Skip to main content

Table 2 Variables and coefficients included in stepwise logistic regression model and logistic regression with LASSO model

From: Prognostic nomogram for 30-day mortality of deep vein thrombosis patients in intensive care unit

Variables Stepwise logistic regression Logistic regression with LASSO
Estimate SE P value Estimate SE P value
Intercept − 5.964 1.158 − 2.716 0.379
 General Condition
  Age
   ≥ 75 1.424 0.320 < 0.001 0.887 0.282 0.002
 Systolic blood pressure (mmHg)
  ≥ 110 − 0.748 0.319 0.019 -0.727 0.281 0.010
 Respiratory rate (1/min)
  ≥ 30 1.172 0.729 0.108
 Acute DVT
  Yes 2.333 0.923 0.012
Comorbidity
 Sepsis
  Yes 1.083 0.370 0.003 1.036 0.325 0.001
 Pulmonary circulation disorder
  Yes 0.673 0.336 0.045
 Neurological disease
  Yes − 0.979 0.491 0.046
 Chronic pulmonary disease
  Yes 1.167 0.346 < 0.001
 Lymphoma
  Yes 1.861 0.827 0.024
 Metastatic cancer
  Yes 1.782 0.414 < 0.001 1.328 0.366 < 0.001
 Obesity
  Yes 1.180 0.553 0.033
Laboratory Indicators
 50-RDW (%)
  ≥ 34.5 − 0.646 0.295 0.028
 Lymphocyte (%)
  ≥ 8.5 − 0.749 0.307 0.015
 Anion gap (mEq/L)
  ≥ 16 − 1.007 0.392 0.010
  14–16 − 0.427 0.401 0.287
 Blood urea nitrogen (mg/dL)
  ≥ 18.5 1.292 0.378 < 0.001 0.911 0.304 0.003
 Log10(creatinine (mg/dL))
  ≤ − 0.1 1.205 0.394 0.002
  ≥ 0.2 0.724 0.385 0.060
 Albumin (g/dL)
  ≥ 3 − 0.864 0.287 0.003
 Log10(lactate (mmol/L))
  ≥ log10(2.25) 0.927 0.311 0.003 0.657 0.270 0.015
 PT (s)
  ≥ 14.5 0.809 0.308 0.009
 Log10(NLR)
  ≥ 1.1 0.872 0.276 0.002
  1. LASSO least absolute shrinkage and selection operator, SE standard error, DVT deep vein thrombosis, RDW red blood cell volume distribution width, PT prothrombin time, NLR neutrophil-to-lymphocyte ratio