USACM Comments on Regulations Governing Protection of Research Participants

By Renee Dopplick, ACM Director of Public Policy
January 7, 2016

USACM submitted comments on the proposed updates to the federal regulations for research involving human subjects. These regulations, known as the Common Rule, are used by multiple federal agencies. They have significance for computing professionals conducting behavioral, analytic, and clinical studies, especially in the subfields of computer security, information assurance, computer networks, computer-human interaction, accessibility, and usability. The regulations also have significance for the sensitive data of individuals participating in federally funded research studies.

USACM supports the overall goals of the proposed updates and welcomes the modernization of the Common Rule to support technological advancements. The proposed changes have the potential to streamline research processes while maintaining appropriate protections for sensitive information. USACM is supportive of the goal of better addressing the complexity of balancing privacy and autonomy concerns with the greater societal value of scientific research.

In its comments, USACM encourages the regulators to consider the following guiding principles:

  • Apply a functional approach to data information protection
  • Consider the potential harms and benefits of re-identification of data
  • Provide flexibility to consider the degree of risk of data reidentifiability
  • Consider the proportionality of data reidentifiability

The comments were developed by the ACM U.S. Public Policy Council (USACM), which serves as the focal point for ACM’s interaction with the U.S. government in all matters of U.S. public policy related to information technology. USACM is comprised of computer scientists, educators, researchers, and other technology professionals. USACM members have experience with privacy, security, data mining, and machine learning algorithms that are used to extract patterns and understanding from large datasets.