In the multivariable model, patients with good QOL were at significantly lower risk for prolonger LOS (OR = 0.96, 95% CI 0.93-0.99; p = 0.007).
After correcting for other covariates in a multivariable model, patients with anaemia at presentation were very likely to remain anaemic after two weeks of therapy (aOR=60.08, 95% CI: 13.36-270.20, p < 0.0001).
In the final multivariable model, patients of whom their first electrocardiogram was performed at a general practitioner (+23.9 min; 95 % CI 9.9 40.8) or hospital (+9.5 min; 95 % CI 2.5 17.3) had an additional treatment delay compared to patients of whom their first electrocardiogram was made in the ambulance (Table 5).
In multivariable models, patients who did not receive CMVIG as part of their prophylaxis were 3 times more likely to develop CMV infection (hazard ratio, 3.4; 95% confidence interval, 1.2-9.5) independent of CMV serostatus.
Likewise, in the time-dependent Cox multivariable models, patients were included in the initial proper risk set, but allowed to move between risk sets (higher risk and lower risk) over time.
In multivariable models, patients selected for treatment were more likely than untreated patients to reside in Southern Europe [incidence rate ratio (IRR) 1.38 (95% CI 1.06 1.82; P = 0.019) compared with Western Europe] and belong to the MSM HIV transmission group [IRR 1.36 (95% CI 1.00 1.83; P = 0.046) compared with IDU].
In a multivariable model of patients with ER-negative tumours who had received chemotherapy, patients with the AGCC/AGCC diplotype had an increased risk of early events compared to patients with the GGGC/GGGC diplotype (adjusted HR = 17.28; 95% CI 1.51 198.51; P = 0.022).
In a multivariable model of patients with ER-negative tumours who had received radiotherapy, the association between the AGCC/AGCC diplotype and the risk of early events was borderline significant (adjusted HR = 5.71; 95% CI 0.98 33.33; P = 0.053).
In separate multivariable models for patients <15 years versus patients ≥ 15 years of age, mortality was significantly higher among patients < 15 years with penicillin-resistant isolates (adjusted HR 1.22, 95% CI 1.04 to 1.43), while the association did not reach statistical significance among older patients (adjusted HR 1.22, 95% CI 0.96 to 1.56).
In a multivariable model of patient characteristics associated with referral to cardiologic evaluation, a positive/inconclusive exercise test and exertional chest pain were associated with referral, adjusted for age and gender (Table 3).
In unadjusted and multivariable models, elderly patients had worse OS (HR, 1.20; 95% CI, 1.09 to 1.31 and HR, 1.17; 95% CI, 1.07 to 1.29, respectively).
