Why electronic medical records are dangerous to older adults

A recent experience with my father-in-law reminded me of something that has concerned me for some time. While EMRs have some benefits for older adults, on balance I believe that they portend more dangers. There are multiple reasons, but the biggest is that health care providers tend to believe everything they read in an EMR. Even if what they read is wrong!

A wise computer programmer once told me that “computer’s are dumb as posts,” they are only as good as the information that human beings load into them. Human beings make mistakes, hence, inaccurate or false information will find it’s way into an EMR. The doctor who reviewed my father-in-law’s EMR prior to seeing him made the first cardinal sin, he believed everything that he read in the record.

The second error was in not directly getting the history from my father-in-law and myself as I sat at the bedside. The final mistake was in doing a cursory examination and forgetting the most important tool a physician has, his own eyes, nose and ears. I’ve often told my patients that the most important thing I ever do is to pay attention to how they look and act when I walk into the exam room. My experience and instincts will sound alarms that then lead me on my search to figure out what I need to do to help them.

It is well known that chess grandmasters think in terms of patterns. They look at the entire chess board and look for recognizable patterns. I believe that geriatricians are similar to grandmasters. If we get caught up focusing on our patient’s diabetes and hypertension, we may miss the more subtle changes that are occurring in their function and quality of life. EMRs have the distinct disadvantage that they are often singularly focused on specific diseases.

The answer is geriatric medicine. Geriatricians are trained to care for the frail elderly. We are trained to look at patterns and the big picture. Keeping a 90-year-old’s blood sugar or blood pressure too low might create problems rather than solve them. The focus on maximized function and quality of life may not mesh with achieving certain laboratory based numerical goals.

One of my favorite stories is about an 88-year-old patient of mine who had been diagnosed with prostate cancer. He was treated with expensive anti-hormonal injections that chemically castrated him. He was getting weaker and finally ended up hospitalized with pneumonia. The urologist was killing the prostate cancer and killing my patient. We stopped the injections and actually placed him on testosterone. The urologist was apoplectic! I asked him what level of PSA was associated with metastases to the bone. He responded, as I knew, that it was over 30. I then said, well, let’s watch his PSA, and if it goes above 20, we can consider treating him again. My patient is now 95-years-old, and still active and functional. His PSA has hovered around 15 for the past 3 years.

Policy makers continue to believe that EMRs will lower costs and improve care. In the realm of the frail older adult, I believe it will do neither, at least not until we develop artificial intelligence software that thinks like a geriatrician. Even then, we must grapple with the issue of human error when inputting data. Until that time, I will continue to take my personal responsibility to heart when I touch, look and listen to a patient.

Michael Wasserman is a geriatrician and can reached on WassDoc.

View 10 Comments >