Physicians and other care providers commonly report deep dissatisfaction with electronic health records (EHRs). Why? We asked two pathologists with expertise in medical informatics to share their experiences and ideas.
Michael Astion, MD, conducted this interview.
There is a significant body of literature on the patient safety risks associated with EHRs. What are some of the bigger risks and how do they relate to laboratory services?
Medication ordering and administration errors are the most common source of patient harm associated with EHRs. In many cases, the root cause can be traced to a system that makes it easy to do the wrong thing, by presenting too many choices or presenting choices that are easily confused. These kinds of errors occur frequently when ordering lab tests. In general, the risk of harm associated with ordering or administering the incorrect medication is higher than incorrect test ordering, so the pharmacy gets a little more attention than the lab (1).
What are some of the specific problems related to laboratory services and EHRs?
Some general, interrelated themes in patient safety cases include information overload, alert fatigue, copy-paste errors, excess complexity, and automation of suboptimal practice. Table 1 lists some examples in which EHR design or implementation problems have enabled provider errors. Problems in EHR design or configuration are a deep source of dissatisfaction to physicians. Making errors is upsetting, and traversing the EHR without error is exhausting. In addition, time spent away from the patient to interact with the computer screen leads to a values conflict, since patient interaction is what many physicians value the most.
How frustrated are physicians?
The design of EHRs often leads care providers to conclude rightly that these systems—including cumbersome lab test ordering and resulting processes—are diverting time away from both patient interaction and higher-level cognitive activities. Furthermore, a variety of studies point out that providers are frustrated by the mismatch between EHR design and clinical workflow, and that they have experienced harmful errors as a result. Perhaps most of all, physicians are frustrated by the amount of additional work EHRs create for them, either that staff would have performed in the past or which represents new effort introduced by EHRs.
What risks in lab services associated with EHRs are of most concern to you?
Failure to retrieve both routine and critical results might be the most pressing safety risk. Routine results can easily be missed as a result of the low signal-to-noise ratio in the lab results section of the EHR. Our confidence is so low in the ability of EHRs to communicate abnormal results that are not immediately life-threatening that many labs use alternative mechanisms to promote results retrieval. These include patient portals and lists of abnormal results that have to be called to providers in a specified time period. For critical results, EHRs have not reduced the need for the person-to-person communication required under CLIA.
How can lab leaders help improve EHRs?
Laboratory leaders should advocate for their institutions to use the Safety Assurance Factors for EHR Resilience (SAFER) Guides (www.healthit.gov/safer/safer-guides) provided by the Office of the National Coordinator for Health Information Technology to evaluate their electronic systems. The SAFER guides provide good direction for evaluating and improving safety in the EHR, and specifically address computerized physician order entry (CPOE), results reporting and follow-up, patient identification, and clinical communication. Laboratory staff and leadership need to work closely with their information technology colleagues in evaluating systems, as knowledge of specific laboratory processes and workflows is required to perform an adequate assessment.
I know that informaticians are prisoners of hope. Can you give us some examples of why we should be optimistic about how future improvements in EHRs will lead to better lab services?
Despite the drawbacks we’ve discussed, EHRs have clearly had their share of successes as well, and we hope that they will continue to improve. The most obvious reason for optimism, because it is achievable and helpful, is non-intrusive clinical decision support for both ordering and interpreting results. Among institutions where the lab participates in building order sets and improving result display, this has already enhanced test orders significantly.
One example of a more speculative improvement might be expert systems to prioritize and synthesize voluminous or complex results. Such systems could increase the value of laboratory testing by emphasizing important data and placing it into the appropriate context for each patient. In addition, expert systems for communicating or escalating important results could help overcome information overload.
Another exciting possibility would be the development of new EHR functionalities and/or workflows that facilitate direct communication between clinicians and both lab staff and pathologists. This could promote teamwork between specialties, one of the major goals of the recent National Academy of Medicine report, Diagnostic Error in Health Care. Improved support for communication could enable the lab to provide its expertise in test selection and interpretation. The value of the EHR in this case could be in routing questions to appropriate personnel based on urgency and context. In addition, the EHR could help the lab detect and respond to ordering errors proactively.
What will it take to realize some of the hopes and dreams you have described?
Improvements are unlikely until incentives are aligned between EHR vendors and users. Unfortunately, it is extremely difficult to switch EHRs and we cannot—at least in the short run—rely on market forces to drive change. One avenue for improvement might be specific measures of usability to accompany, or in many cases replace, existing metrics for evaluating meaningful use of clinical systems. Another opportunity comes from emerging EHR communication standards such as Fast Health Interoperability Resources (FHIR) that allow institutions and/or vendors to develop “apps” that can communicate with EHRs. In addition, if labs, providers, and vendors collaborate to translate continuous quality improvement activities into design, efficiency, and workflow improvements inside the EHR, efficiency gains are certainly within reach.
- The Joint Commission. Investigations of Health IT related deaths, serious injuries, or unsafe conditions. https://www.healthit.gov/sites/default/files/safer/pdfs/Investigations_HealthIT_related_SE_Report_033015.pdf (Accessed March 30, 2015).
Noah Hoffman, MD, PhD is associate professor and associate director of medical informatics at the University of Washington Department of Laboratory Medicine in Seattle.
Patrick Mathias, MD, PhD is a clinical informatics fellow at the University of Washington Department of Laboratory Medicine in Seattle.
CLN's Patient Safety Focus is sponsored by ARUP Laboratories