Premium
Survey Definitions of Gout for Epidemiologic Studies: Comparison With Crystal Identification as the Gold Standard
Author(s) -
Dalbeth Nicola,
Schumacher H. Ralph,
Fransen Jaap,
Neogi Tuhina,
Jansen Tim L.,
Brown Melanie,
Louthrenoo Worawit,
VazquezMellado Janitzia,
Eliseev Maxim,
McCarthy Geraldine,
Stamp Lisa K.,
PerezRuiz Fernando,
Sivera Francisca,
Ea HangKorng,
Gerritsen Martijn,
Scire Carlo A.,
Cavagna Lorenzo,
Lin Chingtsai,
Chou YinYi,
Tausche AnneKathrin,
da Rocha CastelarPinheiro Geraldo,
Janssen Matthijs,
Chen JiunnHorng,
Cimmino Marco A.,
Uhlig Till,
Taylor William J.
Publication year - 2016
Publication title -
arthritis care and research
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.032
H-Index - 163
eISSN - 2151-4658
pISSN - 2151-464X
DOI - 10.1002/acr.22896
Subject(s) - gout , medicine , gold standard (test) , hyperuricemia , rheumatism , tophus , logistic regression , physical therapy , uric acid
Objective To identify the best‐performing survey definition of gout from items commonly available in epidemiologic studies. Methods Survey definitions of gout were identified from 34 epidemiologic studies contributing to the Global Urate Genetics Consortium (GUGC) genome‐wide association study. Data from the Study for Updated Gout Classification Criteria (SUGAR) were randomly divided into development and test data sets. A data‐driven case definition was formed using logistic regression in the development data set. This definition, along with definitions used in GUGC studies and the 2015 American College of Rheumatology (ACR)/European League Against Rheumatism (EULAR) gout classification criteria were applied to the test data set, using monosodium urate crystal identification as the gold standard. Results For all tested GUGC definitions, the simple definition of “self‐report of gout or urate‐lowering therapy use” had the best test performance characteristics (sensitivity 82%, specificity 72%). The simple definition had similar performance to a SUGAR data‐driven case definition with 5 weighted items: self‐report, self‐report of doctor diagnosis, colchicine use, urate‐lowering therapy use, and hyperuricemia (sensitivity 87%, specificity 70%). Both of these definitions performed better than the 1977 American Rheumatism Association survey criteria (sensitivity 82%, specificity 67%). Of all tested definitions, the 2015 ACR/EULAR criteria had the best performance (sensitivity 92%, specificity 89%). Conclusion A simple definition of “self‐report of gout or urate‐lowering therapy use” has the best test performance characteristics of existing definitions that use routinely available data. A more complex combination of features is more sensitive, but still lacks good specificity. If a more accurate case definition is required for a particular study, the 2015 ACR/EULAR gout classification criteria should be considered.