The influence of volume and intensive care unit organization on hospital mortality in patients admitted with severe sepsis: a retrospective multicentre cohort study

Table 3 Organizational characteristics of the ICUs and their association with risk-adjusted hospital mortality

ICU characteristic	Descriptive^a	Odds ratio (95% confidence interval)	P value
Number of admissions with severe sepsis per year	72.8 ± 44.0, 65	0.997 (0.995–1.000)	0.027
Total number of admissions per year	522.2 ± 255.0 (488)	1.000 (1.000–1.001)	0.759
Number of ICU beds	15.0 ± 9.8 (12)	0.997 (0.984–1.010)	0.676
Number of hospital beds	561.7 ± 357.6 (510)	1.000 (1.000–1.001)	0.586
Intensivist responsible for ICU treatment	96.4 (27)	^b	^b
Intensivist available on weekdays 7–18 hours	96.4 (27)	^b	^b
Intensivist available in evening and weekend	67.9 (19)	1.237 (0.878–1.741)	0.224
Number of intensivists per ICU bed	0.30 ± 0.12 (0.33)	1.164 (1.010–1.341)^c	0.036
Number of nurses per ICU bed^d	3.6 ± 1.01 (3.8)	0.956 (0.861–1.063)	0.406
Staffing
General physician^e	60.7 (17)	0.981 (0.741–1.298)	0.891
Residents	96.4 (27)	^b	^b
Fellows in training for intensivist	21.4 (6)	1.014 (0.749–1.374)	0.927
MCU as a step-down facility	42.9 (12)	1.261 (0.990–1.608)	0.061
24-hour recovery in hospital	28.6 (8)	1.118 (0.878–1.425)	0.365

^aValues are expressed as mean ± standard deviation (median) for continuous variables and percentage (n) for dichotomous variables. ^bVariable not taken into account in regression analysis because of lack of variation. ^cOdds ratio per 0.1 increase in intensivist-to-bed ratio. ^dvalues based on 24 ICUs. ^eGeneral physician: physician working temporarily at the ICU, not in training for specialist. ICU, intensive care unit; MCU, medium care unit.

ISSN: 1364-8535