Data Design Diabetes

In their own words...

Data Design Diabetes Question: LetŐs start at the beginning. How did your story begin? What was the catalyst for your solution?

MediSapien Answer: ZyDoc chairman James Maisel, MD, a practicing retinal specialist for over 30 years, has worked with thousands of diabetic patients, as well as numerous diabetic educators, endocrinologists, internists, nephrologists, and other diabetes health professionals in dealing with real world diabetic problems. Realizing first-hand the seriousness of diabetes and its complications, he was motivated to work with other practitioners and organizations to find better ways to prevent, treat, and control the disease.

In 1993, after founding ZyDoc, Maisel and his team saw the potential of using speech recognition and NLP to capture, structure, and share data from patient encounters. The vision of MediSapien was born as a web-based platform that would be designed to facilitate interoperability and enable access to rich datasets for healthcare givers and application developers. Diabetes presented an ideal challenge to see if MediSapien could provide structured data useful for specific disease management from the millions of unstructured transcribed records as a prelude to using an organizationŐs records for diabetic management.

This yearŐs challenge is distinct from previous years in its focus on the intersection of data and healthcare. How has access to data empowered your team and shaped your approach?

Given that MediSapien's prime mission is unlocking hidden data, the emphasis of this year's Data Design Diabetes presents a timely opportunity for us to work with the diabetes care community regarding the use of previously unavailable data to further patient care and healthcare knowledge.

MediSapien enables the extraction of highly granular detail from the individual patient content-rich Ňhealth story,Ó and allows the aggregation of data from patient population segments. Rather than being an end-user of data, MediSapien is focused on generating, storing, and supplying structured data that can be consumed and shared by the healthcare community, including diabetes caregivers, and knowledge workers such as researchers and application developers.

Who are some of the game-changing people and companies in health tech that you most admire? How have they caught your eye?

The essence of the game-changing people that we admire (info borrowed from web sources) includes Drs. Lawrence Weed, Carol Friedman, and Farzad Mostashari. Dr. Weed is known as the Ňfather of the problem-oriented medical record (POMR),Ó which he developed along with subjective, objective, analytical, and planning (SOAP) progress notes. The POMR is a key concept in medical informatics, providing a way to audit medical records by focusing not only on what was done, but why. The POMR and SOAP notes have been adopted by countless healthcare professionals and major medical institutions.

Dr. Friedman is a luminary in clinical NLP and the developer of MedLEE, a widely used and studied NLP engine. Dr. Friedman is Professor of Biomedical Informatics at the Columbia College of Physicians and Surgeons, Columbia University.

is the National Coordinator for Health Information Technology at the U.S. Department of Health and Human Services. To this role, Dr. Mostashari brings a rich body of experience to continue leading the nation in constructing a healthcare system that uses information and technology to empower individuals, and improve both the health and healthcare of Americans.

The companies or organizations we admire are Healthcare Open Systems and Trials (HOST) and its merger organization, HIMSS. HL-7 has also done very much to promote standards in the medical informatics industry. Also, the College of American Pathologists (CAP) should be admired for its work with SNOMED CT that has empowered medical informatics. "Systematized Nomenclature Of Medicine Clinical Terms" (SNOMED CT) is a reference terminology standard from the Unified Medical Language System (UMLS). SNOMED consists of concepts, terms, and the interrelationships between them. It standardizes the way healthcare terminology and data is recorded. It aims to facilitate the coding, retrieving, analysis, aggregation, indexing, and the exchanging of clinical information across health care entities.

If they call your name on June 4th as the challenge winner, how will you make use of the $100,000 prize?

Since our objective is to partner with companies and providers within the diabetes healthcare community, itŐs an easy call to make in deciding how the use the $100,000. Our MediSapien platform functions as a middleman between data providers (those who produce or store unstructured text) and those who will make use of the structured data once MediSapien has performed its functions. Many of the downstream users of the structured data produced by MediSapien will be fed such data through various channels; the money would be used to create an efficient and easy-to-use means of data exchange, so MediSapien can be used by the full variety of members of the community.

We had originally planned to feed the structured data via the new healthcare standard documents (CCRs, CCDs and other CDAs) - our experience to date is that potential recipients of such data are not equipped to handle these. Therefore, we would use the $100,000 to create one or more new interface alternatives, that we would tailor based on discussions with members of the community, particularly some of the startup companies and small vendors who are creating innovative applications. This is not to say that we would avoid the needs of larger players (e.g., an ACO); for these more sophisticated players, creating a standard web services infrastructure may be quite appropriate. But for the smaller players, we would seek to create a data exchange approach commensurate with their technology level and development pocketbook.