Edward Ng1, Saskia E Vanderloo2, Linda Geiss3, Jeffrey A Johnson4. 1. Health Analysis Division, Statistics Canada, Ottawa, Ontario, Canada. Electronic address: edward.ng@statcan.gc.ca. 2. Clinical Epidemiology Program, Ottawa Hospital Research Institute, Ottawa, Ontario, Canada. 3. Division of Diabetes Translation, US Centre for Disease Control, Atlanta, Georgia, United States of America. 4. School of Public Health, University of Alberta, Edmonton, Alberta, Canada.
Abstract
OBJECTIVE: There are 2 major forms of diabetes mellitus: types 1 and 2. A major limitation of most current population-based diabetes surveillance systems is the classification of diabetes types. Our objective was to examine the concordance of self-reported diabetes type with a previously developed classification algorithm, using a nationally representative survey sample. METHODS: Self-reported data were available from 2544 adults with self-reported diabetes, aged ≥20 years and older, who responded to the diabetes component of the 2011 Survey of Living with Chronic Diseases in Canada. We examined the concordance of self-reported diabetes type with an algorithm based on self-reported, but objective, respondent characteristics, such as age of diagnosis and treatment patterns. Concordance was measured using kappa coefficients. Sensitivity, specificity and positive and negative predictive values were calculated using the algorithm as the reference "standard." RESULTS: Approximately 11% of the estimated population did not self-report diabetes type; almost all of these respondents would be classified as having type 2 diabetes by the algorithm. Of those self-reporting diabetes type, we found moderate overall agreement between the algorithm and self-reported type (kappa, 0.52; 95% confidence interval [CI], 0.52 to 0.53). Perfect agreement was noted in the youngest age group (kappa, 1.0; 95% CI, 1.0-1.0) but agreement was poor for the oldest age group (kappa, 0.20; 95% CI, 0.19 to 0.20). CONCLUSIONS: An algorithm based on self-reported, objective characteristics related to diabetes diagnosis and treatment patterns may have the potential to overcome limitations of simple self-report diabetes type for the classification of diabetes type in older adults.
OBJECTIVE: There are 2 major forms of diabetes mellitus: types 1 and 2. A major limitation of most current population-based diabetes surveillance systems is the classification of diabetes types. Our objective was to examine the concordance of self-reported diabetes type with a previously developed classification algorithm, using a nationally representative survey sample. METHODS: Self-reported data were available from 2544 adults with self-reported diabetes, aged ≥20 years and older, who responded to the diabetes component of the 2011 Survey of Living with Chronic Diseases in Canada. We examined the concordance of self-reported diabetes type with an algorithm based on self-reported, but objective, respondent characteristics, such as age of diagnosis and treatment patterns. Concordance was measured using kappa coefficients. Sensitivity, specificity and positive and negative predictive values were calculated using the algorithm as the reference "standard." RESULTS: Approximately 11% of the estimated population did not self-report diabetes type; almost all of these respondents would be classified as having type 2 diabetes by the algorithm. Of those self-reporting diabetes type, we found moderate overall agreement between the algorithm and self-reported type (kappa, 0.52; 95% confidence interval [CI], 0.52 to 0.53). Perfect agreement was noted in the youngest age group (kappa, 1.0; 95% CI, 1.0-1.0) but agreement was poor for the oldest age group (kappa, 0.20; 95% CI, 0.19 to 0.20). CONCLUSIONS: An algorithm based on self-reported, objective characteristics related to diabetes diagnosis and treatment patterns may have the potential to overcome limitations of simple self-report diabetes type for the classification of diabetes type in older adults.
Authors: Jourdyn A Lawrence; Kellee White; Jason L Cummings; James W Hardin; Myriam E Torres Journal: J Racial Ethn Health Disparities Date: 2019-05-07
Authors: Kai McKeever Bullard; Catherine C Cowie; Sarah E Lessem; Sharon H Saydah; Andy Menke; Linda S Geiss; Trevor J Orchard; Deborah B Rolka; Giuseppina Imperatore Journal: MMWR Morb Mortal Wkly Rep Date: 2018-03-30 Impact factor: 17.586
Authors: Sarah S Casagrande; Sarah E Lessem; Trevor J Orchard; Kai McKeever Bullard; Linda S Geiss; Sharon H Saydah; Andy Menke; Giuseppina Imperatore; Keith F Rust; Catherine C Cowie Journal: BMJ Open Diabetes Res Care Date: 2020-12
Authors: Jennifer G Nooney; M Sue Kirkman; Kai McKeever Bullard; Zachary White; Kristi Meadows; Joanne R Campione; Russ Mardon; Gonzalo Rivero; Stephen R Benoit; Emily Pfaff; Deborah Rolka; Sharon Saydah Journal: J Clin Transl Endocrinol Date: 2020-07-03