Digitizing medical prescriptions and ensuring data ownership rests with the patient could open up a large market spanning healthcare decision support, wellness and insurance industries in India. There is an opportunity for entrepreneurs to use IndiaStack to activate this market.

According to data from Medical Council of India, there are close to 8,00,000 active doctors in India. Assuming 80% of them are in urban areas, that is about 6,40,000 doctors available in urban areas, where 28% of India’s population resides. A key artifact of a doctor-patient interaction is the medical prescription. An estimated 3 billion prescriptions are written by doctors across urban India every year. And almost all of them are on paper! A typical medical prescription contains doctor and patient identifiers, diagnosis, tests required, medication, dosage and other details. It is in an unstructured format and is mostly written in individual handwriting styles on letterheads. Digital Prescriptions or Electronic Health Records (EHRs) aren’t really mainstream in India. What does it take to create a digital prescription ecosystem and what demands can be unearthed?

To answer that, let us analyze the workflow and the decision making that goes through in a doctor-patient interaction. Let us take an example of a patient Shyam. Shyam is a junior marketing professional with a technology firm (his first job after finishing his MBA), is in his early 20s and has been experiencing chest pain for the last few weeks. He maintains an active lifestyle and is a fitness enthusiast. Shyam visits a nearby general physician to understand the possible cause of this pain and discomfort. Sure enough, he has spoken about this to his friends and family, and searched for possible causes by doing a generic Google search. And that has only made him more anxious! And now, he visits a doctor.

As with a typical doctor-patient interaction, doctor asks for patient’s symptoms, medical history, type of pain, triggers, among several other factors in order to better ascertain the cause of the pain. And the doctor has narrowed down to 2 or 3 possible causes and wants Shyam to get a series of tests. Doctor prescribes the tests on a medical prescription document. Shyam visits a diagnostic center nearby, gets the tests done and gets medical diagnostic reports back to the doctor. Doctor goes through the reports, and is able to identify the cause and provides advice, medication and dosage to help address the issue and sets a next visit date to check on the progress.

While this is an over-simplified version of a doctor-patient interaction workflow, let us look at the scenario from Shyam’s point of view.

Shyam is anxious about the possible causes and for good reason. He does have an insurance cover but is uncertain of the coverage or terms of coverage. He needs to know how much these drugs and tests cost so that he can plan for short-term liquidity. He might have to call a friend for help. He has to plan to take time-off from his office and inform his boss. He has to respond to emergency work calls. He has to postpone any of his existing commitments to get to the root of this diagnosis. Post the diagnosis and commencement of treatment, he has questions on any lifestyle adjustments he has to make. What should he eat or avoid eating? What activities should he indulge in and what not to? What is the impact of accidental lapses in taking the medication? It is a stressful situation to be in till he gets back to normalcy. And the stress is caused due to lack of information at key decision points in this journey. He wants to access help in some of these decision points, or maybe even all of these decision points. But, Shyam wants to stay in control of his data and share specific data points with information or service providers on a need to know basis.

Now, you are an entrepreneur and see this problem statement from Shyam’s point of view. What system can you build that can help Shyam? Could it be that the most common artifact of a doctor-patient interaction, a medical prescription, can help unleash a portfolio of services to help Shyam in the various decisions that he has to take? And could it be done in a scalable way using digital prescriptions? And through all of this, can Shyam stay in absolute control of his own data?

Digital Prescriptions as a concept have been around for quite some time and have taken various avatars. They have been met with varying degrees of success in developed economies. The idea was and is simple at a concept level. Equip doctors and clinics with computers or tablets. Doctors would type in prescriptions into these devices instead of writing on a letterhead. This provides clean digital records that can be accessed anytime, and maybe even reduce prescription errors, if the doctor is trained to enter information in a structured format!

In the Indian context, there are quite a few challenges to this idea. First, access to a computer in clinics is not a common sight. Second, there is a cost to changing the behavior of a doctor from writing to typing. Third, doctors don’t yet see a tangible benefit for themselves in making this hard change from writing paper prescriptions to typing digital prescriptions. However, if somehow prescriptions were digitized and were always available, the following scenarios depict the benefits to the doctor:

1. 1. Scenario 1: Patient visits the same doctor twice in a row: In this scenario, doctor pulls up patient’s past records from his/her digital library of issued prescriptions by searching for patient identifier. This is how hospital systems run and the database index is usually a patient ID given to each patient. Patient has to remember this ID or can retrieve this ID using mobile number in some establishments.
   2. Scenario 2: Patient visits 2 doctors sequentially for a consultation. This could happen due to multiple reasons. Maybe, the patient wants a second opinion and chooses to visit 2 doctors. Maybe, Doctor 1 is a general physician and Doctor 1 refers patient to a specialist, Doctor 2. In each of these cases, the base workflow is the same. Patient visits Doctor 1. Doctor 1 writes a digital prescription to a patient. Doctor 1 owns the record. Patient visits Doctor 2. Doctor 2 accesses medical record of patient from Doctor 1, and is thus better equipped with patient history to make a more accurate diagnosis.

While Scenario 1 might still be applicable for a handful of large hospitals, it is largely irrelevant to clinics in India. Private practitioners running their clinics are rarely in the habit of assigning unique ID numbers to patients they see. They do however identify and remember details of their repeat patients with a remarkable precision. As always, this extraordinary human ability diminishes at scale.

And the interoperability depicted in Scenario 2 is virtually non-existent within most hospitals. And is absolutely absent across 2 different clinics!

Meanwhile, there are other digital prescription innovations that came into the market. One such innovation was the idea of digital pen. Doctors use the digital pen to write prescriptions on their regular letterheads, and the pen would transmit the written text via Bluetooth or other similar technology to a nearby computer or smartphone. It is a great leap from asking doctors to type on a computer, but it still has its challenges. The pen is a combination of hardware and software and requires maintenance. Any additional overhead to a doctor is at the mercy of a single break in user experience. So, we will have to wait and watch how this innovation shapes up.

The focus of both the above ideas has been on amending doctor behavior in some way or the other to enable digitization of medical prescriptions and making the doctor the primary owner of a medical prescription. And the data primarily resides with the doctor or the healthcare establishment. But, what if we flipped the responsibility of digitizing a prescription from a doctor to a patient. What if a patient could own a digital health record, choose whom to share it with and whom not to, and even get behavioural, dosage and financial decision support information, all arising out of information sourced from a digital medical prescription.

Now, you are an entrepreneur and want to learn the possibilities and the markets it opens up for a startup. Let us look at a possible scenario through the eyes of Shyam. Once Shyam visits his doctor, he scans the medical prescription through a mobile app powered by a startup, which takes advantage of some of the key IndiaStack APIs. Upon scanning, the following set of activities can be triggered to help Shyam with various decision points that we outlined earlier:

1. 1. 1. Shyam’s prescription is uploaded onto his DigiLocker account and he digitally signs by using his digital signature registered with a eSign Service Provider (ESP) (via the eSign API specifications) in order to self-attest the original prescription
      2. Startup requests Shyam for access to the prescription using Digilocker Requestor API
      3. Shyam provides consent to the startup in order to access his documents guided by the specifications in the Electronic Consent Framework. Shyam can also choose what programs are relevant to him and what are not relevant
      4. A handwriting recognition program identifies doctor’s handwriting with a precision that increases with use (using Machine Learning methodologies) and parses the medical prescription into structured data fields. So, now the medical tests recommended as well as the drugs and dosages recommended are in legible typed characters. The startup could provide a hybrid service that combines human as well as machine intelligence to identify a medical prescription’s content
      5. A cost estimation program within the app environment now uses this information to fetch the estimated cost of getting the diagnostics done and buying the medicines. Shyam has the estimated costs handy before buying. And this can be linked to registered online pharmacies who can deliver prescription drugs. Shyam can pay the online pharmacy via his UPI account
      6. An insurance coverage check program can ascertain insurance coverage of the upcoming set of procedures and purchases and provide (even indicative!) information to Shyam to make any key decisions in this regard. This can also help insurance companies by reducing their claim processing timelines since Shyam is most likely to claim only when he knows that his plan covers these expenses (Startup can build a revenue stream from insurance companies)
      7. Shyam requests the app to check for short-term credit options and terms based on Shyam’s credit scores, which could optionally be made available to the app (Startup can build a revenue stream from lending companies)
      8. A lifestyle management program can pull out information from a central database that identifies common dos and don’ts with the condition Shyam has been diagnosed with. (Optionally, it could also be linked to other wearable devices if there is relevance!). This advice can be securely loaded back onto Shyam’s Digilocker account

The central points here are two-fold — Shyam owns the digitized data, not the doctor or the hospital or the insurance company. Shyam can choose to give or not give consent for specific services and can give consent only for specific time periods. Shyam is thus in control of the situation and is not obliged to take on a bundle of services forever, when all he wants is one specific service for a limited period. And Shyam can get all the decision support with a few taps on his smartphone screen.

How would you go about building a startup that provides this service? What obstacles do you see for yourself as you begin on this journey? We would love to hear your thoughts.

This research was developed as part of the Bharat Inclusion Initiative.