R L Kruse1, B G Ewigman, G C Tremblay. 1. Center for Family Medicine Science in the Department of Family and Community Medicine, School of Medicine, University of Missouri-Columbia, 65212, USA.
Abstract
OBJECTIVE: This report describes a method for linking separate confidential data sets that contain personal identifying information while preserving required anonymity. METHODS: Research data were linked with child abuse and neglect (CAN) report data by an independent "safe" analyst using an identical set of unique identifier codes assigned to each case in both data sets after all personal identifiers had been removed. RESULTS: The research team never learned CAN report status of individuals, the state agency never saw the research data, and the desired analyses were completed using the merged data set. CONCLUSIONS: The method was successfully used to merge data from separate sources without divulging confidential information.
OBJECTIVE: This report describes a method for linking separate confidential data sets that contain personal identifying information while preserving required anonymity. METHODS: Research data were linked with child abuse and neglect (CAN) report data by an independent "safe" analyst using an identical set of unique identifier codes assigned to each case in both data sets after all personal identifiers had been removed. RESULTS: The research team never learned CAN report status of individuals, the state agency never saw the research data, and the desired analyses were completed using the merged data set. CONCLUSIONS: The method was successfully used to merge data from separate sources without divulging confidential information.
Authors: Stephen B Johnson; Glen Whitney; Matthew McAuliffe; Hailong Wang; Evan McCreedy; Leon Rozenblit; Clark C Evans Journal: J Am Med Inform Assoc Date: 2010 Nov-Dec Impact factor: 4.497