| OCR Text |
Show Current uses/developments The value of probabilistic record linkage has been substantiated by several studies in the area of public health and injury surveillance. An excellent example is the Crash Outcome Data Evaluation System (CODES) project which involves several states including Utah. Similarly, Australian researchers have effectively analyzed outcomes of road crashes by linking together data routinely collected by various authorities dealing with road crash casualties (Rosman, 1995). Huntsman Cancer Institute in Salt Lake City, Utah has conducted cohort and inter-generational studies by linking birth and death data. Other examples include surveillance of sexually transmitted diseases (Calzavara et al., 1990), analysis of hospital readmission rates (Henderson, Goldacare, & Simmons, 1989), linking maternity and neonatal records of hospital discharge (Smalls, Pickering & McKeller 1987), and linking vital statistics birth/infant death records and hospital discharge records for epidemiological studies (Herrchen, Gloud & Nesbitt, 1997). Linking mortality data to morbidity data, tracking cases longivity, and developing survey frames are some other applications (Jaro, 1995). Data linkage and confidentiality Since one or more unique or "quasi-unique" identifiers are involved in data linkage which can permit a misuse of information, the importance of individual confidentiality immediately becomes self evident. When unique identifiers are present in the data with sensitive information obtained for matching, an extreme caution should be used in storing it. The unique identifiers retained for the purpose of linkage should be stripped off from the linked files being made available for access. References Calzavara, L., Coats, R., Craib, K., et al.(1990). Under-reporting of AIDS Cases in Canada: A Record Linkage Study. Canadian Medical Association Journal, 142(1), 36-39. Fellegi, I.P.& Sunter, A.B. (1969). A Theory for Record Linkage. Journal of the American Statistical Association, 64, 1183-1210. Henderson, J., Goldacare, M.J., & Simmons, H.M. (1989). Use of Medical Record Linkage to Study Readmission Rates. British Medical Journal, 299(6701), 709-713. Herrchen, B., Gloud, J.B., & Nesbitt T.S.(1997). Vital Statistics Linked Birth/Infant Death Record Linkage for Epidemiological Studies. Computer and Biomedical Research, 30, 290-305 Jaro, M.A.(1995). Probabilistic Linkage of Large Public Health Data Files. Statistics in Medicine, 14, 491 -98. Nanan, D.J., White, F. (1997). Capture-recapture: Reconnaissance of a Demographic Technique in Epidemiology. Health Canada-CDIC, 18(4), 1-7 Newcombe, H.B., Kennedy, J.M., Axford, S.J., & James, A.P. (1959). Automatic Linkage of Vital Records. Science, 130,954-59 Newcombe, H.B., & Kennedy, H.M. (1962). Record Linkage. Communications of the Association for Computing Machinery, 5, 563-6. Rosman DL. (1995). The Linkage of Hospital and Police Information on Road Crash Casualties: An Investigation of Alternative Methods. Road Injury Information Program Report Series No. 7. Nedlands (Australia), Department of Public Health, University of Western Australia. Smalls, M., Pickering, R. & McKeller, D. (1987). Linking Maternity and Neonatal Discharge Records. Community Medicine, 9(2), 171-175. 129 |
