Table 2. Prediction rules for skin/environmental shedding among 44 patients receiving versus not receiving CDI therapy at the time of the cultures of whom 14 (31.8%) had positive skin and/or environmental cultures for C. difficile.

Prediction rule	Sensitivity to identify patients with positive skin and/or environmental shedding cultures	Specificity to identify patients with negative skin and environmental shedding cultures	Proportion of all patients correctly predicted as shedders or non-shedders	OR (95% CI)
Patients receiving current CDI therapy (N = 15)	N = 4	N = 11	N = 15
CDI therapy with metronidazole (vs. vancomycin taper)	4/4 on metronidazole (100.0%)	11/11 on vancomycin taper (100.0%)	15/15 (100.0%)	Perfect prediction
Patients not receiving current CDI therapy (N = 29)	N = 10	N = 19	N = 29
Incontinence present (vs. absent)	4/10(40.0%)	18/19(94.7%)	22/29(75.9%)	14.6(1.1, 129.4)
Incontinence OR decreased mobility (Braden score <4) * present (vs. both absent)	9/10(90.0%)	15/19(79.0%)	24/29(82.8%)	33.8(3.2, 351.0)

^*Braden mobility score (1 = Completely immobile; 2 = Very limited; 3 = Slightly limited; 4 = No limitation).