Application of a record linkage software to identify mortality of enrolees of large integrated healthcare organisations. (10th November 2022)

• Record Type: Journal Article
• Authors: Zhou, Yichen; Liang, Zhi; Glenn, Sungching; Chen, Wansu; Xie, Fagen
• Is Part Of: International journal of business intelligence and data mining. Volume 22:Number 1/2(2023)
• Page Start: 264
• Page End: 285
• Publication Date: 2022-11-10
• Subjects: data cleaning -- data standardisation -- data matching -- mortality linkage; 006.312
• Languages: English
• ISSNs: 1743-8187

Abstract

Information on mortality is important for the improvement of public health and the conduct of medical research. Healthcare organisations typically lack complete and accurate information on mortality. This paper proposes a comprehensive process to link the records of the enrolees of a healthcare organisation with the death records of 2015 obtained from the California State via a commercial data linkage software. The developed linkage process has successfully identified 23, 628 and 21, 009 death records of health plan enrolees from the state file after the initial and second post-linkage, respectively. Validation of the linkage process against the deaths records documented in the internal systems of the organisation achieved a sensitivity of 97.5% and a positive predictive value of 88.7% at the time of initial linkage but increased to 99.4% in three years using more information available later. The linkage process demonstrated high accuracy and can be utilised to support various business needs.