Safe use of electronic health records and health information technology systems: trust but verify.

OBJECTIVES: We will provide a context to health information technology systems (HIT) safety hazards discussions, describe how electronic health record-computer prescriber order entry (EHR-CPOE) simulation has already identified unrecognized hazards in HIT on a national scale, helping make EHR-CPOE systems safer, and we make the case for all stakeholders to leverage proven methods and teams in HIT performance verification.
METHODS: A national poll of safety, quality improvement, and health-care administrative leaders identified health information technology safety as the hazard of greatest concern for 2013. Quality, HIT, and safety leaders are very concerned about technology performance risks as addressed in the Health Information Technology and Patient Safety report of the Institute of Medicine; and these are being addressed by the Office of the National Coordinator of HIT of the U.S. Dept. of Human Services in their proposed plans. We describe the evolution of postdeployment testing of HIT performance, including the results of national deployment of Texas Medical Institute of Technology's electronic health record computer prescriber order entry (TMIT EHR-CPOE) Flight Simulator verification test that is addressed in these 2 reports, and the safety hazards of concern to leaders.
RESULTS: A global webinar for health-care leaders addressed the top patient safety hazards in the areas of leadership, practices, and technologies. A poll of 76 of the 221 organizations participating in the webinar revealed that HIT hazards were the participants' greatest concern of all 30 hazards presented. Of those polled, 89% rated HIT patient/data mismatches in EHRs and HIT systems as a 9 or 10 on a scale of 1 to 10 as a hazard of great concern. Review of a key study of postdeployment testing of the safety performance of operational EHR systems with CPOE implemented in 62 hospitals, using the TMIT EHR-CPOE simulation tool, showed that only 53% of the medication orders that could have resulted in fatalities were detected. The study also showed significant variability in the performance of specific EHR vendor systems, with the same vendor product scoring as high as a 75% detection score in one health-care organization, and the same vendor system scoring below 10% in another health-care organization.
CONCLUSIONS: HIT safety hazards should be taken very seriously, and the need for proven, robust, and regular postdeployment performance verification measurement of EHR system operations in every health-care organization is critical to ensure that these systems are safe for every patient. The TMIT EHR-CPOE flight simulator is a well-tested and scalable tool that can be used to identify performance gaps in EHR and other HIT systems. It is critical that suppliers, providers, and purchasers of health-care partner with HIT stakeholders and leverage the existing body of work, as well as expert teams and collaborative networks to make care safer; and public-private partnerships to accelerate safety in HIT. A global collaborative is already underway incorporating a "trust but verify" philosophy.