Study: Placebo is Safe, Ethical for Patients in Short-term Studies of Hypertension Drugs
DURHAM, N.C. - The controversial practice of studying new drugs for hypertension by comparing them against an inactive "placebo" is ethical and safe in short-term studies, even if it means withholding effective treatment for a short time, according to a study conducted at Duke University Medical Center.
Results of the quantitative review, called a meta-analysis, are published in the June 15 issue of the journal Science.
In fact, the general practice of using placebo to test the effectiveness of new drugs is an acceptable and useful research method under the right conditions, said Dr. Robert Califf, Director of the Duke Clinical Research Institute and one of the study authors. Other authors include Dr. Sana Al-Khatib, Dr. Jeremy Sugarman, Dr. Vic Hasselblad, Dr. John Alexander and Dr. Douglas McCrory, all from Duke.
Specifically, clinical trial investigators must be able to show that withholding active therapy from patients for a short time is unlikely to result in physical harm; patients must give voluntary, informed consent; and investigators must closely monitor patients in these studies, said Al-Khatib, lead author of the study.
In a retrospective review of 25 hypertension studies conducted in the late 1990s, the researchers found that there were no more adverse health effects among patients who received placebo than there were among patients receiving active treatments. Adverse events among the 6,409 patients included death, stroke, myocardial infarction and congestive heart failure.
The results are particularly relevant in the midst of what that the researchers call a heated debate over the acceptability of using placebo in clinical trials instead of using established and proven therapies as a "control." A "control" is the old drug or placebo against which a new drug is being tested to see if the new drug is safe and effective.
Opponents of placebo use in research claim it is dangerous and unethical to withhold effective treatments for diseases where effective therapies exist. In particular, they say, hypertension is so prevalent a disease and has been so clearly linked to stroke, coronary artery disease and premature death that withholding treatment by giving placebo instead of active therapy is potentially life-threatening, according to the article.
Placebo proponents, on the other hand, say that placebo use, under certain conditions, is safe and provides important information that drug-to-drug comparisons simply cannot.
Most notably, says Califf, testing a new drug against placebo produces a purer, more accurate assessment of the drug's effectiveness, because the drug is being compared to no active treatment at all. In contrast, drug-to-drug comparisons rely on previously-collected data about the existing drug -- data that could be erroneous or may not apply to the current study conditions and patient population. If data about the existing drug is not well-established or relevant to the current study, then comparing it to the new drug will not produce accurate and usable results.
The Duke researchers also point out that studies comparing placebo to an experimental therapy generally requires a smaller sample size to attain statistical significance than do studies comparing the experimental therapy to another treatment. As a result, trials may be conducted faster and at a lower cost, exposing fewer patients to the potential risks of the experimental therapy.
Third, if researchers use placebo only when it is likely to be safe to do so, then health risks to patient-subjects participating in trials are generally minimal, say the study authors. And any clinical trial, using placebo or established drugs, carries some risk with it, they said.
In the 25 hypertension studies reviewed, the study investigators were careful to abide by safety measures that would not expose subjects to undue risk, said Sugarman. First, each of the combined studies were short in duration. With such limited exposure to placebo, the risk of serious adverse events was quite low. Second, only patients with mild to moderate hypertension, either with no history of significant medical illness or with no recent history of myocardial infarction or stroke, were enrolled in these studies. Third, patient-subjects were closely monitored during these studies, and they were withdrawn if they had persistently elevated blood pressure or were determined to be at risk of serious harm.
"Thus, when patient-subjects are carefully selected and adequately monitored, limited exposure to placebo in clinical trials of hypertension is not associated with increased risk of serious adverse events," the authors wrote. "Using placebo controls in such trials, therefore, would seem ethically permissible."
Finally, say the authors, when it comes to decision-making regarding the use of placebos in clinical research, general edicts should give way to a more nuanced consideration of their true risks, benefits and appropriate use. "In this way, the rights and interests of patient-subjects enrolled in research will be protected while useable scientific data are gathered."
The study was supported by the Duke Clinical Research Institute.