Abstract

Deviations are the differences which are measured between the expected or normal
values and the observed values for a product or process condition from a procedure
or a documented standard. Deviations occur almost every day in the pharmaceutical
industries, handling of deviations and minimising the recurrence of them has very
critical considerations in the quality management system of the pharmaceutical
industry. This article explains the procedure for handling of deviations by identifying
the deviation, understanding its criticality, conduct root cause analysis and
suggesting corrective and preventive action for it. This is explained through various
case studies such as out of calibration equipment’s, facility modification, temperature
excursions, out of specification during the manufacturing process, with a detailed
process about the handling of deviations.

Keywords
Deviations, Root cause, Criticality, Corrective and Preventive Action

“Deviations are the measured differences between the observed and expected or
normal values for a product or process condition or a departure from a documented
standard or procedure”. A deviation may occur during testing and sampling of
finished products and raw materials acceptance and manufacturing. Deviations can
also be found out by the complaints given by the customers or comments given by
the customer when the company’s standards do not meet the critical quality
attributes as per the requirements. For the sake of continuous improvement and
compliance to Good Manufacturing Practice (GMP), if any deviation occurs from the
official procedures, then it must be documented. Food and Drug Administration
(FDA) part 211.192 requires a thorough investigation of any deviation including
documentation of conclusions and follow up. Quality Risk Management (QRM)
principals employed in the firm should make sure that all the deviations occurred are
rectified and recorded [1].

Deviation Handling
Among the crucial elements of the quality management system (QMS), handling the
deviations plays an important role to assure the quality of the product by
continuously improving the quality of it. As a part of Corrective and Preventive action
(CAPA), if the deviation is once detected, then it needs immediate action (i.e.,
corrections), the root cause analysis should be done and systemic actions need to
be implemented (i.e., corrective actions) to prevent non-conformances in the future.
Recently, QRM is being used to prevent the risk of deviations in the pharmaceutical
industry. International standards like International Conference on Harmonisation
(ICH) guideline Q9 (ICH Q9) and World Health Organization (WHO) recommends
using QRM system in the pharmaceutical industries. If any deviation occurs, how the
personnel reacts to it is the main challenge to a system. This mainly depends on the
level of training, qualification, commitment and support from the higher authorities of
the company [2].

In order to record, classify and investigate the events based on their risk, decision
tree will be used so that the person can make proper decisions regarding it. The
decision tree explains the simplified assessment of risk which answers the following
questions (fig. 1).

Figure 1: Classification of deviations

Whether the events affect the quality of the product? Do the approved specifications
or written procedure which is examined, conflict with a requirement? [2]
Types of Deviations[3]
Deviations are of two types: Planned deviations, unplanned deviations

Planned Deviations:

Any deviation from a standard procedure selected intentionally for a short period to
avoid undesirable situation without affecting the safety and quality of the product or
procedure.

E.g.: Batch executed with lower input due to non availability of raw materials [4].

Unplanned Deviations:

Unplanned deviations are the accidental non conformance observed after or during
the implementation of an activity. Unplanned deviations may occur due to the
following reasons: Equipment breakdown, Interruption of Power supply, Site
Accidents, Utility Breakdown

Errors during documentation. Deviations may be further categorized into 3 types

based on the impact of the deviation on the product quality, safety and validation
state of the facility and process [5]: Critical, major, minor

Critical deviation:

The deviation will have a notable impact on the critical attributes of the product.

For Example: Usage of contaminated raw materials and solvents. Integrity failure of
high efficiency particulate air filters [5].

Major Deviation:

The deviation will or may have a notable impact on critical attributes of the product.
For Example: Critical process and in process parameter failure. Significant variation
from standard output range.

Minor Deviation:

The deviation will not have any direct impact on the quality of the product.

For Example: Weights not replaced properly after use. Equipment and measuring
device malfunction [5]

Procedure For Handling Of Deviation

When to raise a deviation:

As soon a deviation occurs it should be immediately reported to Quality assurance

(QA) within one working day from the time when the deviation has occurred or as per
the Standard operating procedure (SOP).

Record the issue in the deviation record, the time of occurrence and name of the
department and the person who observed it should be documented. Standard
against which the deviation occurred should be recorded [6]. The general flow for
handling of deviations in depicted in fig. 2.

Figure 2: Flow chart for deviations

Initial details:

All the initial details regarding the deviation should be mentioned. All technical details
and critical process parameters should be recorded. Track the progress [6].

Investigating deviations:
Establish what happened, Understand the events, Use investigation tools, Identify
the causes and check for any common causes. Check if any other materials,
components, batches and equipment are affected? [6]

Root Cause Analysis

Root cause analysis is a systematic process for identifying “root causes” of problems
or events and an approach for responding to them. Root cause analysis can be
identified and classified as errors caused by Men, Material, Machine, Method and
Mother Nature.

Tools for identifying the Root cause

Fishbone analysis [6]:

A fishbone diagram, also called a cause and effect diagram, is a visualization tool for
categorizing the potential causes of a problem to identify its root causes. A fishbone
diagram is useful in product development and troubleshooting processes to focus the
conversation. After the group has brainstormed all the possible causes for a
problem, the facilitator helps the group to rate the potential causes according to their
level of importance and diagram a hierarchy. The design of the diagram looks much
like a skeleton of a fish. Fishbone diagrams are typically worked right to left, with
each large “bone” of the fish branching out to include smaller bones containing more
detail (fig: 3).

Figure 3: Fish bone analysis

5 whys analysis [7]:

5 Why analysis is used as a tool in root cause analysis. It is a set of five questions to
find out the base of the problem. Sometimes, it is necessary to find out by asking
more than 5 questions. Ask a question ‘why’ repeatedly to know the root of the
problem until you find out the correct root cause.

For Example: Why the machine was stopped suddenly? Answer: Due to human
error, Why did the human error occur? Answer: Human suddenly pressed the stop
button. Why did the human press it? Answer: The label on the button was not visible.
Why was the label not clearly?

Answer: Because it was covered with dirt. Why was it covered with dirt? Answer: It
was not properly cleaned. Therefore, here we understand that the root cause is no
proper cleaning practice [9].

Fault Tree Analysis (FTA) [8]:

Fault tree analysis is a tool to find out the root cause analysis for the deviations. This
helps to evaluate the failure of system one at a time and sometimes, by identifying
the casual chain of events, multiple causes can be combined (fig. 4).

Figure 4: Fault tree analysis

The results of these events are represented pictorially in the tree form. FTA is used
to investigate the deviations and complaints to understand the root cause and to
make improvements so that it does not lead to further problems [9]. Few other tools
for identifying the causes are: Pareto charts, brainstorming, flowcharting, change
analysis [5]

Closure of the Deviations

Prepare an investigation report for the occurred deviation and explain what
happened and why did it happen. Identify the Root cause for the deviation. Write
corrective and preventive action. Finally, Quality Assurance shall review the reports
and proceed further for the closure of the deviation. [fig. 5]

Figure 5: Closure of deviations

Case Studies on Handling of Deviations
CASE STUDY 1: Employment out of Calibration Equipment/Instruments in the
manufacturing process [10]

Details of a Deviation-Out of calibration Equipment was employed in the

manufacturing process. Type- Unplanned Deviation

Immediate actions taken after identifying Deviation

Whether operations were stopped-No. As it is identified after the production of a few
batches. Whether QA personnel was informed-Yes. It was immediately informed to
Quality Assurance Personnel. Whether the affected material/batch numbers (part
quantity or full quantity) was identified and segregated-Yes.

Root cause analysis

Out of Tolerance:

The first thing to be considered is to check the calibration certificate when out of
tolerance occurs. You must understand what went wrong and check the data in the
calibration report and check what the reason for out of tolerance is. You must
analyse the risks. For example, Weighing balance: Let us say that if the weighing
balance reading is not zero before weighing and you will weigh a product then it
causes deviation. If you want to weigh 40 g of sample and the weighing balance is
showing 0.05 g in the starting and not zero then the product you weighed will not be
40 g. It will be 39.95 g. Hence, it causes a deviation.

When did this happen?

Finally, check when was the measurements shown by the instrument accurately and
the due date for calibration and previous calibration date.

Where is the instrument used?

You have to check where the instrument is used, such as production department or
warehouse, etc. check whether the logbook is maintained for the instrument and is it
maintained properly and entered after every use.

How is it used?

The last step is to identify how the out of tolerance instrument was being used.
Determine what all measurements were being made at the location. This information
will likely be found in the operator’s work instructions or end users procedures or an
engineering specification. The objective of this step is to determine that whether the
out of tolerance instrument could have affected any of the products manufactured or
the services that are provided by this instrument, in this time frame, in this location,
for these measurements. This can be achieved by reviewing the documentation
process.

Corrective and preventive action

Corrective action:

Products which are not shipped must be separated instantly for testing. Re-
inspection should be done for the products which are stored in the warehouse. If a
product is already shipped, then it should be recalled. If a customer utilizes the
product as the raw material and has not yet used it for production, then you can test
or inspect in their site itself. If there is no proper equipment for testing then you have
to recall.

Preventive action:

You have to calibrate the instrument regularly at periodic intervals. Calibration data
must contain the following data: Is the calibration within the range? Is it out of
calibration? Can calibration be extended for a longer time? [11]. Whenever it’s
calibrated, is it within the specified limit?

CASE STUDY 2: Facility modification [12]

Details of a deviation

Type: Planned deviation: The secondary packaging material stored in the

warehouse I is shifted to warehouse II due to the flooring modification carried out in
warehouse I. During this period, the following will not be followed in the warehouse I:
Area cleaning will not be performed as per the SOP and will not be recorded. The
temperature monitoring for packaging material storage area will not be done.

Dispensing of packaging material will not be done at the warehouse. The newly
received materials which need to be stored in the warehouse I is shifted to
warehouse II.

Reason/justification for deviation:

To carry out flooring work at the warehouse I. Area cleaning for the warehouse will
be done. Temperature monitoring will be done as per the SOP. Dispensing of the
materials will be done as per the SOP in warehouse II. Transfer of all materials from
warehouse I to warehouse II will be done in a closed container.

Final disposition of deviation by QA department:

This planned deviation was for storing of secondary packaging materials in
warehouse II due to flooring modification in warehouse I. As per this deviation:
Training was given to all the personnel involved to execute the deviation.

Before shifting material, the area was cleaned and the temperature was monitored
and recorded as per SOP. During this period following was not done in the
warehouse I. Area cleaning and temperature monitoring were not done. Impact
assessment: Warehouse II was cleaned and the temperature was maintained before
the transfer of material from warehouse I. Appropriate conditions were maintained.
Materials were shifted and stored with care. After the completion of flooring
modification in warehouse I, the area was cleaned and the materials were shifted
back from warehouse I to warehouse II in a closed container. The planned deviation
is completed.

CASE STUDY 3: Handling out of specifications [13]

Description

During in process reaction monitoring, one of the process impurities was observed in
higher levels against the predetermined specification. The details are given below,
Obtained result: 0.65 %, Limit: Not More Than 0.50 %.

Investigation details

Selection of investigation tool or method for investigation

Based on the nature of the deviation, the following investigation techniques can be
used to identify the root cause. Genchi Genbutsu (Go & See) technique. Brain
storming technique. 5-WHY technique. Six-M Framework (Ishikawa diagram/Fish
bone diagram) technique. Among all the above techniques, based on the nature of
deviation, “Six-M Framework” technique is employed to identify the probable root
cause as depicted in fig. 3.

Machine (Equipment)

Reviewed the preventive maintenance records of all the equipment engaged in the
said batch manufacturing and noted that all the equipment is meeting the
requirement. No abnormalities were noted during preventive maintenance.
Calibration status of the entire instrument like gauges, temperature sensors and
temperature indicator which are affiliated with respective equipment was reviewed
and noted that all the instruments are calibrated and no out of calibrations were
reported. Cleaning status of respective equipment was reviewed and noted that
entire equipment train was cleaned as per respective cleaning records and line
clearance was acquired from Quality Assurance department before batch charging.
Based on the above data, it is understood that the Equipment/Instruments involved
in batch operations were fit and there are no breakdowns reported during batch
execution.

Material (Input materials)

All the input materials used in the said batch were reviewed and noted that all are
meeting the predetermined acceptance criteria and are suitable for usage. Key
starting material details were reviewed and noted that the obtained source was
approved facility and quality of the batch was meeting the specification. Quantity of
all the input material was charged as per the approved batch manufacturing record
(BMR). No abnormalities were identified with respect to input materials quality and
quantity.

Men

All the personnel involved in the batch operations were well aware of the
manufacturing process as per the batch manufacturing record. All are well
experienced and trained on procedures. All the batch operations were carried out as
per the BMR. No abnormalities were noted with respect to operations and testing of
the batch.

Method (Process)

All the process parameters were followed as per the BMR. However, during in
process reaction monitoring step, the impurity content was obtained 0.66 % which is
above the acceptance limit i.e. not more than 0.50 %. As per the BMR, if the initial
sample does not complies, it is recommended to repeat the addition of respective
reagent with the reaction mass at specified temperatures. Hence, addition was done
and 2nd sample was sent to Quality control (QC) to check the impurity level. The
obtained result was 0.65 % which is still above the required limit. Upon further
discussion with Research and development (R&D), after few hours of maintenance
at specified temperature, 3rd sample was sent to QC to check the impurity level and
found to be 0.58 % which is still above the limit of not more than 0.50 %. Hence, the
reaction monitoring was terminated and proceeded for subsequent operations. To
have a better over view on the process operations, comparison was made between
pre and post executed batches.

Based on the comparison data, it is identified that the addition of reagent to the
heterogeneous reaction mass was done in shorter time period when compared to
previous and subsequent batches. As there is no standard instruction in the BMR,
operator completed the reagent addition activity in shorter time period. By nature,
this is an exothermic reaction, due to fast addition there is a sudden shoot up in the
temperature beyond the designed space. As this is a heterogeneous mass, this
sudden shoot up in the temperature affected the rate of reaction and led to in
process non-conformance even after addition of excess reagent. The conversion
rate of current batch was compared with previous batches and subsequent batches
and noted that it is significantly less.

Mother Nature (environment)

No contamination occurs with Raw materials in the process.

Measurement

All the input material (solid & liquid) are charged as per the quantities mentioned in
the BMR. Testing was carried out as per the approved method of analysis. Hence,
there is no abnormalities identified related to measurement.

Root cause

Due to faster addition of reagent to the heterogeneous reaction mass, there is a

sudden shoot up in the temperature beyond the designed space. This sudden shoot
up affected the conversion rate and led to the in process non-conformance.

Corrective and Preventive action

Correction:

Reaction monitoring was terminated and proceeded for subsequent operations.

However, same impurity was monitored during complete analysis of Active
Pharmaceutical Ingredient (API) with the same limit. Quality of the batch was
reviewed and found meeting the predetermined specification.

Corrective action and Preventive action:

To avoid the recurrence in future batches, the batch manufacturing record of
respective API was revised by providing instruction for slower addition of the reagent
to the reaction mass at respective operation within the specified time period.

CASE STUDY 4: Handling temperature/Relative Humidity excursions [14,15]

During the review of Supervisory Control and Data Acquisition (SCADA) report, it
was noted that there is above 24 h excursion of relative humidity (RH) in powder
processing area of manufacturing block clean room area.

Investigation details

Review of data:

Relative humidity was out of limit, where the limit is 60 % and it was found to exceed
the specified limit.

From the data given in Table 1, it is noted that the relative humidity was out of
acceptable limit. However, temperature was well within the acceptance limits.

Temperature (º) RH (%)

Day Time (Limit: NMT 25º) (Limit: NMT 60 %)

Monday 00:00 21.3 65.5

01:00 22.0 62.3

02:00 23.5 66.3

03:00 24.2 67.2

04:00 20.1 66.6

05:00 23 67.3

06:00 22.2 65.1

07:00 21.2 70.3

08:00 22.5 71.5

09:00 22.7 71.2

10:00 22.0 70.5

11:00 23.4 74.2

 Temperature (º) RH (%)
Day Time (Limit: NMT 25º) (Limit: NMT 60 %)

Tuesday 12:00 23.1 78.2

13:00 23.5 74.1

14:00 23.7 75.1

15:00 23.2 73.2

16:00 23.4 73.4

17:00 24.0 71.2

18:00 24.2 78.1

19:00 24.3 77.5

20:00 24.8 75.4

21:00 24.9 72.1

22:00 24.2 76.1

23:00 24.6 73.2

00:00 24.8 70.1

Table 1: Relative Humidity Data

Summary of investigation

Based in the occurred discrepancy, probable causes were identified and analysed.
Details are listed below.

Temperature of chilled water system:

The data of chilled water temperature to Air Handling Unit (AHU) system checked
and found there was no abnormality.

Clean room doors:

Checked the doors open condition and found there were no door gaps and all doors
are in closed condition during that period.

Chilled water strainer choke:

There were no chokes found in the chilled water strainer to AHU system.

Chilled water coil choke:

Cleaned the chilled water coil and restarted the AHU system. There was no change
in the RH reading.

AHU filter choke:

Checked the AHU filter and there was no blockage in the filters.

AHU drain line choke:

Drain line was checked and found that water was not draining properly. So, the line
was replaced immediately.

Working condition of solenoid actuator valve of chilled water and hot water:

Solenoid actuator valve of chilled water and hot water to AHU system was checked
and found the hot water solenoid actuator valve was not operating properly.
Immediately the solenoid actuator valve issue was attended and resolved. The RH
found to be in controlled state after replacing it.

Inference

Based on the above investigation and analysis, it is concluded that the occurred
discrepancy in RH is due to the malfunction of hot water solenoid actuator valve of
AHU system. The issue was immediately attended and ensured that the RH is under
controlled state. Preventive maintenance was also carried out as per the approved
schedule and found satisfactory. However, respective clean rooms were idle during
this said period of time. Hence, there is no impact on the product quality.

Root cause:

The malfunction of hot water solenoid actuator valve of AHU system is the root
cause for inconsistency in the relative humidity.

Corrective and preventive action:

Correction:

The hot water solenoid actuator valve of AHU system was immediately rectified and
ensured that RH is under controlled state (NMT 60 %).
Corrective Action and Preventive Action:

The checking of solenoid actuator valve functioning is already a part of preventive

maintenance activity and was carried out as per the approved schedule. However, to
prevent the recurrence, it is proposed to provide a dedicated chilled water line for
AHU’s Booster pump from the chilled water main line to provide constant flow of
chilled water.

CASE STUDY 5:

At specified operation, reaction mass was stirred at 38º instead of 40º to 45º.

Investigation details

Review of raw material quality and quantity:

Analytical results of the respective input materials were reviewed and found meeting
the predetermined specification. All the raw materials were charged as per the
standard quantity provided in the approved Batch Manufacturing Record.

Review of batch manufacturing record

Process parameters:

All the process parameters of respective batch were reviewed from the executed
batch manufacturing record and noted that all the process operations were executed
as per the BMR. However, at specified operation, the reaction mass was stirred at
38º instead of 40° to 45º. At specified operation, the process solvent used was
‘Dichloromethane’ and the boiling point of dichloromethane was around 39°. Hence,
the reaction mass was maintained below 39°.

In process controls:

In process controls are reviewed and found all are meeting the pre-determined in
process specification as per the BMR. No abnormalities were identified during
reaction monitoring.

Personnel Evaluation:

Concerned personnel involved in the batch manufacturing were inquired with respect
to the temperature discrepancy and found no operational deviations occurred.
Training records of respective personnel involved for the execution of the batch were
reviewed and observed that they are enough experienced and trained on
manufacturing process of specific API stage.

Summary of investigation

Based on the above review it is noted that the reaction was maintained below 39°
due to the boiling point of the process solvent i.e. Dichloromethane. However, further
operations were reviewed and no abnormalities noted and all the in process samples
met the specification without any abnormalities.

Root Cause

Due to the boiling point of process solvent i.e. Dichloromethane. Reaction mass was
stirred at 38º instead of 40º to 45º.

Impact on other batches

The output and quality of subject batch were reviewed and found to be well within
the limits. No abnormalities identified with respect to other process operations.
Hence, there is no impact on the subject batch quality due to said discrepancy.
However, there were no further manufacturing of said API stage. Hence, there is no
impact identified on other batches.

Details of Corrective & Preventive action

Correction:

Reaction mass was maintained at 38º instead of 40º to 45º based on the boiling
point of the process solvent.

Corrective and preventive action:

As the occurred deviation is due to process solvent boiling point, it is proposed to

revise the batch manufacturing record of said API stage by modifying the
temperature at specified operation as 35° to 40° instead of 40º to 45º.

Conclusion
Deviation management plays a key role in maintaining product quality and leading to
continuous improvement among the essential elements of a well-established Quality
Management System (QMS). A thorough knowledge and experience about handling
of deviations, analysing the deviations and implement Corrective and Preventive
actions plays a very dynamic role in better and successful functioning of a
pharmaceutical industry. Proper analysis of a deviation would provide the
pharmaceutical organization, methods and procedures to be established and
implemented to reduce the errors and deviations in future, providing better
assurance and benefits to the organization.

References