Risk of thrombotic events and other complications in anticoagulant users infected with SARS-CoV-2: an observational cohort study in primary health care in SIDIAP (Catalonia, Spain)

Table 3 Incidence rate ratios from the fitted negative binomial regression

	Non-exposed to OAC			Exposed to OAC
Outcome	sum of events	sum of person-days	Incidence rate (events/1000 person-day)	sum of events	sum of person-days	Incidence rate (events/1000 person-day)	IRR	95% CI^a	P-value	PS matched samples IRR	95% CI^a	P-value
Pneumonia	4577	20,955,070	0.22 (0.21–0.22)	242	696,364	0.35 (0.3–0.39)	0.98	(0.85–1.14)	< .001	1.04	(0.84–1.3)	0.710
Stroke and pulmonary embolism	764	20,955,070	0.04 (0.03–0.04)	188	696,364	0.27 (0.23–0.31)	5.27	(4.22–6.58)	< .001	1.80	(1.06–3.06)	0.030
Hospital admission	9478	20,955,070	0.45 (0.44–0.46)	804	696,364	1.15 (1.08–1.24)	2.12	(1.74–2.58)	< .001	1.16	(1.03–1.29)	0.013
Death	8193	20,955,070	0.39 (0.38–0.4)	1318	696,364	1.89 (1.79–2)	3.39	(2.80–4.11)	< .001	0.60	(0.55–0.65)	< .001

^a95% CI and P-values are derived using robust standard errors as suggested in Long J. S., Ervin L. H. [25]
Adjusted IRR by age, gender, smoking habits, comorbidities and comedications
OAC oral anticoagulants, IRR incidence rate ratio

ISSN: 2731-4553