A Randomized Trial to Assess the Impact of a Package of Diagnostic Tools and Diagnostic Algorithm on Antibiotic Prescriptions for the Management of Febrile Illnesses Among Children and Adolescents in Primary Health Facilities in Burkina Faso
Kiemde F., Valia D., Kabore B., Rouamba T., Kone AN., Sawadogo S., Compaore A., Salami O., Horgan P., Moore CE., Dittrich S., Nkeramahame J., Olliaro P., Tinto H.
Abstract Background Low- and middle-income countries face significant challenges in differentiating bacterial from viral causes of febrile illnesses, leading to inappropriate use of antibiotics. This trial aimed to evaluate the impact of an intervention package comprising diagnostic tests, a diagnostic algorithm, and a training-and-communication package on antibiotic prescriptions and clinical outcomes. Methods Patients aged 6 months to 18 years with fever or history of fever within the past 7 days with no focus, or a suspected respiratory tract infection, arriving at 2 health facilities were randomized to either the intervention package or standard practice. The primary outcomes were the proportions of patients who recovered at day 7 (D7) and patients prescribed antibiotics at day 0. Results Of 1718 patients randomized, 1681 (97.8%; intervention: 844; control: 837) completed follow-up: 99.5% recovered at D7 in the intervention arm versus 100% in standard practice (P = .135). Antibiotics were prescribed to 40.6% of patients in the intervention group versus 57.5% in the control arm (risk ratio: 29.3%; 95% CI: 21.8–36.0%; risk difference [RD]: −16.8%; 95% CI: −21.7% to −12.0%; P < .001), which translates to 1 additional antibiotic prescription saved every 6 (95% CI: 5–8) consultations. This reduction was significant regardless of test results for malaria, but was greater in patients without malaria (RD: −46.0%; −54.7% to −37.4%; P < .001), those with a respiratory diagnosis (RD: −38.2%; −43.8% to −32.6%; P < .001), and in children 6–59 months old (RD: −20.4%; −26.0% to −14.9%; P < .001). Except for the period July–September, the reduction was consistent across the other quarters (P < .001). Conclusions The implementation of the package can reduce inappropriate antibiotic prescription without compromising clinical outcomes. Clinical Trials Registration clinicaltrials.gov; NCT04081051.