WEBVTT

1
00:00:00.000 --> 00:00:04.331
For this project, ask: could inadequate
mixing within the intended batch

2
00:00:04.481 --> 00:00:07.873
range lead to non-uniform product before
transfer?

3
00:00:08.023 --> 00:00:11.585
This is more useful than asking whether
the vessel is risky.

4
00:00:11.735 --> 00:00:15.981
The question identifies a possible
failure, product consequence and

5
00:00:16.131 --> 00:00:17.731
operating context.

6
00:00:17.881 --> 00:00:22.083
Our diagram shows the possible chain from
lost mixing to variable

7
00:00:22.233 --> 00:00:25.284
concentration and a potentially affected
dose.

8
00:00:25.434 --> 00:00:29.317
It is not measured process data and does
not prove that any real

9
00:00:29.467 --> 00:00:30.981
vessel performs poorly.

10
00:00:31.731 --> 00:00:34.206
A stopped agitator is a failure mode.

11
00:00:34.356 --> 00:00:39.177
Non-uniform concentration may be a
product-quality effect, depending on

12
00:00:39.327 --> 00:00:41.269
the formulation and process.

13
00:00:41.419 --> 00:00:44.811
An incorrect delivered dose may be a
patient consequence.

14
00:00:44.961 --> 00:00:49.377
These links require scientific
understanding; they are not automatic for

15
00:00:49.527 --> 00:00:50.613
every liquid.

16
00:00:50.763 --> 00:00:55.200
The exercise asks you to explain the chain
and list the evidence needed

17
00:00:55.350 --> 00:00:56.438
to judge it.

18
00:00:56.588 --> 00:01:00.022
Avoid assigning a score before you
understand the scenario.

19
00:01:00.772 --> 00:01:06.340
ICH Q9(R1) links risk decisions to
scientific knowledge and

20
00:01:06.490 --> 00:01:07.770
patient protection.

21
00:01:07.920 --> 00:01:12.379
The degree of formality should reflect
risk, uncertainty, importance and

22
00:01:12.529 --> 00:01:13.612
complexity.

23
00:01:13.762 --> 00:01:19.340
For our case, a small development study
may not represent the full batch range.

24
00:01:19.490 --> 00:01:23.330
Record that limitation and plan how to
reduce uncertainty.

25
00:01:23.480 --> 00:01:27.234
A low number based on a guess is not
stronger evidence than a

26
00:01:27.384 --> 00:01:28.878
clearly stated unknown.

27
00:01:29.627 --> 00:01:35.008
Invite people who understand the process,
equipment and potential consequences.

28
00:01:35.158 --> 00:01:39.382
Ask each person which assumptions could be
wrong and what evidence would

29
00:01:39.532 --> 00:01:40.897
change the decision.

30
00:01:41.047 --> 00:01:45.655
If the project deadline is influencing
scores, make that bias explicit.

31
00:01:45.805 --> 00:01:50.115
In the activity, write one risk question
and one cause-to-harm chain.

32
00:01:50.265 --> 00:01:54.314
Keep the language specific enough that
another reviewer can disagree with

33
00:01:54.464 --> 00:01:57.536
it constructively and identify what needs
checking.