OpenEvidence answers clinical questions in natural language with cited, evidence-grounded responses. It is free for verified US healthcare professionals, and it has signed content agreements with NEJM, JAMA, Nature, NCCN, Cochrane and others, plus partnerships with societies including ACC, AAFP, ACOG and ACEP. For looking something up, it is excellent and it is free. Pogosh does a different job: it returns structured, coded next steps that a chart or a product can act on.
FACTS VERIFIED AGAINST OPENEVIDENCE INC. SOURCESAn earlier version of this page implied that a signed BAA covering PHI was a reason to choose Pogosh over OpenEvidence. That was unfair. OpenEvidence's security page states that it complies with HIPAA and that covered entities may transmit protected health information under its standard business associate agreement. What remains true is narrower: OpenEvidence's published API terms say developers may not use the API to process PHI.
Ask OpenEvidence a question and you get a cited answer in prose, written for a clinician to read and translate into action. Give Pogosh a note and you get structured, coded output: named labs with LOINC codes, imaging with RadLex codes, the specialty to involve and how urgently (adapted to setting, age and sex, including what happens after the emergency department), escalation criteria, a strict emergency check, relevant calculators offered but never auto-run, and a note draft. An EHR or a product can place those items directly; prose has to be read and retyped. Neither is better in general. They answer different needs, and the output from both is advisory.
We would rather you picked the right tool than picked ours. These lists are meant to be used honestly.
Only rows that can be checked against a public source. Rows where Pogosh CDS is the weaker option are included on purpose.
| Pogosh CDS | OpenEvidence | |
|---|---|---|
| What it returns | Coded orders (LOINC labs, RadLex imaging), routing, escalation, note draft | A cited, natural-language answer to a clinical question |
| Who consumes it | A clinician, or software placing the orders | A clinician reading the answer |
| Licensed premium content | None; Pogosh is not an evidence library | Agreements with NEJM, JAMA, Nature, NCCN, Cochrane and others |
| Cost to an individual clinician | Free to try at /try; plans from $149 per month | Free for verified US healthcare professionals |
| Developer API | Documented REST API; evaluation key issued directly | API Terms of Service exist; no public endpoint documentation or pricing found |
| PHI | BAA available; PHI endpoint runs on AWS under the AWS BAA | Security page offers covered entities a BAA; API terms bar PHI through the API |
| Calculators | Offered from the note, never auto-run | Not evaluated here |
| Independent validation | None published yet | Not evaluated here |
Verified against OpenEvidence Inc. public sources. Where a figure is not published by the vendor we say so rather than estimate. Products change — if something here is out of date or unfair, email support@pogosh.com and we will correct it.
If what you need is to look something up, you probably should not; OpenEvidence is free for verified US clinicians and very good at that job. Pogosh is bought by people who need structured output another system consumes: an EHR that has to place coded orders, a practice that wants the workup inside the chart, or a developer integrating decision support into a product.
According to its security page, covered entities may transmit PHI under OpenEvidence's standard business associate agreement. Its separate API Terms of Service say developers may not use the API to process PHI. If you are integrating through an API with patient data, that distinction matters.
Yes, and that is a sensible setup: OpenEvidence for the question you want to read an answer to, Pogosh for the coded orders and routing your chart needs to act on. Both are advisory; the clinician decides.
Paste a real note and judge the reasoning yourself.