A study performed by two Drexel University School of Medicine doctors shows that most of the guidelines doctors use when determining how to treat infectious diseases are based on very limited evidence. More than half of the recommendations are based on evidence from individual case studies or on expert opinions. Few are based on evidence from clinical trials.

The manner in which infectious diseases are treated is based a lot more on guesswork than either patients or doctors would like it to be.

Case studies are in-depth analyses of an individual or group over a long period of time. While they can generate a great deal of information, they don't compare the effects of different types of treatments and so give a very low quality of evidence about which treatment(s) are best. Clinical trials, on the other hand, test how different treatments stack up against each other and are considered to give considerably better evidence about the effectiveness of a particular treatment.

One reason for the problem is that it's extremely difficult to conduct good clinical studies on infectious diseases. Many would be unethical because they would cause needless pain, suffering or death, with people going un-treated or undertreated just to test a treatment approach.

The time course and low occurrence of many infectious diseases cause added design problems. But patients would likely benefit if more clinical trials were conducted.

The researchers looked at 41 different guidelines set between 1994 and May 2010 by the Infectious Diseases Society of America. These included guidelines for treating Lyme disease, bacterial meningitis and community-acquired pneumonia. All told, these guidelines contained 4,218 individual treatment recommendations. The researchers set out to find out exactly what evidence each of these recommendations were based on.

Fourteen percent were based on evidence from at least one randomized clinical trial; 31 percent were based on evidence from one well-conducted study, even though it wasn't a randomized controlled trial. But more than half of the recommendations (55 percent) were based only on the opinions of individual experts or panels, or on case descriptions.

This says that the manner in which infectious diseases are treated is based a lot more on guesswork than either patients or doctors would like it to be.

For example, current guidelines for treating many bacterial infections generally call for antibiotics to be taken for seven days. If there was good evidence that three or five days of treatment was just as effective, patients would experience fewer side effects, save money and there would likely be a decrease in bacteria acquiring antibiotic resistance. But there currently is no such evidence.

On the brighter side, the researchers did find that newer guidelines tend to have more literature references than the older ones did, indicating that more studies are being done. And travel medicine guidelines tend to be backed up by good supporting evidence.

The researchers worry in particular about younger doctors looking at the guidelines as if they were written in stone, while they're only educated guesses.

Guidelines are generalized treatment recommendations, not laws. They could be thought of as tips for how to treat a patient in a vacuum. Individual cases require individualized treatment. Specific patient information may trump the recommendations called for by guidelines and suggest an entirely different type of treatment. Guidelines are not rules for how to treat an individual patient.

An article on the study appears in the January 10, 2011 issue of Archives of Internal Medicine.