One Health drivers of antibacterial resistance: quantifying the relative impacts of human, animal and environmental use and transmission

AbstractIntroductionAntimicrobial resistance (AMR), particularly antibacterial resistance (ABR) is a major global health security threat projected to cause over ten million human deaths annually by 2050. There is a disproportionate burden of ABR within lower- and middle-income countries (LMICs), but it is not well understood how ‘One Health’ drivers, where human health is co-dependent on the health of animals and environmental factors, might also impact the burden of ABR in different countries. Thailand’s “National Strategic Plan on Antimicrobial Resistance in Thailand” (NSP-AMR) aims to reduce AMR morbidity by 50% through a reduction of 20% in human antibacterial use and a 30% reduction in animal use starting in 2017. There is a need to understand the implications of such a plan within a One Health perspective that mechanistically links humans, animals and the environment.MethodsA mathematical model of antibacterial use, gut colonisation with extended-spectrum beta-lactamase (ESBL)-producing bacteria and faecal/oral transmission between populations of humans, animals and the environment was calibrated using estimates of the prevalence of ESBL-producing bacteria in Thailand, taken from published studies. This model was used to project the reduction in human ABR (% reduction in colonisation with resistant bacteria) over 20 years (2020-2040) for each potential One Health driver, including each individual transmission rate between humans, animals and the environment, exploring the sensitivity of each parameter calibrated to Thai-specific data. The model of antibacterial use and ABR transmission was used to estimate the long-term impact of the NSP-AMR intervention and quantify the relative impacts of each driver on human ABR.ResultsOur model predicts that human use of antibacterials is the most important factor in reducing the colonisation of humans with resistant bacteria (accounting for maximum 72.3 – 99.8% reduction in colonisation over 20 years). The current NSP-AMR is projected to reduce the human burden of ABR by 7.0 – 21.0%. If a more ambitious target of 30% reduction in antibacterial use in humans were set, a greater (9.9 – 27.1%) reduction in colonisation among humans is projected. We project that completely limiting antibacterial use within animals could have a lower impact (maximum 0.8 – 19.0% reductions in the colonisation of humans with resistant bacteria over 20 years), similar to completely stopping animal-to-human transmission (0.5 – 17.2%). Entirely removing environmental contamination of antibacterials was projected to reduce the percentage colonisation of humans with resistant bacteria by 0.1 – 6.2%, which was similar to stopping environment-human transmission (0.1 – 6.1%).DiscussionOur current understanding of the interconnectedness of ABR in a One Health setting is limited and precludes the ability to generate projected outcomes from existing ABR action plans (due to a lack of fit-for-purpose data). Using a theoretical approach, we explored this using the Thai AMR action plan, using the best available parameters to model the estimated impact of reducing antibacterial use and transmission of resistance between populations. Under the assumptions of our model, human use of antibacterials was identified as the main driver of human ABR, with slightly more ambitious reductions in usage (30% versus 20%) predicted to achieve higher impacts within the NSP-AMR programme. Considerable long-term impact may be also achieved through increasing the rate of loss of resistance and limiting One Health transmission events, particularly human-to-human transmission. Our model provides a simple framework to explain the mechanisms underpinning ABR, but further empirical evidence is needed to fully explain the drivers of ABR in LMIC settings. Future interventions targeting the simultaneous reduction of transmission and antibacterial usage would help to control ABR more effectively in Thailand.