Is there a way to programatically (or manually) download the biologic license applications data in a machine readable format?

Here is a PDF that at the bottom has the 2015 approvals, https://www.fda.gov/downloads/Drugs/DevelopmentApprovalProcess/HowDrugsareDevelopedandApproved/DrugandBiologicApprovalReports/NDAandBLAApprovalReports/UCM486442.pdf but I don't see those BLA applications in this data (link at bottom of page). https://www.fda.gov/drugs/informationondrugs/ucm079750.htm

Thanks!

bumped to the homepage by Community 2 days ago

This question has answers that may be good or bad; the system has marked it active so that they can be reviewed.

i downloaded the Drugs@FDA dataset and was able to find the features the pdf contains. I just used a text editor and a search application but did not go into the semantics of the data.

With the BLA NUMBER or APPLICATION NUMBER (e.g. BLA 125504/0.0) serving as key you can get all your desired infos. However you need to strip down the prefix BLA and the suffix /0.0 to be able to do so.

Beginning with the file Applications.txt you can find the ApplType (BLA) and the Applicant in the field SponsorName.

Within Products.txt you can find the Proprietary Name and the Proper Name in the fields DrugName and ActiveIngredient. But be advised that there can be more than 1 product per ApplNo so you should probably use the product with the ProductNo = 001.

To get the Approval Date you have to take a look into Submissions.txt. There you should filter for SubmissionType = ORIG to find the data in the field SubmissionStatusDate. Also the Review Classification (whether S for Standard or P for Priority) is listed there in the field ReviewPriority.

The last piece of information (the O in Review Classification for Orphan Designation) can be retrived via SubmissionPropertyType.txt. There a record with the corrsponding ApplNo and SubmissionType = ORIG could be found labeling the SubmissionPropertyTypeCode as Orphan.

In summary the data is available in a relational nature. I think your best bet is to import the data and construct the records shown in the pdf by joining it together (e.g. by using a relational database and applying SQL). As i mentioned at the beginning i did not go into the semantics of the data so there is a good chance i missed something.

Edit

Here a little proof of concept in SQL Proof of Concept

  • from what URL did you download the data? Looks slightly different than mine. – Sam Shleifer Oct 11 at 2:46
  • I used the download link you provided. Yeah I noticed that my result looks slightly different than the report you linked to your question. My answer is not a final solution but should guide in the right direction. – randomDude1001 Oct 11 at 7:46

Your Answer

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

Not the answer you're looking for? Browse other questions tagged or ask your own question.