Seeing through health information technology: the need for transparency in software, algorithms, data privacy, and regulation*
Bonnie J. Kaplan
Abstract
The paper discusses transparency issues related to the four areas of software and algorithms, data and privacy, contracts and liability, and intellectual property through the lens of health information technologies. Arguing that an interdisciplinary approach makes the scope of issues more apparent and can lead to multi-faceted recommendations for addressing these issues, the paper contributes by bringing together scholarship and studies in law, medical informatics, ethics, and other domains. Software (algorithms, clinical decision support and electronic record systems, and popular commercial applications and devices) is too opaque to be used with confidence. Data and processes on which software is based are unknown, and, like system contracts, shrouded in intellectual privacy protections that prevent disseminating knowledge of software problems. Regulatory distinctions (e.g., between clinical and research data in HIPAA and the Common Rule) and the multiple agencies (e.g., FDA, FCC, FTC) governing different aspects of health-related information technologies make it difficult to understand what is regulated and how. Click-through privacy policies and end-user agreements for commercial applications are not generally comprehensible and make informed consent impossible. Most people are unaware of how data is gathered, aggregated, re-identified, and used in ways they may not sanction. Because big data makes all data health data, existing regulatory boundaries are increasingly irrelevant, people are affected in all aspects of their lives without their knowledge, and privacy and autonomy are compromised. The paper suggests potential remedies so that transparency and a multi-faceted interdisciplinary approach can inform robust public discussion and corrective regulatory, technological, organizational, educational, and social actions concerning information technologies.