Skip to main content
. 2020 Dec 14;15(12):e0243741. doi: 10.1371/journal.pone.0243741

Table 3. Predictors of accurate empirical therapy of the index ESBL-UTI episode.

Predictora Univariate analysis Multivariate analysis
Non accurate N = 80 Accurate N = 205 P value OR 95%CI P value
Ageb 62 [47.5–73.75] 60 [29–72] 0.136 0.994 0.983–1.005 0.296
Gender 0.002 0.037
 • Female 30 (37.5) 119 (58) Ref.
 • Male 50 (62.5) 86 (42) 0.528 0.289–0.963
Comorbidity 0.520
 • No 20 (25) 44 (21.5)
 • Yes 60 (75) 161 (78.5)
Interval between paired isolates (months)b 3 [2–5] 3 [2–5] 0.673
Reasons of admission 0.001 0.018
 • Medical reasons 55 (68.8) 176 (85.9) Ref.
 • Surgical reasons 25 (31.3) 29 (14.1) 0.451 0.234–0.870
Presence of urological/surgical condition 0.084 0.839
 • No 35 (43.8) 113 (55.1) Ref.
 • Yes 45 (56.3) 92 (44.9) 0.939 0.511–1.725
Negative Intervening culture 0.288
 • No 72 (90) 192 (93.7)
 • Yes 8 (10) 13 (6.3)
Type of ESBL producer 0.591
 • E. coli 65 (81.3) 172 (83.9)
 • K. pneumoniae 15 (18.8) 33 (16.1)

Abbreviations: OR, odds ratio; CI, confidence interval.

a All data expressed as n (%) of participants unless otherwise indicated.

b Data described as median [Interquartile range].