OBJECTIVES: In the Democratic Republic of Congo and other low-resource countries, community-acquired pathogens are increasingly resistant to most locally available antibiotics. To guide efforts to optimize antibiotic use to limit antibiotic resistance, we quantified healthcare provider-specific and community-wide antibiotic use.
METHODS: From household surveys, we estimated monthly healthcare visit rates by provider. From healthcare visit exit surveys, we estimated prevalence, Defined Daily Doses (DDD), and Access/Watch/Reserve distribution of antibiotic use by provider. Combining both, we estimated community-wide antibiotic use rates.
RESULTS: Of 88.7 (95%CI81.9-95.4, 1588/31221 person-months) healthcare visits per 1000 person-months, visits to private clinics (31.0, 95%CI 30.0-32.0, 418/31221) and primary health centres (25.5, 95%CI 24.6-26.4, 641/31221) were most frequent. Antibiotics were used during 64.3% (95%CI 55.2-73.5%, 162/224) of visits to private clinics, 51.1% (95%CI 45.1-57.2%, 245/469) to health centres, and 48.8% (95%CI 44.4-53.2%, 344/454) to medicine stores. Antibiotic DDD per 1000 inhabitants per day varied between 1.75 (95%CI 1.02-2.39) in rural Kimpese and 10.2 (95%CI 6.00-15.4) in (peri-)urban Kisantu, mostly explained by differences in healthcare utilisation (respectively 27.8 versus 105 visits per 1000 person-months), in particular of private clinics (1.23 versus 38.6 visits) where antibiotic use is more frequent. The fraction of Watch antibiotics was 30.3% (95%CI 24.6-35.9%) in private clinics, 25.6% (95%CI 20.2-31.1%) in medicine stores, and 25.1% (95%CI 19.0-31.2%) in health centres. Treatment durations <3 days were more frequent at private clinics (5.3%, 9/169) and medicine stores (4.1%, 14/338) than at primary health centres (1.8%, 5/277).
CONCLUSIONS: Private healthcare providers, ubiquitous in peri-urban settings, contributed most to community-wide antibiotic use and more frequently dispensed Watch antibiotics and shortened antibiotic courses. Efforts to optimize antibiotic use should include private providers at community-level.