Skip to main content

Table 4 Logistic regression models to evaluate the risk factors for longitudinal cognitive decline as indicated by IQCODE score ≥ 3.5 in COVID-19 patients

From: Post-infection cognitive impairments in a cohort of elderly patients with COVID-19

Variables Univariable ORs (95 %CI) P value Multivariable ORs (95 %CI) P value
Age, year 1.023(1.009–1.037) 0.001   
Sex, male 0.934(0.756–1.157) 0.530   
Education, year 0.959(0.936–0.982) 0.001 0.968(0.944–0.993) 0.011
BMI, kg/m2 1.007(0.947–1.072) 0.821   
Severe, vs. non-severe 3.616(2.719–4.810) < 0.001 2.833(2.065–3.888) < 0.001
ICU admission, vs. no 5.100(3.122–8.331) < 0.001   
High flow oxygen therapy, vs. no 0.828(0.635–1.079) 0.163 0.421(0.302–0.586) < 0.001
Delirium, vs. no 5.578(3.597–8.650) < 0.001 5.480(3.292–9.122) < 0.001
Coexisting disorders, vs. no
 Hypertension 1.825(1.470–2.266) < 0.001 1.661(1.320–2.091) < 0.001
 Diabetes 1.344(1.037–1.743) 0.026   
 Hyperlipidaemia history 1.180(0.835–1.666) 0.348   
 Stroke history 1.590(1.027–2.463) 0.038   
 Coronary heart diseases 1.800(1.341–2.415) < 0.001   
 COPD 2.368(1.686–3.326) < 0.001 2.005(1.398–2.877) < 0.001
  1. Dependent variable: IQCODE ≥ 3.5
  2. Independent variables: age, sex, education, severity, ICU admission, high flow oxygen therapy, delirium, hypertension, diabetes, stroke, coronary heart disease, COPD
  3. Abbreviations: BMI body mass index, ICU intensive care unit, COPD Chronic Obstructive Pulmonary Disease