PODLOD-interlab Ver1a

PODLOD interlaboratory calculation program, version 1, dated 2019-09-18
Download: www.wiwiss.fu-berlin.de/fachbereich/vwl/iso/ehemalige/wilrich/index.html
PODLOD interlab program, version 1, dated 2019-09-18, for estimating the POD function and the LOD of
For more information see the sheets 'Info' and 'Example'. For details see: JARVIS, WILRICH & WILRICH: Estima
from an interlaboratory experiment, Journal of AOAC International, 2019, 102(5) 1617-1623. This Excel program is distr
www.wiwiss.fu-berlin.de/fachbereich/vwl/iso/ehemalige/wilrich/index.html. There you can also subscribe to a newsletter in
General information on the experiment

Name of the experiment
Microorganism Matrix Date of experiment
Sample size A0 in g or ml No. of laboratories No. of inoculation levels
Note: To obtain reliable results, 5 inoculation levels should be used. These should comprise 3
informative levels as well as a blank and a positive control. For details, see the sheet 'Info'.
Program by Wilrich & Wilrich 1/19 Printing date: 02/05/2020

POD function and the LOD of a binary microbiological measurement method from an interlaboratory experiment
VIS, WILRICH & WILRICH: Estimation of the POD function and the LOD of a binary microbiological measurement method
23. This Excel program is distributed in the hope that it will be useful, but without any warranty. It can be downloaded at:
also subscribe to a newsletter informing about updates of the program. For assistance contact: wilrich@wiwiss.fu-berlin.de.
How to use this program (Macros have to be enabled):

1. For information on the design of the experiment and evaluation of the results, see the sheet 'Info'.
2. Enter the sample size, the no. of laboratories (up to 30) and the max. no. of inoculation levels (up to 30) per laboratory i
► The tables for your input data will be generated automatically below.
3. Enter your data into the yellow cells of the input tables generated according to step 2.
4. Press 'Ctrl+b' to start the calculation or use the button 'Calculate Results' (see top right).
► The results will be shown in two blue tables below. A figure for the POD curves will be shown in the sheet next to this
5. You can change the sample size or the no. of laboratories or inoculation levels in row 11 at any time (while the other inp
6. You can also change the data in the input tables at any time.
► Changes of data in the yellow tables cause the results tables to be deleted but you can re-calculate as per step 4 at a
7. You can print the tables (with a dynamically adjusted print area) and the figure using the respective 'Print' buttons (see t

laboratory experiment Calculate Due to compatibility problems, the buttons are disabled in
Print Figure
Results Excel 2000 and older. 'Print Figure' is disabled also in Excel
measurement method 2003 because no figure is generated (see also the sheet
can be downloaded at: Print Tables Print Tables 'Info'). In any case, it is possible to calculate by pressing
b/w color 'Ctrl+b' and to print normally via the File-menu.
rich@wiwiss.fu-berlin.de.
the sheet 'Info'.

ation levels (up to 30) per laboratory into the yellow table.
p 2.
right).
will be shown in the sheet next to this sheet.
ow 11 at any time (while the other input data can be preserved).
you can re-calculate as per step 4 at any time.

ng the respective 'Print' buttons (see top right).

he buttons are disabled in

ure' is disabled also in Excel
rated (see also the sheet
e to calculate by pressing
the File-menu.

For more information see the sheets 'Info' and 'Example'. For details see: JARVIS, WILRICH & WILRICH: Estima
Information on the design of the experiment:

In order to obtain reliable results, 5 inoculation levels should be used. These should comprise 3 informative levels which p
correspond to a low, a target and a high inoculation level. In addition, a blank and a positive control should be used but th
1 tube each. An example of such a design is shown below.
Input data
Laboratory 1 Designation: Example design
Inoculation level in cfu/g or cfu/ml No. of inoculated tubes No. of positive tubes
d n y
d = 0 = blank control 1
d = low level 8
d = target level 8
d = high level 8
d = positive control 1
Information on the evaluation of the results:

It is recommended that a coordinating laboratory evaluates the results and checks them for plausibility. If results seem to
criteria below or for other reasons, e.g. reversed results, the coordinating laboratory should request clarification or repetit
respective laboratory. If the results cannot be improved, the coordinating laboratory should consider whether the results o
should be excluded from the input data in order not to compromise the mean results.
Problem:
If a laboratory obtained a positive result for the blank control, this is clearly an error and the test
should be repeated after checking for possible causes, e.g. contaminated medium or faulty
technique.
If a laboratory obtained no positive result for any of the inoculation levels, this suggests that the
tubes were not inoculuted or that the medium does not support growth of the test organism or
that the primary sample did not contain the test organism. The test procedure should be carefully
checked for possible errors. The coordinating laboratory should ensure that correct results are
provided so that the mean results are not compromised.
If a laboratory obtained only positive results for all inoculation levels (except for the blank
control), this suggests that preparation of the inoculation levels was not done properly. The
procedure should be carefully checked for possible errors. The coordinating laboratory should
ensure that correct results are provided so that the mean results are not compromised.
If a laboratory obtained a negative result for the positive control, this is clearly an error and the
test should be repeated after checking for possible causes e.g. failure to inoculate the medium
or wrong level of test organisms in the inoculum.

Required version of Excel:

prise 3 informative levels which preferably should This program should run on Excel 2007 and newer. It al
ve control should be used but these require only runs on Excel 2003, however, no figure is generated (du
compatibility problems).
Furthermore, the buttons in the sheet 'Program' might n
work in older Excel versions. In any case, it is possible t
start the calculation with 'Ctrl+b' and to print normally via
File-menu.
Correct display of figures:

In order to display the figures, the computer must be
connected to a printer (driver). This can also be a conve
pdf or xps.
Revisions of the PODLOD interlab program

Version 1, dated 2019-09-18, is a program for estimatin
POD function and the LOD of a binary microbiological
measurement method from an interlaboratory experime
or plausibility. If results seem to be incorrect based on the

ld request clarification or repetition of the tests from the
d consider whether the results of the respective laboratory
Protocol used in this program:

No calculation is carried out for the respective
laboratory and its data are not included in the
calculation of the mean results.
The data of the respective laboratory are

included in the calculation of the mean results.



laboratory experiment
measurement method
can be downloaded at:
ion of Excel:
uld run on Excel 2007 and newer. It also
3, however, no figure is generated (due to
ems).
buttons in the sheet 'Program' might not
l versions. In any case, it is possible to
n with 'Ctrl+b' and to print normally via the
y of figures:
the figures, the computer must be
nter (driver). This can also be a converter to
he PODLOD interlab program:

019-09-18, is a program for estimating the
the LOD of a binary microbiological
hod from an interlaboratory experiment.

For more information see the sheets 'Info' and 'Example'. For details see: JARVIS, WILRICH & WILRICH: Estim

Name of the experiment Reference method
Escherichia Coli Meat 29.11.2018
25 10 5
Note: The values for d refer to 1 g or 1 ml and not to the sample size as given in row 11.
Input data
Laboratory 1 Designation: Lab 1
d n y
0.0 1 0
0.04 8 5
0.08 8 6
0.16 8 8
0.32 1 1

d n y
0.0 1 0
0.04 8 4
0.08 8 6
0.16 8 8
0.32 1 1

d n y
0.0 1 0
0.04 8 7
0.08 8 7
0.16 8 8

0.32 1 1

d n y
0.0 1 0
0.04 8 7
0.08 8 8
0.16 8 8
0.32 1 1

d n y
0.0 1 0
0.04 8 6
0.08 8 7
0.16 8 8
0.32 1 1

d n y
0.0 1 0
0.04 8 4
0.08 8 7
0.16 8 7
0.32 1 1

d n y
0.0 1 0
0.04 8 5
0.08 8 6
0.16 8 8
0.32 1 1


d n y
0.0 1 0
0.04 8 5
0.08 8 6
0.16 8 7
0.32 1 1

d n y
0.0 1 0
0.04 8 1
0.08 8 7
0.16 8 8
0.32 1 1

d n y
0.0 1 0
0.04 8 2
0.08 8 3
0.16 8 8
0.32 1 1

Example sheet (to use the program go to the sheet 'Program'):

This example 1 corresponds to the data for the reference method as given in the above mentioned publication
JARVIS, WILRICH & WILRICH, Journal of AOAC International, 2019, 102(5) 1617-1623.
Results
LOD50% = 50% level of detection in cfu/g
Estimate of Estimate of Estimate of Estimate of Estimate Lower 95%

log SD of log SD of log lab confidence
mean effect mean effect mean effect effect limit
F ̂_0 μ ̂ s_μ ̂ σ ̂ d0.5 d0.5,LCL
0.835 -0.180 0.088 0.319 0.033 0.027
The results are based on the data of laboratories 1, 2, 3, 4, 5, 6, 7, 8, 9 and 10
Results for the individual laboratories

Laboratory
Log SD of log
No. Designation Lab effect lab effect lab effect
i Laboratoryi F ̂_i fi sfi
1 Lab 1 0.881 -0.126 0.282
2 Lab 2 0.784 -0.244 0.278
3 Lab 3 1.462 0.380 0.319
4 Lab 4 2.256 0.814 0.388
5 Lab 5 1.233 0.209 0.302
6 Lab 6 0.714 -0.337 0.276
7 Lab 7 0.881 -0.126 0.282
8 Lab 8 0.690 -0.372 0.275
9 Lab 9 0.648 -0.434 0.275
10 Lab 10 0.420 -0.867 0.282

measurement method
t can be downloaded at:
he above mentioned publication

17-1623.
%
= 50% level of detection in cfu/g or cfu/ml
Upper 95% Lower 95% Upper 95%

confidence expectation expectation
limit limit limit
d0.5,UCL d0.5,LEL d0.5,UEL
0.041 0.016 0.070
boratories
Remarks

Estimated probability of detection, POD
Estimated mean POD curve, its 95 % confidence band, the 95 % expectation band and the POD curves of the labs
Reference method • Escherichia Coli in Meat • 29.11.2018
1
0.8
0.6
0.4
0.2
0
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4
Inoculation level d in cfu/g or cfu/ml
Mean 95% confidence limits 95% confidence limits
For more information see the sheets 'Info' and 'Example'. For details see: JARVIS, WILRICH & WILRICH: Estim

Name of the experiment Alternative method
Escherichia Coli Meat 29.11.2018
25 10 5
Note: The values for d refer to 1 g or 1 ml and not to the sample size as given in row 11.
Input data
d n y
0.0 1 0
0.04 8 7
0.08 8 6
0.16 8 7
0.32 1 1

d n y
0.0 1 0
0.04 8 5
0.08 8 6
0.16 8 7
0.32 1 1

d n y
0.0 1 0
0.04 8 3
0.08 8 5
0.16 8 8

0.32 1 1

d n y
0.0 1 0
0.04 8 5
0.08 8 8
0.16 8 8
0.32 1 1

d n y
0.0 1 0
0.04 8 5
0.08 8 6
0.16 8 8
0.32 1 1

d n y
0.0 1 0
0.04 8 6
0.08 8 6
0.16 8 7
0.32 1 1

d n y
0.0 1 0
0.04 8 5
0.08 8 6
0.16 8 8
0.32 1 1


d n y
0.0 1 0
0.04 8 6
0.08 8 7
0.16 8 8
0.32 1 1

d n y
0.0 1 0
0.04 8 7
0.08 8 8
0.16 8 8
0.32 1 1

d n y
0.0 1 0
0.04 8 5
0.08 8 8
0.16 8 8
0.32 1 1

Example sheet (to use the program go to the sheet 'Program'):

This example 2 corresponds to the data for the reference method as given in the above mentioned publication
JARVIS, WILRICH & WILRICH, Journal of AOAC International, 2019, 102(5) 1617-1623.
Results
LOD50% = 50% level of detection in cfu/g
Estimate of Estimate of Estimate of Estimate of Estimate Lower 95%

log SD of log SD of log lab confidence
mean effect mean effect mean effect effect limit
F ̂_0 μ ̂ s_μ ̂ σ ̂ d0.5 d0.5,LCL
0.938 -0.064 0.090 0.162 0.030 0.024
The results are based on the data of laboratories 1, 2, 3, 4, 5, 6, 7, 8, 9 and 10
Results for the individual laboratories

Laboratory
Log SD of log
No. Designation Lab effect lab effect lab effect
i Laboratoryi F ̂_i fi sfi
1 Lab 1 0.844 -0.169 0.280
2 Lab 2 0.690 -0.372 0.275
3 Lab 3 0.606 -0.500 0.275
4 Lab 4 1.359 0.307 0.311
5 Lab 5 0.881 -0.126 0.282
6 Lab 6 0.762 -0.271 0.277
7 Lab 7 0.881 -0.126 0.282
8 Lab 8 1.233 0.209 0.302
9 Lab 9 2.256 0.814 0.388
10 Lab 10 1.359 0.307 0.311

measurement method
t can be downloaded at:
he above mentioned publication

17-1623.
%
= 50% level of detection in cfu/g or cfu/ml
Upper 95% Lower 95% Upper 95%

confidence expectation expectation
limit limit limit
d0.5,UCL d0.5,LEL d0.5,UEL
0.036 0.019 0.045
boratories
Remarks

Estimated probability of detection, POD
Estimated mean POD curve, its 95 % confidence band, the 95 % expectation band and the POD curves of the labs
Alternative method • Escherichia Coli in Meat • 29.11.2018
1
0.8
0.6
0.4
0.2
0
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4
Inoculation level d in cfu/g or cfu/ml
Mean 95% confidence limits 95% confidence limits

PODLOD-interlab Ver1a

Uploaded by

Copyright:

Available Formats

PODLOD-interlab Ver1a

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

PODLOD-interlab Ver1a

Uploaded by

Copyright:

Available Formats

PODLOD interlaboratory calculation program, version 1, dated 2019-09-18

General information on the experiment

Sample size A0 in g or ml No. of laboratories No. of inoculation levels

Program by Wilrich & Wilrich 1/19 Printing date: 02/05/2020

How to use this program (Macros have to be enabled):

Program by Wilrich & Wilrich 2/19 Printing date: 02/05/2020

the sheet 'Info'.

you can re-calculate as per step 4 at any time.

Program by Wilrich & Wilrich 3/19 Printing date: 02/05/2020

he buttons are disabled in

Program by Wilrich & Wilrich 4/19 Printing date: 02/05/2020

Information on the design of the experiment:

Information on the evaluation of the results:

Program by Wilrich & Wilrich 5/19 Printing date: 02/05/2020

Required version of Excel:

Correct display of figures:

Revisions of the PODLOD interlab program

or plausibility. If results seem to be incorrect based on the

Protocol used in this program:

The data of the respective laboratory are

The data of the respective laboratory are

The data of the respective laboratory are

Program by Wilrich & Wilrich 6/19 Printing date: 02/05/2020

he PODLOD interlab program:

Program by Wilrich & Wilrich 7/19 Printing date: 02/05/2020

General information on the experiment

Laboratory 2 Designation: Lab 2

Laboratory 3 Designation: Lab 3

Program by Wilrich & Wilrich 8/19 Printing date: 02/05/2020

Laboratory 4 Designation: Lab 4

Laboratory 5 Designation: Lab 5

Laboratory 6 Designation: Lab 6

Laboratory 7 Designation: Lab 7

Laboratory 8 Designation: Lab 8

Program by Wilrich & Wilrich 9/19 Printing date: 02/05/2020

Laboratory 9 Designation: Lab 9

Laboratory 10 Designation: Lab 10

Program by Wilrich & Wilrich 10/19 Printing date: 02/05/2020

Example sheet (to use the program go to the sheet 'Program'):

Estimate of Estimate of Estimate of Estimate of Estimate Lower 95%

Results for the individual laboratories

Program by Wilrich & Wilrich 11/19 Printing date: 02/05/2020

he above mentioned publication

Upper 95% Lower 95% Upper 95%

Program by Wilrich & Wilrich 12/19 Printing date: 02/05/2020

General information on the experiment

Laboratory 2 Designation: Lab 2

Laboratory 3 Designation: Lab 3

Program by Wilrich & Wilrich 14/19 Printing date: 02/05/2020

Laboratory 4 Designation: Lab 4

Laboratory 5 Designation: Lab 5

Laboratory 6 Designation: Lab 6

Laboratory 7 Designation: Lab 7

Laboratory 8 Designation: Lab 8

Program by Wilrich & Wilrich 15/19 Printing date: 02/05/2020

Laboratory 9 Designation: Lab 9

Laboratory 10 Designation: Lab 10

Program by Wilrich & Wilrich 16/19 Printing date: 02/05/2020

Example sheet (to use the program go to the sheet 'Program'):