• No results found

Furthermore, it was hypothesized that the quality of the implementation process would act as a moderator on the relation between the degree of devolution reforms and the quality of elderly care

N/A
N/A
Protected

Academic year: 2021

Share "Furthermore, it was hypothesized that the quality of the implementation process would act as a moderator on the relation between the degree of devolution reforms and the quality of elderly care"

Copied!
76
0
0

Bezig met laden.... (Bekijk nu de volledige tekst)

Hele tekst

(1)

Devolving healthcare services to municipalities and the impact on the quality of elderly care: A comparative study of the Netherlands and Sweden

Chantalle Jacoline van Zanten Student number: 2189607

University of Groningen Faculty of Economics and Business MSc International Business and Management

January 18th 2017

Supervisor: drs. A. Visscher Co-assessor: dr. R.W. de Vries

(2)

2

ABSTRACT

This study researches the influence of the devolution reforms on the quality of elderly care, while also taking into account the influence of diversity of care arrangements and the quality of the implementation process. This research compares the Netherlands, where the devolution took place in 2015, and Sweden, where the devolution took place in 1992. It was proposed that the degree of devolution reforms and the diversity of care arrangements would have a positive effect on the quality of elderly care, and the degree of devolution reforms would also have a positive effect on the diversity of care arrangements. Furthermore, it was hypothesized that the quality of the implementation process would act as a moderator on the relation between the degree of devolution reforms and the quality of elderly care. A sample size of 69 responses in the Netherlands, and 5 responses in Sweden was collected. After applying statistical tests with the use of the statistical program SPSS, the hypotheses that the degree of devolution reforms and the diversity of care arrangements have a positive effect on the quality of elderly care have been supported for the combined dataset and the dataset of the Netherlands. None of hypotheses were supported for the dataset of Sweden. However, the results must be interpreted with care due to the small sample size. Therefore, future research is necessary to gain more knowledge on this intriguing and relatively recent topic.

Word count: 14949 words (excluding references and appendices)

Keywords: devolution; municipalities; quality of care; elderly; nursing homes; diversity;

decentralization; implementation

(3)

3

ACKNOWLEDGEMENTS

With this thesis, I have reached the end of my Master International Business and Management at the University of Groningen. I would like to thank my supervisor drs. A.

Visscher for his insights, feedback and suggestions. I would also like to express my gratitude to dr. H.J. Drogendijk for translating my survey to Swedish. Lastly, I would like to thank all who have participated in the study, and everyone who has supported me while writing my thesis.

(4)

4

TABLE OF CONTENTS

1. Introduction ... 6

2. Theoretical Considerations ... 12

2.1 Degree of devolution reforms ... 12

2.2 Diversity of care arrangements ... 16

2.3 Quality of the implementation process ... 18

3. Methodology ... 20

3.1 Data collection ... 20

3.2 Measurements ... 22

3.2.1 Devolution of healthcare services ... 22

3.2.2 Quality of elderly care ... 23

3.2.3 Diversity of care arrangements ... 24

3.2.4 Quality of the implementation process ... 24

3.3 Method of analysis ... 24

4. Results ... 25

4.1 Socio-demographic results ... 25

4.2 Assumptions ... 26

4.2.1 Skewness and kurtosis ... 28

4.2.2 Linearity ... 29

4.2.3 Normality ... 29

4.2.4 Multicollinearity ... 30

4.2.5 Homoscedasticity ... 31

4.3 Outcomes ... 32

4.3.1 Hypothesis 1: the devolution reforms on the quality of elderly care ... 32

4.3.1.1 Interpretation hypothesis 1 ... 34

4.3.2 Hypothesis 2: the devolution reforms on the diversity of care arrangements ... 34

4.3.2.1 Interpretation hypothesis 2 ... 35

4.3.3 Hypothesis 3: the diveristy of care arrangements on the quality of elderly care 36 4.3.3.1 Interpretation hypothesis 3 ... 37

4.3.4 Hypothesis 4: the quality of the implementation process as a moderator ... 37

4.3.4.1 Interpretation hypothesis 4 ... 40

5. Conclusion and Discussion ... 42

5.1 Conclusion ... 42

5.2 Discussion ... 44

6. Limitations and Future Research ... 46

6.1 Limitations ... 46

6.2 Suggestions for future research ... 47

References ... 49

Appendix 1: Socio-Demographic Results ... 54

Appendix 2: Statistical Outcomes ... 56

2.1 Multicollinearity testing ... 56

(5)

5

2.2 Homoscedasticity testing ... 57

Appendix 3: Survey English ... 61

Appendix 4: Survey Dutch ... 67

Appendix 5: Survey Swedish ... 72

(6)

6

1. INTRODUCTION

On January 14th 2014, van Rijn, the State Secretary of Health, Welfare and Sport, has sent the new law for Social Support (Wet Maatschappelijke Ondersteuning) to the House of Representatives. With this law, municipalities get more responsibilities for organizing support at home for people who cannot participate in the society on their own. The goal of this law is to make it possible for people to stay at home longer, and to better participate in society.

Municipalities are supposed to get €3.9 billion for the new responsibilities. The new law for Social Support took effect on January 1st, 2015 (Rijksoverheid, 2014).

On July 4th 2016 one of the headlines in the newspaper Trouw was: ‘Care for elderly is inadequate’. The Inspection of Healthcare performed a research involving 150 nursing homes.

The outcomes indicate that patient safety is worrisome in 38 nursing homes, and of those 11 cases were even extremely worrisome (Kuiken, 2016). Due to the fact that elderly have to stay at home longer, the care in the nursing homes becomes more complex. By the time elderly go to nursing homes, they require complex care for which most staff has not been educated. The 11 extremely worrisome cases have issues related to unclarity about responsibilities, the absence of overviews of medications, and carelessly applied restraints (Kuiken, 2016). Clearly, there are issues which have come up since the decentralization of elderly care. This research will amplify these issues, and it will contrast it with those aspects that were brought forth after the decentralization of elderly care in Sweden.

The Netherlands is not the only country where decentralization of healthcare issues to municipalities or to counties was on the agenda, or where it led to difficulties. The last thirty years, decentralization policies have gained domain, and they have been implemented throughout the developing world on a large scale (Mitchell, & Bossert, 2010; Madon, Krishna, & Michael, 2010). According to Rondinelli and Cheema (2007) these policies have included political, fiscal and administrative decentralization. Political decentralization aims at providing greater policy-making power to citizens and representatives, who have been elected locally. Fiscal decentralization renders greater authority, meaning the use and the collection of revenues which can be spent at the local level, to local jurisdictions. Lastly, administrative decentralization expands the role the local level plays in delivering public services (Rondinelli, & Cheema, 2007). Over the years, external as well as internal pressures, in both

(7)

7

low- and middle-income countries, have blended these three types of decentralization policies into ordinary practices (Mitchell et al., 2010).

The gaining popularity of decentralization is based on the logic behind this concept.

According to Saltman, Bankauskaite, and Vrangbaek (2007) larger organizations are less accountable and agile than the smaller organizations, as long as these smaller organizations are properly organized and guided. Therefore, there is a great attraction in the possibility of establishing institutions which operate locally and are also locally responsible (Saltman et al., 2007). Decentralization is believed to stimulate the improvement of how services are delivered, but also to better assign resources to the population’s needs, to involve the community in priority setting, and to aid in reducing inequities in health (Østergren, Boni, &

Kaarbøe, 2007). Several studies have been conducted in developing countries, and these indicate that particular indicators of health were improved after decentralization, such as the mortality rate and the infant mortality rate (Montero-Granados, de Dios Jiménez, & Martín, 2007). However, Yasar (2011) found with his assessment of the Health Transformation Programme (HTP), which was announced in 2003 in Turkey, and which was partly based on decentralization, that the ultimate goals set by the government were far from realized. These goals consisted of improvement in health status, protection against financial risks, and satisfaction with healthcare. The study did find some improvements in health system performance with the implementation of HTP, however, the ultimate goals were not reached (Yasar, 2011). Another study conducted by Langran (2011) in the Philippines, found that problems like inequity, financial risks, and a lack of participation, were still encountered.

Different studies, conducted in different countries, show very controversial results about the influence of decentralization, whereby the expected outcomes of decentralization and the actual outcomes do not coincide. One of the explanations for these striking results could be that there is a substantial difference between countries, which could lead to different outcomes. Many countries have undergone healthcare reforms, however not much comparative research between countries has been done (Esteves, 2012). According to Esteves (2012) comparative research is important to shed light on shady corners related to healthcare reforms.

(8)

8

This research will focus on the differences between countries, the way they cope with decentralizing elderly care to municipalities, and what influence the degree of decentralization has on elderly care.

As described by Hendriks, Spreeuwenberg, Rademakers and Delnoij (2009) and Schäfer et al. (2010), the Dutch healthcare system has been replaced by a more managed competition in January 2006. As part of this reform, Dutch residents are obliged to have a basic health insurance. Furthermore, rejecting citizens for a basic healthcare package is not allowed. Besides, it is not permitted to adapt the price of healthcare packages based on favourable risks, or to make a distinction according to risk. Health providers and health plans negotiate with each other on price, how care is organized, and what the basic package entails.

Health plans do not have to set up a contract with every health provider. Moreover, collective arrangements can be offered to the insured for a reduced premium. The aim with these reforms is to make it easier to switch health plans, and it is supposed to result in a healthcare system that is more demand-driven, less expensive, and of a better quality (Hendriks et al., 2009; Schäfer et al., 2010). Furthermore, according to Schäfer et al. (2010), with this reform the government has changed its role of directly steering the healthcare system to being a safeguard of the process. And with it, the responsibilities are transferred to insurers, providers and patients. The devolution of the responsibility of home care services to municipalities has increased the diversity of care arrangements. Besides, the government still acts as a watchdog for quality, accessibility and affordability of healthcare services, to prevent undesired effects of this new ‘managed competition’ system. Nowadays, private healthcare providers are the ones who are primarily responsible for providing healthcare services. Healthcare can be subdivided into four categories. Public health services are mainly providing preventive care.

Prevention, promotion and protection are all part of the responsibilities of municipalities. In total, 29 municipal services carry these tasks out for all 443 municipalities. Primary care is mainly provided by GPs (general practitioners), midwives, pharmacists, physiotherapists and psychologists. The GP acts as a gatekeeper, because hospital care and specialist care (except in case of emergencies) are only accessible with a referral from a GP. Secondary care entails the care that is only accessible with a referral from a GP. It includes hospital care and mental

(9)

9

care. The last type of care entails long-term care. This is mostly provided by home care organizations, and residential and nursing homes.

The Dutch health insurance system can be subdivided into three partitions. The first entails the insurance every inhabitant of the Netherlands is obliged to have for long-term care.

This insurance provides continuous care for the chronically ill under the Exceptional Medical Expenses Act (Algemene Wet Bijzondere Ziektekosten, AWBZ). This act is based on income-dependent contributions. Before being able to access this, an assessment takes place, and the provision of care goes via care offices, which are independent entities, but closely aligned with health insurers. The second partition consists of basic health insurance, which is regulated by the Health Insurance Act (Zorgverzekeringswet, Zvw). Inhabitants of the Netherlands contribute to this partition via two ways. Firstly, they pay a premium to their chosen health insurer. And secondly, an income-dependent fee is contributed via employment, by deducting part of their payroll. These contributions come together in a fund which is allocated among health insurers based on a risk-adjustment system. The last partition is a complementary voluntary health insurance. These can include services that are not covered under the Acts of the first two partitions. Furthermore, prevention and social support are financed through general taxation. And lastly, every insured has an own risk of minimum

€385, varying up to €885 (van de Ven et al., 2013; Rijksoverheid, 2016).

Since the reforms of 2006, the payment of healthcare providers has changed. GPs now have a combination of fee-for-service and capitation fees. Furthermore, hospitals and mental care institutions now work with Diagnosis and Treatment Combinations (DTCs), which is an elaborate diagnosis-related groups-type system. Besides, the intensity of care needed for a patient determines the payment for long-term carers (Schäfer et al., 2010).

Moreover, recent reforms have decentralized the responsibilities of caring for elderly and people with disabilities, to municipalities (Dijkhoff, 2014; Rijksoverheid, 2014).

According to Anell, Glenngȧrd, and Merkur (2012) this decentralization of elderly care to municipalities can also be found in Sweden. In the Health and Medical Services Act of 1982 it is specified that the county councils/regions and municipalities are responsible for providing access to good healthcare for everyone, which includes taking care of funding. The main responsibility for delivering healthcare services can therefore be found at the level of the

(10)

10

national government and the 21 county councils/regions, while the 290 municipalities are mainly responsible for care of elderly and disabled people (Anell et al., 2012; OECD, 2013).

The decentralization of elderly care to municipalities happened in 1992, while the care for physically disabled people followed a few years later, as well as the care for the long-term mentally ill. The decentralization of elderly care unintentionally co-occurred with an economic recession, which led to additional difficulties regarding financing the public services. This occurrence resulted in rapidly decentralized elderly care (Johansson, 1997;

Anell et al., 2012; European-Observatory, 2016).

According to Anell et al. (2012), there are three basic principles which apply to healthcare in Sweden. The first entails human dignity, which means that all inhabitants of Sweden are entitled to equal healthcare, regardless of their status. The second one involves the principle of need and solidarity, which entails that patients in greatest need are prioritized in medical care. Lastly, the principle of cost-effectiveness indicates that the relation between costs and effectiveness should be reasonable regarding improvement of health and quality of life.

The financing of the healthcare system mainly comes from central and local taxes.

Both county councils as well as municipalities have the right to generate income-related taxes.

The central government is responsible for providing subsidies for prescription drugs. The government further provides financial support to lower levels through grants to smooth out geographical inequalities. Local income taxes account for 70% of the budgets of county councils and municipalities. The remainder is supplemented by the government grants and user charges. Only a small part of the population has an extra health insurance, about 3 to 4 percent and in most cases these insurances are paid for by employers. This private insurance provides faster access to care (Anell, 2008; Anell et al., 2012).

The payment of healthcare providers differs between county councils. However, payments based on global budgets or a mixture of those, case-based and performance-based are very common in hospitals. While payments for primary care providers are mostly based on capitation for registered patients, this will be complemented by fee-for-service and performance-based payments. Furthermore, county councils are responsible for paying the

(11)

11

full costs for inpatient drugs. And for prescription drugs, the county councils receive grants from the government (Anell et al., 2012).

In Sweden, the government still acts as a watchdog. If county councils and municipalities do not use their budgets correctly, the government can lower the grants as a financial penalty. Moreover, most healthcare providers are both owned as well as operated by the county councils and municipalities. Besides, most private providers still work with county councils and most are bound to them with a contract. In this way, the costs can be controlled at a local level (Anell, 2008).

Even though these healthcare systems differ between the countries, they do show a few similarities, and both systems use decentralization of elderly care to municipalities. Both countries have focused their attention mainly on elderly care (Schäfer et al., 2010; Anell et al., 2012). Therefore, this research will focus on the decentralization of care for elderly. The healthcare systems of both the Netherlands and Sweden are quite comparable, which leads to the following research question:

Does the quality of elderly care improve in the Netherlands and Sweden after devolving healthcare services for elderly to municipalities?

The next section provides a theoretical background based on the available literature.

Afterwards, the methodology will be explained, and subsequently the findings will be discussed in the results section. This paper will present a short summary and the limitations of this study in the conclusion and discussion part. Furthermore, theoretical and practical implications will be provided and this paper will conclude with propositions for future research.

(12)

12

2. THEORETICAL CONSIDERATIONS

Decentralization can be seen as a socio-political process whereby responsibilities and central authorities are moved or redistributed to local governments in the belief that they will be better in meeting the population’s needs and interests (Regmi, Naidoo, Pilkington, &

Greer, 2009; Regmi, 2014). According to Mills, Vaughan, Smith and Tabibzadeh (1990) there are different types of decentralization, consisting of four main types: deconcentration, devolution, delegation, and privatization. Deconcentration entails handing over administrative authority to offices of central government ministries, which are locally-based. Devolution is the strengthening or formation of local authorities, known as local government or subnational levels of government, which have their own defined set of functions, and are considerably independent of the national level. Delegation is mostly used in the management of teaching hospitals. It consists of transferring managerial responsibilities to organizations that are not part of the central government structure. These organizations are only indirectly controlled by the central government. Finally, when voluntary organizations or private profit-making, or non-profit-making enterprises receive government functions, then it is called privatization. In this case, there is still some degree of government regulation (Mills et al., 1990).

In this paper, decentralization is defined as: “a rather dramatic devolution, whereby responsibilities for funding, as well as quality control in the health sector are delegated from the central government to district authorities and private institutions” (Kristiansen, & Santoso, 2006: 248). The transfer of authority and responsibilities to municipalities or counties needs to be an interplay between fiscal, political and economic aspects to make it a successful transition (Regmi, 2014). However, this interplay is not always present. Therefore, even though many countries engage in devolving healthcare services to lower levels, there are very contradictory outcomes measured about the effect of devolution reforms on elderly care (Montero-Granados et al., 2007; Yasar, 2011; Langran, 2011). These contradicting outcomes can be the result of countries engaging in different degrees of devolution (Hyde, 2016).

2.1 Degree of devolution reforms

Countries implement devolution in an attempt to allow patients more autonomy to decide where, by whom, and what kind of treatment they want to receive (Peckham,

(13)

13

Exworthy, Powell, & Greener, 2008). Such involvement is seen as the right of any patient, while also establishing greater satisfaction among patients, because they receive healthcare services that better correspond with their needs. Furthermore, it has the potential of improving healthcare services due to managed competition between healthcare providers and by augmenting the quantity of services that are offered (Peckham et al., 2008).

According to Costa-Font and Rico (2006) and Dao and Lodenstein (2011) when healthcare services are centralized, there is a lack of accountability and a lack of perceiving users as customers, thus leading to little incentives for improving the quantity and quality of care. Devolving healthcare services to municipalities is the perfect form of downward accountability. Besides, it is a big motivator for improving the quality of service delivery (Costa-Font, & Rico, 2006; Dao, & Lodenstein, 2011). Furthermore, it is important to have a proper workforce in the right place for improving the quality of care, which entails that policies become very relevant to make this coincide (Maslin-Prothero, Masterson, & Jones, 2008). Therefore, if the right degree of devolution is chosen, and the municipalities receive enough autonomy to form new policies, it is assumed that it will have a positive effect on the quality of elderly care.

Quality of care is obtained when the health services that are offered increase the chances of desired results for patients, and which are in accordance with the present-day professional knowledge (Institute of Medicine, 2001).

According to Maarse and Jeurissen (2016) the reform that took place in 2015 is one of the largest overhauls of the healthcare sector in the Netherlands. The focus of the devolution is put on individual responsibility, restructuring the financial side of the healthcare sector, the involvement of both healthcare providers and municipalities, and cutting expenditures. The municipalities are required to draw up policies, which can cause inequalities in the access to long-term care between different districts. Furthermore, the Dutch healthcare culture can be described as rather egalitarian, which leads to equal access, however this system can also make it more difficult to receive extra care if necessary. Besides, the devolution was implemented in a short period of time, which led to many uncertainties and risks, mainly for patients (Maarse, & Jeurissen, 2016). However, according to Janssen, Jongen and Schröder- Bäck (2016) long-term care reforms contributed to person-centred care delivery, since patients are entitled to the freedom and autonomy to choose healthcare services of their preference. Moreover, new initiatives are arising to further ameliorate the quality of elderly

(14)

14

care (Janssen et al., 2016). The quality of care is an important issue on the agenda, and a fairly high percentage, namely 69.5% of the Dutch population, perceives elderly care as good (Alders, Costa-Font, de Klerk, & Frank, 2015).

The devolution in Sweden was a much longer process than in the Netherlands. It took ten years for the implementation of the devolution to happen (Johansson, 1997; Anell et al., 2012). According to Fotaki and Boyd (2005) the focus of the reforms was mainly put on user autonomy, whereas equal access or equity of opportunity were emphasized less, and market mechanisms became more prominent for delivering services.

Devolving elderly care was economically advantageous in the short term, however it turned out to be rather expensive in the long run. Furthermore, the interests of elderly were emphasized less than some other groups, which led to negligence of investing in elderly healthcare services. Moreover, due to cutting costs and limiting patients’ choice of healthcare providers the quality did not improve (Fotaki, & Boyd, 2005).

In general, according to Johansson (1997), the users of elderly healthcare services are quite positive, and the quality of care is rated very high. However, there have also been indications of the opposite. In a study conducted nation-wide, 18 percent of the elderly were not receiving services which met their needs, while this was only 6 percent in a similar study carried out in 1988 (Johansson, 1997).

Therefore, when the emphasis is placed on improving the quality of care, and giving autonomy to patients to choose their own provider, the quality of the healthcare services will improve. Sweden has placed much less emphasis on the autonomy of patients, and less emphasis on elderly care in general. In the Netherlands, however, more emphasis was placed on improving the quality of care, and it became one of the highest priorities. Furthermore, the devolution process in Sweden has been a long ongoing process, and the level of decentralization kept on changing over the years. Sweden thereby, lost control over the way they wanted to improve the quality of care. In the Netherlands, a higher degree of devolution can be detected by the higher amount of autonomy and the higher financial power the municipalities received. Therefore, the following hypothesis can be formed:

H1: the higher degree of devolution reforms in the Netherlands in comparison with Sweden, will have a larger positive effect on the quality of elderly care.

(15)

15

When a national healthcare plan is still in place, a strong tendency towards equity and pressures for uniformity occur (Peckham et al., 2008). However, with devolving healthcare services, diversity arises which could lead to divergent local services (Peckham et al., 2008).

Devolution of healthcare services changes the construction of the status quo by allowing alternations to organizational forms and the way the services are delivered (López- Casasanovas, 2007). Devolution therefore encourages municipalities to take initiative in providing healthcare, while also stimulating innovation. This in turn, leads to more diversified healthcare service provisions (López-Casasanovas, 2007; Costa-Font, 2010). Political devolution increases the probability of diversified healthcare services, due to the ability to experiment with policies (Exworthy, 2001; Peckham et al., 2012).

Elderly care has changed over time due to the ageing of the population (Trahan, &

Caris, 2002). This ageing leads to specific long-term care needs for elderly. Difficulties in the ability to meet the needs of this part of the population is partly due to changes in the type of healthcare services required. Devolution enables municipalities to manage these changes, and to adapt the services to suit the elderly demands (Trahan, & Caris, 2002).

Diversity of care arrangements are diversified healthcare services that are utilized to meet the needs of local elderly. The diversified services also fill the gaps in service provision that existed before the care was devolved to municipalities (Martin, Peet, Hewitt, & Parker, 2004).

To adapt services to the needs and requirements of elderly, municipalities need a certain amount of autonomy. In the Netherlands, municipalities receive a certain degree of freedom in the way they deliver their healthcare services. However, there are minimum requirements set by law (Schäfer et al., 2010). Furthermore, the long-term care adds up to a large share of the total healthcare system, and the costs take up 38% of the total healthcare budget (Schäfer et al., 2010). This entails that long-term care is very important in the Netherlands. Moreover, especially elderly care is eminent on the agenda of municipalities, due to the longer life expectancy of the population, and the ageing (CBS, 2014). Attributable to this, Dutch municipalities focus on providing appropriate care to elderly, leading to diversified healthcare services since the devolution (Schäfer et al., 2010; Worrall, &

Chaussalet, 2015). Additionally, according to Smith and Häkkinen (2007), the Netherlands did not concede for the top-down approach most countries use when implementing a new healthcare system. An alternative approach, the bottom-up approach, was used to convey the

(16)

16

ability to municipalities for quality improvement initiatives. The top-down approach entails the development of a framework in which the design of the national information system is communicated via a centralized approach to support decentralization. The bottom-up approach however, relies on healthcare organizations and professionals to engage in healthcare improvement initiatives (Smith, & Häkkinen, 2007). This bottom-up approach allows municipalities more autonomy to diversify the healthcare services. According to Vorst, de Vries, and Gussekloo (2011) this is also the only way to implement a new healthcare system. They propose that healthcare is too complex for a top-down approach. They argue that it will lead to long waiting lists and higher controlling costs. Moreover, they advocate that self-steering regional networks will be best at implementing optimal healthcare services (Vorst et al., 2011).

In Sweden, the Health and Medical Services Act of 1982 was designed to administer a considerable amount of autonomy to municipalities to organize their health services (Anell et al., 2012). The reforms in 1992 also yielded an increase in diversity and volume of healthcare services (Anell, 2005). Sweden used the same bottom-up approach as the Netherlands (Smith,

& Häkkinen, 2007), which also allowed Swedish municipalities more autonomy to diversify healthcare services. However, the increase in diversified services was soon levelled out by the traditional hierarchical management, also known as centralization (Arnell, 2005). The past decades, the Swedish devolution has been fluctuating. The traditional hierarchical management was later re-introduced by developing regional policies instead of local policies (Anell, 2005).

When a higher degree of devolution is reached, and municipalities receive an appropriate amount of autonomy, it will have a positive effect on the diversity of healthcare services. Attributable to the traditional hierarchy that levelled out the diversity, it can be said that Sweden has a lower degree of devolution reforms and that it led to less diversified healthcare services. This leads to the following hypothesis:

H2: the higher degree of devolution reforms in the Netherlands will have a larger positive effect on the diversity of healthcare services than in Sweden.

2.2 Diversity of care arrangements

The study conducted by Regmi, Naidoo, Greer and Pilkington (2010) suggests that devolution leads to an increase in access to healthcare services. Furthermore, it augments the

(17)

17

utilization of services, and lastly, the delivery of the services show an improvement (López- Casasanovas, 2007; Regmi et al., 2010). As described earlier in this paper, devolution stimulates diversity by innovating healthcare service provisions. Diversified and innovated services are better able to meet the needs of the population (Trahan, & Caris, 2002).

Moreover, due to increased competition between healthcare providers and the freedom of the population to choose their own provider, diversity of services of a high quality is required (Peckham et al., 2008). Furthermore, diversity can improve the efficiency of healthcare services by providing a greater range of services, while better meeting the needs of the population and improving the quality of care (Akin, Hutchinson, & Strumpf, 2005;

Masanyiwa, Niehof, & Termeer, 2015). In a research conducted in Bolivia by Faguet (2004), the outcomes show that municipalities invest more funds into services that have received the highest priority. The services were better adjusted to the population’s needs, and these investments led to the provision of diversified services which in turn led to a better quality of care.

In the Netherlands, municipalities received enough autonomy during the devolution to adapt their services to the needs of elderly. Partly due to the autonomy, and partly due to increased competition between providers, there is an expansion in care arrangements (Schäfer et al., 2010). Furthermore, there is a focus on making the services for long-term care client- tailored (Maarse, & Jeurissen, 2016). This entails diversifying services, and improving the quality of the offered services.

Not many Swedish municipalities have opened up to increase the competition among healthcare providers (OECD, 2005). Furthermore, in Sweden the municipalities have received less autonomy compared to the Netherlands (Anell et al., 2012), and the received autonomy was levelled out at later times (Anell, 2005). Therefore, there has not been an enormous increase in the diversity of healthcare services (Anell, 2005). Moreover, municipalities are not allowed to employ doctors, leading at times to the inability to provide appropriate medical care (OECD, 2005).

Due to the focus the Netherlands have on elderly care, diversifying services, and client-tailored long-term care, it is proposed that the greater diversity of services in the Netherlands will improve the quality of care more, leading to the following hypothesis:

H3: the diversity of healthcare services in the Netherlands is greater than in Sweden, leading to a larger positive effect on the quality of elderly care.

(18)

18

2.3 Quality of the implementation process

Østergren et al. (2007) have conducted a research on the implementation of devolution in Norway, the Russian Federation, and Italy. They classified the reforms regarding three aspects. The first was if the reform was comprehensive or narrow. With this they wanted to determine whether the whole healthcare sector was affected by the devolution, or only small parts. The second aspect was if the reform was radical or cautious. Hereby they classified to what extent the reform broke with the past. The final one looks at the way the blueprints are defined, or if there is room for interpretation. They found ongoing devolution in Norway, even though the government itself was pushing for a radical change. The implementation process is going slow, and they experienced a few problems. However, these were minor in comparison with the Russian Federation. Here the implementation process was radical, and this led to a failing devolution. Italy endured difficulties as well, however, they found that the devolution went better when the local government was stronger (Østergren et al., 2007).

Based on their findings, it can be said that devolution often comes with complications.

The devolution of healthcare services to municipalities was announced in 2014 in the Netherlands. The reform itself took place in 2015 (Rijksoverheid, 2014). In an opinion piece in De Correspondent, Schepers (2014) wrote that municipalities only had six months to prepare themselves for the devolution of elderly care. According to Dijkhoff (2014), with these types of devolutions, municipalities need several years before they are able to manage, and turn into practice, their social responsibilities in the way intended. Due to the little amount of time municipalities were allowed to prepare for the responsibility of elderly care, this reform could be seen as rather radical. Sweden on the other hand, has followed a more cautious approach with devolving bit by bit (European Observatory, 2016). Furthermore, the Swedish devolution of care was already announced in 1982, while it took place in 1992 (Anell et al., 2012). The devolution in Sweden can therefore be qualified as rather incremental.

According to Schäfer et al. (2010) and Anell et al. (2012), the devolution in both countries only affected small parts of the healthcare system. Furthermore, in the Netherlands, there has been room left for interpretation, so municipalities could adjust the blueprints into local policies, while Sweden is familiar with a stricter use of blueprints (European Observatory, 2016). Receiving autonomy to adjust blueprints, is important to increase the

(19)

19

effectiveness of policies (Greer, 2007). Hoque, Davis and Humphreys (2004) agree with this.

They describe that autonomy is desirable due to the positive effect on the municipalities’

ability to deliver healthcare services superior to the ones offered before the devolution.

It is hypothesized that a higher degree of devolution reforms will have a positive effect on the quality of elderly care. The quality of the implementation process exists of the received autonomy, if the process was implemented quickly, and which part of the healthcare sector was affected. According to Dijkhoff (2014) an adequate amount of time should be available to reach municipalities’ goals and for the implementation process to be successful. Therefore, a rather radical devolution will affect the quality of care. Furthermore, the received amount of autonomy will influence the degree of devolution, and with it the quality of care (Greer, 2007;

Hoque et al., 2004). Based on these findings, the following hypothesis is formed:

H4: the quality of the implementation process will act as a moderator on the relation between the devolution reforms and the quality of elderly care.

All proposed relations are visually represented in Figure 1. It has been suggested that the degree of the devolution reforms and the diversity of care arrangements will have a positive effect on the quality of elderly care. The degree of devolution reforms will also have a positive effect on the diversity of healthcare services. Lastly, it is suggested that the quality of the implementation process has a moderating effect on the relation between the degree of devolution reforms and the quality of elderly care.

Figure 1

(20)

20

3. METHODOLOGY

The goal of this research was trying to gain insights into the effect of devolution reforms of healthcare services for elderly to municipalities, and the influence on the quality of care. It was hypothesized that the quality of care is influenced by the degree of the devolution reforms. Furthermore, it was theorized that the diversity of care arrangements acts as a mediator between the degree of devolution reforms and the quality of elderly care. Lastly, it was proposed that the quality of the implementation process has a moderating effect on the relation between the degree of devolution reforms and the quality of elderly care. This following section describes the data collection, the measurements, and the method of analysis.

3.1 Data collection

The aim of this research was to find an answer to a ‘does’ question, and therefore a quantitative research approach is appropriate (Saunders, Lewis, & Thornhill, 2009).

According to Creswell (2014) quantitative research examines relationships between variables and thereby tests objective theories. The variables are mostly measured using instruments, which results in data that can be statistically analysed. Quantitative research is a form of deducting, and also of being aware of biases, controlling for other explanations, and being able to replicate as well as generalize the outcomes (Creswell, 2014). Furthermore, quantitative methods are directed at the number or amount of qualities that can be researched with qualitative methods (van Aken, Berends, & van der Bij, 2012). During this research, attention has been paid to aspects suchlike controllability, validity and reliability. According to van Aken et al. (2012) to reach an inter-subjective agreement on the outcomes of the research, the first requirement is controllability. It is a prerequisite for the validity and reliability of a research. Controllability entails giving a detailed explanation of the execution of the research, so the study can be repeated by others to check for the same outcomes.

Reliability has been reached when a particular test or tool, for example a questionnaire, produces similar outcomes in a different situation, while assuming other conditions did not change (Roberts, Priest, & Traynor, 2006; Heale, & Twycross, 2015). Validity in a quantitative study is subdivided into three categories. The first is content validity, which entails if the items measure the content they are intended to measure. The second is predictive

(21)

21

validity, which answers the question if the results correlate with other outcomes. The last one is construct validity to see if items measure hypothetical concepts or constructs (Creswell, 2012).

For this research, different types of data have been gathered. First, the available literature on outcomes of devolution of healthcare services has been studied. Special attention has been paid to research conducted in the Netherlands and Sweden by the European Observatory and the European Healthcare Management Association to gain more insight into the workings of the healthcare systems of both countries.

Moreover, to research the proposed relations between the factors, a survey has been conducted. The statements in the survey were based on existing literature, and the target group was asked to fill in their level of agreement with the statements on a five-point Likert- scale.

The online questionnaire has been send to homes for elderly in the Netherlands and in Sweden. Homes for elderly were the target group of this research, since they see the indirect effects of the devolution in the care for the elderly. The nursing staff and the management staff from elderly homes in both the Netherlands and Sweden were asked to fill in the survey.

The survey was set up in English (Appendix 3), and first tested among five people to make sure that the questionnaire was understandable. Hereafter however, it was translated to Dutch (Appendix 4) and Swedish (Appendix 5) to reach as many people as possible.

Furthermore, the survey has been distributed via social media, namely Facebook and LinkedIn. Various Facebook groups of nursing homes in both the Netherlands and Sweden were targeted with the request to post the survey in their group. Moreover, several institutions were contacted for possible tips and/or contacts. Most institutions unfortunately were not able to help further distributing the survey.

In the Netherlands, 175 nursing homes for elderly were approached via email, and another 175 in Sweden. They were kindly asked to distribute the survey to staff members.

This was done in an attempt to increase the number of respondents. With a population size, which entails all management and nursing staff in homes for elderly, larger than 10.000 and an alpha of 0.10, then according to the table as proposed by Bartlett, Kotrlik, and Higgins (2001) a sample size of 83 was desirable. It was strived for to reach this sample size in both

(22)

22

the Netherlands and Sweden to get a representative sample of the target group. The survey ran for three weeks, between mid-November and early December, and after one week a reminder was sent to the 175 nursing homes in both the Netherlands and Sweden.

3.2 Measurements

To be able to conduct the research, the concepts first had to be operationalized. First, the devolution of healthcare services has been defined, followed by the operationalization of degree of the devolution reforms. Furthermore, the measurements of these concepts were explained as well.

3.2.1 Devolution of healthcare services

Cascón-Pereira, Valverde and Ryan (2006) researched papers of different authors for an indication of devolution, and they concluded that devolution can be seen as a multifaceted phenomenon. These authors identified the following dimensions that devolution may incorporate: tasks/responsibilities, decision-making power, financial power, and expertise power. According to this research, the degree of devolution is higher when a higher number of dimensions are devolved. The lowest level of devolution is considered to be the transfer of tasks without decision-making power, and the transfer of expertise without financial power (Cascón-Pereira et al., 2006).

The devolution of tasks/responsibilities will be measured by the reallocation of tasks or activities that were formerly undertaken by the central government (Brewster, & Holt Larsen, 1992).

Finkelstein (1992: 506) defined power as “the capacity of individual actors to exert their will”. He conceptualized decision-making power into four types: structural power, which encompasses the formal position someone holds within the organization, such as the level, titles and compensation; ownership power entails shareholdings and the relationship with the top management team; furthermore, expert power involves the contacts and relationships that make others turn to someone for their advice; and lastly, prestige power is the power someone holds with their status, powerful friends or a certain background (Finkelstein, 1992). In this research, power has been defined as the capacity municipalities received to exert their will.

(23)

23

This entailed the power that is received from the government, based on the four types of power as described by Finkelstein (1992).

Financial power is the devolution of a budget, which is accompanied by enhancing responsibilities and functions on a lower governmental level. It also includes greater autonomy in financial decision-making and budget making (Ghaus-Pasha, Pasha, & Khan, 2000; Cascón-Pereira et al., 2006). Research will be conducted to gain insights into the financial power that municipalities received after the devolution.

With the devolution of expertise power, not only tasks are devolved, but skills and knowledge to execute these tasks as well (Cascón-Pereira et al., 2006). It has been researched how much knowledge and skills were transferred to municipalities after the devolution in both countries.

3.2.2 Quality of elderly care

The Institute of Medicine (2001: 44) has defined quality of healthcare as “the degree to which health services for individuals and populations increase the likelihood of desired health outcomes and are consistent with current professional knowledge”. Stelfox and Straus (2013) follow the definition of healthcare quality indicators (QIs) as described by the National Library of Medicine, which states that QIs are direct measures, both qualitative and quantitative, like norms, standards, and criteria, which are used in the determination of quality of care. Janssen et al. (2016) adopted three principles in their study to measure the quality of long-term care. These three principles consist of affordability, availability and person- centeredness, which are distinguished by the European Quality Framework for Long-Term Care Services to indicate what quality in long-term care consists of (European Partnership for the Wellbeing and Dignity of Older People, 2012). The quality of elderly care was therefore covered in the questionnaire using questions regarding the norms, standards, criteria, availability, affordability and person-centeredness according to the professional expertise of the target group.

(24)

24

3.2.3 Diversity of care arrangements

As described earlier in this paper, diversity of care arrangements entails the utilization of healthcare services to meet the needs of elderly, and filling the gap in the provision of services before elderly care was devolved to municipalities (Martin et al., 2004). In this study, therefore, the diversity of care arrangements has been measured in the survey according to the perceived increase in diversified care arrangements for elderly according to the target group.

Furthermore, the amount of resources elderly homes have at their disposal was also covered, as well as questions regarding care being adapted to one’s unique personal needs.

3.2.4 Quality of the implementation process

The implementation process as described by Østergren et al. (2007) looks at the devolution in terms of which part of the healthcare sector is affected, how radical the reform is, and how much freedom is received to adjust blueprints. In this study, the quality of the implementation process has been measured according to the following principles: firstly, it was considered how big a part of the healthcare sector was affected by the reform.

Furthermore, the target group has been asked to give an indication of how radical the reform was in their opinion. And lastly, the autonomy received by the municipality to adjust blueprints has been taken into account.

3.3 Method of analysis

The survey has been distributed via an online tool called Qualtrics. This tool ensured that the responses were saved, and it provided an overview of the responses. The responses were thereafter entered into a statistical program named SPSS. This software aids in analysing large amounts of data, and it is therefore very suitable for analysing quantitative data. Via the program, statistical tests have been conducted to test the hypotheses.

(25)

25

4. RESULTS

In this section, the responses of the questionnaire were analysed with the help of the statistical program SPSS. Based on the outcomes, the hypotheses were either supported or not.

4.1 Socio-demographic results

In total 125 people started the online questionnaire, of which 69 finished the survey.

Of these 69 responses, 64 were Dutch, and five were Swedish. In total 74 responses have been recorded. The other five were received by physically going to nursing homes in the Netherlands. In total 55 questionnaires were physically distributed, of which five were filled out. The largest part of the respondents was reached via email, namely 45, while 24 were reached via social media, and five by physically going by with hard copies of the survey.

In the Netherlands 8.7% of the nurses are male and 91.3% is female (Werknemers in de Zorg, 2016). Of the participants in the Netherlands, 21.7% was male and 78.3% was female. In Sweden 9% of the nurses are male and 91% is female (Persson & Wallin, 2004).

Of the participants in Sweden, 20% was male and 80% was female.

The mean of the age categories was 2.88 (SD = 0.832), which means that most respondents in the Netherlands fell into the age category of 41 – 55 years. The same goes for Sweden with a mean of 3.00 (SD = 0.707).

The distribution of inhabitants over the provinces in the Netherlands is shown in a pie chart in Appendix 1 (Klein, 2016). As can be seen by comparing the pie charts, the distribution of the respondents is reasonably representative. The same is done for Sweden (Statistics Sweden, 2016), however, here the distribution is not representative. The respondents are residents in three out of 21 regions (Iän), namely Stockholms Iän (40 percent), Skåne Iän (20 percent), and Östergötlands Iän (40 percent). The representation of this is also shown in pie charts in Appendix 1.

Moreover, 2.9% of the respondents in the Netherlands had VMBO as the highest degree received. While 5.8% had HAVO as the highest degree. Most of them either had MBO as the highest degree (34.8%) or HBO (43.5%). The last 13% had finished a Master. In Sweden 80% had a degree from a university of applied sciences, which in Swedish is known as Högskola, and which is equivalent to the Dutch HBO. The other 20% had finished a Master as well.

(26)

26

Most respondents were employed, namely in the Netherlands it was 94.3%. Of the other respondents, 4.3% was a student and 1.4% was self-employed. In Sweden, all respondents were employed.

4.2 Assumptions

The first step taken was to combine the data from Sweden and the Netherlands into one database. This was done to reach a general conclusion on the hypotheses. After a general conclusion was obtained, the same tests were conducted for both countries separately, to gain insights into the differences between the countries.

The Netherlands and Sweden combined

After merging the data into one dataset consisting of 74 responses, the dataset was checked for negatively stated statements. A few questions had to be converted to positively stated statements, in accordance with the other statements. Hereafter, the correlation between variables was examined and sum variables were made from the ones that correlated with each other.

As described by Bartholomew, Steele, Moustaki and Galbraith (2008), and Symeonaki, Michalopoulou and Kazani (2015) many consider the type of data as obtained with the Likert-scale as ordinal data. However, for Likert scales where at least a five-point scale is used, the ordinal categories can also be understood as interval data. Therefore, one would be able to perform statistical tests with these pseudo-interval variables (Bartholomew et al., 2008; Symeonaki et al., 2015). Because the type of data that was obtained during this research is a five-point Likert-scale, the data will be treated as interval data (Allen, &

Seaman, 2007; Brown, 2011), and therefore the correlation between variables was examined using a Pearson Correlation Coefficient test.

Based on the outcomes of the Pearson test, some questions had to be taken out due to non-significant correlations. After leaving out some questions, four sum variables were computed for the quality of the implementation process, the degree of the devolution reforms, the quality of elderly care, and for the diversity of care arrangements.

After computing the sum variables, the internal consistency was tested using Cronbach’s alpha. According to Gliem and Gliem (2003), internal consistency is excellent when it is higher than 0.9, it is good when it exceeds 0.8, and it is still acceptable when it is

(27)

27

0.7 or higher. When Cronbach’s alpha is 0.6 or higher, the internal consistency is questionable, and everything lower than 0.6 is poor or unacceptable. In table 1 the Cronbach’s alphas for the sum variables are shown. As can be seen in the table, the diversity of care arrangements has the weakest alpha with 0.692, which is considered to be questionable. The rest of the sum variables all have an alpha that exceeds the 0.7, which makes all acceptable.

Table 1: Cronbach’s alpha

The Netherlands

The same steps have been followed for the dataset with the responses from the Netherlands, which consists of 69 responses. The next step was to view the correlations with the Pearson test between the questions and to form sum variables. With the dataset of the Netherlands, the newly made sum variables consisted of exactly the same questions as the sum variables of the combined dataset of the Netherlands and Sweden. However, the Cronbach’s alphas, as can be seen in table 1, do differ slightly from the combined dataset, due to a slightly different set of responses.

Sweden

With the last dataset, the same approach was used as before. The dataset of the responses from Sweden consisted of five responses. Here the Pearson Correlation Coefficient test was used as well. Due to the smaller sample size, the correlations between the questions were different than from those of the dataset of the Netherlands or the combined dataset. Less

Sum variables

The Netherlands

and Sweden The Netherlands Sweden

Quality of elderly care 0.870 0.865 0.909

Quality of the

implementation process

0.887 0.874 0.889

Diversity of care arrangements

0.692 0.719 0.909

Degree of devolution reforms

0.718 0.730 0.889

Referenties

GERELATEERDE DOCUMENTEN

Denk eens aan andere uitdagingen voor kinderen. Een glijbaan of schommel kan je maar op enkele manieren gebruiken. De grootste uit- dagingen voor kinderen zijn die met de

implementation plan; plan for assessment of implementation; study design for measuring impact of intervention; explains how method was applied; internal and external validity)

The Netherlands has a universal mandatory social health insurance scheme, which covers a broad range of long-term care services provided in a variety of care settings.. 6 |

The present text seems strongly to indicate the territorial restoration of the nation (cf. It will be greatly enlarged and permanently settled. However, we must

The vitamin C MUPS capsule formulations, with AVG and SLS as functional excipients, showed an improvement in transport across the intestinal epithelium compared

In this overview of the nature of the contemporary effective school principalship,, elements of wide-ranging diversity have been identified. The role of a principal is found to

The effect of column height on the bubble properties, such as bubble velocity, local void fraction, interfacial area and equivalent diameter, will now be

(2014) found that internal initiatives are less successful with respect to autonomy in large developing economies compared to small and remote developed economies.. This