Skip to main content

Table 1 C-statistics for coronary artery disease for full population and stratified by sex and age group (younger and older than 55 years of age)A,B

From: Polygenic risk score improves the accuracy of a clinical risk score for coronary artery disease

C-statistic (95% CI)

 

All

Men

Women

Participants aged < 55 y

Participants aged ≥ 55 y

Participants not receiving lipid-lowering treatment at baseline

A. White British ancestry

 Participants, no

272,307

124,155

148,152

102,330

169,977

235,172

 Cases, no

7036

5093

1943

1276

5760

5091

 Polygenic risk score

.64 (.634–.646)

.643 (.636–.651)

.641 (.629–.654)

.69 (.626–.705)

.632 (.625–.639)

.646 (.638–.653)

 Pooled cohort equation

.718 (.713–.723)

.663 (.656–.67)

.706 (.695–.717)

.749 (.736–.761)

.665 (.658–.671)

.73 (.724–.737)

 Polygenic risk score + pooled cohort equation

.753 (.748–.758)

.714 (.708–.721)

.741 (.73–.751)

.793 (.781–.806)

.705 (.699–.712)

.766 (.76–.772)

B. African ancestry

 Participants, no

6753

2901

3852

4528

2225

5896

 Cases, no

88

46

42

42

46

63

 Polygenic risk score

.542 (.485–.6)

.574 (.494–.654)

.6 (.511–.634)

.548 (.46–.637)

.543 (.462–.624)

.534 (.464–.604)

 Pooled cohort equation

.714 (.659–.769)

.674 (.595–.753)

.734 (.653–.815)

.657 (.572–.742)

.721 (.656–.787)

.698 (.628–.768)

 Polygenic risk score + pooled cohort equation

.716 (.665–.768)

.695 (.622–.768)

.732 (.654–.81)

.679 (.597–.761)

.696 (.629–.763)

.707 (.64–.774)

  1. ACox proportional hazard models for CAD using recalibrated polygenic risk score, pooled cohort equations, and both combined models
  2. BC-statistics shown for combined European meta-analysis + Japan Biobank PRS methods. Results are presented for both White British and African ancestry populations