Real-world evidence in India: what it is, the rules, and how studies are built

A clinical trial answers a narrow question well: does this drug work better than the alternative, in these patients, under these conditions? It cannot answer most of what doctors, payers and drug makers need to know afterwards. How is the drug actually used? In which patients? With what results over five years? How does it compare with what Indian patients really receive?

Those questions are answered from routine care. This article explains what real-world data and evidence are, what Indian rules say, and how a study built with clinicians in India works. It is written for medical affairs, HEOR and clinical development teams, and for clinicians who want to take part.

Real-world data and real-world evidence

The definitions most widely used come from the US Food and Drug Administration’s 2018 framework for real-world evidence:

• Real-world data (RWD) are “data relating to patient health status and/or the delivery of health care routinely collected from a variety of sources”: electronic health records, claims and billing data, product and disease registries, data patients generate themselves, and data from devices.
• Real-world evidence (RWE) is “the clinical evidence about the usage and potential benefits or risks of a medical product derived from analysis of RWD.”

Regulators now use RWE in earnest. In the US, the 21st Century Cures Act of 2016 required the FDA to build a programme for using it to support new indications and post-approval requirements. In Europe, the European Medicines Agency set up DARWIN EU in 2022, a network that has brought in about 40 data partners, each converting their data to a common format so the same analysis can run across all of them.

What RWE can answer

• Treatment patterns: who receives a therapy, at what dose, for how long, and what they receive before and after it.
• Effectiveness: how a treatment performs in routine practice, including in patients a trial excluded.
• Safety: adverse events in larger and more varied populations, over longer periods.
• Natural history: how a disease progresses, which matters most for rare and chronic conditions.
• Burden and cost: visits, admissions, tests and spending around a condition.
• External comparators: a well-characterised group of routine-care patients to compare with a single-arm study.

What Indian rules say

The New Drugs and Clinical Trials Rules, 2019 do not use the phrase “real-world evidence”. They do draw the distinction that matters: “efficacy” is a drug’s effect in a controlled clinical setting, while “effectiveness” is its “ability to achieve the desired effect in a real world clinical situation after approval of the drug.”

Their Fifth Schedule, on post-market assessment, sets out three routes:

1. Phase IV trials, run under an approved protocol with the protections and compensation rules that apply to clinical trials.
2. Post-marketing surveillance, observational or non-interventional studies, which need the Central Licencing Authority to approve the protocol but are not subject to the provisions that govern clinical trials of a new drug.
3. Periodic safety update reports (PSURs): every six months for the first two years after approval, then annually for the next two.

A draft amendment published in June 2026 proposed extending PSUR reporting for some products, including biologics. It had not been notified as final at the time of writing.

We are not aware of specific CDSCO guidance on real-world evidence as of October 2026. An industry position paper in 2022 noted that RWE studies are not specifically defined in the Rules. Check the current position with CDSCO before relying on it.

The Clinical Trials Registry - India (CTRI), run by ICMR, has been mandatory for clinical trials since 15 June 2009, and registration must happen before the first participant is enrolled. For other research, including observational studies, ICMR describes registration as voluntary. CTRI’s own guidance is “when in doubt, register”, and it advises that even retrospective studies register before data collection starts.

ICMR’s National Ethical Guidelines for Biomedical and Health Research Involving Human Participants (2017) apply to almost every RWE study:

• every research proposal goes to an ethics committee, and the committee, not the researcher, decides whether review can be exempt, expedited or full;
• registries set up for research, or whose data will be used for research, need the ethics committee’s prior approval;
• consent can be waived only in defined cases, such as retrospective studies where participants are de-identified or cannot be contacted, or research on anonymised data, and only where the risk is less than minimal;
• personal information in records used for research should be coded or anonymised, with access limited.

The Digital Personal Data Protection Act, 2023 has a research exemption: section 17(2)(b) says the Act does not apply to processing for research, archiving or statistical purposes when the data is not used to make a decision about the individual and the processing follows prescribed standards. The DPDP Rules, notified in November 2025, set those standards (lawful, purpose-limited, minimised, accurate, time-limited, secure and accountable processing). They take effect around May 2027, with the Act’s main obligations. Until then, and alongside them, consent and ethics approval remain the basis for most studies.

ABDM. The Ayushman Bharat Digital Mission links records to a patient’s ABHA and moves them between providers with consent. By August 2026 the government reported 96.43 crore ABHA numbers and more than 110 crore health records linked. The draft Health Data Management Policy (April 2022) provides for sharing anonymised, aggregated data for research, with access procedures to be set by the National Health Authority; we could not find those procedures published.

Where Indian real-world data comes from

• Hospital and clinic records, where they are digital and structured.
• Disease registries, such as ICMR’s National Cancer Registry Programme, and specialty registries run by professional societies.
• Pharmacy and laboratory data.
• Patient-reported data, collected with consent.

The challenge is not volume, which India has in abundance. It is that much routine care is recorded on paper or in free text, records are split across providers, and data quality is rarely measured. A study that ignores this produces numbers nobody should trust.

How a study built with clinicians works

For questions that existing data cannot answer, the most reliable approach is to build the dataset prospectively with clinicians who see the patients. This is how we run a commissioned study:

1. The question. A sponsor brings a question: treatment patterns in a disease, outcomes on a therapy, the natural history of a condition.
2. Feasibility. We assess whether routine Indian care can answer it: how many patients, at which kinds of centres, with which data recorded.
3. The clinician network. We recruit specialists in the therapy area as investigators, at centres that see the right patients.
4. Protocol and approvals. A protocol with a defined data set and outcome definitions, ethics committee approval at each site, CTRI registration where appropriate, and regulatory approval where the study falls under the post-marketing rules.
5. Consent. Patients are told what is collected and why, and their consent is recorded.
6. Data capture. Data is recorded as structured fields in the course of care, to the study’s definitions, rather than abstracted from free text later.
7. Quality checks. Completeness, conformance and plausibility are checked continuously, and queries go back to the site while the patient is still remembered.
8. Analysis and reporting. De-identified data is analysed to the protocol, and results are reported to the sponsor and, by agreement, published with the clinicians.

The independence of the clinicians and the integrity of the data matter more than speed. A study whose findings the sponsor controls entirely, or whose data cannot be audited, will not persuade a regulator, a payer or a doctor.

Where 92health fits

92health is an EMR used by clinics, hospitals, pharmacies, labs and optical shops in India. Because clinicians record care in it as structured, coded data (diagnoses in ICD-10, lab results validated as FHIR R4, records linkable to ABHA with consent), the same records that run the clinic can, with patients’ consent and ethics approval, answer research questions.

We work with life-sciences teams to commission evidence in a therapy area: scoping the question, recruiting the clinician network, and running the study as described above. Book a call to talk about a question.

Sources

• US FDA. Framework for FDA’s Real-World Evidence Program, December 2018.
• European Medicines Agency. DARWIN EU.
• CDSCO. New Drugs and Clinical Trials Rules, 2019 (consolidated text, as amended in 2022).
• Clinical Trials Registry - India and its FAQ.
• ICMR. National Ethical Guidelines for Biomedical and Health Research Involving Human Participants, 2017.
• MeitY. Digital Personal Data Protection Act, 2023 and DPDP Rules, 2025.
• PIB. ABDM factsheet, August 2026.