Checks at ordering
Interactions, allergies and duplicate medicines, checked as each order is written in OPD, wards and ICU.
IN DEVELOPMENT · PRE-VALIDATION
A second look at every prescription.Aurora is being built to give one.
From OPD to discharge, Aurora is being built to check each order against the patient’s allergies, kidney and liver function and other medicines, send concerns to a pharmacist, and keep abnormal results open until someone acts.
THE DECISION STAYS WITH THE CLINICIAN
VERSION 1 · IN DEVELOPMENT
Version 1, from OPD to discharge.
- OPD
- WARD
- ICU
- LABS
- PHARMACY
- DISCHARGE
Dose flags from the label
Kidney and liver function checked against each drug’s label. Aurora flags; it never calculates a dose.
Antibiotic stewardship
48-hour time-outs, de-escalation when cultures return, stop dates and surgical prophylaxis duration.
Pharmacist review
Apart from a short high-risk list, recommendations go to a pharmacist’s queue, not straight to the doctor.
Discharge reconciliation
Medicines before admission, in hospital and at discharge compared, with every gap sent for review.
Results followed through
Abnormal lab and health check-up results stay open until someone acts on them.
NOT IN VERSION 1
Diagnosis, triage, dose calculation and ICU prediction. Aurora informs; the doctor makes every decision.
How a flag travels.
- 01
At ordering
Aurora checks each order as it is written. Only a short list of high-risk problems interrupts the prescriber.
- 02
To the pharmacist
Everything else goes to a pharmacist’s review queue, with the rule, its source and the patient values that triggered it.
- 03
The doctor decides
The pharmacist recommends; the doctor accepts or overrides. Aurora never changes an order.
PHARMACIST REVIEW QUEUE
5 OPENILLUSTRATIVE · SAMPLE DATA, NOT REAL PATIENTSTHE DOCTOR DECIDES
Quiet by default
Alerts that fire for everything get ignored, so only high-risk problems interrupt. The rest wait for review.
Every flag shows its source
Fixed rules decide what gets flagged. AI only explains a flag in plain language and cites where it comes from.
Built for Indian prescriptions
Being built on India’s Common Drug Codes, so Indian brand names resolve to their molecules, and on each hospital’s own antibiotic policy.
Validated before use
Not for clinical use until it has been validated in Indian hospitals.
The published evidence Aurora builds on.
INDEPENDENT RESEARCH — NOT AURORA’S RESULTSAurora is in development and pre-validation. These figures describe published research on the problem and on comparable approaches; they are not Aurora’s performance, and Aurora is not a certified medical device.
- 2.85%1
- of hospital admissions caused by an adverse drug reaction: the median across 21 prospective Indian studies
- 12% fewer2
- high-risk drug combinations given when alerts were limited to high-risk pairs, in a randomised trial across nine ICUs
- 12.5% → 44%3
- antibiotic de-escalation after a stewardship programme with review by a physician or pharmacist, at four Indian secondary-care hospitals
These are measures of prescribing, not of patient outcomes: a review of 16 trials found decision support did not change mortality.4 Aurora’s own results will be published once it has been validated.
- 1. Patel & Patel, Current Drug Safety, 2016. Systematic review of prospective studies. doi.org/10.2174/1574886310666150921104523
- 2. Bakker et al., The Lancet, 2024. Cluster randomised stepped-wedge trial. doi.org/10.1016/S0140-6736(23)02465-0
- 3. Zacchaeus et al., Antimicrobial Stewardship & Healthcare Epidemiology, 2023. doi.org/10.1017/ash.2023.171
- 4. Moja et al., American Journal of Public Health, 2014. Systematic review and meta-analysis. doi.org/10.2105/AJPH.2014.302164
Follow Aurora from first principles to the bedside.
OPENS WHATSAPP · dr.jys@xumanai.com