The evidence needs to be beside the doctor.
FromA video call→ToA purpose-built clinical environment
Video consultation works well for many kinds of care. For the cases where physical evidence matters, a second layer designs the consultation from the start for a doctor and patient in different places: the evidence in view, more of it a request away, and knowledge within reach.
“Let’s listen to the lungs again”
The doctor reviews the brief. Something doesn’t sit right.
“Let’s listen to the lungs again.”
Two minutes later, at the Point of Care, the operator places the digital stethoscope. The sounds arrive on the doctor’s screen. The doctor listens, and asks for one more reading. illustrative
The examination and the consultation have become one conversation.
Two environments, each built for its purpose
The patient environment is at the Point of Care: private and comfortable. The patient is examined, measured, imaged, tested and prepared.
The doctor environment is a professional workspace built around evidence, reasoning, decision, not a laptop and a webcam. It opens with everything in view:
- Patient: chief complaint, history, medication.
- Evidence: vitals, findings, images, heart and lung sounds, ECG, tests.
- Context: previous reports and consultations.
Clinical intelligence alongside. AI can flag missing information, organise the case and surface relevant medical knowledge. The clinician always makes the decision.
An interactive consultation
The doctor can request more evidence at any point: “Let’s check the lungs again.” “Please take another image.” “Let’s repeat the blood pressure.” “I need an ECG.”
- Doctor
- request
- Point of Care
- evidence
- doctor
- decision
Knowledge at the point of need. Facing an unfamiliar case, the doctor doesn’t have to leave the consultation to search the internet. The environment can surface relevant guidance, similar cases, expert reasoning and learning, right when it matters.
Learning from reasoning, not asking for answers. The system doesn’t put a young doctor on the phone to a specialist. It shows how an experienced clinician approached a similar case: what they considered, what evidence mattered, what they decided and what happened next. The doctor learns, then returns to their own patient.
Go deeper
A closed loop
Physical examination, clinical brief, evidence review, patient conversation, more examination if needed, relevant knowledge, clinical reasoning, decision, prescription, tests or referral, follow-up and outcome. The outcome matters, because the system can learn what happened after the consultation.
Who it helps. Patients get a consultation built on real evidence, with the doctor able to check what they need in the moment. Doctors get the conditions to think: complete information, the ability to ask for more, and knowledge within reach. Young doctors gain confidence, supported by the experience of others.
What the guidelines already require
The Telemedicine Practice Guidelines set the frame this consultation has to fit. The patient’s consent is needed: implied when the patient starts the consultation, explicit when a health worker or caregiver does, and the doctor records it[1]. Prescribing follows the guidelines’ lists of medicines by type of consultation, and Schedule X drugs and narcotic or psychotropic substances may not be prescribed[1]. This shift adds no new prescribing powers.
Getting it right
The doctor decides. AI organises and informs. It never makes the clinical decision, and every suggestion is clearly marked as support.
Clinical safety first. The consultation follows the national guidelines, with clear pathways for in-person care whenever it’s needed[1].
Reliable connections. Evidence has to arrive complete and fast, including where connectivity is weak.
Patient consent. Patients know who is examining them, who is reviewing their case and how their information is used.
What could go wrong
- Over-reliance on the tool.A polished workspace could make doctors defer to its suggestions, so AI support stays clearly marked as support.
- Too many prompts.Alerts could crowd out thinking, so the workspace shows what the case needs and no more.
- Bandwidth.Heart sounds, images and ECG traces are heavy, so the workspace has to degrade gracefully on a weak network.
- Longer consultations.Asking for more evidence takes time, so time to decision is measured, not assumed.
How we’d test it
The pilot Points of Care connected to a small group of doctors working from a purpose-built workspace, starting with interactive examination.
- The test
- Interactive examination
- What we would measure
- Measure how often doctors request additional evidence, time to decision, referral appropriateness, patient outcomes and doctor confidence. Add knowledge support and similar-case learning next.
- What would end the idea
- Doctors rarely use the option to ask for more evidence, or it adds time without improving decisions.
Sources
- [1]Board of Governors in supersession of the Medical Council of India, Telemedicine Practice Guidelines (25 March 2020), sections 1.4.2.3, 3.4 and 3.7.4Checked; primary text
Spotted a source we missed or a rule we misread? Tell us what we got wrong.
A doctor who can see the whole case will also meet cases that stretch them. The next shift is about learning in that moment.