Still Rx'ing Antibiotics for Bronchitis and Non–Group A Strep Sore Throat

A presentation at IDWeek 2013 on October 3, 2013, by Jeffrey A. Linder, MD, (Harvard Medical School and Brigham & Women’s Hospital, Boston) covered two sessions:

• Barnett M, Linder J. Antibiotic Prescribing for Adults With Sore Throat in the United States, 1997-2010. Abstract #962.

• Barnett M, Linder J. Antibiotic Prescribing for Adults With Acute Bronchitis in the United States, 1996-2010. Abstract #963. 

Overuse of antibiotics may seem to be primary care’s problem, but solving it will require a complete re-imagining of reimbursement for primary care. And we have a long way to go.

Harvard’s Jeffrey Linder, MD, presented data showing little progress in the war against unnecessary antibiotic use for routine viral upper respiratory tract infection (URI), and though it wasn’t the subject of his study, he divided blame between physician perception of patient demand (overestimated by physicians), and the time needed to explain why antibiotics aren’t needed (underestimated by our system). Without longer visits, radical realignment of incentives, and correction of physician and patient expectations, overuse of antibiotics will persist.

Dr Linder’s presentations at IDWeek 2013 underscore the enormity and unfortunately, the persistence of the problem. Since the late 1990s, we’ve made modest progress with sore throat, but none at all for acute bronchitis (which should have an antibiotic prescribing rate of 0%).

IDWeek 2013 is an infectious disease (ID) conference, so this was cause for much hand-wringing and doom saying, and not without reason. First, the bad news, and then, let’s think through what can be done about this-because without action, we’re going to see more and more C difficile colitis and highly resistant bacterial strains.

Let’s start with sore throat visits, where we’ve made modest progress. Dr Linder’s group conducted a cross-sectional analysis based on data from the CDC’s National Ambulatory Medical Survey and National Hospital Ambulatory Medical Survey (NAMCS/NHAMCS). The study was based on a nationally representative sample of adults with sore throat who presented to primary care physicians and emergency departments between 1997 and 2010, in 2-year periods. Antibiotic prescribing rates were calculated for first-line agents for group A streptococci (GAS), second-line agents, and non-recommended antibiotics.

The bad news: physicians prescribed an antibiotic in 60% of these visits. Over the study period, the prescribing rate did not change. Penicillin prescribing rate remained stable over the course of the study, at 9% of visits. Azithromycin prescribing increased during the study period from below the threshold of measurement, to 15% of visits in 2010 (P < .001). There was no change in the prescribing rates for amoxicillin (17%) and non-recommended antibiotics (15%).

Unfortunately, the prevalence of GAS in the context of sore throat is about 10%. Dr Linder stressed that GAS has not developed penicillin resistance, so penicillin is the only drug that should be prescribed here-in practice, it is not. Dr Linder took some pains (and he was justifiably pained by this) to explain that GAS has developed resistance to azithromycin preparations such as the Z-Pak, now prescribed in 15% of these visits. At the same time, he allowed that the observed 60% prescribing rate represents an improvement nationally, even though the “correct” rate is 10%. Why? NAMCS/NHAMCS survey showed an 80% prescribing rate in 1990, and a 70% prescribing rate in 2000. This is slow, but still dismal progress.

Results from Dr Linder’s adult bronchitis study are worse, using the same methodology. Physicians prescribed an antibiotic in 73% of these visits, and this did not improve over the 13 years of the study. It’s a similar prescribing rate to that seen in the 1980s. A quick Google search will confirm that many people still think acute bronchitis should be treated with antibiotics; ID specialists will tell you that the prescribing rate should be zero.

Dr Linder says he’s disappointed because he and many of his ID colleagues had high hopes for progress to be reflected in these studies. I’m not sure why-we’re still working under the same system that brought us to this place.

What’s to be done? This was the thinnest part of the discussion, because we are faced with a system problem, not an ID problem. Dr Linder agreed that current CDC- and IDSA-endorsed guideline promulgation and educational efforts should continue, although his studies show they’ve had little, if any effect. This is an opportunity for primary care to help avert a public health catastrophe. Physicians need to take ownership over this issue and demand more time for URI patients-E&M criteria do not capture what is lost when these visits are viewed simply as “Level 2 Office Visit, CPT 99212.”

Primary care must demand a system change-as the patients’ most important public health advocate. Read the full publication, which Dr Linder says is forthcoming in JAMA-Internal Medicine, and make a change.