Comparative effectiveness research — a proposal for a new NHMRC funding stream

Evidence-based medicine underpins high-quality health care.1 The use of evidence in the practice of medicine informs appropriate decision making, reduces variability in clinical practice and helps ensure improvement of patient outcomes. Most evidence relating to new knowledge is derived from randomised clinical trials.2,3

However, evidence from clinical trials necessarily involves small and very carefully selected populations with particular demographics and disease characteristics. Hence, the results of clinical trials are often not directly relevant to clinical reality. In addition, the scientific basis of “standard therapy” against which new interventions are compared has not necessarily been rigorously tested. Uncertainty about how best to use such new evidence versus the range of alternative options available in the “real world” may create variable practices and impair patient outcomes.

Paradoxically, evidence from these trials is used by the Medical Services Advisory Committee and the Pharmaceutical Benefits Advisory Committee to make decisions about approval and reimbursement of health interventions.

Furthermore, while there are obvious ways by which governments and health insurance funds can constrain the costs of new technology, new services and new drugs before their reimbursement, their capacity to influence the use of existing, funded services in a manner that improves the quality of care is limited.

In this article, I outline why comparative effectiveness research — examining the applicability of marketed and/or approved products, services and technologies to clinical reality — is critical to the future of health care in Australia. Such research should be funded by the existing health portfolio, as it would not only help reduce clinical uncertainty but, in an era of increasing cost constraints, could improve the efficiency of health expenditure within the framework of existing funded services.

It might be assumed that funding from the current National Health and Medical Research Council (NHMRC) competitive grants scheme is sufficient for gathering comparative effectiveness evidence. However, this scheme tends to be focused on innovation, track record and scientific merit, but comparative effectiveness research may not be seen to be as novel as research that seeks to redefine the boundaries of new knowledge. In the United States, a focus on comparative effectiveness research has emerged as a high priority of the current administration.4

Given the limited pool of funds available for medical research in Australia, I propose a new funding program that is ring-fenced for comparative effectiveness research but administered by the NHMRC, with applications assessed on the basis of scientific merit, innovation, improved health and patient outcomes, and cost offsets. The new funding stream should encourage research that:

• rigorously tests scientifically based variations in approved or standard health interventions that are publicly funded

• directly compares the relative roles of approved, publicly funded components of treatment algorithms.

A separate funding stream dedicated to clinical research that tests standard health interventions would benefit the community and the government — it would provide a means of optimising the utility of previously funded and approved clinical programs. A simple example of the type of project that such funding might realise is the comparison of variations to approved chemotherapy regimens, most of which are derived from historical practice patterns. For instance, the Short Course Oncology Therapy (SCOT) trial — comparing 6 months of chemotherapy (standard care)5 to 3 months of chemotherapy6 for the adjuvant treatment of operable bowel cancer (to determine whether less chemotherapy is not only less toxic and costly but also equally efficacious) — is currently underway in Australia and Europe.7

In addition, research that compares the components of treatment algorithms — conducted within or across disciplines in relevant populations — would enable optimisation of existing health practices, many of which required no specific approval process to have been introduced into standard treatment algorithms. (Modifications to treatment algorithms are often funded by being absorbed into the Medicare Benefits Schedule [MBS] and/or health insurance funds or by state-based programs.) For example, postoperative radiotherapy as adjuvant treatment for locally advanced gastric cancer is considered one standard of care in Australia and the US.8 Implementation of this modality required no formal approval process and the costs of adding radiotherapy to treatment algorithms were subsumed into the costs of standard care. However, because other studies suggest that perioperative chemotherapy (without radiotherapy) may suffice,9 a comparative effectiveness trial would result in considerable insights into the relative utility of the two approaches, with initial trial-related cost-savings providing substantial offset to the cost of the research.

Furthermore, such research can do more than determine the role of radiotherapy in the treatment of operable gastric cancer. For example, in the Trial of Preoperative Therapy for Gastric and Esophagogastric Junction Adenocarcinoma (TOPGEAR), the study design process required agreement on evidence-based surgical standards, standardised pathology examination and reporting, and standard chemotherapy regimens for this disease — none of which relate specifically to the experimental intervention.10 Hence, the research process resulted in clinician engagement in a change process that directly affected quality of care. By reducing variability in clinical practice outside the remit of the clinical trial, the research methodology will help ensure high-quality standards for patients, regardless of whether they are enrolled in the study. Similarly, the process of identifying and prioritising a potential question would help facilitate a change process inside and outside the specific research endeavour. The importance of such downstream effects is fundamental to understanding the potential impact of clinical research on the quality of the health system.

Other types of more sophisticated research initiatives that test the interface between imaging, surgery and other techniques or practices, as well as the roles of pharmaceuticals, would undoubtedly emerge if the proposed funding stream were created. In so doing, such funding would enhance the capacity of Australian clinicians to promote the health care quality agenda in a cost-effective manner and across various components of the health system, such as the MBS and the Pharmaceutical Benefits Scheme (PBS). It would ensure that comparative effectiveness research initiatives do not have to compete against projects designed to develop new knowledge.

A funding stream for comparative effectiveness research would invite research applications which are primarily designed to promote research into existing, funded health practices. Key criteria in funding decisions could include the likelihood of the research resulting in substantial impacts on the quality of patient care, in large numbers of patients, and the potential of the research to engage the broader health care community in a change process. The funding would not only help establish the relative roles of individual PBS- and MBS-funded interventions, but, from a broader viewpoint, would also show the government how clinical research represents a critical component of health care delivery. It would bring a quality focus to funding agencies concerned with the cost-effective delivery of medical services to the Australian community.

Funds from the existing health portfolio (MBS, PBS, etc) could be directed to a new NHMRC-administered funding stream of $100 million per annum for comparative effectiveness research. To start the program, the National Institute of Clinical Studies could facilitate development of a list of priority areas in three or more areas of health practice (eg, oncology, gastroenterology and rheumatology) by consultation with professional organisations, government agencies and consumer groups — similar to the list of priorities compiled by the US Institute of Medicine.11 This list could be used to pilot the funding program and could subsequently be extended to cover all areas of health practice. While the priorities within each health area could help inform the peer-review process, the call for applications should not discourage unanticipated ideas. In addition, the program should not be designed to provide a dedicated funding source for health economists — rather, it should encourage collaboration between practising clinicians from various disciplines, including experts in fields such as health economics and health system design.

A funding stream dedicated to comparative effectiveness research would improve health care quality, foster clinical practice research, and increase efficiency and effectiveness in the health system. It would also highlight the value of clinical research.