Accurate detection of antimicrobial resistance (AMR) depends on adequate laboratory capacity. We aimed to document key weaknesses hindering AMR detection at various tiers of laboratory networks in 14countries in sub-Saharan Africa, and analyse their significance in AMR surveillance and policies. In this analysis, we obtained retrospective data on AMR and antimicrobial consumption from 14countries participating in the Mapping Antimicrobial Resistance and Antimicrobial Use Partnership between May 1, 2019, and June 30, 2020. We established the number, capacity for AMR detection, and other characteristics from all bacteriology laboratories within the national network from existing national health system compendiums, combined with a self-applied scored survey tool. We calculated AMR detection readiness scores at a facility and country level and analysed the pertinence of national strategies to address system gaps for AMR detection and surveillance. Of the 53 770listed laboratories, 675 (1%) were formally assigned to deliver bacterial testing and invited to complete the survey tool. Of the 504 (75%) facilities that returned the completed survey, 393 (78%) reported antimicrobial susceptibility testing (AST) capacity and collectively provided geographical access to less than 50% of the general population in seven countries. Continuous access to water was reported by 347 (88%) laboratories, power sources by 341 (87%) laboratories, and the presence of qualified laboratory scientist or technologist by 377 (96%) laboratories. By contrast, ISO15189 accreditation was reported by 90 (23%) laboratories and use of an electronic laboratory information system by 54 (13%). Reference laboratories were associated with higher AMR detection readiness scores than were district laboratories (odds ratio 4·7 [95% CI 1·3-10·2; p=0·014). Private, not-for-profit laboratories were associated with higher scores than were government-affiliated facilities (9·2 [1·6-53·8; p=0·014). Designated national AMR sentinel sites were associated with higher scores than were non-sentinel sites (5·8[2·9-11·8; p<0·001). Laboratories processing between 1001and 3000bacterial cultures annually were associated with higher scores than were those processing less than 200cultures annually (4·8 [1·7-13·7]; p=0·0040). Strengthening bacterial testing and capacity for AMR detection represented less than 20% of the proposed interventions in 12of the 14national AMR action plans. AMR action plans and other relevant national strategies should prioritise the scale-up of bacterial testing services to improve access to care and promote quality AMR surveillance. Interventions that democratise AST to lower laboratory tiers, formally designate AMR sentinel sites, and implement whole-of-network laboratory information and quality management systems are urgently needed. Fleming Fund, UK Aid, and US Centers for Disease Control and Prevention through the Training Programs in Epidemiology and Public Health Intervention Networks.
Read full abstract