Anisfield - Investigating

Effective Investigations
and
Corrective Actions (CAPA)
P
Proper
Investigation
I
i i off Quality
Q li E
Events
Michael H. Anisfeld
Globepharm Consulting
What Do You Do When The Bad Stuff Happens?
Pretend It Never Happened
Ignore
It
I
!#&*!$%...
Curse
Advise Boss
Advise Your Buddy

Walk Away
Fix It
Its a One-Off
Investigate + Fix
Blame Someone Else
Investigate
Investigate, Fix + Document
Copyright, Globepharm Consulting, 2008
Failure/Deviation System
The purpose of
Th
f a failure/deviation
f il
/d i i system is
i
to assure that each failure/deviation does not
adversely impact product quality and that
effective corrective action is taken to reduce
the probability of such failure in the future
Contributors to Failure
Materials
Facilities
Equipment
Instrumentation
People
Processes
Procedures
Failure / Deviation
Definition
F il
Failure
Product does not meet specification.
Failure may be detected in/by
Production
Laboratory
Deviation
Change in procedures, equipment, materials,
personnel related to product manufacture having
an impact on product quality
6
Definitions
Immediate Cause
Situation directly causing the problem
Root Cause
Basic Causal Factor, which if corrected or removed prevents a
repeat of the problem
Intermediate Cause
Reason for problem at more fundamental level than the
immediate cause, but not the root cause [why-2, why-3]
Root Cause Analysis
Structured questioning process enabling identification of
underlying beliefs and practices that result in poor quality
7
Causes of Failure
Common
C
mm n Cause
C
[End mi S
[Endemic
Systematic]
t m ti ] (Variation)
(V i ti n)
Originates from the basic elements of the process:
Machines
Materials
Methods
Manpower
Measurement
Assignable Cause [Special] (Variation)

An unplanned variation originating outside expected
operating process conditions [special cause variation]
8
Failure/Deviations
Measures
There are three key steps in this process and each of these should
be monitored regularly
Recording the failures and deviations
Total number
Number by department
Action on product batch(es) involved
Audit by sampling to verify that corrective action was taken
Corrective action on root causes
There should be an elimination of failures/deviations due to causes

for which corrective actions have been implemented
Regulators Expectations
Comprehensive,
C
h
i
H
Honest A
Approach
h To
T
Investigate OOS Results/Failures
Evaluation Using Scientifically Valid Principles
Learning From Experiences
Permanent Solutions To Problem
Authorities Not The Ones To Find The
Problem
10
11
What to Do
When the Unexpected Happens
UNEXPECTED?
OOS
Laboratory
Manufacturing
OOT
OOE
OOY
Water System
Reconciliation Yield Check

Reconciliation Packaging Material
Deviation
Discrepancy
Unusual Event
12
13
What Do The US Regulations Say?
Investigation
In
ti ti n specifically
p ifi ll stated
t t d in the
th cGMPs:
GMP :
211.22
211.125
211.170
211.180
211.186
211.188
211.192
211.198
211.204
Responsibilities of quality control unit.

Labeling issuance.
Reserve samples.
General requirements.
Master p
production and control records.
Batch production and control records.
Production record review.
Complaint files.
Returned drug products.
14
What Does The EU Guide Say?
I
Investigation
ti ti specifically
ifi ll stated
t t d iin th
the GMP
GMPs:
1.3 Quality Assurance
1.4 Quality Control
2.7 Production/QC Responsibilities
5 39 Production
5.39
5.56 Packaging Operations
6.2 Quality Control
8.1 Complaints
15
Future Regulations
An international
i
i
l harmonized
h
i d
approach to
Risk Management
ICH-Q9
16
ICH-Q9
Risk Management
initiate
risk management process
risk assessment
Risk man
nagement tools
s
risk identification
risk analysis
risk evaluation
unacceptable
risk control
risk reduction
risk
i k acceptance
t
risk communication
risk communication
output/results of the
OK
risk acceptance
review event
17
CAPA
19
How do you know

whats happening?
R
Recognize
i
that
h there
h
is
i a problem:
bl
Capture The Data
Analyze The Data
Trend The Data
Then
hen Fix the problem
problem:
Investigate the Cause
Attempt to Get to Root Cause
Implement Effective Fixes
Monitor the Fix
20
10
21CFR820.100 - CAPA
820.100 Corrective and preventive action
(a) Each manufacturer shall establish and maintain procedures
for implementing corrective and preventive action. The
procedures shall include requirements for:
(1) Analyzing processes, work operations, concessions,
quality audit reports, quality records, service records,
complaints, returned product, and other sources of quality
data to identify existing and potential causes of
g p
product,, or other quality
q
y problems.
p
nonconforming
Appropriate statistical methodology shall be employed
where necessary to detect recurring quality problems;
(2) Investigating the cause of nonconformities relating to
product, processes, and the quality system;
(3) Identifying the action(s) needed to correct and
prevent recurrence of nonconforming product and other
quality problems;
21
21CFR820.100 - CAPA
820.100 Corrective and preventive action (continued)
(4) Verifying or validating the corrective and preventive action
to ensure that such action is effective and does not adversely
affect the finished device;
(5) Implementing and recording changes in methods and
procedures needed to correct and prevent identified quality
problems;
(6) Ensuring that information related to quality problems or
nonconforming
f mi product
d t iis disseminated
di
mi t d to
t th
those di
directly
tl
responsible for assuring the quality of such product or the
prevention of such problems; and
(7) Submitting relevant information on identified quality
problems, as well as corrective and preventive actions, for
management review.
(b) All activities required under this section, and their results,
shall be documented
22
11
CAPA Analyzing Sources

of Quality Data
Environmental
E i
l
Monitoring
Complaints/
Returns
Repairs/
Ser icing
Servicing
Process
Deviations
or Failures
Acceptance
Activities
CAPA
Stability
Data
Calibration/
Maintenance
Audits
Yields/Scrap
23
What To Track
T
Track
k By:
B
Product
In-process, release, stability, validation
Event
Reason for investigation
Operator/Analyst
If cause is operator/analyst error
Instrument/Equipment
If cause instrument related
Cause
Investigation conclusion
Corrective Action
Remedy and prevention
Timeliness
Time to complete investigations

24
12
CAPA Data Collection

Incidents Reported - 2003
Question:
250
200
Are we Getting Better as a Company?

Or Dropping Off in Capturing Incidents?
210
186
178
165 171
150
143
Incidents
132 130
100
119
98
85
92
50
0
Jan Feb Mar Apr May Jun
Jul
Aug Sep Oct Nov Dec

25
Run Charts: Trend Charts
Environmental Data Purified Water System
Specification
Upper Limit
OOT? / OOE?
Specification
Lower Limit
26
13
Tracking / Trending
Paper/Pen
Spreadsheet (Excel, Oracle)
Dedicated Computer Systems
Trackwise
(Sparta Systems: http://www.sparta-systems.com)
CATSweb
(AssurX: http://www.assurx.com/catsweb.html)
p
MetricStream
(Metric Stream: http://www.metricstream.com)
SmartCAPA
(SmartCAPA: http://www.pilgrimsoftware.com)
Others ..
27
SmartCAPA
(www.pilgrimsoftware.com)
28
14
29
Purpose of Investigation
The
Th purpose of
f an investigation
i
i
i it
i to
determine the cause of the failure.
Even if the batch is rejected based on the OOS
result, the investigation is necessary to determine
if the result is associated with other batches of
the same drug product or other products
FDA: Draft OOS Guideline, September 1998

30
15
Failure / Deviation What To Do
Failure/Deviation
Detected
F il
/D i ti
D t t d
I f
Inform
QA
24 hours
Perform Investigation
Batch Disposition
Immediate Fix
Root Cause Analysis
30 days
Communicate to Management
Corrective/Preventive Action
Assess Effectiveness of Fix
31
Investigations:
Youve Got to Be A Detective!
32
16
Investigations 101
P p
Purpose
off ann In
Investigation
ti ti n
Obtain facts as to cause
Common
Special - Assignable
Process of an Investigation
Complete
p
description
p
of problem
p
Validity of inputs
Review key process variables
Determine scope of investigation
Document investigation
33
Learn How To Be A Detective
What Equipment,
Equipment
Machine, Tool?
What Is Wrong?
What Is the
Complaint?
What is the Undesired
Behavior?
34
17
Wh is
Who
i Involved?
I
l d?
Staff
Consultants
Vendors
Visitors
Name names/positions!
35
When does
Wh
d
the
th problem
bl
occur?
Day
Date
Time
Shift
Phase of operation
When in equipment life
cycle
Are there time patterns?

36
18
Whi h
Which:
Unit
Area
Department
Line
Machine
Location of defective
item, or where on
defective item
37
How is the what

what or who
who
impacted?
Injury / Death
Shut-Down /Start-Up
Damage
Type
Classification of Defects
How Much
How Many
How Big An Issue
What resources are needed

to handle/resolve the issue!
38
19
Change Analysis
PROBLEM?
E
COMPARISON?
WHAT?
What is the object or

process with the problem?
What is wrong?
What similar object or

process doesnt have the
problem? What else could be
wrong?
WHERE?
Where does the problem

occur
Where else could the

problem occur?
WHEN?
When did the problem

happen?
When did the problem not

happen?
How large is the problem?
How large could it be?
HOW BIG?
39
Investigation Documentation
E
40
20
Investigation Documentation
Date
D
t p
problem
bl m occurred
d
Statement of the Problem
Listing of batches impacted
List of personnel interviewed
Statement of each persons opinions and memory of event
Discussion of potential causes
Discussion of potential impact
Discussion of extent of problem
Recommendations for fixes
Statement of actions taken to fix
Final decisions regarding batch
Management approvals
41
42
21
Root Cause Analysis
Corrective
C
ti actions
ti
tto observations
b
ti
((usually
ll symptoms)
t
)
have limited effect on sustained improvement or
elimination of recurrence of the failure
Corrective actions need to address the underlying
causes and root causes of a problem in order to
eliminate the situation from recurring
43
Root Cause Analysis

Equipment
Design, Capability
Design
Maintenance
Operator Error
Procedures
No Procedure
Wrong Procedure Used
Procedure Not Available
Procedure Difficult To Use
Training
None
When
Effectiveness
Quality Control
Inspection Required
Inspection Performed
Management Systems
Qualified Supervision
Audits
Feedback
Prior Corrective Actions Taken
Communication
Planning
Process Capability
Adequate Time To Perform Task
Standards
Manpower
Information Available
Adequate
Timely
Personnel Problems
Training
Fail/Safe Processes
Environment
Lighting
Noise
Fatigue
44
22
Corrective Action
What
level
Wh is
i the
h lowest
l
l
l at which
hi h we
can do something to prevent the
problem from re-occurring?
45
46
23
Investigation Tools
Brainstorm
B
i t
th
the iissues
Flowcharting
Fishbone Analysis Cause/Effect diagrams
Why-Why Analysis
Pareto Charts
Run Charts
Force Field Analysis
Six-Sigma Analysis - MAIC
Kepner-Tregoe
47
Brainstorming
Possible
P
ibl ttargett questions
ti
to
t serve as the
th focal
f
l point
i t of
f
the brainstorming process:
What would solve the problem?
What strategy could resolve the root cause?
What solutions have already been thought of?
What solutions have not been thought of?
How can we prevent the situation recurring?
What different methods might work?
What crazy ideas might help?
48
24
Fishbone Analysis
Manpower
Management
People too busy

People not paying attention
Batch
Record
Errors
Methods
Machines
49
Why-Why Analysis
Oil Drops
D
On
O Fl
Floor
Dripping Engine
Oil Can to Collect
Funnel in Oil Can
Raise Funnel
Unless youve gone
5 levels of questions
you probably have
not got to root cause!
50
25
Pareto Analysis
Pareto: 80 20 rule
Batch Record Problems

87
90
80
70
60
50
40
30
23
20
18
10
12
0
Signatures
Data
Errors
Other
51
Process Shift, F-440 Particle Size
26
APR Annual Product Review
53
Force-Field Analysis
The
Th Positive/Negative
P iti /N
ti
Analysis for Solutions
can stimulate thinking

just as a the force-field
format did in the past.
Since we are looking for
solutions, the two
columns play better
better
against worse.
What
Wh t would
ld make
k
the problem
better?
1.
2.
3.
4.
5.
6.
7.
8.
What
Wh t would
ld make
k
the problem
worse?
1.
2.
3.
4.
5.
6.
7.
8.
54
27
Six Sigma Problem Solving MAIC

Measure
(1) Describe the
Problem
A nalyze
(4) Identify Potential
Causes
Improve
(5) Analyze
Existing Data
Control
(9) Determine Best Solution
(12) Implement
Solution
0.251
0 251
IS
IS NOT
Control Plan
0.250
What
Item
Function
(2) Lid
Prevents Spills
Failure Mode
Lid falls off due
to looseness
0.249
Where
Leaks from drinking hole
0.248
When
Extent
12.5
6
Jan '87
Jul '87
Jan '88
Jul '88
Month
Number
Code
Spill when removing lid
(10) Pilot Solution
Burn
Go-no go gauge
Body diameter
to small
Burn
Go-no go gauge
Surfaces too slick
Burn
Sampling Plan
Hole too large
Stain
Pin gauge
Hole has too

direct of path
Stain
Lid diameter
too small
Burn
(above)
Body diameter
to large
Burn
(above)
Material catches
Burn
Sampling Plan
Cause 1 O
Cause 2 X
Cause 3 O
C
Cause 4 X
Cause 5 O
X X O A
O A AO
O AA O
A X O O
O OO X
(11) Verify Solution Works

LSL
15
USL
20
25
30
35
Seal Strength
OC Curve
70
T M B
(8) Collect Additional Data

Until Root Cause Identified
(3) Measure Problem

Magnitude
T M B
Fact 1
Fact 2
Fact 3
Fact 4
Fact 5
FACTS
(1) All Machines
(2) Second Shift
(3) Certain Codes
(4) Started 8/22
(5) Steadily Worse
(6) All Operators
19
T M B
Effects Control
Lid diameter
too large
(7) Compare
Causes to Facts
(6) Construct List of

Verified Facts
(2) Determine When

Problem Started
T M B
Causes
T
O
R
Q
U
E
10.2
K = 20
60
1.0
Upper Control Limit
10.0
50
K = -20
Pa
Radius
9.8
30
9.6
36.0
37.0
38.0
0.5
Lower Control Limit
40
10
Subgroup
15
20
0.0
Percent Defective
10
55
Copyright 2000, Taylor Enterprises Inc.
56
28
ICH-Q9
Risk Management
initiate
risk assessment
Risk man
nagement tools
s
risk identification
risk analysis
risk evaluation
unacceptable
risk control
risk reduction
risk
i k acceptance
t
risk communication
risk communication
output/results of the
OK
risk acceptance
review event
57
Tools To Prioritize
Fixes and Efforts
FME
FMEA
- Failure
F il
M
Mode
d Eff
Effect Analysis
l i
FTA
- Fault Tree Analysis
HACCP - Hazard Analysis,

Analysis
Critical Control Points
58
29
Bibliography
Root Cause Analysis:
- Kepner and Tregoe, The New Rational Manager

- Paradies and Unger, Taproot: The System for Root Cause Analysis,
Problem Investigation & Proactive Improvement
- Ammerman,
Ammerman The
The Root Cause Analysis Handbook
Handbook
- Pokras, Systematic Problem-Solving and Decision-Making
Risk:
McDermott et al., The Basics of FMEA
Human Error:
Lorenzo, ABS Consulting, A Managers Guide to Reducing Human Error
g
Managements
Role:
Harbour and Kieffer, Managing the Quality System
FDA:
Investigating Out of Specification (OOS) Test Results for

Pharmaceutical Production (Guidance), September 1998
Quality Systems Approach to Pharmaceutical GMPs, September 2004
59
Stay In Touch
Michael
H. Ani
Anisfeld
Mi h l H
f ld
Globepharm Consulting
313 Pine Street
Deerfield IL 60015, USA
Phone:
Fax:
E-mail:
Website:
USA +1+847 914 0922

USA +1+847 914 0988
manisfeld@globepharm.org
www.globepharm.org
60
30
61
Questions?
62
31

Anisfield - Investigating

Hochgeladen von

Dokumentinformationen

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

Anisfield - Investigating

Hochgeladen von

Copyright:

Verfügbare Formate

Effective Investigations

What Do You Do When The Bad Stuff Happens?

Pretend It Never Happened

Advise Your Buddy

Blame Someone Else

Investigate, Fix + Document

Copyright, Globepharm Consulting, 2008

Copyright, Globepharm Consulting, 2008

Copyright, Globepharm Consulting, 2008

Assignable Cause [Special] (Variation)

Copyright, Globepharm Consulting, 2008

Action on product batch(es) involved

Audit by sampling to verify that corrective action was taken

Corrective action on root causes

There should be an elimination of failures/deviations due to causes

Copyright, Globepharm Consulting, 2008

Reconciliation Yield Check

Copyright, Globepharm Consulting, 2008

What Do The US Regulations Say?

Responsibilities of quality control unit.

Copyright, Globepharm Consulting, 2008

What Does The EU Guide Say?

Copyright, Globepharm Consulting, 2008

Copyright, Globepharm Consulting, 2008

How do you know

Copyright, Globepharm Consulting, 2008

Copyright, Globepharm Consulting, 2008

CAPA Analyzing Sources

In-process, release, stability, validation

Reason for investigation

If cause is operator/analyst error

If cause instrument related

Remedy and prevention

Time to complete investigations

Copyright, Globepharm Consulting, 2008

CAPA Data Collection

Are we Getting Better as a Company?

Aug Sep Oct Nov Dec

Run Charts: Trend Charts

Environmental Data Purified Water System

Copyright, Globepharm Consulting, 2008

(Sparta Systems: http://www.sparta-systems.com)

(Metric Stream: http://www.metricstream.com)

Copyright, Globepharm Consulting, 2008

FDA: Draft OOS Guideline, September 1998

Copyright, Globepharm Consulting, 2008

Failure / Deviation What To Do

Root Cause Analysis

Copyright, Globepharm Consulting, 2008

Learn How To Be A Detective

Copyright, Globepharm Consulting, 2008

Learn How To Be A Detective

Learn How To Be A Detective

Are there time patterns?

Copyright, Globepharm Consulting, 2008

Learn How To Be A Detective

Learn How To Be A Detective

How is the what

What resources are needed

Copyright, Globepharm Consulting, 2008

What is the object or

What similar object or

Where does the problem

Where else could the