Co-payments for general practitioners in Denmark: an analysis using two policy models

This study was conducted as a literature review, which aimed at finding experiences with OOP payments. The search was based on a combination of the following search terms: out-of-pocket payment, co-payment, general practice, general practitioner, doctor, cost-sharing, demand, utilization of services. The search was conducted in three databases: PubMed Medline, EMBASE and Cochrane Library. The search included articles in English, Swedish, Norwegian and Danish, which have been published during the last ten years. Furthermore, the search was limited to studies from the Nordic countries, “Finland, Sweden, Denmark, Norway, and Iceland” [8], in order to make sure that the experiences were comparable to the Danish context. All searches were conducted during the last week of October 2014. In total, 1821 articles were found. The main inclusion criteria were that the articles discussed OOP payments for GPs and that they were from Nordic countries. The selection process resulted in five articles being included in the analysis.

Based on the chosen articles, a chain search was conducted, where articles that were possibly relevant were found from the list of references in the chosen articles. This made sure that a lot of relevant material was found and provided a quality control, since included articles had been approved by previous research [9, 10]. The same inclusion criteria were applied, and the chain search found six articles.

The results were analysed using The Health Policy Triangle (HPT) and The Kingdon Model. The HPT is a theoretical framework that consists of four factors: actors, content, context and process. Actors refer to individuals or organisations who are either involved in the policy-making or affected by the policy. When defining possible actors, it is important to note what power and interest each actor possesses. Content refers to the actual content of the policy [11]. This is not applicable to this study, since the article is discussing the possible implementation of OOP payments and not the details about the payment itself. The context of a policy can be divided into four categories: situational factors, structural factors, cultural factors, and international factors [12]. Situational factors refer to situations in the surroundings, while structural factors describe the structure of the policy system. Cultural factors refer to how the surrounding culture can affect the policy, and international factors focus on how international conditions affect the policy. Finally, the process describes how the policy is initiated, developed, implemented and evaluated [11]. The latter two factors are outside the scope of this study, and will not be discussed further.

When discussing the potential implementation of OOP payments, the Kingdon Model is used. According to this, possible policies are likely to reach the agenda if a policy window is created. A policy window is created when three streams run together. The problem stream describes whether the general perception in the population is that the government should deal with the problem, and whether the government is able and willing to do so. The policy stream covers the debate in the media and among politicians, analysing the support of a solution. Finally, the politics stream describes whether the opinion among politicians is in favour or against the policy, which can be influenced by the national mood, campaigns or even replacement of the government [11].

Denmark has faced a financial crisis since 2008, which represents the situational factors. This means that the unemployment rate has increased, causing reduced income among some Danes [44]. Because of this, the Danish population might be more resistant to co-payments, as they will further decrease their income.

The Danish Constitution has two rights that represent the structural factors. The first is freedom of speech, which allows the population to express any non-discriminating opinion [45], and the second right ensures that Danes can meet in groups [46]. Because of these factors, the public debate is potentially influencing politicians.

The cultural factors can be perceived as a barrier to the possible implementation of OOP payments policy. The Danish healthcare system has mainly been publicly funded since 1973, meaning that Danes are used to GP visits being free of charge. The fact that visits to the GPs have been publicly funded for so many years is generally seen as an obstacle for implementing co-payments, since the payment structure to GPs has an established history [26].

In contrast, international factors may promote implementation, since Danes may be influenced by neighbouring countries and international organisations. Co-payments to GPs have been an integral part of the health systems in Sweden and Norway since the establishment of the systems. Swedish citizens pay between 100 and 200 SEK (€ 10 and € 20) each time they visit a GP, with a yearly maximum of 1100 SEK (€ 115). Children and adolescents under the age of 20 do not pay for using GPs [47]. In Norway, citizens pay 127 DKK (€ 17) each time they visit a GP, with the exception of children below the age of 16 [48]. The yearly maximum is 1712 DKK (€ 230) [49]. The fact that Sweden and Norway have implemented co-payments for GPs might convince the Danish population that cost-sharing could be a good way to reduce public expenditure for health. Furthermore, a report published by OECD concluded that Denmark should consider introducing co-payments to GPs in an effort to lower public expenditure [50]. When international organisations publish studies in favour of OOP payments in Denmark, it gives the government some of the needed support, and thereby promotes implementation.

The authors of a Danish literature review found that co-payments may lead to decreased demand, meaning that citizens will visit their GPs less often. More specifically, through a literature search in EconLit and a chain search, 51 quantitative studies published between January 1990 and December 2011 on the effects of OOP payments were included in the review. The authors highlighted that the vast majority of studies found that OOP payments reduced the number of services demanded, including the number of visits to GPs [51]. An economic analysis by CEPOS estimated that introducing a co-payment of 127 DKK (€ 17) per GP visit for every citizen above the age of 16, with an annual limit of 1,712 DKK (€ 230), would save society 2.3 billion DKK (€ 310 M), assuming a decreased demand of 10% [49]. Even though CEPOS is an organisation in favour of OOP payments, and this saving might be overestimated, the study suggested that OOP payments would reduce public health expenditure, which had also been observed in other countries that had introduced cost-sharing practices. The decrease is mainly observed in the use of pharmaceuticals and is supported by extensive literature, including a literature review that found that a fixed co-payment reduces drug use even when the co-payment is small [52].

OOP payments can lead to increased inequality [26]. A study based on the 2009 EU SILC survey showed that there is a link between co-payment and having unmet healthcare needs. 24.9% of the participants noted financial reasons as the main reason for their unmet medical needs. Furthermore, OOP payments would decrease accessibility to GPs, because payments would act as a barrier. According to a Norwegian study, the number of visits to GPs decreased when the accessibility decreased [53]. It can be assumed that the accessibility will decrease the most for the lowest socio-economic groups, because these groups have less money. This decreased accessibility is inconsistent with Danish health law, where it is said that there should be easy access to healthcare [54]. This discrepancy between Danish health law and OOP payments may hinder implementation. Moreover, a literature review showed that the vulnerable groups in society are more affected by OOP payments than the less vulnerable, richer groups [51]. A Danish nationwide survey established that immigrants and their descendants make lower use of services with OOP payments compared to ethnic Danes [55]. These studies showed that the lower socio-economic classes of Denmark might reduce their use of GPs the most, putting them at risk of ill health in the long term and creating inequalities. Increased inequality is a serious matter, because it can decrease overall life expectancy (LE) and can have economic consequences for society [56].

Even though LE in Sweden and Norway is higher than in Denmark [57], in the long term fewer visits to GPs may deteriorate health, as diseases will be diagnosed at a later and probably more costly stage, i.e. in hospital. To our knowledge, no studies have calculated the additional cost of treating patients at a more costly stage due to OOP payments to GPs in a Danish context. The fact that no studies exist may reflect the lack of support for OOP payments in Denmark.

For OOP payments to GPs to have no negative effects on health, it would require that the population is capable of distinguishing between necessary and unnecessary visits. A literature review on drug use and co-payment showed that payments for drugs decreased the use of both necessary and unnecessary drugs [58]. This study indicates that citizens might not be capable of making the right choice, due to the parameter of asymmetric information, thereby risking deterioration of their health in the long term because of OOP payments.

The problem stream consists of the increasing health expenditure for GPs because of increased pressure on the healthcare system. In Denmark, it is the task of the government to make sure that public costs are kept within the approved limits [15], meaning that it is accepted by the population that the government has responsibility for dealing with the problem. However, since the cultural factors in the context of the policy have shown that Danes have been used to free access to GPs for decades, it might not be generally accepted that the government is entitled to introduce OOP payments.

Regarding the policy stream, the analysis of the media debate has shown that the media in general has been publishing stories against OOP payments. This means that even though OOP payments to GPs were generally accepted as a solution to decrease public expenditure on health, the support might be low. Furthermore, the analysis of the Danish population and GPs as stakeholders has shown that these actors are against introducing co-payments. This could mean that there is no policy stream, because the population and the media do not support OOP payments and can use the possibilities of expressing their opinion to influence the politicians. However, the analysis of the international factors has shown that the politicians can find support from Sweden, Norway, and the OECD.

Overall, it can be argued that the problem stream and the politics stream run together, because OOP payments are generally accepted to be a means to decrease demand, and there is some support within the Parliament. However, all three streams are very unlikely to run together, because of the resistance towards OOP payments to GPs in the general population, the media, the GPs and parts of the Parliament. This means that it might be difficult to implement OOP payments in Denmark.

This study was a literature search, meaning that the strength of the analysis is dependent on the material found. The quality of the chosen literature was in general high, because all the chosen articles were found through scientific databases [10]. Additionally, the majority of the chosen articles have been published in leading scientific journals, meaning that the majority of them are high quality, peer-reviewed articles [59]. However, because our inclusion criteria only accepted studies from the Nordic countries, similar studies from outside the Nordic context were excluded. None of the included studies discussed the possibility that introducing OOP payments that are below the market value may in fact lead to an increased demand in some population groups. Studies discussing this may have been excluded because of the inclusion criteria. Consequently, this paper may overestimate the impact of OOP payments on the reduction in demand.