Bad research? Flawed conclusions from Harvard-based EMR study?

Recently I read a post over at Medwire News citing a that investigated whether the copy-and-paste methodology could have detrimental effects on diabetes control.  Since I previously blogged on this topic, of course I found the study very interesting.  In their study, the authors used a software program to correlate hemoglobin A1c reductions (a marker of diabetes control) with how often copy-and-paste was likely used.  The software guessed at whether parts of a note met enough similarity to previous notes to suggest that they might have originated by copy-and-paste methods.  Because the diabetic patients being followed in the copy-and-paste group did not have lower hemoglobin A1c levels over time, the authors concluded that copy-and-paste had a role in causing bad patient outcomes.

Although interesting, this study is concerning for several reasons.  It places a bad view (real or not) on doctors who use copy-and-paste responsibly and effectively.  Since I do this all the time and have not noted any particularly bad outcomes in my diabetes patients, I have to question whether the conclusion is valid.  There is so much potential for error in this type of statistical correlation research that I think big disclaimers should be noted.  What kind of counseling was being done in the subgroup of diabetics that had an allegedly poorer outcome?  Was that group of diabetics different from the ones with a good outcome?  What approaches to treatment were used?  What years were they being treated?  Who was treating them?  Etc, etc.  Perhaps the most important two questions in my mind are:  Who are the authors responsible for the study?  What are the authors’ personal biases?

The authors conclude that “These results lead us to question whether copied electronic documentation is a reliable representation of patient care,” in a letter to the Archives of Internal Medicine. “If it is not, it could be either an honest mistake or deliberate falsification.  In the latter case, copied documentation that does not reflect the actual events is a serious breach of medical ethics. In either case, it carries a significant financial and legal risk.”  There seems to be such a negative slant here that I have to again ask about the personal biases of the authors and how this may have affected their study design.

What would be my motivation to enter documentation that said I did  things that I didn’t actually do?  That could also be an important question with multiple possible answers.  How much time does the doctor have?  Is the doctor a resident trainee?  How protective of the practice does the doctor feel?  Does quality of electronic notes in general (which is highly varied) directly correlate with patient outcomes?  Maybe, since the data from Brigham and Women’s Hospital suggests falsification in the author’s eyes, they should take a good, long look at the context of their research methodology (who was involved in writing the clinical notes and under what circumstances) rather than relying on a possibly very flawed method to generalize negative study results to all doctors using reasonable and responsible documentation methods.  Or maybe I’m being too rash?

Dr. West is an endocrinologist in private practice in Washington, DC. He completed fellowship training in Endocrinology and Metabolism at the Johns Hopkins University School of Medicine. Dr. West opened The Washington Endocrine Clinic, PLLC, as a solo practice in 2009.  He can be reached at

About the author

Dr. Michael West

Dr. Michael West

Dr. West is an endocrinologist in private practice in Washington, DC. He completed fellowship training in Endocrinology and Metabolism at the Johns Hopkins University School of Medicine. Dr. West opened The Washington Endocrine Clinic, PLLC in 2009. He can be contacted at


  • I currently get my care through the VA medical system since loosing my private insurance. I have found one provider consistanty copy and pasted that I was reporting my sugars to him, when I did not have a meter. When I asked him about this he told me I had told him my levels, when I asked for a meter to check he documented that my other meter was taken away from me for unknown reasons (I am in my 50’s have 2 grad degrees, worked in hospitals as a clinican).

    The VA system and the EMR has such great potentials, to pull data, to improve care outcomes, research potentials. I ran across a recent publication by the VA where they were addressing sucide in Vets and looked at last medical contact to see if the vet reported feeling suicidial. The pulled data from the EMR, and paper records. The outcome in brief was that the system needed to find a way to have Vets feel more comfortable in speaking about suicidial feelings. It is interesting that when I reviewed my records, there was documentation that I had denied sucidial feelings, however, I have never been asked that quesiton.
    There are many interesting issues that get presented with the EMR, one of them is the time it takes for a provider to type out the note compared to being able to dictate-potentaily the beast may have created many issues which will negatively impact care.

  • Kevin, it sounds like that was a fraudulent doctor. I don’t know why they do such wacko things. It’s not hard to document accurately. Unfortunately, there are bad eggs in the medical system. Good for you for pointing out the inaccuracies though. I’ve had patients do so for me occasionally, and then I correct their files by adding addenda.

Click here to post a comment