Digitization of Health Services - Part I
Decoding the Ayushman Bharat Digital Mission (ABDM)….
Conversation between Dr Akshay S Dinesh (ASD), Dr Amit Mishra (AM), Prof. Sundeep Sahay (SS) and Prof.T Sundararaman (TS)
ASD: There are a a lot of official documents that explain the objectives of the Ayushman Bharat Digital Mission (ABDM), but let me begin with the 2017 National Health Policy . To quote: “Recognising the integral role of technology […] in the healthcare delivery, a National Digital Health Authority (NDHA) will be set up to regulate, develop and deploy digital health across the continuum of care. […] The policy aims at an integrated health information system which serves the needs of all stakeholders and improves efficiency, transparency, and citizen experience. Delivery of better health outcomes in terms of access, quality, affordability, lowering of disease burden and efficient monitoring of health entitlements to citizens, is the goal…. Establishing federated national health information architecture, to roll-out and link systems across public and private health providers at State and national levels ….. will be supported by this policy.” The other key document is the National Health Authority’s “Brief Guide on Ayushman Bharat Digital Mission and its various Building Blocks, version 1.1” published December 2021.
In brief the idea is that there will be interlinking of systems, enabling people to exchange their records and data across multiple providers. And for this we need to have an ecosystem built through a central authority. The authority would prescribe common health data standards, developing core modules such as registry of health facilities, healthcare professionals, and this will allow interoperability and also accelerate digitization of healthcare institutions and dovetailing various resources. I don’t exactly understand what dovetailing various resources and digitization processes mean, but we can come to that.
So, my first question is whether this is a correct summary of what the Ayushman Bharat Digital Mission is all about? And if so, where has it reached?
SS: Thanks Akshay. I’ve been reading these documents and I think on the outset it seems extremely complex and maybe even a little utopian to try to build this architecture at such scale and diversity. I am interested to know what is the current status of this whole thing and what may be the metrics you’re using to assess the progress in this whole effort. The documents look very neat, much like a PhD student’s assignment paper on what should be a national architecture, but there should be a lot of messiness involved. So unless we know more of what exactly is being rolled out and its status, it is difficult to comment.
AM: So Akshay, what you have just summarized is fine, but there is a history behind it. This concept of a national agency to build a digital health architecture, can be traced back to an NHSRC study done in 2011-12 to analyse the status of integration of public health IT systems. And then when we worked on open data standards for healthcare domain, which was called metadata and data standards (MDDS), also, a part of the recommendation was to have an authority to build this digital architecture. This concept was then included into the draft national health policy,2015. But at this point, the digital architecture requirements of national health insurance also came in, and in the final policy as adopted that took the priority.
But the point that I want to make here is that there are two approaches to develop a national digital health architecture. One is insurance based and the other is public health system based. Though ABDM documents talk of both, what is implemented is more aligned to an insurance-based model where the focus is on exchange of health records for care coordination across public and private sector and for insurance-type reimbursement, which is not the same as creating a health information exchange for public health where both public and private providers share data.
TS: I agree with both assessments. As of now, the ABDM has very little to offer to the public health management information systems which continue to grow and proliferate without reference to ABDM. The current direction of digital architecture building under ABDM is to support health insurance initially and then later healthcare delivered through “integrated care management networks.”
So Sundeep, when you ask “What are the metrics on which they are assessing progress?” the reply is quick and ready.. and relates to the registration of facilities, health professionals, service-users and service delivery.. Just look at the official ABDM dashboard: ‘As of now the system has registered with verification 3.53 lakh facilities and 5.34 lakh health professionals and 45.76 lakh health records and 70 crore individuals with an ABHA ID. And this data is available by state and by multiple categories within. The ABDM has also created 106 “microsites” in many cities, across 17 states. where the registration will achieve saturation coverage and ability to exchange health records between facilities and providers would be established. These are in themselves very impressive figures. Another very positive feature of the ABDM, is that the design, as stated, does not insist for all providers and care managers to use only the software applications provided by the authority. The authority will approve any applications for compatibility with its standards [Metadata and Data Standards (MDDS) and Electronic Health Records (EHR)] and make them available in the cloud. But that is easier said than done and whether both these standards are able to ensure interoperability is an open question. But having said that, I think they are going ahead and evolving their own norms for approving interoperability in what they have called the “M1, M2 and M3 levels of a sandbox”
SS: Though they have written in many places in the document that at the local level, people can use whatever software they want to use, practically that’s not the case.
The centre is insisting for states to use only earmarked systems coming from the centre and not used other software. HISP India has been working in Himachal on DHIS 2 compatible, Hospital Management Information Systems based on Open MRS platform for the last 15 years and they abruptly pushed us out because now they say you should only use the CDAC system. The issue is not only about the software being dismissed, but long-term efforts of relationship and solidarity building between the implementing agency and the state hospital staff, which was the foundation to build capacity and sustainability, also being discarded. Technical aspects of software can always be upgraded, but not such social capital. This is an old problem. The centre ought to insist only on the data it needs, and not on the software states use. But it always does the latter. Despite a welcome recognition of this issue in the ABDM documents, the problem persists.
ASD: Can you elaborate on how that can happen? Is it based on the idea that your software is not compliant with certain standards that they are prescribing? I know they are prescribing global standards like FHIR and SNOMED and LOINC and all of those. Are they putting extra criteria on top of that as to who can participate in the ABDM?
SS: No, I think our experience in Himachal was that they just asked us to pack up and go. As the State said they are now mandated to use CDAC’s software. It’s quite unclear whether the CDAC or NIC software are compatible with all those different standards because it’s not really open source in that sense, where the source code can be put to public scrutiny. NIC never allows anyone to look at their source code. And anyway, I do not think their software is compatible. And this M1, M2, M3, I don’t think it holds for NIC or CDAC or even the very many public information systems like IHIP which are being rolled out to the whole country (IHIP refers to the Integrated Health Information Platform which is the national information system for disease surveillance).
AM: And as far I know there are no standards implemented on the public health management information side at all. There is a general call to be compliant with SNOMED which no one really understands what it means.
TS: Many states’ publicly funded insurance schemes have their own digital systems which have matured over 10 to 15 years, and with which the users and programme managers are familiar. Many such schemes are under pressure to close down their systems and instead start using only the NHA-mandated central system. This seems in clear violation of their own guidelines. Perhaps at a technical level the central systems being developed are not designed for interoperability, and therefore this short-cut of closing down and replacing. That’s a shortcut for NHA, but an immense increase in work burden and decrease in operational efficiency for everyone else.
ASD: Let us take this facility registration component where many states have shown progress. Amit, you were involved in a facility registration earlier. Is it the same approach or is this different and if it’s so, why?
AM: ABDM is working on a different solution, from what we were working on in the 2014-17 period. That was called the NIN or National Identification Number for all public health facilities. Creation of a facility registry was a part of follow-up to the creation of the MDDS. We took facilities as they appeared in two different health management information systems (HMIS and Mother and Child Tracking System, MCTS) and mapped these to each other to identify the unique facilities in both the systems, then put these on a portal to get these verified by states and district-level users. As part of this process, a lot of facilities were eliminated, and others were added and we arrived at about 200,000 facilities each with its NIN. But unfortunately, that data was never put to use. I think during COVID, there was some effort made, but it did not last. That system is completely dysfunctional now and ABDM begins all over again without limited or no reference to the earlier efforts. The problem with the ABDM approach is that it is voluntary reporting – and so some states report tens of thousands and others much less. Then there is a stage of verification that lags far behind. A recently published paper says that only 21% of the facilities were verified by assigned central or stage agencies. Further, this system is not dynamic and fails to capture major changes to the attributes of each facility over time. We observed that while the name and the ID remain constant, facilities are continuously being upgraded and some even shutting down – and there is no mechanism to record these.
Let me also add that in addition to the above two we also have a facility registry under the Clinical Establishment Act database, the Insurance Regulatory and Development Authority of India (IRDAI) runs a private hospital registry called ROHINI, and there is another health resource repository which was a mega project being done by Bill & Melinda Gates Foundation (BMGF) in partnership with the ISRO to locate all the facilities and resources in the country. Perhaps that too is discontinued. And as we know it there is very limited or no interface between any of these systems.
SS: With reference to attributes of public facilities, was it referring to IPHS? And if you allow voluntary sort of registration from facilities, isn’t there the danger that this multiplicity of nomenclature would get more expanded (than standardized) without adequate verification?
AM: I think that was one of the attributes, but I am not sure. When I was working with the public sector facilities, we also realized that this facility nomenclature is a real problem because you have a variety of facilities available in the country and putting them into a standard structure is extremely difficult. From their nomenclature it was difficult to identify whether this is a primary care facility, outpatient unit, or a dispensary etc. So there has to be some level of standardization required in that space which did not happen. The challenges in categorising private health facilities are much more and we are still not aware of these. I think once you have done, say, enrollment of a thousand facilities, you have to go back to the classification system, the attributes and review it to find a solution to the problems of making the database dynamic. I don’t think that exercise has been done till now.
ASD: Let us now shift from facility registration to patient registration or the ABHA number. One of the main parameters of progress is ABHA enrolment or registration. A unique patient ID is seen as essential for insurance transactions. But is it adding value? Closely related is the pressure to increase ABHA registration. This pressure is reaching absurd proportions! NMC, it seems, has told all medical colleges that all patients should have ABHA registration, otherwise their clinical numbers will not be counted and as per NMC rules, a certain minimum patient attendance is mandatory for recognition. I heard privately that in some medical colleges, students are being persuaded to provide their own details in lieu of patients’ ABHA so as to show a higher achievement on this parameter. That is only an hearsay example and it may be an outlier. The general point is that such pressures lead to entry of poor quality and unverified data. And does the end justify these means?
TS: The interface with publicly funded insurance varies across states. All state insurance systems have their own unique number which is not synonymous with ABHA number. Some states continue with their own unique IDs, but add on an ABHA number. But this is patchy and not everyone with an insurance number gets an additional ABHA number. Others like Tamil Nadu limit giving ABHA numbers to those who are not covered by the CMCHIS which is about 30 percent of the population. Some states have made an ABHA number part of out-patient registration and can deny or delay treatment if this is not made. Getting ABHA registration requires the Aadhar card and there is a significant part of the population who do not have the card or due to quasi-technical glitches cannot use it. On paper of course the Aadhar card is voluntary, but in the implementation it’s mandatory. Despite all these problems there is progress on coverage with ABHA!
The real problem is that such coverage is not linked to any entitlement. On paper ABHA is best used to retrieve one’s health records. But adequate health records have not started up, and often there is no record to be retrieved. Some programmes, notably the Nikshay scheme for TB patients are better linked. It earlier required an Aadhaar card, and now it is an ABHA card – which is just an additional requirement since ABHA requires another identity card and in practice Aadhar is often insisted on. In effect therefore ABHA, as it stands now, works as a citizenship penalty. It doesn’t give you any entitlement. There is ongoing effort to use ABHA to facilitate out-patient appointments, but having a unique ID is not the main determinant of access to the out-patient care. Access requires organizational improvements on the supply side and different forms of support on the demand side. When it comes to healthcare sickness is the functional ID for a person seeking care. Further unlike in other welfare schemes a unique ID cannot be justified on grounds of preventing irrational use or wastage. Given this lack of immediate benefit, the pressures on enhancing ABHA registration from the top are needless and will tend to b counterproductive, as poor-quality data will be fed in by lower levels to avoid reprimand and punishment. This dynamic is also seen with other initiatives where targets are set to enforce total coverage.
I would say that we have this obsession with ID systems and unique IDs. Amit listed four independent efforts at facility registration. I think we would be able to enumerate more than eight or nine for unique IDs. While giving an ABHA number to every individual is progressing remarkably well, the larger question is “So what?” Or are we asking that question too early? Should we give it more time to see whether it will work?
ASD: So, the question is, will the ABDM reach there? I mean, could we just say, “okay, some things take time and maybe five years down the line this will be a success once every one – patient, provider and facility – is registered, it will work! There may be gaps, but we should not throw out the baby with the bathwater!” Is that a meaningful stance or should we recommend to the government to think about the barriers in some other way? What do you suggest?
SS: I think we have to examine the purpose of ABHA from multiple perspectives. From the government side its processes are currently an end in itself and success is based on how many people have registered etc. But if we take it from the patient perspective, is it easing their access to health services? I don’t think so, because I think it’s just adding a huge level of bureaucracy. Now in Himachal, after working more than 10 years in hospital information systems and many lakhs of patients being registered, we found the most common ID which was being used to register was the telephone number. Because people had it, they remembered it, it was with them always. So I think the criteria for saying it is working or not, should relate to making it easier for the patient to access services, which I don’t think ABHA is doing. Amit, I think you want to add something?
AM: It is difficult to say. As of now there is no use case for ABHA. And one of the reasons that they have not been able to put it to use is that the insurance exchange platform is not yet ready. And we don’t know when it will be ready because they are saying that this is in the final stages and testing is going on and an agency is building it, but it is not there. So probably in a year or so, we are told the exchange would be ready and we would be able to exchange records. But once records are exchanged, then we would be able to understand whether the ABHA and the ABDM design has real value and whether or not it is solving any problem. Until any record based on ABHA is exchanged, there is no value as such. From public systems’ point of view, I would suggest that if we are able to post a record from Nikshay (the TB control portal) to the RCH portal or the other way using ABHA, it would be a step forward. But I think that use case has not been considered yet. It is also uncertain whether the state publicly funded health insurance programmes (PFHI) are using or requiring the ABHA number. The transaction portals that are in place for PFHIs, enable insurance claims and reimbursement, and these use formats and scanned documents that enable a third-party agency to review the insurance claim and then to make the payment. Each of these have their own unique patient ID numbers and facility identifiers. We also know that 100+ microsites were created to have saturation coverage of facility and provider registration and enable health record exchange, including for private sector facilities of the microsite. But only 2 or 3 have taken off with record exchange, and even here we are unsure of outcomes. So, it’s still in the initial phases of development.
ASD: Even if we assume ABHA is not the solution, could we go ahead with other existing IDs like the mobile phone number and move on with the other components of digitization like digital records creation, human resources and infrastructure?
AM: Let me clarify. I don’t think with mobile numbers we can exchange records. To exchange records and to build a registry you need to have a proper identification system. Many countries are not able to do it only because they don’t have a unique identification system. So ABHA in that sense, from a technology point of view is in the right direction but the problem is in the infrastructure and then the use case that they need, they have not been able to build. But without a standard ID, you cannot exchange your records.
TS: Amit, I want to question this wisdom. I was visiting a PPP model in rural Meghalaya accompanied by a senior management consultant. Despite good service delivery, they were not using any unique ID. So the consultant with me said, “but how do you manage to identify people?” Now the provider, thinking this was a trick question, scratched his head a little bit and then ventured “Name. I identify him by the name”. I think if we use the name, age and sex and then any one ID number, that should be fit for purpose. In this transaction, neither the patient nor provider is hiding his identity, nor are there any problems due to duplication – so why this need for creation of yet another ID. The problems with access and continuity of care lie in the organization of services. Patient identification is not a barrier. Digitization has, at best, a minor supplementary contribution to make
AM. I do not quite agree. You are right in saying that for searching a patient in the database of a facility or even a small state like Meghalaya, any ID should suffice. But for the purpose of transfer of the data or exchange of record from one place to another place, across different districts, states and different IT systems, or for building a longitudinal record of a patient, you would require a unique identifier. If there is a limited variability in situations, any ID would work. But when the patient arrives into the clinics and hospitals, they come with different backgrounds and different situations and in these cases a locally decided permutation combination of identifiers doesn’t work.
TS: I would submit that the requirements of such portability are small, and for most service users, the record is with one provider and one care pathway. At any rate it is a limited set of records. The number of healthcare transactions that will need a more sophisticated portability arrangement are a small proportion of the total healthcare transactions within a health system. For handling this minority of healthcare transactions, we can make use of available identifiers. We have address details of the patient. We have ration card numbers, and Aadhar numbers and other health insurance numbers. We even have fingerprints and iris scans on record. If a patient says that he visited a certain hospital which has a digital record and is willing to transfer the record the lack of an ABHA ID will not come in the way. But if the hospital is not willing to transfer it, then ABHA is of little help.
To illustrate: If someone walks into AIIMS outpatient clinic, from a primary care centre in a state, the real problem is not identity of the patient, but whether the primary care centre was maintaining a health record, and whether it is willing to share it, the process of authorizing and implementing the transfer and later receiving feedback. The non-digital institutional barriers of record maintenance and transfer overwhelm the digital dimensions. The digital dimensions can be coped with the name and any identifier, but the non-digital dimensions need institutional changes including standard protocols for reference and care pathways etc. But I’ll go with your final conclusion that we need to build some use cases at this level of a district. But the other consensus we have is that ABDM is not playing any role currently or in the foreseeable future in the process of digitizing the health management information systems. Do we agree on these?
AM: I agree. And one point that I wanted to add to what you said is for persons visiting from primary care to tertiary centre, we don’t have set protocols for transfer of the patients or for referrals. And many patients do not follow the optimal pathway and are searching for the appropriate provider. While digital pathways and exchange of data is important the immediate challenge is to create the human bridge. IT is a tool – it must follow the norms we create.
SS: In our Himachal experience many patients bypass the pathways altogether and arrive straight at the tertiary hospitals. The proportion requiring transfer of records is very small. One of my students just completed her doctorate on HIV care seekers. There is not only fragmentation of the care process, many patients due to social stigma are giving anonymized names, so as not to be identified too widely. I find the TB digitization system is more user-friendly for enabling follow up and continuity of care. So, I think, our whole perspective must be clear on who we are trying to support through these systems. I don’t think these systems are making it easier for the patients to access care. I think right now it is very much the perceived information needs of bureaucracy and the state and not the patients’.
ASD: Well. That begs the question, if it is not the immediate concerns of patients, what is the administrators’ rationale for this current ABDM design?
TS: Let me submit to you that the choices made for the ABDM design are absolute requirements only if you are expecting a third party to transfer a substantial number of patient records from one provider to another provider with or without consent of the clients as would happen if one insurance company or health management organization (HMO) and their clients are purchased by another HMO, or if government is outsourcing care of a population with ABHA cards to such an agency. For a sick patient seeking health care, an unique ID is not essential. But for such a transfer or records across providers it is. For over 15 years now a model called “strategic purchasing of care from integrated care networks” has been advocated by many “policy entrepreneurs” and a few policy committees. Such policy entrepreneurs are often, but not necessarily, associated with international consultancies and corporate influence. Often this “integrated care network” is projected as the solution to the problem of fragmentation due to current insurance schemes which do not cover outpatient care and primary health care. In such a model, a network of providers, and facilities, private and public are organized by a corporate-led Health Management Organization (HMO) to provide free care for a given number of clients, who could be identified by their ABHA numbers. In this model, the government directly or through an insurance company, purchases care from these integrated care networks, using an appropriate process of tendering and contracting. Every element of the ABDM as designed now would make sense for such a model. There is however no use case available and the expectation that such a model can be built is to my mind rather ambitious. But there are many efforts in this direction- and who is to say it will not succeed. There are some who would argue that such HMO/integrated care networks need not be under corporate ownership and these could be government owned, but to my mind that is rather disingenuous.
ASD: I think we should address civil society concerns on privacy and confidentiality issues. Again, I see the ABDM 2021 document says all that needs to be said. It provides for a consent manager, it has the facility as the provider of the records, it has the consent of the patient required for the transfer of information. But is that, as an architecture, adequate? Will it be operationalized? What are your comments?
SS: In general, public concerns for privacy are minimal and I see people coming into the hospitals, leaving their documents around and documents thrown around on the table. The exception is HIV and diseases associated with social stigma. I’m not sure whether a formal consent process would work as trust in the government is quite low. I think where privacy is a concern people will use their own local means of being anonymized to the system by giving different names or going to different clinics for different treatments etc.
AM: There is a technical problem with many consent mechanisms. Often these are based on receiving OTP on your mobile and you give the OTP to the provider and the provider uses that OTP. In a hospital and sick patient context along with network and connectivity challenges this transaction could be very difficult. It could interfere in the care and in the creation of the record. So there has to be some alternative mechanism that has to be thought of in the Indian context for addressing the privacy issue.
TS: What we have seen in this regard are operational guidelines which at best could reflect policy. But for ensuring privacy, these guidelines need to be supported by law. And legislation on privacy and confidentiality for this context is non-existent. Without legislation there is a poor chance of implementation. I think many of these systems are designed to make the patient records transparent to insurance agencies and HMOs- but these would not ensure that the decisions of insurance agencies are transparent to service users in defense of their entitlements. So there is need for much more caution and we need to see how this plays out in practice.
TS: Before we close, Akshay, what are your comments on this because you have thought about this issue?
ASD: Yeah, it’s a bit confusing to me at the moment because if they enforce ABDM, for example, for HIV care, or enforce ABHA as the unique ID, then how do service users even skip the system? Would they have to go out of the system altogether? And now that raises its own bag of worms. And then if you also talk about agency, is it okay to mandate ABHA for a healthcare entitlement? It’s not just about confidentiality and privacy, but about the agency of an individual to choose whether to link their record or not. There are even more basic questions here.
But to sum up our discussion so far:
- The National Health Policy 2017 called for a digital health architecture for two purposes – one was for support to exchange of health records across providers and the other was for public health information needs. ABDM prioritizes the former and, as it stands now, little to contribute to the latter.
- The ABDM architecture is ambitious and it remains to be seen as to whether it is feasible. Its full utilization seems designed to require a HMO style re-organization of the health sector much advocated by market-theory influenced policy entrepreneurs. But for institutional and political reasons such re-organization is even more challenging than its digitization, and even less feasible and desirable. It is not even clear whether such re-organization is on any official policy agenda. It is definitely not on the Ministry of Health’s agenda.
- There are design issues that plague the facility and provider registers that are under development. These need to be designed as dynamic registers and as inter-operable with other parallel registers.
- Given all the above problems and the lack of any tangible perceivable benefit from an ABHA number, the ABHA should not be made mandatory. Further the government must ensure, through both policy and legal tools, that there is no delay or denial of care due to ABHA concerns.
- There should be no pressure to close down existing digital systems of creation and exchange of health records and health management information systems that are in use by hospitals, state public health departments and state government led publicly funded health insurance programmes, and their replacement by a centrally designed system. ABDM can however create an exchange platform with clear connectivity links to which existing systems can become inter-operable. Financing to enable such interoperability has enabled such linkage in most countries.
- Reiterating that digitization of health records has considerable potential benefits for continuity of care, could ABDM explicitly focus on this objective and build a few clearly functioning models of record creation and exchange, using its current approach at some 10 to 15 district sites across the country- before it scales up?. Further in these districts, the exchange of health records need not be linked to insurance- and could also serve two-way referrals along care pathways outside the insurance framework. At least some of these models should be based on providing interoperability with existing hmis system Culling out public health relevant information from these health records should be technically possible, but for the current actors, who have limited cognition and/or priority for public health needs, that may be an impossible “ask”.
- Given that the whole push to digitization and to integrated care network models are justified in the name of efficiency, we need to have good cost-benefit and cost-effectiveness studies to assess these options. A huge amount of valuable provider time and management efforts are being sunk nationwide into the current compulsions of the ABDM roll-out. But is the expense and effort cost-effective? Could we have achieved digitized continuity of care arrangements at less costs and efforts. We think so. In this context the ABDM should also sanction a few district wide continuity of care alternatives which take a more incremental, participatory and pluralistic approach to building a digital plus institutional architecture required for continuity of care – and that could act as comparators and learning sites.
But now let us move to public health information systems. I notice that with a possible exception with Nikshay, the consensus in this panel is that though ABDM arose in part as part of the need for interoperability and reliability of public health information systems, as currently designed, it does nothing for those concerns. In our next part II of this conversation, we will be discussing the challenges that public health management information systems face and the possible ways to address these, without reference to the ABDM and its priorities.
Acknowledgements: Our thanks to Ms Roubitha David, for her assistance in transcription and editorial coordination. And to a number of others who shared their experience with ABDM. We note that as the programme is evolving at a different pace in different centres, some of the information here may be outdated, or more likely incomplete. So hoping that this conversation continues…
About the Participants:
Dr Amit Mishra is a healthcare professional with a master’s degree in health administration from TISS, Mumbai, and a diploma in health and social care data sciences from the University of Edinburgh. For over the past 15 years, he has been associated with the design, development, and implementation of health information systems. He was part of the HMIS reforms team established within NHSRC which supported the design and implementation of HMIS under the National Rural Health Mission in its first few years. Later, he worked with the Ministry of Health & Family Welfare to implement several digital health projects. He was the nodal resource person from NHSRC who served on the core team that led the development of metadata standards for integrating public health information systems in India. He also led the implementation of a project to build a registry of public health facilities in India called NIN. Currently, he is working with the Ministry of Public Health in Qatar to build data systems for non-communicable diseases.
Dr Akshay S Dinesh (MBBS, FHM, PGDMLE) is a generalist straddling public health and technology for a progressive society. Akshay is the co-founder of Action for Equity and is also associated with JeevaRaksha as an emergency medicine trainer, with IPH Bengaluru as an honorary associate, with SOCHARA as a digital archivist, with DemTech.ai as the head of engineering, with Nivarana as the Technical Editor, with Bahutva Karnataka, Sarvatrika Arogya Andolana – Karnataka, and free software, free knowledge, open data communities in India as a volunteer and contributor. You can read more about Akshay here.
Sundeep Sahay is a professor in Informatics at University of Oslo, and has affiliate positions in the Faculty of Medicine in Oslo and Sheffield University, UK. Over the last 25 years, he has been engaged with research, policy and practice across multiple countries in Afirca and Asia, with a strong focus on the public health system in India. He has also been engaged in various policy engagements with WHO, ADB, UNICEF and other. He has been one of the pioneers in establishing the R&D network HISP (Health Information Systems Programme) (dhis2.org) in 2000 at the Department of Informatics, and also set up and nurtured HISP India, an NGO, over more than 20 years in an honorary capacity. Between 2007 and 2009, he was seconded to NHSRC, Delhi, where he led the national Health Management Information System reform process under NRHM. It was in this period, and with a significant contribution from the NHSRC team that the District Health Information System-2 (DHIS2). Sundeep led this effort. To fully appreciate his contribution, one needs to understand DHIS-2 better. DHIS 2 (dhis2.org) is a free and open-source digital platforms for HMIS which supports the decentralized use of information and provides the flexibility to be customized and adapted to multiple use cases with limited core computer science expertise. DHIS2 was first born in Kerala India in 2005, involving collaborative development between developers from HISP India and University of Oslo, Norway students in 2005, and was evolved and taken into states of Kerala, Gujarat, Jharkhand and Madhya Pradesh. The development of this platform was subsequently taken up by a development team at the University of Oslo who have further evolved it into a robust global public good which has been adopted by 80+ countries and also by WHO in the framework of a Collaborative Centre agreement with Oslo. Unfortunately, even as DHIS2 has been adopted globally, it gradually lost its roots in India. And much of this was due to the way central systems replaced more decentralised, user-friendly state hmis systems.
Sundeep Sahay is one of the leading scholars in the research fields of Information Systems and ICT for Development, with nearly 300 peer-reviewed publications and has been the primary supervisor for 30+ doctoral candidates. His current research focus is on Antimicrobial Resistance, Environmental AMR and Climate and Health.
Thanks.
should have referred to the national health stack,its evolution the digital health blue print which formed the basis and give digital architecture , building blocks for NDHM. As well as health systems for new india- building blocks 2018
Thank you. I found this very useful to understand what’s actually happening on the ground
Two things seem to jump out for me: 1.Digital & non- digital info systems have to integrate seamlessly – for this we have to go use-case by use-case ( rather than from a overarching architecture, down to individual use-cases)
2. Capacity building is key – at all levels, including the very top in health departments
Thank you,
Neethi
You should have referred to the national health stack,its evolution the digital health blue print which formed the building blocks for NDHM
Reply: Thanks for pointing this out. It is certainly useful to be reminded that, just like the public health informatics had an evolution, the current focus of the NDHM on EHR and record transfers has also evolved over a period- and this development of concepts included the papers on national health stack, and the digital health blueprint> Point noted and thanks again,
Sundar
This conversation unravels the enormous challenges in monitoring access to health care.
We might at least focus and ensure accuracy in measurement of health outcomes at the population level – much simpler at least in terms of cause-specific mortality – without which, gauging the value of health system inputs such as access will remain a subject of conjecture..
Thanks. This exposes the disheartening mess the digitalization initiative is in. It is certainly underfunded. It also hopes to ride on the existing system of healthcare which is incoherent. Simply mandating something as simple as an eprescription could have been easier to implement at scale and across states. We will also need the innovation system to be coordinated keeping in mind future UHC architectures rather than reacting to current reality. These are our thoughts shared in the book, esp the concept of Health Innovation Parks. We also need a health council on the lines of the GST council. Politics is shredding any attempt to create a solid sustainable architecture. The sooner we face up to this fact and fix it, the better it will be for the creation of a NextGen healthcare system that serves the patient well.
Read this very interesting discussion.
Being unaware of these developments, I have a few questions:
1. How do you ensure patient confidentiality?
2. How much digitisation of patient records has happened?
3. Are the I T systems compatible?
4. What about the private sector corporate hospitals, smaller clinics etc… many of them have their own IT systems.
5. Would it not have been better to have a full architecture in place in a few sites and execute piece by piece over a period of time instead of a full scale roll out at one go
Thanks
Reply: Thanks for the queries. I attempt a response:
A 1. There is a workable plan for privacy by a in place. However, what we point out is that without support from legislation, there is very little likelihood of implementation. Confidentiality is a related concern- but it is distinct from privacy. I do not find any plan for confidentiality as distinct from privacy. But I
A.2 Many health encounters are recorded. However These could be called health records. However in the form of longitudinal record of many health encounters from a number of different providers and available for exchange- use cases seem to be few.
A.3. We do not think that the major IT systems in operation are compatible. There are plans to address this, but the effort is sub-threshold. It may happen though. The current approach is to get more users to use a CDAC or NIC developed platform.
A4. There is a huge ongoing effort to register private hospitals of all sorts, especially in the micro-sites. But this is registration. But are these hospitals finding any value addition from this- has recording work become easier, or are they able to obtain records of patient care from other hospitals etc etc? ? To the extent we know this is not taking place- at least not on a significant scale. And most such hospital IT systems are not interoperable with other systems, even ABDM sponsored.
A.5. That was our suggestion too- that It should have been a simple matter to construct a proof of concept in a few districts and also some comparators- alternatives, estimate costs and benefits and then proceed.
From:
Quite interesting Sir,
Your point of saying, for the exchange of few records, we are trying to burden the entire Health system is the crux of the problem. Also-if hospital and patient is willing , we can share the records across the world through mail in seconds
Thanks for sharing
Excellent discussion.
Hope there will be some light at the end of the tunnel..if those at the helm of policymaking read and understand these issues,