Weighing Data Integrity

White Paper
3 Ways to Ensure Compliance

Ensuring the reliability and integrity of data generated across the entire pharma produc-
tion chain is fundamental to regulatory compliance around the world. Also weighing pro-
cesses – particularly quality critical ones – need to follow those principles. Moving from
paper-based systems to hybrid paper-electronic systems or fully automated, networked
data-capture can help reduce documentation errors and optimize processes.

Weighing – though sometimes oversimplified or ne-

glected – is an important process step in pharmaceuti-
cal manufacturing. Dispensing raw materials, dosing
material into a tablet coater, checking granulation or
tableting all require weighing and documentation for
complete, compliant batch records. As such, Good
Documentation Principles apply to these activities as
stated in relevant Good Manufacturing Practice (GMP)
guidelines.

Automating all or a part of this data gathering with

electronic solutions can help make sure documenta-
tion is accurate, traceable and stored safely. In this pa-
per, we will compare manual transcription to docu-
mentation with three automated or partially automated
solutions, assessing the pros and cons of each.
Table of Contents
Regulations and Definitions
Weighing Data Management
Option 1: Adding a Printer
Option 2: Automated Data Transfer
Option 3: Compliant Networked Solutions
Summary
 Regulations & Definitions

The reliability and integrity of pharmaceutical data is sues have risen significantly, especially outside the US
fundamental to regulatory compliance and patient (Figure 1, Reference 1). A similar picture is true for oth-
Weighing Data Integrity

safety. Having confidence in generated data and being er regulatory authorities.

able to reconstruct production activities are key. All
Number of warning letters citing data integrity
major regulatory agencies are currently focusing their
activities in terms of inspections and publications on 15
15

the topic of data integrity. Under this close scrutiny,

many pharmaceutical companies worldwide have 10
10
been warned by inspectors because they allegedly fal-
sified or altered data, or because they failed to keep
5
5
accurate information trails about how they produced or
tested their drugs.
0
0
2013 2014 2015
When analyzing warning letters issued by the FDA be-
Figure 1: Warning letters that cite data integrity violations in
tween 2013 and 2015, letters citing data integrity is- plants in the US ■ and outside US ■ .

Principles of Data Integrity

Recent guidelines on this topic use similar terms and definitions. What are the most important aspects to con-
sider, and how are common technical terms defined?

• Data integrity – Data is complete, consistent and accurate in all paper and electronic forms.
• ALCOA principle – Data should be attributable to the person generating it, legible and permanent, contemporaneous,
the original record or a true copy and accurate. Recently ALCOA+ has been introduced that adds the following attributes:
complete, consistent, enduring and available.
• Data criticality – Defining how important data is to quality, safety and efficacy decisions, determined by considering
the type of decisions influenced by the data.
• Data risk – Reflecting the vulnerability of particular data to unauthorized deletion or amendment and the opportunity to
detect such alterations during a routine review.
• Raw data – Original records retained in the format in which they were generated (paper or electronic) or a true copy.
Raw data must permit the full reconstruction of activities that generated the data. In the case of basic electronic equip-
ment which does not store electronic data or only provides printed output, the printout constitutes the raw data.
• True copy – Copy of an original recording of data that has been verified and certified to confirm it is an exact and com-
plete copy that preserves the entire content and meaning of the original record.
• Metadata – Data that describes the attributes of other data and provides context and meaning. These are data that de-
scribe structure, data elements, interrelationships and other data characteristics. It also permits data to be attributed to
an individual or to the original data source. Metadata forms an integral part of the original record.
• Audit trails – Metadata that are a record of critical information and that permit the reconstruction of activities. Comput-
erized systems should always provide for the retention of audit trails to show all changes to the data while retaining
previous and original data. It should be possible to associate all changes to data with the person making the changes.
• System user access – The use of access controls to ensure that people have access only to functionality that is ap-
propriate for their job role and actions taken are attributable to a specific individual.
References 2, 3, 4

White Paper Weighing Data Integrity

2 METTLER TOLEDO © 11/2016 Mettler-Toledo GmbH
 Weighing Data Management and Solutions

Manual recordkeeping and transcription is still a popu-

Frequently observed data integrity issues in phar-
lar way of documenting weighing data. It is also one of
Weighing Data Integrity

maceutical manufacturing:
the major sources of error and other data integrity vio-
• Not recording activities contemporaneously
lations. Mixing up numbers, leaving out important in-
• Data backdating
formation or not recording values in the proper forms
• Data fabrication (falsification)
are examples of trouble spots. In severe cases, data is
• Copying existing data as new data
backdated or even fabricated. Not keeping raw data
• Re-running samples
and not documenting routine testing or calibration data
• Discarding data
are other examples.
• Not being able to produce raw data

Lack of training and fatigue can cause errors such as

these to occur more frequently in cases where docu-
mentation is manual. Manual recordkeeping is also
time consuming and keeps operators from performing
other value-added activities. If the 4-eye-principle ap-
plies, additional personnel must check to see if data
has been captured accurately, adding to the time re-
quired and potentially affecting productivity.

Data Management:
A typical weighing process

A typical weighing process requires the following data-

management activities (workflow, Figure 2):
• Creating and transferring the recipe or procedure to
be followed
• Carrying out the actual weighment
• Transferring the results to the batch record
• Transferring this data to a MES system for analysis
and storage

In the pages that follow, we will compare how these

actions are handled manually against three potential
solutions for improving the accuracy and ease of data compliance to help pharmaceutical manufacturers
transfer during typical weighing processes. Specifical- avoid warnings and citations while maintaining the
ly, we will look to see where gains can be made in productivity required in today’s competitive pharma-
terms of data accuracy, transfer speed and regulatory ceutical marketplace.

Further Analysis
Recipe Data Data Capture
Weighing Storage
SOP Transfer/Import & Transfer
Reporting

Figure 2: Weighing process – data acquisition, generation and transfer.

White Paper Weighing Data Integrity

3 METTLER TOLEDO © 11/2016 Mettler-Toledo GmbH
 Option 1: Adding a Printer

A basic way to reduce transcription errors is to use Data integrity Checklist

the print capabilities of your balance or scale. Adding
Weighing Data Integrity

Traceability ensured (✔)

a printer is a simple way to document and store data.
Low risk of transcription errors (✔)
An example data flow (Figure 3) at a weighing station
Efficiency of process (✔)
used to capture tablet weights after coating might look
like this: Transfer of metadata (✔)
1. Results generated by the scale are printed on pa-
Audit trail (system & data) ✖
per. Some metadata maintained in the instrument
can be added. Centralized data storage ✖
2. Printed records are then used to manually tran-
Paperless production ✖
scribe figures into reports.
3. Reports are checked for errors by a second person Electronic signatures ✖
following the “4 eyes” principle.
FDA 21 CFR Part 11 / ✖
4. Data may also be manually transcribed into an EU Annex 11 compliant
ERP system.

Container Scale Batch Record Report

& Products Form

Print-out Approval

ERP

Figure 3: Data flow with weighing results documentation and storage on a printout eliminates one source of transcription error. Additional
steps are still manual.

Manual Transfer Automatic Transfer

Printouts
A typical printout contains material name, batch num- Batch No.: 0000000
ber, date and time, weight value/unit, and operator Product No.: 0000000
name or ID. This data is considered raw and needs to Product name: xxx
Lot No.: 000000000
be maintained and archived. A label creator e.g. Net: 000 kg
DatabICS software available for our ICS terminals al- Tare: 0.00 kg
Gross: 00.00 kg
lows the generation of custom labels that meet regula- Operator: xxx
tory requirements that can be applied to containers or Date: 00.00.00
placed in records. Time: 00.00.00

White Paper Weighing Data Integrity

4 METTLER TOLEDO © 11/2016 Mettler-Toledo GmbH
 Printing eliminates the error risk inherent when we read and note results off a terminal screen.
However, this is still a manually intensive process that requires a significant amount of time
and attention.

Pros: Cons:
Weighing Data Integrity

Direct printing is a simple way to document and store Information is not ready for digital processing and an-
measurement data without transcription errors. It is alyzing. Manual transcription is still necessary. Print-
cost-effective and easy to implement in existing pro- outs also have the tendency to fade over time and can
cesses. be lost. For critical data, the “four eyes” principle ap-
plies, which requires additional resources.

Additional features for enhanced data integrity

Terminal user-management functionalities allow the identification of the
person carrying out the weighing task and limits access rights to critical
functions as required and checked during data-integrity inspections.

Digital material identification using barcode scanners improves automat-

ed identification of containers or raw materials. A RS232 or USB-enabled
barcode reader can be connected to weighing equipment for easy data
capture. Material ID data can be printed on strips or labels. This allows
materials identification and transfer to the scale to recall target weights
or user instructions.

White Paper Weighing Data Integrity

5 METTLER TOLEDO © 11/2016 Mettler-Toledo GmbH
 Option 2: Automated Data Transfer

To eliminate manual steps and related transcription er- Data integrity Checklist
rors, a scale or balance can be directly connected to a
Weighing Data Integrity

Traceability ensured (✔)

computer allowing unidirectional data collection using
a PC program. Data is submitted as a text protocol. Low risk of transcription errors (✔)

Efficiency of process ✔
An example data flow at a weighing station used to
capture container weight before and after granulation Transfer of metadata ✔
might look like this:
1. Results are generated by the scale, then trans- Audit trail (system & data) ✖
ferredto a computer with the help of data collection Centralized data storage ✖
software.
2. This data is then transferred to Excel or printed. Paperless production ✖
3. Reports are double-checked by a second person Electronic signatures ✖
following the “4 eyes” principle.
FDA 21 CFR Part 11 /
4. After that, data may be manually transcribed into EU Annex 11 compliant
✖
an ERP system.

Container Scale Comuter Report

& Products

Approval

ERP
Bar code reader Printer
Figure 4: Direct connection or transfer via USB stick. Transcription errors are eliminated and data quality is improved. Additional measures
are needed to ensure compliance.

Manual Transfer Automatic Transfer ----  Alternativ data transfer

Most common interfaces for data transfer to computer:

Depending on requirements such as data rate and volume, network type, environment/distance from data
source or network structure, different scale interfaces are available.

Serial interfaces such as RS232/422/485 Ethernet TCP/IP via cable or WLAN are the Wireless interfaces such as Bluetooth or
or USB allow point-to-point communica- most common interface type for PC net- WLAN allow communication with mobile
tion between printers and barcode readers works. scales or in locations where cabling is
or PCs. complicated.

White Paper Weighing Data Integrity

6 METTLER TOLEDO © 11/2016 Mettler-Toledo GmbH
 Data Communications
Data communication software such as DatabICS or
Collect+ allow simple data transfer including some
metadata such as value and unit, time and date, ma-
terial ID, net/gross weight and tare. Layouts are config-
Weighing Data Integrity

urable and input masks and calculations are possible.

Further measures such as printing reports, checking data accuracy and obtaining signatures
need to be implemented via SOPs. System validation is needed if critical data is processed.

Pros: Cons:
Direct transfer from the scale to a computer eliminates There is no centralized data storage. Not all metadata
manual transcription. Adding simple communication is captured and values can still be changed or omitted
software allows transfer of the value along with part of when reporting without traceability. This lack of an au-
the metadata. Data transfer is enabled via USB or in- dit trail potentially reduces traceability, making it im-
terfaces such as serial or Ethernet. possible to find out what data was changed, when
and by whom.

Alibi data storage

Many weighing terminals contain a protected internal memory. Original-
ly required by OIML weights and measures regulations, data retained
consists of the device serial number, date & time of measurement, net
weight and tare. METTLER TOLEDO adds additional data points for ex-
panded data sets. Data cannot be altered in the scale itself. It is provid-
ed in CSV format which can be used for further analysis. Data can be
retrieved via Ethernet, USB or WLAN depending on scale type.

White Paper Weighing Data Integrity

7 METTLER TOLEDO © 11/2016 Mettler-Toledo GmbH
 Option 3: Compliant Networked Solutions

Recipe-weighing software or quality-control soft- Data integrity Checklist

ware offer strong workflow management as well as
Weighing Data Integrity

data capture and reporting. These software types Traceability ensured ✔

provide several significant benefits to data accuracy Low risk of transcription errors ✔
and integrity over simple data-collection software.
Efficiency of process ✔
For example, at a formulation station where multiple de- Transfer of metadata ✔
vices are connected and multiple raw materials and rec-
ipes are handled, data is highly critical for product qual- Audit trail (system & data) ✔
ity. Similarly when looking at quality control processes
Centralized data storage ✔
to monitor fill levels or tablet uniformity, weighing or oth-
er measurement steps take place at different worksta- Paperless production ✔
tions, calculations need to be carried out and data ana-
Electronic signatures ✔
lyzed in real time to react quickly to product deviations.
FDA 21 CFR Part 11 /
✔
EU Annex 11 compliant
In both cases compliant manual documentation pro-
cesses become resource-intensive. Software can sim-
plify and speed up these processes, capturing savings
that offset initial implementation costs.

• Full traceability: instrument, date, time, method, • All data is available for immediate reporting or di-
raw material information and calibration history is rect analysis and can be approved by a supervisor.
recorded. Robust systems include enhanced user- • All data is stored securely in the database and can
rights management. be transferred to the ERP system.
• Guided SOPs and automatic data capture reduces
waste and rework.
• A complete log and audit: not only is data record-
ed, but any interaction with the system is logged
into the audit trail, telling an auditor or reviewer
“who did what, when”.

Container Multiple Scales or Devices Network Server

& Products

Approval
Report
Storage

ERP

Figure 5: Data flow in digital production with automated transfer of data from the scale to the network server.

Manual Transfer Automatic Transfer

White Paper Weighing Data Integrity

8 METTLER TOLEDO © 11/2016 Mettler-Toledo GmbH
 When considering true data integrity in regulated environments, full integration of weighing
equipment into compliant software is highly desirable.

Pros: Cons:
Data is automatically transferred to eliminate transcrip- Complete solutions such as FormWeigh.Net or Free-
Weighing Data Integrity

tion errors and increase productivity. Processes and Weigh.Net cost more than paper records or direct-to-
calculations are guided to ensure each user carries PC solutions at the outset. However, initial expense is
out the same procedures, helping to eliminate rework usually made up quickly by error reduction, time sav-
and waste. Metadata is linked to results, supporting ings and traceability improvements.
full traceability. A variety of import or export options
allow data transfer into existing networks, data ar-
chives or ERP systems.

Figure 6: Statistical quality control software such as FreeWeigh.Net integrates data capture and analysis for many devices including check-
weighers and analytical instruments.

Audit trail functionality

The audit trail in FreeWeigh.Net or FormWeigh.Net records all changes
to data records, indicating the change itself along with user informa-
tion and date and time in full compliance with 21 CFR Part 11. The
audit trail thus ensures that all changes made to the system can be
traced. Once activated, the audit trail cannot be switched off, elimi-
nating the potential for audit trail tampering or loss.

White Paper Weighing Data Integrity

9 METTLER TOLEDO © 11/2016 Mettler-Toledo GmbH
 Summary

Humans are capable of error, even when stakes are certain level of security, error mitigation and time sav-
high. At the same time, pressure on production costs ings. Whether you choose to add a printer, connect to
Weighing Data Integrity

means resources need to be spent wisely. While paper a higher-level MES system or provide full integration
logs have historically been the norm and tend to be with a workflow and data management system such
low-cost, automation of results capture and storage as FormWeigh.Net or FreeWeigh.Net, capturing data at
can help reduce error risk, enhance productivity and the point of origin is the foundation of any data-integri-
reduce waste or rework for more accurate, compliant ty improvement effort. This timely automated capture
processes. avoids transcription errors, captures relevant metadata
and frees up operators for other value-added tasks.
There are three levels of non-manual data capture that
production units can take advantage of. Each offers a

Weighing Data Transfer: Solutions Snapshot

Traditional: Option 1: Option 2: Option 3:
Manual data Adding a Automated data Compliant net-
recording printer transfer worked solution
Gateway Eyes Printer, Computer Network, software
Pen Bar code USB stick such as FormWeigh.
reader Net, FreeWeigh.Net

Traceability ensured (✔) (✔) (✔) ✔

Low risk of transcription errors ✖ (✔) (✔) ✔

Efficiency of weighing process ✖ (✔) ✔ ✔

Automated transfer of metadata ✖ (✔) ✔ ✔

Audit trail (system & data) ✖ ✖ ✖ ✔

Centralized data storage ✖ ✖ ✖ ✔

Paperless production ✖ ✖ ✖ ✔

Electronic signatures ✖ ✖ ✖ ✔

FDA 21 CFR Part 11 / EU Annex 11 compliant ✖ ✖ ✖ ✔

✔: full support; (✔): only partially possible or only reachable with additional measures

White Paper Weighing Data Integrity

10 METTLER TOLEDO © 11/2016 Mettler-Toledo GmbH
 References

1. FDA Warning Letter Database

http://www.fda.gov/ICECI/EnforcementActions/WarningLetters/default.htm
Weighing Data Integrity

2. MHRA GMP Data Integrity Definitions and Guidance for Industry, March 2015
https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/412735/Data_integrity_definitions_
and_guidance_v2.pdf

3. WHO Good Data & Records Management Practice, June 2016

http://www.who.int/medicines/publications/pharmprep/WHO_TRS_996_annex05.pdf

4. FDA Draft Guidance “Data Integrity and Compliance with cGMP”, April 2016
http://www.fda.gov/downloads/drugs/guidancecomplianceregulatoryinformation/guidances/ucm495891.pdf

METTLER TOLEDO Product Information Further Reading

Industrial Weighing Regulatory Compliance Whitepaper
www.mt.com/industrial www.mt.com/ind-pharma-regulations
Industrial Terminals Data Integration Whitepaper
www.mt.com/terminals www.mt.com/ind-data-integration
Printers: Ideal Dispensing Station Application Note
www.mt.com/printers www.mt.com/ind-dispensing-station
ICS Scales
www.mt.com/ics689
DatabICS
www.mt.com/ind-databics
Collect+
www.mt.com/collectplus
FormWeigh.NET
www.mt.com/formweigh
FreeWeigh.Net
www.mt.com/freeweigh

Mettler-Toledo GmbH
Industrial Division
www.mt.com
For more information
CH-8606 Nänikon, Switzerland
Phone + 41 44 944 22 11

Local contact: www.mt.com/contacts

Subject to technical changes

© 11/2016 Mettler-Toledo GmbH
MTSI 30321357 / Marcom Industrial