Analytics: A possible way to domesticate wild health data

Analytics offer some parallels to the rise of agriculture about 12,000 years ago, according to Brian Dixon, assistant professor of health informatics at Indiana University and research scientist with the Regenstrief Institute. With new standards and tools, the days of hunting and gathering data are over. All that unstructured information sitting in healthcare data centers can be tamed and put to work, he says in an article published at

"Normalizing" raw patient data by mapping it to LOINC (Logical Observation Identifiers Names and Codes) and SNOMED CT (Systematized Nomenclature of Medicine--Clinical Terms), using natural language processing and tools such as the Notifiable Condition Detector can harness data for research--not just storing it for regulatory purposes.

Just last week, Regenstrief Institute experts took the lead in a discussion paper released by the Institute of Medicine advocating using data from everyday doctor visits to improve healthcare for the population at large.

The article focuses on six analytics use cases for healthcare IT. In a move toward evidence-based medicine, Boston's Beth Israel Deaconess Medical Center, for instance, is rolling out a smartphone app to offer healthcare providers access to 200 million data points about two million patients. It's encoding physician free-text notes to SNOMED CT, which makes them more searchable.

The article also advocates pressing vendors for their strategy with service-oriented architecture (SOA), which enables linking contemporary data sets to legacy IT architecture. Mark Dente, managing director and chief medical officer for MBS Services, says that SOA provides the capability to host today's data sets with those to come – and that organizations might not yet realize they need. Be wary if a vendor doesn't have an SOA strategy, Dente says.

Oregon Health & Science University this week announced a partnership with Intel to create "next-generation computing technologies" for sorting through reams of biomedical data in a search for genomic clues to cancer. The New York Times noted a similar "arms race" against cancer at sites including Mount Sinai's medical school, Memorial Sloan-Kettering Cancer Center and Harvard Medical School.

Still, translating between standards and databases remains difficult. Massachusetts General Hospital computer scientists this week noted the complexity of translating between the Health Quality Measures Format (HQMF) and Informatics for Integrating Biology and the Bedside (i2b2), a task they found incredibly time consuming.

To learn more:
- find the article


Fuel Top Line Growth and Increase Membership

In this webinar, payers can learn how to accelerate the sales pipeline and grow membership across all health insurance market segments using comprehensive sales technologies.