Accepted for/Published in: Journal of Medical Internet Research
Date Submitted: Aug 4, 2020
Date Accepted: Dec 18, 2020
Warning: This is an author submission that is not peer-reviewed or edited. Preprints - unless they show as "accepted" - should not be relied on to guide clinical practice or health-related behavior and should not be reported in news media as established information.
Key variables for developing effective e-health intervention designs for those with, and without, mental health disorders: navigational depth, number of hyperlinks per page, use of a popup menu, reading level, page length, participant abilities with the technology.
ABSTRACT
Background:
E-health applications offer the potential to increase service convenience, appeal, and responsiveness, while also providing an avenue to expand the ability to tailor services in order to improve individual relevance, engagement, and use. It is critical in order to achieve these goals that the designs of e-health applications are intuitive to use. Limited research exists on designs that work for those with a severe mental illness, many of whom have difficulty traveling to treatments, reject or infrequently seek treatment, and tend to discontinue treatments for significant periods of time.
Objective:
The purpose of this study was to evaluate the influence of 12 design variables (e.g., website depth, reading level, use of navigational lists) on the usability of e-health application websites for those with, and without, mental health disorders (including severe mental illness).
Methods:
A 212-4 fractional factorial experimental design was used to specify the designs of 256 e-health websites. This systematically varied the 12 design variables. The final destination contents of all websites were identical, only the designs of the navigational pages varied. The 12 design elements were manipulated systematically to allow assessment of combinations of design elements rather than only one element at a time. Participants (n=222) sought the same information on 8 randomly selected websites, from the 256. Classification and regression tree analyses were used to identify effects among the 12 variables on participants’ abilities to locate information, for the sample overall and each of three diagnostic groups of participants (schizophrenia spectrum disorders--SSD, other mental illnesses, and no mental illness). Mixed effect regressions, which accounted for the dependency of the 8 observations within participants were used to test for main effects and interactions.
Results:
The best and worst designs were identified for each of the three groups, and the sample overall. The depth of a website’s navigation, that is, the number of screens/pages users needed to navigate through to find desired content, had the largest influence on usability (ability to find information) and efficiency (time to find information). The worst performing designs for those with SSD had a 9% success rate, the best a 51% success rate: the navigational designs made a 42% difference in usability. For the group ‘other mental illnesses’ the design made a 50% difference, and for those with no mental illness a 55% difference. The websites with the highest usability all had several key similarities, as did the websites with the poorest usability.
Conclusions:
It is possible to identify evidence-based strategies for designing e-health applications that result in markedly higher levels of usability, even for those who are less savvy with technology. These improvements in design benefit all users. For those with SSD, or other severe mental illnesses, there are designs that are highly effective. The best designs have key similarities, but are able to vary on some characteristics. Clinical Trial: Not applicable.
Citation
