Despite the fact that doctors are key to delivering health care they are seldom involved in research and far less often cited as leaders on research teams. The relationship between the research organisations and doctors is the key to understanding their limited involvement in innovation. ‘Good’ research is a painstaking science in which clearly defined research questions are articulated, appropriate methods are applied, data is efficiently collected and appropriate analysis is conducted to craft conclusions that take into account the limitations and strengths of the study. Seldom, if ever, does a single study, no matter how large, offer robust conclusions that will lead to change in practice. The design and execution of high quality research requires expertise which takes many years of further training and experience. The acquisition of these skills may take doctors out of clinics and at a significant personal opportunity cost.
The subject of clinical research, i.e. patients, must give informed consent before they can be included in a study. This is more complicated than working with uncomplaining rats in a sanitised laboratory. In practice limited control over research subjects means that most clinical research cannot be generalised and is therefore less likely to be published in high impact journals. Most research is also conducted at universities, directly or indirectly. Universities and medical schools have to generate a surplus income to grow in size and influence. Very little research in primary care or public health has a commercial value, therefore to profit from clinical research universities rely on government funding. The government agenda may be driven by political imperative. Therefore a government minister unveiling shiny new machines makes for a far more voter friendly photo opportunity than one launching a more efficient way to rehabilitate people with mental illness or manage incontinence in general practice.
Therefore funding is heavily weighted towards biomedical sciences. Here the focus is on cure rather than prevention or more efficient service delivery. Genetic research, nano particles and the study of prions, is therefore more likely to get generously funded than research on system design that would allow people to die in comfort in their own homes.
In 2010 the Australian National Health and Medical Research Council divided its research funding so that 39% of the funds were awarded to preventive medicine and public health. At the same time the majority of government funding on health care in practice is on so-called primary care services.24
For universities the return on investment does not favour clinical research, so that laboratory-based research on a cure for cancer makes a far more compelling case than research involving therapists in the community or models of disease self-management. And yet, in the scheme of things, research in how to deliver an equitable health service is going to make more of an impression on the community in the short term than research on a cure for cancer that may be 20 years away.
Academics understand that universities are financially rewarded for adopting this paradigm by a system that is driven by priorities related to a return on investment. Given the competitive nature of those who enrol in medical school this is a considerable disincentive and drives clinicians out of research. As if that was not sufficient disincentive, there are major challenges to recruiting participants in clinical practice.25
Patients do not seek help from doctors only to spend most of their consultation negotiating an opportunity to participate in research that may or may not benefit them directly. When the patient is paying for the doctor’s time, as is the case in many countries, doctors have no incentive to introduce distractions to that consultation. In reality many of the patients in clinical practice are excluded from research designs which usually favour young, articulate, English speaking, literate, relatively healthy people and not those living with the conditions for whom the evidence has apparently been generated.26