‘Must I take all my antibiotics?’ Why infectious disease experts are rethinking this longstanding rule

You’ve probably heard it for years, every time you get a prescription for antibiotics: “Make sure to take all the pills.”

Traditionally, clinicians have given two reasons for these strict instructions: So the infection doesn’t come back, and so you don’t end up with bugs so hearty that antibiotics can’t kill them.

But these days, clinicians like me are starting to rethink this approach.

As a physician specializing in infectious diseases and the medical director of the the University of Virginia’s antimicrobial stewardship program, I’ve spent much of my career focused on using antibiotics wisely and encouraging colleagues to be thoughtful about how they choose and for how long they prescribe these medicines.

The evidence around how long patients should take antibiotics is more nuanced than the traditional advice to “finish the course” suggests. Yes, it’s still a good idea to follow your doctor’s advice – but not because stopping your medicine early would breed bacteria that are resistant to antibiotics.

A mistaken cause of microbial resistance

When physicians began to use penicillin widely in the 1940s, they noticed something curious in the first few hundred patients. Some improved dramatically in just a few days, while others required longer courses to cure their infection and avoid relapse.

Studies eventually showed that the type of infection a patient has, the specific bacteria causing it and how well an antibiotic can penetrate into infected areas in the body all play a role in this variation.

But the message that patients must finish whatever course of antibiotics they were given – which was promoted by the Centers for Disease Control and Prevention and World Health Organization – was based on a misconception that if you don’t kill all the infecting organisms right away, the ones that come back and cause a relapsed infection will be resistant to the original antimicrobial drug.

However, there is little evidence of that. Take strep throat, for example. This bacterial infection sometimes comes back after a short course of antibiotics, but that’s not because of resistant bacteria. Rather, it’s because the patient may get reinfected – for example, from the bug circulating at home or in school – or because an antimicrobial did not penetrate the infected tissue fully, leaving some straggler bacteria in place to regrow.

In fact, the bacteria that cause strep throat have never developed resistance to penicillin, even after decades of use or in the setting of a relapsed infection.

Balancing antibiotics’ pros and cons

Scientists now know that it’s not shortened courses of antibiotics that can promote antimicrobial resistance, but that antibiotic exposure itself creates selective pressure for resistant bacteria to emerge.

What many clinicians, including myself, worry about is that most antibiotics do not just target the bacteria causing the infection. They also affect the many bacteria people naturally carry in their gut, mouth and elsewhere in the body including the beneficial ones.

The longer someone takes an antibiotic, the more it can disrupt these microbial communities – often called the microbiome.

Antibiotics begin affecting the microbiome as soon as they are taken. By killing or suppressing bacteria that are susceptible to the drug, antibiotics can give resistant bacteria a survival advantage, allowing them to make up a greater proportion of the remaining bacterial population. Thus, each additional day of treatment provides more opportunity for this selection to occur.

This is why researchers have increasingly focused on finding the shortest course that reliably cures an infection by providing enough antibiotic to treat the infection successfully, while avoiding unnecessary exposure that can further disrupt the microbiome and give unwanted bacteria a chance to thrive.

Durations in flux

Over the past decade, several clinical trials involving infections including bladder infections, pneumonia and even bloodstream infections have found that shorter courses of antibiotics can work just as well as longer ones, often with fewer side effects.

Thanks to this evidence, doctors today prescribe much shorter courses of antibiotics than they did a decade ago. What used to be a 14-day prescription is now often three, five or seven.

But shorter is not always better. Some complicated infections require weeks of treatment. An infection may be contained in a pocket, such as an abscess, or involve tissue that antibiotics have difficulty reaching. Or, a person might have a weakened immune system, making it harder for their body to clear the infection.

Ideally, your healthcare provider would take these factors into account and use the most up-to-date evidence for deciding which antibiotic to prescribe and for how long. Following their advice is still your best bet.

Their medical judgment may be imperfect, simply because the full picture on how long infections must be treated is still coming into view. Researchers have not studied all types of infections, and few studies have looked at patients with underlying health issues or those in hospitals.

What’s important to remember, however, is that the goal isn’t simply to take fewer antibiotics but to use them for the shortest amount of time that has been shown to safely and effectively treat a particular infection.

Can you stop when you feel better?

Since about 2020, researchers began asking whether patients could use their improvement in symptoms as a guide for when to stop taking antibiotics and thus limit unnecessary antibiotic exposure.

In theory, this could work for some patients some of the time for some infections – thereby reducing unnecessary antimicrobial exposure – but in practice, it’s not that simple. While certain infections could be treated with shorter courses of antibiotics than doctors have traditionally prescribed, there’s little to no research on how the length of time taking antibiotics aligns with when a person’s symptoms improve. For some infections, a person may feel better but still require additional antibiotics to fully treat and keep the infection from coming back.

What’s more, feeling better is subjective, so using a patient’s sense of when they’re feeling better as a guide means different people might stop taking their antibiotics at different times.

A better way to cut down on unnecessary antibiotics use is for clinicians to be more mindful not to prescribe them unnecessarily – which many still do. A 2025 review of 412 studies showed that roughly a third of antibiotics prescriptions worldwide are inappropriate.

This is especially true for colds and other infections in the nose, throat and sinuses, many of which are caused by viruses and don’t need antibiotics at all. When patients take antibiotics for these ailments and start to feel better after a few days, it’s likely not because the antibiotic improved their condition but just because their body successfully fought off the infection on its own.

Another piece of the discussion is trust. The “complete the course” message has long served as a simple, clear way to get people to take their medications seriously. Changing that message, especially when trust in public health is already fragile, needs to be done carefully.

Even with this in mind, however, the data is clear that a shorter unfinished course is not the cause of antibiotic resistance.

And for us physicians, it’s our job to prescribe antibiotics only when they’re truly necessary, to rely on up-to-date evidence to guide duration and to talk to our patients about what to expect: when they should start feeling better, when to worry about recurrence and what to do if they’re not sure whether to keep taking their medication.