Naturalistic study of guideline implementation tool use via evaluation of website access and physician survey

Clinical guidelines support decision-making at the point-of-care but the onus is often on individual users such as physicians to implement them. Research shows that the inclusion of implementation tools in or with guidelines (GItools) is associated with guideline use. However, there is little research on which GItools best support implementation by individual physicians. The purpose of this study was to investigate naturalistic access and use of GItools produced by the American Academy of Neurology (AAN) to inform future tool development.

Website accesses over six months were summarized for eight AAN guidelines and associated GItools published between July 2012 and August 2013. Academy members were surveyed about use of tools accompanying the sport concussion guideline. Data were analyzed using summary statistics and the Chi-square test.

The clinician summary was accessed more frequently (29.0%, p < 0.001) compared with the slide presentation (26.8%), patient summary (23.2%) or case study (20.9%), although this varied by guideline topic. For the sport concussion guideline, which was accompanied by a greater variety of GItools, the mobile phone quick reference check application was most frequently accessed, followed by the clinician summary, patient summary, and slide presentation. For the sports concussion guideline survey, most respondents (response rate 21.8%, 168/797) were aware of the guideline (88.1%) and had read the guideline (78.6%). For GItool use, respondents indicated reading the reference card (51.2%), clinician summary (45.2%), patient summary (28.0%), mobile phone application (26.2%), and coach/athletic trainer summary (20.2%). Patterns of sports concussion GItool use were similar between respondents who said they had and had not yet implemented the guideline.

Developers faced with resource limitations may wish to prioritize the development of printable or mobile application clinician summaries, which were accessed significantly more than other types of GItools. Further research is needed to understand how to optimize the design of such GItools.