There’s been a quiet revolution in medicine over the past five years as doctors have shifted away from scribbled notes and moved much of their work to their computer screens.
However, the adoption of electronic medical records, aided by billions of dollars in federal incentives, has had some unintended consequences, from life-threatening errors to less eye contact with patients.
Putting a stethoscope on some of those downsides is Dr. Robert Wachter, a University of California, San Francisco, medical professor, author and hospitalist. In his new book, “The Digital Doctor: Hope, Hype and Harm at the Dawn of Medicine’s Computer Age” (McGraw-Hill, $30, 320 pages), Wachter offers a doctor’s-eye view of how the computerized world of medicine is changing.
Wachter, who is interim chairman of the UCSF Department of Medicine, talked with us at a recent medical conference and by phone. Here’s an excerpted version:
Q: Most would say electronic medical records have helped patients with scheduling appointments, seeing lab results and emailing their doctor. What are the downsides?
A: The book flowed from my own wonderment that some of this is really screwed up. Absolutely no one is happy with the status quo, patients or physicians. … It’s not just the Luddites … even good folks are beginning to crumble under the weight of technology in medicine. It’s a completely new way of practicing medicine. If you don’t approach it that way, you can get it wrong and then you start to see burnouts and errors.
Q: You discuss the “breathtaking errors” that can occur from computerized systems, such as the teenage patient given an antibiotic dose that was 39 times the recommended amount – despite a computer alert. The boy suffered a grand mal seizure, but survived.
A: That was at UCSF, a fabulous hospital with the best computer system money could buy. Physicians receive about 1,000 medication alerts each day and the pharmacists about 5,000. … When adding new protections that didn’t exist before, some are a double-edged sword. If you’re firing medication alerts every few minutes, eventually doctors, nurses and pharmacists will ignore them. It’s the same as when you click out of “terms and conditions” on a website. If the last 99 medication alerts you’ve seen haven’t been useful or caused you to change your prescription, the 100th could get ignored. … We put inordinate amounts of trust in technology. But in health care, the tools are so early and primitive, there’s a danger we’ll put too much trust in the technology that can lead to errors.
Q: You’ve written that modern medicine has become “dehumanized” by the computer’s arrival in the exam room. Are doctors becoming too focused on the so-called iPatient and losing sight of their real patient?
A: It’s not that we were uniformly great at this before. It’s very easy to romanticize that we were absolutely perfect at making eye contact, handholding and involving patients in their (health care). … There’s always been a challenge in the exam room between the person in front of you and the records that need to be filled out. When it was all on paper, it had a different flavor. You could concentrate on the patient and scribble a few notes. Today, the tension between documentation – the clerical aspect of medicine – and the human interaction is more stark.
In my book, there’s a crayon drawing done by a 7-year-old girl after a visit to her pediatrician. It shows the patient, her mother and a nurse at the end of the exam table … at the other side of the drawing, the doctor is hunched over his computer with his back to everyone. The only thing the girl got wrong is that the doctor has a smile on his face. Doctors don’t want to be data-entry clerks.
Q: With doctors spending more hours on their computers, inputting notes, diagnoses, answering patient emails, etc., how much of a burden has it become?
A: Filling out “a few more things” has morphed into 50 more things. On average, doctors are spending 40 to 50 percent of their time on the computer, either reading or entering things. That’s probably half the time of a 30-minute encounter with a patient. … Our average primary doctors say when they get home after seeing patients, they spend an additional three hours on the computer. Every time any of your patients gets a blood test or sees a specialist, it comes to your inbox. For a primary care doctor following 1,800 patients, that just does not work. It doesn’t mean the information isn’t valuable, but … we need to think deeply about the the workflow and people problems in the systems. You know something’s wrong when an Arizona hospital advertises for a physician emphasizing that the hospital does not have electronic medical records.
Q: What’s the solution?
A: Now that it bothers patients and doctors more deeply, it’s not a trivial concern. The cleverest solution is medical scribes (who take notes for a doctor), but it’s not the ultimate solution. The technology will get better and more unobtrusive … like voice-activated technology that can generate a doctor’s notes. … In the last several years, Silicon Valley has gotten very interested in health care. With all the new data and tools, Silicon Valley could be our savior here … and may be the biggest threat to traditional health systems if they can deliver health care safely and less expensively.
Q: One of your concerns is that doctors may self-censor their diagnostic notes, for fear that patients reading them will be unduly alarmed. You said four conditions were especially susceptible to doctors tiptoeing around their words: cancer, mental health, substance abuse and obesity.
A: I worry about self-censorship. Only 5 million patients have access to open notes, so it’s still rare. But it’s inevitable that it will happen. Patients have a legal right to own their medical records, but it has side effects. If you come in with possible pneumonia, I can’t rule out cancer or lupus so I write that in my notes because I want to prompt myself next time I see you or for a colleague if you came into the emergency room. But I probably won’t write that if you’re going to look at the (notes). … I’m going to be worried that the patient is going to be wigged out, even if it’s only a 0.5 percent chance of cancer.
Q: You mention that the digital age means some medical fields – like radiology, for instance – could diminish in importance, if not disappear altogether.
A: For the next five or 10 years, that’s not going to happen. It’s very complicated looking at an X-ray and putting it in context with a patient’s medical history. But over 20 years, I’d bet on the computer for this one. There is field after field where technology has disrupted the practice. … In medicine, it will play out in patchy ways. If you’re a radiologist or dermatologist, a lot of what you do is visual recognition, pattern recognition. That strikes me as not a problem for artificial intelligence (to take over). If I was making a 20- or 30-year bet, I’d be betting against diagnostic radiology, dermatology and pathology that utilize visual pattern recognition.
Q: What’s your overall diagnosis of medicine’s digital age?
A: Even today with all the problems, we’re better off with (electronic medical records) than without them. It’s easy to get lost in the disappointments, but there is no question that people are getting better care and we are preventing more errors than when we were on paper.