After weeks of listening to some very educated and experienced Wharton MBAs involved in the big data space, I am convinced that big data may not actually be that big at all. This isn’t meant to downplay big data, but to stress the importance of identifying the appropriate data within the bigger data in order to make good decisions.

In health care, these decisions are necessarily made where care and its payment intersect. That is where health care analytics companies will provide true value.

1. Analytics needs to be actionable for a recommended intervention for a clinician, for instance, and at the point of care.

2. The information needs to be in a form that is easily accessible and useful to the user—not in a form that requires a lot of work for a busy clinician to figure out.

3. The data needs to answer a question or solve a problem a clinician is having. An example is: What is the recommended care for a patient who has been readmitted twice in the past three weeks for congestive heart failure and whose lungs continue to fill with fluid, despite what is believed to be appropriate home or therapeutic care?

As clinicians and care givers become increasingly at risk financially for patient populations, cost-effectiveness must become a major consideration.

4. The integrity, reliability and security of the data need to be ensured. If clinicians do not have confidence that the data is correct, they in turn rely less on the programs that assemble it.

5. Lastly, and likely most importantly, big data needs to be easily translatable if it is to become a truly ubiquitous tool.  Because data comes in different forms from different sources, a “common” language is going to be required. The question of who takes ultimate responsibility for this language in health care is extremely important. While the federal government appears to be taking the lead on this, it is very slow going.

Presentations by W. Michael Long, CEO of Lumeris, at our health care alumni conference (courtesy of Wharton Health Care Management Alumni Association board member, Jeff Smith, WG’99, among others) and at the Health Informatics meetings held at the offices of Duane Morris in New York (courtesy of Peter Fishman, WG’07, and Mitch Goldman, WG’75,  along with panelists Dr. Steve Davidson, WG’89; David Wierz; and Ryan Cochran, W’07) provided these very good insights into the ingredients for success of big data companies.

Those health care big data companies that are addressing questions at the point of care, providing a complete and integrated picture of what is occurring, delivering just the needed amount of analytics, and ensuring appropriate oversight will be the ones that likely provide significant value to the health care system and emerge as winners.

This is a very interesting race and space with no clear front runners yet.  Stay tuned.