Table 3.

Single variable analysis of risk factors for SARS Co-V infection among hospital staff cases and controls, Hanoi, 2003.

				SARS cases		Controls
Activities During Index Patient's Hospitalization	OR*	95% CI	p-value†	n = 22	%	n = 45	%
Touched index patient	2.8	0.9–8.5	0.085	9	(41)	9	(20)
Talked to or touched index patient without mask‡ (ever)	1.9	0.6–5.9	0.363	7	(32)	9	(20)
Came within 1 meter of index patient	9.3	2.8–30.9	<0.001	17	(77)	12	(27)
Came within 1 meter of index patient, without mask‡ (ever)	5.4	1.8–16.3	0.003	14	(64)	11	(24)
Spoke with index patient	3.5	1.2–10.4	0.028	11	(50)	10	(22)
Entered patient room	20.0	4.1–97.1	<0.001	20	(91)	15	(33)
Spoke with index patient in his room	3.7	1.1–12.6	0.052	8	(36)	6	(13)
Saw (viewed) index patient	14.0	3.6–55.3	<0.001	19	(86)	14	(31)
Visited patient room when patient was not there	3.7	1.3–10.9	0.027	12	(55)	11	(24)
Touched visibly contaminated surface	7.8	2.3–25.9	0.001	12	(55)	6	(13)
Entered general ward	8.0	1.7–38.4	0.005	20	(91)	25	(56)
Other characteristics§
Upper respiratory infection w/in prior 6 months	0.2	0.04–0.9	0.039	2	(9)	15	(33)
'Other' non-clinical job¶	0.2	0.03–0.7	0.011	2	(9)	18	(40)
Direct patient care activities	2.0	0.7–5.6	0.298	13	(59)	19	(42)
Sanitation/kitchen job	2.2	0.7–7.0	0.223	7	(32)	8	(18)

* Mantel-Haenszel odds ratio; odds ratios refer to affirmative responses.

† Fischer's exact (two-sided).

‡ Respirator or surgical mask; N95 respirator masks were not widely available at Hospital A until March 12^th.

§ A proportion of enrollees (n = 19 cases, n = 8 controls) were asked about a history of heart disease, lung disease, diabetes, and smoking. None of these factors had a statistically significant association with SARS cases status.

¶ Other non-clinical jobs include administration, reception, security, maintenance, operations (see Table 1).