Publication Information (EuropePMC) | |
Title | Polygenic Scores Help Reduce Racial Disparities in Predictive Accuracy of Automated Type 1 Diabetes Classification Algorithms. |
PubMed ID | 36745605(Europe PMC) |
doi | 10.2337/dc22-1833 |
Publication Date | April 1, 2023 |
Journal | Diabetes Care |
Author(s) | Deutsch AJ, Stalbow L, Majarian TD, Mercader JM, Manning AK, Florez JC, Loos RJF, Udler MS. |
Polygenic Score ID & Name | PGS Publication ID (PGP) | Reported Trait | Mapped Trait(s) (Ontology) | Number of Variants |
Ancestry distribution GWAS Dev Eval |
Scoring File (FTP Link) |
---|---|---|---|---|---|---|
PGS000024 (GRS2) |
PGP000014 | Sharp SA et al. Diabetes Care (2019) |
Type 1 diabetes (T1D) | type 1 diabetes mellitus | 67 | - |
https://ftp.ebi.ac.uk/pub/databases/spot/pgs/scores/PGS000024/ScoringFiles/PGS000024.txt.gz |
PGS000023 (AA_GRS) |
PGP000013 | Onengut-Gumuscu S et al. Diabetes Care (2019) |
Type 1 diabetes (T1D) | type 1 diabetes mellitus | 7 | - |
https://ftp.ebi.ac.uk/pub/databases/spot/pgs/scores/PGS000023/ScoringFiles/PGS000023.txt.gz |
PGS Performance Metric ID (PPM) |
Evaluated Score |
PGS Sample Set ID (PSS) |
Performance Source | Trait |
PGS Effect Sizes (per SD change) |
Classification Metrics | Other Metrics | Covariates Included in the Model |
PGS Performance: Other Relevant Information |
---|---|---|---|---|---|---|---|---|---|
PPM018536 | PGS000023 (AA_GRS) |
PSS011012| Multi-ancestry (including European)| 39,820 individuals |
PGP000477 | Deutsch AJ et al. Diabetes Care (2023) |Ext. |
Reported Trait: Type 1 diabetes | — | AUROC: 0.781 | — | — | — |
PPM018532 | PGS000024 (GRS2) |
PSS011012| Multi-ancestry (including European)| 39,820 individuals |
PGP000477 | Deutsch AJ et al. Diabetes Care (2023) |Ext. |
Reported Trait: Type 1 diabetes | — | AUROC: 0.875 | — | — | — |
PPM018533 | PGS000024 (GRS2) |
PSS011009| Multi-ancestry (including European)| 57,643 individuals |
PGP000477 | Deutsch AJ et al. Diabetes Care (2023) |Ext. |
Reported Trait: Type 1 diabetes | — | AUROC: 0.822 | — | — | — |
PPM018534 | PGS000024 (GRS2) |
PSS011014| European Ancestry| 34,939 individuals |
PGP000477 | Deutsch AJ et al. Diabetes Care (2023) |Ext. |
Reported Trait: Type 1 diabetes | — | AUROC: 0.888 | — | — | — |
PPM018535 | PGS000024 (GRS2) |
PSS011014| European Ancestry| 34,939 individuals |
PGP000477 | Deutsch AJ et al. Diabetes Care (2023) |Ext. |
Reported Trait: Type 1 diabetes | — | AUROC: 0.858 | — | — | — |
PPM018537 | PGS000023 (AA_GRS) |
PSS011009| Multi-ancestry (including European)| 57,643 individuals |
PGP000477 | Deutsch AJ et al. Diabetes Care (2023) |Ext. |
Reported Trait: Type 1 diabetes | — | AUROC: 0.817 | — | — | — |
PPM018538 | PGS000024 (GRS2) |
PSS011011| European Ancestry| 16,663 individuals |
PGP000477 | Deutsch AJ et al. Diabetes Care (2023) |Ext. |
Reported Trait: Type 1 diabetes | — | — | PPV (+PRS): 100.0 % PPV (reference): 86.0 % |
eMERGE type 1 diabetes algorithm | — |
PPM018539 | PGS000023 (AA_GRS) |
PSS011011| European Ancestry| 16,663 individuals |
PGP000477 | Deutsch AJ et al. Diabetes Care (2023) |Ext. |
Reported Trait: Type 1 diabetes | — | — | PPV (+PRS): 97.0 % PPV (reference): 86.0 % |
eMERGE type 1 diabetes algorithm | — |
PPM018540 | PGS000024 (GRS2) |
PSS011010| Multi-ancestry (excluding European)| 40,980 individuals |
PGP000477 | Deutsch AJ et al. Diabetes Care (2023) |Ext. |
Reported Trait: Type 1 diabetes | — | — | PPV (+PRS): 93.0 % PPV (reference): 71.0 % |
eMERGE type 1 diabetes algorithm | — |
PPM018541 | PGS000023 (AA_GRS) |
PSS011010| Multi-ancestry (excluding European)| 40,980 individuals |
PGP000477 | Deutsch AJ et al. Diabetes Care (2023) |Ext. |
Reported Trait: Type 1 diabetes | — | — | PPV (+PRS): 86.0 % PPV (reference): 71.0 % |
eMERGE type 1 diabetes algorithm | — |
PPM018542 | PGS000024 (GRS2) |
PSS011014| European Ancestry| 34,939 individuals |
PGP000477 | Deutsch AJ et al. Diabetes Care (2023) |Ext. |
Reported Trait: Type 1 diabetes | — | — | PPV (+PRS): 97.0 % PPV (reference): 71.0 % |
eMERGE type 1 diabetes algorithm | — |
PPM018543 | PGS000023 (AA_GRS) |
PSS011013| Multi-ancestry (excluding European)| 4,881 individuals |
PGP000477 | Deutsch AJ et al. Diabetes Care (2023) |Ext. |
Reported Trait: Type 1 diabetes | — | — | PPV (+PRS): 83.0 % PPV (reference): 53.0 % |
eMERGE type 1 diabetes algorithm | — |
PGS Sample Set ID (PSS) |
Phenotype Definitions and Methods | Participant Follow-up Time | Sample Numbers | Age of Study Participants | Sample Ancestry | Additional Ancestry Description | Cohort(s) | Additional Sample/Cohort Information |
---|---|---|---|---|---|---|---|---|
PSS011012 | — | — | 1,511 individuals, 36.0 % Male samples |
Mean = 46.9 years Sd = 16.3 years |
Native American, Asian unspecified, Oceanian, Other | — | MGBB | — |
PSS011013 | At each site, a trained medical reviewer performed manual record review for all individuals identified as having type 1 diabetes by the eMERGE algorithm. To confirm a diagnosis of type 1 diabetes, participants had to meet all of the following criteria, modified from (13): Diagnosis confirmed by an endocrinologist or primary care physician Current use of basal-bolus insulin or pump No secondary cause of diabetes listed in the medical record: gestational diabetes, checkpoint inhibitor use, glucocorticoid-induced diabetes, cystic fibrosis diagnosis, hemochromatosis, pancreatogenic diabetes, posttransplantation diabetes, maturity-onset diabetes of the young, or diagnosis of type 1.5 diabetes | — | 2,101 individuals, 37.0 % Male samples |
Mean = 52.1 years Sd = 16.3 years |
African American or Afro-Caribbean | Self-identified race = Black | MGBB | — |
PSS011013 | At each site, a trained medical reviewer performed manual record review for all individuals identified as having type 1 diabetes by the eMERGE algorithm. To confirm a diagnosis of type 1 diabetes, participants had to meet all of the following criteria, modified from (13): Diagnosis confirmed by an endocrinologist or primary care physician Current use of basal-bolus insulin or pump No secondary cause of diabetes listed in the medical record: gestational diabetes, checkpoint inhibitor use, glucocorticoid-induced diabetes, cystic fibrosis diagnosis, hemochromatosis, pancreatogenic diabetes, posttransplantation diabetes, maturity-onset diabetes of the young, or diagnosis of type 1.5 diabetes | — | 1,269 individuals, 34.0 % Male samples |
Mean = 46.4 years Sd = 16.1 years |
Hispanic or Latin American | Self-identified race = Hispanic | MGBB | — |
PSS011013 | At each site, a trained medical reviewer performed manual record review for all individuals identified as having type 1 diabetes by the eMERGE algorithm. To confirm a diagnosis of type 1 diabetes, participants had to meet all of the following criteria, modified from (13): Diagnosis confirmed by an endocrinologist or primary care physician Current use of basal-bolus insulin or pump No secondary cause of diabetes listed in the medical record: gestational diabetes, checkpoint inhibitor use, glucocorticoid-induced diabetes, cystic fibrosis diagnosis, hemochromatosis, pancreatogenic diabetes, posttransplantation diabetes, maturity-onset diabetes of the young, or diagnosis of type 1.5 diabetes | — | 1,511 individuals, 36.0 % Male samples |
Mean = 46.9 years Sd = 16.3 years |
Native American, Asian unspecified, Oceanian, Other | Self-identified race = Other | MGBB | — |
PSS011014 | At each site, a trained medical reviewer performed manual record review for all individuals identified as having type 1 diabetes by the eMERGE algorithm. To confirm a diagnosis of type 1 diabetes, participants had to meet all of the following criteria, modified from (13): Diagnosis confirmed by an endocrinologist or primary care physician Current use of basal-bolus insulin or pump No secondary cause of diabetes listed in the medical record: gestational diabetes, checkpoint inhibitor use, glucocorticoid-induced diabetes, cystic fibrosis diagnosis, hemochromatosis, pancreatogenic diabetes, posttransplantation diabetes, maturity-onset diabetes of the young, or diagnosis of type 1.5 diabetes | — | 34,939 individuals, 47.0 % Male samples |
Mean = 59.1 years Sd = 16.9 years |
European | Self-identified race = White | MGBB | — |
PSS011009 | At each site, a trained medical reviewer performed manual record review for all individuals identified as having type 1 diabetes by the eMERGE algorithm. To confirm a diagnosis of type 1 diabetes, participants had to meet all of the following criteria, modified from (13): Diagnosis confirmed by an endocrinologist or primary care physician Current use of basal-bolus insulin or pump No secondary cause of diabetes listed in the medical record: gestational diabetes, checkpoint inhibitor use, glucocorticoid-induced diabetes, cystic fibrosis diagnosis, hemochromatosis, pancreatogenic diabetes, posttransplantation diabetes, maturity-onset diabetes of the young, or diagnosis of type 1.5 diabetes | — | 16,663 individuals, 48.0 % Male samples |
Mean = 51.9 years Sd = 14.8 years |
European | Self-identified race = White | BioMe | — |
PSS011009 | At each site, a trained medical reviewer performed manual record review for all individuals identified as having type 1 diabetes by the eMERGE algorithm. To confirm a diagnosis of type 1 diabetes, participants had to meet all of the following criteria, modified from (13): Diagnosis confirmed by an endocrinologist or primary care physician Current use of basal-bolus insulin or pump No secondary cause of diabetes listed in the medical record: gestational diabetes, checkpoint inhibitor use, glucocorticoid-induced diabetes, cystic fibrosis diagnosis, hemochromatosis, pancreatogenic diabetes, posttransplantation diabetes, maturity-onset diabetes of the young, or diagnosis of type 1.5 diabetes | — | 11,443 individuals, 39.0 % Male samples |
Mean = 48.4 years Sd = 14.1 years |
African American or Afro-Caribbean | Self-identified race = Black | BioMe | — |
PSS011009 | At each site, a trained medical reviewer performed manual record review for all individuals identified as having type 1 diabetes by the eMERGE algorithm. To confirm a diagnosis of type 1 diabetes, participants had to meet all of the following criteria, modified from (13): Diagnosis confirmed by an endocrinologist or primary care physician Current use of basal-bolus insulin or pump No secondary cause of diabetes listed in the medical record: gestational diabetes, checkpoint inhibitor use, glucocorticoid-induced diabetes, cystic fibrosis diagnosis, hemochromatosis, pancreatogenic diabetes, posttransplantation diabetes, maturity-onset diabetes of the young, or diagnosis of type 1.5 diabetes | — | 19,524 individuals, 37.0 % Male samples |
Mean = 50.3 years Sd = 15.3 years |
Hispanic or Latin American | Self-identified race = Hispanic | BioMe | — |
PSS011009 | At each site, a trained medical reviewer performed manual record review for all individuals identified as having type 1 diabetes by the eMERGE algorithm. To confirm a diagnosis of type 1 diabetes, participants had to meet all of the following criteria, modified from (13): Diagnosis confirmed by an endocrinologist or primary care physician Current use of basal-bolus insulin or pump No secondary cause of diabetes listed in the medical record: gestational diabetes, checkpoint inhibitor use, glucocorticoid-induced diabetes, cystic fibrosis diagnosis, hemochromatosis, pancreatogenic diabetes, posttransplantation diabetes, maturity-onset diabetes of the young, or diagnosis of type 1.5 diabetes | — | 10,013 individuals, 46.0 % Male samples |
Mean = 55.9 years Sd = 13.9 years |
East Asian, South East Asian, Native American, South Asian, Other | Self-identified race = Other | BioMe | — |
PSS011010 | At each site, a trained medical reviewer performed manual record review for all individuals identified as having type 1 diabetes by the eMERGE algorithm. To confirm a diagnosis of type 1 diabetes, participants had to meet all of the following criteria, modified from (13): Diagnosis confirmed by an endocrinologist or primary care physician Current use of basal-bolus insulin or pump No secondary cause of diabetes listed in the medical record: gestational diabetes, checkpoint inhibitor use, glucocorticoid-induced diabetes, cystic fibrosis diagnosis, hemochromatosis, pancreatogenic diabetes, posttransplantation diabetes, maturity-onset diabetes of the young, or diagnosis of type 1.5 diabetes | — | 11,443 individuals, 39.0 % Male samples |
Mean = 48.4 years Sd = 14.1 years |
African American or Afro-Caribbean | Self-identified race = Black | BioMe | — |
PSS011010 | At each site, a trained medical reviewer performed manual record review for all individuals identified as having type 1 diabetes by the eMERGE algorithm. To confirm a diagnosis of type 1 diabetes, participants had to meet all of the following criteria, modified from (13): Diagnosis confirmed by an endocrinologist or primary care physician Current use of basal-bolus insulin or pump No secondary cause of diabetes listed in the medical record: gestational diabetes, checkpoint inhibitor use, glucocorticoid-induced diabetes, cystic fibrosis diagnosis, hemochromatosis, pancreatogenic diabetes, posttransplantation diabetes, maturity-onset diabetes of the young, or diagnosis of type 1.5 diabetes | — | 19,524 individuals, 37.0 % Male samples |
Mean = 50.3 years Sd = 15.3 years |
Hispanic or Latin American | Self-identified race = Hispanic | BioMe | — |
PSS011010 | At each site, a trained medical reviewer performed manual record review for all individuals identified as having type 1 diabetes by the eMERGE algorithm. To confirm a diagnosis of type 1 diabetes, participants had to meet all of the following criteria, modified from (13): Diagnosis confirmed by an endocrinologist or primary care physician Current use of basal-bolus insulin or pump No secondary cause of diabetes listed in the medical record: gestational diabetes, checkpoint inhibitor use, glucocorticoid-induced diabetes, cystic fibrosis diagnosis, hemochromatosis, pancreatogenic diabetes, posttransplantation diabetes, maturity-onset diabetes of the young, or diagnosis of type 1.5 diabetes | — | 10,013 individuals, 46.0 % Male samples |
Mean = 55.9 years Sd = 13.9 years |
East Asian, South East Asian, Native American, South Asian, Other | Self-identified race = Other | BioMe | — |
PSS011011 | At each site, a trained medical reviewer performed manual record review for all individuals identified as having type 1 diabetes by the eMERGE algorithm. To confirm a diagnosis of type 1 diabetes, participants had to meet all of the following criteria, modified from (13): Diagnosis confirmed by an endocrinologist or primary care physician Current use of basal-bolus insulin or pump No secondary cause of diabetes listed in the medical record: gestational diabetes, checkpoint inhibitor use, glucocorticoid-induced diabetes, cystic fibrosis diagnosis, hemochromatosis, pancreatogenic diabetes, posttransplantation diabetes, maturity-onset diabetes of the young, or diagnosis of type 1.5 diabetes | — | 16,663 individuals, 48.0 % Male samples |
Mean = 51.9 years Sd = 14.8 years |
European | Self-identified race = White | BioMe | — |
PSS011012 | — | — | 34,939 individuals, 47.0 % Male samples |
Mean = 59.1 years Sd = 16.9 years |
European | Self-identified race = white | MGBB | — |
PSS011012 | — | — | 2,101 individuals, 37.0 % Male samples |
Mean = 52.1 years Sd = 16.3 years |
African American or Afro-Caribbean (Black) |
— | MGBB | — |
PSS011012 | — | — | 1,269 individuals, 34.0 % Male samples |
Mean = 46.4 years Sd = 16.1 years |
Hispanic or Latin American (Hispanic) |
— | MGBB | — |