My post on Amalga Hospital Information Systems generated a fair amount of interest and I was asked to expand somewhat on why I believe this is an important initiative. Let me tell you a story. I was working on the in-patient medicine service late last year when I noted a new patient had been admitted through the night and was now on my list. She was diagnosed with a viral gastroenteritis and was placed on a short term observation status. I normally have 15 or 20 patients to see every day and by necessity I have to prioritize who gets seen first, and who can wait until later in the day. Patients admitted under observation status are expected to be discharged within about 24 hours because their illnesses are usually fairly mild. As I scanned my work list I made a mental note of my new patient, reviewed her data as contained within the hospital’s Electronic Health Record, and planned to see her in the late morning or early afternoon. There was no indication whatsoever that this patient was suffering from anything other than a fairly mild illness.
As I was making rounds mid-morning a “Code Blue” (cardiac arrest) was broadcast over the hospital’s intercom. While the patient names are never revealed, their room numbers are by necessity. I knew right away who it was. My patient suffering from a presumed mild illness had just “coded”. The nursing and medical staff did their very best for her, but unfortunately she died a short time later. I hadn’t even seen this poor woman, yet I felt sick with the sense that we had somehow failed her.
In the days that followed I reviewed her case in detail and it was apparent that while the admitting diagnosis was reasonable given the presenting findings, her condition had subtly changed over the hours since her admission. The fact that this wasn’t recognized wasn’t anyone’s fault, rather it’s the nature of the current state of in-patient health care in general. While thankfully rare, events such as this occur in hospitals all over the United States. Why? Because of our current asynchronous health information systems. Patient vital signs are checked manually by a nurse or assistant, scribbled on a piece of paper, and manually entered into either a paper chart or an Electronic Health Record. The data then sits in either the paper or electronic repository until viewed and interpreted by the attending physician. There is no active support system that flags patients who may be changing for the worse.
The failure to detect subtle changes in medical condition occurs because of the lack of real time clinical decision support and patients and their families suffer for it. As I noted previously, this is exactly what the MEWS score, as provided within an electronic clinical dashboard, would address. It’s time we married the software resources and expertise of corporations such as Microsoft with the real world clinical needs of our nurses and physicians.
Disclosure time: For obvious medico-legal reasons, the above case is a very real compilation of my experience as a hospitalist rather than one specific patient.
