Taking the mystery out of medical research
Michelle Lynde, ClH
Are medical research studies a mystery to you? We show you how to read a research study to figure out whether it's a credible source of information.
New medical research studies on natural foods, herbal remedies, and supplements appear in the popular media almost daily. Media coverage of the latest medical marvel may grab our attention, but how can we be sure that these often frustrating snippets of publicity are credible and based on sound research?
Well, you don’t have to shoot the media messenger. Instead, a healthy dose of information about how new therapies are studied and published can help take the mystery out of medical research.
A Range of Studies
Basically, research is divided into two categories: observational research that includes case-control and cohort studies, and experimental research or clinical trials.
Case-control studies investigate the potential causes of disease, rare diseases, and serious side effects. Researchers follow two groups of people: one group (the cases) who have already developed the disease that is being studied; and another group (the controls) who have not contracted the disease. The medical histories of both groups are reviewed to ascertain differences in their experiences. The characteristics of the case group before the onset of the disease are then compared to the control group. Case-control studies are fast and relatively inexpensive, but to achieve unbiased results, both groups must have very similar traits, except for the disease itself.
Cohort studies observe groups of people before they develop a particular disease or outcome. For instance, researchers select one group of healthy individuals who are routinely exposed to second-hand cigarette smoke and a similar healthy group who are not exposed. Both groups are followed over a long period of time. The results are then measured to detect the different outcomes of the exposure, including calculating the risk of developing the disease, in this case, lung cancer. Cohort studies can be costly, and because of the time involved, participants may drop out of the study before completion.
Experimental research is generally conducted in phases. In preliminary phase I clinical trials, the treatment is administered to a small group of volunteers to evaluate its safety and establish an appropriate dosage. In phase II, the treatment’s safety is tested on a larger group of participants. Researchers document the efficacy of the treatment and any side effects. If the outcome looks promising, then the treatment proceeds to a full-scale phase III trial. These expensive and long-term clinical trials can involve thousands of participants and are considered the best method of determining if a treatment is safe and effective for human use as compared to other treatments, including placebo.
A phase III trial is commonly known as a randomized clinical trial (RCT). A well-designed RCT follows a detailed protocol that describes the research procedures and ethical issues and addresses the safeguards of the participants who will be involved. A computer program randomly assigns participants to an active treatment group or a control group. The control group receives a placebo, an inert preparation that is otherwise identical to the active treatment being tested. Randomization ensures that the trial results are accurate and unbiased. Blinding also helps minimize the chance of bias influencing the results. In a single-blind trial, participants do not know if they are getting the active treatment or placebo. In a double-blind study, neither the participants nor the researchers know who has been assigned to which study group.
Meta-analysis combines the results of independent studies to derive an overall conclusion. For example, when independent studies on the relationship between weight reduction and blood lipid levels showed inconsistent results, scientists pooled the data from over 70 similar studies and found significant decreases in total cholesterol and other blood lipids due to weight loss. Meta-analysis is not always perfect, however. Data from flawed studies or from studies that use different methods to measure variables may be included, resulting in a comparison of apples to oranges.
Parts of a Peer-Reviewed Article
Scientific journals require that the findings of research studies and clinical trials go through a peer-review process before publication. Experts who are not connected to the study examine the data and decide whether the trial followed protocol and has merit. Only the studies that pass peer review are published.
A peer-reviewed article is composed of six sections, beginning with the abstract, which briefly outlines the study, its results, and the conclusions. The reader can review the abstract to decide whether the study is of interest without reading the full text. Note that abstracts do not provide enough detail to properly assess the validity of the study.
The introduction section discusses the potential importance of the research and how the research will be conducted.
The methodology section explains in detail exactly how the research was carried out. Nobody will deny that reading a study up to this point can be hard work, but don’t despair; the best is yet to come. The answers, or what scientists like to call the statistical analyses of the data, are included in the results section. This section compares the participants to determine whether existing differences might have influenced the results. It also includes a descriptive analysis of the outcome data. Tables, charts, or graphs often accompany the outcome data, offering an at-a-glance view of the overall results of the study.
The discussion or conclusion section of an article is usually the most interesting. It is based on the author’s interpretation and opinion and can shed new light on the meaning of the results. Sometimes an alternative explanation of the results and its implications is also presented.
The article finishes with a reference list of other published papers that the author reviewed for background information. References also provide a good resource for exploring the topic further. However, references should not be more than 15 years old, unless the article is historical in nature.
Where to Read Scientific Studies
Large public and university libraries maintain subscriptions to respected scientific journals such as the New England Journal of Medicine, Canadian Medical Association Journal, and The Lancet. These and other journals are sometimes available on the journals’ websites free of charge or for a nominal fee. Some full-text medical journals can be found at
freemedicaljournals.com and findarticles.com. A wide variety of alternative medicine topics are available from the Townsend Letter for Doctors and Patients at townsendletter.com.
Another reliable tool for locating research is PubMed (ncbi.nlm.nih.gov/entrez), a free website provided by the US National Library of Medicine that covers a wide range of articles on conventional and veterinary medicine, life sciences, and complementary and alternative therapies. PubMed provides access to more than 4,000 journals, containing over 16 million entries or citations (the pertinent information needed to find the full text of a journal article). Knowing how to read a citation helps to locate it on a website or at a library.
What Makes a Study Relevant
The next time you read or hear about a new medical study, keep these questions in mind. Are the findings of the study relevant to you? Were the participants the same age, sex, and ethnic background, with the same health concerns? What was the size and duration of the study? Generally, a long-term study on large groups of participants gives more valid results. Is the purpose of the study clear and concise? If the study is a randomized clinical trial, are blinding procedures used? Are the conclusions consistent with the objectives of the study? Were treatment side effects found to be harmful? Who provided the funding and did they stand to gain or lose from the outcome of the study? Be skeptical of the results of a single study. The findings of one study often need to be duplicated by other researchers to further substantiate the results before being considered an acceptable medical treatment.
Far more published research is available today than was available a few years ago. Ongoing studies in medicine are crucial, especially in the fields of nutrition, herbal medicine, homeopathy, acupuncture, and other integrative therapies, if they are to be recognized and accepted by the scientific community and the public. Getting to the bottom of valid health-related information can be a challenge because many researchers frequently study the same topic in different ways, making the results appear contradictory and confusing. Further, scientists do not always agree on what constitutes scientific evidence in a study.
The key is a better understanding of how the world of research works. So take up the challenge; with a little practice, you can interpret the latest medical studies with a more discerning eye.