The year of transparency: measuring quality of cardiac care

National quality measurements are initiatives from government, supervision institutions, insurance companies and patient organisations. From the perspective of the individual hospital and/or cardiology department these initiatives can be judged to be external requests for accountability.

The Dutch Healthcare Inspectorate (Inspectie voor de Gezondheidszorg, IGZ) has an important task, as described in Article 36 of the Healthcare Insurance Act, to verify if hospitals meet the minimum level of quality according to general healthcare acts and the professional standards as defined by the different medical specialists [4]. Verification is achieved by surveillance of compliance to the law, regulations, professional standards and guidelines. The Dutch Healthcare Inspectorate focuses on surveillance of the highest risks by mostly collecting process and structure indicators as delivered by the healthcare providers. For ST-segment elevation myocardial infarction, outcome, structure and process indicators (number of PCI procedures, in-hospital or 30-day mortality, door-to-needle time or door-to-balloon time and the percentage of patients referred for cardiac rehabilitation) are acquired. For pacemaker and implantable cardioverter defibrillator implantations the number of procedure-related complications within 90 days has to be registered [15]. The Safety Management System (Veiligheidsmanagementsysteem, VMS) is a Dutch patient safety program started by the Ministry of Health, Welfare and Sport and supported by all hospitals, primarily initiated to reduce avoidable patient injuries during hospital admission. These VMS indicators are now (partly) incorporated in the indicators collected by IGZ.

As described in the introduction, the Dutch healthcare system changed in 2006, aiming to reduce rising healthcare costs while improving quality of care. In order to achieve this, the Dutch government introduced a regulated healthcare market. Two new important acts were introduced: the Healthcare Insurance Act (Zorgverzekeringswet) and the Act of Regulation of Healthcare (Wet Marktordening Gezondheidszorg) [4, 16]. In the new system the health insurance companies play a central role, positioned between patients and caregivers, with a shared responsibility to ensure good quality and cost-effective care. For the first time it became possible for the insurance companies to selectively contract. Additional to the responsibility in limiting the rising healthcare costs, insurance companies are required to analyse and interpret quality of care provided by caregivers. Article 14 of the Healthcare Insurance Act, and the general directorial based on this, states that insurance companies share the responsibility for efficient and timely healthcare of good quality, based on professional standards defined by the scientific professional organisations and healthcare providers. The explanatory memorandum of the act states that more information on outcome of caregivers will be available in the future [17]. Currently, however, more attention is given to the volume and cost agreement than to the provided quality of care [18, 19]. The Dutch Healthcare Authority (Nederlandse Zorgautoriteit, NZa), with its task of overseeing the regulated healthcare market, is positive about the increased attention to quality of care in contracting during recent years [20]. A recent report of the Council for Public Health and Healthcare (Raad voor de Volksgezondheid en Zorg, RVZ) concluded that health insurance companies have to be more transparent about the criteria used for contracting care, which caregivers are contracted and how patients were involved in the process of contracting [21].

When focussing on cardiology in the Netherlands, there are three large national registries: (1) the National Cardiovascular Data Registry (NCDR), (2) the Supervisory Committee for Heart Interventions in the Netherlands (Begeleidingscommissie Hartinterventie Nederland, BHN) and (3) Meetbaar Beter. These registries are all initiated by the healthcare professionals involved and funded by the participating hospitals. From the perspective of the hospital and cardiology department, these initiatives can be regarded as internal quality initiatives, primarily meant to improve the internal quality of the individual healthcare provider. Increasingly, on request of the government, Healthcare Inspectorate, insurance companies and patient organisations, these registers are used for external accountability as well. The NCDR was initiated by the Netherlands Society of Cardiology (Nederlandse Vereniging voor Cardiologie, NVVC) and organised in steering committees to develop different databases for every area in cardiology [22]. Furthermore, the NCDR data are sent to the national Implant Register (Medisch Implantatenregister), recently initiated by the Ministry of Health, Welfare and Sport [23]. The NCDR is NEN7510 certified, an information security certificate. Currently, data of more than 250,000 device patients are registered and almost all hospitals (85) participate in NCDR. The BHN, a collaboration of cardiothoracic surgery, cardiology, anaesthesiology and paediatric cardiology, is a national registry of cardiac interventions. The BHN includes data of all 16 cardiothoracic centres since 2007 [24]. Meetbaar Beter, initiated by the Catharina Hospital, Eindhoven and the St. Antonius Hospital, Nieuwegein, is a collaboration of currently 12 cardiothoracic centres in the Netherlands. In the near future Meetbaar Beter will also include data from PCI centres without on-site heart surgery. The Meetbaar Beter initiative raises a new Dutch concept of transparency reporting: a patient-oriented and physician-driven registry. Meetbaar Beter registers patient-outcome data in order to optimise clinical processes. Annually, these outcome indicators are analysed and when required, improvements are established [25]. Besides these national initiatives, in 16 out of 25 regions in the Netherlands, cardiologists collaborate with general practitioners, emergency services and patient representatives in a regional context which is called NVVC Connect. NVVC Connect aims to optimise regional care for myocardial infarction, atrial fibrillation and heart failure patients [26].

The first initiatives to register the quality of cardiac care on a national basis were started in Sweden and the United States. Currently, the Swedish registry SWEDEHEART collects data from all 74 hospitals in Sweden. In the United States, the American College of Cardiology (ACC) initiated the National Cardiovascular Data Registry (NCDR®CathPCI), which contains information on 12 million patients from 1577 participating centres [27]. England and Wales collect information in the Myocardial Ischaemia National Audit Project (MINAP)/National Institute for Cardiovascular Outcomes Research (NICOR) database on all patients with acute coronary syndrome, which contains data from all 236 hospitals [28, 29]. These registries are useful instruments for addressing important clinical questions by retrospectively selecting patients for a randomised trial [30]. An overview of other international AMI registries is given in Tab. 2. It emphasises that sufficient funding is important to ensure a solid registry.

Since registration is a time-consuming process, it is important to ascertain whether the used quality indicators actually provide the desired effect of improving quality of care. Chatterjee et al. have described three mechanisms by which registrations can help to improve patient outcomes [31].

First of all, reporting about quality of care in cardiology itself can lead to an incentive for hospital leaders and clinicians for improvement. In order to achieve this, it is important that results can be shared safely. Studies show that by paying attention internally to quality of care, improvement in outcomes of healthcare can be observed, which is called the Hawthorne effect [32].

Public reporting can also be a powerful incentive for clinicians and hospital leaders to improve. Besides, transparency increases confidence of patients in the healthcare system. However, public reporting of quality indicators in the United States also demonstrated some disadvantages. First of all, some studies comparing reporting states and non-reporting states show no differences in outcome [33]. A further concern of public reporting is that it will lead to risk aversion among physicians, deferring patients with more complex pathology, as is demonstrated in the literature. For example, in the United States, the majority (89 %) of interventional cardiologists have reported that the decision to intervene in critically ill patients was influenced by participating in the reporting of quality measures [33]. A registry confirmed this trend in practice, showing that patients in reporting states (e.g. New York) were less likely to undergo a PCI procedure if they were in shock [34]. Public reporting of CABG mortality in New York led to an increase of sicker patients being referred to the adjacent state Ohio [35]. Although transparency in quality indicators is increasing in the Netherlands, currently results cannot be linked to individual caregivers. The Society for Cardiothoracic Surgery in Great Britain and Ireland, in collaboration with the National Health Service, provides open access to information on treatment results of all individual cardiothoracic surgeons [36]. To improve transparency and to help patients in making informed decisions in the Netherlands, the aim of the Dutch government is to publish results of quality of care measurements at a national website for patients; www.​kiesbeter.​nl [3].

Pay-for-performance is the newest quality improvement mechanism, which is gaining attention from healthcare leaders and healthcare insurance companies as a strategy for maximising quality while controlling costs. Pay-for-performance implies a shift in paying for quality healthcare instead of volume of care, which can be a strong stimulus to improve quality [31, 37].

As addressed before, the results of any measurement must be relevant for the different stakeholders in healthcare. In the current use and development of quality indicators it is also important to take into account that indicators are scientifically acceptable. The indicators should therefore be reliable and valid [38].

Reliable means that the indicator provides the same result on repeated measures and that the dataset is as complete as possible with uniform datasets which are collected in a uniform way. Also the Dutch Federation of University Medical Centres (Nederlandse Federatie van Universitair Medische Centra, NFU) points out in their report on a central vision on registration of care the value of a uniform standardised dataset. They aim to develop a uniform structure of elementary data elements and the use of a unified medical language based on international standards [39]. The use of universal definitions is encouraged by the International Consortium for Health Outcomes Measurement (ICHOM), an international non-profit organisation with the aim of transforming healthcare systems by measuring and reporting patient outcome [40].

Validity means that the indicator measures what it is intended to measure. This requires a good methodological quality, taking into account potential differences in case-mix and random variation. A common remark heard by doctors is that they worry about case-mix correction and that a negative outcome compared with others can be explained by the more difficult patient population served. A good case-mix correction applied in crude data could change the compared clinical outcome and is important to avoid unintended consequences [35, 41, 42]. Furthermore, it is important to remember that quality indicators are just an indication of the real quality of care. Therefore, the indicators should give appropriate coverage of the quality of care of a department and be in line with the crucial aspects of current strategies to improve quality of care.

On a national level, the current registries have to establish a unified collaboration and define useful indicators. NCDR, Meetbaar Beter and BHN are currently formalising this. It is intended that as of January 2016 NCDR and Meetbaar Beter will collaborate on the PCI database. On a regional level, hospitals should work together to acquire and exchange follow-up data, which requires commitment by the hospital’s board of directors in measuring quality of care on a local level. Possibly this can be achieved by appointing a board member on quality of care, as proposed by the Dutch government. Secondly, feasibility is of major importance. Currently, hospitals and their departments deliver between 600 and 1000 quality indicators to external parties each year. Since possibilities for automatic computation are still limited in most hospitals, this task is extremely time consuming. This leaves little time to use these externally reported quality indicators for internal improvement programs. Indicators should be easy to register in daily practice and ideally automatically obtainable. For this, integration of hospital electronic patient file systems with the national registries is essential. To ensure a long-term and reliable registry, solid funding is crucial. However, next to this national funding, sufficient funding on a departmental level is vital, especially if implantation is expected to become mandatory. Finally, it has to be decided what level of transparency is useful to improve the quality of care.