Dutch General Hospital Building Investment Decision Model

(1)

Dutch General Hospital Building Investment Decision Model

The New Gelre Hospital, Source: Royal Haskoning

Master Thesis Technology Management

_________________________________________________________________________

Author: Joppe Bijlsma

joppebijlsma@gmail.com Student number: 1333577

Study program: Technology Management

Faculty of Economics and Business

University of Groningen

Company: Deloitte Consulting B.V. Supervisor Deloitte: Drs. P. Princen

Supervisors University: Dr. ir. H. van de Water Dr. B. Crom

(2)

ii

ABSTRACT

The new reimbursement system starting in 2009 will change building investment decisions for hospitals. The hospital building investment decision will then be the responsibility of the General Hospital Management. New criteria for assessing the hospital building investment decision are being developed and the importance of these criteria is as yet unknown. This thesis is written in order to construct a hospital building investment decision model. This model is a formal

conceptualization of the relationship of the various factors that are relevant in decision-making and the influence of these factors on the decision.

In literature we did conclude that building investment decisions can be evaluated from three perspectives: strategic, functional and financial perspective. Criteria are found for the different perspectives. In this literature study three problems are identified: criteria reported in literature still need to be validated, it is unclear how cash flow should be predicted and it is also unclear how discount rate should be determined. A Delphi study within Deloitte Consulting is conducted to address these problems. The criteria are validated and adapted with the Delphi study. AHP Analytic Hierarchy Process(AHP) can be used to rank the different criteria. The AHP is performed within Deloitte and a general hospital. The results are used to revise the criteria and limitations are discussed. No consistent ranking of the different criteria can be determined in general, but the model can be used to facilitate a more objective discussion about the hospital building investment decision and to check the completeness of the hospital building investment business case. Finally suggestions for further research are given.

(3)

iii

It is not the goal but the way there that matters--and the

harder the way, the more worthwhile the journey

1

.

My study can be seen as an exciting journey and it was fun because of my fellow travelers: Three groups of travelers were with me during the complete journey: Nitro, BMD and C&D. Some other groups and persons which took a short trip with me on the way are the BIG Board 04//05, my Hong Kong Group, Klaas-Jan in Eindhoven and my TM friends.

For my last journey I want to thank a couple of people within Deloitte: Peter Princen, the participants of the research and my fellow interns Desiree, Jan, Erik, Martijn, Jasper, Joep and Rachelle. The supervisors of the university Hen van de Water and Ben Crom. Finally, Maud for all her support.

My journey could also be described by if you don't know where you are going, any road will take you there2. Therefore I want to especially thank my parents, who always agreed when I was changing my destination on the way.

Amsterdam, 20 October 2008 Joppe Bijlsma

1_{Sir Wilfred Thesiger, British explorer (1910-2003)}

2

(4)

iv

1 INTRODUCTION

“Construction of new Isala Hospital postponed.

The Board of the Isala Hospital decided after consultation with the Medical Staff to postpone the construction of the new Isala hospital. The Board wants to get more insight in the implications of future regulation changes and the financial consequences of constructing a new hospital.” (Isala Klinieken, 2007)

The Netherlands is currently undergoing a market-oriented health care reform which started in 2005 (Schut, 2005). This health care reform will influence the reimbursement system for General hospitals, especially the reimbursement of hospital buildings. In this chapter the current and future hospital reimbursement systems for Dutch General hospitals will be explained.

Ninety percent of the revenue for General Hospitals is patient care reimbursement according to Price Waterhouse Coopers (2007). Patient care reimbursement is currently split in two parts: the direct patient related reimbursement and hospital building reimbursement. From 2009 onwards the hospital building reimbursement will be integrated with direct patient care reimbursement (Ministry of Health, 2008a). First the direct patient care related reimbursement will be explained. Direct patient care revenues are currently allocated by DBCs. A DBC3, Diagnose Treatment Combination, defines the whole of the hospital and medical specialist activities and services per year arising from the demand for care by a patient consulting a specialist in a hospital for one defined illness (Custers, 2007). Direct patient care revenue is registered in two categories: the A segment and B segment. Both are registered in the same way, but the revenue allocation drivers and the regulation is different between the two segments. A segment DBC‟s revenue is allocated based on the budget parameters and B segment DBC‟s revenue is allocated based on the

production of DBC‟s. The revenue allocation drivers of the different segments are described in more detail below.

3

(7)

2 The A segment revenue is based on fixed budget parameters and variable budget parameters. Hospitals reimbursement based on fixed budget parameters is independent of the produced amount of DBC‟s. The fixed budget parameters are based on three drivers: the size of the

population residing in the hospital‟s clinical catchments‟ area, the number of authorized beds and medical specialist units. Reimbursement based on variable parameters is negotiated upfront with health care insurers on a yearly basis. The variable budget parameters are based on four drivers: hospital admissions, inpatient days, outpatient visits and daycare visits (Custers,2007). The DBC reimbursement rates for the A segment are set centrally by the Dutch Health Care Authority (NZA) and the health insurers have to negotiate contracts with every hospital (Schut,2005). Hospitals are in this way guaranteed a certain amount of income, because the A segment revenue is partly independent of production and health insurers have to negotiate contracts with every hospital using a centrally set DBC reimbursement rates.

The B segment revenue is directly allocated based on the DBC‟s produced by the hospital and is dependent on production. Regulation for the B segment is different from the A segment in two additional aspects: health insurers don‟t have to negotiate a contract with every hospital. Health insurers can choose the hospitals which they want to negotiate a contract with. In the B-segment the reimbursement rates are negotiable instead of fixed centrally (Schut, 2005). This makes the revenue for the B segment more uncertain.

(8)

3 Table 1: Patient care financing for hospitals in 2008

There are important regulation changes impacting the reimbursement of Dutch General Hospitals in the next five years. First more of the registered DBC‟s and accompanied revenues will be shifted from the A to the B segment. This is shown graphically in Figure 1. Therefore, more of the hospital revenue will be financed by the B segment, depending on the actual production. This will make the hospital revenue more uncertain.

Patient care reimbursement drivers for hospitals from 1 January 2008

Direct patient care Hospital building

A segment B segment

Fixed budget parameters

 the size of the population  the number of beds  medical specialist units Variable budget parameters

(9)

4 Figure 1: Percentage of reimbursement in the B segment Dutch Patient Care based on Ministry of Health, 2008a, 2008d

A second regulation change is in the hospital building reimbursement. Currently hospitals are reimbursed for their interest and deprecation at the end of the year. According to the Ministry of Health this reimbursement system doesn‟t give hospital incentives to be cost-conscious. Hospital management tries to construct a hospital building within the standards instead of constructing the most efficient building (Ministry of Health, 2005b). According to the Ministry of Health a new system is needed that will give hospitals an economic incentive to be efficient. A new

reimbursement system will be introduced in 2009. The DBC tariffs will also include

reimbursement for the hospital building. This new tariff which includes building reimbursement in combination with the current DBC tariff is called full DBCs tariffs4. This new tariff is a markup over the current DBC tariff. This new situation is described in Table 2. Actual hospital building costs are largely fixed costs. Hospitals are currently reimbursed for these costs

independent of their DBC production. In the new system for every DBC produced a markup is given for reimbursement of the hospital building. If less DBC‟s are produced hospitals will get less building reimbursement than their hospital building costs. If hospitals produce more DBC‟s they will get more reimbursement than the actual hospital building costs. This new

reimbursement system will also influence the interest rate hospital financers will ask for their hospital building financing. Because the repayment of debt becomes more uncertain probably the financers of hospital debt will ask for a higher interest rate (Ministry of Health, 2008b).

4

In Dutch: integrale tarieven 0% 10% 20% 30% 40% 50% 60%

B-segment DBCs as percentage of total DBCs registered

(10)

5 According to the Council for Public Health and Health Care (RVZ, 2006a) there are two risks when implementing this new system. The accessibility of health care could be reduced when hospitals close (temporary) because they are unable to fulfill their building debt payments. The quality of hospital buildings can fall, because not enough money is invested in maintenance. The Council for Public Health and Health care states that the benefits of the new system will

outweigh the cost, but this needs to be proven.

This markup tariff will be set on central level and will be an average of the current

reimbursement of all hospitals. The interest and deprecation cost of hospital buildings decrease over time. Hospitals that recently were constructed or are constructing a new building have a disadvantage over hospitals that have older buildings. The Ministry of Health (2007) researched the financial impact of this new reimbursement system and concluded that no individual hospital will be forced to close in this new reimbursement system. This needs to be proven when the new system is applied.

The current Ministry of Health policy will introduce a law in 2012 which allows hospitals to make profit, but this law needs approving of the Dutch House of Representatives. Most hospitals currently have a non-profit character (Ministry of Health, 2008e). An overview of regulation in 2009 is shown in Table 2.

Patient care financing drivers for hospitals from 1 January 2009

Direct patient care Hospital building

(11)

6 Fixed budget parameters

 the size of the population  the number of beds  medical specialist units Variable budget parameters

 hospital admissions  inpatient days  outpatient visits  daycare visits Number of DBC‟s in current year Markup on DBC tariff A and B segment

Table 2: Regulations for General Hospitals from 2009

This change in regulation doesn‟t only change the reimbursement of hospitals. Financers are also demanding changes in the hospital design, because hospital income now depends on production (Bouwcollege, 2007). Hospitals are currently designed as integral buildings. By dividing a hospital in different layers some parts of the hospital could be used by external parties when production is less than expected. The hospital beds for instance can be used as hotel rooms. The question is if a hospital can combine patient care and hotel activities at the same time. Hotel activities are not part of the primary process of a hospital. It could bring practical problems when a hospital is used for two separated functions at the time. This combination of activities is not implemented yet in daily practice; feasibility of this combination is therefore not yet known.

The Netherlands Board for Healthcare Institutions (NBHI) describes four layers to be used in a hospital: hot floor, hotel, office and industry. Hot floor, comprises the high-tech, capital intensive functions that are specific for hospitals. The hotel comprises all functions for accommodation of patients. The functions for diagnostics and simple examinations and treatments are

accommodated in the office. The office also accommodates specific office facilities, such as staff accommodation, accounting and management. Finally industry accommodates all medical

supporting and facilitating functions e.g. the laboratory (Bouwcollege,2007).

(12)

7 In this new situation hospital building investment should be carefully evaluated. First, hospitals have to analyze the financial consequences of their investment decisions, because their

depreciation and interest costs are not being reimbursed directly. Second, financers could ask for other design requirements for hospital buildings, for instance making the hospital suitable for external parties.

(13)

8

2 RESEARCH DESIGN

2.1 Problem statement

There are 85 General Hospitals in the Netherlands (PWC, 2007). Prior research by Kamath and Oberst (1992) indicated a difference in the assessment of investments between profit and non-profit hospitals. At the start of this research interviews were conducted with consultants working within Deloitte Consulting who had been involved in advising General Hospitals about building investment decision. One of the questions asked was whether hospital management teams judge building investment decisions from a profit or a non-profit perspective. According to the Deloitte experts hospital managers judge building investment decision from a non-profit perspective. This thesis is therefore written for non-profit general hospitals.

Currently hospitals provide a building proposal to the Netherlands Board for Healthcare

Institutions. The NBHI decides whether a hospital building may be build and then guarantees the financing of the building independent of patient output. In the new situation managers of Dutch General Hospital have to make the hospital building investment decision themselves instead of making a proposal. Reiter and others (2000) state that there is little systematic information available about the current capital budgeting practices of health care providers. This lack of information increases the possibility of ill-advised decisions. This research addresses the criteria that the hospital management could use. Assessing hospital buildings from a financial perspective could be based on new and perhaps other criteria. Different (qualitative) factors play an important role in hospital investment decisions, but the relative importance of different criteria is unknown (Smith et al., 2006; Reiter et al., 2000). This problem will be addressed in the second part of the research.

(14)

9 The results of this research will support Dutch General non-profit hospital management in their building investment decisions.

Investment is defined in this research as an asset acquired for the purpose of producing income and/or capital gains for its owner.

The research context of this thesis is Deloitte Consulting part of the Deloitte Holding. Deloitte Consulting advices Dutch General Hospitals. Deloitte currently advices hospitals on the financial costs and benefits of a building investment, but this tool is limited in that it only covers the financial perspective. Kamath and Oberst (1992) state that “the primary goal of a non-profit hospital is traditionally thought to be that of meeting the needs of the community which it serves by providing quality health care at affordable prices. The overall objective stated in this fashion not only discourages a hospital from evaluating its capital investments from a strictly financial standpoint but also introduces other qualitative aspects of hospital operations”. Kamath and Oberst found in their research that 77 percent of the hospitals in United States let qualitative factors determine their decisions. Given this large percentage of hospitals that evaluate their decision from a non-profit perspective, a decision model which includes financial and non financial aspects is wanted. This is the assignment of this research.

The assignment is to design a decision model for Dutch General non-profit hospitals for hospital building investment decisions.

A decision model is defined as a formal conceptualization of the relationships of the various factors that are relevant in decision-making and how they can be applied in the actual decision making process (Siegel et al., 2005).

(15)

10 Criteria and constraints

The model and research are constrained. This model is designed at Deloitte Consulting, the client of this research. At the start of the research different employees at Deloitte Consulting were interviewed to define core requirements for the model. These were identified as follows:

 the model should provide a theoretical foundation, based on existing literature, if appropriate literature can be found in the literature research

 the model should cover both financial as well as qualitative criteria

2.2 Research approach

Hospital managers have to make building investment decisions themselves. The question is which criteria they should use. Existing literature will be studied to identify criteria for these building investment decisions.

But as stated in the problem statement new criteria can emerge. For instance, financers could ask for adaptability of the hospital building. The question arises whether in literature all current relevant criteria for hospital building investment decisions are incorporated. Therefore, it makes sense to let experts validate the criteria for the hospital building investment decision, but also to stimulate them to add or remove relevant criteria. Different staff levels are involved in the hospital building investment decision. Experts of managerial level as well as of subordinate level who have a direct hierarchical relation, could be involved in the same study. Since a judgment without hierarchical influences is preferred, a method that allows for anonymous participation is used. A method that can validate criteria, that allows experts to alter the outcome and can guarantee anonymity is the Delphi method (Skulmoski et al., 2007; Pivo, 2008). This Delphi method allows us to capture the knowledge of experts who worked on several building investment projects.

(16)

11 judgment. A second way is to study that entity and relate this to other similar entities by making comparisons. According to Saaty, to make a judgment a person must compare an actual

impression with impression in memory of similar stimuli. The method used for ranking should therefore use pair wise comparison. A method that uses pair wise comparison in a limited amount of time is Analytic Hierarchy Process (AHP). This method will be used in this research.

Two different experts groups will be interviewed: consultants and hospital managers. Consultants mainly advice about financial aspects of hospital building investment decisions, it could be that consultants put more emphasis on financial aspects of hospital building investment decisions. Hospital managers could put more emphasis on strategic considerations. In order to avoid overrepresentation of the financial aspect or strategic aspect in the results ranking of experts groups are compared. The ranking of the two different experts groups are compared. If differences are found in the ranking research, needs to be done to factors explaining these differences and the influence of these factors on the ranking.

This research consists of three parts a literature study, a Delphi Study and Analytic Hierarchy Process. The different parts will now be described in more detail.

2.2.1 Literature study

(17)

12 After the literature study the Hospital Building Investment criteria are defined. Three gaps remain from the literature study. These three gaps are the validation of the criteria found to asses hospital building investment appraisal, cash flow prediction and the discounted rate determination (Smith et al., 2006; Reiter et al., 2000).

2.2.2 Delphi consensus study

“The Delphi method is an iterative process used to collect and distill the judgment of experts using a series of questionnaires interspersed with feedback.” (Skulmoski et al., 2007)

A typical Delphi process consists of several steps (Skulmosik et al., 2007 ; Verhagen et al., 1998):

1. Selecting participants

2. Develop, release and analyze Delphi Round One Questionnaire

3. Develop, release and analyze Delphi Round Two, where the participants are confronted with the answers from the previous round and are invited to explain why their point is different from the other participants. This process continues until there is agreement among the participants in Round X or no new arguments are put forward in Round X. In the last case the criteria cannot be validated with Delphi study.

An issue of the Delphi method that is much debated is how to define an expert operationally. Some Delphi studies define an expert as anyone with relevant input. This has resulted in the use of house wives in a study on future communication services. Experts can be classified into three groups according to their intensity of knowledge: awareness, reading about the item or actually working in that field (Mullen, 2003). In the Delphi study conducted for this research only experts working in the field are interviewed.

(18)

13 interviewed within Deloitte is a former Hospital Board Director and we specifically asked him to use the insights from his former position to validate the criteria from a hospital perspective.

All experts advised recently two or more Dutch general hospitals during investment selection of their hospital buildings. The experts are selected by screening the Deloitte projects involving building investment for general hospitals. A list was constructed of people advising in two or more projects. Three people were identified and asked to name more experts. One more expert was named. Four is the minimum panel size used in health care settings (Mullen, 2003).

First the participants are confronted with the model as constructed from literature (see Appendix 4 for the questions). The participants are then asked if they can agree with the criteria used and if they can suggest any criteria to be added or deleted. The author analyzes the differences in opinions and prepares questions for the next round. The participants are confronted with the differences in their answers. This process continues until consensus is reached or no new arguments are put forward in the last round.

2.2.3 Analytic Hierarchy Process Study

Analytic Hierarchy Process is used to rank the criteria. The Analytic Hierarchy Process breaks a problem down into its smaller constituent parts and then calls for only simple pair wise

comparison judgments to develop priorities in each hierarchy.

The Analytic Hierarchy Process consists of three steps according to Saaty (1994).

1. Problem decomposition: the problem is decomposed into elements (which are grouped on different levels to form a chain of hierarchy) and each element is further decomposed into sub-elements.

2. Comparative analysis: the relative importance of each element at a particular level will be measured by a procedure of pair wise comparison. The decision makers provide

(19)

14 matrix, called a reciprocal matrix, has positive entries everywhere and satisfies the

reciprocal property 𝑎_𝑗𝑖 = _𝑎1 𝑖𝑗

Numerical Weight Verbal judgement

1 Equal importance

3 Moderate importance

5 Strong importance

7 Very strong importance

9 Absolute importance

2,4,6,8 Intermediate values

Reciprocal s of above If activity i has one of the above nonzero numbers assigned to it when compared to activity j, then j has the reciprocal value when compared with i.

Table 3: Numerical weight and verbal judgment (Saaty,1994)

3. Priorities: the priorities can be obtained by adding values in each row and column divided by the sum of all judgments.

Forman and Gass (2001) report that all the axioms of AHP, except the synthesis axiom,

correspond with real-world situations encountered. The synthesis axiom states that importance of elements higher in the hierarchy does not depend on lower level elements. In real world situations it might occur that there is dependence between lower and upperlevel elements.

(20)

15 C1

C2

Figure 2: The influence of lower level on higher level elements

The first method involves using a „supermatrix‟ calculation. Saaty and Vargas (2006) describe extensive how to apply a supermatrix. For feedback between adjacent levels only, this is equivalent to deriving priorities with respect to each criteria. This will results in different rankings for each criteria. Each ranking gives the relative contribution of the other criteria to the criteria assessed. For example the relative influence of Criteria 1 and 2 on Criteria 3. The

(21)

16 An example is the influence of other criteria on criteria 2 (Figure 3).

Objective

Criteria 2

Criteria 1 Criteria 3

Figure 3: An example of dependence between criteria

The relative influence of Criteria 1 and Criteria 3 with respect to Criteria 2 is captured in Table 4. In the last column the priority is calculated. When this is done for all three criteria, three 2x2 matrixes are obtained.

Criteria 2 Criteria 1 Criteria 3 Priority

Criteria 1 1 5 0.83

Criteria 3 1/5 1 0.17

Table 4: The relative weight of the Criteria 1 and 3 with respect to Criteria 2

The calculated priority for Criteria 2 is inserted in the second column. A value of zero is used for the middle value because it is assumed that Criteria 2 does not influence itself. If the rankings for the other criteria are also obtained, the Interdependence Matrix can be filled. The

(22)

17 The second method which can be used when dealing with dependence is to make judgments for lower levels of the hierarchy before the upper level or to reconsider judgment at the upper level levels after making judgments at the lower level. The brain performs the feedback function by considering what was learned at other levels of the hierarchy when making judgments for this level (Forman and Gass, 2001).

The above methods deal with situations where the synthesis axiom does not apply. Does the synthesis axiom apply in this research? It is unclear in this research if lower level elements influence upper level elements in the model5. Theoretical the synthesis axiom could not apply in this research. The second method is used to take into account interdependency if it would occur. Participants are allowed to reconsider their judgments when using the AHP.

Forman and Gass (2001) point out that the AHP is criticized for different issues, such as transitivity and rank reversal when adding irrelevant alternatives.

The axiom of transitivity is that: If A is preferred to B and B is preferred to C, then A is preferred to C (e.g. if A is preferred two times over B and B is preferred three times over C, A is preferred six times over C). AHP allows for intransitive relationships (e.g. the decision makers prefers A eight times over C, while A is preferred two times over B and B is preferred three times over C). But according to Forman and Gass (2001) the AHP doesn‟t ignore transitivity but provides a measure of consistency, the consistency index, so that the decision maker can proceed

accordingly. The consistency ratio is used in this thesis (see Appendix 2 on how to calculate this consistency ratio).

Another issue is whether rank reversal an addition or deletion of an alternative can result in changes of relative ranking of other alternatives. The issue of rank reversal is not deemed important in this thesis, since the same criteria are compared and no alternative is deleted or added.

5

(23)

18 The Analytic Hierarchy Process is applied as follows. From the literature study perspectives and criteria are derived. These are confirmed and altered by the Delphi study. The experts of Deloitte and the hospital experts are asked to rank the criteria pair wise. The possibility of feedback could exist. This will be prevented by letting participants review the upper level elements after filling in the lower level elements. Participants can in this way adjust an upper level if it is influenced by a lower level element. The pair wise comparison was checked on consistency by using the

consistency ratio. A consistency ratio of ten percent or less was considered acceptable

(Saaty¸1994). If the consistency ratio was more than ten percent the pair wise comparisons were reviewed.

(24)

19

2.3 Conclusion

This research design can be summarized as follows (see Figure 4). First in order to find hospital building investment criteria existing literature is studied to find criteria. New criteria could also be emerging, for instance the requirement to make hospitals adaptable. Delphi is used to give participants the opportunity to validate and alter the criteria from the literature study. After the Delphi study a question that remains is how the different criteria are weighted; this is evaluated by the Analytic Hierarchy Process. Afterwards the results are analyzed and a decision model and its applicability is presented. In the next paragraph the first part of the research, the results of the literature study, will be discussed.

Literature study: Gather criteria

Delphi: Validate and alter

criteria AHP: Rank criteria Research Steps Overview of criteria from literature Criteria after emperical validation Building Investment Decision Model Products

(25)

20

3 LITERATURE STUDY

3.1 Hospital investment models

Capital budgeting in hospitals is a unique complex decision due to the fact that funds are not directly linked to output, to the existence of multiple objectives encompassing both quality and quantity and to setting priorities that both satisfy the medical staff as well as the board (Kamath et al., 1992). Research indicates that investment decisions in hospitals are largely based on financial cost-benefit analysis in combination with qualitative criteria (Graham et al., 2001; Kamath et al., 1992). Reiter et al. (2000) states that criteria investment decisions in hospitals are driven by financial cost-benefit analysis, the mission or strategic plan and the medical staff. The medical staff also participates in the decision making process in Dutch General Hospitals (Putters, 2003). The model should support these three elements and should provide information about the

financial cost and benefits, the contribution of the project to the strategy and the opinion of the medical staff.

3.1.1 Lyons’ Balanced Scorecard

(26)

21 some invalid assumptions about the relationship amongst these areas. When organizations use the Balanced Scorecard they should be aware of the fact that these performance indicators may be and could result in sub-optimal performance.

The Balanced Scorecard is also applied to hospitals. Lyons and others (2003) conducted a research where a Balance Scorecard was used for capital investment in a hospital. The Balanced Scorecard was developed as an ex-post evaluation and not as evaluation framework ex-ante, though Milis and Mercken (2003) and Lyons (2003) have argued for the Balanced Scorecard as integrated evaluation framework ex-ante. It has limited uptake in practice as a capital budgeting tool. Future studies should explore whether the Balanced Scorecard can be used as an ex-ante tool (Alkaraan,2006). Because it is developed as a framework post and not for investment ex-ante it is not applied in this research.

3.1.2 Dewulfs Real Estate Model

Another model for the hospital investment decision that provides information about the financial costs and benefits, the contribution of the project to the strategy and the opinion of the medical staff is Dewulfs Real Estate model. Dewulf and others (1999, 2004) argue that in every

organization, governmental or private, the building investment should be viewed from the

(27)

22 Figure 5: Aspects of real estate decision, based on Dewulf et al., 1999

3.2 Strategic perspective

The strategic perspective is the evaluation of the investment against the contribution to the hospital mission. The strategic perspective describes the requirements from the hospital board for investment decisions of buildings. Dewulf and others (1999) give examples of criteria used by governmental organizations to measure the strategic perspective, for instance the cultural value of a building. Governmental organizations have a mission to fulfill the needs of the whole Dutch society. Hospitals have another mission, they serve the local community and are focused on providing health care (Kamath et al., 1992). This different mission also implies different criteria for the investment in a hospital building. In studies found in literature the importance of the strategic perspective is mentioned, but not described in detail. Smith and others for example states (2006) “CFOs evaluated projects against both the strategic plan and the mission of the organization, with the later being very important”

Gibson (2004) describes criteria used by the board members for capital budgeting. They asked board members which criteria they used to allocated resources to implement their mission. Board members used eight criteria for evaluating projects against the mission. The criteria are

summarized here below.

(28)

23 - Alignment with external directives: a project applies to identified existing government

mandates and legislated obligations as relevant considerations for setting priorities.

- Academic commitments: the degree in which a project contributes to clinical programs in education of future health care professionals and in facilitating integration of these

activities and research and the role of academic health science centre‟s in establishing best practices standards.

- Clinical impact: was defined as the project contribution to the quality of the service provided.

- Community need: a project contribution to health services needs of patients in the local catchments‟ area of a hospital.

- Partnerships: the project supports formal agreements and commitments with other organizations in coordinating delivery of health care to defined populations.

- Interdependency: described the coordination and collaboration between clinical services within the organization. This involves the degree of coordination and collaboration of the medical staff.

- Resource implications: the extent to which a decision impacts other regular operational (clinical) activities provided.

(29)

24 strategy as serving their local community. This overlaps with the criterion community need, which is the project contribution to health services needs of patients in the local catchments‟ area of a hospital. Strategic fit measures something which is also measured by other criteria and is therefore excluded from the criteria (Ottowa Hospital, 2008; Saskatoon Health Region, 2008; Kingston General Hospital, 2008).

The conclusion is that Gibson‟s criteria can be used for the strategic perspective, because these criteria describe the contribution to the mission. The academic commitment criteria and strategic fit are questionable and therefore excluded as criteria for the hospital building investment model. The other remaining criteria are summarized in Table 5

Strategic perspective

External directives A project compliance to existing government mandates and legislated obligations

Clinical impact A project contribution to the quality of the service provided

Community need A project contribution to the health services needs of patients in the organization local catchment area

Partnerships Project fulfillment of the formal agreements and commitments with other organizations in coordinating delivery of health care to defined populations

Interdependency The degree of collaboration and coordination of the medical staff with a decision

Resource implications The extent of impact of a decision on the other regular (clinical) activities provided..

Table 5: Strategic perspective (based on: Gibson, 2004)

3.3 Financial perspective

(30)

25 Currently Dutch hospitals most frequently use the Payback Period, a Naïve method, (Asselman, 2008)

Kamath and Oberst (1992) discuss several financial Naïve and Net Present Value methods for capital budgeting in hospitals: Payback period, Discounted Payback period, Acounting Rate of Return, Net Present Value, Internal rate of return and Profitabilty index.

The Payback Period (henceforth PP) calculates the period required to recapture the investment. A shorter payback period is prefered over a longer payback period, because the money invested in the project is recaputtered sooner. The two most seriously disadvantages of the PP method are that it does not take into account the returns after the payback period and it ignores the timing of returns. It is still widely used because it is simple to understand, calculate and use. Another argument for its widely use is that decision makers use it in order to recieve confirmation that they have made a good choice (Lefley,1996).

(2)

where Investment = the cost of the project

Cash Flow = the forecast incremental net spendable income from an investment, determined by deducting all operating and fixed expenses from the gross income. If expenses exceed income, a negative cash flow is the result.

The Accounting Rate of Return (ARR) is calculated by dividing the average profit by the investment. A larger ARR is prefered over a smaller ARR, because the profitbalitity of the investment with a larger ARR is higher. The limitation of this method is that it does not take the time value of money into account (Drury, 2004).

(3)

(31)

26 The Discounted Payback Period (DPP) calculates the period to recapture the investment with the help of discounted paybacks. A smaller DPP is prefered over a larger DPP. DPP overcomes the limitation of timing of returns, but it ignores the returns after the payback period (Lefley,1996)

DPP =_{𝐷𝑖𝑠𝑐𝑜𝑢𝑛𝑡𝑒𝑑 𝑐𝑎𝑠ℎ𝑓𝑙𝑜𝑤}𝐼𝑛𝑣𝑒𝑠𝑡𝑚𝑒𝑛𝑡 (4)

where Discounted cashflow = the forecast incremental net spendable income from an investment, determined by deducting all operating and fixed expenses from the gross income on time t

divided by (1+r)t

r

= cost of capital or discount rate

The Net Present Value methods are Net Present Value, Internal Rate of Return and Profitability Index. Net Present Value (NPV) is calcultated by discounting the future cashflows minus the investment divided by the investment. A higher NPV implies that more money is left over after project and a higher NPV is preffered over a smaller NPV. NPV uses time value of money and cannot easily be manipulated (Drury, 2004).





t t T t r C NPV  



0 1 (5)

Where C = the forecast incremental net spendable income from an investment, determined by _t deducting all operating and fixed expenses from the gross income on time t

T = project life

r =the opportunity cost of capital or the discount rate.

Net Present Value is considered the most advanced method because it includes all cashflow and also considers the time value of money (Drury, 2004).

(32)

27 positive, negative and again positive cash flows. If the IRR percentage is above the opportunity cost of capital the project can be selected. When this technique is used as a selection tool for mutual exclusive investment projects, risks are not accounted for. The project with the highest IRR might get selected in spite of being much more riskier than a project with a slightly less IRR (Millis et al. , 2004). Considering the large amount of money involved with the hospital building investment, this decision is mutually exclusive and therefore this method cannot not be used.





t t T t r C  



 1 0 0 (6) Where the IRR is the r that makes the NPV zero

Profitability index(PI) is the NPV dived by investment. A higher NPV means that a project is more profitable. One limitation is that it does not take project revenue into account. It prefers projects with a smaller project size and a larger PI over projects with a large projects size with a smaller PI, in spite of the fact that more money is received by the larger project (Drury, 2004). PI= 𝑁𝑃𝑉+𝑖𝑛𝑣𝑒𝑠𝑡𝑚𝑒𝑛𝑡

𝐼𝑛𝑣𝑒𝑠𝑡𝑚𝑒𝑛𝑡 (7)

All above methods have as limitation that they are static. Two methods overcome this limation: Real Option Reasoning and Decision Tree Analysis. These two methods are both based on Net Present Value.

Real Option Reasoning is derived from finance literature and views investement projects

analogous to financial options. The option value stems from the managerial flexibility of a investment decision when a project is undertaken (Verbeeten, 2006). An example is when the management decides to postpone a project. Imagine a project with a cost € 100 and a single cashflow of € 120. When the management decides to postpone the project one year the cost is still € 100 but the cashflow will be €130 due to a larger market.

NPV1 = -€100 + €120/1.1= € 9.09 NPV2= -€100/1.1 + €130/1.12 = €16.53

(33)

28 In the current situation hospitals hire construction companies for the building activities. This gives hospitals a limited ability to change the project once started because of contractual

boundries (Twynstra Gudde, 2006). Due to these contractual boundaries managerial flexibility is limited and option value can not be used.

Decision Tree Anaylisis (DTA) is a dynamic version of discounted cash flow and can incorporate future external uncertainties. Future events can be modeled by using a probability tree (Rivet, 1980; Van Horne, 2004). The different branches are the possible outcomes of an investment. An example is given below. In Year 0 an investment of 10 euro is made. This has a probability of 50% of generating a cash flow of € 7 in year 1. If €7 cash flow is generated in the first year the chance is 60% of generating a cash flow € 7 and 40% of generating a cash flow € 5 in Year 2.

- € 10 € 7 (0.5) € 5 (0.5) _{€ 5} € 5 € 7 € 3 (0.4) (0.6) (0.5) (0.5)

Year 0 Year 1 Year 2

Figure 6 : Example of Decision Tree Analysis

(34)

29 branch orignated from that node, and the highest is selected. The expected outcome of that node is carried back to the next change or decision point by multiplying it by the probabilities

associated with branches it travels over.

In order to use the model appropiate measures of probability have to be established. Hospitals have to determine the probability of events occuring. Estimatitions by hospitals have to be

checked in the light of previous experiences. For example management thinks that the probability of extra building cost occuring is 25%. The hospital manager has to eximane prior building decisions and see if in 25% of the prior cases the hospital building investment had to deal with extra building costs.

Net Present Value using DTA is debated around two issues. A first issue is how one should select the set of alternatives of the decision tree. There is a wide range of decisions and possible consequences. Therefore the choices are restricted at each decision point by allowing a limited possible decisions and consequences. It is possible to break the same problem down using different alternative sets to see the impact of different alternative sets for the net present value. Managers use for instance extra building cost as a possible alternative in Model 1. The NPV of Model 1 is than compared with the NPV of Model 2, where two possible alternatives: extra material cost and extra labour costs are used instead of the extra building cost. A second issue is the particular timing of the sequences. Decisions and results take place in a continuum of time. The decision tree breaks the continuum of time into limited discrete timeslots. By breaking down the Decision Tree in different timeslots it is possible to observe the sensitivity of net present value to different timeslots. For instances the hospital can use months instead of years to calculate the influence of the different factors.

Hospitals could apply Net Present Value with Decision Tree Analysis as follows:

1. First the hospital should define and describe significant interdependencies among factors. 2. Hospitals should make probability estimates of events affecting the hospital building

(35)

30 3. The decision tree should be checked for sensitivity for different decision points and

different time intervals used.

The model can now be used to calculated the Net Present Value.

Hospital building projects are dealing with external uncertainty of cash flows and costs. It could be that for instance the building costs will be higher. Because future cashflows and costs can differ NPV in combination with DTA is used. Because of the limited managerial flexibility Real Options is not suitable. The project can be ranked by giving the building with the highest NPV with DTA the largest score. There is a difficulty with applying the NPV with DTA. According to Reiter and others (2000), it is unclear how the discounted rate should be determined and how cash flows should predict for analyzing the financial cost-benefits of investment.

Discount rate and cash flow determination

The NPV with DTA discussed above, uses the cash flow in combination with the discount rate. The discount rate consists according to the Capital Asset Pricing Model (CAPM) of costs of capital adjusted by the risk of the investment (Graham et al., 2001). There is currently no consensus in the United States about how hospitals should determine the discount rate. The two most common methods to determine the discount rate use the weight average costs of capital (WACC) or the inflation rate as a discount rate (Reiter et al., 2000). The WACC is calculated as follows.

(8) Where E = total market value of equity and equity equivalents

D = total debt and leases Re = Cost of equity Rd = Cost of debt Tc = Corporate tax rate

(36)

31 at the hospital level (see for instance Adam et al. 2003). Managers estimates are still the most widely used methods for cash flow determination. A method for determining cash flows subjectively is the Delphi method. This method makes it possible to identify relevant

assumptions and information from experts in various departments. This method is already used in practice successfully (Ang et al., 1979) . Hospitals could presently use Delphi for estimating cash flows until mathematical models are developed.

3.4 Functional perspective

Dewulf (1999) describes six criteria for the functional perspective: Quantity is the ratio of square meters available for the different functions, such as outpatient, inpatient and office areas. The criterion quantity incorporates how many required square meters are fulfilled by a project. Functional aspects are defined as the extent in which buildings facilitate the primary process efficiently. Visual aspects are aspects covering the building appearance. Technical aspects are the average lifetime of the materials used in the building. Sustainability aspect focus on the extent of the environmental impact done by the materials that are used and location as the extent in which users of the building can access the hospitals. According to Dewulf the measurement of these criteria can be done by letting users rate these criteria.

As described in the introduction new hospitals could follow a common trend. They could be designed modular and be divided in layers. The layers can be used by other organizations. This is called adaptability. The adaptability is the extent to which a location is easily adaptable to

provide another function or can be used by another organization (Bouwcollege, 2007).

Functional perspective

Quantity Square meters fulfilled by the project for the different functions

Functional aspects The extent in which buildings facilitating the primary process of a hospital efficiently

(37)

32

Sustainability aspects The extent in which the environment is impacted by the materials are used,

Location The extent in which the users of the building can access the hospitals

Visual aspects The building appearance

Adaptability The extent to which a location with a certain function can easily be adapted to provide another function or be used by an external organization

Table 6: Functional perspective (based on Dewulf 1999)

3.5 Conclusion

Dewulfs model can be used for assessing hospital building investment decisions. However, some adaptations seem warranted. These can be illustrated by discussing three different perspectives.

Strategic perspective

The strategic perspective cannot be measured by Dewulfs criteria because they are measures of governmental missions. The strategic perspective is better measured by using Gibson criteria that describe a project contribution to the hospital mission: Alignment with external directives, Clinical impact, Community need, Interdependency, Resource implications and Partnerships.

Financial perspective

The financial perspective of Dewulf measures only the costs but not the benefits of a building. A hospital building also generates revenue, therefore another method is needed. From the financial methods NPV with DTA is theoretically the most sophisticated, because it includes the time value of money. Therefore, the NPV is added to account for financial perspective. However, it remains unclear how cash flows for hospitals should be predicted and which discount rate should be used for property investment projects.

Functional perspective

(38)

33 Strategic perspective Alignment with external directives Clinical impact Community need Interdependency Resource implications Partnerships Hospital Building Investment Functional perspective Quantity Visual aspects Functional aspects Technical aspects Sustainability aspects Location Adaptability Financial perspective

Net Present Value with DTA

Figure 7: Hospital Building Investment Criteria

A limitation of Dewulfs real estate model is that the interests of patients are not covered. This should be included in the model. In the strategic perspective both the community need and the clinical impact are considered in the investment decision. The community need considers the health need of the patient in the local catchments‟ area and the clinical impact considers the quality of service. The patient interest can thus be covered in the model.

The original model is adapted, but three questions remain. First, the adoptions of the model have to be validated. Secondly, it is questionable if in literature all the criteria for the hospital building investment have been described. Thirdly for the financial perspective it is unclear how the discount rate should be determined and how cash flow should be predicted (Reiter et al., 2000). These gaps will be researched using Delphi consensus study (see Appendix 4).

(39)

34

4 DELPHI AND AHP

The results are divided in two parts: in the first paragraph the results of the Delphi study and in the second paragraph the results of the AHP study are discussed.

4.1 Delphi consensus study

The aim of this Delphi consensus study is to validate the criteria to asses hospital building investment appraisal, determine how cash flows should be predicted and how the discount rate should be determined. The detailed results can be found in Appendix 5.

4.1.1 Overall perspectives

In Delphi Round 1 all experts agreed with the different perspectives and did not want to add or delete perspectives. No second Delphi Round was conducted for the overall perspective, because no disagreement existed among participants.

4.1.2 Strategic perspective

All experts agreed with the different criteria and did not want to add or delete criteria for the strategic perspective.

Participant A wanted to rephrase the partnership criterion. Participant A stated that hospital building investment was a long-term investment. For a long-term investment, only long-term partnerships needed to be considered. Participant A rephrased the criterion in “Project fulfillment of the long-term formal agreements and commitments with other organizations in coordinating delivery of health care to defined populations”.

(40)

35 4.1.3 Financial perspective

The participants were asked if they agreed with using the Net Present Value with Decision Tree Analysis for analyzing the financial perspective and how it should be used. All participants agreed in the first Delphi round how to analyze the financial perspective for the hospital building investment decision. The financial perspective was captured in a simple model using Net Present Value without Decision Tree Analysis.

The participants suggested using a discount rate for analyzing building costs and revenues. They suggested using the interest rate for debt. They suggested that the future need of equity for a hospital should be determined and the discount rate for equity should be set to be able to fulfill this need. The future equity is determined on the solvability ratio requested by the financers and discount rate for debt is based on the interest rate by the financers.

The predication of cash flow was considered difficult by the participants, because it is uncertain which DBC‟s will be registered in the B-segment instead of the A-segment and when. The participant handled this uncertainty in future cash flow by estimating incremental price and volume increases above the current cash flow. By estimating the cash flow uncertainty will still exist. The participants all suggested the use of sensitivity analysis to take the different factors that influence the cash flow into account. Three variables were identified using different building cost (e.g. other material or changes in building cost), different full cost markup tariffs (e.g. x% above the current DBC tariffs) and different patient totals (e.g. x patients extra/year).

(41)

36 it does not estimate the financial impact of the investment decision. Therefore, the Net Present Value is used with DTA.

Since consensus was evident in the first Delphi round no second round was necessary for this item.

4.1.4 Functional perspective

Three participants agreed with the criteria used for the functional perspective. Participants were asked if they had other criteria to add to the model. One criterion which is mentioned by

Participant B is medical technology. Medical technology can influence the number of times a patient visits a hospital.

In Delphi Round 2 the participants are confronted with the difference in opinion about the medical technology. Participant B who raised the issue was confronted that the other participants did not mention that criterion. The participant stated that medical technology should be a

separated criterion, because it could influence hospital design. The other three participants were confronted with an extra criterion of medical technology for the functional perspective.

According to the other three participants Medical technology can change the demand for capacity for instance the numbers of hospital bed, but it is considered an influence on the capacity of the hospital, not a separate criterion.

In the next round, the participants were once again confronted with their differences. The

participant who added medical technology agreed with the overlap and changed his opinion. The other participants stayed with their argument. All participants now agree with not adding the medical technology criterion.

4.1.5 Delphi Results

(42)

37 and use sensitivity analysis to give more insight in the factors influencing the cash flow.

Sensitivity can be useful to get insight in the different factors that influence the result of the investment, but cannot be used to determine the result of the investment. Net Present Value with DTA is used in the model because it can determine the result of the investment and gives insight in the risks. The discount rate is determined by using the interest rate for the debt and the

estimated future equity using the solvability required by financers.

4.2 Analytic Hierarchy Process

Since the criteria are now validated, AHP is used to rank the different criteria. The AHP is conducted within Deloitte and within a General Hospital(Hospital 1). Data of the AHP ranking is shown in Appendix 7.

Managers of another General Hospital (Hospital 2) were interviewed, but were only willing to discuss rankings for a specific case. That hospital was currently developing their Long Term Housing Plan. This could have discouraged the discussion of the ranking of the criteria in general. The results of the AHP ranking from Hospital 2 are excluded from this report because it is not a ranking in general but for a specific case.

4.2.1 AHP Results

(43)

38 Overall AH P De loi tt e AH P GH Strategic perspective 66% 25% Financial perspective 22% 25% Functional perspective 11% 50% Total 100% 100% CR 0,00000 0,0000 Strategic perspective

Alignment with external directives 20% 3% Clinical impact 38% 22% Community need 14% 45% Interdependency 13% 4% Resource implications 5% 13% Partnerships 13% 12% Total 100% 100% CR 0,09284 0,06935 Functional perspective Quantity 16 % 19% Location 39% 20% Functional aspect 21% 36% Visual aspects 7% 4% Adaptability 8% 14% Technical aspect 7% 4% Sustainability aspect 2% 4% Total 100% 100% CR 0,09561 0,072369

Table 7: Overall ranking of AHP

4.3 Conclusion

(44)

(45)

40

5 ANALYSIS OF DELPHI AND AHP

5.1 Criteria

The Deloitte experts and General hospital 1 confirmed the criteria. The Deloitte experts rephrased the partnership criterion. The General Hospital 2 experts mentioned the requirement of renting a hospital building. It could be that other hospital uses the financing of the building as a criterion. It should be further evaluated if this is a requirement for this specific case or a general criterion.

5.2 Cash flow prediction and discount rate determination

Participants agreed that hospitals should only take the interest of debt and future equity into account when determining the discount rate. This confirms the results of Reiter and others (2000). Reiter suggests that the discount rate should be equal to the cost of capital and the rate of growth in equity necessary to fulfill the institutional goals. By using the interest rate and

solvability as input variable a discount rate can be determined.

The cash flow is estimated using subjective management information, because no mathematical models are available at this moment. This is the only option currently available to hospitals. This method is also frequently used in practice in the research of Reiter and others (2000) and Graham and others (2001).

5.3 Ranking

The ranking of the General Hospital Experts 1 and the Deloitte Experts of the perspectives and different criteria are now discussed. The financial perspective only has one sub criterion and cannot be compared. The financial perspective is not discussed separately.

5.3.1 Overall perspectives

There is a difference in the importance given to the different perspectives. Hospital experts consider the functional perspective the most important perspective and the financial and strategic perspectives equally important. The Deloitte experts considered the strategic perspective the most important perspective, than the financial perspective and finally the functional perspective.

(46)

41 These ratings confirm that the financial perspective is not the most important aspect for the

hospital investment decision. This is also shown in the research of Kamath and others (1992). From theory the strategic perspective is ranked most important and the financial perspective is considered the second most important criterion (Smith et al., 2006). The ranking would be Strategic, Financial and Functional perspective.

5.3.2 Strategic perspective

The criteria „alignment with external directives‟ and „interdependency‟ are considered less important by the hospital experts than by the Deloitte Experts. This could be because hospital experts are only interested in satisfying these criteria and do not try to maximize these criteria. The clinical impact is considered less important by the General Hospital experts, but nevertheless is ranked as second most important criterion. Community need is ranked much more important by the hospital experts. These experts see this as the primary criterion representing their mission. Resource implication is ranked slightly higher by hospital experts. This could be because of the high impact of hospital construction on the organization. Partnerships are ranked equally by both groups.

5.3.3 Functional perspective

Hospital experts rank the functional aspect higher than the location. Deloitte ranked the criteria the other way around. Hospital experts have to work in the building and therefore rank functional aspects high. The adaptability aspect was considered more important by hospital experts. An explanation for this could be pressure from the financers of the hospital. The quantity criteria is rated almost the same by both groups. The visual, technical and sustainability aspect are deemed less important by both expert groups.

5.4 Applying the model

(47)

42 be used to check the completeness of the business case of a hospital building investment decision. The model can be used as a checklist which aspects should be taken into consideration.

The participants agreed with the general criteria, but they use only a few criteria for the selection of an investment when it is applied to a specific case. This could imply that the model consists of two types of criteria. Criteria that have to be satisfied and criteria used to score the different investments. The model could be applied in two phases. First, the projects are filtered for the criteria that can be satisfied or not. For instance, projects where the location is considered too far away from the patient population are eliminated. In a second phase AHP can be applied to

determine the best project. The General Hospital 2 who applied the criteria to a specific case was enthusiastic about the pair wise comparison offered by AHP. AHP is used before for investment appraisal (Angelis et al., 1996; Forman and Gass, 2001).

Figure 8: Phases of the Hospital Building Investment Model

If this two stage approach is used the criteria could be divided in two groups. The first group consists of the criteria only used for screening the project and the second group consists of the criteria used for both screening and calculating the maximum score. The first group could consist of the low ranked criteria: the projects only have to satisfy these criteria. These requirements are the visual, technical and sustainability aspects of the functional perspective and the alignment with external directives and the interdependency of the strategic perspective. All other criteria fit into the higher ranking group. This model should be applied to a case to see if these criteria can be split in two groups.

Possible Hospital Building Investment Projects

Possible Hospital Building Investment Project who do

not satisfy the criteria

Hospital Building Investment Project with maximum value

(48)

43

5.5 Conclusion

In the Delphi rounds the criteria collected from the literature study are confirmed by the Deloitte experts as well as the Hospital experts. During the Delphi Study some criteria were altered and/or better formulated by the Deloitte experts.

During the Analytic Hierarchy Process ranking no general consistent ranking can be determined. From literature the ranking Strategic, Financial and Functional perspective seems to make the most sense. It could be concluded that there is substantial variation in the ranking of criteria in investment decision-making by the Deloitte experts and the hospital experts. One reason could be that the client population of Deloitte could not be representative for the hospital population. The Deloitte hospital clients could for instance deal with other issues than the hospital population as a whole. It could also be possible that Deloitte rates the criteria structural differently than the hospital experts. In this case it could be possible that Deloitte suggest no optimal solutions to the hospitals, because Deloitte does not focuses on certain issues are draws other conclusions. Further research needs to be done to find out if Deloitte values the criteria structural differently than the hospital population as a whole and what the reasons are of this difference.

(49)

44

6 DISCUSSION

The Delphi method was conducted with experts who advised in two or more hospital building investment projects. Consensus about the criteria was received easily. This does raise the

question of the critical approach taken by the experts to the different criteria. The experts advised mainly about the financial consequences of hospital building investment. It could be that since Deloitte experts mainly advice about financial consequences, their knowledge of qualitative criteria is limited. This could limit the validation of the different criteria. The hospital experts also valued the different criteria, but also these respondents only advised in the hospital building investment decision. A limitation of this research is that only advisors for the hospital investment decision are involved in the empirical part of this research.

Another limitation of this research is that the experts were asked to validate more than one different perspective or criterion at the same time. Verhagen and others (1998) asked for each individual criterion if it should be included or not, using a 5-point Likert scale. This could have stimulate more discussion.

In this thesis a non-profit perspective is taken. Hospitals could make profit from 2012 onwards. This model is designed for nonprofit hospitals. This could have two implications. In a for profit hospital more importance is paid to the financial criteria. Also, the discount rate could be higher because reimbursement for using equity could be requested by shareholders (Kamath et al., 1992). Because hospitals buildings last between 20-40 years it can be argued that a more for profit perspective should be taken.

The patient interest is not directly incorporated in the model. In paragraph 3.5 it is described that community need and clinical impact cover the most important interests of the patient. But during this research patients were not interviewed to validate the criteria from a patient perspective.

(50)

Dutch General Hospital Building Investment Decision Model