Table 3 Predictors and coefficients of the risk predictive model of retinal arteriosclerosis for females over 45.

From: Development and temporal validation of five year retinal arteriosclerosis risk prediction in high risk adults from an East China cohort

Characteristic

Coefficient (SE)

p value

Age (years)—RCS(45.0, 51.3, 96.8)

  

Spline base variable 1

0.253 (0.019)

< 0.001

Spline base variable 2

−0.174 (0.020)

< 0.001

Body mass index (kg/m2)

0.130 (0.019)

< 0.001

Waist circumference (cm)

0.016 (0.007)

0.030

Hip circumference (cm)

−0.027 (0.009)

0.002

Systolic pressure (mmHg)

0.011 (0.003)

< 0.001

Diastolic pressure (mmHg)

0.012 (0.004)

0.003

Hypertension

 

< 0.001

no

Ref.

 

yes

0.460 (0.071)

 

Diabetes

 

<0.001

no

Ref.

 

yes

0.477 (0.095)

 

HDL (mmol/L)—RCS(0.48, 1.49, 3.64)

  

Spline base variable 1

−0.542 (0.163)

0.001

Spline base variable 2

0.569 (0.220)

0.010

log2 Scr (µmol/L)—RCS(4.88, 5.86, 7.44)

  

Spline base variable 1

−0.849 (0.257)

0.001

Spline base variable 2

0.847 (0.304)

0.005

  1. Abbreviations: RCS: restricted cubic spline; SE: standard error.
  2. Note: RCS (k1, …, kj) indicates restricted cubic spline with knots at positions k1 to kj, corresponding to the following base variables for risk predictive factor x: spline base variable 1 = x; spline base variable i + 1 = (max[(x – ki)3, 0] – (kj – ki) × max[(x – kj − 1)3, 0]/(kj – kj−1) + (kj – 1 – ki) × max[(x – kj)3, 0]/(kj –kj − 1))/(kj – k1)2; i = 1, …, j – 2.
  3. The LP structured from the derivation set is as:.
  4. LP = hypertension + diabetes + 0.253*Spline base age1 − 0.174*Spline base age2 + 0.130*BMI + 0.016*WC − 0.028*HC + 0.011*SP + 0.012*DP − 0.542*Spline base HDL1 + 0.570*Spline base HDL2 − 0.849*Spline base log2(Scr)1 + 0.847*Spline base log2(Scr)2, in which: hypertension: no = 0, yes = 0.460; diabetes: no = 0, yes = 0.477.
Back to article page