Currently submitted to: JMIR mHealth and uHealth
Date Submitted: Feb 23, 2026
Open Peer Review Period: Feb 23, 2026 - Apr 20, 2026
(currently open for review)
Warning: This is an author submission that is not peer-reviewed or edited.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.
Patient Trust and Adherence in an AI Skin Cancer App: a randomised-controlled experimental study on the effects of Teledermatologist Supervision and Diagnostic Risk Level
ABSTRACT
Background:
Given the rising incidence of skin cancer, efficient triaging of skin lesions is ever-more critical to ensure healthcare systems can cope. To this end, skin cancer mobile healthcare apps can be leveraged, but it is essential to understand how patients’ trust of and adherence to the apps’ advice can be optimised such that they are neither underused nor overused such that unnecessary appointments are minimised without compromising detection.
Objective:
To test whether human-teledermatologist supervision in an AI-driven skin cancer app increases patient trust and adherence intentions, compared with AI-only advice, and how this is influenced by patient motivational context (curiosity or concern) and risk assessment level.
Methods:
Randomised controlled crossover online experiment. Conducted in May–June 2025 as a Single-centre study among Dutch adults via an academic hospital-based patient panel. Of the 2,707 patient panel members, 879 participated (response rate: 32.5%). Participants were aged 18–93 (mean age: 62.5), 50.4% female, 21% with prior skin cancer. Participants were randomly allocated to motivation conditions (concern vs curiosity). Participants completed four simulated mobile app trials varying in advice source (AI-only vs hybrid teledermatologist+AI) and skin cancer risk assessment (high vs low). After each scenario, trust and adherence intention were rated. Trust (0–100 scale) and intention to adhere to advice (recoded ordinal variable). Models included advice source, risk assessment, motivation, interactions, and participant random intercept.
Results:
Trust was significantly higher for hybrid risk assessments (mean=76.9, SE = 0.73) compared to AI-only (mean=65.68; p<.001; partial η² = .22). The odds of adherence were 1.7 times greater with hybrid versus AI-only (OR=1/0.59=1.7, 95% CI [1.28, 2.22], p<.001). High-risk assessments increased adherence (OR=2.9, 95% CI [2.2, 3.8], p<.001), with a moderated effect by motivation (p=.033), showing stronger adherence with concern versus curiosity. The effect of advice source on adherence was fully mediated by trust.
Conclusions:
Teledermatologist supervision in AI-driven skin cancer apps robustly increases patient trust and adherence intentions, especially for high-risk advice and concerned users. Integrating human supervision with AI supports maximising patient adherence to advice, and thus improving triaging efficiency. Clinical Trial: The study was preregistered on the Open Science Framework (https://osf.io/64pjn/) on 26th May 2025.
Citation
Request queued. Please wait while the file is being generated. It may take some time.
Copyright
© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.