Years ago, I attended a health policy conference where Newt Gingrich proclaimed, “paper kills.” In those days, we all thought that digitizing medical records was going to revolutionize healthcare. It was guaranteed, we thought, to make care safer by replacing hand-written, often illegible, prescriptions and visit records with tidy digitally recorded ones. Built-in clinical decision support would also reduce or eliminate errors and help clinicians remember to provide all the needed preventive services. Further, digitized records would be easier to share between various sites of care.
Today’s electronic health records have accomplished some of these goals, but there are also downsides that aren’t always discussed by those advocating for their use. In his new book, “The Digital Doctor: Hope, Hype and Harm at the Dawn of Medicine’s Computer Age,” Dr. Bob Wachter, Chief of Hospital Medicine at University of California San Francisco and the “Father of Hospitalist Medicine” shares some of the dark side of health information technology (HIT).
During his sabbatical year, Wachter traveled the country interviewing close to 100 people involved in one way or the other in the digitization of healthcare. Typical of all things tech, women were woefully underrepresented in this list of luminaries making up less than 10% of the interviewees if you don’t count the people involved in the case gone wrong that he weaves through the book.
One patient’s story
That being said, it is about my only criticism of this well-written, well-researched, and entertaining book. Wachter illustrates the points he wants to make with real-life stories of people impacted by health IT or the lack thereof. The most important story he tells is that of a teen, Pablo Garcia, who lives with a rare condition called NEMO deficiency syndrome, a condition with multiple manifestations including recurrent infections with bacteria, such as Staph and Pneumococcus, and various gastrointestinal disturbances.
Pablo was hospitalized at UCSF’s Benioff Children’s Hospital for a colonoscopy to evaluate an intestinal polyp, a fairly routine medical procedure. Due to a series of coinciding events, Pablo was given a massive overdose of the antibiotic Septra. Wachter traces the genesis of this error from the time the drug order was entered into the electronic health record through Pablo’s swallowing of 38 1/2 pills. The boy suffered a grand mal seizure as a result of the overdose, but thankfully he ultimately recovered.
The problem, it turned out, was how the Epic EHR was configured for pediatric medication orders. The resident physician who ordered the drug was alerted by the pharmacist that she needed to make a change to the original entry so that the correct dose would be 160 mg. The system was configured to default to the unit of measurement (mg versus mg/kg) last used.
So, when she typed in 160 mg thinking that was the total dose to be dispensed, the system changed it to 160 mg/kg as that was the unit used in the original order. A series of seemingly minor events, such as having a float nurse who was on an unfamiliar ward, order processing via a robot instead of the pharmacist, and Pablo’s mom telling him he would likely have to take a lot of pills prior to his procedure all contributed to this gigantic dose of pills being administered.
In Pablo’s medical error case, the individuals involved apparently relied on the computer more than their common sense. It’s a reminder that we still have to think about what we are doing and ask questions if something doesn’t pass the “sniff test” even though it is what the computer said to do.
Cut and paste, and more…
Other issues Wachter explores include residents being glued to their computer screens instead of interacting with patients and ward staff, unhappy physicians (a hospital recruiting docs added NO EMR to its ad), the demise of radiology rounds, and the lack of user-centered design that makes many doctors yearn for their paper charts.
The “cut and paste” features make it easy for notes to be copied over and over again—even if the information in them was not correct to begin with. And to add to that, templates that allow doctors to quickly check off answers to different parts of the history and physical exam—heart sounds normal (check), breath sounds – no wheezes or rales (check), abdomen non-tender, no masses (check again)!
It was exactly these features that got an old friend of mine into trouble when he counter-signed a resident’s note using the template designed for this purpose. Failure to read it carefully after pasting it, led him to document in the chart that he had, “examined the patient and concur with the resident’s findings.” The problem was he hadn’t examined the patient—in fact, the patient went to surgery before he could examine her—but he countersigned based on the resident having presented the case to the subspecialist.
Of course, when the case went south, my friend was left holding the bag because he had “lied in the medical record.” Sloppy documentation, yes. He agrees. But it wasn’t lying and, by the way, the problem occurred during the surgery anyway.
I highly recommend you take the time to read The Digital Doctor. Not only does Wachter spin a great tale, but he makes you think about electronic medical records in a whole new light.