Table 3.

Cox proportional hazards model with ICU mortality as the endpoint. Current ICU caseload was defined as the highest number of patients admitted to an ICU in 8-h time blocks divided by the total number of beds reported for the respective ICU and year

	HR	95% CI		p
SAPS 3 [per point]	1.05	1.05	1.06	< 0.0001
Patient gender
Female	1.00
Male	0.91	0.89	0.93	< 0.0001
Hospital level
Primary care hospital	1.00
Secondary care hospital	1.03	0.91	1.16	0.675
Specialised care hospital	0.73	0.62	0.87	< 0.0001
Tertiary care hospital	0.91	0.74	1.11	0.351
ICU type
Medical	1.00
Surgical	0.73	0.65	0.83	< 0.0001
Year of discharge
2013	1.00
2014	0.92	0.86	0.99	0.028
2015	0.88	0.79	0.98	0.016
2016	0.98	0.86	1.11	0.696
2017	0.97	0.87	1.08	0.575
2018	1.00	0.89	1.13	0.938
2019	0.93	0.82	1.05	0.256
2020	1.12	0.99	1.26	0.077
2021	1.13	0.99	1.28	0.064
2022	1.03	0.92	1.15	0.575
Time of day
08:00–15:59	1.00
16:00–23:59	0.90	0.86	0.94	< 0.0001
00:00–07:59	0.49	0.47	0.52	< 0.0001
Calendar day
Working day	1.00
Non-working day	0.89	0.86	0.92	< 0.0001
Current ICU caseload (bed occupancy) [%]
≤ 50	1.09	1.02	1.16	0.008
51–75	1.10	1.05	1.15	< 0.0001
76–100	1.00
> 100	1.06	0.99	1.15	0.110
Current ICU workload (TISS-28) [points]
≤ 20	≤ 20	0.88	0.78	0.99
21–30	1.00
31–40	1.03	0.96	1.10	0.455
> 40	0.96	0.85	1.08	0.512

Current ICU workload was defined as the median of TISS-28 scores of all other patients present at an ICU within 8-h time blocks. Current ICU caseload and Current ICU workload were modelled as time-dependent covariables

95% CI 95% confidence interval, HR hazard ratio, SAPS 3 Simplified Acute Physiology Score, TISS Therapeutic Intervention Scoring System