Patrick James Darragh*, University of Toronto, Toronto, Canada Mat Trudel, University Health Network, Toronto, Canada Howie Abrams, University of Toronto, Toronto, Canada Dante Morra, University of Toronto, Toronto, Canada Track: Research Presentation Topic: Mobile & Tablet Health Applications Presentation Type: Poster presentation Submission Type: Single Presentation Last modified: 2013-09-25

If you are the presenter of this abstract (or if you cite this abstract in a talk or on a poster), please show the QR code in your slide or poster (QR code contains this URL).

Background: Clinical practice guidelines (CPGs) offer a useful tool to help health care professionals apply the
current standard of care for a particular disease. Currently, CPGs are disseminated mainly as large, complex
documents, which are difficult to access and navigate, and cumbersome to cross-reference for patients with multiple
comorbidities. With powerful mobile devices offering ubiquitous computing and powerful application development
platforms, there is potential to improve how clinicians utilize CPGs. While guideline apps are available for mobile
devices, they do not offer significant improvement in user experience compared to the current standard.

Objective: To develop a prototype iOS application that displays relevant recommendations from multiple clinical
guidelines when provided clinical data for a patient with comorbid chronic diseases .

Methods: The following Canadian guidelines were identified using MEDLINE, Google Scholar and directed
internet search: coronary artery disease, chronic kidney disease, congestive heart failure, diabetes mellitus, chronic
obstructive pulmonary disease, hypertension, dyslipidemia, atrial fibrillation. The most current guidelines were
included. Recommendations from guidelines were extracted by an internal medicine resident and included if
they were actionable and applicable to chronic disease management. A graphical decision tree that integrated
these recommendations was created in Omnigraffle and later translated into computer language through a coding
model. A prototype of the iOS application was created using Omnigraffle, Keynote, and XCode. The concept and
application prototype was presented to physicians, allied health professionals and internal medicine residents, and
feedback was collected informally.

Results: Over 20 Canadian guidelines were considered and 8 were identified as the most recent guidelines for
the identified chronic diseases. Sixty-two recommendations were extracted from these guidelines and resulted
in 32 patient management actions and 52 patient data inputs. The combination of inputs, logic (and/or/not),
recommendations and actions resulted in an integrated decision-tree with minimum 4-levels per decision. A
prototype iOS application based on the decision tree was presented to 33 individuals, and feedback on the model
included: target users of the application are trainees, physicians, allied healthcare professionals, and nurse
practitioners; standardization of guideline selection and interpretation should be considered; this application could
be used as teaching tool for trainees and knowledge translation mechanism for content providers; and there is
potential to integrate with electronic medical records.

Conclusion: The complexity involved in translating recommendations from clinical guidelines into computer code
illustrates why guidelines are difficult to implement in current clinical practice. It is feasible to develop such an iOS
application; however, further studies are needed to examine its impact on improving guideline adherence and patient
outcomes.