A recent study found a way to ensure that secure EHR data can be exchanged and linked across multiple sites, improving patient care and further benefitting future research.
Published in the Journal of the American Medical Informatics Association, the study’s authors “developed and distributed a software application that performs standardized data cleaning, preprocessing, and hashing of patient identifiers to remove all protected health information.” From there, the authors “linked individual records using a central honest broker with an algorithm that assigns weights to hash combinations in order to generate high specificity matches.”
While the study was relegated to the Chicago metropolitan area, results showed that by using software that standardizes the assignment of a unique seeded hash identifier merged through a third-party, facilities “can enable large-scale secure linkage of EHR data for epidemiologic and public health research.”
While there has been an increased push for secure HIE use across the nation, the report’s authors explained that many areas lack central HIEs “that meet the needs and concerns of local healthcare systems, researchers, and networks.”
To overcome this problem, we successfully developed and implemented an IRB-approved approach using a distributed software application that enabled multiple, otherwise unaffiliated (and competing), healthcare institutions to aggregate longitudinal clinical data on approximately 5 million residents of a large United States city…With the rapid increase in adoption of EHRs as part of routine clinical care, a secure method of aggregating records across care sites may present an efficient complement to prospective data collection for research or public health purposes.
The authors used pre-existing EHR data, which came from either local data warehouses or medical record systems.
“The data proposed for use in the HealthLNK research project included none of the elements defined as PHI under federal HIPAA regulations other than patient ZIP codes (included to enable later mapping of disease distribution),” the report stated.
Health data privacy and security concerns were also addressed. Certain patient information, such as names and dates of birth, was combined with other patient information to create uniformly distributed information groups, the authors explained.
“To protect against dictionary or ‘rainbow-table; attacks, we seeded the hash algorithm by requiring users of the hashing application to enter a passphrase and passcode distributed by a team member not involved in managing the inbound hashed files,” the report said. “Seeding the hashing algorithm ensured that only users knowing the seed could use the HashIDs to link individual records.”
The authors added that there can be an improved framework for accurate measuerment of a population’s health with “a combination of distributed query methods with secure and encrypted identity disambiguation.”
Overall, the authors were optimistic about using such options in the future, especially as EHRs increase in popularity and healthcare officials seek out ways to improve care. Using secure EHR data that can be linked to multiple sites presents researchers and public health officials with “an opportunity to efficiently re-purpose existing clinical data to generate new insights and guide regional interventions,” the authors stated.
View the original content and more from this author here: http://ift.tt/1BD4D92
from health IT caucus http://ift.tt/1GA42Rr
via IFTTT
No comments:
Post a Comment