Book at chat
By Richard Pope —

What the NHS Single Patient Record can learn from India’s Ayushman Bharat Digital Mission

Part of India’s National Health Authority, the Ayushman Bharat Digital Mission was founded in 2021 to design, build, operate and scale digital public infrastructure for India’s health system. Among the products and platforms it is responsible for are the Unified Health Interface, which provides open protocols for linking medical records, making bookings and managing consent; and the ABHA App which lets patients maintain a copy of their health records, access services and manage consent.

Since 2021 it has scaled its platforms to over 400,000 health facilities, registered over 810 million digital accounts and linked nearly 740 million health records. It did this with an initial 5 year budget that roughly equates to about £150 million.

In this 2023 talk, Kiran Anandampillai, Technology Advisor at the National Health Authority describes the approach.

In the UK (or more specifically, England) the Single Patient Record is a new initiative that aims to remove “the need for patients to repeat their medical history unnecessarily at every appointment”, and manage and “read, update, and share care plans with clinicians”. It is not the first time joining up healthcare data has been attempted.

The Indian and British health care systems are very different, but what might the NHSlearn from the successful implementation of the Ayushman Bharat Digital Mission’s services and platforms?

1. Starting small is the route to scale

The platforms and services were initially launched in 6 ‘Union Territories’ which are directly governed by the central government. After about 18 months, 173,369,087 Ayushman Bharat Health Accounts have been created and 10,114 doctors and 17,319 health facilities had been registered. Only then were the platforms and services opened up to health facilities and patients nationwide.

While those 6 areas are in themselves very large, at least by UK standards, the principles restricting the launch of a service by geography is similar to the one deployed by the UK’s Universal Credit, which initially launched in a single postcode area and scaled from there

In addition to location, the Ayushman Bharat Digital Mission restricted the types of data that the Unified Health Interface platform would accept. Rather than attempt to digitise everyone in one go, they started with common well known things like vaccination records, discharge summaries and wellness records.

The NHS should emulate this approach with the Single patient Record by starting at a small number of health facilities or postcode districts, then growing over time. It should also limit the number of initial data types supported, only adding more as they understand what is most valuable.

As this video shows, patients can create their accounts and link them to a hospital at the the point of care by scanning a QR code. Once they have done this, any records for that visit will be linked to their account and automatically available to them. As this video shows, clinicians can also request access to records from other hospitals at a consultation, with the patient granting consent. Patients are the glue that joins the system together.

This idea of a user building up the longitudinal record over time, visit by visit is very different to the one-time opt-in approach that the UK has tried in the past. Previous attempts have tried to get the public to opt-in once and for a wide range of use-cases. Even if the UK government eventually legislates to make all data available to patients by default (presumably removing some need for opt-in) these types of interaction are still valuable. They make it clear to people what is happening and why and provide a ‘service design moment’ in a user’s journey to hang other information from (for example sign posting next steps).

3. Large scale adoption does not require mandates (if you focus on the needs in the system)

The Unified Health Interface is optional for patients and healthcare providers. Instead of mandates, the Ayushman Bharat Digital Mission used a range of interventions to get adoption. These included creating a sandbox and developer documentation for software providers to test their integrations, codeveloping data standards, paying private providers per data point linked, and creating a certification scheme.

The avoidance of mandates is not uncommon for successful public sector digital platforms. In Estonia, the X-Road data exchange platform was not mandated, it was just the simplest way for delivery teams to comply with the one-only principles. In the UK, GOV.UK Notify scaled rapidly without any mandate because the team, did the work the pricing model and onboarding process worked best.

Mandates give scale, but to not guarantee outcomes.

Rather than starting with mandates, the Single Patient Record should start by understanding what it will take to make it simple for healthcare facilities and software providers to adopt.

4. Pragmatism is more important than the perfect data model

The data models used by the Ayushman Bharat Digital Mission allow for a mix of structured data and PDFs. While the ideal is structured data, it also allows for a ‘path of least resistance’ option of just attaching a PDF of a prescription or a discharge note. Over time the amount of structured data will, presumably, increase and the amount of PDFs will decrease - sometimes you have to take an incremental approach to digitisation.

This is the same approach that India’s government wallet app DigiLocker takes. Government ministries can issue digital credentials for things like licences and permits as structured data, if they have the capacity to do so, or as PDFs if they do not.

In approaching the Single Patient Record, the UK should allow for similarly permissive data structures. This would mean that a test result or information about an appointment would, in the best case scenario, appear in the NHS app as text and data. But PDFs of letters would be there as a fall back.

5. Simple open protocols, not monolithic systems

The Unified Health Interface and the ABHA app allow patients to book appointments with registered clinicians. It does this using simple open protocols, rather than monolithic platforms.

Previous attempts at common booking platforms in the UK have never got off the ground. For example, when the UK Ministry of Justice tried to create one, they “concluded a single‘booking platform’ wasn’t really something that could exist” because “the back-ends that handle these bookings are also already tightly integrated into other parts of the organization”. The NHS should not repeat that mistake and instead use a combination of open protocols and the NHS app to surface bookings to patients.

6. Digital public goods, not IT procurement

Common with India’s other big digital bets like Aadhaar and UPI, the National Health Authority uses the language of the language used is Digital Public Goods and Digital Public Infrastructure.

The concept of DPI is less common in the UK, but the concept is gaining ground. The recent Blueprint for Digital Government, for example, included the following when describing the future state of the UK public sector:

Public sector organisations will be connected, not fragmented. Modern digital public infrastructure will make public organisations more integrated

Digital Public Infrastructure is a radically different model from the outsourcing model the NHS is used to - focusing on public value creation, public values and internal capacity. With the Single patient Record, the NHS should embrace the principle and practices of Digital Public Infrastructure.