Residents at various stages of training assess approximately 1,400 patients per year on our consultation-liaison service at the Mayo Clinic in Rochester, Minn. Given the variety of demands of any busy teaching service, we felt it would be helpful to reduce documentation time, while preserving adequate quality of communication via the notes, to increase the time available for education and other clinical pursuits.

To accomplish this, we created a standardized self-entry electronic consultation note that would assist in quickly documenting a thorough consultation-liaison psychiatry interview. We hypothesized that using electronic templates to type notes directly into the electronic medical record (EMR) would decrease documentation time, increase legibility, and improve the immediate availability of documentation by residents. We felt these improvements would be noted despite increasing the amount of data recorded by most residents. Further, we hypothesized that such a system would be welcomed by the residents, who valued efficiency.

To implement this system, we utilized a commercial software program, ShortHand (1). This program allows for “word expansion” wherein an entire word or phrase will appear after a keyword—often only two or three letters—is typed. More powerfully, this program allows for the creation of templates, which are customizable lists of commonly used words and phrases that are included in the final document with a simple mouse click. For example, if the user typed “ROS” into the EMR, a series of questions querying the status of the standard Review of System would appear in the form of check boxes highlighting common answers and an area allowing for free text entry to describe each of the systems. Similarly, a template outlining a thorough Mental Status Exam, including the most common descriptors of each subset thereof, would appear when the user typed “MSE,” allowing the user to click the descriptors quickly and move to the next section of the exam. The symptom lists were designed with the “Unremarkable” finding at the top of the list, and unusual results in order of descending frequency thereafter. If a resident were to omit part of an examination, a descriptor labeled “Not assessed” would be offered, allowing the resident to move to the next section of the template while reminding him or her of the expectation that this area be examined. As a commercially available program, ShortHand was designed to be compatible with many currently available software applications and works in almost any application into which you can type. This program was customized to meet the specific requirements of a psychiatric consultation note (2–4). We gave specific attention to diagnoses common on our consultation-liaison service (e.g., delirium, capacity evaluations, and affective disorders in typically complex medical patients) to permit rapid documentation. This involved the creation of templates to allow for simple point-and-click inclusion of findings, diagnoses, and recommendations most commonly seen on our service. Further, standard phrases, such as our contact information, were included automatically in all documentation, requiring no additional effort on the part of the resident. Importantly, completion of all required aspects of each particular component of the examination was mandatory prior to moving to the next screen, ensuring more thorough notes.

Residents were taught to use this note system in a 60-minute session. Note entry could be made directly into any computer workstation throughout the hospital. After training, residents were able to contact us for further clarification as questions arose. Six months after institution of this program, residents (PGY-2 through PGY-6) were asked via e-mail to estimate the time required to produce a handwritten versus an electronic note. In that same e-mail, we solicited comments regarding the use of this system compared with the traditional written note. A second copy of this e-mail was sent to the residents 3 months later to increase the response rate. This process was not reviewed by the Institutional Review Board, as our intent was to gather “quality control” data to support the use of this technology.

We sent the e-mail questionnaire to 30 residents. Twenty-three replied, providing a 76.7% response rate. Five had never handwritten a note and had only used the electronic documentation system. Ten had never entered notes electronically, as they rotated on the service prior to application of this technology. Eight had documented their consultation-liaison interviews both electronically and via handwriting. Of those eight, the average estimated time to handwrite a note was 45 minutes, whereas the average estimated time to document electronically was 27.5 minutes, a difference of 17.5 minutes per note. When comparing all responses, the average handwritten note took 43.89 minutes, and the average electronic note took 35.7 minutes. However, three residents documented electronically but chose not to use ShortHand. Their average documentation time was 55 minutes, which skewed the results for those typing their notes. When calculating the average time to type a note for those using ShortHand, the average was 30.45 minutes (Table 1).

The difference between the traditional method of documentation (handwriting) and documentation using the templates created for this purpose was 13.44 minutes per note on average. Among those who had the opportunity to utilize both formats, the average handwritten note took 45 minutes, and electronic entry took 27.5 minutes, a difference of 17.5 minutes per note.

In addition to requesting estimates of documentation time, feedback regarding the use of the ShortHand system was sought. The feedback obtained was overwhelmingly positive. Notes entered using ShortHand were more immediately available, more legible, more thorough and “data-inclusive,” completed more quickly, and more uniform in presentation and content.

Residents learned the system quickly and provided valuable feedback to make the templates even more efficient. Importantly, the standardized nature of the note ensured that all details required in a full consultation were performed. Despite this, most saved time by using this system, making documentation more efficient, which is consistent with available literature from other fields (5, 6). As notes were entered directly into the electronic medical record, they were immediately available via any computer to the primary service, which decreased systematic inefficiency. Additionally, most attendings felt that this system allowed for more efficient use of their time as well, as notes were more thorough, allowing subsequent documentation to be more brief and targeted. Although we felt that decreasing the time to write a note could allow for extra teaching time, we did not directly assess this outcome because we believed that the allocation of “extra” time would be dependent upon the trainee, the supervising physician, and the situation each day. Despite this limitation, the use of templates has the educational benefit of reinforcing the thorough collection of data and key components of the mental status examination, as well as expanding the residents’ vocabulary for potential descriptors. A second limitation of this study is that only eight residents used both the handwritten and typed methods, due to the timing of the introduction of the templates to the service. Upon introduction, it was required of all residents on the consultation-liaison service to type their notes, so the data were limited to those residents assigned to the service in the 6 months prior to and the 6 months after the institution of these templates. Another limitation was the reliance on retrospective estimations by residents as to the amount of time it took to generate a “thorough consultation note.” Given the subjectivity of the question, there is significant potential for bias.

Future research in this area should include formal measurement of documentation and whether the potential increased availability of time results in more education. Additionally, as templates can be adapted in response to new governmental documentation regulations, assessment of compliance with these regulations will be needed, which we did not assess. In all, implementation of this electronic system was well received, quickly learned, led to more efficient and thorough documentation, and was more immediately available than transcribed notes, all of which are consistent with extant literature (5–9). Interestingly, our findings suggested a time savings (17.5 minutes per note among the residents who had used both typing and handwriting), though two earlier studies demonstrated no difference in time spent on documentation (5–8). Formal evaluation of documentation time in each medium is needed.