Skip to main content
. 2020 Apr 7;17(7):2536. doi: 10.3390/ijerph17072536

Table 3.

Goodness-of-fit assessment.

Studies Mortality Risk Combined Fatal and Non-fatal Risk CVD Markers
CHD CVD Stroke CHD CVD Stroke Carotid Atherosclerosis Disease Hypertension Pulse Blood Pressure QT Prolongation
Log-linear dose-response association model
Deviance a 19.40 22.58 15.98 13.04 7.06 18.54 2.99 20.27 14.02 4.97
Degrees of freedom b 17 22 17 9 4 11 5 21 6 4
p-value c 0.306 0.426 0.526 0.161 0.133 0.070 0.702 0.504 0.029 0.291
R2 0.320 0.258 0.027 0.537 0.798 0.134 0.844 0.230 0.066 0.185
Adjusted R2 0.280 0.225 −0.031 0.486 0.748 0.055 0.813 0.193 −0.089 −0.019
AIC 0.17 −6.77 6.58 −0.56 1.26 −2.22 3.38 −4.36 4.55 5.58
Non-linear dose-response association model (restricted cubic splines)
Deviance a 17.28 22.81 15.39 5.83 3.94 17.48 1.71 12.94 9.16 3.44
Degrees of freedom b 16 21 16 8 3 10 4 20 5 3
p-value c 0.367 0.354 0.496 0.666 0.267 0.064 0.789 0.880 0.103 0.328
R2 0.373 0.620 0.035 0.512 0.564 0.110 0.892 0.199 0.292 0.435
Adjusted R2 0.297 0.584 −0.085 0.390 0.273 −0.068 0.838 0.118 0.008 0.058
AIC 29.95 5.89 23.86 12.34 10.37 16.45 13.75 23.55 12.16 11.43

CVD: cardiovascular disease; CHD: coronary heart disease. a: Measure of the total absolute deviation between reported and predicted log-relative risk taking into account the covariance structure of the residuals. b: Degrees of freedom from the deviance statistic. c: p-value from test for model specification. AIC: Akaike’s information criterion.