FROM CURE TO
PREVENTION: WHY
DON’T WE GO?
MASTER
THESIS:
P
OLITICAL
E
CONOMY
MEREL NOTENBOOM
UNIVERSITY OF AMSTERDAM1 JUNI 2017
Abstract
The aim of this research was to identify the influences on the preventive healthcare expenditures as a percentage of total healthcare expenditures in the Dutch healthcare system. In order to understand the complexity of preventive healthcare in the Dutch healthcare system, it was explained how the different elements interfere and influence the system surrounding preventive healthcare. The public policies towards preventive healthcare are influenced in particular by powerful stakeholders such as, medical professionals, health insurance companies, governmental health organizations, and political institutions like laws and regulations. These public policies set the opportunities and constraints for the preventive healthcare expenditures. A positivist mixed method research was performed on said important questions that could make it possible to stimulate preventive healthcare. This research was performed with a quantitative research to test the relationship between national income and preventive healthcare as a percentage of total healthcare expenditures to know whether national income has an effect on preventive healthcare expenditures. Then, experts were interviewed who are involved in the process of stimulating preventive healthcare for the qualitative research part. To answer the research question “Why are the expenses of preventive healthcare in the Netherlands limited to 4% of the total healthcare budget?” this research has investigated the influence of national income on preventive healthcare expenditures as a percentage of total healthcare expenditures, investigated the supply and demand related mechanism that might affect preventive healthcare expenditures, investigated the political institutions that might affect public policies that limit preventive healthcare expenditures and investigated the political power of professionals that might affect public policies that limit preventive healthcare expenditures. From my research, it can be concluded that national income does not have a significant influence on preventive healthcare expenditures as a percentage of total healthcare expenditures. Furthermore, it can be concluded that price is partly influencing preventive healthcare expenditures. Price affects preventive healthcare expenditures by being an obstacle for new initiatives, but not for being an influence on the income of medical professionals. Moreover, it can be concluded that political institutions affect the public policies that limit preventive healthcare expenditures by limiting the possibilities for organizations to engage in stimulating of financing preventive healthcare projects and initiatives. Lastly, it can be concluded that political power is more working as a political influence system of the professionals. This influence on public policies regarding preventive healthcare is exercised by collaboration with the Ministry of VWS, national campaigns, and publishing reports. However, the healthcare system in the Netherlands positioned the insurers in a center position (see figure 10). The expert interviews showed that health insurers are not in favor of increasing preventive healthcare for their organization, since preventive healthcare does not fit completely in the “Law health insurers”.
However, to increase preventive healthcare expenditures as a percentage of total healthcare expenditures, a change in the offering of preventive healthcare by the center actor in the healthcare system is needed. Therefore, the “Law health insurance” should change to enlarge the possibilities for preventive healthcare for the center actor. Since, health insurers have a center position in the healthcare system and their strong influence to determine therapy options through their healthcare financing policies. In sum, it might be said that the way to go to a higher percentage of preventive healthcare expenditures as a percentage of total healthcare expenditures is, change the (political) institutions, in particular the “Law health insurance” and involve the medical professionals of the healthcare sector more in the development of public policies regarding preventive healthcare. The (political) institutions are holding back the stimulation and financing of preventive healthcare initiatives, so a change is needed to create possibilities for actors in the (preventive) healthcare system. This research found that medical professionals are aiming for more preventive healthcare and their knowledge and expertise can help to work towards the goal of creating a healthcare system more focused on preventive healthcare.
Tabel of Contents
Introduction ... 6
Theoretical framework ... 10
Preventive healthcare ... 10
Supply and demand related explanations ... 11
(Political)Institutions explanations ... 15 Political power ... 18 Research design ... 20 Data ... 21 Methodology ... 22 Quantitative methodology ... 23 Qualitative methodology ... 25 Case selection ... 25
Participants and setting ... 28
Data generation ... 29
Data analysis ... 30
Results ... 31
Case study... 36
Dutch healthcare system ... 37
Preventive healthcare in the Netherlands ... 39
Objectives of preventive healthcare in the Netherlands ... 41
Supply and demand related explanations ... 42
(Political) institutions ... 44
Dutch healthcare financing system ... 45
Political power ... 49
How to make a step towards more preventive healthcare? ... 51
Conclusion ... 52 Discussion ... 53 Methodology ... 54 Hypotheses ... 54 Limitations ... 56 Further research ... 57 References ... 58
Reports ... 63 News articles ... 64 Appendix 1 ... 65 Appendix 2 ... 67 Appendix 3 ... 96 Appendix 4 ... 120 Interviews ... 121 Interview GGZ ... 121 Interview Achmea ... 123 Interview EY ... 127 Interview LHV ... 130 Interview GGD ... 134 Conversation NHG ... 137
Introduction
“Focus on prevention: wellness not sickness.” (Hilary Clinton, 2008)
For writing this master thesis, I followed an internship at EY at the Financial Accounting & Advisory Services – Healthcare department. EY often faces questions from their clients (e.g. health care institutions), social entrepreneurs and governmental organizations regarding preventive healthcare. Relevant questions from society are: how does preventive healthcare influence the total healthcare expenditures, how can healthcare organizations move in the system full of laws and regulations influencing preventive healthcare, what is the influence of other actors in the health system regarding preventive healthcare, and how can new initiatives be financed. These questions have led to a need for EY for a research that matches my interests in political health enquiries.
To start a research concerning preventive healthcare, it is important to outline the definition used in this research for preventive healthcare. This thesis adopts the following definition: “Preventive healthcare aims at reducing the likelihood and severity of potential illnesses by protection and early detection” (Zhang et. Al, 2009)
Important enquiries concerning preventive healthcare are; how preventive healthcare can solve issues regarding the rising age of the population and the more expensive medical need that increase the total healthcare spending (Kim et. al, 2014). Furthermore, preventive healthcare could also be an answer to the rising questions about how to deal with overweight and obesity (Gutierrez et. Al., 2010)
Since EY needs a research focusing on the Dutch (preventive) healthcare system, this research has a strong focus towards the Netherlands. The Dutch government wrote several policy papers that focus on preventive healthcare in order to formulate the vision of the Dutch government on preventive healthcare policies (VWS, 2013; VWS, 2016). A Dutch governmental organization, Rijksinstituut voor Volksgezondheid en Milieu (RIVM) defines preventive healthcare as the healthcare aimed at keeping people healthy, promote people’s health and to protect people’s health (RIVM, 2017). Furthermore, the Dutch preventive healthcare is aimed at preventing diseases to occur and to prevent disease complications (RIVM, 2017). Furthermore, the Dutch government has included policies in the preventive healthcare programs to discourage the use of alcohol and cigarettes and policies to avoid depressions and diabetes (VWS, 2013; VWS 2016). These preventive healthcare interventions are
focused on avoiding and preventing worse health issues. These interventions can help to increase the health condition of the Dutch population and therefore will lead to lower healthcare expenditures. Preventive healthcare does not only lower healthcare spending but also increases that number of people that work, lowers the number of sick days and lowers the need for caregivers.
Preventive healthcare can help to control the Dutch healthcare spending because research has shown that preventive healthcare is more cost effective than curative healthcare as we take the total costs for society into account (VWS, 2013). This is a focus point for the Dutch society in order to have fiscal policy discipline and is needed for economic growth according to many scholars that are in favor of the Washington Consensus (Williamson, 2000). Furthermore, cost effectiveness in healthcare will point focus towards the quality of the Dutch healthcare system.
In order to understand the complexity of preventive healthcare in the Dutch healthcare system, I will explain how the different elements interfere and influence the system surrounding preventive healthcare. The public policies towards preventive healthcare are influenced in particular by powerful stakeholders such as, medical professionals, health insurance companies, governmental health organizations, and political institutions like laws and regulations. These public policies set the opportunities and constraints for the preventive healthcare expenditures.
In recent years, many scholars have focused on theories that explain the complexity and working of healthcare systems. The first theories focus on supply and demand related explanations to explain how opportunities and constrains affect preventive healthcare expenditures initiated by public policies. Several scholars focused on the relation between income and health expenditures (Newhouse, 1977; Leu, 1986; Gerdtham et. al., 1992). However, an extensive literature review has concluded that this method to test the relationship between national income and healthcare expenditures has not been used to test a comparable relationship between income and preventive healthcare. Other scholars have shown a general relationship between price and quantity of preventive healthcare (Hey and Patel, 1983; Meredith, 2013). This relationship is not yet tested in a
single case study of the Netherlands.
A different approach for explaining the low level of preventive healthcare expenditures is given by the political institutions theory (Immergut 1990; Immergut, 1992). This theory gives reason to belief that preventive healthcare expenditures are influenced by the political institutions in a country. This theory has also not been tested in a single case study of the Netherlands. Another theory, the political power explanation could help to understand the low level of preventive healthcare expenditures (Immergut 1990; Immergut, 1992). This could implicate that social or professional
groups have a considerable influence on political decision-making processes concerning preventive healthcare and financing preventive healthcare initiatives.
These mentioned scholars show how many theories might explain the influence of powerful groups and political institutions on the opportunities and constrains for the public policies regarding healthcare, and the influence of these public policies on the behavior of actors in the healthcare systems by supply and demand related explanations. However, these scholars have not yet explained how these theories can be applied for preventive healthcare in the Netherlands. When these theories would hold for preventive healthcare in the Netherlands, it would be possible to know what influences the public policies concerning preventive healthcare and what opportunities and constraints are present in the current healthcare system for preventive healthcare. In order to increase the focus in the healthcare system from cure to prevention, the expenditures on preventive healthcare as a percentage of total healthcare expenditures have to increase. The ratio between preventive healthcare and curative healthcare needs to increase.
In the current situation, many new initiators for preventive healthcare face the difficult consideration how to finance their project (EY, 2016). Since initiators have the idea that the Dutch healthcare system does not provide possibilities to finance their idea, they face difficulties with launching their initiatives. To stimulate preventive healthcare, it should be clear what the best financing system would be for an innovative health provider. When there is an initiative for a project that could help to prevent the need for expensive healthcare, it should be good to have a clear view on the best way to go. New initiators explore alternative financing systems, however, these alternatives financing systems are not “one size fits all”. To increase the preventive healthcare expenditures new initiators, need to know what the possibilities are in the current healthcare systems and whether they can better turn to alternative healthcare financing systems. The results of this research will help to increase the succeeded initiatives.
To explore the possibilities in the current healthcare system for moving from cure to prevent, it is important to know how the current healthcare institutions and healthcare policies act on preventive healthcare. Then we should get to know whether and what institutional and policy changes can contribute to the objective of increasing the level of preventive healthcare. When these institutional and policy changes are not possible or are most likely not able to help achieving a higher preventive healthcare expenditure level, we need to look at alternative systems. We should know what alternative financing systems are available to stimulate preventive healthcare and furthermore we need to know what conditions are set by stakeholders in the healthcare system in order to increase their levels of preventive healthcare.
The questions above form the alignment for the problem statement: currently, the healthcare system in the Netherlands is focused on curative healthcare. The benefits of a healthcare system focused on preventive healthcare are reason to make a change, however the healthcare expenditures as a percentage of total healthcare expenditures do not change, so the ratio between preventive healthcare and curative healthcare does not change. The lack of research findings on why this change is not set in motion might play an important role in this complex path. In the context of the complexity of the Dutch healthcare system regarding preventive healthcare, the main research question arises: “Why are the expenses of preventive healthcare in the Netherlands limited to 4% of the total healthcare budget?” Therefore, the aim of this research is to identify the influences on the preventive healthcare expenditures in the Dutch healthcare system.
With the results of this research it will be possible to understand how the different actors influence public policies regarding preventive healthcare, to give advice to the government how to stimulate preventive healthcare and to advise future (social) entrepreneurs what type of financing system would fit their project. This will hopefully lead to higher spending on preventive healthcare. Most importantly, this stimulation of preventive healthcare could help to sustain their fiscal policy discipline in the future, because of having a better possibility to restrain the Dutch healthcare costs. I will conduct a positivist mixed method research on said important questions that could make it possible to stimulate preventive healthcare. I will start by performing a quantitative research to test the relationship between national income and preventive healthcare as a percentage of total healthcare expenditures to know whether national income has an effect on preventive healthcare expenditures. Then, I will interview stakeholders who are involved in the process of stimulating preventive healthcare for the qualitative research part.
This research continues with the theory section, in which I will clarify and elaborate about the key concepts of this research. Then, in the method section will I will explain the method used in order to gather the necessary information.
Theoretical framework
In this literature review, I will start by exploring the literature several theories that can help to answer the question “Why are the expenses of preventive healthcare in the Netherlands limited to 4% of the total healthcare budget?” The first theories will be combined in supply and demand related explanations. This will give a more economic theoretic view on the level of preventive healthcare expenditures. Then I will go further into the literature that explains healthcare expenditures from a more social sciences view. This will give institutional and policy explanations which are found by various important scholars that are named in this section.
Many scholars have explained with theories how healthcare expenditures are influenced. In this research, the question is whether these theories will also hold for preventive healthcare. Since preventive healthcare is part of healthcare, it might be expected that the “behavior” of preventive healthcare expenditures is similar to the “behavior” of healthcare expenditures. The “behavior” is the way (preventive) healthcare reacts to influences. The influences on preventive healthcare expenditures could therefore result in similar outcomes compared to influences on healthcare expenditures.
Preventive healthcare
Preventive healthcare can be defined as the healthcare aimed at keeping people healthy, promote people’s health and to protect people’s health: “Preventive healthcare aims at reducing the likelihood and severity of potential illnesses by protection and early detection” (Zhang et. Al, 2009). There are three reasons why a health system should focus more on preventive healthcare. The first reason is population aging, the second reason is that medical costs are rising, the third reason is that there is an underuse of preventive healthcare for older adults (Kim & Kawachi, 2017).
The first framework was designed by Caplann in 1964. “The goal of primary prevention, as it was defined by Caplan and further refined by Cowen (1977, 1980), is to use risk reduction strategies to prevent the onset of specific diseases before any symptoms arise. Primary prevention efforts target either the entire population or subgroups with known vulnerabilities (e.g., raising alcohol prices on military bases so that they are equal to the prevailing price in the community). Secondary prevention efforts are designed to identify symptoms of disorders early in order to reduce their duration (e.g., screening and intervention for depression) or their sequelae (e.g., screening and intervention for suicidal risk). Tertiary prevention is designed to prevent long-term disability and rehabilitate individuals with a disorder and to return them to their productive capacity as quickly as possible (e.g.,
a pain management program aimed at preventing the sorts of disability that prevent a return to duty)” (Warner et. Al., 2014).
Later, the Gordon framework was established where he uses three different categories for preventive healthcare (O’Connell, 2009). There are three different types of preventive healthcare that are described in the Gordon framework are; Firstly, universal preventive healthcare that targets risk factors for the total population, this includes targets mentioned in the introduction as, national policies on less alcohol, less smoking, less depressions, less diabetes and more exercising (Koopmans et. al., 2012; VWS, 2013; VWS 2016). Furthermore, national vaccination programs that includes the whole population is part of this type of preventive healthcare. Secondly, there is selective preventive healthcare, targeted at high risk groups in society, for example flu vaccination program for elderly people or people with asthma; thirdly, there is indicated preventive healthcare, targeted at specific persons that do not yet have an indicated disease but have a high risk of developing this disease for example, when there is a heritable disease in the family; fourthly, health related preventive healthcare targeted at people that have an indicated disease as for example online platforms for people with a disease (Koopmans et.al., 2012). This type of preventive healthcare can be seen in the initiative “Luchtbrug”. This is an online platform for children with Asthma where they can find online help by connecting online to Asthma nurses, find information about their lung tests and find information about what to do when a worsening of their tightness happens (EY, 2016).
Supply and demand related explanations
There are several theories in social sciences that help to explain the healthcare expenditures. In this section is explained how the existing theories regarding healthcare expenditures could also hold for preventive healthcare. According to Vatter and Refli (2003) there are two approaches dominating this debate about healthcare expenditures, the supply-side and the demand-side approach. This approach and explanation is clarified by Newhouse (1977). Newhouse states that over 90% of the variance of healthcare expenditures can be explained by per capita healthcare expenditures. He was the first scholar that clarified this relationship, but he was followed by many others in the years after (e.g. Duan et. Al, 1982; Cucker & Sisko, 2013). Relating this to preventive healthcare expenditures, we can find various explanations by taking a supply – demand approach. Since the article by Newhouse, many scholars (including Duan et. Al, 1982; Cucker & Sisko, 2013) followed him in his approach on explaining healthcare expenditures.
Newhouse (1977) found an income elasticity of healthcare expenditures between 1.15 and 1.31 for 13 OECD countries with data from 1970, with a variation of per capita Gross Domestic Product (GDP) explaining around 90% of the variation in per capita healthcare expenditure. Other scholars in the 1980s and beginning of 1990s found similar results that healthcare could be seen a luxury good (Leu, 1986). This means that when people or a nation becomes wealthier, so an increase in GDP, the expenditures will also increase.
Gerdtham et. al. (1992) also found in their systematic analysis of relationships across 19 OECD countries that per capita GDP contributes significantly to the explanation of the health care expenditure variation between countries. Where Newhouse did not add institutional and policy explanations, Gerdtham has added these perspectives to his research.
To know whether these theories would hold for preventive healthcare expenditures, it is important to know first how these causal mechanisms work for healthcare expenditures. When national income increases, the budget of the government also increases. When the total budget of the government increases, the share that is spend on healthcare might also increase. This increase might be in absolute numbers as well as in relative numbers. The absolute number increases when the total budget increases and the percentage of expenditures/share remains the same. The relative increase might also occur, when its decided to increase the percentage of expenditures. There might be an increase in this percentage since there is a high demand for budget for healthcare. When other departments do not need more money, this part can shift to the healthcare department. This results in the conclusion that national income influences healthcare expenditures positively. This causal mechanism is concluded in several studies (Newhouse, 1977; Gerdtham et. al., 1992). However, these mechanisms might not work the same for preventive healthcare. Preventive healthcare is a percentage of total healthcare expenditures. As we can see in the table below (table 1), the preventive healthcare expenditures as a percentage of total healthcare expenditures are not moving considerable. Therefore, the absolute expenditures will increase when national income increase, since the absolute healthcare expenditures increase and the share of total healthcare expenditures increases. Furthermore, the preventive healthcare expenditures as a percentage of total healthcare expenditures will not increase, by definition, when national income increases. It remains a percentage of total healthcare expenditures. The ratio between preventive healthcare and curative healthcare is not influenced, by definition, by an increase of national income. An increase will more likely happen when there is demand for a change in the ratio between preventive healthcare and curative healthcare.
Dataset: Health expenditure and financing Year 2010 2011 2012 2013 2014 Country Australia 1,8 2,0 1,8 1,9 .. Austria 1,8 2,0 2,0 2,0 2,1 Belgium 1,6 1,7 1,7 2,1 2,1 Canada 5,8 5,9 6,0 6,1 6,1 Chile .. .. .. .. .. Czech Republic 2,5 2,3 2,1 2,7 3,2 Denmark 2,3 2,3 2,2 2,5 2,4 Estonia 3,4 3,3 3,5 3,0 3,0 Finland 3,3 3,3 3,3 3,3 3,3 France 2,1 2,0 2,0 2,0 1,9 Germany 3,4 3,2 3,2 3,1 3,2 Greece 1,2 1,1 1,0 1,1 1,4 Hungary 3,9 3,4 2,9 2,8 2,6 Iceland 1,5 2,4 2,5 2,5 2,3 Ireland .. .. .. 2,8 2,7 Israël 0,6 0,5 0,4 .. .. Italy 2,9 2,8 2,8 2,9 4,1 Japan 3,0 2,6 2,6 2,5 .. Korea 3,2 3,2 3,4 3,5 3,6 Latvia 2,9 3,3 2,9 0,7 2,0 Luxembourg 1,9 2,3 2,1 2,1 2,1 Mexico 4,0 4,1 3,0 3,4 3,2 Netherlands 4,0 3,7 3,7 3,6 3,7 New Zealand .. .. .. .. ..
Norway 2,6 2,7 2,8 2,9 2,9 Poland 2,1 2,1 2,0 2,6 2,6 Portugal 2,1 2,1 2,0 1,8 1,8 Slovak Republic 5,7 2,8 4,2 2,1 1,9 Slovenia 3,7 3,9 3,8 3,8 3,1 Spain 2,3 2,2 2,1 2,0 2,0 Sweden 3,5 2,9 2,9 3,1 3,0 Switzerland 2,4 2,3 2,2 2,2 2,2 Turkey .. .. .. .. .. United Kingdom .. .. .. 4,2 4,1 United States 3,3 3,1 3,1 3,0 3,0
Table 1: OECD data set: 2010 -2014. Preventive healthcare expenditures as a % of current health expenditures. Source: World Bank
The quantitative part of this study will test the relationship between national income and preventive healthcare as a percentage of total healthcare expenditures. This possible causal mechanism results in the following hypothesis:
Hypothesis 1:
The level of national income does not explain the low level of preventive healthcare expenditures.
Hey and Patel (1983) suggest that “the quantity of preventative care purchased by the government unambiguously increases with a fall in the price of preventative healthcare”. Meredith (2013) supports this older study by concluding “as in many other recent studies, we find that demand for preventative health products is highly price sensitive.” Since this was a case study in Kenya, a low developing country, it would be interesting to test whether this conclusion also holds in a developed country as in this single case study of the Netherlands.
When looking deeper into the literature concerning preventive healthcare, we can find several specific demand explanations concentrated at preventive healthcare. Henkel (1994) states that the demand for preventive healthcare increases when the health insurance coverage increases.
When looking at the institutional problem of wrong incentives in the healthcare system, we can see that this is also a demand related problem. “The case of fully insured individuals is an exception, in which demand equals the entire population, since individuals face no cost of purchase. An insurer covers preventative healthcare for the entire population only if effectiveness and the discount factor are high enough, and nobody if not” (Vandoros & Carmen, 2011).
The demand for new preventive healthcare initiatives is therefore not very high. Furthermore, in general health providers as medical professionals often get paid on production base (EY, 2017). This means that a doctor has a higher income when he has more patients. This would mean that when looking at demand and supply of preventive healthcare, doctors could face a decrease in patients when there is a higher percentage spend on preventive healthcare since it prevents people from getting sick or getting sicker. This could lower the demand from doctors for preventive healthcare. Furthermore, it might be possible that doctors would increase the price of preventive healthcare to compensate for fewer patients and to lower the attractiveness of preventive healthcare. This is exactly the opposite of the effect preventive healthcare is trying to achieve. These explanations are from a demand related approach but this is a result of institutional problems for preventive healthcare.
The theories on the relationship between price and the demand and supply for preventive healthcare expenditures are tested in the qualitative single case study in the Netherlands. The theories result in the following hypothesis:
Hypothesis 2:
The price of preventive healthcare does explain the low preventive healthcare expenditures in the Netherlands.
The academic literature on demand and supply related explanations give a broad overview on the general explanations, however the academic scholars did not focus on whether these general explanations hold for a single case study in the Netherlands. Therefore, this study would be interesting for knowing whether the Dutch healthcare system reacts the same to demand and supply influences in relation to preventive healthcare.
(Political)Institutions explanations
enduring collection of rules and organized practices, embedded in structures of meaning and resources that are relatively invariant in the face of turnover of individuals and relatively resilient to the idiosyncratic preferences and expectations of individuals and changing external circumstances” (March, 1989). “There are constitutive rules and practices prescribing appropriate behavior for specific actors in specific situations. There are structures of meaning, embedded in identities and belongings: common purposes and accounts that give direction and meaning to behavior, and explain, justify, and legitimate behavioral codes. There are structures of resources that create capabilities for acting. Institutions empower and constrain actors differently and make them more or less capable of acting according to prescriptive rules of appropriateness” (Rhodes & Rockman, 2008). The neo-institutionalist approach describes that policy differences are largely due to variances in the design of political institutions and to variances in the strategies of the interdependent collective actors (Immergut, 1990; Immergut, 1992). These variances in institutions as described by said scholars result in large differences in healthcare systems. Vatter and Ruefli (2003) state that institutions are of key importance in the evolution of welfare systems from a policy viewpoint. Tibandebage and Mackintosh (2005) explain that rules and sanctions in such a system are usually complemented by implicit agreements between stakeholders, and they emphasize the influence of these implicit agreements on performance of shared understandings. Bloom et al. (2008) characterize the arrangement of formal and informal rules and understandings as an implicit contract, which underpins the effective functioning of a health system. “These arrangements make the agreements possible that are necessary for translating expert knowledge into trusted and trustworthy services and managing social arrangements for healthcare finance” (Bloom et. al, 2008). In the absence of these arrangements, health sector actors rely predominantly on relationships shaped by the market and informal rules (Bloom, 2011).
The most central political institutions that are a major influence on the healthcare system are explained by numerous scholars (Bloom et. Al., 2008; Immergut, 1990; Immergut, 1992; Nørgaard, 1996; Vatter & Ruefli, 2003). The most important function of political institutions is the ability to ensure stability in policy outcomes and institutional arrangements through mechanisms that permit a core of political representatives to veto legislative changes and proposals (Immergut, 1990). The political institutions form the voting system in a country which is often funded on a system of majority voting. Only a majority in a parliament and senate can change legislation. This difficult system to pass new legislation ensures stability in policy outcomes and institutional arrangements. The culture of the national political systems and institutional structure determine the interaction within the system and determine the processes of policy development and implementation, their strategic limits and options (Nørgaard, 1996). Changing healthcare institutions is dependent on the presence of veto opportunities, because these opportunities are a critical aspect of decision
structures (Immergut, 1990). She points out that many scholars have explained that it is difficult to obtain majority votes for new policy proposals because of these veto opportunities. “For almost any proposal that can garner a majority of votes within a particular political arena, an alternate proposal can be found that will attract an equally large number of votes. It is particularly difficult to draft a proposal that can prevail over the previous status quo” (Shepsle 1986). The system is also designed that radical political changes (as we can also describe the change towards a more preventive healthcare system) are impeded due to the political institutions (Immergut, 1990). Most voting systems are based on majority voting. This implies that new legislation needs to have a majority in parliament and in the senate. The system prevents radical changes to pass easily through a system. For preventive healthcare, this implies that (most often) the government must be in favor of preventive healthcare. The opinion of the government is formed by the largest party or a coalition of parties. These parties must have a positive view on preventive healthcare in order to have a majority in parliament and the senate that will vote positively for legislation and that will stimulate preventive healthcare and the financing of preventing healthcare.
Morone (1990) states in his article that most scholars give “the mixture of interest-group muscle and a cultural dread of government as the explanation for slow health reform in the United States, but he explains that focusing on the political institutions theory gives an alternative. “National health reform has been wrecked by the organization of our government. The fragmentation of checks and balances, rooted in the Constitution and exacerbated by each generation of political reform, creates a policy apparatus that is maladroit at securing broad policy changes. Our government produces distributive benefits to narrow claimants; big reform efforts generally result in stalemate” (Morone, 1990). Daly (2011) states about the healthcare system in the United States that "For decades our healthcare system has been designed to treat patients once they're sick,". This is the opposite from preventive healthcare. Therefore, in order to increase the preventive healthcare, the system should change to prevent people from getting sick. Financing of healthcare should be changed, compensation for medical professionals and health insurers should change. Daly gives a very concise overview of why he thinks the system does not work anymore in the United States, including statement by politicians, however, he does provide concrete solutions. Another institutional problem with the current healthcare is that system is focused on a short time horizon while preventive healthcare requires a medium and long-time horizon when assessing healthcare programs. (Coretti & Rugeri, 2015).
The above explained theories lay down the system wherein healthcare needs to be changed. It is important to understand how these theories and mechanisms relate to healthcare expenditures and
Medisch Genootschap Nederland plead for more laws and regulations in the institutions in order to be able to make a more effective change towards preventive healthcare. This indicates that laws and regulations need to be made, to be able to make a move towards a system focused on preventive healthcare. It explains the mechanism with an influence of (political) institutions on public policies. Without institutions, a change is often not made since a change most often require means. Organizations need to have a stimulus to make a change. Institutions form the common purpose, direction and action for increasing preventive healthcare expenditures. These institutions are made by the legislators in a country, thus in order to have institutions that guide to preventive healthcare, this subject needs to be high on the agenda of the legislators. These legislators can be influenced by various parties and pressure groups. To test the influence of (political) institutions on laws and regulations regarding preventive healthcare, a qualitative study is performed in a single case study of the Netherlands. The theories result in the following expectations.
Expectation 1:
(Political) institutions influence the laws and regulations regarding the stimulation of preventive healthcare in the Netherlands.
Expectation 2:
(Political) institutions influence the laws and regulations regarding the financing of preventive healthcare in the Netherlands.
Political power
As a result of the institutions, there are possibilities in the system that allow for powerful groups to influence the political system. Freidson (1986), notes that “to gain insight into the full range of professional powers, we must move outside the workplace and into the broader political economy. . . . [and look at] those who are in a position to influence the policies of the state on which the special position of the profession depends.” Peterson states that professional power was established in the healthcare sector because there is an information asymmetry for society (Peterson, 2001). Furthermore, he states that professionals in the healthcare sector “have specialized knowledge that the average person, lacking the profession’s formal training, cannot acquire or easily interpret. Gladstone (1995) states “The power of the medical profession lies in its success in having secured by political means a legal monopoly over the practice of healing in contemporary society. This made the doctor the official expert on health and illness in modem society, a title enshrined in written law”. As captured by the old cliché, however, knowledge is power—not only power evident in the application
of a profession’s knowledge about particular decisions in its standard workplace, but influence over the entire social structure that defines and regulates the environment in which that work is accomplished” (Peterson, 2001). This means that society has less information about healthcare practices, treatments and policies, so society has given their trust to the professionals according to Peterson. The influence (power) of professionals has risen since society has given their trust to the professionals. Immergut (1990;1992) states that in the healthcare system, the medical practice has this strong political power. This political power is the result of the monopoly of the medical practice according to Immergut. When doctors would receive less income because of decreasing patient numbers due to changes in the healthcare system, the medical practice is likely to use political power to stop these changes.
These articles might explain the mechanism of an influence of powerful groups on public policies. The professionals might influence choices of the government regarding preventive healthcare. The articles provide no information on the single case of the Netherlands, but they do provide explanation that I can use to test these explanations for political power of profession groups in the Dutch case. Furthermore, these articles are only focused on political power, without placing it in a more general view on the problems with changing the healthcare system. To test the degree of political power and the influence of that political power on laws and regulations regarding preventive healthcare, a qualitative study is performed. The theories result in the following expectations.
Expectations 3:
The professionals in the healthcare system in the Netherlands have a strong political power.
Expectation 4:
The professionals in the healthcare system prevent the healthcare system to change towards a preventive healthcare system.
The literature does not show an area of disagreement, but it shows that there are no scholars that focused in their research on a single case study in the Netherlands. I can combine these different theories from previous scholars to find out why the Dutch government only spends 4% of the total health budget on preventive healthcare. I can look at this problem from a demand-side approach, a supply-side approach, from a political institutions approach and from a political power approach. These theories will give a broad perspective on the problem in the Netherlands, however, it will be more concise than existing literature because it focuses on one country.
To summarize the literature on demand and supply related explanations, political institutions, and political power, is that preventive healthcare expenditures are likely to behave according to a demand and supply model that influence the opportunities and constrains for public policies concerning preventive healthcare. Income, price and quantity might all have an influence on preventive healthcare expenditures. Furthermore, the political institutions might influence this model by influencing these public policies. Political power that is obtained by several professional groups is also influencing public policies.
From this literature review, the two hypotheses and four expectations mentioned in this theoretical framework are formulated based on the found academic literature. The hypotheses substantiate the main research question of this paper, “Why are the expenses of preventive healthcare in the Netherlands limited to 4% of the total healthcare budget?”.
Research design
To answer my stated research question “Why are the expenses of preventive healthcare in the Netherlands limited to 4% of the total healthcare budget?” I will perform a positivist mixed method research. I will try to find a causal relationship between income and preventive healthcare expenditures and between the problems in the healthcare system and the preventive healthcare expenditures.
In my literature review I found several theories that might explain low preventive healthcare expenditures, a supply explanation, a demand explanation, a political institutions explanation and a political power explanation. These different theories have leaded to four hypotheses:
Hypothesis 1:
The level of national income does not explain the low level of preventive healthcare expenditures.
Hypothesis 2:
The price of preventive healthcare does explain the low preventive healthcare expenditures in the Netherlands.
The other theories are not suitable to put into hypotheses. It is not possible to define one independent variable of these theories that has a causable relation on the dependent variable. However, I can state that I expect that the political institutions theory explains the low preventive healthcare expenditures in the Netherlands. This is possible via:
Expectation 1:
(Political) institutions influence the laws and regulations regarding the stimulation of preventive healthcare in the Netherlands.
Expectation 2:
(Political) institutions influence the laws and regulations regarding the financing of preventive healthcare in the Netherlands
Furthermore. I can state that I expect that political power explains the low preventive healthcare expenditures in the Netherlands. This is possible via:
Expectations 3:
The professionals in the healthcare system in the Netherlands have a strong political power.
Expectation 4:
The professionals in the healthcare system prevent the healthcare system to change towards a preventive healthcare system.
Hypothesis 1 will be tested by performing a quantitative study. I will try to find the relationship between national income and preventive healthcare expenditures as a percentage of total healthcare expenditures in OECD countries. Then I will continue to focus on a single case, the Netherlands, to test hypothesis 2 and the expectations 1 - 4. I will do expert interviews with an employee from the Ministry of VWS (Ministry of Volksgezondheids, Welzijn en Sport, Ministry of Health, Welfare, and Sport), employees of municipalities, employees of health insurers, employees of hospitals, employees of EY and representative of medical professionals.
Data
For the first hypothesis concerning the relationship between national income and preventive healthcare expenditures as a percentage of total healthcare expenditures, I will use data from the
OECD and of the World Bank on preventive healthcare expenditures as a percentage of total healthcare expenditures and national incomes of OECD countries.
For the second hypothesis and the expectations, I will gather data by expert interviewing and reviewing of reports and policy documents. Some of these sources can be labelled as secondary literature. Secondary literature is literature that explains or discusses information given at the original source. The secondary literature is more likely to generalize and critical address primary literature. The different information sources will provide information that will be the data for my single case study. The information coming from the interviews will be supported by academic literature, reports and policy documents on the single case of the Netherlands.
Methodology
In this study, there are several hypotheses to be tested. In order to test these hypotheses, I will conduct a mixed-method research. A mixed-method research is a research method where quantitative and qualitative methods are combined in one single research (Bryman, 2007). A mixed methods design is used because this method gives the opportunity to get a broad international view on a possible large influence of national income on preventive healthcare expenditures and the opportunity to have an in-depth case study by interviewing actor in the Dutch healthcare system. One of the hypothesis can only be tested by performing a quantitative study, the hypothesis that states that the level of national income does not explain the low levels of preventive healthcare expenditures in the Netherlands. The other hypothesis, that the price of preventive healthcare does not explain the low level of preventive healthcare expenditures in the Netherlands, and the expectation that political institutions and political power do explain the low level of preventive healthcare expenditures can be tested by conducting interviews, analyzing report and analyzing policy documents. The various sources and a critical reflection of all sources will result in a funded analysis.
The hypotheses and expectations are tested with the mixed method. The weight of the qualitative research method is larger since it will test one hypothesis and two expectations and is focused on the case study of the Netherlands.
For this research, I will use the explanatory sequential design (Bryman, 2007), what is explained in the following figure. I will start to test whether there is a correlation between national income and preventive healthcare expenditures as a percentage of total healthcare expenditures in an international analysis of 29 OECD countries. The data set preventive healthcare expenditures as a
percentage of total healthcare expenditures is chosen because this is more interesting than preventive healthcare expenditures as a percentage of national income. Since preventive healthcare expenditures are part of the total healthcare expenditures and I want to know whether there is a shift from curative to preventive healthcare and how we can achieve this shift, it is interesting to study the proportion of preventive healthcare expenditures to the total healthcare expenditures and see whether there is a shift due to a higher national income. Then I will look in depth in the results of the Netherlands, and use those /this result for a qualitative study to test the other hypotheses.
Figure 1: Explanatory sequential design Source: Bryman, 2007
Quantitative methodology
In order to know whether there is a correlation between national income of OECD countries and preventive healthcare expenditures of OECD countries, I will conduct a research that examines this influence. I will study the possible effect of changes in national income of OECD countries on preventive healthcare expenditures as a share of current expenditure on health. This is a quantitative study on this possible effect.
The first step was collecting the data on preventive healthcare expenditures as a share of current expenditure on health from OECD countries. This data could be found in a data set provided by the OECD on http://stats.oecd.org/Index.aspx?DataSetCode=SHA. The countries that did not have available data or countries that had only data available for a few years were removed.
Data on GDP was taken from the World Bank
(http://data.worldbank.org/indicator/NY.ADJ.NNTY.PC.KD.ZG?locations=OE. The data set was first adjusted to the available OECD countries of the data set with preventive expenditures. Then the data set was adjusted to the same available years as the data set with preventive healthcare expenditures. These years were for all countries between 2000 and 2015. The countries that did not have sufficient data of have did not have data at all were removed. The chosen criteria for sufficient data was ten years per country. The data set contained after removal of said countries, 29 countries (appendix 1).
The data was analyzed in two steps: first the correlation between national income and preventive healthcare expenditures as a percentage of the total healthcare expenditures was calculated for each country, then the correlation coefficients were analyzed by calculating the descriptive statistics. This two-step approach was chosen because I want to study whether the average correlation between national income and preventive healthcare expenditures as a percentage of the total healthcare expenditures is different from zero. Therefore, the first step of calculating the correlation coefficients was needed to be able to take the second step of studying the average influence.
Per country the variables national income and the share of preventive healthcare expenditures were imported in Eviews and a correlation test was performed. The correlation coefficients are defined as sample coefficients (Howitt, 2007). Next, per country a Scatterplot was made of the correlation between the two variables national income and preventive healthcare as a percentage of total
healthcare expenditures (Appendix 2).
Furthermore, the correlation coefficients were collected in a new Excel file in order to generate a new data set to be able to test whether the correlation coefficients are on average significantly different from zero. The value of the correlation coefficients (r) can be in the range -1 < r < +1. Level of significance is p < 0.05
After collecting all correlation coefficients between the variables national income and preventive healthcare expenditures, the descriptive statistics of this data were calculated in Eviews.
As the descriptive statistics showed that the Skewness and Kurtosis are not in the acceptable range of one point below or one point above the levels of Skewness and Kurtosis of the perfect normal distribution, it was concluded that this sample of correlation coefficients was not a normal distribution. This entails that I could not perform a standardized Z-test. This implies that I need another test that is able to test whether data is on average significantly different from zero. In order to be able to test whether there is an influence of national income on preventive healthcare expenditures as a percentage of total healthcare expenditures, there were two hypotheses formulated. The null-hypothesis is that the correlation coefficient mean of the total sample has a correlation of zero. The alternative hypothesis states that the correlation coefficient mean of the correlation is not equal to zero.
𝐻0: 𝜇(𝑟) = 0 𝐻1: 𝜇(𝑟) ≠ 0
To be able to test these hypotheses, a one sample t – test could be performed. Since the alternative hypothesis stated that the correlation is not equal to zero, the expectation was not that this is
one-sided, so this was a two-tailed one sample t-test. In Eviews, the function “simple hypothesis tests” was used for this test. The mean that was tested was the mean of the sample with a mean of zero, as is stated in the null hypothesis.
After the one sample t-test was performed, the degrees of freedom of this sample was determined. Since there were 29 observations in this sample, the degrees of freedom were 28. To be able to test the significance at a 5% level, the critical values were found for these values. This gave critical values of – 2,048 and 2,048.
Since the said test does not include the effect of time on the preventive healthcare expenditures as a percentage of total healthcare expenditures, a second test was performed. The data set was corrected by taking GDP annual growth (%) into account. The data set with GDP annual growth (%)
was taken from the website
http://data.worldbank.org/indicator/NY.GDP.MKTP.KD.ZG?end=2014&locations=OE&start=2000. The preventive healthcare expenditures as a percentage of total healthcare expenditures were multiplied with the annual GDP growth (%) to exclude the possibility that the preventive healthcare expenditures as a percentage of total healthcare costs have risen in the same proportion as GDP has increased. The new data set was used to perform the same test as stated above. First the correlation coefficients were calculated by using Eviews (appendix 3). This was followed by calculating the descriptive statistics. Next, the one student t-test was performed to test the above stated hypotheses.
Qualitative methodology
The second part of this study uses a qualitative research design. The qualitative part of this study focuses on a single case study of the Netherlands.
Case selection
Gerring (2004) states that there are several criteria for the case study. It means that its method is qualitative and that there is a small-N (Yin, 1994), that it is “clinical, participant-observation, or otherwise in the field” (Yin, 1994), that the study researches the properties of a single case (Campbell & Stanley, 1963), or that that “the research investigates a single phenomenon, instance, or example” (Gerring, 2004). He concludes that the case study is best defined as “an intensive study of a single unit with an aim to generalize across a larger set of units” (Gerring, 2004). He also states that case studies are more useful “when insight into causal mechanisms is more important than insight into
In this research, the Dutch case is studied as a case of the international study on the influences on preventive healthcare expenditures. The case study selection has one desire; “to be a representative sample” (Seawright & Gerring, 2008). Applying this desire to the case study selection in this research, this criterium would be met since the Netherlands is one of the countries in the world that has the highest healthcare costs. Therefore, the Netherlands faces large challenges in financing their healthcare. This is less of a problem for countries with low healthcare expenditures, so the Netherlands is representative for countries that have a high interest in changing their healthcare system to a more sustainable one, based on preventive healthcare.
Figure 2: Health expenditures per capita, 2013 (or nearest year) Source: OECD
The case selection of the Netherlands would also meet the second desire of being useful for the variation on the dimensions of theoretical interest. Since the Netherlands is an outlier concerning correlation between national income and preventive healthcare as a percentage of total healthcare expenditures, in the international study of the influence of national income on preventive healthcare expenditures as a percentage of total healthcare expenditures. The explanation for being an outlier will give an extension of the theoretical information on the explanations for the low preventive healthcare expenditures in countries. Since I will study the outlier case, this is a deviant case selection method (Seawright & Gerring, 2008)
Figure 3: Correlation coefficients without annual growth (%) taken into account.
Figure 4: Correlation coefficients with taking annual growth (%) into account
Furthermore, the Dutch case is of substantive importance because the assignment for this research is given by EY Netherlands. A practical reason for choosing the Netherlands is that the research is done to graduate at the University of Amsterdam, a Dutch university.
The case study is based on interviews, secondary literature as policy documents from EY, Ministry of VWS (Ministry of Health), RIVM and ZonMW. These policy documents will match the period that is studied in the quantitative section, 2000 – 2015. These combined information channels will give the best overview on the Dutch case. This will be the most effective method to find the answer to the research question. -1.00 -0.75 -0.50 -0.25 0.00 0.25 0.50 0.75 1.00 2 4 6 8 10 12 14 16 18 20 22 24 26 28 correlation coefficients -1.00 -0.75 -0.50 -0.25 0.00 0.25 0.50 0.75 1.00 2 4 6 8 10 12 14 16 18 20 22 24 26 28 correlation coefficients The Netherlands The Netherlands
Participants and setting
Since this research is a single case study of the Netherlands, the participants for the expert interviews are selected from organizations in the Netherlands. In order to have a sample of experts that can give the best information about the views and situation from the perspective of their organization, the experts were selected from diverse organizations. The GGD Amsterdam was selected since it is an organization within the municipality. The National General Practitioners Association was selected since this organization represents a large share of the general practitioners in the Netherlands and they are an important performing of preventive healthcare. Together with the Dutch General Practitioners Society, the next organization selected, they are a strong advocate of preventive healthcare in the Netherlands. The Dutch General Practitioners Society represent also a large share of the general practitioners in the Netherlands, and is therefore able to speak on behalf of these general practitioners. Achmea was selected, since it is the largest health insurer of the Netherlands. As a health insurer, they are a large influence in the health system. The GGZ was selected since it is an organization that already made a change towards more preventive healthcare. Furthermore, the GGZ is a large organization that has a large influence on the healthcare expenditures in the Netherlands. The final organization that was selected for these interviews, is EY. Since healthcare and healthcare financing is a focus point of EY and therefore many organizations seek advice from EY, it is a major influence in the healthcare sector.
The participants for the expert interviews were selected for their position at one of the following organizations.
GGD Amsterdam (Geneeskundige en Gezondheidsdienst Amsterdam) GGD Amsterdam – Medical and Health Service
Landelijke Huisartsen Vereniging
National General Practitioner Association Nederlands Huisartsen Genootschap
Dutch General Practitioner Society Achmea
GGZ (Geestelijke Gezondheidszorg) Mental Health Organization EY
The experts were interviewed at the office of the organization or during an interview by phone. The interview questions can be found in appendix 5. The interviews were recorded with the voice recorder of a laptop and notes were made in order to ask more in-depth questions when needed.
Data generation
The data was gathered with the coded expert interviews, policy documents and reports. As said, the interviews were recorded and after the interview, the recording was transcribed by using the program MAXQDA. The transcribing process focused on the parts of the interview that contained important, useful information. The data was coded according to the beforehand set labels. These labels were selected based on the question that will be used for the data analysis. These questions can be found in the next section “Data Analysis”. The data was selected based on the labels that can be found in the following table.
Labels Randvoorwaarden Beleid Rol organisatie Preventieve zorg Wet- en regelgeving Verdienmodel Financiering Partijen Bereidheid verandering Politieke invloed Zorgverzekeraar Prijs Figure 5: Labels
Data analysis
The data was analyzed by using the labelling method. This means that the interview transcripts were read and that certain words, phrases or sentences were given a label. All coded parts of the interview were collected and the overall view on a particular matter is given in the analysis. These general views and perspectives were used to make the statements about the hypotheses. The theory section was used as the base for the analysis. The expectations that come out of the theory and literature were used to analyze the interviews. The questions that were base for the analysis were:
Does price of preventive healthcare have an influence on the demand or supply of preventive healthcare in the Netherlands?
Does the health insurance coverage have an influence on the demand for preventive healthcare?
Is the production based payment system for medical professionals an influence on the support for preventive healthcare?
What political institutions form a constrain for preventive healthcare?
Are there any groups in the political system that can veto changes towards preventive healthcare?
Is there a national willingness to change the healthcare system into a system that is more focused on preventive healthcare?
What actors are constrained to finance preventive healthcare and preventive healthcare initiatives?
Results
The question of the quantitative research part of this study asked whether there is a relation between national income and preventive healthcare expenditures as a percentage of total healthcare expenditures. In the literature review is explained that a relationship between national income and total healthcare expenditures is often found by numerous scholars, however a relationship between national income and preventive healthcare expenditures was not yet tested. In this study, I tested the first hypothesis.
Hypothesis 1:
The level of national income does not explain the low level of preventive healthcare expenditures.
This hypothesis was transformed for the quantitative part of this research. 𝐻0: 𝜇(𝑟) = 0
𝐻1: 𝜇(𝑟) ≠ 0
This graph shows the descriptive statistics of the correlation coefficients of the 29 OECD countries. As can be seen in the graph, there is a wide variaty of correlation coefficients of the different countries. This graph gives a first impression that that is a correlation between said national incomes and preventive healthcare expenditures as a percentage of total healthcare expenditures.
Figure 6: Distribution and descriptive statistics
Observations: 29 Mean: 0.281867 Median: 0.421423 Maximum: 0.939475 Minimum: -0.921766 Std. Dev.: 0.564810
To test whether this first impression is right and whether there is a significant correlation between the two variables, I performed a one sample t-test in Eviews. The results are given in the table below. Hypothesis testing for Series
Sample 1 29
Included observations: 29
Test of hypothesis mean: 0.000000
Sample mean = 0.281867 Sample Std. Dev. = 0.564810
Method Value Probability
t-statistic 2.687.458 0.0120
Table 2: Student t-test result
The table shows that the t -statistic of this test is 2,687458. As was explained in the methodology section, the critical values of this test are – 2,048 and 2,048. The t-statistic falls in the critical region and therefore, I can reject the null hypothesis and can conclude that there is a significant positive correlation between national income and preventive healthcare expenditures as a percentage of total healthcare expenditures. This is the same correlation that was found by numerous scholars for the relationship between national income and total healthcare expenditures.
As was stated in the methodology, the same tests were performed including a correction for the annual growth of national income. To exclude this influence on the results, the same annual growth rate was applied to the preventive healthcare expenditures as a percentage of total healthcare. The following graph shows the distribution of this second test. We can see in the descriptive statistics that the mean is much closer to zero, and the minimum and maximum are around the same values.
Figure 7: Distribution and descriptive statistics
To know whether there is a significant correlation between national income and preventive healthcare as a percentage of the total healthcare expenditures, I conducted a student t-test for the second time. The results of the student t- test are shown in the data below.
Hypothesis testing for Series Sample 1 29
Included observations: 29
Test of hypothesis mean: 0.000000
Sample mean = 0.042381 Sample Std. Dev. = 0.626974
Method Value Probability
t-statistic 0.364014 0.7186
Table 3: student t-test result
We can see that these results show now that the t-statistic is 0,364014. As was explained before in the methodology section, the critical values of this test are – 2,048 and 2,048. The t-statistic does not fall in the critical region and therefore, I cannot reject the null hypothesis and can conclude that there is no significant correlation between national income and preventive healthcare expenditures
Observations: 29 Mean: 0.042381 Median: 0.172819 Maximum: 0.880798 Minimum: -0.958215 Std. Dev.: 0.626974
