For nearly ten years, the medical community has been investing massive resources in Comparative Effectiveness Research (CER), a form of clinical research which helps to draw comparisons that are of use for medical practice as they are much more closely related to the daily life of practitioners (unlike more conventional clinical research tools which prefer randomised trials which use placebos for comparison, doctors rarely prescribe placebos). CER considers the medical interventions closest to everyday practice, comparing the effectiveness of different medical treatments (which is rarely done in the literature), even drawing comparisons with alternative non-medical treatments. The results of this research – which is usually independent and state funded – should be widely adopted by the medical community and have the support of patients. However, a recent study by the Rand Corporation, by Justin Timbie et al. published in the October 2012 issue of Health Affairs (abstract only available free online in English), shows that these studies are unfortunately slow to change clinical practice and attempts to analyse the underlying causes for this. It defines five factors which affect the North American health system but which are often found in European health systems.
1. Perverse financial incentives, such as fee-for-service payment, that may militate against the adoption of new clinical practices
2. Ambiguity of study results that hamper decision-making by doctors
3. Cognitive biases in the interpretation of new information
4. Failure of the research to address the needs of end users
5. Limited use of decision support (such as medical records) by patients and clinicians.
The authors make three recommendations for improving this situation to enable CER to change doctors’ obsolete decision-making practices. Firstly, policies should be formulated to encourage the development of consensus objectives, methods and evidentiary standards before studies get under way to avoid the results being rejected once the studies have been completed. Secondly, the studies must be aimed at a wider public to prevent the results being adopted by one medical speciality and rejected by others. Thirdly, alternative payment mechanisms should be found to reward performance and effectiveness rather than the number of services (which may be costly) which are rewarded by fee-for-service payments.