex prob- lems. Systems archetypes facilitated understanding of the full range of feedbacks operating in the system. They helped to realize that the fundamental solution (i.e., sharing waiting lists) can help to minimize side effects caused by quick fixes.

To understand system behavior over time, it is necessary to visualize the dynamics of system behavior using real data. Avail- able data can be used to quantify the model and observe real world systems via computer simulation. The simulated system model can be validated by initializing it at past points in time and comparing the results it produces with historical data. Afterward, it can be used to simulate various “what-if ” scenarios before their actual implementation (Hirsch, 1979; Morecroft, 2007).

ACKNOWLEDGMENTS

The authors would like to thank managers and employees of Koraal Groep who contributed considerably to this study.

REFERENCES

van Ackere, A., Larsen, E., & Morecroft, J. (1993). Systems thinking and business process redesign. European Management Journal, 11, 412–423.

Akcali, E., Coté, M., & Lin, C. (2006). A network flow approach to optimizing hospital bed capacity decisions. Health Care Management Science, 9, 391–404.

Braun, W. (2002). The systems archetypes. Retrieved from http://wwwu. uni-klu.ac.at/gossimit/pap/sd/wb_sysarch.pdf

Carter, M. (2002). Diagnosis: Mismanagement of resources. ORMS Today, 29, 26–32.

Checkland, P. (1981). Systems thinking, systems practice. Chichester, UK: John Wiley & Sons Ltd.

Davenport, T. H. (2009). Make better decisions. Harvard Business Review, 87, 77–87.

Drucker, P. (1993). The new realities. New York: Harper & Row. Dumas, B. (1985). Hospital bed utilization: An implemented simulation

approach to adjusting and maintaining appropriate levels. Health Services Research, 20, 43–61.

Forrester, J. (1961). Industrial dynamics. Cambridge, MA: MIT Press. Forrester, J. W. (1980). Information Sources for Modeling the National

Economy. Journal of the American Statistical Association, 75, 555– 566.

Haines, S. (2007). Strategic and systems thinking: The winning formula. San Diego, CA: Systems Thinking Press.

Harper, P. (2002). A framework for operational modeling of hospital resources. Health Care Management Science, 5, 165–173.

Harper, P., & Gamlin, H. (2003). Reduced outpatient waiting times with improved appointment scheduling: A simulation modelling approach. OR Spectrum, 25, 207–222.

Hirsch, G., & Miller, S. (1974). Evaluating HMO policies with a computer simulation model. Medical Care, 12, 668–671.

Hirsch, G. (1979). System Dynamics modeling in health care. ACM SIGSIM Simulation Digest, 10, 38–43.

Homer, J., & Hirsch, G. (2006). System dynamics modeling for public health: Background and opportunities. American Journal of Public Health, 96, 452–458.

Jackson, M. (2009). Fifty years of systems thinking for management. The Journal of the Operational Research Society, 60, 24–32.

Jansen, D., Krol, B., Groothoff, J., & Post, D. (2004). People with intellec- tual disability and their health problems: A review of comparative studies. Journal of Intellectual Disability Research, 48, 93–102.

Kim, D. (1999). Shifting the burden: Moving beyond a reactive orienta- tion. The Systems Thinker, 10, 5–6.

Kommer, G. J. (2002). A waiting list model for residential care for the mentally disabled in the Netherlands. Health Care Management Science, 5, 285–290.

Lane, D., Monefeldt, C., & Rosenhead, J. (2000). Looking in the wrong place for healthcare improvements: A system dynamics study of an accident and emergency department. The Journal of the Operational Research Society, 51, 518–531.

Levin, G., & Roberts, E. (1976). The dynamics of human service delivery. Cambridge, MA: Ballinger Pub.

van Merode, G., Groothuis, S., Schoenmakers, M., & Boersma, H. (2002). Simulation studies and the alignment of interests. Health Care Management Science, 5, 97–102.

Morecroft, J. (2007). Strategic modelling and business dynamics: A feedback systems approach. Chichester, UK: John Wiley & Sons.

Mot, E. (2010). The Dutch system of long-term care. CPB Netherlands Bureau for Economic Policy Analysis. Hague, the Netherlands. Retrieved from http://www.cpb.nl/en/publication/dutch-system- long-term-care

Plsek, P., & Wilson, T. (2001). Complexity science: Complexity, leader- ship, and management in healthcare organisations. British Medical Journal, 323, 746–749.

Polder, J., Meerding, W., Bonneux, L., & van der Maas, P. (2002). Health- care costs of intellectual disability in the Netherlands: A cost-of-illness perspective. Journal of Intellectual Disability Research, 46, 168–178.

Reynolds, M., & Holwell, S. (2010). Systems approaches to managing change: A practical guide. London: Springer.

Richmond, B. (1994, July). System dynamics/systems thinking: Let’s just get on with it. Paper presented at the 12th International Conference of the Systems Dynamics Society, Stirling, Scotland.

Rohleder, T., Bischak, D., & Baskin, L. (2007). Modeling patient service centers with simulation and system dynamics. Health Care Manage- ment Science, 10, 1–12.

de Savigny, D., & Adam, T. (2009). Systems thinking for health systems strengthening. Geneva: WHO.

De Schipper, J. C., & Schuengel, C. (2010). Attachment behaviour towards support staff in young people with intellectual disabilities: associations with challenging behaviour. Journal of Intellectual Dis- ability Research, 54, 584–596.

Schuengel, C., Kef, S., Damen, S., & Worm, M. (2010). People who need people: Attachment and professional caregiving. Journal of Intel- lectual Disability Research, 54, 38–47.

Senge, P. (1990). The fifth discipline: The art and practice of the learning organization. New York: Bantam Doubleday, Dell Publishing Group, Inc.

Smits, M., & van der Pijl, G. (1999, January). Developments in hospital management and information systems. Paper presented at the 32th Hawaii International Conference on System Sciences, Maui, Hawaii.

Sterman, J. (2000). Business dynamics: Systems thinking and modeling for a complex world. New York: McGraw-Hill.

Sundaramoorthi, D., Chen, V., Rosenberger, J., Kim, S. B., & Buckley- Behan, D. (2009). A data-integrated simulation model to evaluate nurse–patient assignments. Health Care Management Science, 12, 252–268.

Journal of Policy and Practice in Intellectual Disabilities Volume 9 Number 2 June 2012

M. Duryan et al. • System Dynamics Modeling

118

Sundaramoorthi, D., Chen, V., Rosenberger, J., Kim, S. B., & Buckley- Behan, D. (2010). A data-integrated simulation-based optimization for assigning nurses to patient admissions. Health Care Management Science, 13, 210–221.

Trochim, W., Cabrera, D., Milstein, B., Gallagher, R., & Leischow, S. (2006). Practical challenges of systems thinking and modeling in public health. American Journal of Public Health, 96, 538–546.

Vissers, J. (1998). Patient flow-based allocation of inpatient resources: A case study. European Journal of Operational Research, 105, 356–370.

Warren, K. (2002). Competitive strategy dynamics. New York: John Wiley & Sons.

WHO. (2008). European ministerial conference on health systems: Health systems, health and wealth. Retrieved from http://www.euro. who.int/__data/assets/pdf_file/0006/78918/E92150.pdf

WHO. (2010). Better health, better lives: Children and young people with intellectual disabilities and their families. Retrieved from http://www.euro.who.int/__data/assets/pdf_file/0003/126570/ e94430.pdf

Wolstenholme, E. (2004). Using generic system archetypes to support thinking and modeling. System Dynamics Review, 20, 341–356.

Yin, R. K. (1984). Case study research: Design and methods. Beverly Hills, CA: SAGE Publications, Inc.

Journal of Policy and Practice in Intellectual Disabilities Volume 9 Number 2 June 2012

M. Duryan et al. • System Dynamics Modeling

119

content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's

express written permission. However, users may print, download, or email articles for individual use.