Data sources
Where the information on Frontlight Research comes from, and how we work with it.
What we do
We search publicly available, relevant and recognised sources: medical and scientific journals, official clinical trial registries and the databases of medicines agencies. We bring their data together by disease, so that you can see in one place what is known about a disease and what is being tested, with the source of every fact.
How we work
Updated every day
We read the trial registries ClinicalTrials.gov and EU CTIS and the new publications in PubMed again every day. The research groups from OpenAlex are updated every week.
Dated
Next to each record you see the date we last checked it.
Arranged by disease
Every trial, medicine and publication is filed under the diseases it is about. A disease often has several names, so we collect its synonyms from the medical vocabularies listed below. The search finds a disease by any of its names, even with a typo.
Support organisations
We find patient and support organisations in the member lists of their European and international umbrella bodies, and we check each one on its own website before we list it.
Machine summary
Some texts are written by a computer from registry and publication data. They are marked “Machine summary”, they can be wrong, and they always link the record they were made from.
How we rate health claims
Every fact check ends with one of four labels. The label comes from the kind of studies we found, not from how the claim sounds. The text under it names its sources.
- True
several randomised, controlled human trials agree.
- Partly true
good results in people, but not confirmed by a randomised trial, or only part of the claim holds.
- Unproven
only animal or lab data, or too little or conflicting human data.
- False
randomised human trials found no effect, or found harm.
Data sources
We read only public, official records. We do not collect contact details of researchers from websites.
Trials and research
The US National Library of Medicine’s registry of clinical studies. We use it for trials, their sites, their status and who can take part, as written by the sponsor.
The European Union’s Clinical Trials Information System, run by the European Medicines Agency (EMA). We use it for trials authorised in the EU and the EEA, their sites and their status. Reused, also commercially, under the EMA Legal Notice, with the credit “Source: EU Clinical Trials Information System (CTIS), European Medicines Agency”. The EMA does not endorse Frontlight Research.
The index of biomedical literature. We use it for the titles, journals and dates of publications on each disease.
An open search service for life-science literature. We use it to find recent animal and lab studies of treatments, and link each study on PubMed.
An open catalogue of research works and institutions. We use it to see which research groups publish on a disease.
The Research Organization Registry. We use it to identify the institutions behind trial sites and research groups.
Approved medicines
The European Medicines Agency. We use its medicine pages, its list of medicines in the EU and its EU orphan designations for treatments approved in the EU and for the approval records under a medicine. Reused under the EMA Legal Notice, with the credit “Source: European Medicines Agency”.
US prescribing labels, published by the National Library of Medicine. Each medicine approved in the US links its label there.
Open data of the US Food and Drug Administration. We use its drug labels and the approval dates in Drugs@FDA to find medicines approved in the US and to show when each was first approved.
The Swiss medicines agency. We use its list of medicines for treatments approved in Switzerland.
The French public medicines database. We use it for medicines approved in France.
The Drug Product Database of Health Canada. We use it to see whether a medicine is on the market in Canada.
Disease names
The Mondo Disease Ontology, CC BY 4.0. We use it for disease names and their synonyms in the search, and to match each disease with its entries in other databases, such as Orphanet, OMIM, ICD and EFO. We combined and filtered the data.
Orphadata, the data of Orphanet, the European reference portal for rare diseases, CC BY 4.0. We use it for the names and synonyms of rare diseases in the search. We combined and filtered the data.
The medical vocabulary of the US National Library of Medicine. We use it for disease names and to search PubMed by disease.
The US clinical version of the disease classification, from the CDC. We use it for disease names.
The knowledge base of the Wikimedia Foundation, CC0. We use it to check that a synonym found in another source belongs to the right disease.
A public–private partnership led by EMBL-EBI, CC0. We use it to cross-check disease synonyms before the search uses them.
Other
IP Geolocation by DB-IP, CC BY 4.0. A copy on our own server guesses your country from your IP address, so a disease page can open with your country chosen. The address is not stored.