Intensified hand-hygiene campaign including soap-and-water wash may prevent acute infections in office workers, as shown by a recognized-exposure -adjusted analysis of a randomized trial

Table 5 Reported-exposure –dependent efficacy of interventions on weekly prevalence of reported symptoms of gastrointestinal infection (GTI)

Weeks with or without reported exposure Arm and number of weeks	Symptoms reported for
	Same week			Following week
	Number and proportion^a	Predictive margin (CI^b)	Risk ratio^c (CI^b)	Number and proportion^a	Predictive margin (CI^b)	Risk ratio (CI^b)
Exposure reported
Controls n = 767	159/0.207	0.164 (0.129, 0.209)	NR	67/0.087	0.061 (0.048, 0.079)	NR
Soap-and-water n = 937	159/0.170	0.206 (0.166, 0.257)	1.256^d (0.878, 1.801)	51/0.054	0.047 (0.036, 0.059)	0.761^e (0.537, 1.060)
Alcohol rub n = 806	143/0.177	0.186 (0.146, 0.239)	1.130^f (0.781, 1.655)	63/0.078	0.056 (0.043, 0.072)	0.904^g (0.634, 1.275)
Exposure not reported
Controls n = 10877	154/0.014	0.015 (0.012, 0.018)	NR	226/0.021	0.021 (0.017, 0.027)	NR
Soap-and-water n = 14107	123/0.009	0.009 (0.007, 0.011)	0.592^h (0.412, 0.847)	210/0.015	0.016 (0.013, 0.020)	0.761ⁱ (0.537, 1.060)
Alcohol rub n = 11180	131/0.012	0.012 (0.009, 0.015)	0.792^j (0.553, 1.145)	193/0.017	0.018 (0.017, 0.027)	0.904^k (0.634, 1.275)

NR not relevant
^aNumber of weeks with reported RTI symptoms representing indicated proportion out of weeks with reported exposure
^bCI, 95% credible interval of predictive margins
^cRisk ratio is the ratio of predictive margin in the intervention arm to that of the control arm
^defghijkProbabilities that the risk ratio is less than 1.0; ^d, 0.10; ^e, 0.95; ^f, 0.26; ^g, 0.72; ^h, 1.0; ⁱ, 0.95; ^j, 0.90; ^k, 0.72
Note! e = i and g = k; no statistically significant interaction was found in the following week data between arm and exposure. Thus, a single estimate was calculated for the entire arm

ISSN: 1471-2334