- Hu, Huiyu;
- Arzika, Ahmed Mamane;
- Sie, Ali;
- Abdou, Amza;
- Maliki, Ramatou;
- Mankara, Alio Karamba;
- Outtara, Mamadou;
- Bountogo, Mamadou;
- Boudo, Valentin;
- Yago-Wienne, Fanny;
- Bamba, Issouf;
- Knirsch, Charles;
- Emerson, Paul;
- Hooper, PJ;
- Lebas, Elodie;
- Brogdon, Jessica;
- Nyatigo, Fanice;
- Oldenburg, Catherine E;
- Lietman, Thomas M;
- O'Brien, Kieran S
Background
To facilitate mass distribution of azithromycin, trachoma control programmes use height instead of weight to determine dose for children 6 months to 15 years old. WHO has recommended azithromycin distribution to children 1-11 months old to reduce mortality in high mortality settings under carefully monitored conditions. Weight was used to determine dose in children 1-5 months old in studies of azithromycin distribution for child survival, but a simplified approach using age or height for all aged 1-11 months old could increase programme efficiency in real-world settings.Methods
This secondary analysis used data from two cluster randomised trials of azithromycin distribution for child mortality in Niger and Burkina Faso. An exhaustive search algorithm was developed to determine the optimal dose for different age groups, using tolerance limits of 10-20 mg/kg for children 1-2 months old and 15-30 mg/kg for children 3-11 months old. Height-based dosing was evaluated against the existing trachoma dosing pole and with a similar exhaustive search.Results
The optimal two-tiered age-based approach suggested a dose of 80 mg (2 mL) for children 1-2 months old and 160 mg (4 mL) for children 3-11 months old. Under this schedule, 89%-93% of children would have received doses within tolerance limits in both study populations. Accuracy was 93%-94% with a three-tiered approach, which resulted in doses of 80 mg (2 mL), 120 mg (3 mL) and 160 mg (4 mL) for children 1-2, 3-4 and 5-11 months old, respectively. For children 1-5 months old, the existing height pole would result in 70% of doses within tolerance limits. The optimisation identified height-based dosing options with 95% accuracy, although this would require changes to the existing dosing pole as well as additional training to measure infants lying flat.Conclusions
Overall, an age-based approach with two age tiers resulted in high accuracy while considering both concerns about overdosing in this young population and simplicity of field operations.