The relationship between medical innovation and health expenditure before and after health reform

Accepted Manuscript

THE RELATIONSHIP BETWEEN MEDICAL INNOVATION AND HEALTH EXPENDITURE BEFORE AND AFTER HEALTH REFORM Songul Cinaroglu PhD , Onur Baser MA, MS, PhD PII: DOI: Reference:

S2211-8837(18)30247-8 https://doi.org/10.1016/j.hlpt.2018.10.001 HLPT 305

To appear in:

Health Policy and Technology

Please cite this article as: Songul Cinaroglu PhD , Onur Baser MA, MS, PhD , THE RELATIONSHIP BETWEEN MEDICAL INNOVATION AND HEALTH EXPENDITURE BEFORE AND AFTER HEALTH REFORM, Health Policy and Technology (2018), doi: https://doi.org/10.1016/j.hlpt.2018.10.001

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

THE RELATIONSHIP BETWEEN MEDICAL INNOVATION AND HEALTH EXPENDITURE BEFORE AND AFTER HEALTH REFORM TITLE: THE RELATIONSHIP BETWEEN MEDICAL INNOVATION AND HEALTH

AUTHOR NAMES AND AFFILIATIONS

Corresponding Author:

AN US

Songul Cinaroglu, PhD Hacettepe University Faculty of Economics and Administrative Sciences (FEAS) Department of Health Care Management 06800 Beytepe Ankara-Turkey +90 546 808 08 18 [email protected] [email protected] [email protected]

CR IP T

EXPENDITURES BEFORE AND AFTER HEALTH REFORM

CE

PT

ED

M

Onur Baser, MA, MS, PhD Chief Scientific Officer, STATinMED Research Associate Professor of Surgery, Columbia University Adj. Professor of Internal Medicine, University of Michigan Professor of Economics, MEF University Editor-in-Chief, Journal of Health Economics and Outcomes Research 145 Hudson Street, Suite 200 New York, NY, 10003 212-603-9565 T 734-646-7991 M [email protected] www.statinmed.com Acknowledgement

AC

This study was supported by a research grant of The Scientific and Technological Research Council of Turkey (TUBITAK) with a grant number 1059B141500020. The sponsor had no role in the study design, collection and analysis of data, the writing of the report or the submission of the paper for publication.

Conflict of Interest Statement The authors declare that they have no conflict of interest.

1

ACCEPTED MANUSCRIPT Highlights  Positive linkage observed between medical innovation and health expenditures.  Significant positive link is stronger in time periods after health reform than before.  Reformist environment moderates the interrelationship between variable groups of

CR IP T

medical innovation and health expenditures.

OBJECTIVES: This study explores the relationship between a group of medical innovations and a set of health care expenditure variables before the introduction of health reform and

AN US

after. METHODS: We assessed two groups of data taken from the databases of the World Intellectual Property Organization and the Organisation for Economic Co-operation and Development (OECD) for the periods of 1990–2000 (before health reform) and 2001–2011 (after health reform). The number of patent publications in relation to medical technology,

M

biotechnology, and pharmaceuticals was considered indicators of medical innovation. The

ED

total public health expenditure, health expenditure per capita, and health expenditure as a share of GDP were considered indicators of health care expenditure. Our dataset included 14

PT

OECD countries that introduced health care reforms between 2000 and 2012. Canonical correlation analysis was used to measure the degree of association. RESULTS: There is a

= 0.91; p < 0.001) and after (rc,

a

= 0.94; p < 0.001) health reforms were introduced.

AC

b

CE

strong positive correlation between medical innovation and health care expenditure before (rc,

Additionally, the degree of this correlation is higher after the health reform period (2001– 2011) than before (1990–2000). CONCLUSIONS: Improved communication channels within societies increase international co-operation, and cost–benefit analysis of medical innovation may help in continuing to improve medical innovation and health care accessibility. We hope that our observations promote the understanding of the relationship

2

ACCEPTED MANUSCRIPT and balance between medical innovation and health care expenditure while fostering an incentive atmosphere for health reforms. Key Words: Medical Innovation, Health Expenditures, Health Reform, Canonic Correlation Analysis INTRODUCTION

CR IP T

In recent years, developed countries have been facing increasing health care costs. Population aging and increased life expectancy escalate health expenditure [1]. Therefore, policymakers implement health care reform for controlling the costs and improving health care outcomes [2]. Providing affordable health care insurance and access for all and simultaneously

AN US

confronting increased health care expenditure has been a major objective of health care reforms [3]. Although cost containment is crucial [4], there is a growing recognition that the cost control alone is inadequate to maintain publicly funded health care systems [5].

M

Innovation and technology are recognized as the drivers of increasing health expenditure [4], and encouraging innovation and nurturing a supportive environment for technological

ED

improvement are the motivations for future strategies of health reform policies [3].

PT

Expenditure on medical technology has rapidly increased since the late 1990s [6, 7]. This

CE

growth is in parallel with the general health expenditure trends in Western economies [2]. Population aging is one of the major reasons for increased health care spending because an

AC

increased number of dependent elderly significantly increases health care expenditure [7]. Moreover, lifestyle variables such as smoking, physical activity, and dietary habits are highly associated with financial burden, and there is little interest in econometric modeling of their health care cost, which is why policymakers feel the need to fight them [3]. Debates on health care costs and new medical technologies have focused on improving the pharmaceutical industry and the quality of health care services [7]. However, Sorenson et al. (2013) [4] state

3

ACCEPTED MANUSCRIPT that the impact of technology on health care costs differs across different types of technologies, diseases, and drugs, including anticancer drugs and invasive medical devices. Medical technology, biotechnology, and pharmaceuticals are the major areas of medical innovation. Medical technology includes medicines, vaccines, and medical devices, which are essential for public health [8]. It has been argued that the increasing cost and use of medical

CR IP T

technology is contributing to the escalating health care costs [9]. Expenditure and sustainability of the health care industry depend on profits to fund future innovations [10]. Smith et al. (2013) [11] are of the opinion that innovation in medical technology is strongly affected by economic crises and price erosions. Supporting this statement, Archibugi et al.

AN US

(2013) [12] state that economic crises cause companies to reduce their investment in innovation. On the other hand, economic crises can be drivers of innovation with respect to providing more innovative alternatives. However, previous studies claim that economic crises

M

lead to a concentration of innovative activities by new firms, and those firms are highly innovative even before a crisis starts [12]. The disquieting effect of economic crisis on

ED

medical innovation in Greece was examined, and a decline was observed in relative biomedical-research productivity starting shortly after the financial crisis began [13].

PT

Although encouraging innovation is key in the development of the health care sector,

CE

promoting innovation needs serious assessment in order to examine the impact of current

AC

economic crises [13].

Biotechnology is another costly area in relation to health care technology. It generally concerns the application of cellular and molecular biology methods to make or modify products or processes [14]. Pharmaceuticals relate to the production of medicines and have played a key role in reducing mortality due to diseases and improving the general health of people all over [15]. In addition, pharmaceuticals are integral to medical technology.

4

ACCEPTED MANUSCRIPT According to Grabowski and Vernon (1994) [16], the pharmaceutical industry is one of the most innovative industries. The number of patents published by the industry helps us understand its nature, determinants, and the level of medical innovation. In addition, the number of patents helps us understand the international patterns in medical innovation. This is because health care technology firms, laboratories, and individuals can apply for a patent to

CR IP T

protect new technology and to prove their technological competence and innovativeness [17]. The importance of patents as contributors to innovation has received support by other studies, and patenting activity has seen a sharp rise of late [18]. A number of previous studies have focused on the number of patent publications, which is influenced by the priority given to

AN US

interventions and the amount of research activity [19]. Although patenting in pharmaceuticals and biotechnology requires time-consuming research and developmental processes, the field of pharmaceuticals plays a major role in patent protection [20], and patenting encourages

M

technological development in health care. Moreover, medical innovation reduces hospital

ED

utilization and the number of hospital stays, thereby increasing life expectancy [21].

Health policymakers are under increasing pressure to control the escalating health care

PT

expenditure. Therefore, the policymakers have become more interested in expensive areas

CE

such as medical innovation than . The policymakers try to strike a balance between the costs and the benefits of medical innovation [22]. As costs of medical innovation continue to rise,

AC

the policymakers become more interested in regulating them. The key question becomes— how do the benefits compare with the costs [23]? This requires us to assess whether and under which circumstances investments in medical innovation produce better value in health care [4]. We must, therefore, examine the relationship between medical innovation and health care expenditure. It is widely believed that new technologies have been a major cause of increasing health expenditure in several countries [24]. Previous studies suggest that although

5

ACCEPTED MANUSCRIPT pharmaceutical innovation is expensive, the benefits of adopting the innovation outweigh the expenses [24]. It is also well-known that innovative drugs can be profitable [25]. Furthermore, the high costs of medical technologies, biotechnologies, and pharmaceutical innovation are reflected in the high prices because of the patent system [24], and we have already discussed how patent applications have played an important role in the development of health care

CR IP T

technology [25]. A supportive environment for interorganizational networks influences the innovative output of pharmaceutical organizations. However, global cohesion among pharmaceutical firms, which is measured using the density of the overall network among pharmaceutical firms, has a negative impact on the innovative performance of a firm [26].

AN US

Advancement in medical technology is proportional to the development level of countries, and some developed countries play a synergetic role in the development of health technologies [27]. It is well-known that the Organisation for Economic Co-operation and

M

Development (OECD) countries are highly developed and industrialized. Because these countries have political, economic, demographic, and health status variables, the level of

ED

innovation and health technologies differ among even the developed OECD countries [27]. Prodan (2005) [28] analyzed the expenditure on research and development and the number of

PT

patent applications among OECD countries during 1981–2001. The study found that the share

CE

of health expenditure as percentage of GDP in relation to total public expenditure has increased significantly over the past 30 years. A number of OECD countries introduced health

AC

care reforms to control costs during that period, and attempts to introduce health care reforms have been undertaken by OECD countries between 2000 and 2012. Table 1 provides a brief summary on the health care reform policies in different OECD countries. The table shows that cost control, care coordination, changing payment, and reimbursement models were the major strategies underlying health care reform policies in the OECD countries from 2000 to 2012 [27].

6

ACCEPTED MANUSCRIPT

[Insert Table 1 here]

Here, we present a brief overview of health care reforms in some of the OECD countries. Austria

is

one

of

the

countries

that

have

introduced

reforms

since

2005.

CR IP T

The major aims of these reforms were to tackle increasing costs and social inequalities, along with improving the application of medical technologies. As a result of these reforms, incomerelated inequality in relation to health care access has decreased since 2005. It should be noted, however, that it remains lower than that in other countries [29]. In 2013, Austria

AN US

introduced legislation to improve efficiency through balancing care provision across health care providers by promoting new primary care models and better coordination of care. Because reform objectives should be achieved by unified decision making across key

M

stakeholders, it will be necessary to have a committed government capable of improving social insurance through targeting overcapacity, promoting efficiency, and achieving further

ED

progress, which is backed up by previous studies [30]. Canada has introduced health reform since 2000. Deber (2003) [31] is of the opinion that health reform in Canada has shaped well

PT

the overall health care organization and has produced an efficient system. Although

CE

substantial progress has been achieved over the last 10 years of health care reform, Canadians currently cannot access information on wait times for necessary medical care [32]. Health

AC

reform started in Chile at the beginning of 2005, and Bitran et al. (2010) [33] analyzed its effect on health outcomes. Their results showed that the reforms have had a positive effect on access to care, treatment outcomes, hospitalization rates, and medical leaves with respect to six chronic diseases. Current studies on the effect of health reform, which includes universal coverage, show that socioeconomic inequalities have reduced following the introduction of health reform [34]. Denmark initiated its health reform in 2007, which includes four major

7

ACCEPTED MANUSCRIPT areas: hospital admission procedures, rehabilitation, health promotion and disease prevention, and mental illnesses. Despite these efforts, studies have shown that health reform was not properly implemented in Denmark because of the lack of political support [35]. Moreover, reports on health reform indicate that patients often complain about the lack of continuity of care in Denmark [36]. Estonia experienced health reform since 2000. Lai et al. (2013) [37]

CR IP T

analyzed the effects of health reform policies in Estonia, and their conclusions support the theory that despite reforms, inequalities in health status and health outcomes remain. Moreover, after the health reform, out-of-pocket health expenditure of households is still distributed among poor population groups. Germany is an OECD country with a very high

AN US

living standard and a very skilled and productive society [38]. Busse and Blümel (2014) [39] hypothesize that policymakers are increasingly interested in regulating and evaluating health care technologies in Germany in recent years, and health care reform has been around since

M

2004. However, the quality of care and inequalities in health care still need to be addressed in spite of reform efforts. In Japan, the medical innovation capability of the health care industry

ED

is high; however, there is a widely held belief that technological advancement is the primary driver of increases in health care costs [40]. The Japanese system works to keep prices low,

PT

particularly in the area of diagnostics. This policy makes diagnostics, especially, available

CE

[40]. In 2006, Japan billed the Medical Law Reform and Pharmaceutical Affairs Act, which was legislated in 2009. Major changes introduced by the revised law are addressing and

AC

classifying over-the-counter drugs according to the risk involved, requiring licensed pharmacists or registered sales persons to sell them, and establishing an informative pharmaceutical environment [41].

Scholars suggest that Japan needs a health care system designed to overcome inequalities with respect to age, employment status, and family situation. The Japan Vision: Health Care 2035

8

ACCEPTED MANUSCRIPT has offered a future perspective on improving health care [42]. In Mexico, the poor are not covered by social security. Furthermore, a high proportion of the population finances health care through out-of-pocket expenses [43]. Mexican health reform started in 2001; Frenk et al. (2006) [43] analyzed its effect and noticed that it produced substantial growth in the public resources devoted to health care and increased health coverage. In 2012, the Mexican

CR IP T

government introduced a public health insurance system called Seguro Popular, which covers a wide range of services. However, analyses show that the implementation of Seguro Popular faced challenges such as limited institutional capacity, limited information systems, and ineffective use of financial resources. Mexico's experience with the implementation of health

AN US

reform may offer several insights for other countries to extend health coverage [44]. The Netherlands performs well in health care compared with most other OECD countries [45]. The Netherlands health care system has been undergoing transformation since 2006. Schafer

M

et al. (2010) [46] state that new medical technologies have been used in the Netherlands with the support of rational policymaking under a health reform policy since the 1990s.

ED

International comparisons show that the Netherlands has lower antibiotic use, fewer number of avoidable hospitalizations, and a lower avoidable mortality. However, growth has slowed

PT

considerably after reverting to more traditional sector agreements on spending [47]. Among

CE

the European countries, Norway can be placed at the top in relation to the development level of its health care system [48]. Ringard et al. (2013) [49] claim that Norway has the highest

AC

level of physician density among the OECD countries. However, lengthy waiting times have been a serious problem in the Norwegian health care system, and several efforts have been undertaken to reduce them. The new treatment choice reform was implemented in 2015. In earlier reforms, less attention was paid to patient rights, whereas current reforms reinforce patient choice and increase the supply of both public and private services. It might be advisable for other OECD countries to tackle high waiting times by following the

9

ACCEPTED MANUSCRIPT developments in Norway [49]. Portugal has experienced health reform since 2005. Barros (2012) [50] states that the net effect of reform policies on health outcomes is not clearly predictable in Portugal. In Slovakia, technology assessment is carried out through a ―categorization‖ process. Pharmaceuticals, medical devices, dietary foods, and diseases are categorized using this system. However, the system is politically controversial and not very

CR IP T

effective [51]. Over the years, several patient groups have formed because of health reform, which are increasingly vocal in policymaking, service provision, and advocacy [52]. Turkey is another developing country and a member of the OECD and has been experiencing health reforms since 2003. The establishment of health reform in Turkey has improved the primary

AN US

health care system, insurance coverage, and provision of incentives for paying for the performance and improvement of family medicine [53]. However, Turkey has witnessed health expenditure increasing dramatically following the health reforms. This has led to

M

increased interest in health technology assessment. Furthermore, the number of studies being undertaken on cost-effectiveness increased after the 2003 reforms in Turkey [54]. The reforms

ED

have shown that financial sustainability has been one of the major challenges of the health system in Turkey. Improved accessibility, technology, and higher expectations from citizens

PT

force a health care system to become financially sustainable [54]. In the U.S., a high level of

CE

health expenditure is one of the major challenges. Accessibility and health care outcomes are the two major causes of disparities in the U.S. [31]. Political institutions, economic forces, a

AC

powerful medical establishment, and a dominant private insurance industry are the priorities of the U.S. health care system transformation [55]. Medical innovation and technology assessment have a long history in the U.S., and several public organizations regulate and control the health technology market. In addition, many large private insurers and pharmacy companies have their own health technology improvement plans [56]. Blank (2012) [55] defends the view that gaps in universal health coverage will continue in the near future in the

10

ACCEPTED MANUSCRIPT U.S. health care system because the U.S. is not constraining the availability of hightechnology medicine and is broadening basic health care services, unlike single-payer Western nations. Furthermore, the U.S. is quickly and erratically moving to transform its health care financing model away from fee-for-service and toward a global payment structure that rewards quality and efficiency over time. The Affordable Care Act (ACA) enables

CR IP T

Americans to begin catching up with universal coverage goals and systems well established in the rest of the developed countries. The ACA's health care delivery system reforms and other innovations are good examples in improving quality, effectiveness, and efficiency [57].

AN US

These brief summaries on health reform programs in OECD countries emphasize that the primary focus of reform programs is to control costs and improve health outcomes. Medical innovation is one of the key drivers of health care expenditure. None of these reform

M

programs prioritize improving medical technology and innovation. In addition, research suggests that the effect of pharmaceutical innovation on health outcomes needs to be better

ED

studied than that of other types of medical technologies [58]. Moreover, Lichtenberg (2015b) [58] defends the view that, to date, research is scant on the effect of biotechnology and

PT

medical technology innovation on health expenditure. There remains a need for further studies

CE

on the effect of innovation on health expenditure and the link between these variables.

AC

Understanding the relationship between medical innovation and health care expenditure is crucial for policymakers to make effective decisions on resource allocation [23]. Studies on health care innovation show that a majority of improvement efforts in the health care sector fail to produce a sustained impact. Lack of managerial support and funding limits innovation in health care [59]. Therefore, health policymakers and economists are faced with increasing costs in health care and medical innovation, which is often regarded as the main factor

11

ACCEPTED MANUSCRIPT explaining the rising health care expenditure [60]. Moreover, research suggests that technological innovation without cost containment also leads to rising health care expenditure [61, 62]. Evidence from developed countries suggests that medical innovation accounts for only half of the increase in health expenditure [60]. However, the correlation between medical innovation and health expenditure remains from clear. In other words, the relationship

CR IP T

between medical innovation and health expenditure is essentially a chicken-and-egg problem. There is a need for a supportive political environment for controlling the costs of health technologies and innovation. One of the long-term aims of health reform policies is to provide the framework for such a supportive environment for innovation and technological

AN US

development. Resource allocation decisions, cost containment plans, and payment mechanism reorganization are the major strategies of a reformist direction in health care. Although it is expected that reform policies will provide a supportive environment for improving health care

M

innovation and technology, no previous studies focus on the relationship between medical innovation and health care expenditure under a health reform policy. We therefore attempted

ED

to uncover the relationship between medical innovation and health care expenditure variables separately before and after health reform periods in certain OECD countries. This makes it

PT

possible to compare the degree of correlation between medical innovation and health

CE

expenditure during two periods that represent time before and after health reform. The following sections provide an explanation on materials and methods, findings of the study,

AC

discussion and policy implementations, and conclusions.

MATERIALS AND METHODS 2.1. Study Design Process Canonical correlation analysis (CCA) was used to measure the degree of association between the set of medical innovation variables and the set of health expenditure variables for two

12

ACCEPTED MANUSCRIPT different periods representing the time before and after health reform. As discussed earlier, the number of patent publications in relation to medical technology, biotechnology, and pharmaceuticals was considered an indicator of medical innovation. Moreover, three different functions of health expenditure were constructed as health expenditure variables. These were public health expenditure (current; US$)), health expenditure per capita (current; US$), and

comprised

central

government

expenditure,

CR IP T

total health expenditure as a share of the GDP (%)). Total public health expenditure state/regional/local

government

health

expenditure, and compulsory contributory health insurance expenditure. A canonical structure for a group of medical innovations and a set of health expenditure variables was analyzed for

AN US

two different study periods, representing before (1990–2000) and after (2001–2011) health reform for 14 OECD countries. These countries underwent reform between 2000 and 2012,

M

whereas 20 countries did not [27].

The canonical correlation results of the two variable sets were examined and compared.

ED

Before performing CCA, logarithmic transformation was applied to both sets of variables to linearize the relationship. A multivariate normality distribution assumption has been provided

PT

for the two sets of variables. Furthermore, Spearman correlation coefficients (rs) showed that

CE

there was no multicollinearity in either set of variables and for all variables in the dataset. 2.2. Dataset

AC

The data came from the World Intellectual Property Organization (WIPO) and OECD statistics for the years 1990–2011. Fourteen OECD countries were selected for this study, each of which experienced health reform during 2000–2012.

2.3. Canonical Correlation Analysis

13

ACCEPTED MANUSCRIPT CCA was performed to examine the link between two sets of variables, medical innovation, and health expenditure. CCA was performed for these two variable groups for two different time periods separately (i.e., 1990–2000 and 2001–2011). CCA is a technique used to find the sets of vectors that maximize the correlation between a set of variables. This analysis was developed by Hotelling (1936) [63] to examine the relationship between different groups of

CR IP T

variables. CCA is one of the common methods of statistics, pattern recognition, and data mining [64]. It is a well-known method for dimension reduction, clustering, and classification [65].

AN US

The main principle of CCA is to find a solution for a linear combination of two sets of variables, with the form: Xi = a1 (predictor 1) + a2 (predictor 2) + … + am (predictor m) and Yi = b1 (outcome 1) + a2 (outcome 2) + … + am (outcome m). The main goal of this method is

M

to identify the coefficients that maximize the correlation between a set of X i and Yi. After the analysis, the first canonical correlation shows the highest correlation between the two sets of

ED

variables. The results show the correlation between the two variable groups [66].

PT

3. FINDINGS

3.1. Descriptive Statistics

CE

Table 2 shows descriptive statistics of the study variables. We present two sets of estimates of

AC

14 OECD countries for two time periods (i.e., 1990–2000 and 2001–2011). The median scores of the variables for medical innovation before health reform were as follows: number of patent publications with respect to medical technology (min. 1; max. 10,905; median 197.5), biotechnology (min. 1; max. 6734; median 180), and pharmaceuticals (min. 3; max. 6431; median 384). For the period after health reform, they were as follows: medical technology (min. 2; max. 24,538; median 99.5), biotechnology (min. 1; max. 15,447; median 71), and pharmaceuticals (min. 1; max. 16,441; median 93). Descriptive statistics for health 14

ACCEPTED MANUSCRIPT expenditure before reform were as follows: health expenditure (public; min. 532; max. 568.159; median 10.786), health expenditure (per capita; min. 144; max. 4.559; median 1.499), and health expenditure (% of total GDP; min. 2.51; max. 12.51; median 7.18). For the period after health reform, they were as follows: health expenditure (public; min. 549; max. 1.229.400; median 24.139), health expenditure (per capita; min. 434; max. 8.171; median

Insert [Table 2] here

AN US

3.2. Canonical Correlation Results

CR IP T

2.727), and health expenditure (% of total GDP; min. 4.73; max. 16.41; median 8.74).

During the analysis, logarithmic transformation showed that the distribution became symmetric, and a multivariate normality assumption was provided for both the set of variables and for both time periods (p < 0.001). Spearman correlation coefficients of the two sets of

M

variables indicated that there was no multicollinearity problem between (rs < 0.70) the two

ED

sets of variables for both time periods (i.e., 1990–2000 and 2001–2011). The results also showed a strong positive canonical correlation between medical innovation and health

PT

expenditure according to the observations representing before health reform (rcb = 0.91;

CE

p < 0.001) and after (rca = 0.94; p < 0.001). A summary of the results on the statistical significance of canonical correlations is presented in Table 3. The null hypothesis is that there

AC

is no relationship between medical innovation and health expenditure for both the time periods. Bartlett's chi-squared test was used to test the null hypothesis. The results showed that the null hypothesis was rejected for the first roots of the two sets of variables for the two time periods, 1990–2000 (

= 281.48; p < 0.001; λ = 0.152) and 2001–2011 (

= 342.14;

p < 0.001; λ = 0.101). Figure 1 shows the number of canonical roots and canonical score plots for the first set of canonical roots. The first set of roots has high statistical and practical significance in 15

ACCEPTED MANUSCRIPT understanding the relationship between medical innovation and health expenditure for before and after health reform. Scatterplots indicate a stronger positive linear relationship between medical innovation and health expenditure after health reform (rc, a = 0.94) than before (rc, b

= 0.91) (Table 3; Figure 1).

AN US

Insert [Figure 1] here

CR IP T

Insert [Table 3] here

DISCUSSION AND POLICY IMPLICATIONS

Medical innovation and technological advancement have enabled significant advances in patient care and improvement in health outcomes [67]. Furthermore, medical innovation and

M

improvements in public health are vital for the prosperity of any country. In other words, the

ED

innovation ecosystem of the medical industry has a potential to improve public health [68]. Lichtenberg (2014) presents a developed country experience and shows that pharmaceutical

PT

innovation increased mean age at death by 0.29 years (3.43 months) from 2000 to 2009 [69]. Although innovation creates additional benefits to the society, it is costly and does not always

CE

result in growth or profitability [70]. Therefore, cost containment is one of the primary policy

AC

areas of financial management to combat rising pharmaceutical costs [71]. The number of patent publications is one of the indicators of medical innovation and one of the costliest areas of innovation. These are policy tools that are used to encourage innovation. They encourage health decision makers to create new knowledge. [72]. Moreover, patents influence the development of and access to new medicines for both the developed and developing countries. The number of patents published is dependent on different natural, scientific, and technological factors [73]. McCormack et al. (2015) [67] note that fostering an open 16

ACCEPTED MANUSCRIPT innovation environment is one of the strategies for improving innovation capabilities and sharing innovative ideas in the medical market. Therefore, stakeholders, values, experiences, expertise, and empirical data shape and inform policy debates on the number of patents. However, the lack of a publicly available dataset to evaluate medical innovation is one of the major limitations for studies on medical innovation. International organizations like the WIPO

CR IP T

publish reports on patent applications in the health sector. These reports follow new trends in policy, business, and technology worldwide [74]. Research on the dynamic nature of medical innovation shows that innovation and technology in health care do not follow a linear model but instead constitute a dynamic process [75]. Such an uncertainty of the outcome of the

AN US

process, however, can make it difficult for policymakers to understand and manage the interrelated process of medical innovation and technology in health care [70]. In addition, the widespread diffusion of new medical technology overwhelms the efforts of policymakers to

M

control increasing costs [61]. Therefore, cost control is one of the promising and popular

ED

strategies of managing pharmaceutical market and health reform policies [4].

Although medical and pharmaceutical companies make innovation efforts more available,

PT

affordable, and accessible in developed countries, it is not the case with developing countries,

CE

where the companies are trying to expand their market. These efforts need support for meeting the needs of low-income groups and for improving the accessibility to medicines

AC

[76]. It is widely believed that medical innovation is expensive and uncertain [77]. The numbers of patent publications for different types of medical technologies are indicators of a supportive environment for the development of health technologies and a supportive political environment in health care. An environment that promotes widespread access to effective medical technologies is one of the primary challenges for policymakers. It is widely accepted that medical innovation involves several expensive activities and is a key driver of rising

17

ACCEPTED MANUSCRIPT health expenditure [68]. To tackle this complexity and the increasing costs in the pharmaceutical industry, a considerable body of literature addresses the cost-effectiveness of medicines and the pharmaceutical market [78]. Although there is a vast literature about controlling costs and policy efforts help to assess and improve health technologies, there is not much information concerning the relationship between medical innovation and technology

CR IP T

and health expenditure [23].

Policymakers in the developed countries are interested in determining whether additional health care expenditure will result in better health outcomes [79]. The OECD countries are

AN US

among the most highly developed and wealthiest countries in the world. Magnussen et al. (2007) [80] speculate that in recent years, financial and political power has been leveraged to improve health outcomes and health expenditure in several European countries. These

M

policies could be the result of fashionable policy trends and transformation programs in health care. To follow this fashion, a number of European countries have experienced health reforms

ED

in recent years. The OECD countries that have experienced health reforms have diverse health care and financial systems. Therefore, the level of health outcomes and health expenditure is

PT

different in these countries. Furthermore, understanding how to tackle increasing costs is one

CE

of the primary aims of health reforms. Ozcan and Khushalani (2016) [27] report that among the 34 OECD countries, 14 countries underwent health reforms between 2000 and 2012,

AC

whereas the remaining 20 did not. Austria, Canada, Chile, Denmark, Estonia, Germany, Japan, Mexico, the Netherlands, Norway, Portugal, Slovakia, Turkey, and the U.S. are the countries that experienced health reforms during this period. Anderson and Hussey (2001) [79] state that the policymakers in OECD countries are highly interested in determining whether additional health care expenditure will result in better outcomes. Sorenson et al. (2013) [4] suggest that the policymakers should focus not only on the costs but also on

18

ACCEPTED MANUSCRIPT whether medical technology innovation results in a better value in health care and decreases in inequalities. The motivation behind this focus is to increase the awareness on how the money spent on medical innovation, technology, physicians, pharmaceuticals, and other health services improves the overall health status of the population. Moreover, cross-national comparisons and analyses will help us understand whether additional health expenditure

CR IP T

results in better health outcomes over time [79].

It is important to emphasize that health expenditure is not the same for all OECD countries. Compared with other OECD countries, the U.S. spends a larger amount of money on health

AN US

care. Moreover, the OECD countries experiencing economic growth witness a proportionally rapid increase in health expenditure [79]. This results in an increase in the number of studies on the efficiency of health systems in the OECD countries. In turn, this increasing interest

M

produces a considerable quantity of research on the efficiency of health systems in the OECD countries [2]. Hakkinen and Joumard (2007) [81] analyzed the efficiency of health systems

ED

and concluded that the OECD countries must focus on improving health outcomes while simultaneously controlling costs to have efficient health care systems. Willeme and Dumont

PT

(2015) [7] concur with Hakkinen and Jourmard (2007) [81] and state that the level of income

CE

and costs are the determining factors in the accessibility to health care services. Moreover, there is vast literature assessing the efficiency of health care systems and the effect of health

AC

reform policies on the system efficiency in developed countries. Ozcan and Khushalani (2016) [27] showed that the OECD countries that improved their public health systems were likely to show improvement in efficiency after a decade of health reforms.

It is clear that medical innovation and technology are the major drivers of increasing health care costs [23]. Understanding the results of medical innovation and the relationship between

19

ACCEPTED MANUSCRIPT innovation and health expenditure is critical in fully assessing the impact of health reform policies [67]. Despite this clarity, there is a need for a detailed analysis on the relationship between medical innovation and health care expenditure under a reformist environment. This paper attempts to fill this deficit in the literature by comparing the relationship between a set of medical innovation and a group of health expenditure variables before and after health

CR IP T

reform periods separately. Our results show a strong positive correlation between a group of medical innovation variables and a set of health expenditure variables before (1990–2000) and after (2001–2011) health reform periods. Moreover, we found that the strength of the relationship between medical innovation and health expenditure is higher after health reform

AN US

period than before. According to our knowledge, this study emphasized the strong linkage between a group of medical innovation and a set of health expenditure variables and observed this relationship while considering two independent time periods that are before and after

M

health reform. Study results ascertain that the relationships between two groups of variables

ED

are stronger for after health reform period than before.

In light of this study, it is suggested for health policymakers to determine health policies to

PT

manage the strong relationship between medical innovation and health expenditure variable

CE

groups while considering reformist atmosphere. The heart of this advice is to foster a supportive environment to improve medical innovation and health technology while

AC

controlling costs. The results of this study take a step further and observe a high relationship between variable groups after health reform period than before. This result confirms the idea that a supportive environment and policies to improve medical innovation are good indicators of increasing health expenditure. Furthermore, under reformist environment, astrong relationship between variable groups is observed than before. In this paper, we contribute to the understanding of the relationship between medical innovation and health expenditure,

20

ACCEPTED MANUSCRIPT which has been a key issue of debate for years. It is anticipated that further studies will contribute to a better understanding of the link between innovation and expenditure under the effect of health reform policies using different kinds of indicators and focusing on countries that have different development levels.

CR IP T

CONCLUSION

It is commonly accepted that technological advances are among the most important indicators of growth in health care spending over time. Moreover, health reform is the background factor and determines the direction of this relationship. To our knowledge, no previous study

AN US

assessed the relationship between medical innovation and health expenditure variables considering the periods before and after health reform, which was surprising. The present study represents the first attempt to analyze this relationship in relation to OECD countries,

M

and our results show a strong link between the variables of medical innovation and health care expenditure. This significant positive link is especially stronger for separately analyzed

ED

datasets representing the time after health reform than those representing the time before. To summarize, it is important to keep in mind that the success of any health reform effort

PT

depends on realistic goals rather than popular policy trends. We hope that our study will

CE

increase policymakers' awareness on the strong relationship between medical innovation and health expenditure, especially in relation to a reformist direction. We also hope that this study

AC

stimulates further research on the relationship between medical innovation and health expenditure. This is particularly important because additional research on this relationship, with a focus on the developing countries, is paramount in ascertaining whether the objectives of health care reform can be achieved on a regular basis, along with controlling costs and improving health outcomes.

21

ACCEPTED MANUSCRIPT REFERENCES 1. Meijer C, Wouterse B, Polder J, Koopmanschap M. The effect of population aging on health expenditure growth: a critical review. EJA 2013;10(4):353-61. 2. Hadad S, Hadad Y, Simon-Tuval T. Determinants of health care system’s efficiency in OECD countries. Eur J Health Econ 2013;14(2):253–65.

CR IP T

3. Fielding JE, Teutsch S, Koh H. Health reform and healthy people initiative. Am J Public Health 2012;102(1):30-3.

4. Sorenson C, Drummond M, Khan BB. Medical technology as a key driver of rising health expenditure: disentangling the relationship. Clinicoecon Outcomes Res.

AN US

2013;5:223–34.

5. Werle J, Dobbelsteyn L, Feasel AL, Hancock B, Job B, Makar L, Manning H, Quigley S, Teppler A, Smith C, Kelly S, Wasylak T. A study of the effectiveness of

M

performance-focused methodology for improved outcomes in Alberta Public Health care. HMF 2010;23(4):169-74.

ED

6. DiMasi JA, Hansen RW, Grabowski HG. The price of innovation: new estimates of drug development costs. J Health Econ 2003;22(2):151–85.

PT

7. Willeme P, Dumont M. Machines that go ―ping‖: Medical technology and health

CE

expenditures in OECD countries. Health Econ 2015;25(3):387-88. 8. Torbica A, Cappellaro G. Uptake and diffusion of medical technology innovation in

AC

Europe: what role for funding and procurement policies? J. of Med. Market 2010;10(1):61–9.

9. Kumar RK. Technology and healthcare costs. Ann Pediatr Cardiol 2011;4(1):84-6. 10. Drummond MF, Griffin A, Tarricone R. Economic evaluation for devices and drugs – same or different? Value in Health 2009;12(4):402-4.

22

ACCEPTED MANUSCRIPT 11. Smith BD, Tarricone R, Vella V. The role of product life cycle in medical technology innovation. J. of Med. Market 2013;13(1):37–43. 12. Archibugi D, Filippetti A, Frenz M. The impact of the economic crisis on innovation: evidence from Europe. Technol Forecast Soc Change 2013;80(7):1247-60. 13. Falagas ME, Bardakas V, Mavros MN. Biomedical research productivity in Greece:

14. Bartholomew

S.

National

systems

of

CR IP T

effect of the financial crisis. Int J Epidemiol 2012;41(4):1206-7. biotechnology

innovation:

complex

interdependence in the global system. JIBS 1997;28(2):241–66.

15. Lichtenberg FR. The impact of pharmaceutical innovation on premature mortality,

Policy (2015-a) ;13(2):207–22.

AN US

cancer mortality and hospitalization in Slovenia, 1997–2010. Appl Health Econ Health

16. Grabowski H, Vernon J. Innovation and structural change in pharmaceuticals and

M

biotechnology. Ind Corp Change 1994;3(2):435–49.

17. Ramani SV, Looze MA. Using patent statistics as knowledge base indicators in the

ED

biotechnology sectors: an application to France, Germany and the U.K. Scientometrics 2002;54(3):319–46.

PT

18. Roper S, Hewitt-Dundas N. Knowledge stocks, knowledge flows and innovation:

CE

Evidence from matched patents and innovation panel data. RP 2015;44(7):1327-40. 19. Cecere G, Corrocher N, Gossart C, Ozman M. Technological pervasiveness and

AC

variety of innovators in Green ICT: A patent-based analysis. RP 2014;43(10):1827-39.

20. Harhoff D, Reitzig M. Determinants of opposition against EPO patent grants-the case of biotechnology and pharmaceuticals. Int J Ind Organ 2004;22(4):443-80.

21. Linchtenberg FR, Pettersson B. The impact of pharmaceutical innovation on longevity and medical expenditure in Sweden, 1997–2010: Evidence from longitudinal, diseaselevel data. EINT 2014;23(3):239–73.

23

ACCEPTED MANUSCRIPT 22. Wahlster P, Goetghebeur M, Kriza C et al. Balancing costs and benefits at different stages of medical innovation: a systematic review of multi-criteria decision analysis (MCDA) BMC Health Serv Res. 2015;15:1–12. 23. Baker L, Birnbaum H, Geppert J. et al. The relationship between technology availability and health care spending. Health Aff (Millwood). Web Exclusives

CR IP T

2003;W3-537–51. 24. Hsieh CR, Sloan FA. Adoption of pharmaceutical innovation and the growth of drug expenditure in Taiwan: is it cost effective? Value Health. 2008;11(2):334–44.

25. Jayadev A, Stiglitz J. Two ideas to increase innovation and reduce pharmaceutical

AN US

costs and prices. Health Aff (Millwood) 2009;28(1):165–68.

26. Guler I, Nerkar A. The impact of global and local cohesion on innovation in the pharmaceutical industry. SMJ 2012;33(5):535–49.

M

27. Ozcan YA, Khushalani J. Assessing efficiency of public health and medical care provision in OECD countries after a decade of reform. CEJOR 2016;25(2):325-43.

ED

28. Prodan I. Influence of research and development expenditures on number of patent applications: Selected case studies in OECD Countries and Central Europe, 1981–

PT

2001. AEID 2005;5-4:5–22.

CE

29. Hofmarcher MM. Austria health system review. Health System in Transition 2013;15:1–331.

AC

30. Hofmarcher MM. The Australian health reform 2013 is promising but requires continuous political ambition. Health Pol 2014;118(1):8-13.

31. Deber RB. Health care reform: Lessons from Canada. Am J Public Health 2003;93(1):20-4. 32. Simms C. Health reform in Canada. IJCP 2010;64(4):426-28.

24

ACCEPTED MANUSCRIPT 33. Bitran R, Escobar L, Gassibe P. After Chile’s health reform: increase in coverage and access, decline in hospitalization and death rates. Health Aff (Millwood) 2010;29(12):2161-70. 34. Cornejo-Ovalle M, Paraje G, Vasquez-Lavin F, Perez G, Palencia L, Borrell C. Changes in socioeconomic inequalities in the use of dental care following major

CR IP T

healthcare reform in Chile, 2004-2009. Int. J. Environ. Res. Public Health 2015;12(3):2823-36.

35. Andersen PT, Jensen JJ. Healthcare reform in Denmark. Scand J Public Health 2010;38(3):246-52.

AN US

36. Braithwaite J, Matsuyama Y, Mannion R, Johnson J, Bates DW, Hughes C. How to do better health reform: a snapshot of change and improvement initiatives in the health systems of 30 countries. IJQHC 2016;28(6):843-46.

Transition. 2013;15:1–196.

M

37. Lai T, Habicht T, Kahur K et al. Estonia: Health system review. Health System

ED

38. Müller-Jentsch W. Formation, development and current state of industrial democracy in Germany. Transfer: European Review of Labour and Research 2016;22(1):45-62.

PT

39. Busse R, Blümel M. Germany health system review. Health Systems in Transition

CE

2014;16(2):1–296.

40. Ikegami N, Campbell JC. Japan’s health care system: containing costs and attempting

AC

reform. Health Aff (Millwood) 2004;23(3):26–36.

41. Matsuo D. Effect of amendment to Japan’s Pharmaceutical Affairs Law. Nomura Research

Institute

(NRI)

Papers,

No.

149,

2009.

Retrieved

http://www.nri.com/global/opinion/papers/2009/pdf/np2009149.pdf. 16.2.2018.

25

Accessed

from: on:

ACCEPTED MANUSCRIPT 42. Miyata H, Ezoe S, Hori M, Inoue M, Oguro K, Okamoto T, Onishi K, Onozaki K, Sakakibara T, Takeuchi K, Tokuda Y, Yamamoto Y, Yamazaki M, Shibuya K. (2015) Japan’s vision for health care in 2035. Lancet 385(9987):2549-50. 43. Frenk J, Gonzalez-Pier E, Gomez-Dantes O. et al. Comprehensive reform to improve health system performance in Mexico. Lancet 2006(9546);368:1524–34.

CR IP T

44. Nigenda G, Wirtz VJ, Gonzalez-Robledo LM, Reich MR. Evaluating the implementation of Mexico’s Health Reform: The Case of Seguro Popular. HS&R 2015;1(3):217-28. 45. OECD

Better

Life

Index,

Netherlands,

Retrieved

from:

AN US

http://www.oecdbetterlifeindex.org/countries/netherlands/, Accessed on: 22.07.2016. 46. Schafer W, Kroneman M, Boerma W et al. The Netherlands Health System Review. Health Systems in Transition 2010;12(1):1–267.

M

47. Kroneman M, Boerma W, van den Berg M, Groenewegen P, de Jong J, van Ginneken E. Netherlands: Health Systems in Transition. 2016;18(2):1-240.

ED

48. Ringard A, Saunes IS, Sagan A. The 2015 hospital treatment choice reform in Norway: Continuity or change? Health Pol 2016;120(4):350-5.

PT

49. Ringard A, Sagan A., Saunes IS. et al. Norway: Health system review.

Health

CE

Systems in Transition 2013;15(8):1–162. 50. Barros PP. Health policy reform in tough times: the case of Portugal. Health Pol

AC

2012;106(1):17–22.

51. Szalay T, Pazitny P, Szalayova A. et al. Slovakia: Health system review. Health System in Transition 2011;13(2):1–203.

52. Dlouhy M. Mental health policy in Eastern Europe: a comparative analysis of seven mental health systems. BMC Health Serv Res 2014;14(42):1-8.

26

ACCEPTED MANUSCRIPT 53. Aktan AO, Pala K, Ilhan B. Health-care reform in Turkey: far from the perfect. The Lancet (London, England) 2014;383:25–6. 54. Tatar M, Mollahaliloglu S, Sahin B et al. Turkey: health system review. Health system in Transition 2011;13(6):1–186. 55. Blank RH. Transformation of the US Health System: Why is change so difficult?

CR IP T

Current Sociology 2012;60(4):415–26. 56. Rice T, Rosenau P, Unruh LY et al. United States of America: Health system review Health System in Transition 2013;15(3):1–431.

57. McDonough JE. The United States Health System in Transition. HS&R 2015;1(1):39-

AN US

51.

58. Lichtenberg FR. The impact of biomedical innovation on longevity and health. Nordic Journal of Health Economics (2015-b) http://dx.doi.org/10.5617/njhe.1290, Accessed

M

on: 21.07.2016.

59. Essen A, Lindblad S. Innovation as emergence in health care: unpacking change from

ED

within. Soc Sci Med 2013;93:203-11. 60. Newhouse JP. Medical care costs: how much welfare loss? J Econ Perspect

PT

1992;6(3):3-21.

CE

61. Bodenheimer T. High and rising health care costs. Part 2: Technological innovation. Ann Intern Med 2005;142(11):932-7.

AC

62. Djellal F, Gallouj F. Innovation in hospitals: a survey of the literature. Eur J Health Econ 2007;8(3):181-93.

63. Hotelling H. Relations between two sets of variates. Biometrika 1936;28(3/4):321–77. 64. Avron H, Boutsidis C, Toledo S. et al. Efficient dimensionality reduction for canonical correlation analysis. The 30th International Conference on Machine Learning, 2013; 1–9.

27

ACCEPTED MANUSCRIPT 65. Sun L, Ceran B, Ye J. A scalable two-stage approach for a class of dimensionality reduction techniques. 16th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2010:313–22. 66. Razavi AR, Gill H, Ahlfeldt H. et al. Canonical correlation analysis for data reduction in data mining applied to predictive models for breast cancer recurrence. STHI.

CR IP T

2005;116:175–80. 67. McCormack B, Fallon EF, Cormican K. An analysis of open innovation practices in medical technology sector in Ireland. Procedia Manufacturing 2015;3:503–09.

68. Daniel GW, Caze A, Romine MH et al. Improving pharmaceutical innovation by

(Millwood) 2015;34(2):319–27.

AN US

building a more comprehensive database on drug development and use. Health Aff

69. Lichtenberg FR. The impact of pharmaceutical innovation on longevity and medical

M

expenditure in France, 2000-2009. Econ Hum Biol. 2014;13:107-27. 70. Zuckerman B, Margolis PA, Mate KS. Health services innovation the time is now.

ED

JAMA 2013;309(11):1113-4.

71. Almarsdottir AB, Traulsen JM. Cost-containment as part of pharmaceutical policy.

PT

Pharm World Sci 2005;27(3):144-8.

CE

72. Gupta H, Kumar S, Kumar RS, et al. Patent protection strategies. J Pharm Bioallied Sci 2010;2(1):2–7.

AC

73. Grabowski H. Patents, innovation and access to new pharmacueticals. JIEL 2002;5(4):849-60.

74. WIPO, WTO (2012) Promoting access to medical technologies and innovation World Health Organization, World Intellectual Property Organization and World Trade Organization, Geneva, Switzerland.

28

ACCEPTED MANUSCRIPT 75. Neumann PJ, Sandberg EA. Trends in health care R&D and technology innovation. Health Aff (Millwood) 1998;17(6):111–9. 76. Hogerzeil HV. Big Pharma and social responsibility-the Access to Medicine Index. N Engl J Med 2013;369(10):896–9. 77. DiMasi JA, Grabowski HG, Hansen RW. Innovation in the pharmaceutical industry:

CR IP T

New estimates of R&D costs. J Health Econ 2016;47:20–33. 78. Barbieri M, Drummond M, Willke R. et al. Variability of cost-effectiveness estimates for pharmaceuticals in Western Europe: Lessons for inferring generalizability. Value Health 2005;8(1):10–23.

AN US

79. Anderson G, Hussey PS. Comparing health system performance in OECD countries. Health Aff (Millwood) 2001;20(3):219–32.

80. Magnussen J, Hagen TP, Kaarboe OM. Centralized and decentralized? A case study of

M

Norwegian hospital reform. Soc Sci Med 2007;64(10):2129–37. 81. Hakkinen U, Joumard I. Cross–country analysis of efficiency in OECD health care

ED

sectors. OECD Economics Department Working Papers, No. 554, ECO/WKP; 2007.

PT

Figure 1. Number of Canonical Roots and Canonical Scores Plot 1990–2000 (Before HR) Canonical Scores Plot (First Canonical Roots)

CE

Number of Canonical Roots Plot of Canonical Correlations

1,1

AC

1,0 0,9 0,8

Value

0,7 0,6 0,5 0,4 0,3

rc, b = 0,91

0,2 0,1 0,0 1

2

3

Number of Canonical Roots

29

ACCEPTED MANUSCRIPT

2001–2011 (After HR) OECD Country

The year reform started

ED

M

AN US

CR IP T

Description of health reform

Canonical Scores Plot (First Canonical Roots)

PT

Number of Canonical Roots Plot of Canonical Correlations

1,0 0,9 0,8

Value

AC

0,7

CE

1,1

0,6 0,5 0,4 0,3 0,2

rc, a = 0,94

0,1 0,0 1

2

3

Number of Canonical Roots

Table 1. Description of Health Reform in OECD Countries 2000–2012

30

ACCEPTED MANUSCRIPT

To change reimbursement for cost containment motives and to improve coordination of health care

Austria

To change primary care system in order to promote cost containment and for

Canada

Study Variables better coordination of health careObservation Period

2005 2000

1990–2000 (Before HR) To increase accessibility of health care with 1990–2000 chronic diseases, and to improve 2001–2011 2005 Degrees of 2 Resource Variable treatment outcomes, hospitalization rates, medical leave rates for chronic diseasesLambda (λ) Bartlett (X ) Probability rc, b NFreedom Min. Max. Median N Min. Max. Median To revamp the hospital sector, and to improve coordination across different public 2007 areas Medical Technology 0.91 281.48 <0.001 0.152 154 9 1 10905 197.5 154 2 24528 99.5 To0.26 improve hospitals and long-term care and to improve quality of care 2000 10.89 4 0.027 0.929 WIPO Bio Technology 154 1 coordination 6734 180 154 1 15447 71 To increase cost sharing and to2001–2011 promote care 2004 (After HR) To promotePharmaceuticals care coordination, to improve of care in 384 hospitals,154 create an1 Degreesquality 154 3of 6431 16411 93 2006 Bartlett (X2) market and cost containmentProbability Lambda (λ) rc, a informative pharmaceutical Freedom Health Expenditure To expand insurance coverage 2001 154 9 532 568.159 10.786 549 1.229.400 24.139 0.94 342.14 <0.001 154 0.101 (Public) (US$) To0.26 manage competition among insurers and promote cost containment 2006 11.68 4 0.019 0.924

Health Expenditure

OECD To develop

CR IP T

Number Variable Chile of Roots Group Removed Denmark 0 Estonia 1 Medical Innovation Germany Number of Japan Roots Removed Mexico 0 Netherlands 1

Health Expenditure

154 care 144 4.559 1.499coordination 154 434 primary care and hospital and to promote care and (Per Capita) (US$) cost containment

Norway

Health Expenditure

8.1712002, 2.727 2001, 2009

To reform primary care with the aim to promote care coordination and cost (As a Share of GDP) 154 2.51 12.51 7.18 154 4.73 containment(%)

Slovakia

To provide compulsory social health insurance and decentralize hospitals

2005

Turkey

To reorganize payment mechanisms, to promote quality through pay for performance, to improve insurance coverage, to improve public-private partnership and to improve family medicine

2003

United States

To widen health insurance coverage to change delivery and reimbursement models

2010

16.41 2005

8.74

M

AN US

Portugal

ED

Source: Ozcan Y.A. Khushalani J. (2016) ―Assessing efficiency of public health and medical care provision in OECD countries after a decade of reform‖ CEJOR, DOI 10.1007/s10100-016-0440-0, p.1–19. Accessed on: 21.07.2016.

PT

Table 2. Descriptive Statistics

CE

Table 3. Summary Results of Canonical Roots Between Medical Innovation and Health Expenditure Variables

AC

rc, b : Canonic correlation coefficient before health reform, rc, a Canonic correlation coefficient after health reform, X2: Bartlett’s Chi-square test

31