How to evaluate media reports about research

Text Size: Normal / Medium / Large
Printer-friendly versionPrinter-friendly version


adapted by Sari Tudiver

Watch the language used and read between the lines!

Much hype about "medical break-throughs" often goes along with media reporting of health research, some of it provided by the pharmaceutical manufacturers. Read critically to see what stage the research is at, what is known about the risks and side effects, as well as benefits.

Is this the first study to show a new finding or relationship?

The results of one study are of interest, but even a well-designed and well-executed study can produce misleading or unreliable results. It usually takes many studies and years of research before definite conclusions can be drawn.

Did the researchers study people like you?

Studies of groups of people who differ from your age, gender, health status, and ethnocultural background may not apply to you.

Was the population studied large enough?

In most cases, researchers need to study a large population over a long period of time to ensure that any observed differences are not simply the result of chance. Sample size should be appropriate to the research question asked.

How strong is the relationship and what is the impact?

Sometimes a research finding of excess risk is expressed as, "people who xxx are at a 100 per cent greater risk of developing cancer." Such a risk may sound much larger than it is. It is important to know something about the size of the baseline risk: if it is one in 1000, doubling the risk (or an increase of 100 per cent) means that the new risk is 2 in 1000, which may not be significant at an individual level.

What about research suggesting that certain conditions or agents cause cancer?

Does the study actually show that the condition or agent causes cancer or does it show an association with the disease? We need studies that follow a large number of cancer-free people over many years (some of whom are subjected to the condition and some of whom are not) in order to determine the number in each group who develop cancer. These prospective (forward looking) studies are more reliable than retrospective or observational studies.

What about studies claiming that certain drugs or therapies prevent cancer?

Do the studies show risk reduction or actual prevention? Demonstrating prevention of cancer involves long term follow-up, since cancers usually take 10-15 years to develop before they are identified. Some clinical trials make the assumption that the number of years are equivalent. However, following 5000 women for 1 year is not biologically equivalent to following 500 women for 10 years (although it is statistically equivalent).

What about animal or lab studies?

Studies using animals or cell cultures in the lab may or may not be relevant to humans; they often suggest important directions for research. Usually, if two studies show an agent is a cause of cancer in laboratory animals, the agent should be considered a potential cause of cancer in humans.

Who funded the study?

If a researcher has received funding from an independent granting agency that uses a peer review process for evaluating and awarding research funds, the research findings will usually be more credible and less subject to bias than if the funding agency has a vested interest in the results. With partnerships among industry, university-based researchers and institutes, this can be a difficult issue to assess.

Do the conclusions follow from the findings?

Do the researchers clearly state what can be known from this and earlier studies and what is still unknown? Do the findings make sense in terms of what is currently known about human biology and the nature of health and disease? Be open to explanations that do not fit existing beliefs or current knowledge because sometimes these are wrong or need modification.

Should you change your behaviour or habits because of research results?

Review the research from more than one study and try to assess the possible risks and benefits to you making a change. If necessary, seek out health providers or other resources to help with a decision. If you are considering being part of a clinical trial, make sure you are clear about your reasons and what is involved.

To Find Medical Research Information

A medical library at a university or hospital, or at a more specialized treatment centre, provides access to a wide range of medical journals and reports. The librarian can help you locate medical articles that meet your needs in print or online through Internet access.

If you live in an area with limited access to specialized resources, a community library might be able to arrange interlibrary loan, or Internet access to information. They can help you find a listing of organizations, and consumer health information services, including toll free numbers.

Sari Tudiver , PhD is editor of A Friend Indeed for women in the prime of life (AFI). The article first appeared in AFI and was adapted with permission from "Finding Out About Breast Cancer Research," a factsheet produced by Breast Cancer InfoLink, Prairies/NWT.

For more information, contact:

A Friend Indeed
Main Floor - 419 Graham Ave.
Winnipeg MB R3C 0M3 Canada
Tel: (204) 989-8028
Fax: (204) 989-8029

Breast Cancer InfoLink
Room AE 168E
1331-29 St. NW
Calgary AB T2N 4N2
Toll-free: 1-800-555-8698