Cost–effectiveness of apixaban and warfarin in the prevention of thromboembolic complications among atrial fibrillation patients

SpringerPlus

Table 2 Risk of modeled health events according to treatment

	ARISTOTLE Rate per 100 patient years¹		Second line treatment Rate per 100 patient years
	Apixaban	Warfarin	ASA^c	No antithrombotic treatment
Ischemic stroke by CHADS₂-score
0–1 (34 % of patients)	0.521	0.458
2 (35.8 % of patients)	0.950	0.934
3–6 (30.2 % of patients)	1.534	1.944
Weighted average/average	0.981	1.021	2.280	2.81²
Hemorrhagic stroke	0.254	0.512	0.388	–
Other intracranial bleed	0.076	0.288	1.455	–
Other major bleed: gastrointestinal	0.680	0.795	1.455	–
Other major bleed: not gastrointestinal	1.110	1.476	1.455	^–
Clinically relevant non-major bleed	2.083	2.995	1.811	–
Myocardial infarction	0.530	0.610	0.616	0.856³
Systemic embolism	0.090	0.100	0.600	0.486⁴
Mortality for the trial duration^a	3.0825	3.3404
Other cardiovascular hospitalizations	–	–
Treatment discontinuation^b	13.177	14.405

^aOther cause mortality after the trial period was estimated by fitting a Gompertz survival function to the Finnish life tables
^bFor reasons other than modeled events
^cThe relative risk estimates from the AVERROES trial (Connolly et al. 2011) were applied to the apixaban event rates in the ARISTOTLE trial (Granger et al. 2011)
¹Dorian et al. (2014)
²The ASA event rates were transformed using a relative risk (RR) reduction of 0.19 (Lip and Lim 2007) for ASA versus placebo
³The ASA event rates were transformed using RR = 0.72 (Yerman et al. 2007) for ASA versus placebo
⁴An RR of 0.19 for stroke was assumed to apply for SE as well