Cecilia L Moore1, Heather F Gidding2, Matthew G Law3, Janaki Amin3. 1. The Kirby Institute, UNSW Medicine, The University of New South Wales, Sydney 2052, Australia. Electronic address: cmoore@kirby.unsw.edu.au. 2. School of Public Health and Community Medicine, UNSW Medicine, The University of New South Wales, Sydney 2052, Australia. 3. The Kirby Institute, UNSW Medicine, The University of New South Wales, Sydney 2052, Australia.
Abstract
OBJECTIVES: To examine the validity of deterministic compared to probabilistic record linkage in the ascertainment of hospitalizations in two linked cohorts. STUDY DESIGN AND SETTING: HIV-negative (HIV-ve) (n = 1,325) and HIV-positive (HIV+ve) gay and bisexual men (n = 557) recruited in Sydney, Australia, were probabilistically and deterministically linked to a statewide hospital registry (July 2000-June 2012). RESULTS: Using probabilistic linkage as the reference standard, deterministic linkage had higher specificity but much lower sensitivity [34.67% (95% confidence interval: 33.44, 35.92)]. A disproportionate number of links missed were individuals with poorer socioeconomic and health indicators, including HIV status. Risk of hospitalization compared to the general male population [HIV+ve standardized incidence ratio (SIR) = 1.45 (1.33-1.59); HIV-ve SIR = 0.72 (0.67-0.78)] was significantly underestimated when deterministic linkage was used [HIV+ve SIR = 0.46 (0.37-0.58); HIV-ve SIR = 0.29 (0.24-0.35)]. The impact of linkage strategy on the calculation of incidence rate ratios (IRRs) was less, but a greater discrepancy in IRRs was seen for diagnostic categories where event rates were low or where the sensitivity of the deterministic linkage was differential between the two cohorts. CONCLUSION: Linkage without proven high sensitivity and specificity should be carefully considered. In circumstances of undetermined sensitivity, SIRs should not be calculated as the extent of underestimation is unknown. The comparison of linked events within or between cohorts is more robust to linkage misclassification; however, selection bias does affect estimates and should be considered before linkage.
OBJECTIVES: To examine the validity of deterministic compared to probabilistic record linkage in the ascertainment of hospitalizations in two linked cohorts. STUDY DESIGN AND SETTING: HIV-negative (HIV-ve) (n = 1,325) and HIV-positive (HIV+ve) gay and bisexual men (n = 557) recruited in Sydney, Australia, were probabilistically and deterministically linked to a statewide hospital registry (July 2000-June 2012). RESULTS: Using probabilistic linkage as the reference standard, deterministic linkage had higher specificity but much lower sensitivity [34.67% (95% confidence interval: 33.44, 35.92)]. A disproportionate number of links missed were individuals with poorer socioeconomic and health indicators, including HIV status. Risk of hospitalization compared to the general male population [HIV+ve standardized incidence ratio (SIR) = 1.45 (1.33-1.59); HIV-ve SIR = 0.72 (0.67-0.78)] was significantly underestimated when deterministic linkage was used [HIV+ve SIR = 0.46 (0.37-0.58); HIV-ve SIR = 0.29 (0.24-0.35)]. The impact of linkage strategy on the calculation of incidence rate ratios (IRRs) was less, but a greater discrepancy in IRRs was seen for diagnostic categories where event rates were low or where the sensitivity of the deterministic linkage was differential between the two cohorts. CONCLUSION: Linkage without proven high sensitivity and specificity should be carefully considered. In circumstances of undetermined sensitivity, SIRs should not be calculated as the extent of underestimation is unknown. The comparison of linked events within or between cohorts is more robust to linkage misclassification; however, selection bias does affect estimates and should be considered before linkage.
Authors: Garrett Prestage; Limin Mao; Susan Kippax; Fengyi Jin; Michael Hurley; Andrew Grulich; John Imrie; John Kaldor; Iryna Zablotska Journal: AIDS Behav Date: 2009-02-06
Authors: Vivienne J Zhu; Marc J Overhage; James Egg; Stephen M Downs; Shaun J Grannis Journal: J Am Med Inform Assoc Date: 2009-06-30 Impact factor: 4.497
Authors: Monique F Kilkenny; Helen M Dewey; Vijaya Sundararajan; Nadine E Andrew; Natasha Lannin; Craig S Anderson; Geoffrey A Donnan; Dominique A Cadilhac Journal: Med J Aust Date: 2015-07-20 Impact factor: 7.738
Authors: Qian Li; Robert J Glynn; Nancy A Dreyer; Jun Liu; Helen Mogun; Soko Setoguchi Journal: Pharmacoepidemiol Drug Saf Date: 2011-05-14 Impact factor: 2.890
Authors: A A Herman; B J McCarthy; J M Bakewell; R H Ward; B A Mueller; N E Maconochie; A W Read; P Zadka; R Skjaerven Journal: Paediatr Perinat Epidemiol Date: 1997-01 Impact factor: 3.980
Authors: Sean Randall; Adrian Brown; James Boyd; Rainer Schnell; Christian Borgs; Anna Ferrante Journal: BMC Health Serv Res Date: 2018-09-03 Impact factor: 2.655
Authors: Boris P Hejblum; Griffin M Weber; Katherine P Liao; Nathan P Palmer; Susanne Churchill; Nancy A Shadick; Peter Szolovits; Shawn N Murphy; Isaac S Kohane; Tianxi Cai Journal: Sci Data Date: 2019-01-08 Impact factor: 6.444
Authors: Amelia Jewell; Matthew Broadbent; Richard D Hayes; Ruth Gilbert; Robert Stewart; Johnny Downs Journal: BMJ Open Date: 2020-07-07 Impact factor: 2.692