Teton Analytics achieved our Expert Determination data model-certification by working closely with Dr. Brad Malin of Vanderbilt University, one of the leading experts in the US on data privacy.
Data Privacy
Safe Harbor
The removal of 18 types of identifiers and that no residual information can identify the individual. With Safe Harbor, a data provider must remove 18 data fields associated with protected health information (PHI) such as zip codes, procedure codes and dates of service. This technique has certain benefits for protecting data, but it also has two key shortcomings.
First, it does not provide a method for linking patient identity across disparate datasets because of the way in which Safe Harbor eliminates patient identity attributes. Second, it does not allow for dates of events or services, which are crucial when creating a meaningful timeline of events or establishing causality for analysis. This is why it is rarely utilized for research-quality data among Life Sciences and Insurance markets.
Expert Determination
Applying statistical or scientific principles and a very small risk that the anticipated recipient could identify the individual.
With Expert Determination, HIPAA provides that a person with appropriate knowledge of and experience with generally accepted statistical and scientific principles and methods for rendering information not individually identifiable can serve as an expert for purposes of a HIPAA review. The role of the expert, as defined by Health and Human Services (HHS), is then to apply such principles and methods so as to determine that the risk is very small that the information could be used, alone or in combination with other reasonably available information, by an anticipated recipient to identify an individual who is a subject of the information.
Furthermore, the expert must document the methods and results that justify such determination. There is often a significant amount of privacy work required to normalize and transform native healthcare data to meet the requirements of an expert determination with the benefit being that the retained attributes enable de-identified patient linkage and the data retains the vast majority of key clinical fields.