Nurse phone follow-up for heart-failure patients in the 90 days after discharge
TL;DR Pulsekeep runs a cardiac-nurse desk that calls heart-failure patients on a fixed schedule for 90 days after they leave hospital, checks weight, breathlessness and medicines, and escalates to their own cardiologist. Cardiac hospitals pay per enrolled patient because the first three months are when patients die or come back.
Heart failure in India hits younger people than in the West, and it kills them early. In Kerala's Trivandrum registry of 1,205 patients, 30.8% had died within a year and 18.1% within the first three months, and the average patient was 61[1]. Three in ten of those discharged were readmitted at least once that year. Only 24.6% of patients with reduced pumping function went home on the full set of guideline medicines[1].
A national registry across 53 hospitals in 21 states found 22.1% one-year mortality[2]. After discharge most families are on their own until the next OPD date, which is often weeks away.
What it does, in regulatory terms: structured follow-up and monitoring under the treating cardiologist's protocol. It does not diagnose or change treatment. Any medicine change is made by the patient's registered doctor under the 2020 Telemedicine Practice Guidelines[3].
The hospital enrols the patient at discharge. Pulsekeep's nurses call on a set rhythm, weekly at first and then fortnightly, and the family logs daily weight and blood pressure from a home scale and cuff through a chat app or a simple app. A two-kilo weight gain in three days, new swelling or breathlessness at night triggers a same-day call to the cardiologist's team. Nurses also check that the prescribed medicines were bought and are being taken, and they nudge patients towards the uptitration visits the cardiologist planned.
The seed pitched AI that predicts chronic-disease trouble. We reframed it, because the best evidence is for plain structured phone support and the thresholds here are rules the cardiologist sets rather than a model.
The patient benefits but the hospital pays. Cardiac hospitals want fewer deaths and readmissions on their books, and they'd much rather see patients back in their own OPD than in a rival's emergency room. We found no public price for HF follow-up programmes in India, so the fee is a guess to test - roughly ₹1,500–3,000 per patient for the 90 days, with families able to extend privately.
India now has multi-state registry data that shows how big the post-discharge gap is[2]. CDSCO finalised its guidance on medical device software in July 2026[4], which makes it clearer how to keep a rules-based follow-up tool out of the higher device classes. The DPDP Rules, notified in November 2025, set out the consent duties for health data[5].
We couldn't verify any published count of heart-failure admissions in India. The firm figures are the registries - 10,850 patients enrolled across 53 hospitals[2], and a one-district registry that found 1,205 admissions from 18 hospitals[1]. Each partner hospital with a busy cardiology unit should give you a few hundred eligible discharges a year (our estimate).
One Indian cardiac-tech firm reads ECGs remotely for Indian clinics, hospitals and public health centres[6], which shows hospitals do buy remote cardiac services, but it doesn't run post-discharge follow-up. A US remote-care company sells home-care and remote patient management to health systems[7]. Big Indian hospitals run their own heart-failure clinics, but we found no independent service that does the calling for them. Take that last finding with a pinch of salt, because our web search was limited.
| Difficulty | Moderate (44/100) |
| Starting capital | ₹1.1–2 Cr |
| Time to launch | 6–9 months |
The software is the easy part. The hard parts are hiring cardiac nurses who stick around, writing escalation protocols cardiologists will sign and getting hospitals to share discharge data fast enough that the first call happens within 72 hours. Night cover for escalations is the cost most founders underestimate.
Regulated. Pulsekeep does monitoring and follow-up under the cardiologist's protocol, not diagnosis, and the rules keep it that way.
Building this outside India? The categories above usually have a local equivalent, so check with your own regulator.
AI must not decide whether a patient is getting worse or needs a medicine change. This is monitoring under a doctor's protocol, not diagnosis, and every escalation goes to the cardiologist.
Want to work out the AI layer for Pulsekeep? Talk to Yash →
I think this is a good fit for a founder with a nursing or cardiology-operations background and hospital relationships. If you want a pure software business then skip it. Prove first that one hospital will enrol 80% of eligible HF discharges and that nurses reach 90% of them within 72 hours.
Plan for roughly ₹1.1–2 Cr, most of it in cardiac nurses, night cover and the software. One hospital pilot is the sensible start.
There's no single licence, but you follow the 2020 Telemedicine Practice Guidelines, CDSCO's device software guidance and the DPDP Act. Doctors make every treatment call.
Yes, the model works wherever hospitals worry about readmissions. Telehealth, nursing scope and health data rules differ, so check with local regulators first.
The free Pulsekeep validation kit has the line-by-line capital breakdown, the effort scoring, the telemedicine, device-software and data-protection checklist, a 14-day validation sprint with pass/fail targets, a 90-day launch plan, 10 interview questions for cardiologists and hospital managers, and 5 assumptions to test. Download the kit →
Thinking about building Pulsekeep or something close to it? The make-or-break work is getting a cardiology department to hand over every discharge and answer escalations the same day, and that's a deal I can help you shape. If you need a hand, you know where to find me. Talk to Yash →
For illustration only. Names are fictional and ideas hypothetical; sourced facts are real but may be out of date. Nothing here is legal, financial or professional advice, and Yash Khatri is not personally liable for any use of it. Read the full disclaimer.