There is a common belief that medicine in all it variety would be better practised according to strict rules and guidelines.
Indisputably, highly technical aspects of medical and surgical care – equivalent to piloting a plane – can benefit from checklists and rules. But for the majority of medical care this does not apply.
For three years in the 90s I chaired the Health Advisory Committee of the NHMRC. Among our responsibilities was the oversight of the development of clinical practice guidelines, CPGs, for all manner of conditions. We also revised a document – ‘guideline for guidelines’ – setting out ways in which available data can be assembled to produce a CPG.
This task had the capacity to swamp our agenda with multiple rounds of consultation and expert advisory panels. I learned much about guidelines and the politics that surround them.
As Tamara Kredo and colleagues from the South African Cochrane Centre reminded us (Int.J.Qual.Health Care 2016 28(1):122-128) the US Institute of Medicine defined CPGs in 2011 as ‘statements that include recommendations intended to optimise patient care that are informed by a systematic review of evidence and an assessment of the benefits and harms of alternative care options.’
They go on to observe that ‘many hundreds [of] health departments, insurers and other health care organizations, professional associations, hospitals, specialty colleges and individuals have attempted to produce recommendations to improve and/or standardize local clinical practices, all using their own interpretations of the best way to construct and write CPGs.’
Cameron Venus and Euzebiusz Jamrozik from Melbourne University, showed recently that many guidelines in common use in Australia don’t have rock-solid evidential bases. https://doi.org/10.1111/imj.14466 In fact, of the guidelines they examined in relation to the diagnosis and treatment of the ten leading causes of death in Australia, less than 20% had top-level evidence supporting the recommendations.
The presence of quality evidence ranged from none in the CPGs for prostate cancer testing, through 5% for colorectal cancer, up to 60% for management of acute coronary syndrome and 40% for chronic obstructive pulmonary disease.
We should not be surprised by this finding. There is much more to CPGs, as there is for health policy in general, than research evidence.
One of the irreducible fundamentals of clinical practice is that the doctor must always cope with individual variation among his or her patients. The expression of a disease may vary, as may individual resilience and defence.
Yes, evidence is indispensable and the higher its quality the better. But there will always be a gap between the applicability of evidence derived from groups to the individual patient sitting in front of you.
In the original definition of evidence-based medicine (EBM), David Sackett – the American-Canadian physician who was the master engineer of EBM – insisted that it included the “judicious use” of evidence in clinical practice, “integrating the best available research evidence with clinical expertise and the patient’s unique values and circumstances” BMJ 1996; 312 doi: https://doi.org/10.1136/bmj.312.7023.71 (Published 13 January 1996) Cite this as: BMJ 1996;312:71
No CPG can take full account of patient preferences or clinical or social realities that may be critically important in the management of the individual patient.
We have come a long way in the development of guidelines and there are were well over 300 fair quality CPGs in 2010 in Australia according to Heather Buchan and colleagues. Med J Aust 2010; 192 (9): 490-494. https://www.mja.com.au/journal/2010/192/9/australian-clinical-practice-guidelines-national-study, nine times as many as a decade previously.
But great care is needed to ensure that we don’t entirely lose clinical perspective amidst all the data, including that which descends on us from myriad scans and laboratory tests and clinical trials.
To see all patients as sitting on the mean in any statistical assessment is to miss the individual, in all their variation. As a statistician once said to me: “I’m not interested in the mean – the outliers are what fascinate me.”
In 2018, Kim Tingley, a science writer, wrote in The New York Times that as long ago as 1927 Francis Peabody, a professor at Harvard Medical School, stated that “the treatment of a disease may be entirely impersonal; the care of a patient must be completely personal”
Writing in the Journal of the American Medical Association, Peabody said: “The significance of the intimate personal relationship between physician and patient cannot be too strongly emphasised, for in an extraordinarily large number of cases both diagnosis and treatment are directly dependent on it.”
Unfortunately, guidelines fail to enshrine this wisdom.
And it’s up to us not to forget that the key element in that ‘personal relationship’ is the joyful recognition of the patient’s individuality.
Stephen Leeder is an emeritus professor of public health and community medicine at the University of Sydney. A version of this article appeared recently in Australian Doctor.