Clearing the health-records hurdle

The U.S. avoided a major domestic public health crisis from the Ebola virus. But the attention paid to the possibility of an epidemic at home got policymakers thinking about how information sharing via electronic health records can help public health responders slow the spread of deadly bugs like Ebola.

Karen DeSalvo, the National Coordinator for Health IT, who is also leading the U.S. response to Ebola as acting assistant secretary of health in the Department of Health and Human Services, said that the possibility of transforming public health is one of the "hidden treasures of health information technology," in a Health IT Policy Committee meeting convened to examine how better EHRs can power responses to epidemics.

There are significant hurdles to clear before the EHRs used in clinical care will be able to really help state, local, and federal health officials track and respond to fast-moving outbreaks in real time. The problem of interoperability and data transfer between EHR systems, medical laboratories, and public health databases is one big issue. More broadly, there is a lack of what experts call "bidirectionality" between health records, so that health officials are prevented either for technical or privacy reasons from accessing individual patient records.

For example, New York City is trying to get a handle on the spread of the chikungunya virus, a mosquito-borne disease that travelers from the Caribbean are increasingly bringing to the United States. Dr. Annie Fine, an epidemiologist with the New York City Department of Public Health, noted that patient addresses are often omitted from test results for the virus that are filed with the city as a matter of law. That makes it harder for public health workers to follow up with infected individuals. Bringing disparate systems that were never designed to work together can be problematic, Fine said. "There are opportunities for failure and data problems at each integration point," she said.

Dr. John Loonsk, chief medical information officer at CGI Federal, echoed this point, noting that while physicians and labs are required to report to state and local officials on identified public health threats, sometimes key clinical information is omitted from lab reports. Underreporting can compound the problem, Loonsk said, citing data from the Centers from Disease Control showing that as few as one in 10 cases of Lyme disease were reported to authorities.

Existing EHRs can be modified to respond to ongoing health crises to support clinical care. In the Ebola situation, EHR vendor Cerner was able to insert clinical questions into its basic emergency room intake interface, to get health workers to ask patients about travel to affected regions and relevant symptoms. Information on exposure and symptoms were used to recommend heightened clinical responses, such as isolation or special care among health workers, where appropriate.

One big takeaway, said Brian Clark, Cerner's GM for emergency solutions, is that CDC and other government content on ongoing epidemics is scattered, and hard to aggregate inside an EHR for use by clinicians. He suggested a "standards-based surveillance app" that could operate on any certified EHR system to supply up-to-date government information and guidance on disease outbreaks.

Janet Hamilton, an epidemiologist with the Florida Department of Health, also noted that better electronic records exchange will minimize the time spent by doctors and health professionals reporting basic facts to officials, and maximize the time they spend treating patients.

Ultimately, broader use of EHRs to detect and respond to epidemics will require changes in technology, but perhaps also changes to health privacy rules at the state and federal level. The passive surveillance of patient EHRs using analytic tools could give greater velocity to detecting not just viral disease outbreaks, but environmental risks, contaminated food and medicine, and other large-scale health problems that are clustered geographically or in certain demographic groups.

That’s not to say epidemiology is lacking in high-tech approaches. New York City, for example, was able to use cell phone location information and subway fare card data to conduct contact tracing on individuals that may have come into contact with the Ebola virus while traveling. However, aggregating that information, and making it available at scale through an EHR platform, appears to be a long way off.