14

time during in-hospital cardiac arrest emergencies, also referred to as “code blue” events. The information requirements during code blue events are different from the information required during regular medical care because code blue emergencies are chaotic, high stress, and time-sensitive. The real-time record of medications and procedures is an essential part of managing patient care. Documentation records also serve an important secondary purpose by supporting administrative functions like patient safety and risk management. This research also brings into focus the tradeoff between structured and unstructured data which underscores the tradeoff between rapid data entry and flexible data entry. It was hoped that visual design improvements would results in improved usability and a simplified the data entry experience. While the redesign effort succeeded in creating a more usable form, the overall impact on data quality and completeness was limited. Overview This research project was conceived in 2008, after the University of Washington hospitals (UW Medicine) had just responded to legal challenges regarding the care during cardiac arrest emergencies. Complaints were leveled in response to patients who were injured as a result of in-hospital cardiac arrest while in-patients at UW Medicine facilities. The hospital risk management staff believed that the physicians and nurses had done everything possible to provide the best care. However, the medical 2 documentation was incomplete. As a result, the documentation could not be used as a comprehensive record of treatment during the resulting medical-legal dispute. In response, physicians, nurses, patient safety representatives, risk management, and hospital administrators people at all levels of the hospital organization began to investigate the question of documentation quality. UW Medicine was in the midst of reviewing its code blue response policy, including an overhaul of its recommended care practices, provider training, code cart equipment, and documentation forms. The code blue documentation form served as the focal point of this research project, which consisted of a needs assessment, a redesign, and an evaluation study. A mixed-methods informatics-based needs assessment, utilizing both qualitative and quantitative methods, was used to thoroughly describe the problem of incomplete documentation. Once the specific problem was described and its underlying causes investigated, the documentation form was redesigned. The redesigned form was then assessed through an evaluation study. Because the later parts of the project relied on the results of the previous sections, this project has been divided into three distinct research activities, each with its accompanying manuscript. Needs assessment The first manuscript provides details about the records review, field observations, user survey, and focus group meetings used to assess the existing code blue documentation form. The needs assessment quantified the problem of missing information. The needs assessment also explored some of the reasons why information was incompletely documented, including a mismatch between the flow of a code blue emergency and the layout of the documentation form. Based on the results of the needs assessment, the most promising solution was a redesign of the code blue documentation form. This redesign would 3 be able to address usability and workflow concerns, while also satisfying institutional interest in updating the form contents to match the current standards of care. Redesign The second manuscript describes the iterative participatory design process used to create a redesigned documentation form. Based on usability principles and visual design principles, prototype forms were created and presented to stakeholders during a series of focus groups. These focus groups were conducted to ensure clinical utility and maximize end-user acceptance. The redesign included changes to the layout of the documentation form, which were implemented to better reflect the way a cardiac arrest event unfolds in real-time. In addition to modifying the layout of the documentation form, the redesign also incorporated feedback from a visual designer. The inclusion of a visual designer was based on literature which indicates that improving the readability of a form can affect the form’s overall usability. Given the real-time nature of these high-stress cardiac arrest emergencies, improved readability was expected to result in more efficient usage of the documentation forms. Evaluation study The third manuscript describes the evaluation study process, including the creation of study materials and the experimental design used to recruit participants and analyze their use of the forms. The results of the evaluation study showed that the redesigned form was more readable, and it demonstrated improved usability. This addressed some of the major complaints attributed to the old code blue documentation forms, such as workflow mismatch and ease of use. Despite this improvement in form usefulness, the quality of the data remained largely unchanged. Some types of data were collected more reliably, but other types of data were still missing. This can be attributed to the tradeoffs when trying to construct a standardized form to collect 

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