What can India learn from other Digital Health Records systems?
TL/DR
India’s digital health records system should enhance its signal-to-noise ratio, incorporate interoperability by design and improve user experience by learning from other such systems globally.
✨India’s journey towards Digital Health Records
There is a lot of excitement in India about digitization of healthcare. The Indian government, over the past few years, has released a series of vision documents and policies enlisting several goals towards digital health, electronic health records and other such innovations. Not stopping at such vision documents and policies, there are also interesting experiments being run including the Regulatory Sandbox started by the government’s National Health Authority for innovators to try out experiments on this technology stack. Here’s an overview of these various documents-
- National Health Stack (NHS): Released in 2018, this document proposed a technology architecture view of the India Health Stack.
- National Digital Health Blueprint (NDHB): In 2019, the government published NDHB that provided further insights into the proposed architecture, adding specific guidance on its implementation. It recommended a federated architecture, instead of a centralised architecture, to be adopted for the management of digital health data to ensure interoperability, technological flexibility and independence across the National Digital Health Ecosystem (“NDHE”).
- Health Data Management policy (HDMP): HDMP, released in 2020, provided granular details of the various components of the proposed NDHE. Here’s a schematic representation of this proposed architecture-
Several healthcare industry players including hospitals, pharma and device organizations and start-ups have welcomed these steps. However healthcare experts have also cautioned against some of these components-
- All of the currently discussed framework sits outside of formal legal structure; NDHM, NDHE and other guideline/frameworks are currently out of any formal legal framework. This of course opens up huge vulnerabilities for patient data privacy, access and usage.
- Basic definitions of ‘personal data’, ‘sensitive personal data’ etc. differ between different documents, There are many interchangeable and/or new terms introduced without details or are ‘forthcoming’; e.g. data retention policy, consequences of non-compliance.
- Finally, several experts have noted lack of focus on patient experiences as well as thinking about the existing challenges faced by healthcare providers including clinicians and their support staff.
With this background in mind, I have summarized three recent research articles examining various elements of digital health records that provide interesting viewpoints of such systems built in other parts of the world. We, in India, can definitely learn more from these and other such experiments.
💡Lesson 1: Let us think about Signal to Noise Ratio
Case in point: This commentary, published in Medical Decision Making, reviewed two articles addressing EHRs’ research potential; these two articles focused on using EHRs to develop prior transition probabilities for the calibration of a model of eating disorders in children and creating research data sets from the EHRs of 2 large health systems in the US.
What does it say?: The commentary nicely summarizes the benefits and shortcomings of use of EHR data, while comparing and contrasting them with data collected more systematically, including US National Health and Nutrition Examination Survey (NHANES) datasets and clinical trial datasets.
The advantages of EHR data are that they reflect the variety of patients in the real world, rather than the carefully selected groups used in trials, and the variety of practice patterns in the real world. The disadvantages come from the same circumstances: the data available differ with the timing, the decisions, and the measurement and recordkeeping practices of each patient/provider pair and the limitations of the particular EHR systems. The article cites several examples of these characteristics of EHR data.
What does that mean for India?:A key vision for India’s healthcare digitization efforts are towards empowering researchers and policymakers with aggregate, anonymized datasets for future research that guides healthcare delivery, quality and access. As Louise Russell puts it in the commentary above, we will need to build a system that promotes growth and evolution of EHR data usage by sharing our experiences with EHR data not just the results but also the problems and solutions, by asking questions such as- Where are the signals strongest? Where is the noise greatest? What approaches work best to reduce the noise and discover the signal? Which research questions can be addressed, and addressed well, with EHRs, and which cannot?
💡Lesson 2: Information sharing is a long game!
Case in point: In this article, published in JAMA Health Forum, authors Turbow, Hollberg, and Ali provide a quick historical perspective of US EHR ecosystem and then focus on how lack of interoperability has become entrenched due to poor system design and incentive structures.
What does it say?: As mentioned above the authors provide a nice short summary of US EHR regulations and resulting effects on its EHR ecosystem-
- 1996-the Health Insurance Portability and Accountability Act (HIPAA): elevated security and privacy of patient data, standardizing what patient information was to be protected and how this had to be achieved.
- 2009 -Health Information Technology for Economic and Clinical Health (HITECH) Act: incentivized EHR adoption , launches the term ‘meaningful use’.
- 2016- The 21st Century Cures Act,-established ways of collaboration between public and private organizations to improve the quantity and quality of information exchange.
- 2020, the ONC Cures Act defined conduct that would constitute information blocking, established technological standards for health IT developers, and identified the patient data that are required to be interoperable.
However, the authors point that patient data have been treated as a commodity owned by EHR vendors, potentially leading to reluctance to share data due to a desire to maintain vendor market share as well as increasing barriers for smaller companies to enter the interoperability space.
What does that mean for India?: Two specific policy goals provided by authors can make a lot of sense in the Indian context too-
- Identifying the key barriers to information exchange across varied health care settings, to understand the tools that health care systems and professionals need to effectively use the data, and to gauge the return on investment of these efforts.
- Providing market and economic incentives for EHR vendors to create user-friendly systems (and for hospitals to choose those EHRs), share data across platforms, and allow health systems to sustainably access and use the data in a meaningful way.
💡Lesson 3: Think about burden on users for all EHR interactions including data entry, retrieval and communication
Case in point: A fascinating article by researchers, including those at Harvard Business School and Stanford University, provides insights into which EHR activities clinicians spend their time doing, the EHR tools they use, the system messages they receive, and the amount of time they spend using the EHR after hours. An accompanying editorial provides further data points about system design and implication from other global examples.
What does it say?: The authors analyzed EHR usage patterns, using the same EHR software (Epic), across 371 health systems, including 348 from the US and 13 from the EU region. Their findings point out several important issues with EHR design and usage-
- US clinicians spent significant more time (90.2 minutes)actively
using the EHR per day compared with the time by non-US clinicians (59.1 minutes). - Most of this additional time was spent notes, while there was less difference in time spent for clinical review.
- In addition, US clinicians compared with non-US clinicians received more messages per day in total and from various sources
- Finally, US clinicians worked in the EHR for a longer time after hours per day than did non-US clinicians.
As the editorial summarizes these results, only the farthest outliers for time on EHR in the non-US sample, at the 99th percentile, spent the same amount of time as a US clinician at the median!
What does that mean for India: The editorial rightly points that US physicians work through more regulatory, compliance, and billing sludge each day than their non-US colleagues, spending more time composing boilerplate documentation designed for billing justification, compliance attestation,
and liability defense and subsequently reading their own and others’ bloated notes; it also cites other past work including one conference report stating- ‘The United States experience was contrasted with those of other nations, many of which have prioritized patient-care documentation rather than billing requirements and experienced high user satisfaction’. In India, we would be better off starting to think about these aspects from now on, rather than realizing them too late!
Summary
While India is still at the very beginning of its digital health records journey, policymakers and implementer both should work together to make sure that the building blocks of the system are being laid by incorporating the learnings from other such systems; this would make a lot of difference for meeting the vision and goals of the system by improving user experience, achieving higher data quality standards and enabling wider reach of the EHR systems.
