From May 13-14, I spent some time at Health:Refactored. Apart from catching up with colleagues and kindred spirits, the event was a good learning venue for healthcare IT developers and designers.
I’m going to broadly capture my notes under data, data analysis (big data), Health UI, and Health API platforms.
Bryan Sivak did a keynote highlighting various Health Data and entrepreneurship programs available at HHS. Among the various data sets, the ones which he highlighted in the talk is NPPES provider database. The update frequency of pillbox data sets was questioned by the audience. By the way, health data consortium maintains a ranked repository of 50 top health related data sets – http://www.healthdataconsortium.org/data-sources and Socrata (http://dev.socrata.com/) offers API access to public data sets. A good place to know more about the HHS lead datasets and initiatives is an upcoming conference – HealthDataPalooza! More on HealthDataPalooza soon, but the conference is doing a half-day data analysis workshop and providing attendees with a license to Archimedes population explorer (videos) and the CMS claims data set (CMS DE-SynPUF). By the way, you should check out the winners of Data design diabetes innovation competition at – http://www.redesigningdata.com/.
In addition to “data” discussion there was talk about “trust data”. Its important to have appropriate data in order to establish a distributed trust ecosystem. A few folks talked about National Strategy for Trusted Identities in Cyberspace (NSTIC) pilots.
Data Analysis (Big Data)
Various data scientist stressed on the need to get started with data analysis with minimalism (you can’t hire quality data scientists – as most of them already work at Facebook, LinkedIn etc.). Following set of tools might be useful:
- Motardata – (http://app.mortardata.com) a “heroku” for deploying and developing Apache Hadoop.
- Spark (http://spark-project.org/) and Shark (http://shark.cs.berkeley.edu/) – these are upcoming faster technologies which are compatible with Hadoop
- R – http://www.r-project.org/
- NLTK (Python) – http://nltk.org/, a natural language processing toolkit which provides easy interface with Wordnet.
- Big Data platforms – Metascale, Teradata, Netezza, Greenplum etc.
Healthcare IT user interfaces and customer interactions need to evolve. The term which stuck with me from the conference was – “lead with love”. Compassion should show through user interface but more so the interface should be enable users to personalize, share, celebrate, and should provide value and voice – according to one of the panelists. I met with folks from various design companies including ideo, madpow etc. It was good to see familiar faces.
Health API Platforms
A personal favorite of mine! A number of strong folks showed APIs at the conference, following are pointers to a few which caught my eye:
- Walgreens API – Announced recently – https://developer.walgreens.com/ (Prescriptions and Rewards API)
- NYeC – A statewide health API – http://nyehealth.org/what-we-do/api-innovation/
- athenahealth – AthenaHealth Marketplace (announced recently): http://www.athenahealth.com/marketplace/
- allscripts – ADP – http://www.allscripts.com/en/company/partners/adp-faq.html
- aetna carepass – https://developer.carepass.com/ (Zipongo announced their carepass connection)
A more detailed analysis of various go to market approaches of various health API platforms would be a candidate for another post. A quick look at above list might leave the reader pondering how are various health care participants like retail/pharmacy (walgreens), government/states (NYeC), electronic medical records (athenahealth, allscripts) and payers (Aetna) extending their markets with data or rewards or “buzz word value” plays using an application programming interface!