Treatment Failure Rates and Hazard Ratios in Each Prognostic Factor Group

No. of Children With Treatment Failure (%) (N = 319)Hazard Ratio for Treatment Failure (95% CI)aP for the Hazard RatioaP for the Interaction (Prognostic Factor × Treatment Group)
Entire cohort101/319 (31.7)
Personal factors
  24–35 mo12/60 (20.0)0.53 (0.29 to 0.96)
  6–23 mo89/259 (34.4)Reference
 ≥4 previous episodes of AOM.57.83
  Yes20/58 (34.5)1.15 (0.71 to 1.88)
  No81/261 (31.0)Reference
Otoscopic signs at enrollment
 Bilateral AOMb.48.76
  Yes44/127 (34.6)1.15 (0.78 to 1.72)
  No55/188 (29.3)Reference
 Severe bulging of the tympanic membrane.03.02
  Yes30/84 (35.7)1.75 (1.07 to 2.85)
  No71/235 (30.2)Reference
 Peaked tympanogramc.02.31
  Yes8/57 (14.0)0.43 (0.21 to 0.88)
  No93/262 (35.5)Reference
Symptoms at entry
 Fever ≥38°C.18.43
  Yes37/110 (33.6)1.32 (0.88 to 1.98)
  No64/209 (30.6)Reference
 Ear pain.46.44
  Yes79/254 (31.1)0.84 (0.52 to 1.34)
  No22/65 (33.8)Reference
 Ear rubbing.29.66
  Yes59/203 (29.1)0.81 (0.54 to 1.20)
  No42/116 (36.2)Reference
 Decreased activity.30.19
  Yes36/135 (26.7)0.81 (0.53 to 1.21)
  No65/184 (35.3)Reference
Severity of illness
 Severe illness according to AAP criteriad.92.88
  Yes63/197 (32.0)1.02 (0.68 to 1.53)
  No38/122 (31.1)Reference
Microbial etiology in nasopharyngeal samplee
S pneumoniae.74.37
  Yes60/191 (31.4)1.07 (0.72 to 1.60)
  No39/125 (31.2)Reference
H influenzae.72.92
  Yes26/79 (32.9)0.92 (0.59 to 1.45)
  No73/237 (30.8)Reference
M catarrhalis.96.12
  Yes72/233 (30.9)0.99 (0.64 to 1.54)
  No27/83 (32.5)Reference
 Any respiratory virus.47.80
  Yes90/276 (32.6)1.29 (0.65 to 2.56)
  No9/40 (22.5)Reference
  • a Hazard ratios and P values are adjusted for treatment group. If the prognostic factor showed an association with treatment failure (ie, age 24–35 mo and a peaked tympanogram) or a significant interaction was found (severe bulging of the tympanic membrane × treatment group), the hazard ratios and P values are adjusted for treatment group, for the other prognostic factors, and for the interaction term.

  • b Data on bilaterality were missing for 4 children in whom an adequate view of the contralateral tympanic membrane was not possible because of thick cerumen.

  • c Peaked tympanogram was defined as a tympanogram with height (static acoustic admittance) ≥0.2 mmho and width ≤300 daPa (ie, type A, C1, or C2 curve). The selected tympanogram was obtained from the ear with most severe otoscopic findings.

  • d Severe illness according to AAP criteria: moderate or severe ear pain or fever ≥39°C.

  • e Data on microbiologial etiology were missing for 3 children because parents refused the nasopharyngeal sampling from their children.