Future of evidence in context

8 minute read

ANZMUSC was born on a chilly Canberra evening in May 2012 to help promote and develop high quality musculoskeletal research and to address the mismatch between the burden of musculoskeletal disease and current research funding.

ANZMUSC was born on a chilly Canberra evening in May 2012.

It was created to help promote and develop high quality musculoskeletal research in Australia and New Zealand, and to address the mismatch between the burden of musculoskeletal disease and current research funding.

ANZMUSC’s vision is to optimise musculoskeletal health through high quality, collaborative clinical research. One of the key mission statements that supports this vision is to “improve the scientific quality of MSK research and its translation into policy and practice”.

While the fundamental aim of ANZMUSC is to facilitate the conduct of the highest quality clinical trials, this will only achieve the overarching vision if the most important clinical research questions are identified, and if the evidence generated is rapidly and robustly translated to practice.

One way in which ANZMUSC is attempting to address the latter is via the concept of “living” evidence. Why living evidence?

Consider the current state of the evidence synthesis that underpins evidence-based medicine: almost unheard of just three decades ago, systematic literature reviews (SLRs) are proliferating. More than 10,000 SLRs are now published every year,1  and this number will continue to rise.

There are clear benefits to this expansion of evidence synthesis, not the least of which is to help summarise the otherwise unmanageable flow of new research literature.

But it is a process that has largely evolved organically. It is therefore not always fit for its purpose, which is to provide high-quality, up-to-date evidence summaries to assist patients, healthcare professionals, and policy-makers to make decisions about healthcare.

The system for updating SLRs has remained mostly ad hoc. Often years, and sometimes decades, can pass between updates on a particular topic. This may be acceptable in areas where the relative benefits and risks of an intervention are clearly established (for example, the Cochrane review of the effectiveness of pool fencing for preventing drowning of children was published more than 20 years ago and the results are so clear that no further updates are required),2 or where no new evidence is being generated. But it is not acceptable where there is a volatile or rapidly evolving evidence base.

Similarly, current methods for evidence-based guideline development are robust and thorough, but are often sluggish and costly. For example, consider the use of opioids for osteoarthritis.

Evidence regarding the balance of benefits and harms of opioids in this clinical setting has been available for several years: the 2014 Cochrane review of opioids for osteoarthritis of the knee or hip3 did not find a clinically-relevant effect on pain, and demonstrated an increased risk of adverse events. And yet, changing practice has been akin to turning the Titanic.

Currently, nearly 10% of all opioids prescribed in Australian general practice are for osteoarthritis pain – equivalent to approximately one million opioid prescriptions for osteoarthritis in Australia in the 2015-16 year, at a direct cost to the healthcare system of approximately $25 million.4

It is predicted that this will rise to 1.5 million opioid prescriptions by 2020-21, at a cost of over $36 million, not counting the enormous indirect cost of opioid-related adverse effects.5

The iceberg is still directly ahead.

So, could we have done better?

The recently published second edition of the RACGP Guideline for the management of knee and hip osteoarthritis6 contains a strong recommendation against offering oral or transdermal opioids for people with knee and/or hip osteoarthritis. However, the previous iteration of these guidelines, which was published in 2009 and remained current on the RACGP website until August 2018, was far more permissive, including a Grade A recommendation in favour of opioid use: “consider prescribing weak or strong opioids with caution for treating at least moderate or severe pain in people with OA of the hip or knee who have not responded to, or are unable to tolerate, other analgesic medications or NSAIDS, and in whom joint replacement surgery is contraindicated or delayed”.

It is clear there has been a large qualitative shift in recommendations around opioid prescribing between 2009 and 2018, but there has been no method for relaying this change in interpretation of the evidence to primary care prescribers in a dynamic way.

If practice had changed at the end of 2014 and resulted in a 20% decrease in prescriptions, we would have expected to avoid 200,000 opioid prescriptions in 2015-16, saving more than $5 million in direct costs, and potentially far more in both direct and indirect costs over the long-term.

What if the “ecosystem” for evidence summaries and clinical guidelines were more responsive to changes in the evidence base, tailored to the local practice environment, and integrated at the point of care?

This vision, based on the emerging field of “living evidence”, is one that is shared by ANZMUSC and a number of other key players in evidence-based medicine. Cochrane, in particular, has been at the forefront of developing novel technologies and methods for generating living systematic reviews which will be the first step in rapidly connecting new evidence with practice guidelines.

Rather than updating a published SLR on an ad hoc schedule, living reviews are designed to be continuously updated as relevant new evidence becomes available.

To quote:

“The concept of living evidence synthesis and related outputs, such as living guidelines, are of increasing interest to evidence producers, decision makers, guideline developers, funders and publishers, as a way to seamlessly connect evidence and practice.”


The best use of increasingly sophisticated technology in medicine is to automate repetitive tasks and free human users to spend more time on contextual decision-making, which is what humans still do better than machines.

Imagine a future in which machine learning and other automated systems were providing continual background surveillance of the medical literature (and perhaps other sources – for example, twitter or online health consumer forums) and feeding new evidence into existing evidence summaries.

Ideally, these living systematic reviews would be hosted on a stable and reputable platform (it is no coincidence that Cochrane is leading the way on this initiative) and would be freely accessible.

The next step would be to establish agile, living guidelines panels within a particular local context (Australia and New Zealand, for example, or APLAR) that would spring into action when the evidence linked to a particular recommendation within a guideline framework is updated, and make a rapid decision about whether or not to update that specific recommendation, based on the new evidence and the local clinical practice environment (including consumer preferences, costs and acccessibility).

Ideally, these guidelines would be easily accessible at the point of care, and any updates to individual recommendations within such a guideline structure would be pushed out to the user as soon as they are updated. That way, clinical practice guidelines become useful, dynamic, evidence-based, locally-tailored aids to clinical decision-making that can assist practitioners and consumers at the point of care to make an optimal shared decision about treatment.

Of course, none of these ideas are necessarily new. In 1753 James Lind, in tackling the scourge of scurvy, recognised both the importance and the scope of the project to connect evidence with practice:

“But as it is no easy matter to root out old prejudices, or to overturn opinions which have acquired an establishment by time, custom, and great authorities; it became therefore requisite for this purpose, to exhibit a full and impartial view of what had hitherto been published on the scurvy.”

Indeed, before this subject could be set in a clear and proper light, it was necessary to remove a great deal of rubbish.

Thus, what was first intended as a short paper to be published in the memoirs of our medical naval-society, has now swelled to a volume. 7

Clearly this is a huge project. It is a global effort with many players.

There is an opportunity for Australia to be a leader in this effort, and ANZMUSC is committed to promoting the position of musculoskeletal diseases within this framework.

We have now registered two protocols for Cochrane SLRs – autologous blood product injections and stem cell injections for knee osteoarthritis – which will serve as pilot projects for the larger endeavour.

The next five years and beyond promise to be an exciting time for evidence systems in musculoskeletal medicine.

Dr Samuel Whittle is a senior consultant rheumatologist at The Queen Elizabeth Hospital, and is a 2018-20 ANZMUSC Practitioner Fellow, researching living evidence


1. Clarke M, Chalmers I. BMJ Evidence-Based Medicine Epub ahead of print. doi:10.1136/bmjebm-2018-110968

2. Thompson DC, Rivara F. Pool fencing for preventing drowning of children. Cochrane Database of Systematic Reviews 1998, Issue 1. Art. No.: CD001047. DOI: 10.1002/14651858.CD001047.

3. da Costa BR, Nüesch E, Kasteler R, Husni E, Welch V, Rutjes AWS, et al. Oral or transdermal opioids for osteoarthritis of the knee or hip. Cochrane Database Syst Rev. 2014; CD003115.

4. Harrison CM, Charles J, Henderson J, Britt H. Opioid prescribing in Australian general practice. Med J Aust. 2012;196: 380–381.

5. Ackerman IN, Zomer E, Gilmartin-Thomas JF-M, Liew D. Forecasting the future burden of opioids for osteoarthritis. Osteoarthritis Cartilage. Elsevier; 2018;26: 350–355.

6. The Royal Australian College of General Practitioners. Guideline for the management of knee and hip osteoarthritis. 2nd edn. East Melbourne, Vic: RACGP, 2018.

7.  Lind J. 1753. A treatise on the scurvy. In: Three parts, containing an inquiry into the nature, causes, an cure, of that disease, together with a critical and chronological view of what has been published on the subject. Edinburgh: Printed by Sands, Murray and Cochran for A Kincaid and A Donaldson.

End of content

No more pages to load

Log In Register ×