Friday, October 07, 2011

This Is Really Quite An Interesting Perspective on Electronic Health Records. Substantial Truth Here I Think.

One clinician explores the reasons take up seems a little slow in the US.
Tuesday, September 27, 2011

Why Doctors Don't Like Electronic Health Records

A physician argues that electronic patient records raise costs, decrease patient visits, and make poor communication tools.
Why are doctors so slow in implementing electronic health records (EHRs)?
The government has been trying to get doctors to use these systems for some time, but many physicians remain skeptical. In 2004, the Bush administration issued an executive order calling for a universal "interoperable health information" infrastructure and electronic health records for all Americans within 10 years. And yet, in 2011, only a fraction of doctors use electronic patient records.
In an effort to change that, the Obama economic stimulus plan promised $27 billion in subsidies for health IT, including payments to doctors of $44,000 to $64,000 over five years if only they would use EHRs. The health IT industry has gathered at this multibillion-dollar trough, but it hasn't had much more luck getting physicians to change their ways.
What is wrong with doctors that they cannot be persuaded to adopt these wondrous information systems? Everybody knows, after all, that the Internet and mobile apps, powered by Microsoft, Google, and Apple and spread by Facebook, Twitter, YouTube, and the iPhone and iPod, will improve care and cut costs by connecting everybody in real time and empowering health-care consumers.
I suspect the answer may lie partly in something essayist E. B. White said about humor. "Humor," said White, "can be dissected as a frog can, but the thing dies in the process, and its innards are discouraging to any but the pure scientific mind." Similarly, humanity withers when it is dissected and typed into an EHR. As Jerome Groopman, a Harvard internist, wrote in How Doctors Think, "Clinical algorithms can be useful for run-of-the-mill diagnosis and treatment ... but they quickly fall apart when doctors need to think outside their boxes, when symptoms are vague, or multiple and confusing, or when test results are inexact."
The computer is oversold as a tool to improve health care, implement reform, cut costs, and empower patients. The reasons are obvious to anyone who treats patients. You cannot look a computer in the eye. You cannot read its body language. You cannot talk to an algorithm. You cannot sympathize or empathize with it. 
We physicians are not Luddites or troglodytes. We are savvy about using the Internet, technology applications, and social media. For us, medicine mixes art and science. What we seek from patients are clues, constellations of signs and symptoms, and stories. We choose not to be reduced to data-entry clerks sorting through undigested computer bytes.
A string of numbers containing demographic, laboratory, and other patient information, no matter how systematically assembled or gathered, is not narrative. It does not tell a story. It contains "just the facts," as Sergeant Joe Friday used to say. That is why an ophthalmologist told me that when he gets an EHR summary, he ignores it: "It does not tell me the patient's story. It does not tell me why the patient is here, what troubles the patient, and what the referring doctor wants me to do."
Richard L. Reece is a retired pathologist and the author of The Health Reform Maze: A Blueprint for Physician Practices. He blogs about health reform, medical innovation, and physician practices at

More here:

This comment really fits with the point I have made on this blog about just how hard it is to actually represent, and then store, the clinical thought processes.

My take is that there is a lot of underestimation of just how hard and complex the EHR task actually is and until that is clearly recognised we will all struggle to make major progress.

We certainly have a fair bit of work to do in this domain before we can be satisfied we have the problem solved.


1 comment:

Oliver Frank said...

Dr. Reece is a little bit naughty in dismissing the ability of electronic information systems to help doctors to do their job better. He admits that:

"Clinical algorithms can be useful for run-of-the-mill diagnosis and treatment"

but then dismisses the value of electronic information systems by going on:

"... but they quickly fall apart when doctors need to think outside their boxes, when symptoms are vague, or multiple and confusing, or when test results are inexact."

A great deal of medical practice does indeed involve not only "run-of-the-mill diagnosis and treatment" (for which our computer systems can help us), but for GPs especially requires us to keep checking to see whether the patient has had various services whose need, or at least the indications for which, are predictable. This includes all routine primary and secondary preventive activities, and planned care of patients with long term health problems, such as diabetes, hypertension, heart disease and so on. Humans find it hard to remember to do all of the recommended routine preventive and follow up activities, but computers, once programmed, can remind the doctor and the patient of planned care which is due or overdue.