Ethical Development of Digital Phenotyping Tools for Mental Health Applications: Delphi Study

Abstract

Journal of Medical Internet Research - International Scientific Journal for Medical Research, Information and Communication on the Internet #Preprint #PeerReviewMe: Warning: This is a unreviewed preprint. Readers are warned that the document has not been peer-reviewed by expert/patient reviewers or an academic editor, may contain misleading claims, and is likely to undergo changes before final publication, if accepted, or may have been rejected/withdrawn. Readers with interest and expertise are encouraged to sign up as peer-reviewer, if the paper is within an open peer-review period. Please cite this preprint only for review purposes or for grant applications and CVs (if you are the author). Background: Digital phenotyping refers to new approaches to measuring behavior, physiological states, and cognitive functioning by applying algorithms, often generated by machine learning, to gather moment-by-moment physiological and biometric data using smartphones or other sensors, such as pulse, keyboard interactions, or voice features. For example, collecting smartphone data regarding screen taps in order to assess an individual’s risk of having a psychotic episode. Digital phenotyping applications to assess behavioral states and mental disorder show promise for medical uses, and also have potential uses in education, the military, insurance, and the criminal justice system. At the same time, digital phenotyping raises novel ethical questions regarding the risks and benefits of collecting massive amounts of highly granular personal data, as well as the collection of sensitive personal information outside of traditional ethical and regulatory frameworks for protecting patients and health data. Objective: Use a modified Delphi technique to arrive at consensus statements regarding key areas of ethical guidance for mental health applications of digital phenotyping. Methods: A modified Delphi technique was used in order to identify the emerging ethical challenges posed by digital phenotyping for mental health applications and formulate guidance for addressing these challenges. Experts in digital phenotyping, data science, mental health, law and ethics participated in the study. Through an iterative process of 3 rounds involving interviews and surveys, the panel arrived at consensus recommendations. The experts focused primarily on clinical applications for digital phenotyping of mental health, but addressed items relevant to other domains. Results: The consensus statements address key ethical areas: privacy and data protections, validation and utility, bias and fairness, and transparency. Conclusions: The Delphi study found agreement on a number of areas regarding the priority ethical issues in the development of digital phenotyping for mental health applications. Standards and guidelines for key areas of digital phenotyping, such as privacy standards outside of health institutions and development of algorithms, are still evolving and thus consensus statements identified general principles regarding ethical digital phenotyping of mental health. Clinical Trial: not applicable