Electronic Medical Records and Symptom Rating Scales: "Solutions" Making Problems Even Worse

Sometimes an effort to solve a problem just creates different, and at times even more serious problems. Take the electronic medical record (EMR) for example. Please. (Apologies to Henny Youngman).

In a clinic I work at that uses an EMR, we are required to fill out a “treatment plan” on each patient. This is - for me at least - a waste of time since my initial psychiatric evaluation note contains all of the information contained in the treatment plan, as well as important contextual information about the patient that is not included in the plan.

The treatment plan is not only redundant, requiring me to enter information twice, but increases the number of notes in the medical record, making it more difficult for anyone reviewing the chart to find specific information they may need to as they attempt to properly care for a patient. 

Sometimes trying to find specific information in the record is like searching for the proverbial needle in a haystack. Since we aren’t given a lot of time to review charts before patients come into see us in the first place, this takes time away from the all-important face time with patients.

A narrative note is always better than a form with boxes, which often does not provide a “box” that adequately addresses an issue for a patient. But having reviewed the notes of other psychiatrists, I can see why some administrator felt that having a formal treatment plan in the EMR was a good idea. A lot of the information inherent in a good psychiatric evaluation was nowhere to be found in the notes from a good percentage of my fellow clinicians.

Not only that, but many of their follow-up progress notes did not mention what symptoms were being addressed by the doctor's treatment. Nor did they mention what symptoms, if any, had responded to any medications that were prescribed, let alone whether or not the symptoms had actually resolved.

The treatment plan template wisely asked for these target symptoms. Recently, however, the plan template was changed to include a question asking the clinician how symptom improvement was to be measured.

The answer should be, “by clinical evaluation.” And that is what I always write. However, I think the designers of the template are looking for some sort of psychological assessment device, as if that were more valid that a clinician’s evaluation. As I frequently rant about on this and other blogs, such measures are typically screening tests consisting of symptom checklists, and as such are NOT meant to be a measure of clinical progress, let alone of a definitive diagnosis.

The question a good psychopharmacologist asks, particularly with major mood disorders, is not only whether the symptoms the patient has have improved. It is whether the patients have returned to their “baseline,” the way their mood was before they developed an episode of, say, major depressive disorder.

The Hamilton Depression Scale (HAM-D) scale for rating depression is one of the most frequently used symptom checklists, both clinically and in research.

It was published fifty years ago Max Hamilton (1912–1988). Interestingly, Hamilton himself was forever pointing out that the HAM-D was not intended to be used to make the diagnosis of depression. It does not have either sensitivity or specificity—there is a great deal of overlap with symptoms of other diagnoses, particularly anxiety disorders. 

Yet somehow it became the gold standard for not only evaluating the efficacy of antidepressants, but for measuring symptom improvement clinically.

I would much rather the administrators of the clinic solved the problem of absurdly vague “progress notes” by spelling out the minimum information they need to contain, not by asking a stupid question on a treatment plan form.