-
Notifications
You must be signed in to change notification settings - Fork 15
Pillbox overview
Pillbox is a resource of the U.S. National Library of Medicine, part of the National Institutes of Health, U.S. Department of Health and Human Services. Pillbox is a United States government resource. This document will be moved to an official Pillbox repo in February 2014.
API documentation
Data documentation
Data downloads
very much a work in progress
(Overview is meant to summarize discussions I’ve had with developers who are first starting to work with this data and frequently have similar questions)
-Challenges
-health care, such as adverse events, adherence, EHR/PHR, patient-centered health care, Blue Button, emergency/disaster response
-working with gov drug data sets like SPL and RxNorm
-Solution=innovators+tools
-drug/drug interactions. why?
-indications, late-stage clinical trials results
One of the main goals of Pillbox is to create a search and reference system where any unique pill you can hold in your hand has one and only one record. This is somewhat different from how the data is organized by manufacturers and distributors in the drug labels. Pillbox is still not there yet. Hopefully this overview will make things a little clearer and help you see where we’re going with the project.
The term pill is a bit of a misnomer. A more accurate term would be “oral solid dosage medication.” This means Pillbox does not contain suppositories, injected, inhaled, or dissolvable medications, and prescription lotions. For the sake of discussion, it’s common to use the word “pill.” There are currently no plans to expand Pillbox beyond oral solid dosage medications, but we hope that through sharing the code and processes that create Pillbox, others will be able to dig deeper into the source data sets and build systems with even more power than Pillbox.
Pillbox data is a combination of two data sources – the FDA’s Structured Product Labels (what most of us call the drug labels) and NLM’s RxNorm. It does not contain the full data set from either source. Rather, we parse data specific to solid oral dosage medications from both and combine them to create Pillbox.
-SPL overview
-RxNorm overview
-unique identifier for each pill (SETID, NDC9, part number)
Labels may contain more than one product, identified in Pillbox by the NDC9. Products may contain different pills, as in the case of birth control pills.
-Data issues
-FDA guidance that isn’t followed
-limits and lack of validation rules
-duplicate and versioning issues
-errors (omission and commission)
-lack of normalization, controlled vocabularies
-standardized vocabularies for inactive ingredients
-Data analytics report (separate wiki?)
This overview is not meant in any way to supersede the FDA’s Structured Product Labeling Resources. It is intended to provided a rough overview of how solid oral dosage medication data is organized in Pillbox and to help developers leverage that data.