Help in evaluating medical news

When a big story about genetics hits the news, it is usually based on an article published in a medical journal by research scientists. If you want to learn the basic facts behind the story, which may have been sensationalized or otherwise distorted by the media, it is best to find the original journal article on which all the 'buzz' is based.

Get the original journal source

Medline is a massive database which catalogues all articles published in the major medical journals. The Medline database can be accessed via 'Webmedline'.

Webmedline allows you to search for a specific article by entering information such as the authors, the journal, and keywords. In addition, you can choose to search for articles published within a certain time period, for example 90 days or one year. Often, you will be able to find some of this information, such as the authors' names or the journal name, from the news source. When searching for an article in Webmedline, enter as much information as possible, to narrow the search. However, if your search does not provide the correct article, you may have to make it less specific, because the search engine is particular about how names of authors and journals are typed.

Your search will provide you with a reference, including the title of the article, the journal it was published in, and the authors. Along with this reference will be a link to the 'abstract' which is an informative summary of the article. If you would like the complete article, you can look for it online at the journal's website (although most sites do not offer the full article unless you are a paid subscriber) or go to the medical library at a local university or hospital. However, be warned that medical journal articles can be quite difficult to read and understand unless you know a lot about the field. For some help interpreting an abstract or article, read the following section.

Help interpreting an original source

Any article published in a major scientific journal is first peer-reviewed, meaning that it is critiqued and edited by other researchers in the field before publication. Even with this review process, many results are later refuted or ammended by fellow scientists or the original authors. In other words, it is very difficult to understand and evaluate the significance of a research development. Even trained researchers do not always agree on the meaning of scientific data. Therefore, it would be impossible to teach you, in the space of a few computer screens, how to fully understand medical literature. However, there are some important concepts which will help you grasp the basic meaning and significance of an article.

Experimental system - Research, even when it is related to human health, is not always done in people, due to the many complexities and limitations. While research done in other animals may also apply to humans, the animal data alone is not suficient proof of relevance to people. In addition, some studies are conducted on human cells or tissues, grown in the laboratory. Results in cells or tissues may apply to people, but they require further proof in human trials.

Sample size - To consider a scientific finding valid, rather than a chance happening, it must be able to be repeated. 'Sample size' refers to the number of cases or people that were involved in the study. The sample size considered necessary for proof varies according to the conditions of the study. For example, the sample size in animal studies can sometimes be fairly small (for example, 10 or 100), because the conditions are very carefully regulated. On the other hand, studies in humans often require sample sizes of 1000s, because there are so many biological and environmental factors which vary between people, that it is hard to establish clear data. There are statistical calculations which can be used to determine whether a result in a certain sample size is significant or not. Some of these calculations result in a 'p-value' which represents the probability of a finding occuring by chance. In general, a p-value of less that .05 (5 percent) is considered significant evidence that the finding is valid.

Clinical trials - Before medical treatments are approved for general use in humans, they are tested in the laboratory or in very small samples (pilot trials) and must then be evaluated in formal clinical trials. There are four phases of trials (1 to 4), with increasingly rigorous criteria. New medical treatments must pass Phase 3 clinical trials to be approved for use. Phase 4 trials are conducted for additional information on safety and efficacy. When results are reported for Phase 1 and 2 clinical trials, the treatments are generally still a long way (months or years) from being generally available. In addition, clinical trials sometimes reveal flaws in treatments, such as side-effects or reduced efficacy, and many treatments which are tested in early clinical trials are never approved for use. Therefore, you should approach results of Phase 1 and 2 clinical trials with hesitation. Phase 3 clinical trials are conducted with large sample sizes, and the results can be considered more reliable, as the treatment has passed through multiple rigourous tests. Patients can sometimes have access to experimental treatments by participating in clinical trials, but they should be aware that the protocols are not yet proven or fully studied.

Related articles - Scientific research is not... well, an exact science! As described above, a published study must show reproducible results for a sufficient sample size. In addition, there are multiple groups of scientists studying related topics. Often, these groups publish contradictory reports about a particular topic, such as the contribution of a particular allele to disease or the frequency of particular alleles in the population. This happens, despite diligent application of good research techniques, because of the great complexity of biological systems. Therefore, to more fully understand the state of research in any field, you will need to read multiple articles.

To find multiple articles related to a topic of interest, you can search the database of journal articles available through Webmedline. Perform a search by entering some descriptive terms related to your topic into the 'keywords' field. You may have to try a few combinations of words to get a specific and helpful result (more than one article and less than 50 is a good place to start). When you find an article which seems to address precisely the question you are interested in, you can also click on 'related articles' for findings directly related to that result.

Reviews - When a number of studies have been published addressing a similar biological question, reviews on that topic are sometimes written. Reviews evaluate all the data in a field and attempt to synthesize the results to form a conclusion. While they do not include detailed information on techniques and individual data, reviews are an excellent resource for everyday readers, because they are usually written in a style that is more accessible, providing more background information and explanations than most journal articles. You can search Webmedline specifically for reviews by specifying the 'Publication Type'.

Vocabulary - There is an extensive dictionary of terms used in genetics research available via The Genomics Lexicon.

Note: Human genetics is an extremely complex topic. This website is meant only as an introduction and overview. If you are concerned about how genetics may affect your health, always consult your physician.