Do Patients with Strep Throat Need to Be Treated with Antibiotics?

Background: Streptococcal pharyngitis is a common presentation to primary care and Emergency Department physicians. Every year, 10 million patients in the United States are treated with antibiotics for pharyngitis. However, less than 10% of these patients actually have strep pharyngitis (Barnett 2013). Prescribing of antibiotics for these patients centers on three arguments:

  1. Antibiotics reduce symptomology
  2. Antibiotics reduce the rate of suppurative complications
  3. Antibiotics reduce the rate of non-suppurative complications (primarily Rheumatic Heart Disease).

So, do patients with strep throat need to be treated with antibiotics?

While treatment of strep throat seems relatively benign, there are significant harms that need to be considered:

  1. 1 in 10 patients will develop antibiotic associated diarrhea (some of these will be C. diff)
  2. Severe allergic reactions occur in 0.24% of patients

This means that out of 10 million patients treated with antibiotics, as many as 24,000 of them will have fatal or near-fatal allergic reactions.

Thus, it’s important to determine if the benefits of antibiotic treatment outweigh the risks.

Argument #1: Antibiotics reduce symptomology

The most important thing to recognize is that in the vast majority of patients, strep throat is a self limited disease that will get better in about 7 days with no treatment (this includes supportive care). The addition of antibiotics provides a modest benefit in terms of symptomology resolution (12-16 hours) (Del Mar 2006). This benefit, however, is in comparison to placebo, which is not standard care. Supportive therapies including acetaminophen and NSAIDs may affect symptom improvement but there are no good studies on this. Finally, we must not forget about the potential additional side effects associated with antibiotics (i.e. diarrhea).

One therapy that has shown remarkable benefits in terms of symptoms is the use of corticosteroids. Hayward et al showed that corticosteroids increased the rate of symptom resolution at 24 hours with an NNT = 4 (Hayward 2012).

Bottom Line: If we are interested in making the patient’s symptoms resolve faster, corticosteroids are our best bet. I typically give 10 mg of decadron IM.

Argument #2: Antibiotics reduce the rate of suppurative complications.

There are a number of potential complications associated with strep throat including acute otitis media (AOM), sinusitis and peritonsilar abscess (PTA).

Yeh 2005 Del Mar 2006
AOM NNT = 25 NNT > 200
Sinusitis NNT = infinity Not Reported
PTA NNT = 28 NNT = 55 – 225

A more recent study demonstrated an overall suppurative complication rate of 1.3% and no difference in patients who received antibiotics versus those that did not (Little 2013).

Bottom Line: It appears that we would have to treat 100’s of patients to prevent one PTA; an easily treatable entity.

Argument #3: Antibiotics reduce the rate of non-suppurative complications.

The two major non-suppurative complications are: Post-strep Glomerulonephritis (PSGN) and rheumatic fever (RF). No study has ever shown that PSGN can be prevented and so, we are left with RF.

Evidence for preventing RF and subsequent rheumatic heart disease (RHD) comes from a series of studies performed in the 1950’s at the Warren Air Force Base. In this military population, investigators found that 2% of patients with strep throat developed RF. With antibiotics, this rate fell to 1% giving an absolute risk reduction of 1% and an NNT of 50-60 to prevent RF (Denny 1950, Wannamaker 1951, Chamovitz 1954, Siegel 1961). The work done by these researchers forms the basis for treatment over the last five decades.

However, we must ask the question of whether these studies apply to our patients today. The rate of RHD in the westernized world is exquisitely low. In fact, the CDC stopped tracking the incidence in 1995 when it fell below 1 per million. Numerous RCTs in

developed countries have shown no cases of RF or RHD in patients treated with placebo (Middleton 1988, De Meyere 1992, Dagnelie 1996, Little 1997, Zwart 2003).

Based on the current incidence of RF in the US, we would need to treat about 2 million patients with strep throat in order to prevent a single case of RF. In addition, only 1 out of every 3 patients who develops RF will subsequently develop RHD. Treating millions of patients with pharyngitis in the pursuit prevention of single digit cases of RHD in the western world makes no sense.

The (lack of) Effect of Antibiotics on Acute Rheumatic Fever -Thanks to Seth Trueger

Many physicians argue that the reason for the decline in RF and RHD is because we treat every patient with pharyngitis for strep. However,epidemiologic data speaks against this. The incidence of streptococcal diseases fell long before the advent of antibiotics but fell concurrently with improvements in public health. It is a far more likely scenario that improvements in sanitation have led to shifts in the serotype of Group A beta-hemolytic streptococcus that causes strep throat in developed countries.

Clinical Bottom Line:

We are far more likely to harm patients with strep pharyngitis by giving antibiotics than to help them in developed countries. This does not apply to developing countries with poor public health (See this post from Casey Parker about treatment in developing areas).

Finally, let’s see what one of the core texts in EM has to say on the topic:

“acute pharyngitis should not typically be treated with antibiotics. The great majority of cases are viral in origin, and suppurative complications following streptococcal infection are both easily treated and too rare to justify routine use of antibiotics. In particular, antibiotics were beneficial in reducing rheumatic fever only during a single military epidemic in the mid-twentieth century, and the decline of rheumatic fever is unrelated to trends in antibiotic use.” (Rosen’s 2014)

Related Posts


  1. Barnett, ML, Linder JA. Antibiotics prescribing to adults with sore throat in the United States, 1997-2010 Research Letter. JAMA Int Med 2013; 174(1): 138-40. PMID: 24091806
  2. Del Mar CB, Glasziou PP, Spinks AB. Should sore throats be treated with antibiotics? (Review). Cochrane Database of Systematic Reviews 2006 Issue 4. CD000023 PMID: 15106140
  3. Hayward G et al. Corticosteroids as standalone or add-on treatment for sore-throat (Review). Cochrane Database Syst Rev 2012. CD 008268 PMID: 23076943
  4. Yeh B., Eskin B. Should Sore Throats Be Treated with Antibiotics? Ann of EM 2005; 45: 82-4. PMID: 15635315
  5. Little P et al. Predictors of suppurative complications for acute sore throat in primary care: prospective clinical cohort study. BMJ 2013; 347: f6867. PMID: 24277339
  6. Denny FW, Wannamaker LW, Brink WR. Prevention of rheumatic fever. Treatment of the preceding streptococcic infection. JAMA. 1950;143(2):151-3. PMID: 15415234
  7. Wannamaker LW et al. Prophylaxis of acute rheumatic fever by treatment of the preceding streptococcal infection with various amounts of depot penicillin. Am J Med. 1951;10:673-94. PMID: 14837911
  8. Chamovitz R et al. Prevention of rheumatic fever by treatment of previous streptococcal infection. NEJM, 1954. 251: p. 466-71. PMID: 13194096
  9. Siegel EE, Stollerman GH. Controlled studies of streptococcal pharyngitis in a pediatric population. NEJM. 1961;265:559-65. DOI: 10.1056/NEJM196109212651202
  10. Middleton DB et al. Standardized symptomatic treatment versus penicillin as initial therapy for streptococcal pharyngitis. Ped Rheum Ther 1988; 113: 1089-94. PMID: 3057159
  11. De Meyere M et al. Effect of Penicillin on the clinical course of streptococcal pharyngitis in general practice. Eur J Clin Pharmacol 1992; 43: 581-5. PMID: 1493837
  12. Dagnelie CF et al. Do patients with sore throat benefit from penicillin? A randomized double-blind placebo controlled clinical trial with penicillin V in general practice. Brit J Gen Pract 1996; 46: 589-93. PMID: 8945796
  13. Little P et al. Reattendance and complications in a randomised trial of prescribing strategies for sore throat: the medicalising effect of prescribing antibiotics. BMJ 1997; 315(7104): 350-2. PMID: 9270458
  14. Zwart S et al. Penicillin for acute sore throat in children: randomised, double blind trial. BMJ 2003; 327(7427): 1324-8. PMID: 14658641
Cite this article as: Anand Swaminathan, "Do Patients with Strep Throat Need to Be Treated with Antibiotics?", REBEL EM blog, January 5, 2015. Available at:
The following two tabs change content below.

Anand Swaminathan

Clinical Assistant Professor of Emergency Medicine at St. Joe's Regional Medical Center (Paterson, NJ)
REBEL EM Associate Editor and Author

Latest posts by Anand Swaminathan (see all)

Like this article?

Share on facebook
Share on Facebook
Share on twitter
Share on Twitter
Share on linkedin
Share on Linkdin
Share on email
Share via Email

Want to support rebelem?

39 thoughts on “Do Patients with Strep Throat Need to Be Treated with Antibiotics?”

  1. Good Day

    Regarding this issue and the studies I agree to some extent. My question is: In terms of physiology, recovery time, health, and costs, do you think Corticoid is the golden goose for adults and children in this situation? Thank you.

    • Lilian – good question. I don’t think there is a magic bullet. Even with steroids, NNT = 4 which means 75% of patients won’t benefit or at least, won’t see the max benefit of the drug. Even the steroid studies have issues. The only magic we really have is time. Sometimes, it’s simply going to take a wait and see approach.

      That being said, I hit most patients with steroids and NSAIDs or acetaminophen (paracetamol) to help with symptomology.

  2. Anand,

    Agree completely with. The only caviet I would add is treatment of sub-groups of your population where the risk is much higher. In my practice Aboriginal and Torres Strait Islanders (Australia’s indigenous Population) have a incidence up to 1.5 per 1000. So you need to treat your sub groups and at risk populations as appropriate.

    • Ray – excellent point. This post truly applies to those living in developed countries with advanced public health. Even in the US, there are pockets of populations (Native American Reservations) where I would treat strep throat because epidemiology differs.

  3. Anand,

    Great topic- thanks for the post.
    I generally agree with your overall thoughts/intents & I generally practice very similarly to you w/regard to strep throat. It’s pretty clear that overall treatment w/abx is beyond over-done in the US. There’s no doubt that risk/benefit is far in favor of not treating when looking at the large all-comer numbers of abx for “pharyngitis” (including treating based of presumption, Centaur, etc)

    I wanna focus for a min on what really seems to be the big issue here: RF/RHD:
    As you know one very common & traditional strategy involves the following:
    Perform a rapid strep, if pos treat, if neg don’t. (neg- send culture for potential FN). Sure there will be FPs w/ this strategy (inherent to test, but also due to “carrier state”, etc) and to what extent I’m not sure. Anyway, David Newman’s numbers address this strategy since he hones in on cases of “confirmed strep”, which while much smaller than the overall #s (your #s), still show a harm/benefit ratio in favor of not treating.

    The most cogent argument out there from “strep-treaters” is the following: the major reason NNT is so low to begin with is BECAUSE of abx treatment for all these years (“vaccine, herd-immunity effect”). I think this is where the crux of the debate is.

    Can you share the most robust evidence you’ve come across in your research to support that the argument that the decline in RF/RHD was due to other epidemiological reasons: better public health/ access to care, decreased crowding, improved sanitation, etc, and NOT as a result of widespread antibiotics use; and that if all of a sudden going forward, generally (w/rare exception) if all cases of confirmed strep throat in the United States were to be treated without abx, that RF/RHD would NOT make a comeback.

    Thanks so much.

    -Sam Ghali (@EM_ResUS)

  4. Excellent post on such an important subject.

    From a patient’s perspecitve, it strikes me that decades of docs prescribing antibiotics for strep (and other unnecessary drugs, unnecessarily) have trained us well. Patients now expect and even demand a prescription when they present with strep. I wrote about this a few years ago in: “When Patients Demand Treatments That Don’t Work” –

  5. Sam – Thanks for the comments. I hear the same arguments all the time. “We’ve obliterated RF/RHD because we treat everyone with strep throat.” The best evidence to refute this is the graphs showing declines in strep deaths and RF/RHD prior to the advent of antibiotics and the fact that the advent of antibiotics didn’t affect these drops. We included one of these diagrams here and you can see more on this in this citation:
    McKinlay JB, McKinlay SM. The questionable contribution of medical measures to the decline in mortality in the united states in the twentieth century. Health and Society 1977 (I can send you a copy if you like).
    Also, see Casey Parker’s post on Broome Docs where he’s got a nice graph (

  6. Carolyn – couldn’t agree with you more. Doctors have done a terrible job of creating this expectation making it much harder for those of us willing to break the pattern based on evidence.

  7. I generally agree about the overutilization of antibiotics and minimal benefit etc… With a couple of caveats:

    1. In the Hayword metaanalysis quoted everyone that received steroids for tonsilitis ALSO received antibiotics, so the benefit seen was from a combination of steroids + antibiotics.

    There is a very distinct possibility that administering steroids ALONE can lead to an increase of supporative complications through the mechanism of immunosupression in a setting of this bacterial infection. As we know with many bacterial infections steroids are relatively contraindicated for that reason unless antibiotics are administered as well.

    I would be really hesitant treating strep throat with steroids alone without the antibiotics without a study looking at steroid treatment ALONE.

    2. I cannot see any indication to give immediate release steroids IM. There are studies after studies that show PO steroids are equivalent to IV steroids. The only thing you get out of administering dexamethasone IM are local complications and increased visit cost.

    A single dose of intramuscularly administered dexamethasone acetate is as effective as oral prednisone to treat asthma exacerbations in young children:

    A Randomized Clinical Trial of Oral versus Intramuscular Delivery of Steroids in Acute Exudative Pharyngitis:

    • Thanks for the comment. You make a number of excellent points.

      Re: steroids. There were a host of patients who did not get antibiotics and no additional adverse events were seen. Could steroids increase the risk of suppurative complications? It’s possible but I’ve not seen it in my practice or heard of it from others. The vast majority of pharyngitis is viral (~ 90%) and I don’t think steroids will convert these to PTA. PTA is much more likely to be a de novo presentation (i.e. not resultant from strep throat).

      Thank you for the references on oral steroids. I think that’s a very reasonable approach instead of IM. I do get a number of patients who have trouble swallowing the prednisone (sore throat + bitter taste) where I use decadron instead but oral steroids seem like a great option.

      Thanks again for reading and sharing your thoughts!

  8. With the most recent post by Washington University EM: Antibiotics for Strep Pharyngitis – The Pediatric Perspective, that does raise the question to me at least, do you give antibiotics to kids? Is there an age cut off?

    • Hello Scott,
      Sorry for the delay in response.

      1. Treatment of Strep Pharyngitis with Abx: If not in an industrialized country then absolutely you treat with abx as the incidence of Rheumatic fever is higher. That being said the incidence is also higher in kids compared to adults in industrialized countries so I do treat kids with abx.
      2. What age cutoff do you use: Most of the adult studies cited are military studies so, the adult population is age 17 and older. Therefore I use a cutoff of 17 years and older for adults, meaning, 16 years and younger in my mind, is a pediatric population.

      Hope this helps.


  9. Thank you for taking the time to put together this nice article. I appreciate the article and the thought process behind it, but unless I’m mistaken, the references you cite to calculate the number of patients needed to treat do not appear to be appropriate sources, since the studies include patients with sore throat due to additional causes besides streptococcus. So, are your conclusions truly valid?

    In general, the standard of care for management of streptococcal pharyngitis is established by authoritative organizations such as the American Heart Association, the Infectious Diseases Society of America and the American Academy of Pediatrics and the American Academy of Family Physicians. All of these groups favor use of antibiotics for proven streptococcal pharyngitis.

    The evidence probably needs to be better before we truly change the current accepted standard of care.

  10. Great review. My question is one that I have not found addressed much on risk/benefit analyses – and that is the reduction in the timeframe for which one is contagious. The classic teaching is that you are no longer contagious 24 hours after starting antibiotics, and one study showed that kids who started antibiotics by 5pm could go to school the next day with no increased risk of spread. In untreated strep, I have seen everything from “stay home until symptoms completely resolve” (which can be 7 days) to “there is potential to spread for 2-3 weeks”. In terms of when parents can send their kids back to school/daycare, this is a huge issue (and even in terms of preventing spread within a family). Taking 2 days off work until kid #1 has been on antibiotics for 24 hours is an entirely different beast than taking 2 weeks off because 3 kids have had overlapping symptoms for 14 days! Not to mention, most parents are going to send their kids back to school when they’re “mostly better” whether or not they’ve had antibiotics, in my mind risking ongoing spread, but seem more likely to abide by the “they can go back after 24 hours of antibiotics” rule. I’d love to hear your thoughts on this!

    • Sarah – this is a great question and one that, I’ll have to admit, didn’t occur to me when I wrote the post. The truth is that we simply don’t know. I’m not clear on how robust the data is showing exactly when you’re no longer infectious but the recent article saying 12 hours after 1st dose antibiotics appears to be the best we have. For now, this may be the only good reason to treat strep throat in developed countries that I can think of. Thanks for sending along your thoughts!


    Unfortunate that evidence-based practices don’t seem to take root and influence our decisions among us doctors.

    I’d like to investigate how many patients visiting AmericanWell get antibiotics from doctors who make a diagnosis by asking questions and looking at patients’ throats on the screen… (I’m all for digital health, this is however not the beneficial kind.)

    Take a look:

    Thanks a lot for this very well-written, important articles.

    The comment about sub-populations could be added to the article, even though it is implied to indirectly.

  12. There are 4 reasons to treat strep pharyngitis.

    1) Prevent rheumatic fever – currently rare disease in US, But still recommended major organizations AAP, CDC re diagnosis and management,

    2) Reduction in transmission – definitely useful. Don’t want my child infected by another with untreated strep pharyngitis.

    3) Rapid improvement in symptoms. If you have ever had strep pharyngitis (severe throat pain, fever, headache, vomiting), and it was treated with narrow spectrum antibiotics (penicillin), and symptoms improved in < 12-16 hours, you would want this therapy.
    Steroids plus antibiotics are the major interventions in the RCTs looking at PO antibiotics alone v PO antibiotics plus steroids.
    The data on steroids alone, especially for non strep pharyngitis show a lack of efficacy.

    Other issues – IMO the rate of severe allergic reactions from oral antibiotics (0.24%) is an overestimate. In a recent publication the rate is much lower 1 out of several million. (Journal of Antimicrobial Chemotherapy, Volume 60, Issue 5, November 2007, Pages 1172–1173,

    J Allergy Clin Immunol. 2010 May;125(5):1098-1104.e1. doi: 10.1016/j.jaci.2010.02.009. Epub 2010 Apr 14. Anaphylaxis epidemiology in patients with and patients without asthma: a United Kingdom database review. 21 / 100,000 person years.

    Test and Treat those at highest risk – Use Clinical Risk (Centor Score) high enough, and then RAT positive,

    • Mike – thanks for the thoughts. I know we’ve had this discussion in the past and your points are appreciated.

      The risk of rheumatic fever in the US (and other developed countries with good public health) is minimal and the benefit of reduction is likely outweighed by the risk of an anaphylaxis based on the best available evidence.

      The reduction in symptoms is minimal. At best, we’re talking about 12-16 hours again based on the best evidence. Even that comparison is against nothing – not against NSAIDs or acetaminophen and definitely not against steroids.

      Reduction in transmission – Agree with this point and may be the only viable reason to treat in the US

  13. I used to be more skeptical of antibiotics for strep pharyngitis until I started working in Kenya and started seeing a lot of RHD. The evidence available suggests that antibiotics do prevent RF – the debate is whether the NNT outweighs the cost and NNH depending on the setting. The difference in rates of strep pharyngitis, RF and RHD between countries has not really been fully explained by income level or hygiene and sanitation alone, or by differing virulence between strains. I don’t think we should entirely dismiss the role of PCN in reducing mortality from RHD in the US from the 1920s-1950s. Improvements in sanitation, hygiene and socioeconomic status can explain historic reductions in mortality for almost any disease, without completely negating the smaller but not insignificant role of medicine.

    One argument for antibiotics not included in this article is to reduce the risk of infecting others – particularly in low income settings where people may be living in close quarters at high risk of transmission. This may still be worth considering in impoverished regions of the US, such as Native American reservations, where rates of GAS and its complications remain high. Perhaps for our treatment guidelines to be effective, we should consider not only the clinical features of an individual patient, but also the socioeconomic status and living conditions of their community.

    More than anything, this highlights the gap in research applicable to LMICs and low-income populations elsewhere – when facing a very common disease that causes more than 33 million cases of RHD per year, why are we forced to make clinical practice decisions based 1950s data from adult males in the US?

    Global, Regional, and National Burden of Rheumatic Heart Disease, 1990–2015
    Antibiotics for the primary prevention of acute rheumatic fever: a meta-analysis
    Cochrane Review: Antibiotics for sore throat

    • Hello Sarah,
      TY for your comments. I agree that individual patient features should be evaluated on a case by case basis. However there have been multiple RCTs well after the 1950’s showing no cases of RF/RHD when treated with placebo as indicated in the post. In my 15 years of practicing medicine, one thing I have certainly learned is there are no absolutes and there are always exceptions. So to your point, yes socioeconomic status and living conditions more than what country you live in is important in this decision process. The final thing I would add, is the rise of shared decision making. This is something I employ with all patients and families on these controversial topics. I state the facts and let them decide which is more important to them.


  14. What advice might you give a mother of small children who tested positive for strep after several miserable days, but was already starting to feel better by the time the test results and prescription came in? How might she weigh the benefits of possible reduced transmission to children and others, vs. the costs of microbiome changes for her and a nursing 1-year-old baby? In general, is there a window during which it makes more sense to take the antibiotics?

    • Hello Mary,
      We appreciate you reading and writing us. Unfortunately we do not provide medical advice on this website. For these types of questions, I suggest you contact your doctor. Best of luck.



Leave a Comment

Time limit is exhausted. Please reload CAPTCHA.