Skip to main content

Table 3 Univariable and multivariable logistic regression analysis with multiple imputation

From: Association of anti-β2-glycoprotein I/HLA-DR complex antibody with arterial thrombosis in female patients with systemic rheumatic diseases

 

Univariate

Multivariate

 

Complete cases (n = 409)

Multiple imputation (n = 704)

 

Conventional cardiovascular risk factors alone (model 1)

The addition of anti-β2GPI/HLA-DR antibody status (model 2)

The addition of anti-β2GPI/HLA-DR antibody status (model 2)

Variables

OR (95%CI)

p-value

OR (95%CI)

p-value

OR (95%CI)

p-value

OR (95%CI)

p-value

Age

1.02 (1.00–1.03)

0.050

1.02 (1.00–1.04)

0.043

1.03 (1.00–1.05)

0.027

1.01 (0.99–1.03)

0.059

Disease duration

1.01 (0.99–1.03)

0.341

1.00 (0.98–1.03)

0.790

1.00 (0.98–1.03)

0.850

1.00 (0.98–1.02)

0.610

Body mass index

1.04 (0.97–1.10)

0.270

1.02 (0.95–1.09)

0.651

1.01 (0.94–1.09)

0.730

1.01 (0.94–1.08)

0.750

Pack-year smoking

1.04 (1.02–1.07)

0.002

1.03 (1.00–1.07)

0.063

1.03 (1.00–1.07)

0.073

1.04 (1.01–1.07)

0.007

Current doses of prednisolone

1.07 (1.01–1.13)

0.014

1.11 (1.02–1.20)

0.012

1.11 (1.03–1.20)

0.009

1.07 (1.01–1.13)

0.018

Dyslipidemia

1.13 (0.68–1.89)

0.632

0.73 (0.35–1.49)

0.387

0.78 (0.37–1.61)

0.500

0.77 (0.42–1.40)

0.401

Arterial hypertension

2.54 (1.56–4.16)

 < 0.001

1.82 (0.93–3.58)

0.081

1.81 (0.91–3.61)

0.092

2.37 (1.33–4.23)

0.003

Diabetes

1.62 (0.83–3.17)

0.155

1.04 (0.42–2.56)

0.932

1.26 (0.51–3.12)

0.620

1.15 (0.52–2.52)

0.723

Anti-β2GPI/HLA-DR antibody status (cut-off = 172.359 U/ml)

5.10 (2.94–8.87)

 < 0.001

4.39 (2.14–9.03)

 < 0.001

5.13 (2.85–9.24)

 < 0.001

  1. Abbreviations: CI, confidence interval; OR, odds ratio
  2. The variance inflation factors were < 5 for all of the independent variables. Compared with the non-arterial thrombosis group, the arterial thrombosis group had a significantly positive anti-β2GPI/HLA-DR antibody cut-off (odds ratio: 5.13; 95% CI: 2.85−9.24) in multivariate logistic regression analysis with multiple imputation. Increased age, a high prednisolone dose, high pack-year smoking, and a history of arterial hypertension were significantly associated with arterial thrombosis