Short‐course antibiotics for acute otitis media

Anita L Kozyrskyj; Terry P Klassen; Michael Moffatt; Krystal Harvey

doi:10.1002/14651858.CD001095.pub2

. 2010 Sep 8;2010(9):CD001095. doi: 10.1002/14651858.CD001095.pub2

Short‐course antibiotics for acute otitis media

Anita L Kozyrskyj ¹, Terry P Klassen ^2,^✉, Michael Moffatt ³, Krystal Harvey ⁴

Editor: Cochrane Acute Respiratory Infections Group

PMCID: PMC7052812 PMID: 20824827

Abstract

Background

Acute otitis media (AOM) is a common illness during childhood, for which antibiotics are frequently prescribed.

Objectives

To determine the effectiveness of a short course of antibiotics (less than seven days) in comparison to a long course of antibiotics (seven days or greater) for the treatment of AOM in children.

Search methods

We searched the Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library 2009, issue 4) which contains the Acute Respiratory Infections Group's Specialised Register, MEDLINE, EMBASE, MEDLINE In‐Process & Other Non‐Indexed Citations, CINAHL, BIOSIS Previews, OCLC Papers First and Proceedings First, Proquest Dissertations and Theses (inception to November 2009); International Pharmaceutical Abstracts, the NLM Gateway, ClinicalTrials.gov and Current Controlled Trials (inception to August 2008).

Selection criteria

Trials were included if they met the following criteria: participants aged one month to 18 years; clinical diagnosis of ear infection; no previous antimicrobial therapy; and randomisation to treatment with less than seven days versus seven days or more of antibiotics.

Data collection and analysis

The primary outcome of treatment failure was defined as the absence of clinical resolution, relapse or recurrence of AOM during one month following initiation of therapy. Treatment outcomes were extracted from individual studies and combined in the form of a summary odds ratio (OR). A summary OR of 1.0 indicates that the treatment failure rate following less than seven days of antibiotic treatment was similar to the failure rate following seven days or more of treatment.

Main results

This update included 49 trials containing 12,045 participants. Risk of treatment failure was higher with short courses of antibiotics (OR 1.34, 95% CI 1.15 to 1.55) at one month after initiation of therapy (21% failure with short‐course treatment and 18% with long‐course; absolute difference of 3% between groups). There were no differences found when examining treatment with ceftriaxone for less than seven days (30% failure in those receiving ceftriaxone and 27% in short‐acting antibiotics administered for seven days or more) or azithromycin for less than seven days (18% failure in both those receiving azithromycin and short‐acting antibiotics administered for seven days or more) with respect to risk of treatment failure at one month or less. Significant reductions in gastrointestinal adverse events were observed for treatment with short‐acting antibiotics and azithromycin.

Authors' conclusions

Clinicians need to evaluate whether the minimal short‐term benefit from longer treatment of antibiotics is worth exposing children to a longer course of antibiotics.

Plain language summary

Short course antibiotics for healthy children with uncomplicated acute otitis media

Acute otitis media (AOM), or middle ear infection, is a common childhood illness, with more than half of all children having at least one infection by the time they are seven. Although otitis media often resolves without treatment, it is frequently treated with antibiotics. The length of treatment varies widely. This review of 49 trials found that treating children with a short course (less than seven days) of antibiotics, compared to treatment with a long course (seven days or greater) of antibiotics, increases the likelihood of treatment failure in the short term. No differences are seen one month later. The amount of gastrointestinal adverse events decreased with a shorter course of antibiotics.

Summary of findings

Background

Description of the condition

Acute otitis media (AOM) is one of the most common childhood health problems, a leading cause of visits to a physician and the most frequent reason children receive antibiotics or undergo surgery. The cost estimates for AOM in the US are between $3 and $5 billion per year. Most children will have at least one episode of AOM, with the peak incidence period being between 6 and 11 months (Rovers 2004).

AOM is defined as the presence of a middle ear effusion in conjunction with signs and symptoms of inflammation in the middle ear, such as otalgia, otorrhoea, fever or irritability. There is a high rate of spontaneous resolution for AOM, but if left untreated it can occasionally lead to complications, such as acute mastoiditis. Hearing loss can lead to behavioural changes and communication delay (Rovers 2004).

Description of the intervention

The standard duration of therapy in North America is 10 days (Froom 1990). However, the optimal duration of treatment is not known and varies worldwide. Half of the practitioners in Great Britain prescribe a five‐day course compared to the majority of practitioners in the Netherlands who prescribe six to seven days of antibiotic therapy. Many opt for no treatment (Froom 1990).

How the intervention might work

The decline during the 1940s and 1950s in suppurative complications of AOM throughout North America and Europe has been attributed to antibiotic therapy (Berman 1995). Evidence suggests that antibiotic‐treated and untreated children living in high‐income countries have similar long‐term outcomes (Burke 1991; Mygind 1981; van Buchem 1981). However, a meta‐analysis pooling the results of four studies suggests that although the spontaneous rate of resolution was high with placebo or no drug (81%) (95% confidence interval (CI) 69% to 94%), antimicrobial therapy increased resolution by 13.7% (95% CI 8.2% to 19.2%) (Rosenfeld 1994). A meta‐analysis by Del Mar documented that compared to placebo, antibiotics reduce pain at two to seven days after the start of treatment (Del Mar 1997; Sanders 2009). The absolute benefit of this finding is 5.6% fewer children experience pain at two to seven days. Long‐term outcomes were similar with and without treatment.

Why it is important to do this review

Practitioners in North America may be hesitant to discontinue prescribing antibiotics, despite findings from the medical literature that support management of AOM without antibiotics. Moreover, the therapy duration for optimal outcome continues to be a question. Expert opinion has recommended five days of antimicrobial treatment for uncomplicated otitis media in children over the age of six years (Paradise 1995; Paradise 1997). The quality of scientific evidence to support a policy for shorter courses of antibiotic treatment has been assessed (Pichichero 1997), but lacks a systematic, quantitative evaluation. Concern regarding resistant bacteria from the overuse of antibiotics (Cohen 1992; Murray 1994) and poor compliance is increasing, as is the cost of health care. Therefore, it is desirable to determine the shortest duration of antibiotic treatment that would result in favorable outcomes. We felt that a systematic review would add some objectivity to this debate.

Objectives

To determine the effectiveness of a short course of antibiotics (less than seven days) in comparison to a long course (seven days or longer) of antibiotics for the treatment of children with AOM. Subgroup analyses of children less than two years old, children with a perforated ear drum, and children with a history of recurrent otitis media were pursued to address concerns that these groups may have less favorable outcomes.

Methods

Criteria for considering studies for this review

Types of studies

For inclusion in the review, we considered randomized controlled trials (RCTs) of the empiric treatment of AOM, comparing two antibiotic regimens of different durations, as described in the title or abstract of the clinical trial. When the duration of antibiotic treatment was not specified in the title or abstract, we retrieved the trial to verify the treatment duration in both arms. We placed no language restrictions on the selection of clinical trials (Gregoire 1995).

Types of participants

We included children aged one month to 18 years, with a clinical diagnosis of AOM and no history of immediate antibiotic use, immune deficiency, chronic disease or head and neck abnormalities.

Types of interventions

We compared antibiotic therapy of a treatment arm for less than seven days (defined as the short course), with a treatment arm greater than or equal to seven days (defined as the long course). The antibiotic may be the same or different in the two treatment arms. Studies using the antibiotic ceftibuten were excluded from this 2009 update due to a lack of clinical relevance as the antibiotic is not commonly used by clinicians.

Types of outcome measures

Primary outcomes

Treatment failure, which included lack of clinical resolution, relapse or recurrence of AOM during a one‐month period following the initiation of therapy. Clinical resolution meant that the presenting signs or symptoms of AOM had improved or resolved.

Secondary outcomes

The cumulative number of treatment failures, relapses and recurrences reported at time of diagnosis and again at a final evaluation point between one to three months. Middle ear effusion was not classified as a treatment failure because of its documented persistence during the course of the disease, regardless of treatment. We sought data on the number of children with persistent middle ear effusion at both of the evaluation points.

Search methods for identification of studies

Electronic searches

The original search was developed and run in 1997 (see Appendix 1 for details of the search). The search was then updated and run in August 2008 for the years 1997 to 2007 (see search strategies in Appendix 2), and updated again in November 2009.

In this update we searched the Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library 2009, issue 4) which contains the Cochrane Acute Respiratory Infections Group's Specialised Register; MEDLINE (1966 to November Week 1, 2009); EMBASE (1974 to November 2009); MEDLINE In‐Process & Other Non‐Indexed Citations (1966 to Week 1, 2009); International Pharmaceutical Abstracts (1970 to August Week 1, 2008); BIOSIS Previews (1969 to November 2009); CINAHL (1981 to November 2009); the NLM Gateway (1998 to August 2008); OCLC Papers First and Proceedings First (1997 to November 2009); ClinicalTrials.gov (1998 to August 2008); Proquest Dissertations and Theses (1861 to November 2009); and Current Controlled Trials (1997 to August 2008). We searched the following databases without any date restrictions in September 2007: the National Research Register; CRISP; the TRIP Database; Scirus; and Google Scholar. We imposed no language or publication restrictions.

We ran the following search strategy in MEDLINE and CENTRAL. The MEDLINE search was combined with the Cochrane Highly Sensitive Search Strategy for identifying randomized trials in MEDLINE: sensitivity‐ and precision‐maximising version (2008 revision); Ovid version (Lefebvre 2008). The search strategy also incorporated the child search strategy developed by Boluyt (Boluyt 2008). We adapted this search strategy to search the other databases.

MEDLINE (Ovid)

1 exp Otitis Media/   2 otitis media.tw.   3 (OM or OME or AOM).tw.   4 (otorrhea* or otorrhoea*).tw.   5 (ear* adj3 (infect* or acute*)).tw.   6 or/1‐5   7 exp treatment outcome/   8 exp Anti‐Infective Agents/   9 (antibacter* or anti‐bacter* or anti bacter* or antibiotic* or anti‐biotic* or anti biotic* or bacteriocid* or antimicrob* or anti‐microb* or anti microb*).tw.   10 (amoxicillin* or amoxycillin* or penicillin* or cefprozil* or clarithromycin* or cefpodoxime* or cefaclor* or ceftriaxone* or azthromycin* cefixime*).tw.   11 or/7‐10   12 6 and 11

Searching other resources

We retrieved clinical trial references and published reviews on the treatment of AOM in order to identify further trials. We then contacted primary authors for information on such trials.

Data collection and analysis

Selection of studies

In the first publication of this review (Kozyrskyj 2000) clinical trials examining the antibiotic treatment of AOM were retrieved from the medical literature search and evaluated by the review authors according to the selection criteria previously described. Decisions to include the trial were made independently. Interreviewer agreement was assessed by the kappa statistic using PC Agree Software (Stephen Walter, McMaster University, Hamilton, Ontario) and consensus was reached regarding trial inclusion.

In this updated review, two review authors (KH, NH) evaluated for inclusion in the review the clinical trials retrieved during the search update. A third review author (TK) was consulted when consensus could not be reached.

Data extraction and management

At least two review authors (KH, LB) independently performed data extraction, followed by a consensus conference to resolve any differences. We developed a standardized form to facilitate data extraction. Data extracted from each trial included:

the number of children with no clinical resolution of AOM;
relapse or recurrence of AOM;
the number of children with persistent middle ear effusion; and
the number of withdrawals from each arm of the trial.

We recorded the data as follows:

Primary outcome: evaluation points included time points until one month after initiation of therapy.

Secondary outcome: evaluation point between one to three months after initiation of therapy.

Important subgroups defined a priori for which data were sought are:

antibiotic use for two days or less;
children aged less than two years; and
perforated tympanic membrane (Appelman 1991; Hendrickse 1988; Kaleida 1991; van Buchem 1981).

Descriptive data collected included treatment site, patient baseline characteristics, co‐interventions, and inclusion, exclusion and outcome criteria to summarise the generalisability of included studies and to facilitate sensitivity analyses. We also recorded drug dose, route and treatment duration.

Change in statistical methods from original version of review

Two studies (Gooch 1996; Hoberman 1997) in the current review contained two arms comparing long‐course versus short‐course antibiotics. In the first version of this review (Kozyrskyj 2000) the short‐course arm was counted twice against each long‐course arm of that study. In this review update, we have modified our methods and combined both long‐course arms into a single comparison against the short‐course arm, as recommended in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2008). This prevents overweighing the short‐course arm of these trials in the meta‐analysis.

Assessment of risk of bias in included studies

We used the Cochrane Collaboration's 'Risk of bias' tool (Higgins 2008) to assess for bias in this updated review. Two review authors (KH, MO) independently evaluated the risk of bias of included trials and then came to a consensus. When consensus could not be reached a third review author was consulted (TK).

Measures of treatment effect

We compared the treatment failure rate following a short course of antibiotics to the treatment failure rate following a long course of antibiotics using odds ratios (ORs) for individual trial outcomes. We determined a summary OR with 95% CI for trials pooled by antibiotic type using the Peto fixed‐effect model.

Unit of analysis issues

We performed analysis on trials grouped by the pharmacokinetic behaviour of the antibiotic used in the short‐course treatment arm, as follows.

Short‐acting oral antibiotics, for example, penicillin, amoxicillin, cefaclor and cefuroxime.
Oral azithromycin.
Intramuscular ceftriaxone.

We also conducted additional meta‐analyses in the short‐acting antibiotic group for treatment duration less than 48 hours and more than 48 hours.

Assessment of heterogeneity

We assessed statistical heterogeneity (Higgins 2008; Laupacis 1988). When significant heterogeneity existed, we examined trials for specific potential clinical differences (Thompson 1994). We also calculated summary ORs using the DerSimonian and Laird random‐effects model (DerSimonian 1986). We determined the summary risk difference (difference in the failure rate between the short and long‐course antibiotics) and a 95% CI for pooled group data.

Sensitivity analysis

We performed a sensitivity analyses to assess whether the results of the primary outcome for all antibiotics was sensitive to blinding, allocation concealment and publication bias. Additionally a sensitivity analysis was undertaken to assess those trials that used the same antibiotic in both the short and long courses of the trial.

Results

Description of studies

See Characteristics of included studies and Characteristics of excluded studies tables.

Results of the search

Updated electronic searches and handsearches from January 2000 through to November 2009 retrieved 4175 additional citations. In this updated review, we included 49 studies containing 12,045 participants. Refer to the Characteristics of included studies for detailed information on individual trials.

Included studies

We included 22 new trials, resulting in 49 eligible trials for this updated review (Adam 2000; Al‐Ghamdi 1999; Arguedas 2005; Arrieta 2003; Block 2000; Block 2003; Block 2004; Catania 2004; Cohen 1998; Cohen 1999; Cohen 2000; Dagan 2000a; Dagan 2000b; de Jose 1998; Dunne 2003; Guven 2006; Hoberman 2005; Kara 1998; Oguz 2003; Pessey 1999; Varsano 1997; Wang 2004). Fourty‐eight of the included studies were classified as randomized controlled trials (RCTs). Randomisation was not mentioned in one study (Mohs 1993). Included sample sizes ranged from 17 participants (Puczynski 1987) to 868 participants (Hoberman 1997). The median sample size was 215 participants (Dagan 2000b). Studies were primarily conducted in North America, South America, Europe and the Middle East (Israel and Turkey). Single trials were carried out in both Egypt (Mohs 1993) and China (Wang 2004). All trials enrolled children; 39 trials enrolled children less than one year old; eight trials began enrolment at two years of age; and two trials evaluated children three years of age or less. The maximum age of enrolment was 14 years (Kafetzis 1997).

The following antibiotics were evaluated in the included studies (number of studies in brackets): amoxicillin or amoxicillin/clavulanate (32); azithromycin (21); cefaclor (8); cefdinir (2); cefixime (2); cefpodoxime (4); cefprozil (2); cefuroxime (3); ceftriaxone (9); clarithromycin (1); penicillin (3); and trimethoprim‐sulfamethoxazole (1). Short‐course interventions were primarily a three to five‐day regimen. Two studies compared a single dose of azithromycin (Arguedas 2005; Block 2003) and two studies compared a short course of the antibiotic regimen at less than or equal to 48 hours (Meistrup‐Larsen 1983; Puczynski 1987).

Short regimens of ceftriaxone or azithromycin were analyzed separately from all other antibiotics. Twenty trials employed tympanocentesis. Of these, four trials (Hendrickse 1988; Pessey 1999; Ploussard 1984; Puczynski 1987) studied short regimens of antibiotics and 15 trials evaluated azithromycin (Arguedas 1996; Arguedas 1997; Arguedas 2005; Aronovitz 1996; Arrieta 2003; Dagan 2000a; Dagan 2000b; Daniel 1993; de Jose 1998; Guven 2006; Hoberman 2005; Khurana 1996; Oguz 2003; Petalozza 1992; Principi 1995).

For the majority of studies, clinical cure or improvement at the end of treatment and at a follow‐up period of 10 days to one month after beginning the study were primary outcomes. Please refer to the Characteristics of included studies table for details on individual study outcomes.

Excluded studies

We excluded two short‐acting antibiotic trials because outcomes were not reported as treatment failures (Bain 1985; Jones 1986). One study (Jones 1986) reported the number of days symptoms were recorded in diaries and the percentage resolution of ear signs, and the other study (Bain 1985) reported eardrum signs and symptoms. Neither of these studies presented data that could contribute to treatment failure as defined in this review. We excluded an additional two studies because the antibiotic used, ceftibuten, was deemed to be not clinically relevant due to its limited use in clinical practice (Roos 2000; Simon 1997), and excluded one study due to the duration of the antibiotic treatment being the same in both arms (Suzuki 2009).

Risk of bias in included studies

A detailed summary of the risk of bias assessment for each study is available in the Characteristics of included studies table.

Allocation

Adequate sequence generation was performed in 12 studies and was unclear in 37. Allocation concealment was adequate in six studies, inadequate in one study and unclear in 42 studies.

Blinding

Appropriate blinding was employed in 15 studies, there was no blinding in 24 studies and unclear blinding in 10 studies. Allocation concealment was also judged as adequate in only one of the appropriately blinded studies (Green 1993).

Incomplete outcome data

In 33 studies incomplete outcome data were appropriately addressed. Six studies did not address the incomplete outcome data appropriately and in 10 studies it was unclear if the data were appropriately reported.

Selective reporting

Thirty‐eight included studies were judged free from a risk of selective reporting bias; five studies had a high risk of selective reporting bias; and six studies were unclear in risk of selective reporting bias.

Other potential sources of bias

Fifteen studies were judged as low risk for other sources of bias. In 31 studies the risk of other sources of bias was unclear, and in three studies the risk was judged as high. Unreported sources of funding were the primary reason for an unclear risk of bias.

Effects of interventions

See: Table 1; Table 2; Table 3

Summary of findings for the main comparison. Short‐acting antibiotic > 48 hours short‐term treatment for acute otitis media.

Short‐acting antibiotic > 48 hours short‐term treatment for acute otitis media
Patient or population: patients with acute otitis media   Settings:   Intervention: Short‐acting antibiotic > 48 hours short‐term treatment
Outcomes	*Illustrative comparative risks (95% CI)**		Relative effect   (95% CI)	No of Participants   (studies)	Quality of the evidence   (GRADE)	Comments
	Assumed risk	Corresponding risk
	Control	Short‐acting antibiotic > 48 hours short‐term treatment
Treatment failure at 1 month or less	Study population		OR 1.34   (1.15 to 1.55)	5093   (16)
	175 per 1000	221 per 1000   (196 to 247)
	Medium risk population
	157 per 1000	200 per 1000   (176 to 224)
Treatment failure at 8 to 19 days	Study population		OR 1.37   (1.15 to 1.64)	3932   (11)
	144 per 1000	187 per 1000   (162 to 216)
	Medium risk population
	117 per 1000	154 per 1000   (132 to 179)
Treatment failure at 20 to 30 days	Study population		OR 1.16   (0.94 to 1.42)	2476   (9)
	203 per 1000	228 per 1000   (193 to 266)
	Medium risk population
	172 per 1000	194 per 1000   (163 to 228)
Treatment failure at 3 months or less	Study population		OR 1.18   (0.98 to 1.41)	2068   (7)
	364 per 1000	403 per 1000   (359 to 447)
	Medium risk population
	364 per 1000	403 per 1000   (359 to 447)
Treatment failure at 90 days	Study population		OR 1.16   (0.65 to 2.06)	207   (2)
	327 per 1000	360 per 1000   (240 to 500)
	Medium risk population
	311 per 1000	344 per 1000   (227 to 482)
Treatment failure at 30 to 45 days	Study population		OR 1.18   (0.97 to 1.43)	1861   (5)
	368 per 1000	407 per 1000   (361 to 454)
	Medium risk population
	364 per 1000	403 per 1000   (357 to 450)
The basis for the assumed risk* (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (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).   CI: Confidence interval; OR: Odds ratio;
GRADE Working Group grades of evidence   High quality: Further research is very unlikely to change our confidence in the estimate of effect.   Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.   Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.   Very low quality: We are very uncertain about the estimate.

Ceftriaxone for acute otitis media
Patient or population: patients with acute otitis media   Settings:   Intervention: Ceftriaxone
Outcomes	*Illustrative comparative risks (95% CI)**		Relative effect   (95% CI)	No of Participants   (studies)	Quality of the evidence   (GRADE)	Comments
	Assumed risk	Corresponding risk
	Control	Ceftriaxone
Treatment failure at 1 month or less	Study population		OR 1.07   (0.86 to 1.33)	1709   (8)
	270 per 1000	284 per 1000   (241 to 330)
	Medium risk population
	254 per 1000	267 per 1000   (226 to 312)
Treatment failure at 3 months or less	Study population		OR 0.89   (0.66 to 1.21)	701   (3)
	402 per 1000	374 per 1000   (307 to 449)
	Medium risk population
	379 per 1000	352 per 1000   (287 to 425)
The basis for the assumed risk* (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (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).   CI: Confidence interval; OR: Odds ratio;
GRADE Working Group grades of evidence   High quality: Further research is very unlikely to change our confidence in the estimate of effect.   Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.   Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.   Very low quality: We are very uncertain about the estimate.

Azithromycin 3 to 5 days short‐term treatment for acute otitis media
Patient or population: patients with acute otitis media   Settings:   Intervention: Azithromycin 3 to 5 days short‐term treatment
Outcomes	*Illustrative comparative risks (95% CI)**		Relative effect   (95% CI)	No of Participants   (studies)	Quality of the evidence   (GRADE)	Comments
	Assumed risk	Corresponding risk
	Control	Azithromycin 3 to 5 days short‐term treatment
Treatment failure at 1 month or less	Study population		OR 1.02   (0.87 to 1.2)	4354   (19)
	185 per 1000	188 per 1000   (165 to 214)
	Medium risk population
	61 per 1000	62 per 1000   (53 to 72)
Treatment failure at 8 to 19 days	Study population		OR 1.27   (1.04 to 1.55)	4347   (18)
	95 per 1000	118 per 1000   (98 to 140)
	Medium risk population
	56 per 1000	70 per 1000   (58 to 84)
Treatment failure at 20 to 30 days	Study population		OR 0.98   (0.82 to 1.17)	2708   (11)
	265 per 1000	261 per 1000   (228 to 297)
	Medium risk population
	273 per 1000	269 per 1000   (235 to 305)
The basis for the assumed risk* (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (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).   CI: Confidence interval; OR: Odds ratio;
GRADE Working Group grades of evidence   High quality: Further research is very unlikely to change our confidence in the estimate of effect.   Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.   Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.   Very low quality: We are very uncertain about the estimate.

Date	Event	Description
9 September 2010	Amended	Contact details updated.
18 November 2009	New citation required but conclusions have not changed	The searches were conducted in November 2009, with a total of 22 new trials added since the original review in 2000 (Adam 2000; Al‐Ghamdi 1999; Arguedas 2005; Arrieta 2003; Block 2000; Block 2003; Block 2004; Catania 2004; Cohen 1998; Cohen 1999; Cohen 2000; Dagan 2000a; Dagan 2000b; de Jose 1998; Dunne 2003; Guven 2006; Hoberman 2005; Kara 1998; Oguz 2003; Pessey 1999; Varsano 1997; Wang 2004). Two of the studies included in the first version of this review (Kozyrskyj 2000) were excluded due to the outcome not being reported as a treatment failure (Bain 1985; Jones 1986) and an additional three studies have been excluded; two were excluded due to use of ceftibuten which has been deemed not to be clinically relevant (Roos 2000; Simon 1997); and one for the duration of antibiotic therapy being equivalent in both arms (Suzuki 2009). The 22 studies that were included in this updated review brought the 30‐day findings closer to equivalence (primary outcome). This would suggest that any new studies would not change the current outcomes and the review authors recommend that it is no longer necessary to update this review.
11 November 2009	New search has been performed	Searches conducted.
22 December 2008	Feedback has been incorporated	Feedback added.
12 March 2008	Amended	Converted to new review format.
30 January 2000	New search has been performed	Searches conducted.

Electronic databases	Search strategies
MEDLINE OVID Version: rel10.5.2   1950 to August Week 2008   Searched: 1 August 2008   Results: 1892 Limits: Date: 1997 to 2008	1. exp Otitis Media/   2. (acute adj5 (OM or otitis media or ear or infection?)).mp.   3. (OM or OME or AOM).ti,ab.   4. otorrh?ea.ti,ab.   5. or/1‐4   6. exp Anti‐Infective Agents/   7. exp treatment outcome/   8. (antibacter$ or anti‐bacter$ or "anti bacter$" or antibiotic$ or anti‐biotic$ or "anti biotic$" or bacteriocid$ or antimicrob$ or anti‐microb$ or "anti microb$").mp.   9. or/6‐8   10. clinical trial.pt.   11. randomized controlled trial.pt.   12. randomi?ed.ti,ab.   13. placebo.ti,ab.   14. dt.fs.   15. randomly.ti,ab.   16. trial.ti,ab.   17. groups.ti,ab.   18. or/10‐17   19. animals/   20. humans/   21. 19 not (19 and 20)   22. 18 not 21   23. and/5,9,22   24. exp Infant/   25. exp Child/   26. Adolescent/   27. Minors/   28. exp Puberty/   29. exp Pediatrics/   30. infant$.mp.   31. infancy.mp.   32. newborn$.mp.   33. baby.mp.   34. babies.mp.   35. neonat$.mp.   36. preterm$.mp.   37. prematur$.mp.   38. postmatur$.mp.   39. child$.mp.   40. kid.mp.   41. kids.mp.   42. toddler$.mp.   43. adolescen$.mp.   44. teen$.mp.   45. boy$.mp.   46. girl.mp.   47. minor$.mp.   48. pubert$.mp.   49. pubescen$.mp.   50. prepubescen$.mp.   51. pediatric$.mp.   52. paediatric$.mp.   53. peadiatric$.mp.   54. or/24‐52   55. and/5,9,22,54   56. limit 55 to yr="1997 ‐ 2007"
Ovid MEDLINE In‐Process & Other   Non‐Indexed Citations   OVID Version: rel10.5.2   1 August 2008   Searched: 1 August 2008   Results: 6 Limits:   Date: 1997 to 2008	1. (acute adj5 (OM or otitis media or ear or infection?)).mp.   2. (OM or OME or AOM).ti,ab.   3. otorrh?ea.ti,ab.   4. or/1‐3   5. (antibacter$ or anti‐bacter$ or "anti bacter$" or antibiotic$ or anti‐biotic$ or "anti biotic$" or bacteriocid$ or antimicrob$ or anti‐microb$ or "anti microb$").mp.   6. (treatment adj5 outcome?).mp. [mp=title, original title, abstract, name of substance word, subject heading word]   7. or/5‐6   8. "controlled clinical trial$".mp.   9. "randomi?ed controlled trial$".mp.   10. "research design$".mp.   11. "clinical research".mp.   12. "random allocation".mp.   13. randomi?ed.ti,ab.   14. ("double blind" adj3 method$).mp.   15. ("single blind" adj3 method$).mp.   16. ((clin$ or control$) adj25 trial$).mp.   17. ((singl$ or doubl$ or trebl$ or tripl$) adj25 (blind$ or mask$)).ti,ab.   18. placebo$.ti,ab.   19. randomly.ti,ab.   20. or/8‐19   21. infant$.mp.   22. infancy.mp.   23. newborn$.mp.   24. baby.mp.   25. babies.mp.   26. neonat$.mp.   27. preterm$.mp.   28. prematur$.mp.   29. postmatur$.mp.   30. child$.mp.   31. kid.mp.   32. kids.mp.   33. toddler$.mp.   34. adolescen$.mp.   35. teen$.mp.   36. boy$.mp.   37. girl.mp.   38. minor$.mp.   39. pubert$.mp.   40. pubescen$.mp.   41. prepubescen$.mp.   42. pediatric$.mp.   43. paediatric$.mp.   44. peadiatric$.mp.   45. or/21‐44   46. and/4,7,20,45   47. limit 46 to yr="1997 ‐ 2007"
EMBASE OVID Version: rel10.5.2 1996 to 2008   Searched: 1 August 2008   Results: 2993 Limits:   Date: 1997 to 2008	1. exp Otitis Media/   2. (acute adj5 (OM or otitis media or ear or infection?)).mp.   3. (OM or OME or AOM).ti,ab.   4. otorrh?ea.ti,ab.   5. or/1‐4   6. exp Antiinfective Agent/   7. exp treatment outcome/   8. (antibacter$ or anti‐bacter$ or "anti bacter$" or antibiotic$ or anti‐biotic$ or "anti biotic$" or bacteriocid$ or antimicrob$ or anti‐microb$ or "anti microb$").mp.   9. or/6‐8   10. exp clinical trial/   11. randomi?ed.ti,ab.   12. placebo.ti,ab.   13. dt.fs.   14. randomly.ti,ab.   15. trial.ti,ab.   16. groups.ti,ab.   17. or/10‐16   18. exp child/   19. exp newborn/   20. exp adolescent/   21. exp puberty/   22. exp pediatrics/   23. infant$.mp.   24. infancy.mp.   25. newborn$.mp.   26. baby.mp.   27. babies.mp.   28. neonat$.mp.   29. preterm$.mp.   30. prematur$.mp.   31. postmatur$.mp.   32. child$.mp.   33. kid.mp.   34. kids.mp.   35. toddler$.mp.   36. adolescen$.mp.   37. teen$.mp.   38. boy$.mp.   39. girl.mp.   40. minor$.mp.   41. pubert$.mp.   42. pubescen$.mp.   43. prepubescen$.mp.   44. pediatric$.mp.   45. paediatric$.mp.   46. peadiatric$.mp.   47. or/18‐46   48. and/5,9,17,47   49. limit 48 to yr="1997 ‐ 2007"
International Pharmaceutical   Abstracts OVID Version: rel10.5.2   1970 to August 2008   Searched: 1 August 2008   Results: 45 Limits:   Date: 1997 to 2008	1. (acute adj5 (OM or otitis media or ear or infection?)).mp.   2. (OM or OME or AOM).ti,ab.   3. otorrh?ea.ti,ab.   4. or/1‐3   5. (antibacter$ or anti‐bacter$ or "anti bacter$" or antibiotic$ or anti‐biotic$ or "anti biotic$" or bacteriocid$ or antimicrob$ or anti‐microb$ or "anti microb$").mp.   6. (treatment adj5 outcome?).mp. [mp=title, original title, abstract, name of substance word, subject heading word]   7. or/5‐6   8. "controlled clinical trial$".mp.   9. "randomi?ed controlled trial$".mp.   10. "research design$".mp.   11. "clinical research".mp.   12. "random allocation".mp.   13. randomi?ed.ti,ab.   14. ("double blind" adj3 method$).mp.   15. ("single blind" adj3 method$).mp.   16. ((clin$ or control$) adj25 trial$).mp.   17. ((singl$ or doubl$ or trebl$ or tripl$) adj25 (blind$ or mask$)).ti,ab.   18. placebo$.ti,ab.   19. randomly.ti,ab.   20. or/8‐19   21. infant$.mp.   22. infancy.mp.   23. newborn$.mp.   24. baby.mp.   25. babies.mp.   26. neonat$.mp.   27. preterm$.mp.   28. prematur$.mp.   29. postmatur$.mp.   30. child$.mp.   31. kid.mp.   32. kids.mp.   33. toddler$.mp.   34. adolescen$.mp.   35. teen$.mp.   36. boy$.mp.   37. girl.mp.   38. minor$.mp.   39. pubert$.mp.   40. pubescen$.mp.   41. prepubescen$.mp.   42. pediatric$.mp.   43. paediatric$.mp.   44. peadiatric$.mp.   45. or/21‐44   46. and/4,7,20,45   47. limit 46 to yr="1997 ‐ 2007"
EBM Reviews Cochrane Central Register of Controlled Trials   2^nd Quarter 2008   OVID Version: rel10.5.2   Searched: 1 August 2008   Results: 425 Limits:   Date: 1997 to 2008	1. exp Otitis Media/   2. (acute adj5 (OM or otitis media or ear or infection?)).mp.   3. (OM or OME or AOM).ti,ab.   4. otorrh?ea.ti,ab.   5. or/1‐4   6. exp Anti‐Infective Agents/   7. exp treatment outcome/   8. (antibacter$ or anti‐bacter$ or "anti bacter$" or antibiotic$ or anti‐biotic$ or "anti biotic$" or bacteriocid$ or antimicrob$ or anti‐microb$ or "anti microb$").mp.   9. or/6‐8   10. exp Infant/   11. exp Child/   12. Adolescent/   13. Minors/   14. exp Puberty/   15. exp Pediatrics/   16. infant$.mp.   17. infancy.mp.   18. newborn$.mp.   19. baby.mp.   20. babies.mp.   21. neonat$.mp.   22. preterm$.mp.   23. prematur$.mp.   24. postmatur$.mp.   25. child$.mp.   26. kid.mp.   27. kids.mp.   28. toddler$.mp.   29. adolescen$.mp.   30. teen$.mp.   31. boy$.mp.   32. girl.mp.   33. minor$.mp.   34. pubert$.mp.   35. pubescen$.mp.   36. prepubescen$.mp.   37. pediatric$.mp.   38. paediatric$.mp.   39. peadiatric$.mp.   40. or/10‐38   41. and/5,9,40   limit 41 to yr="1997 ‐ 2007"
CINAHL Plus with Full Text (EBSCOhost) 1981‐present   Searched: 1 August 2008   Results: 532 Limits:   Date: 1997 to 2008 Age Groups: Infant, Newborn 0‐1 month, Infant, 1‐23 months, Child, Preschool 2‐5 years, Child, 6‐12 years, Adolescence, 13‐18 years	# Query   S7( S5 and S1 )   S6( S5 and S1 )   S5( S4 or S3 or S2 )   S4( antibacter* or anti‐bacter* or "anti bacter" or antibiotic or anti‐biotic* or "anti biotic" or bacteriocid or antimicrob* or anti‐microb* or "anti microb*" )   S3(MH "Treatment Outcomes+")   S2(MH "Antiinfective Agents+")   S1(MH "Otitis Media+") or ( acute w5 OM OR acute w5 "otitis media" OR actue w5 ear OR acute w5 infection ) or ( ottorrh?ea or OM or OME or AOM )
BIOSIS Previews^® (ISI Web of Knowledge^SM [3.0]) 1969 to present   Searched: 1 August 2008   Results: 1161   Limits: 1997 to 2008	Sets   History   #6 #5 AND #4   #5 TS=clinical trial* OR TS=research design OR TS=comparative stud* OR TS=evaluation stud* OR TS=controlled trial* OR TS=follow‐up stud* OR TS=prospective stud* OR TS=random* OR TS=placebo* OR TS=(single blind) OR TS=(double blind)   #4#3 OR #2 OR #1   #3 TS=ottorrhea   #2 TS=(OM OR OME OR OME OR AOM)   #1 TS="otitis media" OR TS=(acute SAME OM OR acute SAME "otitis media" OR acute SAME "ear infection")
OCLC PapersFirst (OCLC FirstSearch) Searched: 1 August 2008   Results: 25 + 3 Limits: 1997 to 2008	(kw: otitis w media) and (kw: antibiotic* OR kw: antibacter* OR kw: bacteriocid* OR kw: antimicrob*)
OCLC ProceedingsFirst (OCLC FirstSearch) Searched: 1 August 2008   Results: 1 Limits: 1997 to 2008	(kw: otitis w media) and (kw: antibiotic* OR kw: antibacter* OR kw: bacteriocid* OR kw: antimicrob*)
ProQuest^® Dissertations & Theses Full Text 1861 to present Searched: 1 August 2008   Results: 62 = 3	(("otitis media" OR "ear infection" or ottorrhea) AND (acute))

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Treatment failure at 1 month or less	16	5093	Peto Odds Ratio (Peto, Fixed, 95% CI)	1.34 [1.15, 1.55]
2 Treatment failure at 8 to 19 days	11	3932	Peto Odds Ratio (Peto, Fixed, 95% CI)	1.37 [1.15, 1.64]
3 Treatment failure at 20 to 30 days	9	2476	Peto Odds Ratio (Peto, Fixed, 95% CI)	1.16 [0.94, 1.42]
4 Treatment failure at 3 months or less	7	2068	Peto Odds Ratio (Peto, Fixed, 95% CI)	1.18 [0.98, 1.41]
5 Treatment failure at 90 days	2	207	Peto Odds Ratio (Peto, Fixed, 95% CI)	1.16 [0.65, 2.06]
6 Treatment failure at 30 to 45 days	5	1861	Peto Odds Ratio (Peto, Fixed, 95% CI)	1.18 [0.97, 1.43]

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Treatment failure at 1 month or less	6	1713	Peto Odds Ratio (Peto, Fixed, 95% CI)	1.37 [1.09, 1.72]
2 Treatment failure at 20 to 30 days	5	1149	Peto Odds Ratio (Peto, Fixed, 95% CI)	1.17 [0.89, 1.54]

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Treatment failure at 1 month or less	4	1298	Peto Odds Ratio (Peto, Fixed, 95% CI)	1.17 [0.73, 1.85]
2 Treatment failure at 20 to 30 days	2	656	Peto Odds Ratio (Peto, Fixed, 95% CI)	1.55 [0.79, 3.04]

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Treatment failure at 1 month or less	9	2955	Peto Odds Ratio (Peto, Fixed, 95% CI)	1.31 [1.09, 1.58]
2 Treatment failure at 20 to 30 days	6	1749	Peto Odds Ratio (Peto, Fixed, 95% CI)	1.25 [0.98, 1.58]

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Treatment failure at 1 month or less	9	3321	Peto Odds Ratio (Peto, Fixed, 95% CI)	1.65 [1.35, 2.01]
2 Treatment failure at 8 to 19 days	6	2153	Peto Odds Ratio (Peto, Fixed, 95% CI)	1.97 [1.54, 2.52]
3 Treatment failure at 20 to 30 days	4	1319	Peto Odds Ratio (Peto, Fixed, 95% CI)	1.27 [0.92, 1.76]
4 Treatment failure at 3 months or less	5	1492	Peto Odds Ratio (Peto, Fixed, 95% CI)	1.24 [1.00, 1.53]
5 Treatment failure at 90 days	2	207	Peto Odds Ratio (Peto, Fixed, 95% CI)	1.16 [0.65, 2.06]
6 Treatment failure at 30 to 45 days	3	1285	Peto Odds Ratio (Peto, Fixed, 95% CI)	1.25 [1.00, 1.57]

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Sensitivity analysis: allocation concealment unclear risk of bias	13	3953	Peto Odds Ratio (Peto, Fixed, 95% CI)	1.30 [1.09, 1.54]
2 Sensitivity analysis: allocation concealment low risk of bias	3	1140	Peto Odds Ratio (Peto, Fixed, 95% CI)	1.45 [1.08, 1.93]
3 Sensitivity analysis: blinding high and unclear risk of bias	12	3927	Peto Odds Ratio (Peto, Fixed, 95% CI)	1.18 [1.00, 1.40]
4 Sensitivity analysis: blinding low risk of bias	4	1166	Peto Odds Ratio (Peto, Fixed, 95% CI)	2.03 [1.48, 2.77]

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Amoxil‐clav. 5 versus 10 days, treatment failure 1 month or less	2	942	Peto Odds Ratio (Peto, Fixed, 95% CI)	1.99 [1.44, 2.74]
2 Excluding amoxil‐clav. 5 versus 10 days, treatment failure 1 month or less	14	4151	Peto Odds Ratio (Peto, Fixed, 95% CI)	1.20 [1.02, 1.42]

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Treatment failure at 1 month or less	8	1709	Peto Odds Ratio (Peto, Fixed, 95% CI)	1.07 [0.86, 1.33]
2 Treatment failure at 3 months or less	3	701	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.89 [0.66, 1.21]
3 Gastrointestinal adverse effects	1	402	Peto Odds Ratio (Peto, Fixed, 95% CI)	2.89 [1.70, 4.91]

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Treatment failure at 25 to 32 days	2	608	Peto Odds Ratio (Peto, Fixed, 95% CI)	1.01 [0.69, 1.47]
2 Treatment failure at 25 to 32 days, =< 2 yrs old	2	368	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.68 [0.43, 1.09]
3 Treatment failure at 25 to 32 days, > 2 yrs old	2	240	Peto Odds Ratio (Peto, Fixed, 95% CI)	1.08 [0.53, 2.20]
4 Gastrointestinal adverse effects	2	658	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.66 [0.45, 0.96]

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Treatment failure at 1 month or less	19	4354	Peto Odds Ratio (Peto, Fixed, 95% CI)	1.02 [0.87, 1.20]
2 Treatment failure at 8 to 19 days	18	4347	Peto Odds Ratio (Peto, Fixed, 95% CI)	1.27 [1.04, 1.55]
3 Treatment failure at 20 to 30 days	11	2708	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.98 [0.82, 1.17]

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Treatment failure at 1 month or less	7	1688	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.96 [0.70, 1.31]
2 Treatment failure at 20 to 30 days	4	740	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.83 [0.57, 1.21]

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Treatment failure at 1 month or less	6	1193	Peto Odds Ratio (Peto, Fixed, 95% CI)	1.26 [0.89, 1.80]
2 Treatment failure at 20 to 30 days	3	688	Peto Odds Ratio (Peto, Fixed, 95% CI)	1.27 [0.86, 1.86]

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Treatment failure at 1 month or less	11	2355	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.74 [0.61, 0.90]
2 Treatment failure at 20 to 30 days	5	902	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.96 [0.70, 1.31]

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Treatment failure at 20 to 30 days	6	1254	Peto Odds Ratio (Peto, Fixed, 95% CI)	1.02 [0.78, 1.34]
2 Treatment failure at 1 month or less	9	2205	Peto Odds Ratio (Peto, Fixed, 95% CI)	1.13 [0.89, 1.44]

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Sensitivity analysis: allocation concealment unclear or high risk of bias	17	4165	Odds Ratio (M‐H, Fixed, 95% CI)	1.04 [0.88, 1.22]
2 Sensitivity analysis: allocation concealment low risk of bias	2	189	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.13 [0.02, 0.90]
3 Sensitivity analysis: blinding high and unclear risk of bias	15	3348	Peto Odds Ratio (Peto, Fixed, 95% CI)	1.15 [0.94, 1.40]
4 Sensitivity analysis: blinding low risk of bias	4	1006	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.80 [0.60, 1.07]

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Treatment failure 1 month or less	8	1747	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.97 [0.78, 1.21]
2 Treatment failure at 8 to 19	7	1667	Peto Odds Ratio (Peto, Fixed, 95% CI)	1.43 [1.10, 1.87]
3 Treatment failure at 20 to 30 days	6	1469	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.93 [0.74, 1.17]

PERMALINK

Short‐course antibiotics for acute otitis media

Anita L Kozyrskyj

Terry P Klassen

Michael Moffatt

Krystal Harvey

Abstract

Background

Objectives

Search methods

Selection criteria

Data collection and analysis

Main results

Authors' conclusions

Plain language summary

Summary of findings

Background

Description of the condition

Description of the intervention

How the intervention might work

Why it is important to do this review

Objectives

Methods

Criteria for considering studies for this review

Types of studies

Types of participants

Types of interventions

Types of outcome measures

Primary outcomes

Secondary outcomes

Search methods for identification of studies

Electronic searches

MEDLINE (Ovid)

Searching other resources

Data collection and analysis

Selection of studies

Data extraction and management

Change in statistical methods from original version of review

Assessment of risk of bias in included studies

Measures of treatment effect

Unit of analysis issues

Assessment of heterogeneity

Sensitivity analysis

Results

Description of studies

Results of the search

Included studies

Excluded studies

Risk of bias in included studies

Allocation

Blinding

Incomplete outcome data

Selective reporting

Other potential sources of bias

Effects of interventions

Summary of findings for the main comparison. Short‐acting antibiotic > 48 hours short‐term treatment for acute otitis media.

Summary of findings 2. Ceftriaxone for acute otitis media.

Summary of findings 3. Azithromycin 3 to 5 days short‐term treatment for acute otitis media.

Treatment failure

Outcome of short‐acting antibiotics in the short course (less than seven days)

i) Short‐acting antibiotics given for less than 48 hours

1.1. Analysis.

ii) Short‐acting antibiotics given for more than 48 hours

2.1. Analysis.

2.2. Analysis.

2.3. Analysis.

3.1. Analysis.

4.1. Analysis.

5.1. Analysis.

6.1. Analysis.

Outcome of ceftriaxone in the short course (less than seven days)

17.1. Analysis.

17.2. Analysis.

Outcome of azithromycin in the short course (less than seven days)

19.1. Analysis.

19.2. Analysis.

19.3. Analysis.

20.1. Analysis.

21.1. Analysis.

Sensitivity analysis

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Treatment failure 1 month or less	11	2616	Peto Odds Ratio (Peto, Fixed, 95% CI)	1.09 [0.86, 1.38]
2 Treatment failure at 8 to 19 days	11	2680	Peto Odds Ratio (Peto, Fixed, 95% CI)	1.08 [0.80, 1.46]
3 Treatment failure at 20 to 30 days	5	1239	Peto Odds Ratio (Peto, Fixed, 95% CI)	1.06 [0.81, 1.39]

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Twice counting ‐ short‐acting antibiotic > 48 hours short‐term treatment, treatment failure at 1 month or less	9	2249	Peto Odds Ratio (Peto, Fixed, 95% CI)	1.43 [1.14, 1.79]
2 Combined control groups ‐ short‐acting antibiotic > 48 hours short‐term treatment, treatment failure 1 month or less	9	2611	Peto Odds Ratio (Peto, Fixed, 95% CI)	1.36 [1.10, 1.67]
3 Twice counting ‐ short‐acting antibiotic > 48 hours short‐term treatment, treatment failure at 1 to 19 days	4	1138	Peto Odds Ratio (Peto, Fixed, 95% CI)	1.52 [1.11, 2.10]
4 Combined control groups ‐ short‐acting antibiotic > 48 hours short‐term treatment, treatment failure 1 to 19 days	4	1327	Peto Odds Ratio (Peto, Fixed, 95% CI)	1.45 [1.08, 1.95]
5 Twice counting ‐ short‐acting antibiotic > 48 hours short‐term treatment, treatment failure 20 to 30 days	7	1632	Peto Odds Ratio (Peto, Fixed, 95% CI)	1.26 [0.97, 1.64]
6 Combined control groups ‐ short‐acting antibiotic > 48 hours short‐term treatment, treatment failure at 20 to 30 days	7	1805	Peto Odds Ratio (Peto, Fixed, 95% CI)	1.22 [0.95, 1.57]
7 Twice counting ‐ short‐acting antibiotic > 48 hours short‐term treatment, treatment failure at 30 to 40 days	2	538	Peto Odds Ratio (Peto, Fixed, 95% CI)	1.07 [0.74, 1.55]
8 Combined control groups ‐ short‐acting antibiotic > 48 hours short‐term treatment, treatment failure 30 to 40 days	2	686	Peto Odds Ratio (Peto, Fixed, 95% CI)	1.12 [0.80, 1.57]
9 Twice counting ‐ short‐acting antibiotic > 48 hours short‐term treatment, treatment failure at 3 months or less	4	745	Peto Odds Ratio (Peto, Fixed, 95% CI)	1.10 [0.80, 1.50]
10 Combined control groups ‐ short‐acting antibiotic > 48 hours short‐term treatment, treatment failure at 3 months or less	4	893	Peto Odds Ratio (Peto, Fixed, 95% CI)	1.13 [0.85, 1.51]

Methods	RCT
Participants	German multicentre trial of children 3 months to 6 years old
Interventions	Cefpodoxime 40 mg to 60 mg twice daily for 5 days   versus   cefaclor 40 mg/kg/day 3 times daily for 10 days
Outcomes	Absence or improvement of signs and symptoms at 3 days and no relapse at 3 weeks after study entry
Notes	—
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Blinding (performance bias and detection bias)   Treatment failure at 1 month or less	High risk
Incomplete outcome data (attrition bias)   Treatment failure at 1 month or less	Low risk
Selective reporting (reporting bias)	Low risk
Other bias	Unclear risk	Funding not reported

Methods	Double‐blind RCT
Participants	German children aged 2 to 14 years
Interventions	Cefixime 8 mg/kg/day for 5 days   versus   same treatment for 10 days
Outcomes	Complete recovery or improvement of symptoms by day 11, clinical response rate at day 6, rate of relapse at day 28
Notes	—
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Blinding (performance bias and detection bias)   Treatment failure at 1 month or less	Low risk
Incomplete outcome data (attrition bias)   Treatment failure at 1 month or less	High risk
Selective reporting (reporting bias)	High risk
Other bias	Unclear risk	Funding not reported

Methods	RCT
Participants	Saudi children aged 6 months to 6 years
Interventions	Single intramuscular dose 50 mg/kg ceftriaxone   versus   40 mg/kg/day Augmentin 3 times daily for 10 days
Outcomes	Resolution of symptoms at 10 days after study entry; improved otoscopic and tympanometric measurements at 60 days after study entry
Notes	—
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Incomplete outcome data (attrition bias)   Treatment failure at 1 month or less	Low risk
Selective reporting (reporting bias)	Low risk
Other bias	Unclear risk	Did not report baselines or funding

Methods	RCT
Participants	Children 6 months to 12 years old in an ambulatory unit in Costa Rica
Interventions	Azithromycin 10 mg/kg once daily for 3 days   versus   amoxil‐clavulanate 40 mg/kg/day 3 times daily for 10 days
Outcomes	Absence or improvement of signs and symptoms at 11 to 13 days, 28 to 30 days and 55 to 60 days after study entry
Notes	—
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk
Allocation concealment (selection bias)	Low risk
Blinding (performance bias and detection bias)   Treatment failure at 1 month or less	High risk
Incomplete outcome data (attrition bias)   Treatment failure at 1 month or less	Low risk
Other bias	Low risk	Industry funding

Methods	RCT
Participants	Children 6 months to 12 years old seen at an ambulatory unit in Costa Rica
Interventions	Azithromyzin 10 mg/kg once daily for 3 days   versus   clarithromycin 15 mg/kg/day twice daily for 10 days
Outcomes	Resolution of symptoms with or without the presence of middle ear fluid at 10 to 11 days and 28 to 32 days after study entry
Notes	—
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk
Allocation concealment (selection bias)	Low risk
Blinding (performance bias and detection bias)   Treatment failure at 1 month or less	High risk
Incomplete outcome data (attrition bias)   Treatment failure at 1 month or less	Low risk
Other bias	Low risk	Industry funding

Methods	Double‐blind, double‐dummy multinational RCT
Participants	Children 6 to 30 months of age in multiple countries
Interventions	Azithromycin single dose 30 mg/kg   versus   amoxicillin 90 mg/kg/day for 10 days in 2 divided doses
Outcomes	Absence or improvement of symptoms at 12 to 14 days and at end of study (day 25 to 28)
Notes	—
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Blinding (performance bias and detection bias)   Treatment failure at 1 month or less	Low risk
Incomplete outcome data (attrition bias)   Treatment failure at 1 month or less	Low risk
Selective reporting (reporting bias)	High risk
Other bias	High risk	Interim analysis, industry funding

Methods	RCT
Participants	US multicentre trial of children 2 to 15 years old
Interventions	Azithromycin 5 to 10 mg/kg once daily for 5 days   versus   amoxil‐clavulanate 40 mg/kg/day 3 times daily for 10 days
Outcomes	Absence or improvement of signs and symptoms at 11 days and 30 days after study entry
Notes	—
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Blinding (performance bias and detection bias)   Treatment failure at 1 month or less	High risk
Incomplete outcome data (attrition bias)   Treatment failure at 1 month or less	High risk
Selective reporting (reporting bias)	Low risk
Other bias	Unclear risk	No funding declared

Methods	Double‐blind, double‐dummy, multicentre trial
Participants	US and Latin American children aged 6 months to 6 years
Interventions	Azithromycin 20 mg/kg/day once daily for 3 days   versus   amoxicillin/clavulanate 45/6.4 mg/kg/day twice daily for 10 days plus additional amoxil 45 mg/kg/day twice daily for 10 days
Outcomes	Clinical response (improvement or cure) among clinically evaluated patients at day 28 to 32; clinical response at day 12 to 16
Notes	—
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Blinding (performance bias and detection bias)   Treatment failure at 1 month or less	Low risk
Incomplete outcome data (attrition bias)   Treatment failure at 1 month or less	Low risk
Selective reporting (reporting bias)	Low risk
Other bias	Unclear risk	Corresponding author from Pfizer, unclear independence of authors, industry funding