Saturday, January 26, 2013

The flu is a virus!

It is winter and a lot of people are sick. Around here, and around the country, there are two big kinds of sick – one is mainly gastrointestinal disease with vomiting and diarrhea as the main symptoms, and the other upper respiratory infections with congestion, cough, and sometimes shortness of breath as the main symptoms. The first (GI) are mostly caused by norovirus in adults and adolescents and rotavirus in small children (and, recent reported, the elderly). The respiratory version is frequently influenza, or other viruses.  Viruses. Not bacteria, which can be treated with antibiotics. Viruses do not respond to antibiotics.

This is not to say that they cannot make you very sick. They can, and do. Especially in the old and very young and immunocompromised, influenza virus can lead to major bacterial complications and death; the swine flu outbreak of 1918 killed more people than WW I. When there are major influenza epidemics, there is a big excess of deaths. This is a really good reason to get the flu shot. Everyone who does not have a firm contraindication (e.g., allergy to eggs, a previous episode of Guillain-Barre syndrome) should receive the vaccine. People should expect that their health care providers have received the vaccine. It is not 100% effective, but it is very effective, and helps make the disease milder even if you contract it, and it decreases transmission.

It is not a reason to get antibiotics for a viral upper respiratory infection or bronchitis. Pneumonia, yes (even though a fairly large percent of pneumonias are viral, it is hard to tell); pneumonia as a complication of influenza, particularly in elderly or immunocompromised people with other chronic diseases (heart, lung, kidney, diabetes, cancer) is very serious. But these people, who have or are likely to have bacterial pneumonia and need antibiotics, represent a tiny fraction of the people treated with antibiotics for viral bronchitis (not to mention even less severe viral upper respiratory infections such as sinusitis, non-strep pharyngitis, and otitis). Bronchitis is no fun. It can make you feel miserable, create chest pain when you cough, and generally make you really sick. It can also last a really long time – 4-6 weeks of coughing is typical. But viruses don’t respond to antibiotics, even if you’ve been sick for a week or a month.

A recent study published on-line-before-print in JAMA-Internal Medicine by Gonzales and colleagues[1] looked at the use of decision support by either paper algorithms or computer systems in reducing the use of antibiotics for acute bronchitis in a very large multi-practice group in rural Pennsylvania (Geisinger Health System). They found basically two things: both the paper and computer assisted decision support tools reduced the rate of antibiotic prescribing about equally and both did so significantly more than in “control” practices that got neither. Unfortunately, the rate dropped from about 80% of to about 68%; that is, a large majority of those presenting with acute bronchitis received antibiotic prescriptions even after the intervention.

In a “Commentary” in the same issue, “Antibiotic Prescribing for Acute Respiratory Infections—Success That’s Way Off the Mark[2] , Jeffrey Linder notes that the problem with the study is that the “success” was very limited; that is, it moved the inappropriate use of antibiotics down, but it was still many times too high. His comparison is to the use of aspirin after heart attack, and how improving the rate from 30% to 40% would have been inadequate; luckily we are now at 94-99%. Another metaphor, more graphic, would be if we were happy that, over 10 years, the number of people killed by the average mass murderer dropped from 15 to 12!  

Since 2005," Linder notes,"a Healthcare Effectiveness Data and Information Set measure for patients aged 18 to 64 years states that the antibiotic prescribing rate for acute bronchitis should be zero. Despite the evidence, meta-analyses, and performance measures, antibiotic prescribing for acute bronchitis in the United States remains at more than 70%.” He is critical of the Gonzales study because, even after its “statistically significant” intervention, “The antibiotic prescribing rate—an event that should never happen for these patients—in ‘successful’ intervention practices was still more than 60%. For individual clinicians…we need to redefine success. Success is not reducing the antibiotic prescribing rate by 10%; success is reducing the antibiotic prescribing rate to 10%.”

Or less. Many people will say “I got antibiotics and I felt better in a couple of days”.  Almost all of these people would have gotten better anyway. There are some studies that show, in large populations, taking antibiotics can shorten symptoms by about a half-day. (This is probably because of some minor bacterial co-infection in some folks, especially those with chronic lung disease). But not by a week, or 2 or 3. Length of time of symptoms is not an indication for antibiotics for a viral illness. And that half day? Linder points out that “5% to 25% of patients who will have an adverse reaction. Worse, at least 1 in 1000 patients who take an antibiotic will wind up in the emergency department with a serious adverse drug event.”  This is, to put it mildly, not good.

Let’s review this: acute bronchitis, much less other “colds”, are viral and viral infections do not benefit in any way from treatment with antibiotics. They can, however, last a long time, and make you miserable. These symptoms are still not indications for antibiotics. The algorithm used by the Geisinger group, and posted on the walls of their examination rooms, is attached. There are some people, particularly the old, immunocompromised, and those with chronic bronchitis (mostly long-time smokers) who can develop pneumonia, which should be treated with antibiotics. They do not have acute bronchitis.

Doctors and other health professionals should know this, and most of them do. Sadly, however, they not only frequently prescribe antibiotics for viral illnesses because their patients “want them”, but also take them themselves for the same non-indications. Doctors, nurses, and others are among the greatest “abusers” of antibiotics (by which I mean taking them when they are not needed). Amazingly, many of these same health care providers are those who do not get the influenza vaccine, which they should be getting! The justification of “I need to stay healthy, and can’t miss work, because I need to care for my patients and don’t want to transmit illness to them” is wrong on 3 counts: 1) Taking antibiotics for a virus won’t make you less sick or shorten the course of your illness, 2) Taking antibiotics won’t prevent you from transmitting a viral illness, and 3) Taking antibiotics for a viral illness increases the risk of superinfections (e.g., yeast vaginitis), drug reactions, and the development bacteria that are resistant to common antibiotics, which you can spread to your patients. By the way, you also don’t need antibiotics for norovirus or other viral (or, indeed many forms of bacterial) gastroenteritis.

This Batman-and-Robin cartoon illustrates the frustration that many of us feel. Obviously, we cannot even think about literally or figuratively treating our patients that way, but I think one of the interesting parallels is that the person asking Batman for antibiotics is not a “regular person” but Robin, a kind of Batman-in-training, and thus Batman’s frustration mirrors that many of us feel when our own trainees (students and residents) inappropriately use antibiotics for themselves.

Linder says “We should address patients’ symptoms, but for antibiotics we need to tell our patients that ‘this medicine is more likely to hurt you than to help you.’” Those of us who are sick and not health care providers need to understand that; those of us who are health care providers have an even greater responsibility.

[1] Gonzales R, et al., A Cluster Randomized Trial of Decision Support Strategies for Reducing Antibiotic Use in Acute Bronchitis, JAMA-Internal Medicine Published online January 14, 2013. doi:10.1001/jamainternmed.2013.1589
[2] Linder, J, “Antibiotic Prescribing for Acute Respiratory Infections—Success That’s Way Off the Mark” JAMA-Internal Medicine Published online January 14, 2013. doi:10.1001/jamainternmed.2013.1984

No comments:

Total Pageviews