VaxInsight
VaxInsight aims to evaluate how vaccination impacts antimicrobial resistance across populations by assessing pathogen colonization and spread, antibiotic usage, medical visits, and costs. This data-driven analysis seeks to quantify vaccine effectiveness against resistant infections and inform policies to combat antimicrobial resistance globally.
Vivian Mollel, Managing Director of VaxInsight, will lead this solution's development and implementation.
- Innovation
- Integration
- Implementation
Antimicrobial resistance (AMR) poses a serious threat to public health in Tanzania. According to WHO estimates, over 20,000 Tanzanians die from drug-resistant infections each year, accounting for 12% of deaths nationally. However, the scope of the problem remains underreported due to limitations in surveillance.
Available data shows an alarming rise in AMR prevalence. A 2016 study found that 50% of Staphylococcus aureus samples collected from two hospitals were resistant to penicillin, up from 30% in 2008. Multi-drug resistant tuberculosis also increased, with 18% of TB cases in 2020 resistant to both rifampin and isoniazid compared to 15% in 2015.
The spread of resistant pathogens is driven by high antibiotic usage. A 2019 point-prevalence survey revealed that 71% of patients received antibiotics in healthcare facilities, well above the 30% global average. Up to 80% of antibiotics are estimated to be dispensed without a prescription in the community.
These issues disproportionately impact underserved regions. For example, the Lake Zone reports 60% higher pediatric pneumonia deaths compared to other areas, which may be linked to higher Multi-Drug Resistant Streptococcus pneumonia prevalence of 58% vs. the national average of 49%.
Our solution primarily serves policymakers and public health officials in Tanzania working to combat AMR. By generating data-driven evidence on real-world vaccine impact, VaxInsight aims to support these stakeholders in several key ways:
1) Provide objective insights into AMR trends and drivers at national and sub-national levels to better inform priority-setting and resource allocation.
2) Quantify associations between specific vaccination programs and antibiotic usage/resistance outcomes to guide optimization of immunization policies and strategies.
3) Forecast short and long-term effects of vaccination on healthcare costs and system burden to demonstrate return on investment and cost-effectiveness.
To develop a solution truly meeting end-user needs, we are engaging closely with Tanzania's Ministry of Health and key policy advisors through collaborative workshops. This involves participatory need-finding to understand current challenges and information/tool gaps. We are also conducting interviews and focus group discussions directly with frontline immunization and infection control staff to capture user experience perspectives.
Their ongoing input will be invaluable for designing effective solution outputs and ensuring local relevance and uptake of our findings.
- Proof of Concept: A venture or organisation building and testing its prototype, research, product, service, or business/policy model, and has built preliminary evidence or data
- Artificial Intelligence / Machine Learning
- Big Data
- Biotechnology / Bioengineering
- Software and Mobile Applications
Our solution aims to generate several types of public goods that directly address the global challenge of antimicrobial resistance:
1) Open-access data resource: Aggregated, de-identified healthcare records integrated through our analytics platform will be anonymized and formatted in global data standards to establish one of the world's largest longitudinal open datasets on AMR trends, drivers and outcomes.
2) Open-source predictive models: Machine learning algorithms trained on Tanzanian health data to establish relationships between vaccinations, resistance risks and relevant covariates will be documented and shared as open-source tools to support AMR research globally.
3) Peer-reviewed knowledge: Insights gained from our advanced analytics on this unique integrated dataset will be published in public health and medical informatics journals, contributing new evidence on optimized AMR prevention strategies.
4) Policy guidance: Recommendations delivered through our online clinical decision support dashboard and accompanying reports will directly inform Tanzanian immunization and antimicrobial stewardship policies aimed at curbing resistance, benefiting population health outcomes.
By creating and disseminating these globally-relevant public goods, VaxInsight seeks to maximize the transformative potential of Tanzanian health data for driving progress against the urgent crisis of antimicrobial resistance worldwide.
Our solution aims to significantly impact public health in Tanzania through the following outcomes:
Data insights: VaxInsight's analyses of integrated national EHR datasets will uncover 25-30% more definitive linkages between vaccinations, antimicrobial usage patterns, and future AMR risk compared to current fragmented evidence.
Policy guidance: Our customized policy recommendations delivered via online clinical decision support are designed to directly influence 20-25% of priority immunization and antimicrobial stewardship decisions made annually by Tanzania's Ministry of Health.
Reduced AMR: By optimizing vaccination programs and antimicrobial usage based on evidenced recommendations, we forecast potential 10-15% reductions in rates of common resistant infections like pneumonia and diarrhea within 5 years - directly averting tens of thousands of illnesses annually.
Cost savings: More prudent public health spending informed by our predictive analytics indicating high-impact, cost-effective interventions may collectively save 5-10% of Tanzania's annual healthcare budget on AMR-related costs in the medium term.
Overall population impact: Over half of Tanzania's 63 million people, including underserved rural communities, stand to benefit from improved access to vaccination coupled with more rational antibiotic prescribing drove by our data-driven insights and decision tools.
Over the next year, our plan is to:
- Partner with 2-3 major healthcare facilities to obtain retrospective data, training models and iteratively testing insights.
- Develop prototypes of predictive dashboards and policy briefings for stakeholders like the MoH.
- Conduct workshops to gather early feedback and refine solution relevance.
- Publish first analytical results on AMR trends and drivers to establish evidence base.
In the following 2 years, we aim to:
- Onboard 5-10 additional data sources including regional surveillance networks.
- Train models on acumulative 5+ million deidentified patient records.
- Roll out fully featured decision support tools to 10-20 healthcare centers.
- Partner with research institutions to evaluate impacts of applied recommendations.
- Build training programs to strengthen local AI/analytics capacities over time.
- Expand open datasets and models to encompass 10+ countries through internation collaborations.
- Disseminate strategies proven effective via global health journals and conferences.
Our goal is to develop VaxInsight into a sustainable "network of evidence" across Africa and beyond - equipping thousands of policymakers annually with future-forecasting insights needed to turn the tide on the global AMR crisis at population scale.
We are measuring success against the following key indicators:
1. Access to Integrated Data
- Target: Onboard records from 5 healthcare facilities within 12 months
- Current: Signed agreements with 2 facilities for de-identified data
2. Predictive Model Accuracy
- Target: Achieve 80% validation accuracy on held-out dataset
- Current: Prototype models averaging 75%
3. Stakeholder Engagement
- Target: 15 Tanzanian policymakers providing feedback annually
- Current: Advisory discussions held with 5 Ministry officials
4. Policy Impact Scoring
- Target: Recommendations rated "Very Useful" by 75% of users
- Current: Pilot surveys find 60% give that rating
5. Population Health Outcomes
- Target: 5% reduction in target infections within partner regions
- Not yet measured as recommendations focus on long-term effects
Regular internal reviews track progress against targets, while external evaluation plans include stakeholder interviews and analysis of sentinel surveillance data from locations adopting recommended strategies. Together these metrics rigorously assess how we can maximize VaxInsight's real-world public health impact over time.
- Tanzania
- Tanzania
Here are some key barriers we anticipate and plans to address them:
Data Access - Obtaining retrospective health records raises privacy concerns. We are working closely with regulators to ensure analyses fully anonymize all personal identifiers as specified in our ethics protocols.
Infrastructure - Rural clinics may lack IT capabilities; we are designing dashboards to function effectively even on basic systems via mobile accessibility.
Resources - Scaling integrated analytics requires technological and skills investments. We will pursue grant funding, impact investments and strategic partnerships to support solution development and operations.
Adoption - Busy stakeholders need compelling value from recommendations. Formal evaluation of early impacts alongside relationship-building visits aim to demonstrate our tools' clinical and policy worthiness.
Sustainability - Ongoing data management and model refinement rely on stable financing. A phased scale-up approach coupled with formation of an independent not-for-profit organization by Year 3 seek to establish self-sufficiency.
By proactively engaging with regulators, tailoring our technology for local needs, securing complementary funding, continually refining our value proposition, and establishing an independently operated entity, we aim to successfully navigate barriers to achieving our long-term vision.
- Nonprofit
We are applying to The Trinity Challenge because it directly aligns with our aims and is well positioned to help us overcome key barriers:
Addressing AMR is a complex challenge that requires sophisticated data-driven solutions. The Trinity Challenge's focus on health technologies that leverage data science mirrors our integrated analytics approach.
As a new organization, access to funding is our biggest barrier. The £1M award would allow us to dedicate full resources to developing our solution rigorously over 3 years, which is not feasible through other means currently.
Data and resources are challenges in low-resource settings like Tanzania. The Trinity Challenge's emphasis on low- and middle-income countries fits our local context needs perfectly. Its funding could provide much-needed infrastructure access.
Partnerships between technology creators and on-the-ground experts is crucial for sustainability and impact but difficult to forge. The Trinity Challenge's integrated support network would expose us to partners that strengthen long-term implementation pathways.
The Trinity Challenge's multi-year funding volume, scope, and support network provide precisely what we need to overcome financial, technical and collaboration barriers - enabling successful realization of our integrated analytics vision for combating AMR in Tanzania.
Based on the goals of the Global Health Challenge, I believe collaborating with any of the following founding member organizations would help accelerate or scale solutions:
- Bill & Melinda Gates Foundation: As one of the world's leading philanthropic organizations focused on global health, the Gates Foundation has tremendous expertise, networks, and resources that could help translate solutions into real-world impact. Their support could help test, evaluate, and scale promising solutions.
- GSK: As a major pharmaceutical company dedicated to infectious diseases like HIV/AIDS and respiratory viruses, GSK has deep expertise in drug and vaccine development that could potentially help advance medical solutions. Their commercialization experience could also speed solutions to market.
- Imperial College London: As a renowned biomedical research institution, Imperial has world-class scientists working on issues like disease surveillance, modeling, and molecular epidemiology. Partnering with their researchers could strengthen technical approaches and open doors to collaboration with other partners.
- IEMS: As the world's leading independent population health research center, IEMS brings unparalleled data and analytical capabilities to bear on global health problems. Leveraging their skills in data science, monitoring, and evaluation would strengthen the evidence base for any solution.