Table 6fComparative Effectiveness of Different Treatment Options for Treating Uncomplicated Acute Otitis Media in Average Risk Children in the 2001 Report and the Present Report

2001 Report2009 Update
ComparisonNumber of trialsSuccess rate difference (95% CI)Number of new trialsTotal number of trialsSuccess rate differenceConclusiona
Drug vs. placebo, wait-and-see, and/or prescription-to-hold
Ampicillin or amoxicillin vs. placebo512% (−22%, −3%)2712%(5%, 18%)Ampicillin or amoxicillin was more successful than placebo
Amoxicillin tid (7d) vs. prescription-to-hold) 20N/A1116% (6, 26)Amoxicillin is more successful than prescription-to-hold (defined as success at day 3)
Antibiotic vs. prescription-to-hold)20N/A113% (−8, 14)Inconclusive (defined as otalgia at day 4–6)
Amoxicillin 90mg/kg/d bid (10d) vs. wait-and-see 30N/A1115% (6, 24)Amoxicillin was more successful (defined as success at day 12)
PcV vs. wait-and-see 30N/A11−3% (−14, 8)Inconclusive (defined as success at day 14)
Drug vs. drug
Ampicillin or amoxicillin vs. Ceftriaxone33% (−8%, 2%)140% (−7%, 7%)Inconclusive
Amoxicillin 50mg/kg/d (bid, 10d) vs. erythromycin 40mg/kg/d (bid, 10d) 40N/A110.6% (−3, 4)Treatments were equivalent (when success defined as freedom from recurrence day 31–40)
Amoxicillin-clavulanate vs. amoxicillin sulbactam 80mg/kg/d; (bid 10d)0N/A110% (−3.3, 3.3)Treatments were equivalent (success d.12–14)
Amoxicillin-clavulanate (>6 yrs old: 250 mg tid x 7d; < 6 yrs old: 125 mg tid x7d) vs. cefaclor (125 or 250 mg tid x 7 d) 50N/A1113% (5, 21)Amoxicillin-clavulanate was more effective than cefaclor (success at day 28–34, as defined by clinical symptoms but not by culture)
Cefaclor vs. trimethoprim-sulfamethoxazole3−6% (−13, 2) (success at less than 14 d)03N/AInconclusive (defined as success at less than day 14); no new data but using MCID
Cefaclor vs. Ampicillin or amoxicillin4−5% (−16, 6) (success at d. 3–7)04N/AInconclusive (defined as success at day 3–7); no new data but using MCID
Cefixime vs. Ampicillin or amoxicillin40.1% (−3.9, 4.2) (success at d. 10–15)04N/ATreatments were equivalent; no new data
Penicillin vs. ampicillin or amoxicillin3−5% (−11, 2) (success at d. 7–14)03N/AInconclusive (defined as success at day 7–14); no new data but using MCID
High vs. Low Dose Treatment
Amoxicillin-clavulanate >60mg/kg/d vs. amoxicillin-clavulanate 40mg/kg/d11.5% (−3, 13)01N/AInconclusive(defined as persistent clinical cure with no recurrence at follow-up)
High-dose amoxicillin bid vs. lower-dose amoxicillin tid1−4% (−14, 7)01N/AInconclusive (defined as success at day 15)t; no new data
Amoxicillin-clavulanate 45/64 mg/kg/day/bid for 7–10 days vs. Amoxicillin-clavulanate 40/10 mg/kg/day/tid for 7–10 days 60N/A110.1% (−4.8, 4.6)Treatments were equivalent (success d. 7–12)
Short vs. Long Treatment Durationb
Ampicillin or Amoxicillin (7–10d) vs. Ceftriaxone (1 dose)33% (−2%, 9%) (success rate at 5–10d)140% (−7%, 7%)Inconclusive
Amoxicillin-Clavulanate (7–10d) vs. Ceftriaxone (1 dose)2No meta-analysis353% (−2%, 7%)Inconclusive
Cefaclor (7–10d) vs. Azithromycin (<5d)1N/A23−1% (−4%, 3%)Treatments were equivalent
Amoxicillin (7d) vs. Azithromycin (1 dose)0N/A111% (−1%, 4%)Treatments were equivalent (defined as no new pain between day 1 and 11)
Amoxicillin-clavulanate (7–10d) vs. Azithromycin (<5d)52% (−1, 5%) (success at 10–14d)49−0.3% (−6%, 6%)Inconclusive
Amoxicillin-clavulanate 45/6.4 mg/kg/d (bid, 10d) vs. azithromycin 10 mg/kg/d (qd for 1 day), 5 mg/kg/d (qd for 4d) 70N/A1126% (6,36)Longer-term amoxicillin-clavulanate is more successful than shorter-term azithromycin (at d. 12–14, when pathogen is H. influenzae)
Amoxicillin-clavulanate 45/6.4 mg/kg/d (bid, 10d) vs. azithromycin 10 mg/kg/d (qd for 3 day) 520N/A11−20% (−34, −6)Amoxicillin-clavulanate was worse than azithromycin (cure defined as negative culture)
Cefaclor 50mg/kg/d; bid 5 d) vs. cefaclor 40mg/kg/d; bid 10d)0N/A110.7% (−3.5–4.9)Treatments were equivalent

Table Notes: bid twice a day; CI confidence intervals; d day(s); kg kilograms (body weight); mg milligrams; NNT number needed to treat; PcV phenoxymethylpenicillin; qd once a day;


Confidence intervals falling within the zone of indifference were considered to establish evidence of no difference, and confidence intervals outside the zone of indifference were considered to establish difference. If the confidence intervals crossed into the zone of indifference, an effect (positive or negative) of the treatment option on the outcome could not be established (inconclusive). For the 2010 systematic review, we used a zone of clinical indifference of +/− 5% for the difference in success rate between two treatment options.


Short vs. long term duration refers to the length of treatment from the patient perspective, rather than from the perspective of drug action.

From: 3, Results

Cover of Management of Acute Otitis Media: Update
Management of Acute Otitis Media: Update.
Evidence Reports/Technology Assessments, No. 198.
Shekelle PG, Takata G, Newberry SJ, et al.

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.