well on corticosteroids alone and drugs like cyclophosphamide should be used only for more severe disease, for example when there is involvement of heart or kidneys. About 80 percent of patients survive five years or more after diagnosis, but there is a wide range of disease severity that affects the prognosis. Kidney involvement and vas-culitis affecting the heart or bowels indicate more severe disease and a potentially worse outcome.
clinical trial An experiment performed on humans to determine the safety and efficacy of a treatment. The clinical trial is a relatively recent and important scientific advance that has allowed scientists to measure and compare responses to a treatment objectively. Before clinical trials became the standard method for evaluating new treatments, these were introduced into clinical practice empirically, using a trial and error approach, and were then propagated among practitioners based on word of mouth and anecdotal experience. This situation is similar to the approach currently used to evaluate many herbal and other nontraditional therapies that are not classified as drugs and therefore do not require proof of efficacy and safety before they are marketed.
unscientific, anecdotal approaches to the evaluation of new therapies led to the introduction of many ineffective or toxic drugs. The controlled clinical trial with its unbiased design and objective measurements revolutionized modern medicine. Anecdotal evidence is based on the responses of one or a few patients and is a seriously flawed method for evaluating a treatment. Anecdotal evidence is similar to testimonials from customers— there is a bias to report only the positive responses. A second major problem with the anecdotal approach to evaluating a treatment is that there is a natural tendency for people taking a new treatment to feel better, even if the treatment is an ineffective dummy called a placebo. This positive response to a new treatment, whatever it is, is called a placebo response. Clinical trial design has evolved to overcome these problems. There are several different types of clinical trials.
Uncontrolled, placebo-control, and active-control trials if a new treatment is studied alone, with no comparison group, it is called an uncontrolled study. uncontrolled studies are difficult to interpret because telling if any response is due to a true drug effect or to the placebo response is impossible. uncontrolled studies, although they can provide useful preliminary information, are not strong evidence in favor of a new treatment, even if the results of the trial are positive.
To decide if a treatment is effective, it is best to compare it with another treatment, and this is called a controlled trial. Controlled studies usually compare the new treatment against an inactive substance, like a dummy sugar pill, called a placebo control, or else against some treatment that is known to be effective, called an active control. Comparing the results of the new treatment with the results of the control treatment makes it much easier to determine if the new treatment is effective.
A placebo-controlled study is generally more efficient in finding out whether a new treatment is effective, but for ethical reasons this design is not always appropriate. Most medical ethicists agree that it is not ethical to treat a patient in a clinical trial with a placebo if an effective treatment is available for that particular condition and if treatment with placebo could result in irreversible, preventable problems for a patient. For example, it is generally considered ethical to evaluate a new drug for pain caused by osteoarthritis against placebo, provided patients have access to adequate pain control if their treatment fails. However, if active rheumatoid arthritis (RA) is not adequately treated for several months it can result in irreversible joint erosions. Therefore, determining how to perform ethical, placebo-controlled studies in RA is more difficult.
Randomized trials if the researchers in charge of a study could choose to allocate patients to either the treatment or the control groups, there could be a tendency for them to allocate those with the worst problems into one or another of the groups. Then, because the patients in one group were sicker, that group could have a worse outcome, even if their treatment was effective. To prevent this happening, everyone enrolled into a study should have an equal chance of being allocated to receive the new treatment or the control treatment—this is called a randomized study. Patients can be randomized into groups in different ways. One is a simple flip of a coin, but generally a computer generates random numbers.
Was this article helpful?