Anaemia in the first week may be associated with long-term mortality among critically ill patients: propensity score-based analyses

Table 2 Cox proportional hazard regressions for estimation of the association between level of week-1 haemoglobin lower than 10 g/dL and one-year mortality in critically ill patients

	Primary population		PSM population		IPTW population		CBPS populaiton
	HR (95%CI)	p value	HR (95%CI)	p value	HR (95%CI)	p value	HR (95%CI)	p value
Model 1	2.950 (2.730–3.187)	< 0.001	1.337 (1.213–1.474)	< 0.001	1.241 (1.076–1.433)	0.003	1.454 (1.304–1.622)	< 0.001
Model 2	2.175 (1.998–2.368)	< 0.001	1.334 (1.209–1.471)	< 0.001	1.253 (1.103–1.424)	0.001	1.356 (1.216–1.513)	< 0.001
Model 3	1.706 (1.564–1.861)	< 0.001	1.308 (1.185–1.443)	< 0.001	1.271 (1.126–1.436)	< 0.001	1.318 (1.178–1.476)	< 0.001
Model 4	1.345 (1.227–1.474)	< 0.001	1.265 (1.145–1.397)	< 0.001	1.276 (1.142–1.427)	< 0.001	1.260 (1.125–1.411)	< 0.001

Model 1. Unadjusted
Model 2. Adjusted for demographic data and comorbidities listed in Table 1.
Model 3. Adjusted for variables in model 2 and critical illness severity/management, including APACHE II score, sepsis, shock and use of mechanical ventilation.
Model 4. Adjusted for variables in model 3 and laboratory data, including white blood cell count, platelet count, albumin, blood urea nitrogen, and creatinine.
Abbreviations: PSM, propensity score matching; IPTW, inverse probability of treatment weighting; CBPS, covariate balancing propensity score; HR, hazard ratio; CI, confidence interval.

ISSN: 1471-227X