The effect of time-varying capacity utilization on 14-day in-hospital mortality: a retrospective longitudinal study in Swiss general hospitals

Table 2 The adjusted effects of cumulative daily exposure to high capacity utilization on 14-day in-hospital mortality without and with IPTW (MSM)

	Without IPTW			With IPTW (MSM)
	Estimate	p-value	Odds Ratio (95% CI)	Estimate	p-value	Causal Odds Ratio (95% CI)
Daily exposure to capacity utilization
≥ 85th percentile per day	0.011	< 0.05	1.01 (1.00 to 1.02)	0.016	< 0.001	1.02 (1.01 to 1.03)
Other adjusted variables
Weekend	0.091	< 0.001	1.10 (1.06 to 1.13)	0.088	< 0.001	1.09 (1.05 to 1.13)
Hospital types
Tertiary care (level 2)	0.007	0.708	1.01 (0.97 to 1.05)	0.009	0.64	1.01 (0.97 to 1.05)
Large basic (level 3)	−0.149	< 0.001	0.86 (0.80 to 0.92)	− 0.141	< 0.001	0.87 (0.81 to 0.93)
Medium basic (level 4)	−0.114	< 0.001	0.89 (0.83 to 0.95)	−0.118	< 0.001	0.89 (0.83 to 0.95)
Small basic (level 5)	0.231	< 0.001	1.26 (1.13 to 1.40)	0.523	< 0.001	1.69 (1.48 to 1.92)
Elixhauser index (Swiss Comorbidity weights)
= 0	1.090	< 0.001	2.97 (2.71 to 3.25)	1.090	< 0.001	2.97 (2.71 to 3.26)
> 0 to < 5	0.847	< 0.001	2.33 (2.10 to 2.60)	0.865	< 0.001	2.38 (2.13 to 2.65)
≥ 5	2.060	< 0.001	7.83 (7.22 to 8.49)	2.080	< 0.001	8.02 (7.39 to 8.71)
Age in 5 years	0.028	< 0.001	1.03 (1.03 to 1.03)	0.028	< 0.001	1.03 (1.03 to 1.03)
Female	−0.267	< 0.001	0.76 (0.74 to 0.79)	−0.270	< 0.001	0.76 (0.74 to 0.79)

Inverse Probability of Treatment/exposure Weight (IPTW) of capacity utilization ≥85th percentile for daily-varying confounders, patient turnover and PCCL (Patient Clinical Complexity Level). The reference categories are (hospital types: university hospitals, Swiss comorbidity weights: < 0). The age groups (five-year) are converted into numeric. Clustering of observations by hospital and patient is accounted for both models

ISSN: 1472-6963