Saving lives by optimizing the MDA vs AMR tradeoff

Mass azithromycin distribution to children under five in high mortality LMICs drastically reduces mortality, but also inflates antibiotic resistance. We will apply advanced analytics and mathematical models to community-level data in order to optimize recommendations for this intervention, ensuring lives are saved while minimizing the risk of increasing antibiotic-resistant infections.

Seth Blumberg MD PhD

The Sustainable Development Goals include reducing preventable under-5 mortality to 25 deaths per 1000 live births by 2030, but communities throughout West Africa are still facing mortality rates four times above the target. Highly effective and feasible interventions to reach that target are needed. Distribution of azithromycin to all children under five has been shown to reduce childhood mortality by 14%, leading the World Health Organization (WHO) to recommend its use in sub-Saharan settings with high childhood mortality. In addition, there are other uses of mass drug administration (MDA), such as for control of neglected tropical diseases. However, more treatment will increase the prevalence of antimicrobial resistance (AMR) in a wide range of bacteria. Thus, the benefit of MDA needs to be carefully balanced against the risk of AMR.

Ministries of health need decision-making tools to help balance the uncertain potential for increasing long-term AMR-related morbidity and mortality against the immediate benefit in reducing child mortality. Scale up of evidence-based MDA programs could save five thousand pediatric deaths annually in Niger alone. Our solution focuses on providing evidence-based decision making tools regarding who to target with MDA, how often to provide treatment, and how long to continue treatment.

Our target audience are global, national, and local public health policy makers who desire more tools for assessing when the potential benefits of MDA outweigh the risks. As addressed in a World Health Organization (WHO) report, support for MDA in areas with high child mortality rates (i.e., > 6% infant mortality) is contingent on close monitoring of AMR.  By providing tools for assessing the level of AMR risk, our solution will expand access to life-saving MDA in vulnerable LMIC communities.

Through our team’s existing work in global health, we have built an extensive network of relevant stakeholders at all levels of governance. To ensure our goals are aligned with public health needs and cultural norms, we will continue our engagement with local, national, and global decision-makers including WHO, ministries of health, and local community leaders in LMICs.

We will engage with all stakeholders to obtain guidance on how our web-based platform can be packaged in as effective a way as possible to facilitate decision-making. We want policy-makers and the community-at-large to feel that our solution incorporates a nuanced perspective and balanced evidence base.

As part of our solution, we look forward to supporting the formation of a consortium of researchers in Africa using advanced analytics and modeling to advance public health. This consortium will serve several broad functions including:

1. Connecting researchers in LMICs and highly-resourced institutions to promote bi-directional exchange of ideas.

2. Building capacity for LMIC-based research and entrepreneurship via exchange of graduate students, postdocs and faculty amongst institutions. The evolving capacity to apply data analytics and modeling to public health problems will have wider impacts on not only AMR, but other infectious disease challenges as well.

3. Engaging with Ministries of Health, local healthcare workers, NGOs and other public health stakeholders to improve the evidence-to-policy pipeline.

Our team is committed to open, reproducible science and FAIR principles. Given our academic affiliations, we are committed to communicating our scientific and public health findings via publications, and conference presentations. We aim to be the most recognized international resource for evaluating the impact of MDA on childhood mortality and AMR prevalence.  We will also look forward to providing insights regarding long term risks of AMR colonization, AMR forecasting, and other relevant topics. We anticipate opportunities to engage with other Trinity Health teams in this regard.

We will create tangible impact by providing MDA in at-risk areas in a safe and effective manner to improve the health of vulnerable children in LMICs. We want to eliminate barriers for children to receive life-saving preventive care afforded by MDA in the short term, whilst also protecting them and their communities from long term AMR accumulation. LMICs have the highest infant mortality rates and also the highest prevalence of childhood diseases that AMR would harm the most (e.g., pneumonia, malaria, and tuberculosis). Thus the need to find the right balance is crucial and technologies that help in this regard will have high impact.

As part of our solution, we will construct a web platform that organizes existing data streams, and provides an interface with open-access models that facilitate public-health decision making. We will model our interface after other tools that have been disseminated for tuberculosis, neglected tropical diseases, and other diseases. Our web platform will also include educational resources for the general public about infectious diseases, the appropriate role of antibiotics, and the risk of mis-using antibiotics.

In the first year, we will focus on Niger and Burkina Faso, where we have the most data from cluster-randomized trials. Our focus will be on understanding current evidence on the benefit of MDA in context of the concern of increasing AMR risk. We will use this evidence to help policy makers refine their goals for MDA distribution. 

In the following year, we will extend our analysis of MDA to other LMICs in Africa, where evidence for MDA is strongest. We will also extend our analysis to consider how the use of MDA to reduce childhood mortality compares with other MDA programs, such as those for reducing neglected tropical diseases such as trachoma.

In the third year, we will consider whether there may be a benefit in implementing MDA to reduce childhood mortality for countries outside of Africa. We will also consider how other types of antibiotic campaigns, such as seasonal malaria chemoprevention, may impact AMR prevalence.

We will use several metrics to monitor success:

Evidence-to-policy guidelines: We will track how many scientific outputs our technological solution generates. These could include specific recommendations such as how to optimize the efficiency of AMR surveillance following MDA, or web-based simulation tools.

Policy change: We will track how many policy documents reference our solution via trackers. We will also maintain qualitative reports via our professional networks of how policy makers are using our solution.

Public engagement: We will monitor metrics for our public-facing website. These include how many unique sites use the website, how often they use it, and what countries they are from. We will use these metrics to learn what analytics features are routinely accessed, and which educational resources are most widely distributed.

Training: We will monitor how many LMIC trainees are involved with this solution, and skills they acquire in the process

Peer-reviewed publications: We anticipate at least two peer-reviewed manuscripts a year as a result of this work, and will keep track of progress

Funding: We view our solution as just one-step in a long road of opportunity for reducing infections and AMR in LMICs.  We will track future funding that is generated from our solution.

Knowledge gaps: We will address knowledge gaps by prospectively validating our modeling solutions.  Our technologies will permit forecasts of the outputs for several ongoing surveillance efforts and clinical trials that we are involved with (e.g community surveillance in ARCH, resistance patterns in AVENIR). By evaluating the accuracy of forecasts we will increase the confidence in our quantitative understanding of MDA and AMR dynamics

Cultural acceptance: Cultural barriers will be addressed with ongoing engagement of stakeholders. This will be accomplished in part via our web platform that provides an interface for accessing data, models, and educational resources. We will also continue to engage with local health care providers, for which we will have a dedicated ‘stakeholder engagement’ working group.

Institutional barriers: Currently, there is limited application of WHO’s guidelines for using MDA to reduce childhood mortality. Curiously, similar MDA has been utilized for reducing trachoma transmission with less institutional reluctance despite comparable risk of increasing AMR prevalence. We will seek to understand institutional context so that we can specifically address current public health, economical, and political concerns. We expect that early involvement of policy makers will improve the uptake of our solution for optimizing the utilization of MDA.

Public health urgency: There is an urgent need to curb the global rise in AMR prevalence, while still providing the life-saving treatments that antimicrobials offer. The visionary Trinity Challenge permits our team to leverage years of research to address this urgent challenge. The potential to save tens of thousands of lives is at stake.

Funding: Our transdisciplinary, unique approach to reducing the impact of antimicrobial resistance and bacterial infections does not fall easily into traditional funding opportunities. The trinity challenge allows our team to pursue all aspects of the evidence-to-policy pipeline so that scientific advances translate into public health gains.

Networking: The Trinity Challenge provides a framework for strengthening and expanding our partners and collaborators. The discussions our team has had about our application has already led to new ideas and connections. This type of transdisciplinary networking is crucial for moving beyond scientific studies towards enacting meaningful evidence-based public health reform.

LMIC training and education: The challenge supports our team’s dedication to improving LMIC research capacity and strengthening educational resources for low-resource settings. The education empowerment of vulnerable subpopulations, healthcare workers, and other invested individuals is necessary to create sustainable, effective solutions to the challenges of AMR.

Public health stakeholders: Beyond our existing connections with individual Ministries of Health, stakeholders that will help translate our scientific discovery into public health solutions include the Africa Centres for Disease Control and Prevention, Clinton Health Access Initiative, and the World Health Organization (Africa branch). We are also excited to connect with the extended network of Trinity Health partners.

Communication specialists: Much of public health relies on individual decision-making. As the COVID pandemic demonstrated, the tendency for misinformation to spread and affect human behavior was stark. To combat AMR, we need to promote evidence-based decision-making and behavior with regard to the use of antimicrobials, health hygiene and One Health stewardship. Additional expertise in this aspect of our solution would help to create impactful change.

Entrepreneurship catalysts: Our solution currently does not incorporate entrepreneurial activities.  As we anticipate our solution may save tens of thousands of lives and avert even more infections, we believe it will create concrete value. We recognize that a solution’s value can often be amplified via monetization of its value. Thus, experts in catalyzing entrepreneurship activities would help scale the impact of our work. Entrepreneurship activities would also help build LMIC capacity and promote sustainability of our solution.