|
Two Types of Sampling Plans
Presence/absence sampling plans
A presence/absence type of sampling plan is a special case of a two-class sampling plan. In general, a two-class sampling plan is defined in terms of the number of samples (‘n’) to be taken from a given lot, the sample size (‘s’), and a threshold concentration (‘m’) above which a sample is considered to be marginally unacceptable. In a presence/absence type of sampling plan, the threshold concentration 'm' is set to 0, and any samples with detectable pathogen result in rejection of the lot.
This type of sampling plan is appropriate when dealing with a pathogen which presents either a “severe direct” threat to health or one which presents a moderate threat coupled with potential for extensive spread (ICMSF, 1986; ICMSF, 2002).
Concentration-based sampling plans
For pathogens presenting a lower threat to health, or for indicator organisms, a concentration-based sampling plan will reduce the likelihood of rejected lots without appreciably increasing health risk. These plans include two-class plans with a non-zero threshold, and three-class plans.
In a concentration-based sampling plan the threshold concentration of concern is denoted by ‘m’ (“little-m”), and rejection of the lot follows only if the number of samples having a concentration higher than this exceeds a pre-set limit for acceptable positives, called ‘c’.
If desired, another parameter ‘M’ (“big-M”) can be defined, creating three-classes. 'M' is the concentration in the sample above which the lot is automatically rejected. Defining M can increase stringency of the sampling plan, as can be seen in Table 1.
Table 1 . Comparison of Presence/absence and Concentration-based Sampling Plans
|
Sampling Plan |
Number of samples (of 10) having concentration: |
Fate of Lot |
|
0 cfu/g |
1- 20 cfu/g |
21 – 100 cfu/g |
≥100 cfu/g |
|
Presence/absence (m = 0 cfu/g) |
10 |
|
|
|
Accepted |
|
9 |
1 |
|
|
Rejected |
|
Concentration-based m = 20 cfu/g; c = 5 samples; M not defined |
|
5 |
5 |
|
Accepted |
|
|
4 |
6 |
|
Rejected |
|
|
5 |
1 |
4 |
Accepted |
|
Concentration-based m = 20 cfu/g; c = 5 samples; M = 100 cfu/g |
|
5 |
5 |
|
Accepted |
|
|
4 |
6 |
|
Rejected |
|
|
5 |
1 |
4 |
Rejected |
|
9 |
|
|
1 |
Rejected |
With a presence/absence sampling plan, any detection of hazard results in rejection of the lot from which the samples are drawn. In a concentration-based plan with no ‘M’ defined (i.e. a two-class concentration-based plan), the only requirement for rejection is that the number of samples with concentration higher than ‘m’ is greater than ‘c’. In contrast, for a three-class concentration-based plan (i.e. where ‘M’ has been defined), lot rejection results when more than ‘c’ samples fall between ‘m’ and ‘M’, and/or where any sample exceeds ‘M’.
Reference:
ICMSF. 1986. Microorganisms in Foods 2. Sampling for microbial analysis: principles and specific applications. University of Toronto Press, Toronto.
ICMSF. 2002. Microbiologcal testing in food safety management. Kluwer Academic/Plenum Publishers,
|