19

discussion and provided multiple viewpoints about each discussion topic. Focus group responses were not audio-recorded, and therefore were not transcribed verbatim. Instead, responses were paraphrased using field notes. These field notes were analyzed to look for patterns and recurring themes, which are presented in the discussion section below. 16 Results Records review results A records review of documentation forms was used to determine which information elements were recorded or omitted. This review was conducted on all 161 records for the 2008 calendar year. Presented here are the most and least available data listed with the percentage of forms containing those pieces of information (Figure 2). Date of Event 98.76% ETT intubation 98.67% Patient name 98.14% Time of first assisted ventilation 50.89% Team leader signature 50.31% Reason resuscitation ended 45.34% Figure 2: The most and least often supplied pieces of information on the 161 code blue forms collected during 2008. Although each emergency event unfolds differently, the fields shown above are standardized elements and should contain data for every emergency event documented. Some data elements were administrative details, such as the date and patient name. Other data elements were used to track patient treatment and whether specific procedures were performed. For example, the use of an AED device should have been marked in all cases, to indicate whether it was used or not. A few data elements were situational and are marked with asterisks, indicating that those data elements were normalized and only tallied in situations when that information was provided as a part of patient care. For instance, the time of ETT intubation was only included when records indicated that an ETT was used on the patient. 17 Field observation results A total of 22 field observations were conducted between March 2008 and March 2009. Observations revealed that the beginning of an emergency event was difficult to document in real-time, because formal documentation only started once the code cart arrived. To compensate, recorders often retroactively documented the initial care steps, and recorders would often spend time after the conclusion of a code blue event finalizing the record. This was corroborated by feedback provided during the focus group sessions. Recorder duties were typically assigned to a nurse as directed by hospital policy, but other care providers (such as the pharmacist) sometimes performed documentation duties instead. This indicates that the pool of recorders is potentially quite large, extended even beyond the nursing staff. The observations also highlighted consistent problems with crowds and noise, with as many as twenty people in attendance. Overcrowding interfered with verbal communications due to excess noise and simultaneous conversations occurring around the patient. It also interfered with the collection of field observation data, because little space was available in the room for non-participants and observers. Survey results The survey generated a total of 36 responses from among the 12 STAT nurses and approximately 300 ICU nurses. These subgroups were the nurses most likely to be involved with the emergency documentation process. The response rate was a small (10%) compared to the subject pool, and this will be considered in the discussion. This sample shows the range of responses for the question about improving the documentation forms (Table 1). 18 Mean SD CI (95%) Information is available 3.78 0.80 (3.52, 4.04) Records are important 4.61 0.77 (4.36, 4.86) Useful in real-time 4.54 0.78 (4.29, 4.80) Helps summarize 4.43 0.88 (4.14, 4.72) Forms are easy to use 2.97 1.11 (2.61, 3.33) Mean SD CI (95%) Pen and paper 3.33 1.17 (2.95, 3.72) Cellphone 2.80 0.93 (2.50, 3.10) Pocket PC 2.72 0.85 (2.44, 3.00) Tablet PC 3.23 0.91 (2.93, 3.53) Computer workstation 3.11 1.30 (2.69, 3.54) Table 1: Survey responses on a scale of 1 (disagree) to 5 (agree) Five of the survey responses indicated that documenters were unhappy with the layout of the existing documentation forms. Responses cited “flow” and “workflow” as points of contention, particularly regarding the large block of checkbox elements near the beginning of the documentation form. “First, the checkmarked sections need to be placed at the end. I realize data collection is important for review, but not helpful during an event. … The forms are very cumbersome and not at all well organized for flow documentation. Most of the information must be filled in after the event.” – Subject 36 When asked about missing elements on the written record, five survey respondents commented that the initial set of checkboxes were often skipped or postponed until after the event had finished. Survey responses indicated that the information in the checkbox section did not always apply to every emergency, nor was it necessarily time-sensitive. “It's different every time and is NOT always predictable. Sequence of events are NOT always the same, it's hard to have a paper with pre-determined lines and boxes for meds.” – Subject 22 The survey results also highlighted the need to identify the information elements that were needed immediately and separate them from the elements that could be filled in after the event. This also suggested that the 

1   2    3    4    5    6   7   8   9   10   11   12   13   14   15   16   17   18   19   20   21   22   23   24   25