TABLE 3.

Simulation results for varying prevalence and contamination level on seeds with and without disinfection, where the detection limit for both pre- and post-production sampling was 2 log₁₀CFU/g

Contamination level, disinfection status	Prevalence (% of positive 25-g samples)	% of samples in which contamination was detected		% of batches contaminated	% of batches reaching the consumer
		Preproductiona	Postproductionb
High (100 to 5,000 CFU/sample), without disinfection	0.0001	0	13.7	13.7	0.0
	0.001	0	80.9	80.9	0.0
	0.01	0.4	100.0	100.0	0.0
	0.1	39.6	100.0	100.0	0.0
	1	100.0	100.0	100.0	0.0
Low (1 to 5 CFU/sample), without disinfection	0.0001	0	12.1	13.7	1.6
	0.001	0	66.2	74.6	8.4
	0.01	0.5	98.2	100.0	1.8
	0.1	35.9	99.9	100.0	0.1
	1	100.0	100.0	100.0	0.0
High (100 to 5,000 CFU/sample), with disinfection (triangular log10 CFU reduction [1, 2.5, 6.5])	0.0001	0	6.4	8.0	1.6
	0.001	0	45.4	49.7	4.3
	0.01	0.2	75.3	79.3	4.0
	0.1	24.1	83.7	85.2	1.5
	1	82.3	87.1	87.7	0.6
Low (1 to 5 CFU/sample), with disinfection (triangular log10 CFU reduction [1, 2.5, 6.5])	0.0001	0	0.1	0.1	0.0
	0.001	0	1.3	1.7	0.4
	0.01	0	4.2	5.0	0.8
	0.1	0.6	8.9	9.8	0.9
	1	6.3	13.7	14.7	1.0

^aIrrigation water from composite samples (50 subsamples of 25 g each) tested after 3 days of sprouting but before the entire batch of 36.3 kg of seeds was placed into production.

^bIrrigation water from 36.3 kg of the batch tested after 3 days of sprouting.