How accurately does the Australian Immunisation Register identify children overdue for vaccine doses? A national...

The accuracy of data recorded in the Australian Immunisation Register (AIR) is important for assessment of population-level vaccine coverage but has not been assessed nationally since 2001. We undertook a cross-sectional study in five states in 2017 using standard criteria to validate AIR records classified as three months overdue for any vaccine at 12, 24 and 48 months. Of 2,000 records selected for audit, 905 were assessable, of which 124 (14%) were misclassified as overdue (errors). Among 563 general practice (GP) records, 91 (16.1%) were errors. Compared with Victoria (1/99; 1%), errors were significantly higher in Western Australia (11/106; 10.4%), Queensland (13/104; 12.5%), South Australia (23/110; 20.9%) and New South Wales (43/144; 29.9%); p < 0.01 for all. Among 165 council and community health centre providers, the overall error rate (17; 10.3%) was non-significantly lower than for GP providers, with an odds ratio (OR) of 0.6 and a 95% confidence interval (95% CI) of 0.3-1.1, and did not differ between states. Records were transmitted to the AIR by paper-based methods in 13 cases, with significantly higher error rates (7/13; 54%) than for practice management software (77/630; 12.2%); OR 9.8 (95% CI 2.8-36.4) or the AIR secure site (23/87; 26.4%); OR 2.6 (95% CI 1.4-4.5). Accuracy is increasingly important, with mandatory reporting to the AIR for all National Immunisation Program vaccines from July 2021, and best achieved by uniform use of practice management software.