Barry Goldlist, MD, FRCPC
Editor in Chief,
Geriatrics & Aging.
We are now definitely in the era of evidence-based medicine. At my teaching hospital, the residents even present a series of noontime rounds that are called 'evidence-based rounds'. If nothing else, these rounds provide definitive evidence that providing a 'free' lunch improves attendance. Add MainCert accreditation and we can get standing room only crowds!
The randomized clinical trials that we read about in peer-reviewed journals are organized in such a way as to maximize the chances of obtaining a positive result. Physicians follow the patients more closely than is usual in normal clinical practice, and often an army of research assistants helps to ensure compliance and meticulously document outcomes. Often, the placebo patients in clinical trials show better results than did those patients who were even relatively recent historical controls. After a rigorous randomized clinical trial, we are often convinced that the treatment is efficacious but are wary about its effectiveness when generalized into normal practice. As well, patients enrolled in clinical trials are often highly selected and might not have the same outcomes as the patients who arrive in our offices. Older persons are generally more heterogeneous than young people, so the issue of patient selection is particularly acute when interpreting the results of trials in elderly patients.
In this issue, Dr. Denis DeSilvey questions the use of thrombolytic agents in the very elderly. Although I do not have access to his references, there is no doubt that others share his view.1,2 Classic teaching from clinical trials has stressed that a similar relative risk reduction will result in more absolute benefit for elderly patients because of their higher baseline mortality and morbidity. This has clearly not been shown, however, when information from large databases in regular practice is analyzed. What should we do as clinicians? I would refer the cautious reader to the excellent commentary that accompanied reference 1 in the May 1, 2001 edition of the CMAJ.3 The authors take a balanced view of the data and, while admitting that further trials are required, make a convincing argument that thrombolytic agents should be still be used in the elderly, albeit with care and judgement.
Does this mean that the era of randomized clinical trials is over? Those who read the two articles in last year's NEJM showing the general concordance of observational studies and randomized trials might wonder. For those doubters of the 'gold standard' of randomized clinical trials, I refer you to a recent editorial in the BMJ.4 The truth is that observational studies and randomized trials should be considered complementary, rather than competitive. Observational studies are weaker at determining causality but better at determining rare, yet important, outcomes. Randomized trials are most often used in studies where the postulated treatment effects are relatively modest. In addition, in instances where observational studies have shown either large harmful effects or large beneficial effects, the use of randomized trials raises some thorny ethical issues.4
The rest of this month's edition of Geriatrics & Aging is just as thought provoking, if somewhat less controversial. Jane Oshinowo writes about pharmacological and non-pharmacological primary prevention of cardiac disease, while Dr. David Spence carefully dissects the available evidence for the treatment of hypertension. There are also reports on newer NSAIDs, andropause, psychotic disorders in the elderly, minimally invasive surgery and breast cancer. We also have a report from the Rotman Conference, truly a world class Canadian event. Cynthia Jackevicius has written an article on how new information filters into clinical practice. I think from this editorial you might say that the answer should be 'carefully'! Enjoy this issue.
References
- Boucher JM et al. Age-Related differences in in-hospital mortality and the use of thrombolytic therapy for acute myocardial infarction. CMAJ May 1, 2001;164(9):1285-90.
- Ayanian JZ et al. Thrombolytic Therapy for patients with myocardial infarction who are older than 75 years: Do the risks outweigh the benefits? Circulation 2000;101:2224-6.
- Thierman DR, Schulman SP. Thrombolytics in elderly patients: A triumph of hope over experience? CMAJ May 1, 2001;164(9):1301-03.
- Ioannidis JPA et al. Any casualties in the clash of randomised and observational evidence? BMJ 2001;322:879-80.
