作者: Alexander Swart , Nicola S. Meagher , Marina T. van Leeuwen , Kun Zhao , Andrew Grulich
关键词:
摘要: Objective: To examine the validity and impact of record linkage using name code compared to full records. Methods: A registry 45,419 opioid substitution clients (1985–2007) was linked with national population-based death cancer registries registrant's name, date birth, sex, state, postcode death. Records were then first two letters given surname (2×2 code). Sensitivity specificity computed regression analysis used identify factors related accuracy. Standardised mortality ratios (SMR) standardised incidence (SIR) estimated. Results: The sensitivity 65.31% 99.91% for records 76.81% 99.89% records. Registrants' age sex accuracy associated risk false linkages. Death risks (SMR 6.98, 95%CI 6.77–7.19; SIR 1.16, 1.08–1.24) significantly under-estimated 4.39, 4.23–4.56; 0.92, 0.85–0.99). Conclusion: Record 2×2 has low but high specificity, resulting in conservative estimates risk. This may translate meaningful differences outcomes.