Impact of 5-HT 3 receptor antagonists on chemotherapy-induced nausea and vomiting: a retrospective cohort study

Table 5 Results of the Poisson regression models on days of hospital/ED associated CINV events

Parameters	Breast cancer (N = 3606)		Carboplatin-treated lung cancer (N = 4497)		Cisplatin-treated lung cancer (N = 1154)
	Regression coefficient (95% CI)	P value	Regression coefficient (95% CI)	P value	Regression coefficient (95% CI)	P value
Age	-0.0032 (-0.0219; 0.0155)	0.7354	-0.0008 (-0.0120; 0.0105)	0.8961	-0.0149 (-0.0325; 0.0027)	0.0968
CCI	0.2529 (0.0572; 0.4486)	0.0113	0.0448 (0.0087; 0.0810)	0.0151	0.0747 (0.0182; 0.1312)	0.0096
Gender (male as reference)
Female	NA		-0.1815 (-0.4202; 0.0573)	0.1363	-0.0132 (-0.3975; 0.3711)	0.9463
Days with cyclophosphamide (in breast cancer) or carboplatin (in lung cancer) treatment	-0.3800 (-0.5600; -0.2000)	<.0001	-0.0483 (-0.0848; -0.0118)	0.0095	0.0113 (-0.0263; 0.0488)	0.5561
Antiemetic regimen (older 5-HT₃ RAs as reference)
Palonosetron	-0.6431 (-1.0391; -0.2471)*	0.0015	-0.3436 (-0.5871; -0.1001)*	0.0057	-0.0373 (-0.4456; 0.3711)	0.8580

* In Poisson regression, the exponential of coefficient indicates the ratio of the comparison groups in outcome. The exponential of the coefficient of antiemetic regimen (-0.6431) was 0.526 in the BC cohort, indicating the average CINV-hospitalization/ED days of palonosetron group was about 52.6% of that of the comparison group (older 5-HT₃ RAs). The exponential of the coefficient of antiemetic regimen (-0.3436) was 0.709 in the LC-carboplatin cohort, indicating the average CINV-hospitalization/ED days of palonosetron group was about 70.9% of that of the comparison group (older 5-HT₃ RAs).