How often have you heard the results of a medical study and wondered, “What does this mean for me or my loved one?” It can be hard to decipher the impact of research findings from an news article or broadcast.

To learn more, you may want to read the original journal article or even just its abstract, a condensed description of the study.

To find a journal article or abstract, check the journal’s website. Many journals charge for the full article, but offer free access to the abstracts. You also can search the National Library of Medicine database PubMed, or check with nearby public, hospital, or university libraries.

Journal articles typically include the following sections:

• Abstract -  a brief summary of the article and results
• Introduction - an explanation of what the authors were studying and why
• Methodology - how the study was conducted, what was measured, who the participants were and how the data was analyzed
• Results - what the authors found at the conclusion of the study
• Discussion - an interpretation of the results that puts the findings in perspective as they relate to previous work in this area

Look closely

Ask the following when evaluating a study:

What type of study was done?

• Randomized controlled trials are considered the “gold standard” of study designs. Participants are randomly assigned to either a treatment or “control” group.
  For example, in testing the effectiveness of a new drug, the treatment group would get a pill with the active drug, while the control group would receive one of the following: 
  
                 1) a placebo (i.e., an inactive pill)
                 2) standard of care treatment; or
                 3) different dose of the active drug
  
  The treatment and control groups should be similar with regard to all characteristics – demographic (e.g., age, sex, race and education) and clinical (e.g., disease stage, tumor size and tumor markers) – except for the study medication.
  
  
• Cohort studies involve two or more groups who have different exposures to some identified factor, such as a gene mutation, and are followed for a period of time to see how many in each group develop disease.
  
  
• Case-control studies select two groups based on their difference in disease status and investigate the occurrence of certain exposures. For instance, a case-control study may include women with breast cancer (the cases) and those without breast cancer (the controls) – and analyze the BRCA1 mutation or a family history of breast cancer in both groups of participants.
  
  
• Cross-sectional surveys collect data at one point in time, but may ask questions that refer to participants’ current and past experiences.

How long did the study monitor participants and was the follow-up comprehensive? 

A study should continue long enough to show some effect of the intervention, but the actual length of time can vary. For example, a clinical trial testing a new drug to prevent nausea and vomiting associated with chemotherapy may only need to follow patients for several days, whereas a trial studying whether vitamins reduce a person’s risk of cancer would require years of monitoring, possibly even decades.

Typically 70 to 80 percent of participants should be followed through the end of the study. If too many people withdraw from the study, the results can be biased. Participants who withdraw from a research study are more likely to have missed medical visits, more likely to have experienced adverse effects, and less likely to have taken a study medication as directed than those who continue to participate in the study. To be valid, a study should have similar follow-up rates for all groups being compared.

If the study examined a drug or medical procedure, were there any side effects?

It is important to note whether side effects were mild, for instance dry mouth, nausea, and diarrhea, or more severe. Additionally, some side effects may aggravate co-existing diseases.

Was the sample size large enough?

A study should include enough people so that it can detect whether an effect, or lack of an effect, was simply due to chance. Smaller studies can also be valuable and provide accurate data; however additional testing is needed to confirm or refute the findings.

Who was excluded?

Many clinical trials exclude patients who have co-existing diseases and take certain medications. Other trials are restricted to people with severe forms of a disease and thus make it difficult to draw conclusions about use of the therapy in milder forms of the disease. Keep in mind that if you do not possess similar characteristics to the individuals in the study, the results may not apply to you.

Where was the research done?

Research conducted at a high volume hospital or medical school may be able to conduct more complex procedures and have more experienced researchers in the field. The research may have been conducted at several different locations; this information can be found in the journal article or abstract following the list of authors.

Who funded the research and published the article?

Consider the study’s financial support, and whether the main sponsor may gain financially from the results. This does not mean that the study is biased; but that the funding source should be among many factors considered when evaluating the study. Funding information is usually in a footnote on the first page or in the acknowledgments at the end of the article.

Check to see whether articles published in the journal are subjected to a “peer review” process. Peer reviewed studies are evaluated by experts who are not affiliated with the study. These reviewers are considered to be unbiased and the study results are believed to be as balanced and accurate as possible.