Soko Setoguchi1, Ying Zhu2, Jessica J Jalbert2, Lauren A Williams2, Chih-Ying Chen2. 1. From the Division of Pharmacoepidemiology, Department of Medicine, Brigham and Women's Hospital, Boston, MA (S.S., J.J.J., L.A.W., C.-Y.C.); Duke Clinical Research Institute and Department of Medicine, Duke University School of Medicine, Durham, NC (S.S.); and Department of Biostatistics, Graduate School of Medicine, University of Tokyo, Tokyo, Japan (Y.Z.). soko@post.harvard.edu. 2. From the Division of Pharmacoepidemiology, Department of Medicine, Brigham and Women's Hospital, Boston, MA (S.S., J.J.J., L.A.W., C.-Y.C.); Duke Clinical Research Institute and Department of Medicine, Duke University School of Medicine, Durham, NC (S.S.); and Department of Biostatistics, Graduate School of Medicine, University of Tokyo, Tokyo, Japan (Y.Z.).
Abstract
BACKGROUND: Linking patient registries with administrative databases can enhance the utility of the databases for epidemiological and comparative effectiveness research. However, registries often lack direct personal identifiers, and the validity of record linkage using multiple indirect personal identifiers is not well understood. METHODS AND RESULTS: Using a large contemporary national cardiovascular device registry and 100% Medicare inpatient data, we linked hospitalization-level records. The main outcomes were the validity measures of several deterministic linkage rules using multiple indirect personal identifiers compared with rules using both direct and indirect personal identifiers. Linkage rules using 2 or 3 indirect, patient-level identifiers (ie, date of birth, sex, admission date) and hospital ID produced linkages with sensitivity of 95% and specificity of 98% compared with a gold standard linkage rule using a combination of both direct and indirect identifiers. CONCLUSIONS: Ours is the first large-scale study to validate the performance of deterministic linkage rules without direct personal identifiers. When linking hospitalization-level records in the absence of direct personal identifiers, provider information is necessary for successful linkage.
BACKGROUND: Linking patient registries with administrative databases can enhance the utility of the databases for epidemiological and comparative effectiveness research. However, registries often lack direct personal identifiers, and the validity of record linkage using multiple indirect personal identifiers is not well understood. METHODS AND RESULTS: Using a large contemporary national cardiovascular device registry and 100% Medicare inpatient data, we linked hospitalization-level records. The main outcomes were the validity measures of several deterministic linkage rules using multiple indirect personal identifiers compared with rules using both direct and indirect personal identifiers. Linkage rules using 2 or 3 indirect, patient-level identifiers (ie, date of birth, sex, admission date) and hospital ID produced linkages with sensitivity of 95% and specificity of 98% compared with a gold standard linkage rule using a combination of both direct and indirect identifiers. CONCLUSIONS: Ours is the first large-scale study to validate the performance of deterministic linkage rules without direct personal identifiers. When linking hospitalization-level records in the absence of direct personal identifiers, provider information is necessary for successful linkage.
Keywords:
Medicare; databases, factual; information storage and retrieval; insurance claim reporting; medical record linkage; medical records systems, computerized; registries
Authors: Qixing Liang; Sarah Ward; Francis D Pagani; Shashank S Sinha; Min Zhang; Robert Kormos; Keith D Aaronson; Andrew D Althouse; James K Kirklin; David Naftel; Donald S Likosky Journal: Ann Thorac Surg Date: 2017-12-19 Impact factor: 4.330
Authors: Fenglong Xie; Lisandro D Colantonio; Jeffrey R Curtis; Monika M Safford; Emily B Levitan; George Howard; Paul Muntner Journal: Am J Epidemiol Date: 2016-09-19 Impact factor: 4.897
Authors: Jeffrey R Curtis; Lang Chen; Aseem Bharat; Elizabeth Delzell; Jeffrey D Greenberg; Leslie Harrold; Joel Kremer; Soko Setoguchi; Daniel H Solomon; Fenglong Xie; Huifeng Yun Journal: Arthritis Care Res (Hoboken) Date: 2014-12 Impact factor: 4.794
Authors: Andrew W Hoel; Adrienne E Faerber; Kayla O Moore; Niveditta Ramkumar; Benjamin S Brooke; Salvatore T Scali; Art Sedrakyan; Philip P Goodney Journal: J Vasc Surg Date: 2017-02-17 Impact factor: 4.268
Authors: Theodoros V Giannouchos; Alva O Ferdinand; Gurudev Ilangovan; Eric Ragan; W Benjamin Nowell; Hye-Chung Kum; Cason D Schmit Journal: J Am Med Inform Assoc Date: 2021-07-30 Impact factor: 4.497
Authors: Soko Setoguchi; Lynne Warner Stevenson; Garrick C Stewart; Deepak L Bhatt; Andrew E Epstein; Manisha Desai; Lauren A Williams; Chih-Ying Chen Journal: BMJ Date: 2014-05-08