By Charles A. Pilcher MD FACEP
November, 2016
Last year Dr. David Troxel of The Doctors Company published an analysis of the role that the Electronic Health Record (EHR) has played in medical malpractice claims. There are few surprises in this report, or an earlier report on the topic by Dr. Troxel, for the experienced med mal attorney, but is presented here for those readers who may find it useful as a reference when evaluating cases of alleged medical negligence involving an EHR.
In his analysis Dr. Troxel first acknowledges that med mal insurers were early to recognize the potential liability risks inherent in EHR’s. For the last 3 years his company has been coding various system and user factors related to EHR’s in their closed claims data. Using retrospective data, they found 97 cases dating back to 20o7 in which the EHR played a role, a rate of just under 1% of the company’s closed claims during that time. User factors were involved in 64% of the claims and system factors in 42%.
System factors included such things as poor design, decision support issues such as alerts, various electronic problems and insufficient scope. Examples of these were:
- The lack of an EHR drug alert when a female patient received “Flomax” instead of “Flonase” after the physician entered only “flo-” as he was searching the prescription database. There was no EHR alert that Flomax was not approved for females.
- A dialysis patient was transferred to a SNF with a continuing order for Lovenox (heparin analog). Because of the lack of a “high risk medication alert” for Lovenox, the patient exsanguinated and died after profuse bleeding during both her first and second dialysis episodes.
- A patient’s bladder was punctured during a proctored “sling” procedure. Although the proctor had approved the sling’s placement, there was no place in the EHR to document the proctor’s presence in the OR by using either a free text narrative or an item in a drop-down menu .
- A child returned from a country with endemic tuberculosis, was evaluated for fever and treated for influenza. The physician documented an absence of tuberculosis exposure, and the system allowed that entry to be copy/pasted into subsequent visit notes without reconsideration or correction. He eventually was diagnosed with tuberculous meningitis.
User factors included entering incorrect information, conversion problems between platforms, pre-population and copy/paste issues, training and other user errors. Examples of these were:
- A physician inadvertently checked a drop-down menu option for morphine 200 mg when a 15 mg dosage was intended, resulting in an ER visit and an overnight hospital stay.
- A pregnant diabetic patient presented in labor and delivered a 4640 gram baby with a shoulder dystocia and brachial plexus injury. The physician, who had planned a C-section if the baby was over 4500 grams, could not access the report of the ultrasound he had done the day before because of lack of both training and a password for the hospital’s EHR.
- A CT scan was done in the ER for a patient with abdominal pain. The consulting gynecologist, who had not been trained on the hospital’s new EHR system, inadvertently viewed the report of a 2 year old scan showing an ovarian cyst and took the patient to surgery. MRSA developed in the wound. No cyst was found because it had been removed 2 years earlier and the current CT scan had not been viewed.
Additional findings of the study included:
- 43% of the events occurred in a clinic or doctor’s office
- 20% of events were for Internal Medicine subspecialties, 16% in primary care, 15% in OB/Gyn, and 14% in Surgery. Surprisingly, only 2% occurred in the ER.
- 27% of allegations were related to diagnosis (failure, delay or wrong) and 19% were medication related
Of note, the Institute of Medicine in 2011 recommended creating a government agency to investigate harm and safety events related to Health Information Technology (HIT). This is apparently still under development.
Dr. Troxel had previously written about the risks and benefits of EHR’s. See the second quarter issue of The Doctor’s Advocate. Among his findings at that time were the following (quoting from the current 2015 report):
- “EHRs improve medical record documentation and legibility. However, point-and-click lists, drop-down menus, templates, canned text, and autopopulation of data fields from personalized or packaged templates (for both procedure notes and the history and physical [H&P]) produce redundant, formulaic information that makes it easy to overlook significant clinical information.”
- “EHRs facilitate medication reconciliation and management and include alerts for improper drug dosages and drug allergies. However, frequent drug-drug interaction alerts lead to “alert fatigue,” sometimes causing physicians to override or disable them.”
- “EHRs improve presentation of data for clinical decision making. However, providers often copy and paste information from prior notes into new notes (and, hopefully, edit as appropriate). This contributes to irrelevant, sometimes erroneous, over-documentation.”
- “Templates with drop-down menus facilitate data entry but are often integrated with automated features elsewhere in the EHR, where errors can be easily overlooked and disseminated. If an item is selected above or below the one desired, for example, “qd” can become ‘qid.’ “
- “Clinical Decision Support (CDS) systems provide alerts, warnings, and reminders for medication and chronic disease management. This information may conflict with a medical specialty’s clinical standards of care or practice guidelines— and with the information in FDA-approved drug labels… A physician should always document why a CDS prompt was overridden.”
- “During discovery, lawyers will request EHR metadata that documents logon and logoff times, what was reviewed and for how long, what changes or additions were made, and when the changes were made. All physician interactions with the EHR are time-tracked and discoverable.”