Skip to main content
. 2016 Jul 15;16:331. doi: 10.1186/s12879-016-1650-8

Table 5.

Multivariate conditional logistic regression analysis of risk factors for in-hospital mortality in patients with healthcare-associated methicillin-resistant Staphylococcus aureus bacteremia in the propensity-matched analyses

Unmatched data set (n = 345) Unmatched data set (IPTW, n = 345) Matched data set (n = 254)
Independent variable OR (95 % CI) P-value OR (95 % CI) P-value OR (95 % CI) P-value
Inappropriate empirical antibiotic therapy 1.26 (0.64–2.48) 0.499 1.33 (0.83–2.12) 0.236 NAa 0.988
Nosocomial infection 2.24 (0.87–5.80) 0.096 2.62 (1.28–5.37) 0.009
Trauma 0.60 (0.18–1.97) 0.398 0.66 (0.28–1.52) 0.329 1.95 (0.50–7.56) 0.335
Surgical operation 1.20 (0.54–2.66) 0.656 1.22 (0.69–2.17) 0.493
Prior exposure to third- generation cephalosporins 1.34 (0.68–2.63) 0.399 1.23 (0.76–2.00) 0.407 0.93 (0.41–2.10) 0.853
Prior exposure to fluoroquinolones 0.51 (0.22–1.22) 0.129 0.66 (0.35–1.24) 0.193 1.24 (0.44–3.51) 0.683
Prior exposure to glycopeptides 2.85 (1.20–6.77) 0.018 2.86 (1.55–5.28) 0.001 3.24 (1.08–9.67) 0.035
CR-BSI 0.92 (0.44–1.92) 0.825 0.79 (0.47–1.32) 0.360
Pneumonia 0.37 (0.14–1.00) 0.050 0.28 (0.14–0.59) 0.001 2.15 (0.51–9.04) 0.296
Severe sepsis or septic shock 3.84 (1.85–7.96) <0.001 3.64 (2.17–6.10) <0.001 5.45 (2.14–13.87) <0.001
Retention of foreign body 1.63 (0.40–6.71) 0.500 1.96 (0.62–6.18) 0.250
Charlson’s comorbidity index (per 1-point increment) 0.66 (0.57–0.77) <0.001 0.68 (0.61–0.75) <0.001 1.52 (1.27–1.83) <0.001
Pitt’s bacteremia score (per 1-point increment) 0.77 (0.64–0.93) <0.001 0.76 (0.66–0.87) <0.001 1.23 (0.99–1.54) 0.067
Age (per 1-year increment) 0.96 (0.94–0.98) 0.001 0.97 (0.95–0.98) <0.001 1.02 (1.00–1.05) 0.105
Vancomycin MICs 0.48 (0.20–1.16) 0.103 0.42 (0.23–0.79) 0.007 1.55 (0.52–4.62) 0.433

CR-BSI catheter-related bloodstream infection, IPTW inverse probability of treatment weighted, OR odds ratio, 95 % CI 95 % confidence interval, MIC minimum inhibitory concentration, NA not available

aThese variables were not available because they displayed perfect marginal homogeneity with respect to each category concerning inappropriate initial empirical antibiotic therapy