Table 4.

Relationship between AIP and incident prediabetes in different models

Variable	Non-adjusted model (HR.,95% CI, P)	Minimally-adjusted model (HR,95% CI, P)	Fully-adjusted model (HR,95% CI, P)	GAM (HR,95% CI, P)
AIP	2.83 (2.67, 3.00) < 0.0001	1.37 (1.28, 1.47) < 0.0001	1.41 (1.31, 1.52) < 0.0001	1.34 (1.24, 1.44) < 0.0001
AIP (quartile)
Q1	ref	ref	ref	1.0
Q2	1.47 (1.39, 1.56) < 0.0001	1.19 (1.12, 1.26) < 0.0001	1.18 (1.11, 1.25) < 0.0001	1.15 (1.08, 1.23) < 0.0001
Q3	1.94 (1.83, 2.05) < 0.0001	1.29 (1.22, 1.37) < 0.0001	1.28 (1.20, 1.36) < 0.0001	1.23 (1.16, 1.31) < 0.0001
Q4	2.40 (2.27, 2.53) < 0.0001	1.33 (1.25, 1.42) < 0.0001	1.34 (1.26, 1.43) < 0.0001	1.28 (1.20, 1.37) < 0.0001
P for trend	< 0.0001	< 0.0001	< 0.0001	< 0.0001

Non-adjusted model: we did not adjust for other covariates

Minimally-adjusted model: we adjusted for gender, age, SBP, DBP, family history of diabetes, drinking status, smoking status, and BMI

Fully-adjusted model: we adjusted for gender, age, SBP, DBP, family history of diabetes, drinking status, smoking status, BMI, TC, LDL-C, AST, ALT, Scr, BUN and FPG

GAM: All covariates listed in Table 1 were adjusted. However, continuous covariates were adjusted as nonlinearity

HR, hazard ratios; CI, confidence interval; Ref, reference; GAM, generalized additive mode; AIP, atherogenic index of plasma