How Safe Are Electronic Health Records?

By Jane Farrell

Four years after publication by the Office of the National Coordinator for Health Information Technology (ONC), voluntary guidelines designed to increase the safety of e-health records have yet to be implemented fully, according to a new survey.

The findings appeared recently in the Journal of the American Medical Informatics Association.

“Less than 20 percent of the recommendations were fully implemented across all the organizations,” said Dean Sittig, Ph.D., the study’s lead author and a professor at UTHealth School of Biomedical Informatics.

Developed by Sittig and Hardeep Singh, M.D., of Baylor College of Medicine, the Safety Assurance Factors for EHR Resilience or SAFER guides entail 140 unique recommendations separated into nine separate guides.

To see if health care organizations willingly implemented the recommendations in the guides, Sittig, Singh and their colleagues asked eight health care organizations in the United States and Australia to conduct a self-assessment. The results were presented anonymously.

Broadly divided into three domains, the recommendations in the “safe health IT (information technology)” domain had the highest adherence rate followed by the “using health IT safely” domain and the “monitoring health IT” domain.

“This is not surprising because the domains were conceived as sequential building blocks,” said Sittig, who is on the faculty of the UTHealth Memorial Hermann Center for Healthcare Quality and Safety. “The safe health IT domain contains many recommendations required for e-health record system certification.”

Examples of the recommendations in the “safe health IT” domain include “Data and application configurations are backed up and hardware systems are redundant” and “EHR downtime and reactivation policies and procedures are complete, available and reviewed regularly.”

While the researchers did not evaluate why the adherence rates varied among the health care organizations in this survey, they speculated that the differences could be related to budgets, personnel skill mix and organizational priorities.

Sittig said the results could be used by other organizations to benchmark their progress toward achieving a safe and reliable e-health record.