Skip to main content
. 2016 Oct 11;36(28):4514–4528. doi: 10.1002/sim.7144

Table 1.

Results from the Atherosclerosis Risk in Communities study: log‐hazard ratios for estimated current SBP (models BCF, LOCF and CA) or SBP intercept and SBP slope (models ORC, RRC and Joint), and C‐indices and Brier scores for 3‐year and 10‐year survival from 9 years on study (standard errors in brackets).

Model
BCF LOCF CA ORC RRC Joint
Survival model
logHR SBP 0.019 (0.001) 0.019 (0.001) 0.022 (0.001) 0.028 (0.002) 0.027 (0.002) 0.030 (0.002)
SBP slope NA NA NA 0.076 (0.040) 0.139 (0.041) 0.155 (0.048)
3‐year CVD risk prediction
C‐index 0.750 (0.021) 0.742 (0.022) 0.751 (0.022) 0.752 (0.022) 0.750 (0.022) 0.751 (0.022)
Change in C‐index Reference  − 0.009 (0.009) 0.000 (0.006) 0.002 (0.006) 0.000 (0.007) 0.001 (0.007)
Brier score 0.026 (0.002) 0.026 (0.002) 0.026 (0.002) 0.026 (0.002) 0.026 (0.002) 0.026 (0.002)
10‐year CVD risk prediction
C‐index 0.728 (0.012) 0.723 (0.012) 0.732 (0.012) 0.733 (0.012) 0.732 (0.012) 0.733 (0.012)
Change in C‐index Reference  − 0.005 (0.005) 0.004 (0.003) 0.006 (0.003) 0.004 (0.004) 0.005 (0.004)
Brier score 0.100 (0.004) 0.100 (0.004) 0.099 (0.004) 0.099 (0.004) 0.099 (0.004) 0.099 (0.004)

Changes in C‐indices are compared with the BCF model. SBP, systolic blood pressure; RRC, risk‐set regression calibration; BCF, baseline carried forward; LOCF, last observation carried forward; CA, cumulative average; ORC, ordinary regression calibration; CVD, cardiovascular disease.