Skip to main content

Quality and costs of primary care in Europe

Final Report Summary - QUALICOPC (Quality and costs of primary care in Europe)

Executive Summary:
4.1.1 Executive summary
In the QUALICOPC study, primary health care systems of 34 countries have been analysed and compared in terms of quality, costs and equity. Results of the study contribute to the evidence on the added value of strong primary care for the performance of health care systems in general. In Europe 26 EU member states are included in the project, as well as Iceland, Norway, Turkey, Switzerland and FYRO Macedonia. Furthermore research institutes from Australia, Canada and New Zealand have joined to implement the study in these countries on their own account. Surveys have been held in each country among samples of General Practitioners and their patients. The data provides insight in the professional behaviour of General Practitioners and the expectations and experiences of their patients. An important asset of the chosen approach is that patients’ answers to the questions in the questionnaire are linked to the information provided by their own General Practitioners in his or her questionnaire. In the QUALICOPC study newly collected data has been combined with existing data sources, such as the Primary Health Care Activity Monitor for Europe database.

This study provides inputs in the form of evidence for health system change. In terms of agenda setting the results of the QUALICOPC-study support the evolution of thinking in primary care (and at a slower pace in health care in general) from disease oriented care to person-focused care. Epidemiological transitions towards chronic disease and multimorbidity have made it necessary to put the goals of people themselves in front. The identification of the improvement potential in the area of person-focused care in the perceptions of patients themselves is a powerful input for setting the agenda for primary care.

This study also supports shifts in the balance between primary care and other sectors of health care. However, this is something that primary care and policies related to primary care in themselves cannot realise. Restricting specialist and hospital care requires also policies that directly address this sector of the health care system. To work out policies in these areas requires strong political will. Reallocation of means to primary care to improve the economic conditions of primary care will create strong reactions.

Primary care policies are context dependent and will have to have a strong local component, next to support in the form of national policies. Designing such policies at local and regional level, needs to take into account the capacity to absorb innovations. The QUALICOPC-study worked through a large network of people involved in primary care. Through this network and through the extensive dissemination activities within the QUALICOPC-study, the study has contributed to the development of the absorptive capacity of primary care systems.
Project Context and Objectives:
4.1.2.1 Context of the project
The 1978 Declaration of Alma Ata stressed the importance of creating and sustaining a s strong primary health care system as part of the health system and, more broadly, in coherence with other sectors. Following this declaration many European countries have taken initiatives towards strengthening primary care. The principles formulated at the Alma Ata conference have had a visible impact on European health systems. In the 1990s decision makers of the former communist countries increasingly supported reforms towards the general practice/family medicine model. Also in other parts of Europe health system reforms were aimed at strengthening primary care. More recently, primary care reinforcement is worldwide still high on the policy agenda, also as an approach to respond to the economic crisis. Policies towards primary care reinforcement often start from the notion that a strong primary care system benefits a nation’s health and health care system. These notions, however, are based on the plausibility of effects rather than on its base of evidence This points to the relevance and importance of health services research to generate sound evidence for decision makers.

What is primary care?
Primary care is the first level of professional care where people present their health problems and where the majority of the population’s curative and preventive health needs are satisfied. Therefore primary care services can best be available close to where people are living with the least possible obstacles to access. Primary care is generalist care, focused on the person in his or her social context, rather than on diseases. The mix of disciplines which make up the primary care workforce may differ from country to country, but general practice or family practice is considered as the core of primary care. Besides general or family practitioners, the most common primary care providers in Europe are general internists, general paediatricians, dentists, pharmacists, therapists (e.g. physiotherapists and speech therapists), and mental health care workers (e.g. community psychiatrists and psychologists) (Health Council of the Netherlands, 2004; Boerma and Dubois, 2006).

Benefits of primary care
Scientific research, both international comparative and within the United States, has provided evidence on benefits of well-developed primary care systems, in terms of better coordination and continuity of care and better opportunities to control costs (Shi et al., 2002; 2005 ; Macinko et al., 2003; Starfield, 1994; Delnoij et al., 2000; Health Council of the Netherlands, 2005). Recent evidence shows that strong primary care is associated with better population health, lower rates of unnecessary hospitalisations and relatively lower socioeconomic inequality. However, overall health expenditures were higher in countries with stronger primary care structures (Kringos et al, 2013).

The QUALICOPC project
The QUALICOPC study has been initiated because the relevance of some of this research for the European situation was limited, more in-depth analyses were needed to corroborate the findings. The variety of models of organisation and provision of health care services found in Europe, are favourable circumstances to undertake sound and comprehensive studies on the merits of primary care for health care systems in general. The rich diversity of regulatory mechanisms, funding schemes and modes of financial and non-financial incentives for providers as well as users of services makes Europe a laboratory for comparative research and a pool of good practices (Groenewegen et al., 2002).

Getting insight in variation and effect of elements of primary care is not a theoretical exercise. The WHO World Health Report 2008, titled ‘Primary health care now more than ever’, has clearly articulated the need to mobilize the production of knowledge on primary care. Despite the broad agreement about the merits of well organised primary care systems, knowledge about its active ingredients was inconclusive. Better international comparative data and analyses of good practices were needed to produce information to policy makers and those responsible for provision of services about the drivers of strong primary care (WHO, 2008; Starfield and Shi, 2002; Rosser and Van Weel, 2004; De Maeseneer et al., 2003). Health reforms in many European countries share the aim to strengthen the first level of care, and as a result there is a demand for comparative information and a growing tendency to learn from foreign experiences. Not just in Central and Eastern Europe, where health care reforms have been much more fundamental than elsewhere, but also in other European countries information on equitable and cost-effective models of primary care can help decision makers to move health sector reforms forward. Health systems research and the exchange of information on health systems continues to be essential in that process.

Primary care and costs
Available evidence suggests that health care systems based on strong primary care spend less and are better able to contain costs. An OECD review of studies on the determinants of international health care cost differences concluded that ‘the role and organisation of ambulatory care is of crucial importance in the overall efficiency and effectiveness of healthcare systems’. Other studies showed that the overall cost of healthcare is generally lower in countries where primary care performs a gatekeeper function. One study showed that health care systems with gate keeping General Practitioners had a less strong increase of health care costs. The remuneration system of General Practitioners plays a mediating role in this relationship: countries with a gate keeping system often have a payment system that is at least partly based on the number of patients on the list (capitation).

Quality and efficiency indicated by avoidable hospitalisation
Avoidable hospital admissions can be used as an indicator of health care performance. An admission is avoidable when a relatively expensive hospital admission for a certain condition could have been prevented by effective or accessible primary health care. As stated earlier, research in the USA has shown that gate keeping systems are able to reduce unnecessary care. The availability of family doctors or General Practitioners is related to lower rates of hospitalization for certain conditions.

Equity
As regards equity, the evidence of a relationship with the structure and strength of primary care at a national level is scarce. Equity is usually studied by analysing large national health interview surveys. An OECD study could not substantiate the relationship with primary care. Concerning the effects of strong primary care on equity results are inconclusive. Until now, no such effects have been clearly demonstrated in international studies

4.1.2.2 Project objectives
The QUALICOPC project aims to evaluate primary care in Europe against criteria of quality, equity and costs. QUALICOPC looks at what strong primary care entails and aims to provide an answer to the question:
What effect does the strength of the primary care structure have on the performance of health care systems?
To this end, the organisation of primary care at the level of the GP practice and national structures for primary care are related to the quality of primary care as perceived by patients, how primary care is organised at the GP practice level and overall health care system goals such as equity and access. The strength of primary care is determined by the degree of development of a combination of primary care functions both at the national level (governance, economic conditions and workforce development) and at the GP practice level (access, continuity of care, coordination of care and comprehensiveness of care) in the context of its health care system (Schäfer et al 2012).

Quality of care
‘Quality’ is a broad concept and can be divided into three features . Firstly, quality can be related to the structure of care (referring to characteristics such as equipment and human resources), the process of care (the actual delivery of care) or the outcomes (consequences of the process in terms of e.g. health status or patients’ evaluations) of it. This division is based on Donabedian’s well-known framework.
Secondly, generic and specific dimensions of primary care quality are distinguished (Campbell et al., 2000). Generic dimensions are those applicable to all health care services; examples are equity, accessibility and user friendliness of services and effectiveness (both clinical effectiveness and interpersonal effectiveness). Specific dimensions are typically applicable to primary care:
- continuity of care (longitudinal care; episodic continuity)
- coordination and integration (with other professionals and levels of care)
- scope of services (broad range of curative and preventive services)
- community orientation

Framework of the study
Based on these notions of quality of care, a conceptual framework was developed for the study that not just shows an ordering of the concepts and the way they are inter-related, but also identifies three levels, ordering the study activities (see Figure 1, attachement).

The study has distinguished the following three levels:
1. Primary care (sub) systems: Primary care is understood as a subsystem of a country’s health care system. In total 34 primary care systems have been included in the study (26 EU Member States, Switzerland, Norway, Iceland, Turkey, Macedonia as well as Canada, Australia and New Zealand). At this level, indicators of design and organization of primary care were used. These relate to financing, regulation and available financial, human, other resources.
2. Provision of primary care services: These are the units of provision within a primary care system, such as health centres, GP practices, policlinics, networks. This level was characterised by its involvement in first contact care; the range of curative and preventive services offered to the population; the conditions for continuity and integration of care and the community orientation of the providers and their organisations.
3. Users of services: the clients or patients of primary care providers: At the users level responsiveness was essential, which relates to physical, financial and psychological accessibility and convenience of facilities and staff.
Elements in figure 1 were inter-related. Design and organization of primary care in a country are hypothesised to influence how tasks and activities in primary care are carried out. Profiles of provision of primary care tasks and activities, in turn, influence the responsiveness of the system as experienced by the users in primary care. Primary care features together relate to overall goals of the health care system, the first of which obviously is good health. Other relevant system goals are: access, equity, costs and quality. At the level of users of primary care services this project studied the perceived quality of care, based on actual patient experiences with primary care and values about the importance of aspects of primary care. Quality as perceived by users is influenced by both the system design and structure and the process quality (provided tasks and activities).

Specific objectives
In order to reach the general aims of the project, some more specific objectives were formulated, including:
- To develop questionnaires for General Practitioners and patients and acquire a qualitatively and quantitatively sufficient response in each country;
- To gain insight in activities and tasks, the process quality and accessibility of General Practitioners;
- To gain insight in quality of care as seen through the patients’ eyes, the perceived access to care and the perceived cost barriers;
-To gain insight into the following relationships:
a. Primary care structures and outcomes related to avoidable hospitalisation;
b. The structure of primary care related to quality of services provision;
c. The structure and process of primary care related to patients’ perceived outcomes;
d. The structure and process quality of primary care related to costs;
e. The structure of primary care related to access and equity;
f. Good practices in integrated primary care in Europe (overarching the other associations).
- To facilitate the cooperation and communication between the project participants;
- To guarantee high quality of output;
- To disseminate the scientific results and the recommendations and implications of the project to relevant EU and national decision makers, international organizations, primary care professionals and the research community.

Project Results:
This chapter describes the main science and technology results from the QUALICOPC project. Four major outcomes of the project will be discussed:

- The study design developed by the QUALICOPC consortium;
- The measurement instruments developed by the consortium;
- The data collected among GPs and their patients in 34 countries;
- Main findings of the project.

4.1.3.1 Study design shared with the forum of researchers
QUALICOPC has used an ambitious methodology integrating different levels of care by the use of existing databases and surveys among GPs and their patients. The use of elements from the 1993 Task Profile study has provided information on changes that have occurred since then and, moreover, an innovative element has been the addition of the patient's perspective, which has increased the chances of meaningful interpretations (Schäfer et al, 2012). Designs like this are still very rare. That is the reason for us to publish the design of the study. The reference of the published design of the study is:

Schäfer WLA, Boerma WGW, Kringos DS, De Maeseneer J, Greß S, Heinemann S, Rotar- Pavlic D, Seghieri C, Svab I, Van den Berg MJ, Vainieri M, Westert G, Willems S, Groenewegen PP. Study protocol: QUALICOPC, a multi-country study evaluating quality, costs and equity in primary care, BMC Family Practice. 2011, 12: 115. (Highly Accessed Publication).

The survey methodology used consists of a multi-actor design, allowing to connect the information on primary care practices with information provided by patients from these practices and system level information. The use of these state-of-the-art methods is expected to serve as a 'model of good practice' for future health services studies.

4.1.3.2 Measurement instruments
The QUALICOPC (Quality and Costs of Primary Care in Europe) study aims to analyse and compare how primary health care systems in 35 countries perform in terms of quality, costs and equity. In order to do this, one of the aims has been to develop questionnaires for General Practitioners and patients

The development of the questionnaires consisted of four phases: a search for existing validated questionnaires, the classification and selection of relevant questions, shortening of the questionnaires in three consensus rounds and the pilot survey. Consensus was reached on the basis of exclusion criteria (e.g. the applicability for international comparison). Based on the pilot survey comprehensibility was increased and the number of questions was further restricted, as the questionnaires were too long (Schäfer et al, 2013).

As a result of this process, four questionnaires were developed: one for General Practitioners, one for patients about their experiences with their General Practitioner, another for patients about what they consider important and a practice questionnaire to be filled out by field workers who visited the practices. The General Practitioner questionnaire mainly focuses on the structural aspects, e.g. on the economic conditions, and the care processes, e.g. comprehensiveness of services of primary care. The patient experiences questionnaire focuses on the care processes and outcomes, e.g. how do patients experience access to care? The experiences questionnaire mainly relates to the consultation that patients just had had. The questionnaire about what patients consider important is complementary to the experiences questionnaire, as it enables weighing the answers from the latter, which is the basis for improvement scores from the patients’ perspective. Finally, the practice questionnaire includes questions on practice characteristics (Schäfer et al, 2013). A description of the development process and the final questionnaires have been published in the following paper:

Schäfer WLA, Boerma WGW, Kringos DS, De Ryck E, Greß S, Heinemann S, et al. Measures of quality, costs and equity in primary health care: Instruments developed to analyse and compare primary health care in 35 countries. Qual Prim Care. 2013 May;21(2):67-79.

4.1.3.3 Data collected among General Practitioners and patients
Second, the project aimed to acquire a qualitatively and quantitatively sufficient response in each country. In the QUALICOPC study (Quality and Costs of Primary Care in Europe), surveys were held among General Practitioners and patients in 31 European countries (EU 27 – except for France- , FYR Macedonia, Iceland, Norway, Switzerland, Turkey) and 3 non-European countries (Australia, Canada, New Zealand). In each country, a nationally representative sample of GPs (target: N= 220 GPs; Cyprus, Iceland, Luxembourg and Malta N=80 GPs) and patients (target: N=2200, respectively N=800) filled in the questionnaires. In Turkey, Spain, Belgium and Canada, larger samples were taken in order to make comparisons between regions. Only one GP per practice or health centre was eligible to participate (Schäfer et al 2011).

GP questionnaires were filled in either on paper or electronically (online or via a tablet computer). In nearly all countries, trained fieldworkers were sent to the participating GP practices to collect patient data using paper surveys. In Belgium, the patients filled in the survey on a tablet computer. In parts of Sweden, Denmark, England, Canada and New Zealand, , the local practice staff instead of fieldworkers was instructed to distribute and collect patient surveys on paper according to the study protocol.

Fieldworkers were instructed to consecutively invite patients 18 years or older, who had had a face-to-face consultation with the GP, to complete the questionnaire until 10 patient surveys were collected. Nine patients in every practice completed the questions about their experiences in the consultation which had just occurred. One questionnaire included questions about the patient’s values in primary care. In addition, each trained fieldworker filled in a short questionnaire about the practice facility, e.g. access to the practice for handicapped. A unique practice identification number links GP responses to the responses of 10 of his or her patients and the fieldworker survey, allowing for multi-level analyses of the data.

Ethical approval was acquired in in accordance with the legal requirements in each country. Both GP and patient surveys were carried out anonymously. Although a standardized data collection procedure across all countries was strongly recommended and strived for, in the actual data collection strategy cultural and ethical requirements for each country were taken into account.

Data collection took place between October 2011 and December 2013. The GP questionnaire was filled in by 6,830 GPs, the Patient Experiences questionnaire by 58,606 patients and the Patient Values questionnaire by 6,742 patients (database version 3.0 January 2014).

4.1.3.4 Development of an international network of primary care researchers
This study was carried out by a consortium of 5 European research institutes from Belgium, Germany, Italy, Slovenia and the Netherlands. As data needed to be collected in 34 countries, the project consortium built an extensive network of involved research institutes. In each country a devoted national coordinator was appointed to coordinate the data collection. This was necessary because the national coordinators could provide insights in the specific features of primary care and their countries and were able to organise the data collection in their country under the supervision of the project consortium. The national coordinators have actively collected the data in their country, but have become involved later on in the project as well as they have analysed their national data, advised the consortium on specific findings of the project and actively participated in dissemination activities. After the official finish of the project, the network will continue to exist and be used for further dissemination activities and possible future research. Below an overview is provided of the institutes and persons involved in the network.

Overview of Qualicopc network developed during the project

Australia: Dr Ian McRae and Prof. Louisa Jorm (Australian Primary Health Care Research Institute and University of Western Sydney)
Austria: Prof. Dr. Manfred Maier, MD and Kathryn Hoffmann, MD, MPH (Department of General Practice, Center for Public Health, Medical University of Vienna)
Belgium: Prof. Sara Willems (Department of Family medicine and Primary Care, University of Ghent)
Bulgaria: Prof. Petko Salchev, MD (Medical University, Sofia)
Canada: Walter P. Wodchis, PhD, MAE, MA, Assoc. Prof. (Institute of Health Policy, Management and Evaluation, University of Toronto)
Cyprus: George A. Samoutis, M.D PhD (University of Nicosia)
Czech Republic: Bohumil Seifert, MD PhD, Assoc. Prof. & Nela Šrámková, MD (Czech Medical Association of J.E.Purkyne (CzMA))
Denmark: Prof. Peter Vedsted and Jacob Reinholdt Jensen, MD (Research Unit for General Practice, Aarhus University)
England: Prof. Niro Siriwardena, MD and Coral Sirdifield (University of Lincoln)
Estonia: Prof. Margus Lember and Kaja Põlluste (University of Tartu, Dep. of Polyclinical and Fam.Medicine)
Finland: Prof. Elise Kosunen (University of Tampere)
F.Macedonia: Katarina Stavric, MD (Center for Family Medicine, University of Skopje)
Germany: Stefan Greß PhD and Stephanie Heinemann (University of Applied Sciences Fulda)
Greece: Prof. Christos Lionis, MD PhD, HonFRCGP (Faculty of Medicine University of Crete, Greece)
Hungary: Imre Rurik, MD (Medical and Health Science Center, University of Debrecen)
Iceland: Ofeigur T. Thorgeirsson, MD EMPH and Jórlaug Heimisdóttir (Reykjavik)
Ireland: Claire Collins, MD (Irish College for General Practice)
Italy: Chiara Seghieri, PhD (StANNA- Sant’Anna School of Advanced Studies)
Latvia: Gunta Ticmane, MD (Rural Family doctors association of Latvia)
Lithuania: Sarunas Macinskas (Lithuanian University of Health Sciences)
Luxembourg: Monique Aubart, MD (Société Scientifique Luxembourgeoise de Médecine Générale)
Malta: Philip Sciortino, MD and Glorianne Bezzina, MD (Malta)
Netherlands: Willemijn Schäfer, MSc and Wienke Boerma, PhD (NIVEL)
New Zealand: Prof. Toni Ashton (Health Systems Section, School of Population Health, University of Auckland)
Norway:Prof. Hasse Melbye, Torunn Bjerve Eide (University of Tromsø, Norway)
Poland: Adam Windak, MD and Marek Oleszczyk, MD (ZiZ Centrum Edukacji Sp. z o.o Krakow)
Portugal: Luis Pisco, MD (Department of Family Medicin, University of Lisbon)
Romania: Dana Farcasanu, MD, PhD (Foundation Center for Health Policies and Services)
Spain: Toni Dedeu, MD (Primary Care Research Institute IDIAP Jordi Gol)
Slovakia: Eva Jurgova, MD (Educational Centre for Health)
Slovenia: Danica Rotar-Pavlic, MD (ULMF)
Sweden: Prof. Tomas Faresjö (Linköpings universitet)
Switzerland: Nicolas Senn, PhD and prof. Thomas Bisschoff, MD (Policlinique Médicale Universitaire)
Turkey: Mehmet Akman, MD, Assoc. Prof. (Turkish Foundation of Family Medicine)

4.1.3.5 Main findings of the project

The added evidence on quality and cost
Multi-level approach
The QUALICOPC study has followed a unique approach. We have distinguished data from three different levels. At system level we have identified indicators of system design and organization of primary care. These indicators are mostly related to financing, regulation and available resources. System level indicators determine the framework and incentives for primary care providers. Therefore, to a certain extent, they determine the behaviour of providers. However, as we assumed while drawing up the study design of QUALICOPC, system level characteristics are able to explain behaviour only to a limited extent. Health services research which is exclusively based on system characteristics loses a lot of information. Therefore, we have to take into account the level of service provision as well. This means we gathered unprecedented data on the involvement of primary care providers in first contact care, the range of curative and preventive services offered to the population, the conditions for continuity of care and the community orientation of primary care.

We have assumed that there is variation at the service provision level although system level characteristics are usually identical for all primary care providers within the same country. We have learned that variation at the service provision level can be more important for the attainment of system goals such as access to care, cost containment, efficiency of care and the perceived quality of services than variation at the system level. By including a third level of analysis, the patients using primary care services, we have been able to gather information about crucial system goals such as access to care and quality of services from the point of view of patients. At the patient level – the users of services – we were able to gather data on responsiveness and physical, financial and psychological accessibility. By analysing all three levels we are able to gather insights about the complex interaction of system characteristics, variation on the practice level and responsiveness on the patient level.

Data on the variation on the practice level is extremely useful for analysing the relationship between characteristic of primary care and system goals. We have found that, although variation between as well as within countries is high, there are also similarities. First, we have analysed the state of primary care in Europe (and beyond) at the time of data collection in 2011 and 2012.

Clusters of countries
By analysing the conditions for the care process in particular practice workforce, practice availabilities, workload and coordinating activities – we were able to determine clusters of countries. The conditions of the care process in these clusters are not identical. However, we found that they are quite similar. We are convinced it is a great advantage that the “clusters of similar primary care characteristics” we have identified are based on empirical analysis on the practice level. It adds additional richness of data to the more conventional clustering of countries on the system level in terms of broad system characteristics, such as Bismarck or Beveridge systems. This traditional clustering has become less meaningful over the past decades with the introduction of elements of competition in both types of systems and the breakdown of the third major group of European health systems that were modelled on the system of the Soviet Union and its direct satellites.

Efficiency through ICT
One particular condition for efficient work in primary care (and health care in general) is access to information and quick communication through the increased use of computers and ICT in general . By now, computers are used in primary care in all countries studied. However, the specific uses still differ widely. Use for clinical functions and record keeping shows large variation between countries. Communication is still an underused function in most countries. This regards both the communication between care providers, in the form of being able to share (information of) medical records and directly exchange information e.g. with pharmacies or hospitals, and the communication between care providers and patients. This latter type of communication is still in its infancy in all countries, witness the low frequencies of email consultation. One step further would be integrating e-health applications to support patients in their self-management in regular primary care treatment. The added value of the overview of ICT use in primary care is that this gives a snapshot of a quickly changing field with a large potential for innovation.

Developing service profiles
Under the conditions the core primary care providers – the GPs – provide their services. Services profiles were measured in four areas. First of all the access role of GPs, i.e. the range of services for which they are the first point of contact; secondly, their role in the management of acute conditions and chronic disease; thirdly, their activities in terms of technical procedures, such as minor surgery; and finally their involvement in preventive care. An important observation is that what GPs do in these areas differs strongly between countries. The biggest differences are in the area of the application of technical procedures. The single area where European GPs show bigger involvement compared to two decades ago, is management of disease. European primary care has responded to changes in health care need, related to ageing, an increase of chronic diseases and multimorbidity. The added value of this information is that it is highly relevant for the substitution potential from specialist and hospital care to primary care. With a view to containing health care costs it is important to be aware of the fact that services that are provided in hospital in one country are mainly done in primary care in another country. Of course the actual possibilities to shift care depend on acceptability among patients, the skills and equipment in primary care and the relationships of primary care with hospital care.

Performance of primary care systems
Second, issues for the evaluation of primary care have been addressed. We refer to care processes that are indicative for strong primary care: continuity, coordination and comprehensiveness; equity in access to care; and two reactions to suboptimal access to or availability of primary care, viz. use of emergency services and hospitalization that could have been prevented by good and timely primary care.

While data on the patient level can be analysed to show variation in terms of perceived quality, we have also used data on the practice level to analyse variation in terms of process quality. The key dimensions of process quality in primary care we have used are informational continuity of care, coordination of care and comprehensiveness of care.

Based on previous research we have assumed that high levels of informational continuity, coordination and comprehensiveness indicate a high level of process quality. As a result, we were able to construct country scores of each of these three indicators. These country scores can be used by policy makers to assess whether the level of process quality in primary care in their country is below or above average in our sample of 34 European and non-European countries. This kind of comparative information has not been available before on the level of service provision. We also were able to cluster countries with similar levels of process quality. We found that – at least regarding countries with an above than average level of process quality – the results are overall consistent with studies which have compared data on the system level (PHAMEU).

Accessibilty and equity
While quality of services is considered an important health system goal, the same is true for access to care. Equal access for equal need means that access to care is not determined by a patient’s socio-demographic characteristics or ability to pay but by his or her need for medical care. From the outset, we considered the analysis of access to be a very important research area within the project and this book. We have used data on the patient level in order to determine to which extent patients did postpone visits to the general practitioner due to financial reasons. In our view, this is a very valid indicator whether the ability to pay determines access to care and therefore overrides the need for medical care. Due to the richness of our empirical data we were also able to determine the influence the patient’s gender, ethnic background, age, family income level and educational attainment on the postponement of visits due to financial reasons. We found that on average less than ten percent of patients within our sample postponed a GP visit due to financial reasons. However, variation between countries is high. In some countries less than two percent of patient did postpone a GP visit due to financial reasons. In other countries this rate was above 20 percent. This finding highlights that postponement of GP care is an actual problem in several countries. In contrast, previous research indicated that access to GP care is more or less equitable in Europe.
Policy makers can use this data to assess whether the level of postponement of visits due to financial reasons within their country is below average or above average of the European and non-European countries in our sample. While other studies have undertaken similar research endeavours, this research has been limited to a small number of mostly wealthy countries. Moreover, characteristics of the patients, such as gender, age and ethnicity, and of the practices, such as opening hours and easy accessibility through open consulting hours, have often been left out.

Bypassing primary care
We have analysed the relationship between the use of emergency department and the accessibility of primary care. The rationale for this approach is twofold. First, emergency department services are more expensive than primary care services. Second, emergency department services are less suited to treat routine conditions and non-urgent complaints than primary care services. As consequence, excessive use of emergency department services leads to an inefficient use of health care resources. Since overcrowding in emergency departments is a common problem in many countries all over the world, possible approaches to reduce excessive emergency department us is an important issue for policy makers across the world in order to reach several important system goals (access, quality, cost containment) at the same time.

Therefore, we assumed that high accessibility of primary care reduces the need for excessive use of emergency department services. The appropriate use of primary care may help to treat health problem in a timely way so that emergency department use can be avoided. By analysing data on the patient level we were able to largely confirm this assumption. Our results show that there are considerable differences of emergency department use between countries. However, variations between practices within countries are even more pronounced. In some countries it may be a good investment to increase capacities for accessible out of hours primary care facilities and to eliminate obstacles to timely primary care such as co-payments or high costs of travelling.

Closely related to the use of emergency services instead of regular primary care is hospital use for conditions that could be well-managed in primary care. Internationally there is consensus about a range of conditions that can be as well or better managed in primary care than in hospitals. Examples of such conditions are asthma and diabetes. There is still a lack of information about avoidable hospitalization in a number of countries included in the QUALICOPC study. However, for the countries that have comparable information it turns out that good accessibility of primary care is related to less hospitalizations for these conditions.

This result is in line with the current literature. The added value of the analysis within the QUALICOPC study is that rates of avoidable hospitalizations could be related to more specific indicators for structure and process aspects of primary care that are prevalent in a country.
Opportunities for improvement

Next, we moved the emphasis from description and analysis to opportunities for change. We have looked at the improvement potential in the area of person-focused care – one of the core values of primary care; at the technical efficiency of primary care by looking at the relationship between inputs and outputs in primary care; and at the possibilities to identify good practices.

From the point of view of policy makers, the potential of health care systems to become more responsive to patients’ needs and expectations is crucial. Therefore, we analysed the expectations and experiences of the patients regarding person-focused care. Person-focused care includes a number of dimensions such as accessibility, continuity, comprehensiveness, communication and patient involvement. By combining patients’ expectations and their actual experiences we were able to calculate improvement scores for each country for each of the five dimensions. The idea behind the improvement scores is that if patients find a certain aspect of care important, but have negative experiences in that same area, there is room for improvement. We calculated improvement scores at country level for each of the five dimensions of patient-focused care. We found that in 24 of 33 countries there is indeed room for improvement on at least one of these dimensions. The potential for patient perceived improvement is highest for comprehensiveness of care. Policy makers can use this information for agenda setting in health policy in their country.
In a second step we endeavoured to determine a relationship between structural characteristics of primary care and patient-focused care on the system level. We found that the potential for improvement regarding several features is lower if the overall structure of primary care is strong. The structural aspects of primary care were derived from the PHAMEU-study which has collected information on the strength of primary care in Europe through available statistics and descriptions of health care systems and through expert opinions from the participating countries. The structure of primary care was divided in issues related to governance (e.g. the existence of explicit policies on primary care), economic conditions for primary care (e.g. primary care expenditures, income of GPs) and primary care workforce development (e.g. the ratio between GPs and medical specialists). These structural dimensions of primary care are typically amenable to policy making. Although we cannot provide simple recipes for improving person-focused care from the patients’ perspective, policy makers who find the responsiveness of care to be an important policy goal, consider improving the structure of primary care.

Efficiency in primary care
The efficient use of scarce resources is an important goal on the health system level. Therefore, we analysed primary care efficiency across European countries. In a first step we used practice level information. We measured efficiency by relating the level of outputs in primary care (such as patients visited per week, contacts per day and procedures performed in the practice) to a given mix of inputs (such as working hours, skill mix and equipment). Our results show that overall input-output efficiency is quite high. However, there are considerable differences between countries and we are able to identify countries which seem to provide care less efficiently than others. We assume that this information is rather valuable to policy makers.
In a second step we related the level of outputs to overall quality perceived by patients. This analysis is a good illustration of the approach of our project since it uses data on the practice level as well as on the patient level. The results of this output-outcome relation are similar to the results of the input-output relation. However, countries performing poorly in the first relations do not necessarily do so in the second relation as well. Still some countries performed worse in the second stage analysis than in the first stage analysis. Policy makers in these countries therefore should investigate how outputs are delivered in primary care.

Good policies and good practices
The concept of good practice or even best practice is a very popular one in heath policy. We showed that the concept is rather ill defined. We therefore suggest to differentiate between good policies on the system level and good practices on the practice level. Our research has shown that there a number of good policies and good practices which may lead to a more effective and more efficient way toward health system goals such as access to care, cost containment, efficiency of care and the perceived quality of services. Now it is up to health policy makers to implement good policies in order to increase incentives for primary care providers to implement good practices. However, as noted at the beginning of this concluding chapter, policy makers should always be aware that system level characteristics are able to explain behaviour only to a limited extent.

4.1.3.6 Implications for health care system change
Changes and reforms in retrospect
Health care systems are constantly in flux. Usually, changes are only seen as the cumulated effects that result from gradual, internal developments. Our knowledge in the area of diagnosis and treatment of disease increases, but also new knowledge of efficient organisation of health care is being developed. The roles of patients, professionals and managers in health care gradually change, as do insights into the role of the preferences of patients in decisions about treatment. As a result of many small and incremental changes the primary health care landscape evolves. If we make up the picture with longer periods of time in between, changes suddenly become visible.

Sometimes bigger changes take place that affect the structure of health care systems. We than speak about health care reforms. Deliberate changes in the structure of the system and the relations between parts of the system. In the past decades the biggest reforms occurred in the countries of Central and Eastern Europe, especially in those that became members of the European Union. In those countries the main health system functions have changed. In reorganizing service provision, a common aim of these countries was to strengthen primary care and to introduce GPs by retraining existing and educating new personnel.

Many of the western European countries have shown only small and incremental changes. Compared to the countries of Central and Eastern Europe there was no urgency to change the structure of the health care system. The challenges of demography and costs develop slowly. Some of the smaller changes, e.g. in the area of remuneration, have had (anticipated or unanticipated) consequences that had wider implications than just the transfer of money. Think for example of pay-for-performance initiatives.

Pressures from the financial crisis
There is now a new urgency to reform health care in a number of countries, especially in the Southern part of Europe, resulting from the financial crisis. Although the primary emphasis is on the reduction of health care costs, there is an awareness that strengthening primary care could be part of the solution. Primary care helps combat the economic crisis in two ways. First, stronger primary care provides better opportunities for cost containment and perhaps even lower costs. However, we should be aware that costs reductions usually start in those parts of health care that are easiest to influence by national politicians. Those parts include primary care in some countries and preventive services and public health in all countries.
Secondly, equitable and accessible primary care might soften the impact of the crisis for the most vulnerable people. Postponement of care in the face of out-of-pocket costs is the first strategy for those who cannot afford much. The result is higher costs of care and incapacity later on. National policies are necessary in this respect, but we should be aware that vulnerable people are usually geographically concentrated, as is the impact of the crisis. This asks for an equally strong emphasis on local solutions in the form of transferrable good practices.

Evidence for agenda setting
The QUALICOPC-study provides inputs in the form of evidence for health system change. The first step is providing evidence for agenda setting. This relates both to international (through international organisations such as the European Commission, the European Forum for Primary Care, WONCA Europe and WHO Europe) and national (through the activities of the network of national coordinators) initiatives. In terms of agenda setting the results of the QUALICOPC-study support the evolution of thinking in primary care (and at a slower pace in health care in general) from disease oriented care to person-focused care. Epidemiological transitions towards chronic disease and multimorbidity have made it necessary to put the goals of people themselves in front. The identification of the improvement potential in the area of person-focused care in the perceptions of patients themselves is a powerful input for setting the agenda for primary care.

Consistent policies requiring political will
The QUALICOPC-study also supports shifts in the balance between primary care and other sectors of health care. However, this is something that primary care and policies related to primary care in themselves cannot realise. Restricting specialist and hospital care requires also policies that directly address this sector of the health care system. To work out policies in these areas requires strong political will. Reallocation of means to primary care to improve the economic conditions of primary care will create strong reactions.

Local capacity to absorb innovations. As mentioned earlier, primary care policies are context dependent and will have to have a strong local component, next to support in the form of national policies. Designing such policies at local and regional level, needs to take into account the capacity to absorb innovations. The QUALICOPC-study worked through a large network of people involved in primary care. Through this network and through the extensive dissemination activities within the QUALICOPC-study, the study has contributed to the development of the absorptive capacity of primary care systems.

Potential Impact:
4.1.4.1 Impact of the project
Starting point for the QUALICOPC study has been that the evidence on the effects of strong primary care is inconclusive. Reforms favouring primary care are based on the plausibility of effects rather than on its base of evidence. The available evidence is from studies with a limited focus, and not representing the diverse situations of health care in the countries of Europe. The QUALICOPC project considerably contributes to this base of evidence and thus advances the state of the art of (primary) health services research. The outcomes of the QUALICOPC project are used to inform the European Union and other international organisations, such as the WHO, but particularly also national governments. The deeper insights, provided by this project, in specific elements of primary care organisation and provision which have a positive effect on performance of health systems in general, will contribute to more effective health policy.

Dissemination by the international network of primary care researchers
The QUALICOPC project has been carried out by an international network of health services researchers in 34 countries. In each country a national coordinator has been appointed to coordinate the data collection. The national coordinators often concerned health services researchers working at a department of primary care or health professionals (GPs) with experience in health services research. This approach lead to the establishment of an international network of researchers and health professionals who are committed to issues related to the outcomes of the project. As a result, this network has also been used as a major “tool” in the dissemination of the project results. To enhance this even further, all national coordinators have received their national data to be used for their own purposes. They have been encouraged to disseminate national project results among scientists, policy makers and relevant stakeholders in their own country. As a result several of the coordinators have, for example, organised national meetings with General Practitioners and other stakeholders and prepared scientific papers for (national) journals. In this way the project increases its impact at the national level.

QUALICOPC in FYR Macedonia, Australia, Canada and New Zealand
Furthermore, publicity and personal contacts with researchers and research institutes in countries outside Europe, especially those with a primary care system comparable to European countries, have resulted in the expansion of the study outside Europe. Research groups in Australia, Canada and New Zealand have been able to raise their own funding to participate. Additionally, in Europe Macedonia FYR has joined the study. As the participation of these ‘new’ countries has resulted in a larger ‘number of observations’ at country level, the analyses to disclose the mechanisms of primary care and their effect on health care system performance are facilitated. Apart from that, a broader international participation has provided the study a deeper insight in the national strategies of PC systems, professional behaviour of health care workers and the expectations and actions of patients around the world. Through the inclusion of the non-European countries, the project consortium has tried to increase the impact of the project outside of Europe.
Impact on health services research

The project has led to the establishment of a network in which researchers in- and outside of Europe have been continuously cooperating in the field of health services research. Health services research is an applied science with a relatively short tradition. QUALICOPC has stimulated the further development and expansion of this science. In some of the participating countries the position of health services research is still weak and its possibilities not well understood. By the involvement of researchers from these countries and the production of important results we have tried to enhance visibility and authority of health services researchers as well as its acceptance among the users of this knowledge. The collaborations within the network and activities carried out related to the use of the data collected in this study will be continued after the official finish of this project period to assure long-term impact of the project.

State-of-the-art methodologies
In this project, we have also tried to increase impact by using sophisticated, state-of-the-art methods of quantitative data analyses. The project has been modelled in such a way that it allows for multilevel statistical models, which are currently used to a limited extent in health services research. The collected data is built up from three levels: patients, nested within GP practices, nested within countries. QUALICOPC has used a survey methodology in a multi-actor design, allowing to directly connecting the information on GP practices with information provided by patients from these practices. Another important, and sparsely used, data analyses technique that has been used is ecometric scale construction. In this technique multilevel analyses are used to construct latent scales in which the hierarchical data structure is taken into account. All scale scores have been constructed centrally using this method. The use of these state-of-the-art methods is expected to serve as a ‘model of good practice’ for future health services studies.

Use of external sources
Finally, we have tried to increase the impact of QUALICOPC by combining previous work (which itself had already a good impact) with new elements. First of all, in our analyses we have used the Primary Health Care Activity Monitor (PHAMEU) database, which was part of research commissioned by DG-SANCO. This database contains national level information on primary care in 31 countries. For the countries which were not included in the database (Australia, Canada, New Zealand and FYR Macedonia), we have collected data on the same indicators using and identical approach.
Next, we have based our instruments for the data collection (the questionnaires) on existing validated questionnaires, such as the Primary Care Evaluation Tool (PCET) used for WHO Euro to analyse primary care in various countries and the well-known Primary Care Assessment Tool (PCAT) developed by Barbara Starfield. Another important existing instrument that was used is the questionnaire from the European study on GP Practice profiles, conducted in 1993. By repeating several questions from this questionnaire, we have been able to compare service profiles of GPs between 1993 and 2012. Currently a paper on this topic is under constructing. Merging these validated approaches into this project has not just been efficient, but it has also offered opportunities to link the QUALICOPC results with those from other studies in which these concepts and instruments have been used.

4.1.4.2 Main dissemination activities
The outcomes of QUALICOPC are aimed to serve various users, being scientists, policy makers, managers, and professionals such as GPs, in a large number of countries, both in- and outside Europe. To reach all target groups, it is chosen to disseminate results, using various methods, including:

- scientific publications;
- presentations at conferences;
- education activities;
- a book;
- a policy brief;
- stakeholder meetings;
- and a website.

The various activities targeting different stakeholders will now be explained further.
Scientific audience
To reach the scientific audience we have aimed at publication of project results in peer reviewed scientific journals. It has been our policy to prefer publication in open access journals, such as BMC Family Practice, in which we have published a description of the study protocol. At this point in time, various papers have been published or submitted to scientific journals. The papers concern international comparative papers on the main study topics of the project: avoidable hospitalisation, the quality of the services provision, patients’ perceived outcomes, costs and access and equity.
Given the wealth of the data collected in the study, the dissemination in the form of scientific articles will continue beyond the scope of the project. While major topics related to the study aims have been addressed in the first papers, more research questions are being addressed in following papers. The articles which have already been published or submitted are presented in section 4.2 (List of scientific publications). Below and overview is provided of planned future publications. Moreover, among the Consortium Partners several PhD candidates are working on various papers in preparation of a PhD thesis. The PhD theses are presented in section 4.2.

Overview of planned publications
1. van Loenen T, van den Berg MJ, Faber MJ, Westert GP. Patients’ propensity to seek care and avoidable hospitalization (Leading partner: RIVM)
2. van Loenen T, van den Berg MJ, Faber MJ, Westert GP. Avoidable hospitalization for diabetes and primary care organization: A country-level study based (Leading partner: RIVM)
3. Van Pottelberge A., Hanssens L., Boerma WGW, Willems S. Social inequity in 34 primary care systems: patients postponing a visit to a GP for financial reasons. Health policy and planning 2014 (Leading partner: Ghent)
4. Hanssens L., Van Pottelberge A., Willems S. Inequity in access to primary care: what can GPs do? Health Affairs 2014 (Leading partner: Ghent)
5. Schäfer WLA, Boerma WGW, Spreeuwenberg P, Schellevis FG, Groenewegen PP. The comprehensiveness of service profiles of European general practitioners between 1993 and 2012: which circumstances explain the developments? (Leading partner: NIVEL)
6. Groenewegen PP, Greβ S, Schäfer WLA. Participation of general practitioners in a large multi-country combined general practitioner – patient survey. (Leading partner: NIVEL)
7. De Rosis S, Seghieri C. Medical records in primary care: use of computer and general practitioners’ behaviors. (Leading partner: St. Anna)
8. Seghieri C, De Rosis S, Groenewegen PP The impact of ICT on continuity of care in Primary care. (Leading partner: St. Anna)
9. Vainieri M,Seghieri C, Murante AM, Kroneman M, W. Boerma, Intended and unintended consequence of remuneration strategies. A multicountry study. (Leading partner: St. Anna)
7. Murante AM, Seghieri C, Schäfer WLA. How does the structure of primary care influence the responsiveness? (Leading partner: St. Anna)
8. Heinemann S, Hofmann W, Groenewegen PP, Greß S. Stadt-Land-Vielfalt in einer nationalen Hausarzt-Stichprobe. Rekrutierungsmethodik der QUALICOPC-Studie in Deutschland.(Leading partner: Fulda)
9. Kübler S, Lederle M, Steinbach A-K, Witzmann S. Qualität in der hausärztlichen Versorgung – Bekommen Patienten was sie erwarten?(Leading partner: Fulda)
10. Armutci M, Brenke G, Rott F. Wie weit weg ist meine Hausarztpraxis und wie nah sollte sie sein? Ergebnisse einer Sekundäranalyse der deutschen Patientenbefragung des QUALICOPC-Projektes. (Leading partner: Fulda)
11. Barthel-Kraus E, Bertsch D, Frank F, Magnussen E-C, Steinmetz S. Untersuchung der Dimension Koordination der hausärztlichen Versorgung in Deutschland im Rahmen des QUALICOPC-Projekts. (Leading partner: Fulda)
12. Neumann M, Riesch J, Schöppner O. Sind die Sprechzeiten in der hausärztlichen Versorgung in Deutschland bedürfnisgerecht für Ältere? Eine Sekundäranalyse der deutschen QUALICOPC-Daten. (Leading partner: Fulda)
13. Köhler E, Schroer K, Becher A. Haben Alter, Geschlecht und Krankheit einen Einfluss auf die Kontinuität in der deutschen Hausarztversorgung? (Leading partner: Fulda)
14 Kert S, Švab I, Sever M, Makivić I, Rotar Pavlič D. Patient access to primary health care in Slovenia during the time of economic crisis (Leading partner: ULMF)
15. Rotar Pavlič D, Sever M, Klemenc-Ketiš Z. The strength of primary care in the eyes of general practitioners in 34 countries (Leading partner: ULMF)

Additionally, the scientific community has been made aware of the project and its results through presentations at scientific and policy related conferences. For example, intermediate results have been presented during a workshop at the bi- annual conference ‘The Future of Primary Care’ organised in 2012 by the European Forum for Primary Care (EFPC). At this conference professionals and researchers involved in the field of primary care came together. Additionally, in a later stage of the project various (conference) workshops highlighting the results from the different work packages have been organised. For example, a workshop at the European Public Health (EPH) conference in 2013 included presentations related to the work packages on equity, patient perceived quality of care and avoidable hospitalisation.

Furthermore, the QUALICOPC Consortium has prepared a book containing an introductory chapter, chapters reporting the results of the analytical work packages and a concluding chapter containing a synthesis and recommendations.

Finally, the lessons learned from the project have been used for teaching purposes for students following Bachelor and Master programmes related to topics studied in QUALICOPC. For example in Germany the project consortium partner has provided courses for the Bachelor and Master Students of “Hochschule Fulda”. Moreover, various master students of these programmes have prepared a final thesis, and in some cases even a (German) scientific paper, on the basis of the data collected in Germany. This is just one example of the various education purposes for which we have used the outcomes of this project. The theses prepared by university students are presented in section 4.2 of this report.

Professionals
Next, the study objectives include answering questions which are relevant to professionals (e.g. General Practitioners). Therefore, we have aimed to reach this community by also publishing papers in professional journals, where possible in the national languages. For example we have prepared an article about the content of the study and this has been translated and published by project partners and national coordinators in various languages (Hungarian, Italian, Czech, Flemish and Turkish). The aim of these translations was to create awareness of the study and to motivate General Practitioners to participate in the study (see section 4.2 ‘List of dissemination activities’). As stated above, the national coordinators are being encouraged to disseminate national level results. At this point, various national level meetings with stakeholders have been organised, e.g. a meeting with all General Practitioners from Italy whom have participated in the study, and national papers are being prepared by both the consortium members and the national coordinators. The articles are aimed to be published in journals for a non-scientific readership, such as health care professionals, politicians and health care managers.

Moreover, we have aimed to specifically reach primary care professionals, by presenting study results at the WONCA Europe and WONCA World conference. This conference brings together thousands of General Practitioners from all over the world. At the WONCA World conference in 2013 an interactive workshop was organised during which the professionals were asked to provide the QUALICOPC consortium with practical feedback on the results. Tailored presentations were provided e.g. about the developments of the GP service profiles between 1993 and 2013 and the audience was invited to reflect on this.

Policy makers
The content of the study is, above all, relevant for policy makers working in international settings. Therefore, results have been translated for the policy world in order to be applied. This societal mission has implied targeted publications and other modes of dissemination to this user group. Therefore, a policy brief has been prepared, including concise and easily accessible conclusions and recommendations. The policy brief is being distributed among organisations and stakeholders in (primary) health care in countries within and outside of Europe. During the project period, policy makers from countries which were not included in the study have also shown interest in the results.

To enhance the visibility of the project, links have been established with international organisations will be established in order to promote publicity for the project and its products, for example WONCA Europe (the European branch of the World Association of General Practitioners / Family physicians), the European Observatory of Health Systems and Policies, the European Forum for Primary Care (EFPC) and the European Public Health Association (EUPHA). Results have been disseminated through newsletters of the organisations and the networks have been requested to assist the consortium in the dissemination of results among relevant stakeholders.

Furthermore, it is aimed to reach policy makers through various methods after the official end of the project. For example, and official request has been made to present major results during the Italian Presidency of the Council of the European Union during the second half of 2014.

Project website
In support to the dissemination activities the project consortium has set up a specific project website. The structure of this website contains information about:

- The background;
- Study objectives;
- Publications and presentations;
- The project partners;
- Meetings;
- National coordinators;
- Contact information.
As much as possible information from the project has been disseminated freely. In all dissemination activities the European Commission and the FP7 context has been acknowledged. Furthermore the Commission has been provided with two interim and a final technical and financial reports.

List of Websites:
The address of the project website is: www.qualicopc.eu For more information on the QUALICOPC project, please contact the following persons at NIVEL in the Netherlands: Wienke Boerma at w.boerma@nivel.nl tel: +31 (0) 30 272 9657 ; NIVEL, Otterstraat 118-124; 3513 CR Utrecht (NL).

Project partners
The institutes which have been part of the QUALICOPC consortium are:
- Ghent University - Department of Family Medicine and primary Health Care, Ghent (BE)
- Hochschule Fulda - University of Applied Sciences, Fulda (DE)
- RIVM - National Institute for Public Health and the environment, Bilthoven (NL)
- StANNA - Sant’Anna School of Advanced Studies, Pisa (IT)
- ULMF- University of Ljubljana, Ljubljana (SI)