In their interesting editorial Juni et al(1) rightly point out that the results of two recent observational studies suggesting cardiotoxicity of both COX2 inhibitors and conventional nonsteroidal antiinlfammatory drugs (NSAID) should be interpreted with caution. As noted by Cuervo and Clarke “Raising the alarm about a potential harm can also do more bad than good if the quality of the evidence or its reporting are poor.“(2) Indeed, what is often overlooked is the need of protection of the public´s health from both adverse effects of drugs and so-called “tomato effect“ also known as type 4 error.(3) This type of error is an overestimation of risks, which leads to rejection of an efficacious therapy.
The potential consequences of such risk in clinical praxis can be illustrated by recent case-controlled study which showed increased risk of acute myocardial infarction in patients who abruptly stopped taking NSAIDs after previous long-term use.(4) If this information is confirmed substantial decline of NSAIDs consumption may have a negative impact on cardiovascular mortality. In order to minimize such risks, reporting of harms from observational studies should be guided by similar standards as were developed for randomised controlled trials (RCT). Moreover, it may not be helpful to publish the results with high risk of bias and confounding, even if there is no better evidence.
Juni et al believe that in addition to the transparent reporting of harms we need a large RCTs like ALLHAT to assess more accurately current controversy regarding cardiovascular risk of NSAIDs. However, complex clinical outcomes are difficult to measure in megatrials and they may sometimes add more to the controversy than they resolve. (5) Hence, we should not neglect the power of observational studies, which may be preferable for identifying rare side effects and in many situations when RCTs would be impractical. (6)
1. Juni P, Reichenbach S, Egger M. COX 2 inhibitors, traditional NSAIDs, and the heart. BMJ 2005;330:1342-3.
2. Cuervo LG, Clarke M. Balancing benefits and harms in health care. BMJ 2003;327: 65-6.
3. Pijak MR, Gazdik F. COX-2 inhibitors and type 4 error. CMAJ 2003;169:190.
4. Fischer, L. M., Schlienger, R. G., Matter, C. M., Jick, H., Meier, C. R. Discontinuation of nonsteroidal anti-inflammatory drug therapy and risk of acute myocardial infarction. Arch Intern Med 2004;164: 2472-6.
5. Pijak MR, Gazdik F, Hrusovsky S. The best type of trial. CMAJ 2004;170:1772-3.
6. Concato J, Shah N, Horwitz RI. Randomized, controlled trials, observational studies, and the hierarchy of research designs. N Engl J Med 2000; 342:1887-92.
About the author:
Dr. Michal R. Pijak is a consultant in rheumatology, allergy and clinical immunology at the University Hospital in Bratislava, Slovakia
Circulated by Article Emporium
Heart Disease Symptom Articles Index
Prevention of “tomato effect“ in reporting of harms in observational studies