Open Agent-Based-Modeling Platform for AMR Stakeholders of Medellín

Open OH-MediRAM platform combines Agent-Based Modeling and computational power to bring out  complexity of AMR in Medellín. It allows to design impact interventions against AMR addressing the specific conditions of Medellín.

OH-MediRAM empowers AMR stakeholders fostering the emergence of citizen-scientist collectives. OH-MediRAM can adapt to the specificities of any LMICs city.

Pilar Donado DVM PhD with more than 15 years in AMR research. She has been member of AGISAR and manages the GHRU Colombia of CGPS (CenterForGenomicPathogenSurveillance) UK.  

We are addressing the understanding gaps on the dynamics of antimicrobial resistance (AMR) in Medellín, Colombia that lead to suboptimal outcomes regarding the targeted strategies to control it.

Medellín Health Institutions have reported pressing antimicrobial resistance concerns vis-a-vis groups of bacteria, prioritized by the WHO at a global scale. Medellín exemplifies AMR risk factors that predominate in cities of low- and middle-income countries (LMIC). It is a densely populated city of 3 million inhabitants. Its social structure is marked by various issues including inequity and poverty. Medellín infrastructure and dynamics favor numerous contacts between human and animal populations within an ecosystem that concentrates AMR selectors, including antibiotic residues.

AMR surveillance systems and research activities in Medellín collect and generate fragmented and disconnected information mainly from hospital settings. The collected data overlooks complexity of AMR. There is a poor representation of citizen- and community-related data. The informality of the economy reaching more than 50% of the total activity, the institutional weakness and the limited available resources compromise the massive collection of accurate and reliable data and metadata. Consequently, the AMR data analytics that are currently run in Medellín are quite precarious.

OH-MediRAM platform targets the individuals, groups and institutions whose objectives are to design and implement interventions to curb the development of AMR in Medellín. The end beneficiaries are the 3 million inhabitants of the city.

Through access to OH-MediRAM, health practitioners could understand and visualize the consequences of their daily practices on AMR and auto-regulate them: e.g. physicians, nurses, veterinarians, responsibles of safety in food chains, pharmacists, wastewater plant managers. Research groups on AMR could use OH-MediRAM for the design and testing of their working hypotheses. The local, regional and national health institutions could run computational simulations to assess the impacts of the different options for their public policies.

In front of the increasing complexity of AMR, the needs of all the targeted audiences of OH-MediRAM relate to bettering the impact assessment of their potential strategies to curb AMR in Medellín. Consequently they could focus the limited ressources on the most effective interventions. 

The flexible open web-application can adapt to the specific needs of the various target audiences. The OH-MediRAM consortium  networks with the key AMR stakeholders in Medellín engaging them in the development of the solution.

OH-MediRAM provides three public goods,  

1. The open access web application will allow AMR researchers, professionals, and organizations to access and use the information transparently, fostering collaboration and advancing knowledge to addressing AMR in Medellín. This will allow improving the decision making process towards development and implementation of public policies to address AMR.

2. OH-MediRAM consortium will issue an annual update of AMR dynamics in Medellín, insuring, as a public good, that information on antimicrobial resistance dynamics stays up to date. This not only directly benefits the consortium but also provides a valuable tool for relevant stakeholders to continuously adjust and improve their practices, thereby contributing to the effective management of antimicrobial resistance. 

3. OH-MediRAM will develop an adaptability protocol, which facilitates the application of the Agent-Based Model (ABM) in other LMICs regions and geographies.  This extends the impact of our solution beyond Medellín, providing other communities with capacity to use and adapt novel technologies to their own needs, considering sociocultural specificities.

Currently the fight against AMR in Medellín is mainly concentrated with following protocols in hospital settings. The AMR dynamics in the OH domain is largely unknown. Unlike in HICs, the AMR situation that is observed in hospital settings may not correlate well with the AMR situation at community level because of the inequal access to quality of health services and because of greater diversity of Social Determinants of Health (SDoH). As a result, a tremendous difficulty exists in the decision-making process about the impact interventions to be implemented to curb the observed trends in AMR increase in Medellín.

By integrating conventional microbiology techniques, genomics bioinformatic analysis, digital epidemiology, and agent-based modeling in OH domain, OH-MediRAM will enable comprehensive surveillance and simulations of AMR patterns facilitating accurate information about AMR dynamics in Medellín. This will allow the development and implementation of effective interventions to mitigate AMR risks. Our solution can guide targeted interventions in healthcare facilities, communities, and food production systems. OH-MediRAM will also benefit vulnerable populations in Medellín, including those living in low-income districts characterized by higher rates of poverty and social inequalities since they are completely overlooked in the current surveillance program.

Currently the AMR stakeholders of Medellín are locked in solutions mainly defined in HICs which do not address the specificities of local conditions. Addressing AMR in countries with strong institutions and low economical informality (<15%) can be delivered by top-down strategies. In LMICs with weak institutions and high economical informality (>50%) these top-down strategies alone are unlikely to have any significant impact. Bottom-up strategies directly involving the field AMR actors are therefore key to complement top-down approaches and to compensate institutional weakness.

 The Open OH-MediRAM platform allows to design both top-down and bottom-up impact interventions against AMR. Through marketing and science communication campaigns, the field AMR stakeholders, including citizens, can be incited to use OH-MediRAM platform to collectively scale out AMR behavioral changes. Through its playful representations of AMR dynamics in Medellín, OH-MediRAM is a powerful sensitization and educational tool for diverse publics involved in addressing AMR: i.e. health professionals, food distributors, retailers and consumers, patients, educators and civil society organization. As a result, it is expected that behavioral changes will minimize AMR risk practices.   

We use the Balance Scorecard approach with Key Performance Indicators along  four dimensions: 

1. OH-MediRAM customer dimension 
   1. number of users 
   2. number of AMR interventions based on the ABM of OH-MediRAM
   3. Outcomes of AMR interventions based on the ABM of OH-MediRAM
   4. Number and outcomes of research programs using the ABM of OH-MediRAM
   5. Number and outcomes of AMR stakeholder sensitization events using OH-MediRAM platform.
   6. Number and outcomes of OH-MediRAM inputs in the National Action Plan against AMR
   7. International expansion of OH-MediRAM
2. Financial Dimension
   • Financial sustainability of OH-MediRAM 
   • Financing attractiveness
   • Revenue stream
3. Internal Process Dimension
   • Number of modelled agents, interactions, and environmental indicators 
   • Model validation  
   • Model proficiency test. 
   • Informative vs uncertainty levels of insights
   • Volume and quality of generated data and information
   • ABM web platform proficiency 
4. Learning & growth dimension 
   • Number of OH-MediRAM consortium members 
   • Research capacity development. 
   • Number and outcomes of collaborations
   • Communication skills development 
   • Critical thinking and problem solving skills

We identified the following possible barriers:

1. Bureaucratic political challenges like negotiating complex political structures and bureaucratic processes, such as sampling permits, can impede the data flow, swift decision-making and hinder the progress of our project.
   Mitigation strategy: we will engage in strategic advocacy efforts, building alliances with civil servants to streamline bureaucratic processes, and agree on the protocols to ensure project deliverables to time.
2. Cultural resistance to change poses a significant challenge, potentially affecting the acceptance and adoption of our solution. Addressing fear within communities related to new technologies or practices is crucial for building trust and facilitating widespread acceptance. Gaps in education about the use of antimicrobials and their resistance may hinder understanding of the necessity of our project.
   Mitigation strategy: we will implement culturally sensitive approaches, involving local communities in the design and implementation phases to address specific cultural concerns.We will develop transparent communication strategies to alleviate fears, emphasizing the positive impact and safety of our solution; through comprehensive educational campaigns, leveraging multiple channels such as workshops, community events, and digital platforms to bridge education gaps. 

The challenge and questions raised by the Trinity Challenge on AMR perfectly fit with the working hypotheses that are at the basis of the OH-MediRAM project.

The highly innovative approach of our solution requires non-conventional visionary financing strategies such as that promoted by the Trinity Challenge.

The back-up of such a visible international organization will provide decisively support to overcome the principal local barriers that we have identified. The Trinity Challenge will enhance our science communication strategy aiming to prompt behavioral changes to curbing AMR in Colombia and globally.

OH-MediRAM consortium would like to establish a collaboration with the Sante Fe Institute (New Mexico, USA), pioneer and indisputable leader in complex system science which is at the core of the innovation of OH-MediRAM project.