Antibiotics for treatment of sore throat in children and adults
Abstract
Background
Sore throat is a common reason for people to present for medical care and to be prescribed antibiotics. Overuse of antibiotics in primary medicine is a concern, hence it is important to establish their efficacy in treating sore throat and preventing secondary complications.
Objectives
To assess the effects of antibiotics for reducing symptoms of sore throat for child and adult patients.
Search methods
We searched CENTRAL 2021, Issue 2, MEDLINE (January 1966 to April week 1, 2021), Embase (January 1990 to April 2021), and two trial registries (searched 6 April 2021).
Selection criteria
Randomised controlled trials (RCTs) or quasi‐RCTs of antibiotics versus control assessing typical sore throat symptoms or complications amongst children and adults seeking medical care for sore throat symptoms.
Data collection and analysis
We used standard methodological procedures as recommended by Cochrane. Two review authors independently screened studies for inclusion and extracted data, resolving any differences in opinion by discussion. We contacted the trial authors from three studies for additional information. We used GRADE to assess the certainty of the evidence for the efficacy of antibiotics on our primary outcomes (sore throat at day three and one week) and secondary outcomes (fever and headache symptoms and incidence of acute rheumatic fever, acute glomerulonephritis, acute otitis media, acute sinusitis, and quinsy).
Main results
We included 29 trials with 15,337 cases of sore throat. The majority of included studies were conducted in the 1950s, during which time the rates of serious complications (especially acute rheumatic fever) were much higher than today. Although clinical antibiotic trials for sore throat and respiratory symptoms are still being conducted, it is unusual for them to include placebo or 'no treatment' control arms, which is a requirement for inclusion in the review.
The age of participants ranged from younger than one year to older than 50 years, but most participants across all studies were adults. Although all studies recruited patients presenting with symptoms of sore throat, few of them distinguished between bacterial and viral aetiology. Bias may have been introduced through non‐clarity in treatment allocation procedures and lack of blinding in some studies. Harms from antibiotics were poorly or inconsistently reported, and were thus not quantified for this review.
1. Symptoms
Throat soreness and headache at day three were reduced by using antibiotics, although 82% of participants in the placebo or no treatment group were symptom‐free by one week. The reduction in sore throat symptoms at day three (risk ratio (RR) 0.70, 95% confidence interval (CI) 0.60 to 0.80; 16 studies, 3730 participants; moderate‐certainty evidence) was greater than at one week in absolute numbers (RR 0.50, 95% CI 0.34 to 0.75; 14 studies, 3083 participants; moderate‐certainty evidence) due to many cases in both treatment groups having resolved by this time. The number needed to treat for an additional beneficial outcome (NNTB) to prevent one sore throat at day three was less than six; at week one it was 18. Compared with placebo or no treatment, antibiotics did not significantly reduce fever at day three (RR 0.75, 95% CI 0.53 to 1.07; 8 studies, 1443 participants; high‐certainty evidence), but did reduce headache at day three (RR 0.49, 95% CI 0.34 to 0.70; 4 studies, 1020 participants; high‐certainty evidence).
2. Suppurative complications
Whilst the prevalence of suppurative complications was low, antibiotics reduced the incidence of acute otitis media within 14 days (Peto odds ratio (OR) 0.21, 95% CI 0.11 to 0.40; 10 studies, 3646 participants; high‐certainty evidence) and quinsy within two months (Peto OR 0.16, 95% CI 0.07 to 0.35; 8 studies, 2433 participants; high‐certainty evidence) compared to those receiving placebo or no treatment, but not acute sinusitis within 14 days (Peto OR 0.46, 95% CI 0.10 to 2.05; 8 studies, 2387 participants; high‐certainty evidence).
3. Non‐suppurative complications
There were too few cases of acute glomerulonephritis to determine whether there was a protective effect of antibiotics compared with placebo against this complication (Peto OR 0.07, 95% CI 0.00 to 1.32; 10 studies, 5147 participants; low‐certainty evidence). Antibiotics reduced acute rheumatic fever within two months when compared to the control group (Peto OR 0.36, 95% CI 0.26 to 0.50; 18 studies, 12,249 participants; moderate‐certainty evidence). It should be noted that the overall prevalence of acute rheumatic fever was very low, particularly in the later studies.
Authors' conclusions
Antibiotics probably reduce the number of people experiencing sore throat, and reduce the likelihood of headache, and some sore throat complications. As the effect on symptoms can be small, clinicians must judge on an individual basis whether it is clinically justifiable to use antibiotics to produce this effect, and whether the underlying cause of the sore throat is likely to be of bacterial origin. Furthermore, the balance between modest symptom reduction and the potential hazards of antimicrobial resistance must be recognised. Few trials have attempted to measure symptom severity. If antibiotics reduce the severity as well as the duration of symptoms, their benefit will have been underestimated in this meta‐analysis. Additionally, more trials are needed in low‐income countries, in socio‐economically deprived sections of high‐income countries, as well as in children.
PICOs
Plain language summary
Antibiotics for adults and children with sore throats
Review question
Are antibiotics effective in treating the symptoms and reducing the potential complications associated with sore throats?
Background
Sore throats are infections caused by bacteria or viruses. Pain or discomfort is the most distinguishing feature. However, fever and headache are also common accompanying symptoms. People usually recover quickly (after three or four days), although some develop complications. A serious but rare complication is rheumatic fever, which affects the heart and joints. Other complications include acute infection of the sinuses, middle ear, tonsils, and kidney. Antibiotics reduce infections caused by bacteria, but not those caused by viruses, and they can cause diarrhoea, rash, and other adverse effects. In addition, communities build resistance to them.
Search date
The evidence is current to April 2021.
Study characteristics
The 2021 update includes 29 trials with 15,337 cases of sore throat. All of the included studies were randomised controlled trials (a type of study where participants are randomly assigned to one of two or more treatment groups) that sought to determine if antibiotics helped reduce symptoms of sore throat, fever, or headache or the occurrence of more serious complications. The included studies were conducted in both children and adults seeking medical care for their symptoms.
Study funding sources
Many of the early studies were funded by the United States Armed Forces and recruited young, adult male military personnel. Later studies were mostly supported by governmental research grants, with a small number funded by private pharmaceutical companies.
Key results
We found that antibiotics reduced the number of people still experiencing headache on the third day of illness. Antibiotics probably reduced the number of people with sore throat after three days and one week, as well as rheumatic fever within two months in communities where this complication is common. Our confidence in the evidence for antibiotic use varied from low to high for other types of complications associated with sore throat.
Certainty of evidence
Overall, the certainty of the evidence from the included studies was low to high. However, there were very few recent trials included in the review, hence it is unclear if changes in bacterial resistance in the community may have affected the effectiveness of antibiotics.
Authors' conclusions
Summary of findings
| Antibiotics compared with control for sore throat | ||||||
| Patient or population: adults and children presenting with sore throat Settings: community Intervention: antibiotics Comparison: control (placebo or no treatment) | ||||||
| Outcomes | Anticipated absolute effects (95% CI) | Relative effect | No. of participants | Certainty of the evidence | Comments | |
|---|---|---|---|---|---|---|
| Risk with control | Risk with antibiotics
| |||||
| Sore throat: day 3 | 660 | 462 (396 to 528) | RR 0.70 (0.60 to 0.80) | 3730 | ⊕⊕⊕⊝ |
|
| Sore throat: 1 week | 190 | 95 (65 to 143) | RR 0.50 (0.34 to 0.75) | 3083 | ⊕⊕⊕⊝ |
|
| Fever: day 3 | 197 | 148 (104 to 211) | RR 0.75 (0.53 to 1.07) | 1443 | ⊕⊕⊕⊕ |
|
| Headache: day 3 | 421 | 206 (143 to 295) | RR 0.49 (0.34 to 0.70) | 1020 | ⊕⊕⊕⊕ |
|
| Rheumatic fever (within 2 months, clinical diagnosis) | 190 | 61 (34 to 110) | Peto OR 0.32 (0.18 to 0.58) | 12,132 | ⊕⊕⊕⊝ | Based largely on risk in pre‐1960 trials |
| Glomerulonephritis (within 1 month, clinical diagnosis) | 1 | 0 (0 to 2) | Peto OR 0.07 (0.00 to 1.32) | 5147 | ⊕⊝⊝⊝ | Sparse data: 2 cases only in the placebo group |
| Quinsy (within 2 months, clinical diagnosis) | 23 | 3 (1 to 11) | Peto OR 0.16 (0.07 to 0.35) | 2367 | ⊕⊕⊕⊕ |
|
| Otitis media (within 14 days, clinical diagnosis) | 20 | 5 (3 to 11) | Peto OR 0.21 (0.11 to 0.40) | 3646 | ⊕⊕⊕⊕ |
|
| The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). | ||||||
| GRADE Working Group grades of evidence | ||||||
| aDowngraded one level for high level of heterogeneity. | ||||||
Background
Description of the condition
Sore throat is a very common reason for people to attend primary care settings (Finley 2018). Sore throat is a disease that resolves spontaneously, that is 'cure' is not dependent on treatment (Del Mar 1992c). Nonetheless, primary care doctors commonly prescribe antibiotics for sore throat and other upper respiratory tract infections. There are large differences in clinical practice between countries, Froom 1990; Tyrstrup 2017, and between primary care doctors (Howie 1971).
Description of the intervention
The administration of antibiotics is likely to shorten the time to the remittance of symptoms and reduce the likelihood of complications in patients whose sore throat has a bacteriological aetiology (van Driel 2016). However, their benefits may be limited in the treatment of sore throat more generally (Reveiz 2013). Doctors have traditionally attempted to determine whether the cause of the infection is bacterial, especially when caused by the group A beta‐haemolytic Streptococcus (GABHS), which can cause acute rheumatic fever and acute glomerulonephritis. However, determining the aetiological agent is difficult (Del Mar 1992b).
How the intervention might work
Antibiotics target bacteria which are potentially responsible for sore throat symptoms and possible subsequent suppurative (pus producing) and non‐suppurative sequelae or complications. Successful eradication of bacteria may promote faster healing and the prevention of secondary complications. However, not all sore throat cases are of bacteriologic origin, and bacteria may resist antibiotic treatment, which could limit the overall effectiveness of the intervention.
Why it is important to do this review
Whether or not to prescribe antibiotics for sore throat is controversial, and it has been estimated that antibiotic prescription rates for sore throat exceed appropriate levels (Smith 2018). The issue is important for several reasons. Sore throat is a very common disease, and differences in prescribing result in large cost differences. In addition, increased prescribing increases patient attendance rates (Howie 1978; Little 1997) which may result in unnecessary visits to primary care for mild illness that does not require medical intervention. Furthermore, antibiotic resistance from overuse is an issue of great concern (Ciorba 2015). This 2021 update was built on an early meta‐analysis, Del Mar 1992a, and is an update of previous Cochrane Reviews (Del Mar 1997; Del Mar 2004; Del Mar 2006; Spinks 2013).
Objectives
To assess the effects of antibiotics for reducing symptoms of sore throat for child and adult patients.
Methods
Criteria for considering studies for this review
Types of studies
Randomised controlled trials (RCTs) or quasi‐RCTs.
Types of participants
Adults and children seeking medical care for symptoms of sore throat.
Types of interventions
Antibiotics or control consisting of either no treatment or placebo.
Types of outcome measures
Primary outcomes
-
Symptoms of sore throat on day three.
-
Symptoms of sore throat at one week (days six to eight).
Secondary outcomes
-
Symptoms of fever at day three.
-
Symptoms of headache at day three.
-
Incidence of suppurative (pus‐producing) complications:
-
quinsy (peritonsillar abscess);
-
acute otitis media (middle ear infection);
-
acute sinusitis.
-
-
Incidence of non‐suppurative complications:
-
acute rheumatic fever within two months;
-
acute glomerulonephritis (inflammation of the kidney) within one month.
-
Search methods for identification of studies
Electronic searches
For this update we searched the Cochrane Central Register of Controlled Trials (CENTRAL) 2021, Issue 2, part of the Cochrane Library (www.thecochranelibrary.com) (accessed 6 April 2021), which contains the Cochrane Acute Respiratory Infections Group's Specialised Register; MEDLINE (January 1966 to April week 1, 2021), and Embase (January 1990 to April 2021). See Appendix 1 and Appendix 2 for details of the search strategy and Appendix 3 for details of previous searches.
Searching other resources
We searched ClinicalTrials.gov (clinicaltrials.gov) and the World Health Organization International Clinical Trials Registry Platform (www.who.int/clinical-trials-registry-platform) on 6 April 2021 for completed and ongoing trials. We hand checked references of selected studies and relevant reviews to find additional studies.
Data collection and analysis
Selection of studies
Two review authors (AS, CDM) independently screened the abstracts of studies identified by the search and retrieved the full‐text articles for those deemed potentially relevant. Two review authors (AS, CDM) examined the full‐text articles and either included studies or excluded studies, providing reasons for exclusion of the latter studies.
Data extraction and management
Two review authors (AS, CDM) independently extracted data from the included studies based on patient‐relevant outcomes, namely the complications and symptoms listed above. We performed data extraction from tables, graphs, and in some cases through contact with the trial authors for raw data (Dagnelie 1996; de la Poza Abad 2016; Little 1997; Zwart 2000; Zwart 2003).
Assessment of risk of bias in included studies
We assessed risk of bias according to the methods described in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). We used the following six criteria to judge risk of bias: adequate sequence generation, allocation concealment, blinding, incomplete outcome data, selective reporting, and other bias.
Measures of treatment effect
All treatment effect outcomes were dichotomous data, reported as risk ratios (RR). We reported the occurrence of complications during the study period for suppurative and non‐suppurative complications. We assessed the presence of symptoms (sore throat, fever, headache) when possible at day three and week one (days six to eight). We also calculated the number needed to treat for an additional beneficial outcome (NNTB) for the primary outcomes.
Unit of analysis issues
The unit of analysis was the individual participant, and all included studies performed randomisation at the participant level, therefore no adjustments were necessary for the effect of clustering.
Dealing with missing data
We performed an intention‐to treat analysis for all outcomes.
Assessment of heterogeneity
We assessed heterogeneity using the Chi2 test, with the significance level set at 0.1. We determined the effect of heterogeneity by the I2 statistic, which indicates the proportion of total variability that can be explained by heterogeneity. We interpreted values of the I2 statistic greater than 50% as indicating substantial heterogeneity, in accordance with the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011).
Assessment of reporting biases
Where possible, we used funnel plots to investigate the possibility of reporting bias. We produced funnel plots when there were sufficient data (10 or more studies).
Data synthesis
Where possible, we combined data in order to perform meta‐analyses to report RR for all relevant outcomes. We used a random‐effects meta‐analytical method (Mantel‐Haenszel) to account for heterogeneity detected using the methods described above. Not all studies were able to contribute data to each of the meta‐analyses performed.
Subgroup analysis and investigation of heterogeneity
We performed a series of subgroup analyses to assess the differences in outcomes across various subgroups within the participant population:
-
treatment with penicillin (omitting other antibiotics);
-
children compared with adults; and
-
positive throat swab versus negative throat swab versus untested and/or inseparable data for group A beta‐haemolytic Streptococcus (GABHS).
Sensitivity analysis
We performed sensitivity analyses to assess the degree to which results were influenced by the following criteria:
-
early (pre‐1975) versus later (post‐1975) studies; and
-
blinded versus unblinded studies.
Summary of findings and assessment of the certainty of the evidence
We created a summary of findings table for the following outcomes: symptoms of sore throat on day three, symptoms of sore throat at one week, incidence of rheumatic fever, incidence of acute glomerulonephritis, incidence of quinsy, and incidence of otitis media (see summary of findings Table 1). We used the five GRADE considerations (study limitations, consistency of effect, imprecision, indirectness, and publication bias) to assess the certainty of evidence as it relates to the studies contributing data to the meta‐analyses (Atkins 2004). We used the methods and recommendations described in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011), employing GRADEpro GDT software (GRADEpro GDT). We justified all decisions to down‐ or upgrade the certainty of the evidence using footnotes, and made comments to aid the reader's understanding of the review where necessary.
Results
Description of studies
See Characteristics of included studies and Characteristics of excluded studies tables.
Results of the search
We considered a total of 64 studies for the review. The PRISMA flow diagram for studies considered in the 2021 update is shown in Figure 1. PRISMA is an evidence‐based minimum set of items for reporting in systematic reviews. Twenty‐nine randomised studies met the inclusion criteria of the review, three of which were added in this 2021 update. Of these, two were historic trials that contributed data to the effect of antibiotics on rheumatic fever incidence (Brock 1953; Houser 1953), and one was a recent trial of various antibiotic prescribing regimens for adults with acute respiratory infections (de la Poza Abad 2016). Not all participants in this trial presented with sore throat at the time of enrolment, hence the authors were contacted to provide a subset of trial data that met the inclusion criteria for the review.
Study flow diagram for the 2021 review update.
Included studies
The included studies investigated a total of 15,337 cases of sore throat. The majority of studies were conducted in the 1950s, during which time the rates of serious complications (especially acute rheumatic fever) were much higher than they are today. Eight studies have been published since 1996, with only four studies occurring since 2000. Although clinical antibiotic trials for sore throat and respiratory symptoms are still being conducted, it is unusual for them to include placebo or 'no treatment' control arms, which was a requirement for inclusion in the review.
The age of participants ranged from less than one year to older than 50 years. The participants of eight early studies were young male recruits from the United States Air Force (Brink 1951; Brock 1953; Brumfitt 1957; Catanzaro 1954; Chamovitz 1954; Denny 1950; Denny 1953; Houser 1953; MacDonald 1951; Wannamaker 1951). Six of the remaining studies recruited children up to 18 years of age only (El‐Daher 1991; Krober 1985; Nelson 1984; Siegel 1961; Taylor 1977; Zwart 2000); four recruited only adults or adolescents aged 15 years or over (de la Poza Abad 2016; Howe 1997; Petersen 1997; Zwart 2003); and nine studies recruited both adults and children (Bennike 1951; Chapple 1956; Dagnelie 1996; De Meyere 1992; Landsman 1951; Leelarasamee 2000; Little 1997; Middleton 1988; Whitfield 1981).
All studies recruited patients presenting with symptoms of sore throat. The majority of studies did not distinguish between bacterial and viral aetiology. However, seven studies included or analysed results for GABHS‐positive patients only (Catanzaro 1954; De Meyere 1992; El‐Daher 1991; Krober 1985; Middleton 1988; Nelson 1984); one study distinguished differences in outcomes between GABHS‐positive and ‐negative patients (Dagnelie 1996); and two studies specifically excluded patients who were GABHS‐positive (Petersen 1997; Taylor 1977).
Excluded studies
The most common reason for exclusion was lack of an appropriate control group (n = 15). Other reasons for exclusion were: irrelevant or non‐patient‐centred outcomes (n = 6), main complaint other than acute sore throat (n = 5), inappropriate or no randomisation to treatment (n = 5), an intervention other than antibiotics was being tested (n = 2), the study tracked natural course of illness only (n = 1), or the study reported previously published data already included (n = 1).
Risk of bias in included studies
The overall risk of bias is presented graphically in Figure 2 and summarised in Figure 3.
Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.
Risk of bias summary: review authors' judgements about each risk of bias item for each included study.
We could produce funnel plots to assess the potential for reporting bias for four outcomes, which are shown in Figure 4 (Symptom of sore throat on day three); Figure 5 (Symptom of sore throat at one week); Figure 6 (Incidence of acute rheumatic fever); and Figure 7 (Incidence of otitis media). Visual inspection of the plots indicated the potential for publication bias related to studies reporting the incidence of sore throat complications.
Funnel plot of comparison: 1 Antibiotics versus control for the treatment of sore throats: symptom of sore throat, outcome: 1.1 Symptom of sore throat on day 3.
Funnel plot of comparison: 1 Antibiotics versus control for the treatment of sore throat: symptom of sore throat, outcome: 1.6 Symptom of sore throat at 1 week (6 to 8 days).
Funnel plot of comparison: 4 Antibiotics versus control for the treatment of sore throat: incidence of complications, outcome: 4.1 Incidence of acute rheumatic fever within 2 months. Rheumatic fever defined by clinical diagnosis.
Funnel plot of comparison: 4 Antibiotics versus control for the treatment of sore throat: incidence of complications, outcome: 4.4 Incidence of otitis media within 14 days. Otitis media defined by clinical diagnosis.
Allocation
In most early studies, participants were randomised to treatment and control groups by methods that could potentially introduce bias (e.g. Air Force serial number, drawing a card from a deck, hospital bed number) or not randomised at all. Allocation methods were generally appropriate in the later studies.
Blinding
Fifteen studies were double‐blinded and seven were single‐blinded, whilst seven studies involved no blinding of participants or study personnel.
Incomplete outcome data
Outcome data were complete for nearly all of the included studies. It was not clear in one study how many participants maintained pain score diaries, and some participants who were initially randomised were excluded due to being GABHS‐positive (Petersen 1997).
Selective reporting
Nearly all of the studies had a low risk of bias related to selective reporting of outcomes.
Other potential sources of bias
The use of antipyretic analgesics was not stated in nine studies, administered routinely in five studies, and prohibited in four studies. The prohibition of analgesics might exaggerate any small symptomatic benefit of antibiotics over control if antipyretic analgesics are usually recommended in normal practice. Funding arrangements contributed another potential source of bias to four studies that were supported by pharmaceutical companies; this may have influenced outcomes for these studies.
Effects of interventions
See: Summary of findings 1 Antibiotics compared with control for sore throat
Primary outcomes
1. Symptoms of sore throat on day three
At day three of the illness, antibiotics reduced symptoms of sore throat (risk ratio (RR) 0.70, 95% confidence interval (CI) 0.60 to 0.80; 16 studies, 3730 participants; moderate‐certainty evidence; Analysis 1.1).
2. Symptoms of sore throat at one week (days six to eight)
At one week (six to eight days), the RR of experiencing sore throat was 0.50 (95% CI 0.34 to 0.75; 14 studies, 3083 participants; moderate‐certainty evidence; Analysis 1.6), although 82% of controls were better by this time.
Secondary outcomes
1. Symptoms of fever at day three
At day three of the illness, antibiotics did not significantly reduce symptoms of fever (RR 0.75, 95% CI 0.53 to 1.07; 8 studies, 1443 participants; high‐certainty evidence; Analysis 2.1).
2. Symptoms of headache at day three
At day three of the illness, antibiotics reduced symptoms of headache (RR 0.49, 95% CI 0.34 to 0.70; 4 studies, 1020 participants; high‐certainty evidence; Analysis 3.1).
3. Incidence of suppurative complications
The incidence of acute otitis media within 14 days was lower in participants receiving antibiotics compared with those in the control group (Peto odds ratio (OR) 0.21, 95% CI 0.11 to 0.40; 10 studies, 3646 participants; high‐certainty evidence; Analysis 4.4). However, antibiotic use made little or no difference to the incidence of acute sinusitis within 14 days (Peto OR 0.46, 95% CI 0.10 to 2.05; 7 studies, 2270 participants; high‐certainty evidence; Analysis 4.6). Data indicated that the incidence of quinsy within two months was also reduced in relation to the control group (Peto OR 0.16, 95% CI 0.07 to 0.35; 7 studies, 2367 participants; high‐certainty evidence; Analysis 4.7).
4. Incidence of non‐suppurative complications
Only 10 studies reported on acute glomerulonephritis as an endpoint, and only two cases occurred, both in the control group. Our estimate of the protection has a very wide 95% CI, including the possibility of no difference between treatment and control groups (Peto OR 0.07, 95% CI 0.00 to 1.32; 10 studies, 5147 participants; low‐certainty evidence; Analysis 4.8), which precludes us from definitively claiming that antibiotics protect sore throat sufferers from acute glomerulonephritis.
Several studies found a benefit of antibiotics for acute rheumatic fever within two months. The incidence of this complication was lower in people treated with antibiotics versus those in the control group (Peto OR 0.36, 95% CI 0.26 to 0.50; 17 studies, 12,132 participants; moderate‐certainty evidence; Analysis 4.1). Few studies examined antibiotics other than penicillin.
A selection of these results is summarised in summary of findings Table 1.
A trial from Thailand was included in the 2003 update (Leelarasamee 2000). This trial is especially important because it is one of the few trials from a non‐Western industrial country. Unfortunately, we were unable to enter data from this study into the meta‐analysis due to the different ways in which data were collected (in particular no data were collected mid‐way through the illness). Nevertheless, the use of antibiotics conferred no benefit (or harm) on symptoms or complications.
Subgroup analysis of symptom reduction and prevention of complications
1. Treatment with penicillin (omitting other antibiotics) for preventing rheumatic fever
Most studies (15 out of 18) assessing the efficacy of antibiotics on preventing rheumatic fever within two months used penicillin. Confining the analysis to these studies alone resulted in less heterogeneity as indicated by the I2 statistic (1%) and a slightly higher estimated protection (Peto OR 0.26, 95% CI 0.18 to 0.40; 14 studies, 8407 participants; moderate‐certainly evidence; Analysis 4.2) than the main analysis for the effect of antibiotics in preventing rheumatic fever (Peto OR 0.36, 95% CI 0.26 to 0.50; 17 studies, 12,132 participants; moderate‐certainty evidence; Analysis 4.1).
2. Children versus adults
Few of the included studies involved children (younger than 13 years of age): only 61 cases in total for when fever was evaluated at day three. There was overlap of the RR 95% CI, so that the trend for children to not experience benefits was not significantly different to adults who did (RR 1.27, 95% CI 0.76 to 2.13; 2 studies, 61 participants; moderate‐certainty evidence, and RR 0.48, 95% CI 0.21 to 1.10; 3 studies, 705 participants; moderate‐certainty evidence, respectively; Analysis 2.3).
3. Throat swabs positive for Streptococcus versus negative for Streptococcus versus not tested and/or inseparable combined data
The probability of still experiencing pain on day three was slightly more than one‐half (RR 0.58, 95% CI 0.48 to 0.71; 11 studies, 1839 participants; moderate‐certainty evidence) for those participants who had positive throat swabs for GABHS, compared to three‐quarters (RR 0.78, 95% CI 0.63 to 0.97; 6 studies, 736 participants; moderate‐certainty evidence; Analysis 1.5) for those with negative swabs. At one week the efficacy of antibiotics was similarly higher for participants with positive GABHS swabs (RR 0.29, 95% CI 0.12 to 0.70; 7 studies, 1117 participants; moderate‐certainty evidence) compared with negative swabs (RR 0.73, 95% CI 0.50 to 1.07; 5 studies, 541 participants; moderate‐certainty evidence; Analysis 1.9).
Sensitivity analysis of symptom reduction and prevention of complications
1. Early (pre‐1975) versus later (post‐1975) studies
We performed analyses to compare the effect of early (pre‐1975) versus late (post‐1975) studies on symptoms of sore throat at day three (Analysis 1.2) and one week (Analysis 1.7). Although there was a trend for earlier studies to have a greater effect on reducing sore throat symptoms as day three (RR 0.62, 95% CI 0.56 to 0.69; 6 studies, 1141 participants; moderate‐certainty evidence) than later studies (RR 0.72, 95% CI 0.60 to 0.88; 10 studies, 2589 participants; moderate‐certainty evidence), the confidence intervals for the two time periods overlapped, meaning this trend was not significant. In contrast, earlier studies found a greater effect on reducing sore throat symptoms at one week (RR 0.14, 95% CI 0.08 to 0.27; 6 studies, 1140 participants; moderate‐certainty evidence) than later studies (RR 0.73, 95% CI 0.55 to 0.97; 8 studies, 1943 participants; moderate‐certainty evidence).
We performed analyses to compare the effect of early (pre‐1975) versus late (post‐1975) studies on the incidence of rheumatic fever within two months (Analysis 4.3), and otitis media within 14 days (Analysis 4.5). Whilst early studies showed that antibiotics probably reduced incidence of rheumatic fever (Peto OR 0.30, 95% CI 0.20 to 0.45; 10 studies, 7617 participants; moderate‐certainty evidence), the effect of antibiotics in later studies could not be calculated due to the absence of cases amongst either antibiotic‐treated or control participants (Peto OR not estimable; 5 studies, 2367 participants; low‐certainty evidence).
There was a subgroup difference between the effectiveness of antibiotics in reducing the incidence of otitis media for early studies (Peto OR 0.25, 95% CI 0.12 to 0.52; 5 studies, 1837 participants; high‐certainty evidence) versus late studies (Peto OR 0.05, 95% CI 0.01 to 0.31; 6 studies, 1923 participants; moderate‐certainty evidence). The certainty of the evidence was lower for later studies due to the wider confidence interval.
2. Blind versus unblinded studies
There was probably little or no difference between blinded and unblinded studies for symptoms of sore throat at day three (RR 0.65, 95% CI 0.54 to 0.78; 12 studies, 2662 participants; moderate‐certainty evidence, and RR 0.81, 95% CI 0.65 to 1.01; 4 studies, 1068 participants; moderate‐certainty evidence, respectively; Analysis 1.3) or at one week (RR 0.62, 95% CI 0.38 to 1.03; 9 studies, 1616 participants; moderate‐certainty evidence, and RR 0.43, 95% CI 0.20 to 0.91; 5 studies, 1437 participants; moderate‐certainty evidence, respectively; Analysis 1.8).
Discussion
Summary of main results
Antibiotics reduce symptoms and the likelihood of complications in the treatment of sore throat. However, the absolute benefits are modest.
Symptoms
Headache symptoms on the third day of treatment were reduced when using antibiotics; throat soreness was probably also reduced with antibiotics. The NNTB to prevent one sore throat was less than six at day three and 18 at one week. Antibiotics were more effective against symptoms at day three and one week if throat swabs were positive for GABHS compared to negative throat swabs.
In the control (placebo and no treatment) groups, after three days symptoms of sore throat and fever had disappeared in about 40% and 85% of participants respectively. Eighty‐two per cent of participants were symptom‐free by one week. This natural history was similar in Streptococcus‐positive, ‐negative, and untested participants. Regarding estimates of the number of people with sore throat who must be treated to resolve the symptoms of one person by day three, the NNTB was about 3.7 for those with positive throat swabs for Streptococcus; 6.5 for those with a negative swab; and 14.4 for those in whom no swab was taken. The last result is difficult to understand, as intuitively one would expect the NNTB value to lie between both the swab‐negative and swab‐positive results. Perhaps participants with less severe throat infections were recruited into the three studies in which swabs were not taken.
Non‐suppurative and suppurative complications
There were too few cases of acute glomerulonephritis to permit a determination as to whether antibiotics protected against this complication. Moderate‐certainty evidence showed that antibiotics probably reduce acute rheumatic fever within two months. However, few recent trials reported this outcome. About 1.7 per 100 placebo participants developed rheumatic fever in trials reporting before 1961. The background incidence of acute rheumatic fever has continued to decline in high‐income countries since then.
Antibiotics reduced the number of people experiencing acute otitis media and quinsy compared to the control group. However, antibiotic use made little or no difference compared to control for protecting against acute sinusitis.
In the included trials, conducted mostly in the 1950s, for every 100 participants treated with antibiotics, there was one fewer case of acute rheumatic fever, two fewer cases of acute otitis media, and three fewer cases of quinsy than in the control group. These figures need to be adapted to current circumstances and individuals. For example, the complication rate of acute otitis media amongst those with sore throats before 1975 was 3%. An NNTB of about 50 to prevent one case of acute otitis media can be estimated from the data. After 1975, this complication rate fell to 0.7%, and applying the odds of reducing the complication with antibiotics from the data table yields an NNTB of nearly 200 to prevent one case of acute otitis media. Clinicians will have to exercise judgement in applying these data to treatment decisions.
Adverse effects of treatment
We were unable to present the adverse effects of antibiotic use because of inconsistencies in the recording of these symptoms. In other studies these were principally diarrhoea, rashes, and thrush (Venekamp 2015). Consideration of the side effects of antibiotics would have been useful in further defining their risk‐benefits.
Overall completeness and applicability of evidence
The majority of trials included in this 2021 review update were conducted prior to 1975, with only four trials published since 2000. The main reason for this is that very few antibiotic trials conducted recently include a placebo control arm. It is therefore unknown whether changes in bacterial resistance and population immunity over time may have altered the applicability of results.
Special risk groups
Acute rheumatic fever is common amongst people living in some parts of the world (Australian Indigenous populations living in low socio‐economic conditions, for example), and antibiotics may be justified to reduce the complication of acute rheumatic fever in these settings. In other parts of the world the incidence of acute rheumatic fever is so low (one estimate is that it took 12 general practitioners' working lifetimes to encounter one new case of acute rheumatic fever in Western Scotland in the 1980s (Howie 1985)) that the risks of serious complications arising from the use of antibiotics for sore throat might be of the same order as that of acute rheumatic fever.
Quality of the evidence
We assessed the certainty of the evidence as low to high. The greatest compromise to the certainty of the evidence arose from non‐clarity in treatment allocation procedures and lack of blinding in some studies. Heterogeneity also played a role in downgrading the certainty of evidence for the effect of antibiotics on sore throat symptoms.
Potential biases in the review process
Non‐reporting of antipyretic use in a high number of studies may have constituted a source of bias in the results. Publication bias may also be considered a potential threat to the validity of results, particularly for the earlier studies.
Agreements and disagreements with other studies or reviews
A recent review analysing the risk‐benefit profile of antimicrobial prescribing for children concluded that antibiotics show little benefit in preventing quinsy following sore throat (Keith 2010). A clinical evidence review of antibiotic treatment for streptococcal pharyngitis concluded that amongst patients with signs and symptoms of positive bacterial infection, a specific diagnosis should be determined by performing either a throat culture or rapid antigen‐detection test, especially in children (Wessels 2011). Antibiotic treatment with penicillin or a first‐generation cephalosporin is then recommended in the case of positive bacteriologic assessment.
Study flow diagram for the 2021 review update.
Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.
Risk of bias summary: review authors' judgements about each risk of bias item for each included study.
Funnel plot of comparison: 1 Antibiotics versus control for the treatment of sore throats: symptom of sore throat, outcome: 1.1 Symptom of sore throat on day 3.
Funnel plot of comparison: 1 Antibiotics versus control for the treatment of sore throat: symptom of sore throat, outcome: 1.6 Symptom of sore throat at 1 week (6 to 8 days).
Funnel plot of comparison: 4 Antibiotics versus control for the treatment of sore throat: incidence of complications, outcome: 4.1 Incidence of acute rheumatic fever within 2 months. Rheumatic fever defined by clinical diagnosis.
Funnel plot of comparison: 4 Antibiotics versus control for the treatment of sore throat: incidence of complications, outcome: 4.4 Incidence of otitis media within 14 days. Otitis media defined by clinical diagnosis.
Comparison 1: Antibiotics versus control for the treatment of sore throat: symptoms of sore throat, Outcome 1: Symptom of sore throat on day 3
Comparison 1: Antibiotics versus control for the treatment of sore throat: symptoms of sore throat, Outcome 2: Symptom of sore throat on day 3: early (pre‐1975) versus late studies (post‐1975)
Comparison 1: Antibiotics versus control for the treatment of sore throat: symptoms of sore throat, Outcome 3: Symptom of sore throat on day 3: blind versus unblinded studies
Comparison 1: Antibiotics versus control for the treatment of sore throat: symptoms of sore throat, Outcome 4: Symptom of sore throat on day 3: antipyretics versus no antipyretics
Comparison 1: Antibiotics versus control for the treatment of sore throat: symptoms of sore throat, Outcome 5: Symptom of sore throat on day 3: GABHS‐positive throat swab, negative swab, untested/inseparable
Comparison 1: Antibiotics versus control for the treatment of sore throat: symptoms of sore throat, Outcome 6: Symptom of sore throat at 1 week (6 to 8 days)
Comparison 1: Antibiotics versus control for the treatment of sore throat: symptoms of sore throat, Outcome 7: Symptom of sore throat at 1 week (6 to 8 days): early (pre‐1975) versus late (post‐1975)
Comparison 1: Antibiotics versus control for the treatment of sore throat: symptoms of sore throat, Outcome 8: Symptom of sore throat at 1 week (6 to 8 days): blind versus unblinded studies
Comparison 1: Antibiotics versus control for the treatment of sore throat: symptoms of sore throat, Outcome 9: Symptom of sore throat at 1 week (6 to 8 days): GABHS‐positive throat swab, GABHS‐negative swab
Comparison 2: Antibiotics versus control for the treatment of sore throat: symptoms of fever, Outcome 1: Symptom of fever on day 3
Comparison 2: Antibiotics versus control for the treatment of sore throat: symptoms of fever, Outcome 2: Symptom of fever on day 3: blinded versus unblinded studies
Comparison 2: Antibiotics versus control for the treatment of sore throat: symptoms of fever, Outcome 3: Symptom of fever on day 3: children compared with adults
Comparison 2: Antibiotics versus control for the treatment of sore throat: symptoms of fever, Outcome 4: Symptom of fever at 1 week (6 to 8 days)
Comparison 3: Antibiotics versus control for the treatment of sore throat: symptoms of headache, Outcome 1: Symptom of headache on day 3
Comparison 3: Antibiotics versus control for the treatment of sore throat: symptoms of headache, Outcome 2: Symptom of headache on day 3: blinded versus unblinded studies
Comparison 4: Antibiotics versus control for the treatment of sore throat: incidence of complications, Outcome 1: Incidence of acute rheumatic fever within 2 months. Rheumatic fever defined by clinical diagnosis
Comparison 4: Antibiotics versus control for the treatment of sore throat: incidence of complications, Outcome 2: Incidence of acute rheumatic fever within 2 months. Penicillin versus placebo
Comparison 4: Antibiotics versus control for the treatment of sore throat: incidence of complications, Outcome 3: Incidence of acute rheumatic fever within 2 months: early (pre‐1975) versus late studies (post‐1975)
Comparison 4: Antibiotics versus control for the treatment of sore throat: incidence of complications, Outcome 4: Incidence of otitis media within 14 days. Otitis media defined by clinical diagnosis
Comparison 4: Antibiotics versus control for the treatment of sore throat: incidence of complications, Outcome 5: Incidence of otitis media within 14 days: early (pre‐1975) versus late studies (post‐1975)
Comparison 4: Antibiotics versus control for the treatment of sore throat: incidence of complications, Outcome 6: Incidence of sinusitis within 14 days. Sinusitis defined by clinical diagnosis
Comparison 4: Antibiotics versus control for the treatment of sore throat: incidence of complications, Outcome 7: Incidence of quinsy within 2 months. Quinsy defined by clinical diagnosis
Comparison 4: Antibiotics versus control for the treatment of sore throat: incidence of complications, Outcome 8: Incidence of acute glomerulonephritis within 1 month. Acute glomerulonephritis defined by clinical diagnosis
| Antibiotics compared with control for sore throat | ||||||
| Patient or population: adults and children presenting with sore throat Settings: community Intervention: antibiotics Comparison: control (placebo or no treatment) | ||||||
| Outcomes | Anticipated absolute effects (95% CI) | Relative effect | No. of participants | Certainty of the evidence | Comments | |
|---|---|---|---|---|---|---|
| Risk with control | Risk with antibiotics
| |||||
| Sore throat: day 3 | 660 | 462 (396 to 528) | RR 0.70 (0.60 to 0.80) | 3730 | ⊕⊕⊕⊝ |
|
| Sore throat: 1 week | 190 | 95 (65 to 143) | RR 0.50 (0.34 to 0.75) | 3083 | ⊕⊕⊕⊝ |
|
| Fever: day 3 | 197 | 148 (104 to 211) | RR 0.75 (0.53 to 1.07) | 1443 | ⊕⊕⊕⊕ |
|
| Headache: day 3 | 421 | 206 (143 to 295) | RR 0.49 (0.34 to 0.70) | 1020 | ⊕⊕⊕⊕ |
|
| Rheumatic fever (within 2 months, clinical diagnosis) | 190 | 61 (34 to 110) | Peto OR 0.32 (0.18 to 0.58) | 12,132 | ⊕⊕⊕⊝ | Based largely on risk in pre‐1960 trials |
| Glomerulonephritis (within 1 month, clinical diagnosis) | 1 | 0 (0 to 2) | Peto OR 0.07 (0.00 to 1.32) | 5147 | ⊕⊝⊝⊝ | Sparse data: 2 cases only in the placebo group |
| Quinsy (within 2 months, clinical diagnosis) | 23 | 3 (1 to 11) | Peto OR 0.16 (0.07 to 0.35) | 2367 | ⊕⊕⊕⊕ |
|
| Otitis media (within 14 days, clinical diagnosis) | 20 | 5 (3 to 11) | Peto OR 0.21 (0.11 to 0.40) | 3646 | ⊕⊕⊕⊕ |
|
| The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). | ||||||
| GRADE Working Group grades of evidence | ||||||
| aDowngraded one level for high level of heterogeneity. | ||||||
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1.1 Symptom of sore throat on day 3 Show forest plot | 16 | 3730 | Risk Ratio (M‐H, Random, 95% CI) | 0.70 [0.60, 0.80] |
| 1.2 Symptom of sore throat on day 3: early (pre‐1975) versus late studies (post‐1975) Show forest plot | 16 | 3730 | Risk Ratio (M‐H, Random, 95% CI) | 0.70 [0.60, 0.80] |
| 1.2.1 Symptom of sore throat on day 3: early (pre‐1975) studies | 6 | 1141 | Risk Ratio (M‐H, Random, 95% CI) | 0.62 [0.56, 0.69] |
| 1.2.2 Symptom of sore throat on day 3: late (post‐1975) studies | 10 | 2589 | Risk Ratio (M‐H, Random, 95% CI) | 0.72 [0.60, 0.88] |
| 1.3 Symptom of sore throat on day 3: blind versus unblinded studies Show forest plot | 16 | 3730 | Risk Ratio (M‐H, Random, 95% CI) | 0.70 [0.60, 0.80] |
| 1.3.1 Symptom of sore throat on day 3: blinded studies | 12 | 2662 | Risk Ratio (M‐H, Random, 95% CI) | 0.65 [0.54, 0.78] |
| 1.3.2 Symptom of sore throat on day 3: unblinded studies | 4 | 1068 | Risk Ratio (M‐H, Random, 95% CI) | 0.81 [0.65, 1.01] |
| 1.4 Symptom of sore throat on day 3: antipyretics versus no antipyretics Show forest plot | 5 | 1137 | Risk Ratio (M‐H, Random, 95% CI) | 0.58 [0.48, 0.70] |
| 1.4.1 Symptom of sore throat on day 3: antipyretics administered | 3 | 455 | Risk Ratio (M‐H, Random, 95% CI) | 0.52 [0.33, 0.81] |
| 1.4.2 Symptom of sore throat on day 3: no antipyretics administered | 2 | 682 | Risk Ratio (M‐H, Random, 95% CI) | 0.62 [0.55, 0.70] |
| 1.5 Symptom of sore throat on day 3: GABHS‐positive throat swab, negative swab, untested/inseparable Show forest plot | 15 | 3600 | Risk Ratio (M‐H, Random, 95% CI) | 0.68 [0.59, 0.78] |
| 1.5.1 Symptom of sore throat on day 3: GABHS‐positive throat swab | 11 | 1839 | Risk Ratio (M‐H, Random, 95% CI) | 0.58 [0.48, 0.71] |
| 1.5.2 Symptom of sore throat on day 3: GABHS‐negative throat swab | 6 | 736 | Risk Ratio (M‐H, Random, 95% CI) | 0.78 [0.63, 0.97] |
| 1.5.3 Symptom of sore throat on day 3: untested for GABHS culture or combined, inseparable data | 3 | 1025 | Risk Ratio (M‐H, Random, 95% CI) | 0.89 [0.80, 1.00] |
| 1.6 Symptom of sore throat at 1 week (6 to 8 days) Show forest plot | 14 | 3083 | Risk Ratio (M‐H, Random, 95% CI) | 0.50 [0.34, 0.75] |
| 1.7 Symptom of sore throat at 1 week (6 to 8 days): early (pre‐1975) versus late (post‐1975) Show forest plot | 14 | 3083 | Risk Ratio (M‐H, Random, 95% CI) | 0.50 [0.34, 0.75] |
| 1.7.1 Symptom of sore throat at 1 week (6 to 8 days): early (pre‐1975) studies | 6 | 1140 | Risk Ratio (M‐H, Random, 95% CI) | 0.14 [0.08, 0.27] |
| 1.7.2 Symptom of sore throat at 1 week (6 to 8 days): late (post‐1975) studies | 8 | 1943 | Risk Ratio (M‐H, Random, 95% CI) | 0.73 [0.55, 0.97] |
| 1.8 Symptom of sore throat at 1 week (6 to 8 days): blind versus unblinded studies Show forest plot | 14 | 3053 | Risk Ratio (M‐H, Random, 95% CI) | 0.57 [0.39, 0.82] |
| 1.8.1 Symptom of sore throat at 1 week (6 to 8 days): blinded studies | 9 | 1616 | Risk Ratio (M‐H, Random, 95% CI) | 0.62 [0.38, 1.03] |
| 1.8.2 Symptom of sore throat at 1 week (6 to 8 days): unblinded studies | 5 | 1437 | Risk Ratio (M‐H, Random, 95% CI) | 0.43 [0.20, 0.91] |
| 1.9 Symptom of sore throat at 1 week (6 to 8 days): GABHS‐positive throat swab, GABHS‐negative swab Show forest plot | 12 | 2524 | Risk Ratio (M‐H, Random, 95% CI) | 0.48 [0.29, 0.80] |
| 1.9.1 Symptom of sore throat at 1 week (6 to 8 days): GABHS‐positive throat swab | 7 | 1117 | Risk Ratio (M‐H, Random, 95% CI) | 0.29 [0.12, 0.70] |
| 1.9.2 Symptom of sore throat at 1 week (6 to 8 days): GABHS‐negative throat swab | 5 | 541 | Risk Ratio (M‐H, Random, 95% CI) | 0.73 [0.50, 1.07] |
| 1.9.3 Symptom of sore throat at 1 week (6 to 8 days): GABHS untested | 3 | 866 | Risk Ratio (M‐H, Random, 95% CI) | 0.35 [0.03, 4.47] |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 2.1 Symptom of fever on day 3 Show forest plot | 8 | 1443 | Risk Ratio (M‐H, Random, 95% CI) | 0.75 [0.53, 1.07] |
| 2.2 Symptom of fever on day 3: blinded versus unblinded studies Show forest plot | 8 | 1443 | Risk Ratio (M‐H, Random, 95% CI) | 0.75 [0.53, 1.07] |
| 2.2.1 Symptom of fever on day 3: blinded studies | 4 | 703 | Risk Ratio (M‐H, Random, 95% CI) | 0.82 [0.54, 1.23] |
| 2.2.2 Symptom of fever on day 3: unblinded studies | 4 | 740 | Risk Ratio (M‐H, Random, 95% CI) | 0.72 [0.43, 1.21] |
| 2.3 Symptom of fever on day 3: children compared with adults Show forest plot | 5 | 766 | Risk Ratio (M‐H, Random, 95% CI) | 0.63 [0.31, 1.26] |
| 2.3.1 Symptom of fever on day 3: children | 2 | 61 | Risk Ratio (M‐H, Random, 95% CI) | 1.27 [0.76, 2.13] |
| 2.3.2 Symptom of fever on day 3: adults | 3 | 705 | Risk Ratio (M‐H, Random, 95% CI) | 0.48 [0.21, 1.10] |
| 2.4 Symptom of fever at 1 week (6 to 8 days) Show forest plot | 4 | 886 | Risk Ratio (M‐H, Random, 95% CI) | 0.91 [0.55, 1.52] |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 3.1 Symptom of headache on day 3 Show forest plot | 4 | 1020 | Risk Ratio (M‐H, Random, 95% CI) | 0.49 [0.34, 0.70] |
| 3.2 Symptom of headache on day 3: blinded versus unblinded studies Show forest plot | 4 | 1020 | Risk Ratio (M‐H, Random, 95% CI) | 0.49 [0.34, 0.70] |
| 3.2.1 Symptom of headache on day 3: blinded studies | 2 | 436 | Risk Ratio (M‐H, Random, 95% CI) | 0.33 [0.09, 1.20] |
| 3.2.2 Symptom of headache on day 3: unblinded studies | 2 | 584 | Risk Ratio (M‐H, Random, 95% CI) | 0.57 [0.45, 0.72] |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 4.1 Incidence of acute rheumatic fever within 2 months. Rheumatic fever defined by clinical diagnosis Show forest plot | 17 | 12132 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 0.36 [0.26, 0.50] |
| 4.2 Incidence of acute rheumatic fever within 2 months. Penicillin versus placebo Show forest plot | 14 | 8407 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 0.26 [0.18, 0.40] |
| 4.3 Incidence of acute rheumatic fever within 2 months: early (pre‐1975) versus late studies (post‐1975) Show forest plot | 15 | 9984 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 0.30 [0.20, 0.45] |
| 4.3.1 Incidence of acute rheumatic fever within 2 months: early (pre‐1975) studies | 10 | 7617 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 0.30 [0.20, 0.45] |
| 4.3.2 Incidence of acute rheumatic fever within 2 months: late (post‐1975) studies | 5 | 2367 | Peto Odds Ratio (Peto, Fixed, 95% CI) | Not estimable |
| 4.4 Incidence of otitis media within 14 days. Otitis media defined by clinical diagnosis Show forest plot | 10 | 3646 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 0.21 [0.11, 0.40] |
| 4.5 Incidence of otitis media within 14 days: early (pre‐1975) versus late studies (post‐1975) Show forest plot | 10 | 3646 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 0.21 [0.11, 0.40] |
| 4.5.1 Incidence of otitis media within 14 days: early (pre‐1975) studies | 5 | 1837 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 0.25 [0.12, 0.52] |
| 4.5.2 Incidence of otitis media within 14 days: late (post‐1975) studies | 5 | 1809 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 0.05 [0.01, 0.31] |
| 4.6 Incidence of sinusitis within 14 days. Sinusitis defined by clinical diagnosis Show forest plot | 7 | 2270 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 0.46 [0.10, 2.05] |
| 4.7 Incidence of quinsy within 2 months. Quinsy defined by clinical diagnosis Show forest plot | 7 | 2367 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 0.16 [0.07, 0.35] |
| 4.8 Incidence of acute glomerulonephritis within 1 month. Acute glomerulonephritis defined by clinical diagnosis Show forest plot | 10 | 5147 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 0.07 [0.00, 1.32] |
