The use of the temporal scan statistic to detect methicillin-resistant Staphylococcus aureus clusters in a community hospital

Table 4 **Multivariable multilevel* negative binomial regression model of variables associated with the rate of MRSA cases**

Variable	Description	IRR	95% CI	P-value
Year	2006	Referent
	2007	1.12	0.73 – 1.72	0.607
	2008	0.89	0.57 – 1.39	0.611
	2009	1.63	1.08 – 2.46	0.021
	2010	2.62	1.76 – 3.89	< 0.001
	2011	2.14	1.13 – 4.06	0.020
Month	January	Referent
	February	1.09	0.70 – 1.71	0.689
	March	1.63	1.04 – 2.57	0.035
	April	1.88	1.22 – 2.91	0.005
	May	1.33	0.83 – 2.15	0.234
	June	0.89	0.53 – 1.51	0.664
	July	1.27	0.79 – 1.06	0.320
	August	1.44	0.92 – 2.26	0.111
	September	1.37	0.87 – 2.17	0.177
	October	1.33	0.84 – 2.10	0.221
	November	1.34	0.85 – 2.12	0.207
	December	1.42	0.91 – 2.23	0.122
Service	Medicine¹	Referent
	Surgery	0.95	0.30 – 2.96	0.924
	Other²	0.42	0.17 – 1.05	0.064

^*Random intercept for ward.
MRSA = Methicillin-resistant Staphylococcus aureus.
IRR = Incidence rate ratio.
CI = Confidence interval.
¹Included the following departments: intensive care (adult and neonatal), oncology, pediatrics, and internal medicine.
²Included the following departments: psychiatry, rehabilitation, hospice, childbirth, and nursery.

ISSN: 1471-2334