A case study in implementation: why is speed so variable?

In every country within the same healthcare economy, some areas adopt new practice quickly, while others do not. Why is the speed of implementation so variable? Medicine is a relationship between the patient and the healthcare provider, whether doctor or nurse. This relationship is based on an ethos adopted over the last 3000 years, and the way in which doctors and nurses interact with their patient cannot suddenly change. This meeting has stressed repeatedly the importance of guidelines. For most healthcare providers, guidelines represent the most important source of information for a clinical situation they have not encountered before. By using guidelines, healthcare providers can feel that they are acting in a way that is reasonable for the patient. Every professional society is now producing guidelines at both national and international levels, and many believe that this is where the focus for change is. The UK is probably one of the most conservative countries in terms of changing practice. Even in the UK, however, there was a rapid change in practice when beta-blockers were introduced into heart failure treatment. A series of three maps shows the pattern of beta-blocker prescription in August 2000 (when all guidelines recommended prescription of beta-blockers), July 2001, and July 2002 (Figure 1). The maps show an increase in the numbers of prescribers over time which spreads out from centers with strong opinion leader support (e.g. Glasgow, Manchester, and Liverpool). By the last assessment, virtually all UK centers were prescribing beta-blockers at a rate that leading centers were prescribing two years before. Therefore, even in a bureaucratic system like the NHS, there can be a rapid change in practice if professionals feel that the evidence is secure. Evidence-based medicine (EBM) is now a formal dogma. In the presence of a clinical problem, the healthcare professional defines it as a simple question, selects the most likely resource to answer that question, designs a search strategy, and summarizes and grades the evidence found. The process is essentially hierarchical and requires a sequence of steps (Figure 2). Although this process is suitable for academics, it is less useful in clinical practice. What most healthcare professionals require is a simple statement that recommends which treatments should be considered for a particular patient. This is how most of us operate in day-to-day practice. As discussed earlier, the concept of economics and prioritization of resources is very important. The population is aware of this and is fearful that doctors and nurses may concentrate more on the cost of treatment than on its benefit to them. For doctors, especially in Germany but also elsewhere, a process is now under way that rewards, but also penalizes, certain types of behavior. This raises the question of how doctors and nurses are managed in their healthcare system. Healthcare systems are struggling, as mentioned earlier, to cope with the increased potential for expenditure and in controlling professionals who were previously very powerful. Within this context, is EBM a force for good? I have worked with organizational sociologists to look at how EBM has changed the way healthcare professionals interact with their patients. Based on these investigations, it appears that in a range of specialties, including cardiovascular medicine and oncology, we should be skeptical that EBM necessarily improves care. EBM probably improves the professional's knowledge of what the evidence is, but it does not necessarily improve the outcome for patients. The explanation is that EBM is only partially implemented (Figure 3). One of the main barriers to implementation of EBM is that the value of the evidence is contested. Healthcare professionals may feel that the patient who presents is not exactly covered by the guidelines and they wish to interpret them. A second barrier is that professionals may continue to demonstrate a significant level of autonomy. One of the definitions of being a professional is having autonomy, and this contrasts with being instructed on what to do in guidelines based on EBM. A third barrier is that clinical knowledge is considered to be developed over time, based on experience in treating patients face-to-face. Looking at the situation in terms of a power balance, EBM preserves the doctor's power because it is based on an objective process for decision-making. EBM also reduces variability in practice and reduces contestability. In another way, however, EBM is anti-professional. While increasing standardization, EBM reduces the value of personal experience and goes against the ethos of the physician's tradition. ‘It used to be that the tertiary centre gave you the answer and then you rolled it out, but you soon realized that this is utterly inappropriate.’ General physician ‘The specialist centres get a highly refined group of patients (often sicker) and what they do the research on is totally inappropriate.’ Cardiologist ‘People don't talk because they have their own little empires in a way … I find it very difficult that there isn't just a simple way … that doctors who deal with heart failure get together, thrash it out round the table, and get it sorted.’ Nurse ‘In my experience it is usually “it wasn't invented here”. People don't take readily to other people's already done work and seem very keen to re-invent the wheel.’ Strategic Health Authority ‘Different groups of people have been persuaded by different things. Clinicians would be very much persuaded by the fact that their peers were doing better, perhaps. Management teams would be persuaded by pressure on them from performance management.’ Strategic Health Authority It is clear from the literature that knowing the evidence (“logos”) is only one part of the process of making changes happen in a large, complex organization. There has also to be a connection to the “ethos” of the professional, and without connection there cannot be communication. The third element is the emotional component (“pathos”), because all of us have much invested in the services we provide (Figure 4). Another issue in implementation is directing care to those areas where there is greatest burden of disease. Generally, heart failure patients are elderly, live outside cities, and do not live near teaching hospitals (Figure 5, left). Analysis of the areas of best care (using beta-blocker prescriptions as a proxy) shows, however, that this is provided in regions where there is least need (Figure 5, right). This is true for many conditions that affect the elderly and represents a real problem for healthcare systems. What are the barriers to implementing change? As Figure 6 shows, there are many. Some are system barriers, such as organizational constraints, lack of reimbursement, and inadequate staffing. In overcoming organization constraints, it is important to provide professionals with structures that enable them to implement what they know to be good care. Reimbursement is a huge issue that accompanies any attempt to introduce new technology. Barriers among professionals have been discussed above and include a lack of awareness or agreement, low motivation, and inability to reconcile guidelines with patient preferences. Among patients, barriers include a lack of awareness, limited access to care, and low levels of adherence. Within this context, it may be that we sometimes concentrate too much on patient concordance rather than on the professional's ability to come forward with guidance on what to do. We also usually spend very little time in trying to influence healthcare systems to ensure these barriers are knocked down. This is a complex process and each of us has to direct efforts to those areas where we have most influence. If we look across the whole of the EU, what are the key factors that identify success and where are the areas that have changed practice quickly to improve outcome? The first factor in implementing change is to recognize that there is a problem, rather than defending current practice. Examples of successful changes in practice include access to echocardiography, titration of therapy, education, communication, and funding. A second factor is to establish a firm evidence base on which to found changes in patient care. There is no shortage of evidence in heart failure, and potentially rather the opposite. Another key issue is local ownership. Do we feel that, within our own area, we have the responsibility to move things forward? An integrated multiprofessional approach is also important. We cannot have separate initiatives from different groups. We have, in addition, to be realistic on what we are trying to achieve, without setting over-ambitious targets but also not setting challenges that are too easy to reach. A further very important factor is alignment with other levers for change within the healthcare system. Reviewing the progress achieved frequently is another obvious need. Finally, patients and carers represent an increasingly powerful force in planning change, particularly in Western democracies through the activities of politicians.