Quality of life, healthcare use and costs in 'at-risk' children after early antibiotic treatment versus placebo for influenza- like illness: within-trial descriptive economic analyses of the ARCHIE randomised controlled trial

Objectives To characterise the quality of life, healthcare use and costs associated with early antibiotic treatment of influenza-like illness (ILI) in 'at-risk' children. Design Economic analysis of a two-arm double-blind parallel group pragmatic randomised controlled trial. Setting Children were recruited from community-based healthcare settings, including general practices, walk-in centres and hospital ambulatory care. Participants Children with risk factors for influenza-related complications, including respiratory, cardiac and neurological conditions, who presented within the first 5 days of an ILI. Interventions Co-amoxiclav 400/57 suspension or placebo. Outcome measures This economic analysis focused on quality of life measured by the EQ-5D-Y, symptoms assessed by the Canadian Acute Respiratory Infection and Flu Scale (CARIFS), healthcare use and costs including medication, hospital visits and admissions, general practitioner and nurse contacts. Outcomes were assessed for up to 28 days post randomisation. Results Information on resource use, EQ-5D-Y (day 28) and CARIFS (day 7) was available for 265 (98%), 72 (27%) and 123 (45%) out of 271 participants, respectively. Average costs in the co-amoxiclav group were 25 pound lower (95% CI -113 pound to 65) pound, but this difference was not statistically significant (p=0.566). The difference in EQ-5D-Y scores between groups was also not statistically significant (-0.014 (95% CI -0.124 to 0.096), p=0.798). However, day 7 CARIFS scores were 3.5 points lower in the co-amoxiclav arm (95% CI -6.9 to -0.1, p=0.044). Conclusions Our findings did not show evidence that early co-amoxiclav treatment improves quality of life or reduces healthcare use and costs in 'at-risk' children with ILI, but may reduce symptom severity though confirmation from further research would be important. Reliable data collection from children's parents/carers was challenging, and resulted in high levels of missing data, which is common in pragmatic trials involving children with acute respiratory tract infections.