Harnessing (and Securing) Meaningful Health Data

By Allison Proffitt

May 3, 2010 | “There isn’t going to be some massive database in the basement of the White House run by Sarah Palin,” promised John Halamka, the CIO of Harvard Medical School, in his keynote at the Bio-IT World Expo. But there are going to be changes in health care IT, and there will be some “federated mechanism that enables us to send data from place to place for a whole variety of purposes for care and research.”

Halamka serves as the Chair of the US Healthcare Information Technology Standards Panel. Of the $30 billion allotted to health care IT in the Obama Administration’s stimulus package, most of it will be distributed to hospitals and clinics after they’ve put health care IT infrastructure in place and are using it wisely. The remaining $2 billion is being distributed by the Office of the National Coordinator for health care IT advances.

“Here’s the strategy,” Halamka said. “Give $2 billion in grants to accelerate the industry. Give the industry a set of standards that are unambiguous for everything from medications, to labs, to quality measurements, to both clinical care and population health… Declare how hospitals and doctors have to use this wisely, and then certify products as being good enough to have the features and functions and capabilities to make this whole thing work.”
The foundations of the efforts are appropriately-collected data and adequate encryption to secure that data. In the next five years as these standards are put into place, doctors and hospitals will be required to collect “meaningful” data and protect that data.

“This is not using a word processor to record data!” Halamka clarified. “This is actually using codified mechanisms so that if you capture medications, problems, allergies, labs, etc. You could use them to inform drug discovery.”

Meaningful data include requiring all orders to be electronic; recording medication and allergy lists for all patients; and recording and updating demographics and vital signs in a timely manner—all using consistent and controlled vocabularies.

There are rules that still need to be clarified, Halamka says. For instance, do you continue to measure height every two months once a patient is 25? What would be the decision support rules for doctors considering ordering expensive radiology images? What is the appropriate mix of structured and unstructured data?

But gathering this meaningful data will immediately begin to enable smarter health care. Offices and hospitals will be able to mine their data and send targeted wellness reminders for preventative care. Patients will have access to a continuity of care document or record, delivered electronically, explaining treatment history, prescriptions, and diagnoses. By 2015, doctors will be required to record allergies down to the exact substance, food, or environmental factor to which the patient reacts.

But with this wealth of data—and wealth of possibilities—comes a huge security responsibility.

“I spend about $1 million a year just protecting the Beth Israel Deaconess records against the nefarious internet. We’re attacked every seven seconds, 24 hours a day, seven days a week,” Halamka said. “Half of the attacks come from Eastern Europe; half of the attacks come from Eastern Cambridge. Every September, 1200 new hackers arrive – they’re called freshmen!”

Jokes aside, Halamka stressed that laptops and thumb drives must be encrypted. Patient privacy must be protected.

Emerging Data
Although about 20% of clinics and hospitals currently have electronic health records, Halamka said all should by 2015. So what will we be able to do with all of that “meaningful” data? Applications are already beginning to emerge. Health data gathered can be aggregated regionally to look at public health trends or build doctor report cards. Google Trends has already predicted an H1N1 outbreak based on regional search terms weeks before it was detected by public health officials.

The SureScripts (representing pharmacies) and RxHub (representing payors) database includes de-identified drug information on 160 million people that can be used to check for drug interactions. And in another example, the Social Security Administration used to spend $500 million a year getting paper records. After moving to electronic records two years ago, a disability claim that took months to adjudicate can be handled in 48 hours.

There are even opportunities to gather rich patient data in the home. Halamka is testing a bathroom scale that calculates his lean body mass and body mass index and transmits the data via XML in real time to Google Health and Microsoft Health. But although an avid blogger and registered technophile, he said he declined the Twitter reporting feature.