Where did I get the idea? From the ID guys with whom I rotated in medical school. The lengths of time are simply made up. Everyone who works in the field with whom I have spoken about it is quite clear on that. Do you honestly think that bacteria work on seven day, three day, ten day and fourteen day schedules? No one knows the precise length of time that is necessary to treat an infection so we make up convenient time periods. Those are very human numbers. They are pretty close to what is needed, but they are not precise numbers exactly tailored through intensive evidence-based investigation. They are quick guides that ease the administration of drug studies.
I'm not implying that they are completely random, just that they are not precise and they are probably often overkill for many patients. But that is what we want for most infections. The fact that they are overkill for some people is why some folks get away with partial courses.
r
Drug trials generally work in arbitrary time periods. That's how they are done. People have gone back afterwards and tweaked the time periods a bit because of this fact and so some drug trials in the past few years have used shorter courses. But they all essentially work on the same paradigm.
You are just defining 'arbitrary' differently than I am, precise as in, to "the minute", or precise as in, will be effective in some acceptable % like 98% of the time? And arbitrary as in 7 days or 8 days or 3 days compared to 10? Maybe 7 is arbitrary within 5 to 9 days, but certainly in blocks of time it is not. Is it practical to study blocks of treatment times in hours? days? and for each infection and population studied? No. But at the same time, for this person who was going to be OK on day 6, are you suggesting people just arbitrarily stop antibiotics? It isn't arbitrary just because 48 or 72 hour blocks of time were the most practical way to determine the standard course of treatment.
The disconnect here is the belief that there is no reason to urge patients to finish a course of an antibiotic. When they are supposed to stop taking it and how they are supposed to know hasn't been addressed.
But lets look at the other issues here.
You say bacteria don't work on a schedule, I beg to differ, they most certainly do. The factors which go into determining dose and the length of treatment are infection site (can affect peak drug concentrations and the resulting half life as well as the typical extent of infection), the antibiotic's mechanism of action, and the rate of reproduction of the organisms. The simplest example is the most common mechanism of action, that of preventing the production of a new cell wall forming thus killing dividing organisms. Length of treatment is simple. How long will it take for the last bacteria to divide and die? That is a known measurable quantity.
If you are treating TB which is an extremely slow growing organism,
for example,
The standard length of treatment for uncomplicated pan-sensitive TB is a minimum of 6 months (26 wk) to include at least 4 months of treatment following sputum culture conversion
Treatment length from some infections can sometimes be prescribed as 48 hours after all symptoms abate but again, that isn't practical if the patient isn't reliable nor in cases where symptoms are not an accurate measure of the absence of infection, so a standard length of treatment is more practical.
But this discussion really is missing the point which I think is the reason there is so much misunderstanding going on. Kelly's question and the discussion is whether or not treating an infection incompletely leads to more risk of antibiotic resistance or not. Treatment length is not determined with the main goal of preventing antibiotic resistance, it is determined by the clinical outcome. And that is determined by research, not by just arbitrarily rolling dice. And a lot of research has gone on looking at shortening antibiotic courses, sometimes with failure, (
A Meta-analysis of Randomized, Controlled Trials Comparing Short- and Long-Couse Antibiotic Therapy for Urinary Tract Infections in Children), and sometimes with success, (
Optimal duration of antibiotic therapy for uncomplicated urinary tract infection in older women: a double-blind randomized controlled trial.). If the length of treatment really were as arbitrary as you implied, then you are saying it isn't evidence based. But since it is evidence based, then stopping treatment prematurely means a larger % of cases will be incompletely treated. And that increases the risk of antibiotic resistance.
Getting back to kelly's question, if you can treat an infection successfully with a short course of treatment, you may also expose less bacteria which are not your intended targets to a small enough dose of antibiotic that you do not reach a threshold which induces antibiotic resistance. While true, that is not always the case and it doesn't change what I said about increasing the risk of drug resistance when you partially treat an infection.
I need to digress here and then get back to the above. Considering how frequently doctors prescribe antibiotics needlessly for respiratory viruses, I can see where it probably has been a common experience for many people to stop antibiotics and seeing no relapse, think nothing happened when they did not follow that urging to not stop treatment mid-course.
I have not been discussing that scenario. I am discussing people who are prescribed the correct drug for the standard course for a bacterial infection. So in this discussion of stopping treatment, I am talking about stopping an antibiotic before the infection is cured, or at least before the standard course is finished which assumes then that a significant risk of not being cured exists. So in this scenario if patients stopped taking antibiotics before the prescribed time, then a good percentage of them would relapse. And relapse of a partially treated bacterial infection is a BIG risk factor for developing antibiotic resistance.
I hope at least that issue is not in question though I get the impression kelly can't accept that basic well known fact without some research. And it is this fact that I find seeking a citation to be absurd.
We also got into a side track about eliminating all of the bacteria of the same strain as was causing the infection or just the bacteria actually in the infected site. Kelly posted research that H-flu carrier state was not eliminated with the usual course of treating OM caused by H-flu. There are other bacteria, research I cited, which are eliminated by a full course of antibiotics. This only goes to show how many factors are involved in infection and drug resistance.
Given that there are multiple factors involved in the development of antibiotic resistance, it is important to notice when someone is using the words 'can', 'may', 'might', 'more likely', and so on. I will again clarify what I have tried to in my last several posts. If you don't take the full course of antibiotics prescribed for a bacterial infection, there is a greater risk of antibiotic resistance developing from the cumulative mutations which increase with exposure to antibiotics.
Here is why: Those mutations may not have been sufficient to allow the bacteria to fully resist the current antibiotics. So the full course would have eliminated the bacteria leaving behind no partially resistant organisms ready to mutate further the next time.
Are all partially resistant bacteria eliminated with every full course of antibiotics? No and I did not say that. Antibiotic resistance is not currently preventable, we can only delay and contain it. Are you better off stopping antibiotics you started if it is determined you didn't have a bacterial infection after all? Yes, and hopefully that is after a minimal exposure so there is less chance of resistant mutations occurring.
But the scenario in the question was stopping a course of antibiotics before the prescribed time. If you want to strictly look at numbers and odds, you can expect that to lead to infections being partially treated more often than with completed courses of therapy. Thus you can expect cumulative mutations to occur more often which increases the rate of drug resistance developing.