When A Flawed EHR Threatens Patients
When Harvard investigators tried to figure out why a patient nearly died, they discovered that even a sophisticated system of e-reminders and alerts isn't always enough.
When assessing a flawed electronic health record system, "interventions should make it easy to do the right thing and hard to do the wrong thing." That's the advice from three Harvard clinicians on how to fix an EHR system that almost allowed a patient to die.
This story begins with a 53-year-old woman who developed life-threatening complications from pneumococcal sepsis because she wasn't given a vaccine to prevent the infection. The patient presented with the sepsis 10 years after a car accident that required the removal of her spleen.
More Healthcare Insights
- The Untapped Potential of Mobile Apps for Commercial Customers
- Automating Hospital Workflows That Maximize Efficiency And Patient Satisfaction
- Redefining Value in Healthcare: Innovation to expand access, improve quality and reduce costs of care
- Enabling Healthcare Transformation with Social Business
The vaccine is considered standard for patients who have had a splenectomy because the absence of the organ weakens their resistance to a variety of pathogens, including the one that causes pneumococcal sepsis.
If this tragedy had happened in the pre-electronic age, the clinicians working the case would have just been admonished to be more diligent about applying clinical guidelines. In today's healthcare system, it required IT managers, physician leaders, and patient safety personnel to figure out whether the hospital's EHR system was at fault, and if so, how to stop such mistakes from happening in the future. The resulting investigation can help inform your next EHR redesign.
The investigation revealed that the patient had been cared for during the entire 10-year period within the same healthcare system. So in theory all the docs had to do was give this patient the vaccine somewhere along the line to prevent disaster. But it turns out that the splenectomy was never noted in the EHR's problem list. Had it been listed, the patient may have been given the vaccine because the system issues reminders to give such patients the drug every five years.
The operative words are "may have." Studies have found that only about six out of 10 patients with the term splenectomy on their problem list actually get the vaccine.
Having limited faith in the reminder approach, the Harvard clinicians considered creating what they call an "active interruptive prompt" that more or less shouted: Give the vaccine! Another option was to send routine reports to all providers on patients who had had the surgery, encouraging them to send letters to those patients about the need for the vaccine.
Unfortunately, even if the system were tweaked this way, it wouldn't have helped our sepsis patient because her primary care physician had implemented the EHR system after the car accident, so the splenectomy wouldn't have been included in her problem list at that time. And it was never added to her record after the fact.
So essentially, IT managers and the clinicians they work with are faced with a twofold dilemma: Many patients who have their spleens removed have that fact listed in the EHR problem list, but no one pays attention. And some patients who have had the surgery never have that information entered in the record.
What's The Solution?
In their New England Journal of Medicine case summary and analysis, Tejal Gandhi, MD, and his colleagues suggest a few partial solutions. They hope that these measures will in fact make it "easy for clinicians to do the right thing and hard for them to do the wrong thing." Among their suggestions:
-- Establish an electronic link between operative notes and the EHR problem lists. Doing so would automatically insert "splenectomy" in the latter.
-- Use billing data to identify patients who have had the surgery and let the billing data "talk" to the physician through the EHR system, prompting the physician to add that fact to the problem list.
I don't think these electronic fixes are enough. The real solution is one that nobody in the healthcare reform camp wants to hear: Short of implanting a microchip in every doctor's head, the simple truth is each clinician needs to see fewer patients and spend more than the typical eight minutes with each of them. And they need more time to manage all the thousands of tasks they have to perform daily. Obviously, this isn't going to happen anytime soon. But that doesn't make it any less true.
Find out how health IT leaders are dealing with the industry's pain points, from allowing unfettered patient data access to sharing electronic records. Also in the new, all-digital issue of InformationWeek Healthcare: There needs to be better e-communication between technologists and clinicians. Download the issue now. (Free registration required.)