HeyHippocratesStewardship, meningitis, endocarditis, HIV — a colleague who cites every call.
Clinical decision support that reasons like an infectious disease physician while you decide: choose and de-escalate the antibiotic, cover the meningitis, apply the Duke criteria, start the HIV plan — with the decision rule and exactly what would change it. Built for verified physicians.
Watch it think — in infectious diseases.
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.
Fever, neck stiffness, altered mental status and a petechial rash — the first move?
Treat suspected bacterial meningitis as an emergency: give dexamethasone with or just before the first dose of empirical antibiotics — ceftriaxone plus vancomycin, adding ampicillin for Listeria in the older or immunocompromised patient. Do not delay antibiotics for the lumbar puncture; image the head first only for a focal deficit, papilledema, seizure, immunocompromise or a depressed level of consciousness. A petechial rash raises meningococcus.
Decision ruleSuspected bacterial meningitis → dexamethasone + empirical antibiotics immediately; LP without delaying treatment; CT first only for focal signs / low GCS / immunocompromise.
Not established from this alone — the CSF, host factors, exposures and local epidemiology change the regimen.
Sepsis from a urinary source, blood cultures now growing E. coli sensitive to ceftriaxone, currently on empirical meropenem — next step?
Now that the organism and susceptibilities are known, de-escalate from the carbapenem to the narrowest effective agent — ceftriaxone here — and secure source control. Set a defined, shortest-effective duration, switch to oral once the patient stabilizes and can absorb, and stop redundant coverage. Stewardship narrows the spectrum without under-treating.
Decision ruleCulture-directed → de-escalate to the narrowest effective agent, source control, shortest effective duration, IV-to-oral switch; stop redundant coverage.
Not established from this alone — the full susceptibilities, source control, host factors and response change the plan.
Illustrative simulations. The physician always decides.
The answers infectious disease physicians 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.
Antimicrobial selection and stewardship — how do I choose and de-escalate?
Start empirically from the likely source, the local antibiogram and the severity, broad enough to cover the probable pathogens without reflex maximal coverage. When cultures and susceptibilities return, de-escalate to the narrowest effective agent and stop redundant drugs. Ensure source control, set the shortest effective duration, and switch intravenous to oral once the patient is stable and absorbing.IDSA · stewardship
Decision ruleEmpirical by source + antibiogram + severity → de-escalate on cultures → source control → shortest effective duration → IV-to-oral when stable.
Missing data: the source and cultures, local resistance, severity and host factors change the choice.
Suspected bacterial meningitis — the emergency sequence?
This is a time-sensitive emergency. Give dexamethasone and empirical antibiotics — ceftriaxone plus vancomycin, with ampicillin for Listeria in patients over 50 or immunocompromised — immediately, without waiting for the lumbar puncture or imaging. Perform CT before LP only when a focal neurologic deficit, new seizure, papilledema, immunocompromise or a depressed consciousness level raises the risk of herniation. Draw cultures, then narrow to the identified organism.IDSA · meningitis
Decision ruleSuspect it → dexamethasone + empirical antibiotics now; LP as soon as safe (do not delay treatment); CT first only for specific risk features; then narrow.
Missing data: the CSF result, host and exposures, and local epidemiology change the regimen.
Infective endocarditis — the Duke criteria and workup?
The modified Duke criteria combine major features — a typical organism in two separate blood cultures and echocardiographic evidence of endocardial involvement — with minor ones such as a predisposing condition, fever, and vascular or immunologic phenomena. Draw at least three sets of blood cultures before antibiotics, get echocardiography (transthoracic then transesophageal), and treat with prolonged targeted therapy. Surgery is for heart failure, uncontrolled infection or a high embolic risk.Modified Duke
Decision ruleApply the modified Duke criteria; ≥3 blood-culture sets before antibiotics; echocardiography (TTE→TEE); prolonged targeted therapy; surgery for HF / uncontrolled infection / embolic risk.
Missing data: the blood cultures, the echo, valve and host factors change the diagnosis and treatment.
HIV — screening, when to start ART, and the initial workup?
Screen broadly with an opt-out antigen-antibody test and confirm a reactive result. Offer antiretroviral therapy to everyone with HIV, promptly and regardless of the CD4 count — rapid start improves engagement. Baseline testing includes the CD4 count, viral load, a resistance genotype, and screening for co-infections and opportunistic risk, with prophylaxis guided by the CD4. Counsel on adherence and prevention.DHHS · HIV
Decision ruleOpt-out screen → confirm; start ART for everyone, promptly, regardless of CD4; baseline CD4/viral load/genotype + co-infection screen; prophylaxis by CD4.
Missing data: the confirmed test, CD4, opportunistic screen and resistance data change the plan.
What it knows in infectious diseases
Your whole practice, one specialist brain — grounded in specialist-written, cited knowledge.
A generalist infectious-diseases colleague across sepsis and stewardship, serious infection, endocarditis, HIV and bone-and-joint infection — cited, non-directive, always with the decision rule.
Deep specialist agents for antimicrobial stewardship, HIV medicine, transplant & immunocompromised infection and travel & tropical medicine — 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 infectious disease physicians 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 Infectious Disease Physician badge.
- First say on what your specialty gets next.
Join the infectious diseases waitlist
Leave your name and work email. We'll reach out the moment infectious diseases opens — with your Founding Infectious Disease Physician offer.
One email at launch. No spam, ever.
Questions infectious disease physicians ask
It reasons in the frameworks you use — the antibiogram, culture-directed de-escalation, the Duke criteria and the meningitis 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 on consults and antibiotic rounds. Access is gated to verified physicians.
Antimicrobial stewardship, HIV medicine, transplant & immunocompromised infection and travel & tropical medicine each have their own deep specialist agent on the Pro plan, on top of the full infectious-diseases 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 infectious-diseases knowledge — traceable to a named source — and always states the decision rule plus what data is missing, instead of a confident guess.