Table 4.

Binary logistic regression showing association between the timing of respiratory virus testing (every 30 min increase) and secondary outcomes

ED	N	>4 hour ED LOS		Patient with a pending RMDT result
		Unadjusted	Adjusted*	Unadjusted	Adjusted†
		OR (95% CI)	OR (95% CI)	OR (95% CI)	OR (95% CI)
A	723	1.58 (1.37 to 1.82)	1.51 (1.28 to 1.79)	1.04 (1.01 to 1.07)	1.06 (1.03 to 1.10)
B	193	1.74 (1.41 to 2.14)	1.70 (1.34 to 2.17)	1.06 (1.01 to 1.12)	1.16 (1.07 to 1.25)
C	301	1.51 (1.29 to 1.76)	1.48 (1.25 to 1.75)	0.99 (0.96 to 1.02)NS	1.02 (0.99 to 1.06)NS
D	530	1.69 (1.48 to 1.93)	1.64 (1.41 to 1.90)	0.99 (0.97 to 1.01)NS	1.02 (1.00 to 1.05)NS
E	239	1.40 (1.21 to 1.61)	1.39 (1.19 to 1.63)	1.00 (0.96 to 1.04)NS	1.02 (0.97 to 1.07)NS
F	182	1.63 (1.28 to 2.07)	1.90 (1.24 to 2.91)	1.01 (0.98 to 1.05)NS	1.05 (1.00 to 1.09)
Overall	2168	1.54 (1.45 to 1.64)	1.51 (1.41 to 1.63)	1.02 (1.01 to 1.03)	1.04 (1.02 to 1.05)

All analyses, except those marked ‘NS’, were significant with a p value of <0.05. The coefficient indicates the likelihood of a given outcome for every 30 min increase in the timing of the RMDT.

*Adjusted for age, triage category, mode of arrival, study ED, patient disposition, test order episode and test result.

†Adjusted for gender, age, triage category, mode of arrival, study ED, patient disposition, test order episode.

ED, emergency department; LOS, length of stay; NS, not significant, RMDT, rapid molecular diagnostic test.