Sensitivity analyses of the model parameters. Dashed lines: Scenario 1 (use of AB only); Solid lines: Scenario 2 (use of non-anti-microbial); Vertical line: baseline values; (a) Percentage of CDAD+ patients that stop AB and start non-AB (baseline = 50%, σ = 0.5); (b) Number of colonized patients admitted per day in the ratio λ
C−
off/λ
C−
on = 2/3 (baseline λ
C−
off = 0.33, λ
C−
on = 0.5); (c) Length of stay of colonized patients in weeks (baseline = 4 weeks, = γ
C−
off/γ
C−
on= 1/28); (d) Hand-washing compliance percentage (baseline = 60%, η = 0.6); (e) Percentage of uncolonized and colonized patients on AB (held equal) that become CDAD+ (baseline = 15%, ω
U−
on = ω
C−
on = −log (0.9893)); (f) Time to symptoms resolution (TSR) of colonized CDAD+ patients on AB in days (baseline = 3.7 days, δ
C+
on = −log (0.73)); AB = antibiotic; CDAD = clostridium difficile-associated disease; non-AB = nonantibiotic; λ
C−
off = CDAD−, colonized, off antibiotic: admission rate; γ
C−
on = CDAD−, colonized, on antibiotic: average length of stay; ω
U−
on = CDAD−, uncolonized, on antibiotic: transit to CDAD+; ω
C−
on = CDAD−, colonized, on antibiotic: transit to CDAD+; γ
C−
off = CDAD−, colonized, off antibiotic: average length of stay; λ
C−
on = CDAD−, colonized, on antibiotic: admission rate.