Best practice for chronic heart failure patients--writing guidelines is not enough

The questionnaire survey of SHAPE (Survey of Heart failure Awareness and Perception in Europe)1 examined attitudes to diagnosis and management of chronic heart failure (CHF) among 6887 European physicians. Overall the findings are important because of the large number of respondents to the questionnaire [2965 primary care physicians (PCPs), 2041 cardiologists, and 1881 internists and geriatricians] from nine European countries. It may come as no surprise to the typical reader of this journal that the authors conclude that CHF guidelines are not being applied consistently by PCPs and are variably applied by internists and geriatricians. Cardiologists appear to be performing well in most aspects of diagnosis but fall short of optimal care for some aspects of management. However, the overall response rate was low (13%) and varied considerably by country (France and Germany were the lowest with 9% and Romania the highest with 23%). The conclusions should therefore be interpreted with some degree of caution in that there is a significant potential for sample bias. The use of various heart failure diagnostic tests was variable in the nine countries studied, except for echocardiography, where cardiologists appear to be almost exclusively the gatekeepers, with 92% performing it routinely in suspected heart failure compared with 71% of internists and geriatricians. Limited access to diagnostic tests is a consistent message from PCPs—80% have direct access to ECG but only 30% have direct access to echocardiography within 3 months. It is not surprising therefore that only 58% of PCPs perceived that ECG was necessary and 64% thought that echocardiography was necessary to make a diagnosis of heart failure. It is possible that these findings were due to lack of knowledge on the part of PCPs, but could also be due to an attitudinal difference related to their inability to access these diagnostic tests as readily as cardiologists and geriatricians. At the time of the survey (2002–2003) only 19% of cardiologists and 13% of internists/geriatricians had access to brain natriuretic peptide (BNP) as a diagnostic test for heart failure. Of further interest to cardiologists, for all nine countries only 11% of cardiologist respondents routinely perform coronary angiography, 19% perform standard exercise testing (only 4% with gas exchange analysis), and only 18% perform a Holter recording routinely in their CHF patients. German cardiologists, however, appear to be enthusiastic users of Holter (59%) and exercise testing (49%), while cardiologists in the UK appear marginally to favour coronary angiography as a routine investigation (18%). The reasons for these differences in approach are not clear but are likely to represent differences in organization and financial aspects of service delivery between nations. The use of various evidence-based drug therapies appears to be significantly lower for PCPs compared with hospital-based physicians. These findings are highlighted by the authors, but there are alternative ways of interpreting these data. Some of the differences in responses could have been due to differences in the typical patient group managed by cardiologists, geriatricians, and PCPs. No data were available to assess this, but we can assume that the cardiologists looked after patients who were younger with fewer co-morbidities and with fewer polypharmacy issues, as described in other single-country studies.2 We might also assume that these younger CHF patients managed by cardiologists were more aware and better educated about the type of treatments that were appropriate for them.3 In the case of β-blockers which were prescribed ‘always or often’ by 40% of PCPs but considered for ‘90% or more of patients’ by 73% of cardiologists, this difference might reflect genuine differences in the types of patients looked after by PCPs and cardiologists. One additional confounding issue in comparing PCPs and cardiologist is that they were asked different questions on the questionnaire, and some of the differences in apparent use of drugs such as β-blockers and angiotension-converting enzyme (ACE) inhibitors could have been due to different interpretations by the physicians of the issue being addressed in the question posed. Furthermore, younger CHF patients might also have been more mobile and better suited to travel for diagnostic tests such as echocardiography and Holter monitoring. Taking these factors into account, the findings may be less surprising than first appears. There is certainly some need to explore these issues further in the proposed second phase of the SHAPE programme. An important finding of concern in this survey is that only 55% of PCPs would refer a patient with mild to moderate heart failure to a specialist. This clearly needs further study and perhaps requires educational programmes with emphasis on encouraging access for all patients to specialists with appropriate diagnostic skills and access to more complex forms of drug and device therapy. A second issue is the finding that few physicians used guidelines as their main source of information about CHF. For cardiologists this may be complacency, rejecting the guidelines in favour of the actual trial evidence itself. For internists, geriatricians, and PCPs this may simply be because they are too busy to read them. This was a finding confirmed in a previous single-country study.4 Guidelines do not appear to be the best single way of ensuring high quality care for CHF patients. They are often written by cardiologists, although there is an increasingly welcome trend to involve other healthcare disciplines and patients themselves. Ideally, guidelines should be developed, launched, and disseminated within the context of a healthcare quality improvement programme with a number of components: dissemination, education, economic evaluation, clinical evaluation, and accreditation. Each guideline should be accompanied by a clinical framework which defines the minimum standard of care to be expected for a patient presenting with CHF in a community or hospital setting. Despite difficulties in agreement between countries and physicians, there should be common ground where standards of care for CHF could be developed using the strongest of recommendations. If agreement could be achieved, it would be a significant step towards common standards of healthcare in the European community. Involvement of patient groups focused on safety and high quality care is one of the foundation stones of this type of approach, and the SHAPE programme would benefit greatly by integrating its efforts with initiatives currently being developed at national and international levels.5,6 In North America, there are a number of programmes currently underway aiming to improve outcomes for heart failure patients in acute and chronic settings based on systematic approaches to diagnosis and management.7,8 Similar small studies in Europe have demonstrated successful outcomes with β-blockers in the elderly.9 Such programmes need to be developed on a much larger scale across Europe if they are to result in significant improvements in outcome. The European Society of Cardiology is in an ideal position to do this, but it is likely to be expensive! One other way of improving implementation of guidelines is by providing incentives to physicians. For CHF this could be achieved by financially rewarding physicians for managing heart failure better. The general medical services contract established in 2003 in the UK has partly succeeded in creating improvements in the quality of clinical care for patients with CHF. PCPs are paid extra money if they initiate and maintain a register of patients with CHF, ensuring that patients have had an echocardiogram, and ensuring that they are taking ACE inhibitors. This is a reasonably good start, but it clearly falls short of achieving the highest quality of care.10 Specialist heart failure nurses have a role to play in improving the quality of care for CHF patients. Many of the perceived difficulties of starting, uptitrating, and monitoring evidence-based therapies is a skilled, time-intensive process well suited to multidisciplinary programmes involving specialist nurses.11 It is hardly surprising, therefore, that in countries where this type of organizational change has been embraced and widely implemented the uptake of evidence-based drug therapies appears better in the SHAPE survey.12 This may reflect more highly developed healthcare services in general rather than services directly related to heart failure. Finally, in attempting to assess physician perception of heart failure management throughout Europe, there are many economic, cultural, and demographic variations that could result in attitudinal differences between healthcare professionals. In addition, there are organizational differences in healthcare services that could result in differences in attitudes and views observed in the SHAPE survey. These issues were not explored by the authors either in the questionnaire itself or in their own discussion. It would be of interest to examine these in the second phase of the study, in particular to understand the potential differences between private and public healthcare systems, and differences between services with and without well developed multidisciplinary care programmes, and to examine whether incentives (financial or other) provided nationally or locally have influenced the physician's attitudes to heart failure diagnosis and management. Such information would be of enormous value in shaping the way we write, disseminate, and implement the contents of all guidelines, not only those for CHF. Conflict of interest: none declared.