Drug interaction not working

Situation
Drug internaction not working tried to add both drug like Ibprufon and Warfrin it did not flag it, can anyone support me in this case.

Tried fresh install same issue

OpenEMR Version
I’m using OpenEMR version: 8.0.0

Browser:
I’m using:Chrome

Operating System
I’m using:Alma cpanel

Hi @mkm1971 ,

I do not have an answer to your issue but I was wondering how did you get to this page (Medications?? / Prescriptions??) I have not paid attention to it from my end.

@moussa Hi so when you upload Rxnorm and Rxcui in the drugs npl i have enabled that so when you add 2 prescprtion or drugs you will be able to see it below.

I am thinking it is related to this, RxNav’s drug/drug interaction features were discontinued:

I will chime in here. The feature worked in the beginning but the company that provided the data. Stopped providing the limited data for free. So, the feature though it exist does not have a data source.

Adding to what @juggernautsei said: NLM switched the interaction API off on 2 January 2024 (their notice). The prescription screen still calls rxnav.nlm.nih.gov/REST/interaction/list.json, gets nothing back, and shows “No interactions found”, so the check is still there with no data behind it.

I wrote a small open-source server that answers that same URL from your own machine, in case it helps anyone here: GitHub - shubhamjoshipromail-svg/localddi: Self-hosted drug-drug interaction API that serves the JSON shape of the discontinued NLM RxNav Interaction API. Deterministic, offline after import, no API keys. Not for clinical use. · GitHub

It runs in Docker, imports RxNorm and the DDInter 2.0 interaction database, and returns the JSON structure NLM used to return. OpenEMR reads the comment, the two drug names, and each interaction’s severity and description, and the response has all of those. On the OpenEMR side the change should be the host name in one line of controllers/C_Prescription.class.php.

Some caveats:

  • I haven’t tested it inside a running OpenEMR, only against the source on master. If someone with a test instance tries it, I’d like to hear what happens.
  • It isn’t for clinical use. Nobody has clinically validated it, and a blank result doesn’t mean a combination is safe.
  • Coverage is partial. On the ONC high-priority list (the “ONCHigh” source NLM used to serve) it finds 721 of the 793 pairs that can still be tested, about 91%. On old NLM responses that came from DrugBank it finds only 29%, because DrugBank listed far more minor interactions than DDInter does.
  • The data licence is non-commercial (DDInter is CC BY-NC-SA 4.0). The code is MIT, and the server can load other data you have a licence for.
  • The description is one generated sentence with the severity level. DDInter’s downloads have no mechanism text.

Two questions for the maintainers: would you take a PR that makes the base URL for that call configurable? And is there a data source the project would rather point people to?

@Shubham_Joshi thanks for posting! Finding a low cost data source is a challenge in the medical industry. Those that have spent the time vetting the data. Have that security mote. I asked perplexity is there an LLM for drug to drug interaction. The answer is below. But this would be for a clinic that is not using a paid eRx service or a country that does now have a national prescription system, imo.

Yes—there are LLM-based systems for drug–drug interactions (DDIs), including specialized research models and general-purpose assistants. The important distinction is between a model that can discuss interactions and a system reliable enough to guide prescribing. Current evidence does not support treating a standalone LLM as the final authority on whether a drug combination is safe.[pubmed.ncbi.nlm.nih][pubmed.ncbi.nlm.nih]

What exists?

Type Examples What it does
General-purpose LLMs ChatGPT, Claude, Gemini Answer interaction questions and explain possible mechanisms; these have been evaluated on clinical case summaries. [pubmed.ncbi.nlm.nih]
Specialized research systems DDI-LLM, DDI-JUDGE Predict interactions using approaches such as molecular-graph information, biomedical information, and task-specific prompting. These are research approaches, not equivalent to validated prescribing tools. [sciencedirect][mendeley]
Biomedical text models BioBERT-based systems Extract reported drug interactions from published text—a different task from deciding whether a particular patient can safely take two drugs. [pubmed.ncbi.nlm.nih]

How reliable are they?

Results are promising but uneven. In one 2025 study, ChatGPT, Claude, and Gemini detected many interactions in pharmacovigilance case summaries, but they also reported interactions that were not present in negative-control cases. Correct answers across all interaction-related questions ranged from 33% to 68%. Those results apply to the particular models and cases tested—not every current version or clinical setting.[pubmed.ncbi.nlm.nih]

That highlights a crucial limitation: identifying a possible interaction is not the same as reliably confirming that there is no interaction, assessing its clinical significance, or choosing the right management. Reviews identify hallucinations, data quality, explainability, and prospective clinical validation as unresolved issues.[pubmed.ncbi.nlm.nih][pubmed.ncbi.nlm.nih]

What makes sense in practice?

For checking a real medication list, a prudent approach is to use an established drug-interaction reference with pharmacist or prescriber review, rather than rely on an LLM alone. An LLM can help explain the findings, but should not independently clear a combination as safe.[pubmed.ncbi.nlm.nih][nature]

If you are building software, a more defensible design is to have the LLM retrieve and explain documented interaction records from a maintained drug database, rather than generate interaction claims from memory. That addresses some—but not all—of the reliability and data-quality problems identified in the research.[pubmed.ncbi.nlm.nih]

Are you looking for a tool to check medications, or a model/API to build into an application?