Design for Use and Acceptance of Tracking Tools in Healthcare

Thesis (Ph.D.)--University of Washington, 2013Tracking tools that collect patient-generated data can have a major impact on health outcomes and patient-clinician communication. The relationship between the patient and the clinician can be fundamentally altered when the patient uses a tracking tool. Clinicians can gain a more holistic understanding of patients instead of relying predominantly on clinic visits for input, and patients can better understand how to manage their condition tool use. Yet, acceptance of tracking tools remains low. In my dissertation work, I investigated patients' use of researcher-driven electronic Patient-Reported Outcome (e-PRO) and patient-driven Personal Informatics tracking tools during cancer treatment. In one study, patients who frequently used PRO tools had lower end-of-study symptom distress than those who used the tool once or not at all. Patient attributes, such as age, gender, and educational attainment, were not found to be an indicator of frequent voluntary use. In a second study, I analyzed self-tracking attitudes and behaviors of 25 women with breast cancer. Results showed that patients' tracking behaviors outside of the research context were fragmented and sporadic, compared to when they were given personal informatics tool. Participants used information they had collected on the tool to view patterns among symptoms, feel psychosocial comfort, and improve symptom communication with clinicians. To better understand the reasons why most patients do not realize the opportunity of using a tracking tool as a path to these benefits and to further inform future tool design, I propose two theoretical models: (1) the Model of Use of Tracking Tools by Patients and Clinicians (MUTT-PC) and (2) MultiTrack. MUTT-PC illustrates factors in symptom communication and feedback in scenarios that use no tracking tools, a patient-driven tracking tools, researcher- or clinician-driven tracking tools, or, in a proposed future scenario, symptom tracking tools that are used collaboratively by the patient and clinician. MultiTrack provides a deeper understanding of tradeoffs in requirements for tracking tool developers, by enumerating multiple dimensions to be considered in design for use and acceptance: (1) the patient, (2) clinician, (3) data collected and presented, and (4) the tracking tool itself. This work contributes to health informatics, health services, human-computer interaction, and information and management science. In this dissertation, I propose the use of a novel framework that separates clinical and personal usefulness of data from the perceived value of the tracking tool itself. Further, incorporating the context of healthcare into tracking tool development considerations promises that both clinicians and patients can realize the value of self-tracking as next-generation tracking tools are deployed