Perceptions regarding antibiotic use may require reevaluation

Risk for antibiotic resistance as an adverse event may be underestimated.

Issue: August 2008

BETHESDA, Md. — Antibiotic overprescription will continue to be a problem if public perceptions about these medications do not change, according to a speaker at the 2008 Annual Conference on Antimicrobial Resistance.

“There is an overestimation of benefits of antibiotics for some illnesses,” John Powers, MD, assistant professor of medicine at George Washington University, told Infectious Disease News.

“People often talk about the tremendous benefits of antibiotics, which is true for some diseases, but they rarely mention the adverse events. There seems to be the attitude that antibiotics are highly effective — period. That statement is not entirely accurate. Antimicrobials are highly effective in serious, life-threatening diseases like meningitis or pneumonia but their effectiveness in self-resolving diseases is still unclear. But very real risks such as skin rashes, allergic reactions, even liver failure have been greatly underplayed, especially when effectiveness is not clearly demonstrated.”

Powers went on to address the idea of resistance as an adverse event. “Resistance is unique in that it is not just an individual problem but a societal problem because it can be passed on to other people. Most people view risk as how it applies to the individual and resistance as a societal issue. They see the two as completely unrelated,” Powers said. “This is not the case at all. Resistance starts at the individual level.”

Evaluation methods

Powers indicated that it is not only misuse of antibiotics that is to blame for antibiotic resistance.

The standard by which the drugs are evaluated requires attention as well, Powers said.

“The heart of the problem is the concept of the noninferiority trial,” Powers said. “We have come to accept the idea that if a new drug is similar to an old one, it will be useful because it might be superior in effectiveness in some situations where it has not been studied. This is based on assumptions, not data. We have gotten into a pattern of being satisfied with showing that the drug is maybe a little worse than current drugs, when what we really want to know is where is the drug better. We have become enamored with mediocrity.”

He addressed the subjectivity and variability of trial standards.

“One problem is that clinical trials involving antimicrobials are relatively small. Studies for cardiac drugs, for example, can involve thousands of participants, whereas studies for infectious disease drugs often involve hundreds or even just dozens,” Powers said. “Fewer people studied means more outstanding questions at the time a drug is approved.”

Powers also addressed evaluation methods for outcomes. “If you just ask doctors to express their opinions about what patients are experiencing, you get tremendous variability and inaccuracy in responses,” he said. “Clinical trials should use standardized measurements that evaluate outcomes that are meaningful for patients, not just whether they have a fever or not.”

Powers discussed the timing of evaluation of outcomes in studies for infectious disease drugs as well. “The timing of our evaluation in these studies is far beyond the time when a person is going to get better anyway,” he said “For example, in trials involving ear infections in children, the data show that most children will get better in three or four days whether they take an antibiotic or not. The timing of the outcome assessment according to the FDA’s 1998 guidelines on otitis was day 21 after administration of the study drugs. If you know that on placebo 80% of children will be better in three days, you can’t measure the effect of a drug if you wait 21 days.”

According to Powers, a better way to assess drugs like these would be to evaluate the time until the symptoms are resolved and measure symptom resolution in a standardized way rather than leaving the decision of whether a person is cured up to clinician judgment.

Responsibility

Responsibility for improving the testing methods falls on the shoulders of all physicians; however, Powers suggested that the government needs to step in and spearhead the process.

“The FDA should be raising issues about noninferiority trials and requiring more stringent standards,” Powers said. “They seem thankfully to be moving in that direction.”

Powers warned of the dangers of becoming too focused on drug approval. “People in the drug companies do, in fact, want to reevaluate the endpoints,” he said. “But companies also want their drug approved right away, so they want to leave it up to the next guy who comes along to answer these questions. This leads to a vicious cycle of never making improvements in testing methods.”

Powers said that antibiotics often perform well in vitro and in animal studies and that they have the highest approval rate of any other class of drugs, which has led to misconceptions about the way antimicrobials are perceived.

But still only 28% of antibiotics that are submitted to FDA are eventually approved, showing that not all drugs that work in a test tube or in mice work in people.

Perception

The prevailing attitudes about risk and benefits of antibiotic use has led to overuse, he said.

“If you go to the doctor with a cough and say you have lung cancer, the doctor would not write a prescription for cancer chemotherapy without extensive diagnosis and testing,” Powers said. “But if you go to a doctor and say you have pneumonia, statistics show that you will walk out with an antibiotic up to 50% of the time. There is an obvious notion that cancer chemotherapy is toxic and that if you do not have cancer, you will not benefit, but there is not this notion surrounding antibiotics.”

Powers noted the perception is that antibiotics can help if you have pneumonia but even if you have a cold; you might as well take antibiotics anyway because they cannot hurt you. “That perception is not supported by the evidence,” he said. “Direct adverse events and antibiotic resistance are important public health problems for individuals and society at large.”

Powers addressed the pervasiveness of this attitude within the clinical community. “Study results show that the longer you have been in practice, the more likely you are to inappropriately prescribe antimicrobials,” he said.

Perhaps a more dangerous perception is the one that all bacteria are harmful – a perception that, Powers indicated, is changing. However, the evolution of the understanding of the symbiotic relationship human cells have with bacterial cells needs to find its way to the clinic, he said. “We need to reach a point where we are leaving the good bugs alone,” Powers said. “We need to figure out how to work from the idea that the only time most bugs are bad is when they get in the wrong place at the wrong time. And when we treat an infection we want to leave the good bugs alone.”

In the end, he encouraged physicians to let the history of antimicrobial drugs inform the way they are viewed in the future.

“We must remember that scientists and physicians 70 years ago, when antibiotics were first invented, were worried most about the adverse events that they caused in patients,” Powers said. “Somehow, those perceptions have been completely flipped.” “Hippocrates said, ‘first do no harm’, not ‘shoot first and ask questions later.’” – by Rob Volansky

For more information:

Powers J. Evaluating the human public health impact of in vitro resistance.

Powers J. Antibiotic resistance as an adverse event and risk perception.

Both presented at: the 2008 Annual Conference on Antimicrobial Resistance; June 23-25, 2008; Bethesda, Md.