2557-OMS-2Tners-80m.XP4 7.10.

2001 14:26 Page 215

8. WORKSHOP – STANDARD
OPERATING PROCEDURES

FOR THE PARTICIPANTS

The workshop is based on 6 SOPs which deal with the same problem : waste disposal. 215
Different departments, all on the same research site and all part of the same overall
organization, have written these SOPs. (See Appendix 4.)
Participants will be expected to do the following :

Review the SOPs

• Read the SOPs and note down any criticisms of the way the SOPs have been
written. Comments should be made page by page to simplify reporting back later
on.

Draw a Flowchart
• From the information contained in the SOPs, construct a single flowchart of waste
disposal at this research centre.

Comment
• Make recommendations to improve the situation.

Your instructor will explain the general situation to you and will comment on flow
charts before you start the workshop.
You will be divided into groups, each of which will be asked to report back to a ple-
nary session at the end of the workshop. So it is important to choose a spokesperson(s)
at the beginning of the session.
The instructor will be available during the workshop to answer your questions and
help you if necessary.
2557-OMS-2Tners-80m.XP4 7.10.2001 14:26 Page 216

TRAINERS Chapter 8 • GLP Training Manual

FOR THE INSTRUCTOR

Introduction to the Workshop

You have at your disposal two different types of flow chart. These should be shown
to the participants and commented on, but do not take more than 15 minutes to do
this. The flowcharts can be distributed to the participants, but underline the fact that
they are just examples of the different types of flowcharts that one sees habitually (see
Appendix 3) :
– The first is a structured chart showing the major steps in the production of a
report. Each major step is subdivided into its constituent parts.
– The second is a “question and answer” flowchart. This one deals with the arrival
216 and receipt of stock at a factory site.

Show the flowcharts and comment upon the usefulness of each. The most important
point is that the flowcharts replace what would otherwise be a long conversation with
someone or a long descriptive text.
Distribute to each participant a copy of each of the six different standard operating
procedures (SOPs) (see Appendix 4). These SOPs are from the same research centre
but written by a variety of departments. They all deal with the same kind of problem,
that of waste disposal.
Tell the participants that these SOPs reflect a real life situation where a manager was
asked to help sort out the problem of waste disposal because the staff responsible for
the collection and disposal of waste found the SOPs difficult or impossible to use.

The Task
Divide the participants into groups. Try not to have more than six people per group.
The groups should be provided a space to work in which is relatively free of interfer-
ence from the other groups.
Tell the groups that their workshop tasks are as follows :
– Read the SOPs and note any criticisms they have of the way the SOPs have been
written. Comments should be made page by page to simplify reporting back later
on.
– From the information contained in the SOPs, construct a single flowchart of
waste disposal at this research centre.
– Make recommendations to improve the situation.
2557-OMS-2Tners-80m.XP4 7.10.2001 14:26 Page 217

Chapter 8 • GLP Training Manual TRAINERS

The group should appoint a spokesperson to report back to a plenary session of the
participants when the task is completed.
The instructor should provide transparencies and pens so that the flowcharts can be
shown by overhead projection.
The instructor will regularly visit each group to answer any questions about the task
and to monitor progress.
Allow two hours for the group work. Some groups work faster than others so stop all
groups working, even if not all the groups have finished, before any one group gets
bored.

217
Reporting
During the plenary session, the spokespersons from each group present their
groups’ comments. Allow 15-20 minutes for each spokesperson to present their group’s
conclusions.

1. Listen first to the comments on the SOP contents. The SOPs are full of mistakes,
they lack coherence, have missing parts, etc. The instructor should listen to the
findings of each group, but to avoid repetition, should ask each group to report
on separate parts, or take the comments page by page, a few pages per group. The
instructor should comment on the points brought out by the groups and always
attempt to underline the cardinal points of GLP. It is likely that the general com-
ments will deal with the necessity to separate out different subjects and make
separate SOPs for them. These SOPs cover two main areas : waste disposal and
medical considerations. They are best dealt with in separate SOPs.

2. When the flowcharts are shown, let the spokespersons explain them as fully as
they wish. The instructor should ask questions about the charts but not make
any negative comments. The points that are important to develop during the dis-
cussion are :
a) before writing an SOP it is useful to draw up a flowchart of the procedure
because this will help you write the SOP in a coherent manner.
b) a good flowchart can replace a wordy SOP ; you do not always need text in an
SOP.
c) flowcharting is a useful tool when reviewing SOPs.
2557-OMS-2Tners-80m.XP4 7.10.2001 14:26 Page 218

TRAINERS Chapter 8 • GLP Training Manual

3. The spokesperson should also present what could be done to improve the situa-
tion. Most comments will probably be based on the need to get the authors of the
individual SOPs together so that they can rework their six SOPs into one or two.
The importance of a transverse review of SOPs (usually by QA) should under-
lined.

GLP WORKSHOP ON CASE STUDIES

Workshop instructions
Below are 11 case studies that have actually arisen during GLP investigations. With
218 your group, please discuss what you would do if faced with these situations, and be
ready to report back to the whole group.
You have about 90 minutes to consider what you would do in these cases. Be as
thorough as you can. Sometimes you may feel that you lack some information to reply
fully. If this is the case, say what extra information you would like and how you would
go about collecting it. Remember that there will be many ways of resolving these situ-
ations. Therefore, if your group comes up with more than one way of dealing with the
problem, please let us know.
When reporting back your group will need a spokesperson. Since each case will be
dealt with separately, you may, if you wish, nominate a separate person for each case.

Instructor’s notes
This workshop session requires groups of participants to discuss actual case studies
that may well occur in GLP studies.
As before, the instructor should designate groups, and after each group has dis-
cussed each case study, the groups should report back in a plenary session where the
ideas of everyone can be aired. So, remind the groups to choose a spokesperson for
each case study and provide transparencies and pens.
You should allow 90 minutes for the groups to decide on their responses to the
cases, and reporting back will take another hour.
Once again, the important point to underline is that there are no “absolute answers”
just ways of approaching these issues, based on the GLP Principles and common sense.
When the groups report back, the instructor should constantly encourage discus-
sion by posing relevant questions and give his/her own point of view based on a com-
plete understanding of GLP.
2557-OMS-2Tners-80m.XP4 7.10.2001 14:26 Page 219

Chapter 8 • GLP Training Manual TRAINERS

To help the instructor, notes for discussion have been added after each case study.

1. You are the study director of a study being conducted to determine the blood
levels of a compound. When the analytical results are reported, it is shown that
the control group of animals has been exposed to the test article at some time
during the study. What should you do ?

Discussion ideas

Is it possible that the analytical results are not correct, that samples have been con-
taminated in the analytical laboratory ? Is the analytical laboratory GLP compliant ?
How bad is the exposure of control animals ? The study director should decide and 219
document the impact on the study. If the study is now invalid, who has to be
informed ?
Is it possible to trace from the raw data the source of exposure ? Have the animals
been wrongly dosed or has there been a mistake when making up the dosing solutions
or in labelling ?
Can an individual(s) be identified as responsible ; in which case there may well be a
need for additional training ? Review of the whole training programme may be
required.

2. During a long-term study, the environmental parameters that are measured on a

daily basis (temperature, humidity, light intensity) have, at various times, been
out of specification. How should a study director deal with this ?

Discussion ideas

How extreme are the variations ? As it is a long-term study, there may be no impact
on study validity. The study director must record his/her view of the impact on the
study.
The final study report must contain a record of this deviation from specifications.

3. The drinking water provided to animals is monitored for quality on a three

monthly basis in your laboratory. In January, the results were satisfactory and
2557-OMS-2Tners-80m.XP4 7.10.2001 14:26 Page 220

TRAINERS Chapter 8 • GLP Training Manual

within specifications. In March, the results were out of specifications for total
germ count. What consequence does this have on the studies conducted between
January and March ? What should happen ?

Discussion ideas

How extreme are the results ? The study director must record his/her view of the
impact on the study.
It is essential to try and find out when and where the contamination started and
why, but this may not be possible. It is possible that the contamination has affected all
the other studies where the animals drink the same water, in which case all the studies
220 will have to be evaluated for the effect that this contamination has had.
Possible actions include : disinfecting the system ; performing more monitoring of
the system to ensure that the contamination has been eliminated ; monitoring on a
schedule that is more frequent than three monthly until you are confident that the
contamination has been completely eliminated.

4. A contract laboratory is performing a GLP study for a sponsor. Part of the study
consists of a hormone analysis for which the contract laboratory has neither the
expertise nor the equipment. A university department close to the contract lab
has the required expert knowledge and equipment. What policy should the con-
tract lab adopt in this situation ?

Discussion ideas

Is the university laboratory performing to GLP compliance ? If so, no problem.

If the university is not GLP compliant, the contract laboratory and the sponsor must
decide whether or not the data are the pivotal part of the study. If so, then there could
be an attempt to render the laboratory as GLP compliant for this assay in the short
term, say in 3-6 months.
If making the laboratory GLP compliant is not a possibility, the university labora-
tory can still perform the assay, but the contract laboratory or the sponsor will have to
monitor very closely the conduct of the assay. The university laboratory could use
some of the contract laboratories SOPs for example, and the university laboratory
could provide its raw data for verification to the contract laboratory QA group, etc.
2557-OMS-2Tners-80m.XP4 7.10.2001 14:26 Page 221

Chapter 8 • GLP Training Manual TRAINERS

The contract laboratory QA group could also inspect the university laboratory in an
on-going phase.
The study director must state the non-GLP status of the university in the final
report. The receiving authority could reject the study if it believes that the hormone
assay was not sufficiently controlled by the contract laboratory or the sponsor, but this
would depend on the importance of the data coming from the assay.

5. During a routine quality assurance audit of an analytical laboratory, the auditor

requests the technician to stop his work because he believes that samples are
incorrectly labelled. What should happen in this case ?
221
Discussion ideas

The question that arises is whether or not it is legitimate for QA to interrupt the
normal work of study staff. There will be different views on this point.
Remember that if the QA person says nothing, the study may be invalidated because
of this possible error in manipulation. As a general rule it is worth calling in a member
of staff superior to the technician to confirm that there is no mistake, or to confirm
that there is a mistake.
QA should be regarded as helping in these situations, not as a hindrance to normal
operations.

6. A computer system has been used in your laboratory for the acquisition of body
weight data for many years. It has, however, never been validated. A regulatory
inspection is expected in about three weeks time. What should you do about this
computerized system ?

Discussion ideas

Three weeks is probably too short to do a complete validation.

Try to write a validation protocol and show this to the inspector (if he requests it)
and ask his opinion on its content. This will show that at least you have decided to val-
idate and have started the process of validation.
Collect as much evidence from the past that the computer system has been per-
forming correctly.
2557-OMS-2Tners-80m.XP4 7.10.2001 14:26 Page 222

TRAINERS Chapter 8 • GLP Training Manual

Be ready to describe to the inspector the systems you have in place for secured
access to the system (passwords, access levels) if he asks for it. The same applies to the
procedure you have for backing up and archiving data.

7. An inspector from your national regulatory authority calls you on the telephone
to say that he will be inspecting your laboratory in two weeks time. How should
you prepare for this “visit” ?

Discussion ideas

222
You are lucky to be informed beforehand.
There is no way, in two weeks, of transforming a non-GLP laboratory into a GLP
compliant laboratory.
What you can do is :
– Check that the organizational chart is up to date.
– Verify the exactness of personnel documentation.
– Make sure that any SOPs which are still being reviewed or revised pass quickly
through the system and are available for the inspector.
– Make sure that the archives are in good order.
– Make sure that personnel are aware of the inspection. Brief them on how to
behave with the inspector.
– Reserve a meeting room to receive the inspector.
– Clear up any obvious mess !
– Find out from him why he is coming. Is he/she coming for a general inspection
of the facilities, for a specific compound and study(ies), or at the request of a for-
eign authority, etc. ?

8. During a QA audit performed at the week-end on a three month iv mouse study,

it is noted that there are no records of training, for intravenous injections into the
tail vein of mice, relating to one of the technicians who has performed this task.
What should QA do ?

Discussion ideas
2557-OMS-2Tners-80m.XP4 7.10.2001 14:26 Page 223

Chapter 8 • GLP Training Manual TRAINERS

During the weekends, persons are sometimes found to be performing work they
have not been trained for. In this case, QA should determine whether or not the person
has really been trained. It may be that the training record is incomplete.
The study director must be informed. He/she should determine whether there is an
impact on the study.
QA would do well to inspect all the training records for all personnel working at
weekends to find out whether this is a general problem or not.
If there is a general training issue this should be reported to management so that all
necessary training can be undertaken.

9. A final report, already signed by the study director, is found to contain some erro- 223
neous data (miscalculation means that one outlying animal was not eliminated
from the calculations when it should have been). What should happen now ?

Discussion ideas

This is unfortunate, as you cannot ever change a final report.

The correction will have to be made by sending an amendment to all the people who
received the original final report.
The amendment will contain all the new calculations and should also explain where
in the original report these calculations supersede what has already been reported.
The study director must sign the amendment.
The amendment must be audited by QA to ensure its accuracy. QA must also pro-
vide a QA statement to accompany the report amendment.
This unfortunate incident underlines the importance of quality control which
should be conducted by the study staff prior to the release of draft reports and the
importance of a QA review prior to the release of the final report.

10. During an inspection of the dosing of dogs, several cases of rejection of formula-
tion (dogs vomiting when replaced in their cages) were observed. However this
was not recorded in the raw data by the technician. How serious is this ? What
should be done about this ?

Discussion ideas
2557-OMS-2Tners-80m.XP4 7.10.2001 14:26 Page 224

TRAINERS Chapter 8 • GLP Training Manual

This is very serious indeed as it could invalidate the study since we no longer have
any reliable idea as to how much of the test item has been administered to the animals.
It is not an infrequent problem, particularly at the start of studies.
You cannot stop the dogs rejecting material, but the study director must be aware of
this, otherwise he/she will not be able to interpret the data correctly.
The SOP for dose administration should be reviewed. It may not indicate that the
dogs must be watched carefully for a period after dosing.
The training of the technicians should be reviewed ; it may be that they are not
aware of their role in this case.
Some attempt should be made to find out if this was a “one-off ” incident. Have there
been records of this in the data during the study so far ?
224 In the worst case scenario, it may be necessary to abandon the study and start again.

11. In the histology laboratory, staff have the habit of sticking non-controlled photo-
copies of SOPs useful for their techniques on the walls. What should you do
about this ?

Discussion ideas

This practice should be discouraged if the SOPs in question are not controlled
copies. This is because it is possible that they will continue to use these “on the wall
instructions” even after the real SOP has been revised.
It may be possible to replace the SOP on the wall with a proper, controlled version.
If so do not forget to ensure that the distribution of this wall copy is recorded, as usual,
in the management system. Remember it will have to be collected in, destroyed and
replaced when any revisions are made to it.