Electronic health record (EHR)-related errors in the ED resulted in more significant patient harm than errors that occurred in the inpatient or ambulatory settings. To reduce risks, EPs can:
- configure the ED’s EHR with end users’ input and in ways that improve care;
- know who to call for help if there’s a problem with the EHR;
- probe further if information about an ED patient seems incorrect.
Electronic health record (EHR)-related errors in the ED resulted in more significant patient harm than errors that occurred in the inpatient or ambulatory settings, according to a recent analysis.1
Seeking to obtain more information on health IT-related problems, researchers analyzed EHR-related cases occurring between 2011 and 2015 submitted to CRICO Strategies’ Comparative Benchmarking System, a database of malpractice cases. Researchers found 420 cases in which the EHR was a contributing factor. Of this group, 50 cases occurred in the ED.
“What’s interesting about the ED cases is that 57% of them ended up in a high-severity outcome,” says Penny Greenberg, MS, RN, CPPS, editor of the report. This compares to 45% in the inpatient setting, and 39% for the ambulatory setting. Greenberg is senior program director of patient safety services for CRICO Strategies, a division of CRICO, which provides medical professional liability coverage for the Harvard medical community. As a group, the EHR-related ED cases cost $18 million to resolve. Here are some case examples:
- A 48-year-old woman’s X-ray was read by the EP as normal, but the radiologist’s report the following day revealed an incidental finding of a pulmonary nodule and recommended a chest CT scan.
“The EHR did not flag the result for the EP to look at it, and it wasn’t reported to the patient,” Greenberg says. Two years later, the patient discovered she had advanced lung cancer and sued, alleging that the cancer should have been caught at the time of the ED visit. The case was settled.
- A man presented to an ED with severe substernal chest pain radiating to the left chest and shoulder. The first troponin level, the CT scan, and the chest X-ray were all normal, and the man’s pain decreased after he received morphine.
The plan was to admit the patient, who was moved from the acute section of the ED to the urgent section because of crowding. This affected the way in which the second troponin level, which came back high, was documented in the EHR.
“Neither the EP nor the ED resident looked up the value, or made the diagnosis of [myocardial infarction], because the value was in the ‘acute’ results section in the EHR, instead of the ‘urgent’ section,” Greenberg explains.
The patient coded and could not be resuscitated. The case was settled for $1.2 million.
- A 39-year-old woman with a history of migraines presented to an ED with severe pain, which resolved before discharge. The following day, she returned with vomiting and worsening pain.
The EP ordered 50 mcg of fentanyl, to be given 25 mcg at a time. However, the EHR was set up for single doses of fentanyl of 25 mcg, or ranges of 50 mcg to 100 mcg. The EP entered a 50 mcg order into the EHR, but did not specify this should be given as two single doses of 25 mcg. “The EHR automatically translated the EP’s order from 50 mcg to a range of 50 mcg to 100 mcg,” Greenberg explains.
Based on the severity of the patient’s pain, the ED nurse administered the higher dose of 100 mcg of fentanyl to the patient. After the fentanyl was administered, the patient used the bathroom, and the pulse oximetry was likely disconnected at this time.
Shortly afterward, the patient was found unresponsive. Resuscitation was successful, but a CT revealed an anoxic brain injury. The patient was removed from life support and expired. The case settled for $450,000.
- A patient with a history of unresectable brain meningiomas was transferred to an ED for evaluation of swelling and bleeding from one of her brain masses. She also reported a headache and left-sided facial numbness.
The ED resident evaluated the patient and prescribed the steroid dexamethasone, selecting the “taper and take” fields in the EHR. The “take” field automatically defaulted to a “1” in the printed prescription.
“The ER resident didn’t know this would occur,” Greenberg says. “It was unclear if this was included in EHR training.” The resident verbally reviewed instructions on how to taper the medication with the patient. “However, the first line of the written instructions, due to the EHR default, were to take ‘1 tab by mouth 1x daily for 35 days,’ which was not intended,” Greenberg notes.
The instructions should have read, “Take 4 tabs q 6 hours x 7 days.” The patient was discharged home, and took the medication according to the incorrect instructions.
Shortly afterward, the patient returned to the ED with continued headaches and left-sided facial numbness. “Her facial droop was unchanged from the prior visit. It was discovered that the patient was not taking the correct dosage of Decadron,” Greenberg says.
The problem was that the EHR default error resulted in the patient’s steroid prescription being too low. “The patient’s symptoms improved once she was taking the correct dose,” Greenberg adds. The case settled for $1,700.
One of Many Factors
It’s easy to simply blame the EHR for the adverse outcomes. However, for every one of the malpractice claims in the database, the EHR was only one of multiple contributing factors. “It’s never just the EHR. Adverse events happen generally because of many factors colliding at once,” says Trish Lugtu, CPHIMS, senior manager of advanced analytics at Minneapolis-based MMIC, a Constellation company.
Assessment failures by the EP, or lack of communication between providers about the patient’s condition, also are to blame.
“The EHR is intended as a tool to support the care team,” Lugtu says. “It is not meant to replace critical thinking or sound medical judgment. It can’t take the place of the provider.”
Malpractice cases involving EHRs close faster than the average claim, Lugtu explains. “After all, EHR is documentation. Wrong or right, it provides attorneys with immense documentation and built-in time stamps.”
Lugtu says that problems with EHRs generally fall into a few simple categories: Information is not seen, information is wrong, or information is missing. Therefore, EPs’ training is very important. “The number one thing is to get emergency providers engaged, to collaborate on workflow design, and to learn how to use the EHR well,” Lugtu offers.
Many EDs have yet to take advantage of the EHR capabilities that can support improvement of care. For instance, automated reports can provide audits of incomplete tests or undelivered results open at the time of discharge. “Not only is it important to configure the EHR well to avoid mishaps, but there is such a great opportunity to leverage the EHR to improve care,” Lugtu notes. Greenberg recommends EPs use these practices to reduce EHR-related risks:
- Know who to call for help if there’s a problem with the EHR;
- Include questions about potential EHR involvement in every event report and root cause analysis;
- Combine information from root cause analyses and incident reports with medical malpractice data;
- Be sure you are in the right record before documenting;
- Use critical thinking.
“If there is information that doesn’t seem correct, ask questions. If the data are not accessible, that does not mean it’s not available,” Greenberg says.
- Greenberg P, Ruoff G. Malpractice risks associated with electronic health records. CRICO Strategies, June 13, 2017. Available at: . Accessed Dec. 4, 2017.
- Penny Greenberg, MS, RN, CPPS, Senior Program Director, Patient Safety Services, CRICO Strategies, Boston. Phone: (617) 450-6850. Email: PGreenberg@rmf.harvard.edu.
- Trish Lugtu, Senior Manager, Advanced Analytics, MMIC, a Constellation company, Minneapolis. Phone: (952) 838-6844. Email: Trish.Lugtu@ConstellationMutual.com.