Skip to Main Content

Winter is coming. With the end of Covid-19 nowhere in sight, the U.S. must brace for the annual surge of influenza, bronchitis, sinusitis, bronchiolitis, and other acute respiratory illnesses. Last winter, the Center for Disease Control and Prevention estimates there were 38 million cases of influenza alone.

One big issue is that the symptoms of these illnesses overlap considerably. A cough, runny nose, sore throat, and fever could mean the flu or other common respiratory illness. But it could also worry people that they might have Covid-19.

We are concerned that the collision of Covid-19 and winter could lead to a rash of inappropriate antibiotic prescribing. That’s a problem because, in addition to being wasteful, taking antibiotics that aren’t needed can lead to serious health issues, including allergic reactions, diarrheal infections, and the development of antibiotic-resistant bacteria that are incredibly difficult to treat.


Antibiotic resistance is such a pressing problem that the National Academies of Sciences, Engineering, and Medicine are holding a three-day workshop on the subject beginning Monday.

Inappropriate prescribing of antibiotics also drives people to get more care in the long term. More on that later.


Long before the pandemic emerged, antibiotics were greatly overused. One in three prescribed antibiotics aren’t appropriate. Acute respiratory infections are generally caused by viruses, which antibiotics don’t kill, yet acute respiratory illnesses account for 75% of all inappropriate prescriptions.

But as bacteria in the body are exposed to antibiotics, they can mutate and develop resistance to these drugs. Antimicrobial resistance causes 35,000 deaths each year in the United States alone and leads to complex diseases that have increased risk of complications and side effects. The health care costs are enormous: The CDC estimates that antibiotic resistance adds $20 billion in costs to the health care system, in addition to a $35 billion loss in productivity each year.

Several forces during the pandemic may lead to a surge in antibiotic use. SARS-CoV-2, the virus that causes Covid-19, is not affected by antibiotics. Antibiotics should be used for people with Covid-19 who also have bacterial pneumonia. Yet some physicians worry that their patients with Covid-19 might be susceptible to bacterial infections and many inappropriately prescribe antibiotics as a precaution.

The pandemic may also drive more adults and children to visit their physicians, which increases the chances they will be prescribed antibiotics. Before the pandemic, sniffles and low-grade temperatures were nothing to worry about. Now people worry that they signal the onset of Covid-19 and seek care to make sure they don’t have it. In addition, schools and day care centers have stricter rules on kids coming in who are ill.

The rapid growth of telemedicine during the pandemic may also increase the overuse of antibiotics. While telemedicine has been a lifeline for people who need care, our research has shown that, among children, telemedicine is more likely to lead to inappropriate antibiotic prescribing than in-person doctor visits. Given that a physician can’t physically examine a child remotely, he or she may not be sure about the diagnosis and prescribes antibiotics under a “better safe than sorry” approach.

While playing it safe is understandable, it represents short-term thinking and does not account for the long-term impact of antibiotics on how people seek care. Say an individual becomes ill with an acute respiratory illness, sees a physician, and is prescribed antibiotics. After a week of taking them, she feels better. Since humans are great at pattern recognition, she intuitively thinks that the antibiotics made her feel better. We know from research that she would have likely gotten better without the antibiotic, but that is not her lived experience. Naturally, the next time she gets an acute respiratory illness, she will go back to the doctor for that supposed cure.

We recently quantified this phenomenon by comparing people who were essentially randomized to low-antibiotic-prescribing physicians or to high-antibiotic-prescribing physicians. Patients who got care from high-prescribing physicians were more likely to seek care when they got sick again and received 14% more antibiotics the following year than those seeing low-prescribing physicians.

This feedforward mechanism compounds the overuse of antibiotics and self-enforces potentially inflated benefits of antibiotic use. This effect also extends across social networks as people learn from their friends and family members. We observed that spouses of people seen by high-prescribing physicians were also more likely to receive antibiotics in the subsequent year than spouses of people seen by low-prescribing physicians.

This matters in the time of coronavirus because in the months ahead, with more anxiety, more people getting sick, more people getting care, and more telemedicine use, doctors may inappropriately prescribe more antibiotics. This could dangerously accelerate the feedforward system, enforcing the misconception that antibiotics are always beneficial and driving even more inappropriate antibiotic use in the future.

Recognizing the risk of a surge of antibiotic prescribing is the first step toward encouraging physicians and patients to use antibiotics appropriately, and not for viral illnesses. It is more important than ever for physicians to take the time to talk to their patients about the appropriate use of antibiotics and reassure them that not prescribing them can be the medically right thing to do. Ideally, this will create a feedback loop such that more judicious antibiotic use in one visit may result in fewer antibiotics into the future, and help establish new norms on when antibiotics are needed.

Jowa (Zhuo) Shi is a medical student in the Harvard-MIT Program in Health Sciences and Technology. Ateev Mehrotra is an associate professor of health care policy at Harvard Medical School and a physician at Beth Israel Deaconess Medical Center.

  • I think the main takeaway is that we need much better diagnosis and data.

    I have the sinuses of a 300 year old. If I get any kind of illness that causes congestion, I have a 10% chance of getting over it on my own. Antibiotics are becoming less effective, but the longer I wait to start treatment, the longer it takes to recover (and the more antibiotic days/types).

    Very little data is gathered on illness course and recovery time compared to treatment given, especially in any kind of personalized healthcare manner.

    And the fact that there is no easy test in 2020 to tell what the nature of a sinus issue (inflammation, virus, bacteria, etc) is insane.

    Hopefully Covid has demonstrated the necessity of this type of data gathering and research. Outside of very broad-brush numbers on over prescribing and recovery without antibiotics, we know next to nothing.

  • If everyone wishes to alleviate their angst about how to treat during fall and winter months, just look into a pre-exposure prophylaxis that is already approved !!! Google A novel plan — PrEP — by Dr Ray Stricker.
    Remember that HIV and malaria still don’t have vaccines! Since we don’t have a Covid vaccine yet, try Dr Stricker’s PrEP plan.

Comments are closed.