Patient preferences for health information technologies: a systematic review

Background: Advances in patient-facing health information technology (HIT) promise to improve health care delivery and patient outcomes. Low utilization of HIT suggests that the preferences of patients may not be adequately guiding the development of these technologies. This systematic review offers an assessment of published evidence regarding patient preferences for HIT. Methods: Articles addressing preferences for HIT from patient and other end-user groups published up through 2020 were identified from PubMed, Web of Science, Scopus and via hand searching. Articles that used quantitative stated-preference methods to explore preferences for HIT were eligible for inclusion. Studies that explored attitudes towards HIT without eliciting trade-offs were excluded. Critical appraisal of study quality was conducted using the PREFS checklist and quality criteria identified by the US Food and Drug Administration including heterogeneity analysis and patient engagement in study design. We conducted thematic analysis of the main preference findings from each study to synthesize patient and enduser preferences for HIT. The review was not registered and authors received no funding to conduct the review. Results: The search yielded 7,299 unique articles, 59 of which were ultimately included in the review. Studies explored preferences for telemedicine (n=30), patient portals (n=12), mHealth (n=9) or multiple HITs (n=8). Preference elicitation methods included direct elicitation (n=26), discrete-choice experiments (n=13), conjoint analysis (n=6), contingent valuation (n=5), and ranking exercises (n=9). Studies had a mean PREFS score of 3.51 out of 5. Forty-two studies conducted preference heterogeneity analysis and only 20 included patients in study design. Thematic meta-analysis indicated that patients prefer HIT that is convenient and lower cost, but does not sacrifice quality, and preferences varied by demographic features such as age as well as depending on the type of health information being communicated. Conclusions: Patient and end-users have distinct preferences for the use of HIT in their medical care. It is timely that researchers and healthcare administrators consider these preferences for HIT given its rapid uptake amidst the COVID-19 pandemic. Although this literature demonstrates that patients can be engaged as participants in preference studies to identify meaningful aspects of HIT, the field was limited in its inclusion of patients in the design of such studies. Future development of HIT should be guided by high-quality preference research that integrates patients in all stages in the design and implementation of HIT. © Journal of Hospital Management and Health Policy. All rights reserved.