Jennifer S Reath, Hasantha Gunasekera, Sanja Lujic, Amanda J Leach, Letitia Campbell, Robyn Walsh, Tim Usherwood, Geoffrey K Spurling, Claudette A Tyson, Deborah A Askew, Kelvin Kong, Chelsea J Watego, Peter Morris, Wendy Hu, Penelope A Abbott
Although our results are numerically similar to those reported in other low-risk populations and no unexpected safety signals were observed in the watchful waiting arm, non-inferiority was not established. Larger, adequately powered trials are required to determine whether watchful waiting is non-inferior in this population.
OBJECTIVE: Determine whether watchful waiting is non-inferior to immediate oral antibiotics for uncomplicated acute otitis media among urban Aboriginal and Torres Strait Islander children.
STUDY TYPE: Non-inferiority unblinded randomised controlled trial.
SETTING AND PARTICIPANTS: Eight Aboriginal Medical Services across three Australian states and territories between 25 August 2014 and 2 June 2023. Children (aged 1.5-16 years) with type B tympanograms and bulging tympanic membrane or acute pain/irritability were randomised by site and age (1.5-6 years and 7-16 years), with stratification using randomly allocated, permuted blocks of four and six in length.
MAIN OUTCOME MEASURES: Watchful waiting compared with immediate oral antibiotics using modified intention-to-treat (using only available data) and per-protocol analyses of Day 7 clinical resolution with non-inferiority threshold set at 10 percentage points.
RESULTS: Children were randomly allocated to watchful waiting (134), six of whom were lost to follow-up or immediate antibiotics (129) with three lost to follow-up. Resolution occurred in 57/106 (53.8%) watchful waiting and 68/113 (60.2%) immediate antibiotic group of those with complete Day 7 data (-6.4 percentage points difference; 90% confidence interval [CI], -17.4 to 4.6) (modified intention-to-treat analysis). Per-protocol analysis similarly demonstrated reduced resolution in the watchful waiting group (49/97; 50.5%) compared with immediate antibiotics (67/112; 59.8%) with -9.3 percentage points difference (90% CI, -20.6 to 2.0). There was less Day 3 diarrhoea in watchful waiting (3/90; 3.3%) than in the immediate antibiotic group (13/93 [14.0%]; -10.7 percentage points difference; 95% CI, -18.6 to -2.7) but no other differences in vomiting, diarrhoea or rash (Days 3-14). Day 7 analgesia use was higher in the watchful waiting (53/107 [49.5%]) than immediate antibiotic group (37/116 [31.9%]; 17.6 percentage points difference; 95% CI, 4.9 to 30.1). There were no intervention-related severe adverse events or perforations.
CONCLUSION: Although our results are numerically similar to those reported in other low-risk populations and no unexpected safety signals were observed in the watchful waiting arm, non-inferiority was not established. Larger, adequately powered trials are required to determine whether watchful waiting is non-inferior in this population.
TRIAL REGISTRATION: Australian New Zealand Clinical Trials Registry ACTRN#12613001068752.