HeyHippocratesReflux, GI bleed, H. pylori, the liver — a colleague who cites every call.
Clinical decision support that reasons like a gastroenterologist while you decide: read the alarm feature, score the bleed, test and treat, and work up the liver — with the decision rule and exactly what would change it. Built for verified gastroenterologists.
Watch it think — in gastroenterology.
Two worked examples, reasoned the way it answers: the relevant clinical checks run, the answer is cited, the decision rule is explicit, and what is missing is flagged.
Typical heartburn and regurgitation, no alarm features, only a partial response to lifestyle change — the approach?
Typical reflux without alarm features does not need endoscopy first — start an 8-week trial of a once-daily proton-pump inhibitor with lifestyle measures. Endoscope now only for an alarm feature: dysphagia, weight loss, GI bleeding or anemia, persistent vomiting or new symptoms in an older patient. If reflux is refractory, optimize the PPI and reassess with endoscopy and pH-impedance testing.
Decision ruleTypical GERD, no alarm features → empirical PPI + lifestyle; alarm feature → endoscopy; refractory → optimize PPI, endoscopy, pH-impedance.
Not established from this alone — alarm features, prior therapy, the response and Barrett risk change the plan.
60-year-old, melena, hemoglobin 8, blood pressure 100/70, taking NSAIDs — Glasgow-Blatchford and next step?
A patient with melena, anemia and borderline pressure has a high Glasgow-Blatchford score, so this is not an outpatient bleed. Resuscitate with a restrictive transfusion strategy (target hemoglobin around 7), start an IV proton-pump inhibitor for a likely NSAID ulcer, stop the NSAID, and arrange upper endoscopy within 24 hours. Suspect a variceal source in a cirrhotic and add a vasoactive drug and antibiotics.
Decision ruleLow Glasgow-Blatchford (0–1) → outpatient; higher → admit, resuscitate (restrictive transfusion), IV PPI, endoscopy <24 h; variceal → vasoactive drug + antibiotics.
Not established from this alone — the exact score, hemodynamics, cirrhosis and anticoagulation change the plan.
Illustrative simulations. The physician always decides.
The answers gastroenterologists actually look up.
The exact questions your field searches for — answered in the format the knowledge base speaks: the answer, the decision rule, what's missing, and the source.
GERD — the approach and when to endoscope?
For typical heartburn and regurgitation without alarm features, treat empirically with an 8-week proton-pump inhibitor and lifestyle change rather than scoping first. Reserve prompt endoscopy for alarm features — dysphagia, odynophagia, weight loss, GI bleeding, iron-deficiency anemia, persistent vomiting, or new symptoms over age 60. Refractory reflux is worked up with an optimized PPI, endoscopy and pH-impedance testing, with Barrett surveillance where indicated.ACG · GERD
Decision ruleTypical, no alarm → empirical PPI + lifestyle; alarm feature → endoscopy; refractory → optimize PPI, endoscopy, pH-impedance; surveil Barrett.
Missing data: alarm features, prior therapy, the response and the endoscopic findings change the plan.
Acute upper-GI bleeding — Glasgow-Blatchford and management?
The Glasgow-Blatchford score uses urea, hemoglobin, blood pressure, pulse, melena, syncope and comorbidity to identify who needs intervention; a score of 0–1 can often be managed as an outpatient. Otherwise resuscitate, transfuse restrictively (hemoglobin target about 7), give an IV proton-pump inhibitor for suspected non-variceal bleeding, and scope within 24 hours. A variceal bleed in cirrhosis adds a vasoactive drug, antibiotics and band ligation.Glasgow-Blatchford
Decision ruleGlasgow-Blatchford 0–1 → outpatient; higher → admit + resuscitate (restrictive transfusion) + IV PPI + endoscopy <24 h; variceal → vasoactive drug + antibiotics + ligation.
Missing data: the score, hemodynamics, the suspected source and anticoagulation change management.
Helicobacter pylori and peptic ulcer — test and treat?
For uninvestigated dyspepsia without alarm features in a younger patient, a test-and-treat strategy is appropriate — a urea breath test or stool antigen (off PPI and antibiotics), then eradication. Choose therapy by local clarithromycin resistance: bismuth quadruple therapy is a robust first line where resistance is high. Confirm eradication after treatment, and stop any offending NSAID in peptic ulcer disease.ACG · H. pylori
Decision ruleDyspepsia, no alarm, younger → test-and-treat (breath/stool off PPI); eradicate by local resistance (bismuth quadruple if high); confirm cure; stop NSAID.
Missing data: age, alarm features, prior antibiotics and local resistance change the strategy.
Abnormal liver enzymes — the workup?
Classify the pattern first: a hepatocellular rise (ALT/AST) versus a cholestatic one (ALP/GGT), and how high and how chronic. A hepatocellular pattern points to viral hepatitis, metabolic (MASLD) or alcohol-related liver disease, drugs, autoimmune hepatitis or hemochromatosis; a cholestatic pattern goes to ultrasound for obstruction, then AMA and MRCP. Stage fibrosis non-invasively (for example FIB-4) and treat the cause.guideline synthesis
Decision rulePattern (hepatocellular vs cholestatic) + degree + chronicity → targeted workup; cholestatic → ultrasound first; stage fibrosis (FIB-4); treat the cause.
Missing data: the pattern and magnitude, chronicity, risk factors and imaging drive the workup.
What it knows in gastroenterology
Your whole practice, one specialist brain — grounded in specialist-written, cited knowledge.
A generalist gastroenterology colleague across reflux, GI bleeding, IBD, hepatology and pancreaticobiliary — cited, non-directive, always with the decision rule.
Deep specialist agents for hepatology, inflammatory bowel disease, pancreaticobiliary and advanced endoscopy — each its own specialist voice and knowledge base — plus your own uploaded materials.
Why not just use OpenEvidence or ChatGPT?
A large language model answers any clinical question with total confidence and no idea whether it is right. A search engine finds papers. HeyHippocrates reasons inside your specialty from a knowledge base a specialist in your field wrote and signed, and tells you the decision rule and what is missing.
Be one of the first gastroenterologists in.
The waitlist is a founding cohort: 40 specialties in build, neurosurgery opens first, and your specialty’s waitlist decides what opens next.
- 1 free month — the whole product, no card.
- Then 50% off for three months.
- A numbered Founding Gastroenterologist badge.
- First say on what your specialty gets next.
Join the gastroenterology waitlist
Leave your name and work email. We'll reach out the moment gastroenterology opens — with your Founding Gastroenterologist offer.
One email at launch. No spam, ever.
Questions gastroenterologists ask
It reasons in the tools you use — the GERD alarm features, the Glasgow-Blatchford score and the H. pylori pathway — and always shows the reasoning, the decision rule and what would change it. The decision always stays your call.
Both. Residents use it to pressure-test a plan and learn the decision rule; attendings use it as a fast, cited second opinion in clinic and before endoscopy. Access is gated to verified physicians.
Hepatology, inflammatory bowel disease, pancreaticobiliary and advanced endoscopy each have their own deep specialist agent on the Pro plan, on top of the full gastroenterology scope on Basic.
No. It is physician-facing clinical decision support — a cited thinking partner. It does not diagnose, does not treat, and gives no patient-facing advice. The decision always stays with you.
Every answer is grounded only in specialist-written, cited gastroenterology knowledge — traceable to a named source — and always states the decision rule plus what data is missing, instead of a confident guess.