Why do you need to finish a course of antibiotics?

As long as I can recall doctors have always warned me whenever I need to take antibiotics to finish the course even if the infection has cleared. Thus if they give you a week’s supply and the infection clears in three or four days the doctor will tell you to keep taking them for the full week.

But why? I’ve never asked until now. Why not just bin the rest, especially given the fact that too much of an antibiotic can make you resistant to it?

I believe it’s two reasons.

First, even if the symptoms have gone done, that doesn’t necessarily mean that the infection is completely gone. You want to make sure the bug is dead, so keep using the antibiotic.

Second, resistance - it’s not that you develop a resistance, but that the bug can develop it. If you use the antibiotic but don’t completely kill the bug, what you’re effectively doing is helping the bug develop a resistance. Think of it as evolution in action. Just by natural variance, some bugs may be more vulnerable to a particular antibiotic, but other bugs may have a bit stronger ability to resist that antibiotic. If you stop early, you may only be killing the weakest ones, but don’t keep going until all the strongest ones are also dead, then you’re helping to select for the stronger ones. Those stronger ones will be more resistant to that antibiotic the next time they encounter it. Stopping early is essentially culling the herd of the weak bugs, and focussing the population on a stronger, antibiotic-resistant group.

Because just because an infection isn’t detectable doesn’t mean that every last microbe has been killed. A few may still be lurking. Continuing the course will kill off everything.

And you don’t become resistant to antibiotics, the microbes do. The last survivors will be the more resistant ones. If you don’t kill them off, they will breed more and more resistant strains.

That’s not how antibiotics work. You don’t get resistant to it, the bacteria evolve to get resistant to it. After three days you might feel better because the only bacteria left are the small percentage with resistance to the antibiotic. Given enough time they will die, because resistance isn’t immunity, but if you stop taking the antibiotic they will recover, giving you nice colony of bacteria especially selected for their resistance to the antibiotic you used.

W00t! I won the triple ninja! :smiley:

The problem is not that you may become resistant, it’s that you are subjecting the bacterial population in your body to strong artificial selection in favor of bacteria that are resistant. If you wipe out all the susceptible bacteria, but leave a small population of resistant bacteria, the resistant population may now expand into a fully resistant infection.

At first glance, it may seem that once you wipe out the susceptible bacteria, continuing the course will do no good, because only resistant bacteria are left anyway. But resistance is not always a binary effect, it will often be quantitative. It may be that most bacteria are killed quickly in a couple of days, but a small sub-population of mutants will only be killed after 4 or 5 days. In addition, even if there are a few fully resistant mutants, keeping the total population of bacteria well under control for a long enough period means that your immune system has much less work to do and can devote its resources to taking are of the small number of resistant bugs.

Here’s a quite impressive (and kind of scary) video showing the evolution of antibiotics resistance in bacteria:

https://www.youtube.com/watch?v=yybsSqcB7mE

By ending a course of antibiotics early, you are essentially creating an environment similar to the early stages in that large petri dish, where the concentration of antibiotics is low enough that some bacteria may survive - and you are doing this while some of the pathogenic bacteria may still be around. So you are helping them to develop resistance for higher doses of antibiotics.

In short: You do only need to take the antibiotics until you’ve killed off all of the germs. But that takes longer than most people think. Your doctor who prescribed the antibiotics knows, though, and that’s how he decided how long of a course to give you.

This x 100,000.

Just because you feel better doesn’t mean the infection’s gone. You want to wipe out ALL of it, or at least get it down to a point where your immune system can take over.

Not finishing a course of ABX is a major cause of ABX resistance.

The doctor doesn’t KNOW anymore than then what the drug manufacturer’s sales rep and associated literature told her. The rest is best guess in a CYA mode.

Antibiotics (like all drugs) are extensively and carefully tested, and the optimum course of administration for various purposes is documented based on that research. Of course the doctor knows how to prescribe any drug because it’s part of his job to read the associated literature and to know the correct protocols. How else would you expect the doctor to know?

In short, what on earth are you talking about? How do you imagine this should work?

Well, the doctor doesn’t have perfect knowledge of the exact count of bacteria in your system, but he does know a lot better than you or I about such things. There’s a reason it takes years to go through med school, after all.

It is of note that there is a move within the infectious disease community towards shorter courses, at least for community acquired pneumonia, as part of antibiotic stewardship.

The conventional reasons for longer courses, as given by others here, seem rational, but in some cases they may not be good reasons after all.

That is a naive view of drug trials.

Drug trials are difficult and expensive and difficult (and difficult). Sometimes, if there is a very good reason to do so, another trial is started, using a different treatment protocol. Sometimes, if there is a good reason to believe the protocol is not optimimum, patients are treated with a protocol other than that tested.

Drug testing is first to see if the protocol is safe. Then to see if the protocol is effective. Sometimes even these limited aims are not achieved. Further trieals may try to identify a better treatement protocol. Only rarely is there any kind of belief that an “optimum” course of administration has been identified.

When antibiotics became cheap and universal enough to be used for unmonitored use, the protocol was “just keep using it for such a long time that we /know/ it is long enough, and we won’t have to bother to check when enough is enough”

Before that, the first uses of antibiotics involved very lmited supply. If there wasn’t enough, it ran out, the patient got sick and died. If your course was very short to start with, it was important to finish it.

It is still (and again) the case that in some situations, your use of antibiotics will be actively monitered, and ceased at the earliest appropriate opportunity.

My first course of antibiotics consisted of one shot of penicillin. If I had been given two, it would have been because I needed two. If I needed two, and decided not to bother, I would have been sick.

Ignorance fought. And evolution is indeed wonderful, and damned scary in this case.

And here you are making the perfect the enemy of the good.

I was addressing a post that implied that a doctor should be doing something other than following documented protocols. There is no better source of information than the protocols developed through research programs. We certainly do not want doctors using their discretion on dosage or course based on informal anecdotal experience.

Even after the clinical trials are over, you could probably come up with some pretty good figures from voluntary noncompliance among patients. That is to say: Patient catches strep throat, patient goes to doctor and gets a course of penicillin, patient stops course early at some point, patient develops strep again and goes back to the doctor, doctor asks patient when patient stopped the course. Now you know that that wasn’t enough. Repeat as many times as needed to get good statistics.

Dr Ian Malcolm’s comments on overuse of antibiotics:

Exactly. The post that Riemann was referring to implied that doctors are doing something much more akin to pharmaceutical shamanism than what they actually do, and that they’re incapable or unwilling to research how a drug is actually used.