washingtonpost.com  > Health > Condition Center > Heart Conditions

Levels of Medical Evidence: How Much Confidence Exists in the Data?

Tuesday, April 19, 2005; Page HE06

Well-designed clinical trials constitute the most scientifically rigorous tests of medical theories and treatments. But medical evidence comes in more than one form.

While less conclusive than clinical trials, these other types of studies do provide valuable information. Sometimes, they are essential in shedding light on questions that -- because of ethical considerations or the difficulty in getting a large enough pool of participants -- don't lend themselves to clinical trials.

_____The Heart_____
Hearts in the Right Place (The Washington Post, Apr 5, 2005)
QUICK STUDY : A weekly digest of new research on major health topics (The Washington Post, Apr 5, 2005)
A Weekly Shot of News and Notes (The Washington Post, Apr 5, 2005)
How Can You Rehab a Broken Heart? Exercise, Diet, Stress Reduction (The Washington Post, Apr 5, 2005)
Hold the Nitro? (The Washington Post, Mar 29, 2005)
More Heart News

Here, in ascending order of reliability, are some common kinds of studies:

A case report is a doctor's detailed account of a patient with a medical complaint or illness. Several such reports of patients with like symptoms constitute a case series.

• An observational study observes and reports patient outcomes without offering any treatment or intervention. The most famous, and longest running, began in 1948 when the U.S. Public Health Service recruited more than 5,000 adults in Framingham, Mass., for a study of coronary heart disease. Over the years, these people have been weighed, poked, questioned and bled; the results of these investigations have changed our view of heart disease. Because of associations that Framingham researchers have found between certain characteristics and disease, we have been advised to exercise, eat less, watch our cholesterol intake, stop smoking and pay attention to our genetic history. Observational research includes case-control and cohort studies.

*A case-control study compares data on a group of people who have a particular condition (the cases) with data on people who don't (the controls). Correlations may suggest a stronger connection than exists between two unrelated sets of data.

*A cohort study follows a group of people (a cohort) who have a particular condition or receive a prescribed treatment over a period of time, comparing outcomes for that group with those for similar people who lack the condition or don't take the treatment. Researcher bias may affect outcomes, as may variations in the groups.

A controlled clinical trial tests a suspected cause-and-effect relationship by assigning participants to a treatment group or a control group in a way that makes an individual's assignment known to the researchers.

A double-blind, randomized, controlled clinical trial, the gold standard of research studies, tries to correct for bias by assigning participants by chance (or randomly) to either a treatment group or a control group; by including a comparison, or control, group that receives a placebo (a sham treatment), another treatment or no treatment at all; and by making sure neither the participants nor the researchers know to which group a person has been assigned until the study's end (double-blinding).

-- Susan Morse

Sources: www.clinicaltrials.gov, www.cancer.gov, www.ahrq.gov, "Epidemiology in Medicine, by Charles H. Hennekens and Julie E. Buring

© 2005 The Washington Post Company


Clinical Trials Center

  •  Cosmetic & Beauty Services

  •  Hospitals & Clinics

  •  Men's Health Care

  •  Women's Health Care