TY - JOUR
T1 - Non-invasive risk scores for prediction of type 2 diabetes (EPIC-InterAct)
T2 - A validation of existing models
AU - Kengne, Andre Pascal
AU - Beulens, Joline W.J.
AU - Peelen, Linda M.
AU - Moons, Karel G.M.
AU - van der Schouw, Yvonne T.
AU - Schulze, Matthias B.
AU - Spijkerman, Annemieke M.W.
AU - Griffin, Simon J.
AU - Grobbee, Diederick E.
AU - Palla, Luigi
AU - Tormo, Maria Jose
AU - Arriola, Larraitz
AU - Barengo, Noël C.
AU - Barricarte, Aurelio
AU - Boeing, Heiner
AU - Bonet, Catalina
AU - Clavel-Chapelon, Françoise
AU - Dartois, Laureen
AU - Fagherazzi, Guy
AU - Franks, Paul W.
AU - Huerta, José María
AU - Kaaks, Rudolf
AU - Key, Timothy J.
AU - Khaw, Kay Tee
AU - Li, Kuanrong
AU - Mühlenbruch, Kristin
AU - Nilsson, Peter M.
AU - Overvad, Kim
AU - Overvad, Thure F.
AU - Palli, Domenico
AU - Panico, Salvatore
AU - Quirós, J. Ramón
AU - Rolandsson, Olov
AU - Roswall, Nina
AU - Sacerdote, Carlotta
AU - Sánchez, María José
AU - Slimani, Nadia
AU - Tagliabue, Giovanna
AU - Tjønneland, Anne
AU - Tumino, Rosario
AU - van der A, Daphne L.
AU - Forouhi, Nita G.
AU - Sharp, Stephen J.
AU - Langenberg, Claudia
AU - Riboli, Elio
AU - Wareham, Nicholas J.
PY - 2014/1
Y1 - 2014/1
N2 - Background: The comparative performance of existing models for prediction of type 2 diabetes across populations has not been investigated. We validated existing non-laboratory-based models and assessed variability in predictive performance in European populations. Methods: We selected non-invasive prediction models for incident diabetes developed in populations of European ancestry and validated them using data from the EPIC-InterAct case-cohort sample (27 779 individuals from eight European countries, of whom 12 403 had incident diabetes). We assessed model discrimination and calibration for the first 10 years of follow-up. The models were first adjusted to the country-specific diabetes incidence. We did the main analyses for each country and for subgroups defined by sex, age (<60 years vs ≥60 years), BMI (<25 kg/m2 vs ≥25 kg/m2), and waist circumference (men <102 cm vs ≥102 cm; women <88 cm vs ≥88 cm). Findings: We validated 12 prediction models. Discrimination was acceptable to good: C statistics ranged from 0·76 (95% CI 0·72-0·80) to 0·81 (0·77-0·84) overall, from 0·73 (0·70-0·76) to 0·79 (0·74-0·83) in men, and from 0·78 (0·74-0·82) to 0·81 (0·80-0·82) in women. We noted significant heterogeneity in discrimination (pheterogeneity<0·0001) in all but one model. Calibration was good for most models, and consistent across countries (pheterogeneity>0·05) except for three models. However, two models overestimated risk, DPoRT by 34% (95% CI 29-39%) and Cambridge by 40% (28-52%). Discrimination was always better in individuals younger than 60 years or with a low waist circumference than in those aged at least 60 years or with a large waist circumference. Patterns were inconsistent for BMI. All models overestimated risks for individuals with a BMI of <25 kg/m2. Calibration patterns were inconsistent for age and waist-circumference subgroups. Interpretation: Existing diabetes prediction models can be used to identify individuals at high risk of type 2 diabetes in the general population. However, the performance of each model varies with country, age, sex, and adiposity. Funding: The European Union.
AB - Background: The comparative performance of existing models for prediction of type 2 diabetes across populations has not been investigated. We validated existing non-laboratory-based models and assessed variability in predictive performance in European populations. Methods: We selected non-invasive prediction models for incident diabetes developed in populations of European ancestry and validated them using data from the EPIC-InterAct case-cohort sample (27 779 individuals from eight European countries, of whom 12 403 had incident diabetes). We assessed model discrimination and calibration for the first 10 years of follow-up. The models were first adjusted to the country-specific diabetes incidence. We did the main analyses for each country and for subgroups defined by sex, age (<60 years vs ≥60 years), BMI (<25 kg/m2 vs ≥25 kg/m2), and waist circumference (men <102 cm vs ≥102 cm; women <88 cm vs ≥88 cm). Findings: We validated 12 prediction models. Discrimination was acceptable to good: C statistics ranged from 0·76 (95% CI 0·72-0·80) to 0·81 (0·77-0·84) overall, from 0·73 (0·70-0·76) to 0·79 (0·74-0·83) in men, and from 0·78 (0·74-0·82) to 0·81 (0·80-0·82) in women. We noted significant heterogeneity in discrimination (pheterogeneity<0·0001) in all but one model. Calibration was good for most models, and consistent across countries (pheterogeneity>0·05) except for three models. However, two models overestimated risk, DPoRT by 34% (95% CI 29-39%) and Cambridge by 40% (28-52%). Discrimination was always better in individuals younger than 60 years or with a low waist circumference than in those aged at least 60 years or with a large waist circumference. Patterns were inconsistent for BMI. All models overestimated risks for individuals with a BMI of <25 kg/m2. Calibration patterns were inconsistent for age and waist-circumference subgroups. Interpretation: Existing diabetes prediction models can be used to identify individuals at high risk of type 2 diabetes in the general population. However, the performance of each model varies with country, age, sex, and adiposity. Funding: The European Union.
UR - http://www.scopus.com/inward/record.url?scp=84890159890&partnerID=8YFLogxK
U2 - https://doi.org/10.1016/S2213-8587(13)70103-7
DO - https://doi.org/10.1016/S2213-8587(13)70103-7
M3 - Article
C2 - 24622666
SN - 2213-8587
VL - 2
SP - 19
EP - 29
JO - The Lancet Diabetes and Endocrinology
JF - The Lancet Diabetes and Endocrinology
IS - 1
ER -