Background

Childhood vaccination is a cost-effective public health intervention with proven positive outcomes on individual, community and global scales as some of the highest return of investment into healthcare.1 India’s vaccination programme is one of the largest in the world, covering 27 million live births per year and 100 million children under age 5 alone, but coverage is far from universal. Recent investments and improvements in India’s vaccine programme are working to improve the rates of immunised children, including the recent Mission Indradhanush.2 However, immunisation data in India is stored in complex fragmented data storage systems that have led to mismatch in resources and system-wide shortfalls. The fragmentation has been due to various innovations being developed in real time in a decentralised manner.3 With the global need for tracking immunisation data due to the COVID-19 pandemic, solutions for improving immunisation data storage are critical. There is a great deal of interest surrounding use cases for blockchain technology in development and healthcare.4 Blockchain, the technology underlying cryptocurrencies such as bitcoin, can be especially useful in improving decentralised and error prone information storage.5 There has been limited use of blockchain technology in the field of healthcare data management despite the enthusiasm surrounding it,4 specifically no prior exploration of the use of blockchain technology in the national storage of healthcare data. In partnership with the Biotechnology and Research Council of India and Gates Grand Challenges India as part of the Immunisation Data: Innovating for Action programme, we explored means for improvement in digitisation of data through the use of blockchain technology.

Needs assessment of Indian immunisation data storage

Exploring the current state of immunisation data storage in India was crucial to avoid increasing further complexity. In-depth interviews with national and state level policy-makers, technological partners and front-line workers were key in elucidating the current needs and opportunities. Themes of data quality, ease of use, mutability of data, infrastructure fault tolerance and cost were most commonly discussed. E-vin is most frequently used to store supply chain data and has had many successes in standardising and improving data collection by focusing on oversight.6 Data size and security have been a challenge with the increasing amount of centrally stored data. On the beneficiary side, the data collection and storage methods varied depending on the state with increasing fragmentation of data storage and frequent difficulty with collating data due to network and central server failures. After a national needs assessment, our prototype work was focused on Gujarat to help connect beneficiary health data from TECHO+, their health data app,7 to supply chain side data. Invested technological partners were key to assuring that the possible solutions generated through this process are able to be implemented in the future.

The lack of communication between the different databases and an inability to easily collate this data was one of the major issues that could be helped by a blockchain-based solution. Blockchain technology is particularly suited to solving problems of storing asynchronous data securely that can be changed from multiple touchpoints. The multiple storage options that already exist in India can thus be connected using a blockchain-based connector system without adding more complexity to an already congested landscape because blockchain technology allows for simultaneous access to data from multiple touchpoints while storing the history of any data changes, thereby ensuring security.

Blockchain as a connector database

Based on an in-depth exploration of the current immunisation data storage landscape, the study team worked with government officials and immunisation stakeholders to highlight the costs and benefits of the use cases of blockchain in immunisation data in India and to prototype a blockchain-based immunisation data connector called ‘Anveshan’. The blockchain-based solution was created to solve the four underlying key problems identified: (1) need for parsimonious solution to data fragmentation, (2) fault tolerance, (3) data sharing with researchers and (4) automatic incentivisation for front-line workers to improve data quality.

Anveshan, the blockchain-based connector database that was developed, connects a beneficiary side and supply side databases (figure 1). After a survey of the technological landscape, we found success with a prototype using the open source Hyperledger blockchain core and GraphQL—a technology developed by Facebook’s research organisation that aims to enable easier more efficient data querying and interaction with complex data stores and schemas.8 GraphQL was able to stitch together different data sets with an overlapping data field allowing Anveshan to connect existing siloed immunisation data stores (figure 2). Data security could be improved by incorporating disincentives for inappropriate data modification directly into the blockchain. Results from data simulations using generated data showed that the Anveshan database was able to decrease data fragmentation by querying across both supply and beneficiary databases. Aggregate analysis capability was increased through development of a dashboard that allowed policy-makers and immunisation programme staff to receive real-time query feedback. These analyses are especially helpful in following the utilisation of immunisation stock at a granular level to allow prediction of immunisation needs by village as well as decreasing vaccine waste and fraud. Anveshan allowed for certain data to be downloaded after deidentification to encourage innovation and research through public private partnerships. After a thorough review of costs and benefits, a private blockchain database was suggested as the optimal solution to balance between benefits like fault tolerance and cost. Each state data repository could be a separate node on the blockchain to ensure quality while ensuring more local control of data to improve data accessibility and upload.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Schematic of data systems connecting with Anveshan.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2

Schematic diagram illustrating data storage structure for proposed private blockchain immunisation data storage for Gujarat, India. Example data IDS are used to illustrate the nodes in the blockchain storage schematic.

Blockchain in immunisation data: costs and benefits

The need for tracking vast amounts of immunisation data is an especially important imperative now with the need for ensuring efficient distribution and utilisation of COVID-19 vaccines in a country as vast as India. Blockchain technology can be particularly helpful in an already complex and fragmented data storage system in India by allowing for greater security by storing multiple copies of immutable data. However, the connecting database and a blockchain solution has not been implemented in practice due to real-world acceptance and scalability problems. Balancing the needs and motivations of a complex landscape of stakeholders remains a challenge to ensure that new technologies can be implemented. Initial buy-in from the primary owners of the immunisation data would have made faster implementation of the technological interventions described here. There are a variety of existing blockchain technologies, and trade-offs are unique to each problem. Blockchains cannot inherently solve the problem of faulty data but can be useful in ensuring the security/integrity of the data after it is stored. Especially in settings with significant variability of reliability of internet connections, distributed fault resistant data storage systems like blockchain can ensure data does not get lost as a result of technology infrastructure issues. Blockchain technology also brings strong data privacy guarantees through the strong encryption schemes that are used in the private–public key-based accounts that are at the core of modern blockchains.

For healthcare data storage, the fundamental trade-off occurs between security and efficiency. Private blockchains are more efficient and cheaper while still preserving a lot of fault tolerance and transparency of their public counterparts. Having multiple servers across India on the national, state, and locality level participating in the blockchain will enable significant fault tolerance that will be able to sustain a subset of servers failing, unreliable internet connections, and other issues plaguing data storage in this vast, low resource environment. We have seen so far that blockchains provide unique benefits and could be powerful as a backbone for a national or state level data, but architecture and configuration matters. A wholistic approach to studying the development and implementation of a new technology in healthcare data storage can be emulated to ensure thoughtful introduction of new technologies as a means to add value to the healthcare landscape.