Ethical open-source digital ID for vaccine delivery

An open-source ‘Global Good’ biometric digital ID to verify vaccine delivery, fight pandemics, and strengthen health systems

Dr. Toby Norman, CEO & Co-founder, Simprints

Vaccines are essential to fight pandemics and prevent future pandemics from occurring. However, the WHO estimates that up to 50% of vaccines may be wasted globally every year and do not reach their intended beneficiaries. In low-income countries, a key part of the challenge is poor quality data on which patients are actually receiving vaccines. In vaccine coverage estimates, there are often huge gaps between government administrative data and independent surveys. For example, in Nicaragua, administrative data estimated an 82% coverage rate for measles immunisations while samples revealed an effective coverage rate of just 50%. In Bangladesh, official estimates near 99% are questioned by published health studies suggesting 54% of children do not receive timely vaccinations.

Furthermore, administrative data often relies on health cards that are frequently lost, damaged, or misplaced: in Nigeria, only 1 in 3 children have vaccine cards to prove their immunisation history. These issues are compounded by what Gavi describes as a “global identity crisis” in health data, with nearly 1 billion people lacking any birth registration or ID. This lack of good quality data makes it extremely difficult to ensure the most vulnerable receive vaccines and other healthcare essential to fighting pandemics. 

Our solution directly serves:

1. Frontline Health Workers. Simprints helps frontline health workers identify patients. Quickly accessing medical history such as immunisation records is essential for health workers fighting a pandemic. However, in low resource settings >1B people lack formal ID, patient names often overlap, and program-issued cards fail when they are lost or damaged. For example, research in Chad found that 59% of mothers lost their infant’s vaccination cards within 2 years. To solve this, Simprints spent over 200 hours with frontline workers co-designing a biometric system that allows them to quickly access the right record at the right time (see Case Study).
2. Public Health Decision Makers. Second, Simprints ensures the data that public health professionals are basing decisions on during a pandemic is accurate. Weak identification leads to duplicate records, over/under- reporting, and gaps in the continuity of care. In a study in Lahore, 35% of records were found to be 'unsatisfactory and inaccurate' and 42.5% of the reports either under or over-reported health data. By biometrically verifying the delivery of every vaccine or treatment, public health professionals can act on accurate coverage data to respond to the crisis. 

While multiple biometric vendors exist, their technologies are proprietary, risking vendor lock-in and preventing digital public goods approaches. According to industry research, vendor lock-in is the #1 complaint of African governments considering patient ID systems,  Furthermore, few technologies are optimised for “last mile” contexts with limited-to-no internet, low-end mobile devices, and non-Western biometric profiles. This can lead to exclusion and lower quality care for the most vulnerable patients.  

With support from the Trinity Challenge, we propose creating the world’s first open source biometric patient ID system for global health. This ‘Global Good’ technology will be openly available under BSD and Apache v 2.0 license for Ministries of Health, researchers, and NGOs leading the fight against pandemics. In addition to open sourcing our core biometric technology, we propose open sourcing our privacy toolkits, training materials, and other knowledge bases essential for effectively deploying biometric digital ID. Our goal is to maintain all of these resources openly under the Digital Public Good approach endorsed by the U.N. Secretary General for technologies that can support achievement of the SDGs.

Biometric IDs create tangible impact by 1) helping frontline workers deliver the right care to the right patient and 2) radically improving the quality of accurate, real-time data for public health decision makers. While there are multiple verticals where biometric IDs can help, a clear use case is vaccines where they can identify patients, prevent duplicate dosing, and ensure precision data on the number of individuals vaccinated, empowering data feedback loops critical for a rapid public health response. A review of over 160 biometric deployments highlighted their ability to increase accuracy, reduce cost, and strengthen data quality. 

Furthermore, global health experts have repeatedly called for biometric digital ID to support pandemic response. After the 2014 Ebola epidemic, the World Bank highlighted these tools as potential ‘game changers’ in fighting future pandemics, as well as helping countries recover faster by “strengthening the mechanisms for service delivery, minimizing waste and maximizing the outcomes of investments made”. Again during the Covid-19 pandemic, organisations like Gavi are once again saying “biometric digital IDs can be a game changer” in the fight. Open sourcing Simprints as a Global Good may reduce the risk of the world repeating this message for a third time.

As a closed sourced technology, Simprints has helped deliver healthcare to >1.2M people across 14 countries.

However, as an open source Global Good, we believe Simprints could radically scale its impact. For example, DHIS2, an open source health platform widely adopted by many ministries of health, has historically helped deliver vaccines to 750 million people. Funded by Gavi, Simprints is integrating with DHIS2 in Bangladesh to use biometrics to support vaccine delivery. As a closed-source solution, this project will likely scale slowly without future donor funding, but as an open source solution this integration would be free to be adopted and deployed by immunisation programmes globally.

Similar arguments extend beyond vaccines to other essential activities in pandemic response like contact tracing and case management. In Ethiopia, Simprints is integrated with CommCare for patient case management by Ministry of Health community health workers. This programme will reach several hundred thousand patients this year, but globally CommCare reached over 350 million patients in 2020 alone. 

By going open source with proper investment in code documentation, privacy toolkits, and training materials, Simprints has the potential to radically increase adoption and help ensure essential health services accurately reach millions of people. 

We measure success against several progress indicators, broadly broken down into 1) technical KPIs and 2) project impact KPIs.

First, in order to help patients, Simprints must be technically accurate, fast, secure, and lightweight (i.e. have a low size and processing footprint for low-end mobile devices common at the last mile). As of our latest 2020.4.0 release, Simprints has >98% average accuracy rate, can search >3,000 records per second, with end-to-end encryption, and an average 261 byte per template size.

Second, in order to measure project impact, Simprints tracks monthly patient enrolments complemented by independent impact evaluations. Our last three cumulative monthly enrolments were 1,240,726 (March), 1,169,815 (February), 1,120,521 (January). See section on ‘evidence’ for examples of independent impact evaluations.   

Successfully open sourcing and maintaining Simprints as a Global Good requires overcoming multiple barriers. According to criteria set forth by PATH Digital Square, the first barrier is updating and adequately documenting code-bases for open source. Here it’s important to distinguish practically between “open source in name only” versus “truly open source”, which is more complex than simply making code public. Truly open source Global Goods have codebases accessible on well known repositories like GitHub, strong technical documentation, and architectures designed explicitly for adoption by technical partners servicing Ministries of Health, researchers, or NGOs. To overcome this hurdle, with support from the Trinity Challenge, Simprints will refactor and publicly maintain our code, in addition to making public open source integrations with leading data collection platforms DHIS2, CommCare, OpenSRP, and ODK.

Second, technology alone is never a silver bullet. As we’ve seen across six years of global health delivery, successful impact requires strong focus on training, systems, and communication. To support organisations deploying Simprints to fight pandemics, we will additionally open source and maintain our training materials, privacy toolkits, social and behavioral change communication (SBCC) materials for community sensitisation, and related technical project management resources.

Simprints works with partners across the public and social sector, including:

• Ministries of Health. Ethiopia, Bangladesh, and Ghana

• NGOs. BRAC, MercyCorps, D-Tree, KhushiBaby
  

• Researchers. Brown University, UNC, LSHTM
  

• Funders. Gavi, USAID, Children’s Investment Fund Foundation, Fondation Botnar, ARM, Cisco, and the Elevate Prize Foundation

Developing robust, ethical, and inclusive digital ID systems before the next pandemic requires intentional and far-sighted investment. We saw calls for biometric digital ID systems after the Ebola pandemic in 2014 and the ongoing Covid-19 pandemic. 

However, the majority of digital health funding is designed around specific project interventions, not preparing for the future. While project-based funding is important, it's often 1) reactive, 2) heavily restricted, and 3) rarely supports building foundational data infrastructure. In contrast, support from the Trinity Challenge can create an open source 'Global Good' for patient ID systems that not only prepares us better for the next pandemic, but strengthens foundational health systems in the interim. 

Trinity Challenge members also represent a coalition of global health implementers, researchers, and experts essential for iteratively developing a new technology. For example, CHAI supports Ministries of Health globally on technology uptake, Johns Hopkins leads research on digital interventions, and Microsoft builds the architectures many of our systems run on. Working together, we could ensure better patient ID systems are thoroughly tested and embedded in digital systems in time for future crises, and not just spinning up after a new pandemic has already started.

Among Trinity Challenge members, we would be excited to partner with:

• CHAI - for collaborating with Ministries of Health in Africa on the testing and implementation of digital health solutions like Simprints (we’ve previously worked with PATH, JSI, and others)

• Imperial College London & Johns Hopkins University - for global health research and evaluations supported by digital health interventions (we’ve previously worked with London Centre for Neglected Tropical Diseases, Brown University, and Cambridge University)

• Microsoft, Google, and Infosys - for technical collaborations and potential engineering support (we currently use both Microsoft Azure and Google Cloud Platform, as well as multiple technical frameworks and programming languages like Kotlin developed by these organisations)

We are currently working with Trinity Challenge Member Zenisys on the eCHIS programme in Ethiopia.

Beyond Trinity Challenge members, we would be excited to extend our partnerships with Ministries of Health and NGOs where an open-source 'Global Good' biometric digital ID could drive real health impact.