Over the past couple of decades, the ever-increasing medical-legal, financial, and patient-safety demands have led to a mushrooming of the amount of documentation needed during patient care, especially in the inpatient setting. With the advent of HM, that burden simply shifted to hospitalists, along with rapidly expanding patient loads as PCPs started delegating inpatient care. The problem was further compounded in teaching hospitals by a gradual decline in resident support as patient caps and work-hour limits were instituted.
Not surprisingly, the quality and accuracy of physicians’ progress notes started to suffer. It became frequently difficult to remember the details of encounters from earlier in the day, and fatigue sometimes led to details essential to continuity of care being left out completely.
A potential—and very intriguing—solution presented itself, however, as we were designing family-centered team rounds. Apparently, the ED group at the adjacent Abbott-Northwestern Hospital in Minneapolis had been using scribes for several years. The scribes captured the providers’ notes during each encounter, and the group had developed a polished system of scribe recruitment and training (mostly pre-med students awaiting acceptance to medical schools).
But would this model work in the inpatient setting where workflow was completely different? This was a total unknown, given that there has been no precedent in the literature about inpatient scribe utilization. Although relatively meager in number, the published articles about outpatient use of scribes were generally favorable, with the authors noting improved productivity and increased patient and staff satisfaction.1,2,3,4
Operationally, the main difference between inpatient and outpatient scribe use lied in the rounding mechanism itself; it was not just a provider-patient interaction, as in the ED, but rather a sometimes cacophonous collection of multiple inputs from families, providers, students, residents, nurses, and ancillary staff. The challenge here was to train the scribes (natural ability played a significant role) to produce a coherent document out of this interaction; the document had to accurately reflect current status and plans of care. The first step was to create a template for the daily progress notes with all the necessary headings, as well as age-appropriate normal exams, with the ability to directly import from the EMR such dynamic data as vital signs, medication lists, and lab results.
Each morning prior to rounds, the scribe was provided a list of patients to be rounded on, and the scribe spent about 30 minutes creating partial notes on each patient, as well as a brief patient identification statement. This proved to be a significant time-saver during rounds and allowed the scribes to keep up with the care team. Each note was pre-pended with the following statements, as recommended by the hospital’s risk-management team:
- I, (scribe name), am serving as a scribe to document services personally performed by (hospitalist name), MD, based on my observation and the provider’s statements to me.
- I, (hospitalist name), MD, attest that (scribe name) is acting in a scribe capacity, has observed my performance of the services, and has documented them in accordance with my direction.
A computer-on-wheels (COW) station was assigned to the scribes and accompanied the rounding team into the patient rooms, making it possible for them to follow the interactions in real time. The scribes were encouraged to hold their hands up and interrupt the conversation at any time if they needed clarification, something that happened rather infrequently. On the other hand, the care team was directed to speak clearly and the residents were instructed to present their cases using the same format as the progress-note template, thus minimizing guesswork on the part of the scribes. The only part of the interaction the scribes could not transcribe directly was the physical examination, and typically the attending physician dictated abnormal findings as well as systems not assessed, if any, at the end. All the scribe-generated notes were forwarded to the rounding provider’s EMR inbox and reviewed by the latter, amended when necessary, and signed electronically. The review of the scribe-generated notes proved to be initially a time-consuming process, typically two hours. However, as the scribes gained experience, the time commitment on the part of the providers decreased rapidly; after about three months, the time spent was about 30 minutes on busy days, even less otherwise, with a concomitant increase in satisfaction with the scribing process.