Table 4. Sensitivity analysis of combined effects of thrombocytopenia and ARDS^†.

cutoff point for platelet count*	Platelet count <cutoff & Non-ARDS		Platelet count ≥cutoff & ARDS		Platelet count <cutoff & ARDS
	AHR†	p	AHR†	p	AHR†	p
50	1.39(0.81,2.36)	0.2286	1.79(1.47,2.18)	6.50E-09	3.62(2.45,5.34)	8.80E-11
60	1.36(0.84,2.19)	0.2107	1.74(1.42,2.13)	6.30E-08	3.91(2.78,5.51)	5.40E-15
70	1.37(0.90,2.07)	0.1407	1.76(1.43,2.15)	5.60E-08	3.43(2.47,4.76)	1.80E-13
80	1.51(1.04,2.19)	0.0308	1.75(1.42,2.16)	1.50E-07	3.40(2.51,4.60)	2.10E-15
90	1.44(1.01,2.05)	0.0459	1.68(1.36,2.08)	1.90E-06	3.66(2.74,4.87)	<1E-16
100	1.42(1.02,1.97)	0.0356	1.68(1.35,2.09)	3.00E-06	3.47(2.62,4.59)	<1E-16
110	1.35(0.99,1.85)	0.0587	1.68(1.34,2.09)	5.20E-06	3.19(2.42,4.19)	1.10E-16
120	1.39(1.03,1.88)	0.0316	1.69(1.35,2.12)	5.00E-06	3.12(2.38,4.09)	2.20E-16
130	1.33(0.99,1.77)	0.0543	1.70(1.35,2.15)	7.20E-06	2.84(2.18,3.69)	1.20E-14
140	1.25(0.94,1.65)	0.1213	1.73(1.36,2.19)	5.80E-06	2.58(1.99,3.36)	1.60E-12
150	1.27(0.97,1.67)	0.0856	1.73(1.36,2.21)	1.00E-05	2.58(1.99,3.34)	8.80E-13

ARDS = acute respiratory distress syndrome; AHR = adjusted hazard ratio.

*Each line represents a multivariate cox regression with adjustment of age, gender, APACHII score, and sepsis; Platelet count (×10³/µL) was dichotomized by cutoff point;

^†AHR was estimated by comparison with non-thrombocytopenia/non-ARDS group.