European Pack Coding Guidelines

European Medicines Verification

System (EMVS)

Version 4.0

July 2017

European Pack Coding Guideline Page 2 of 11

Version 4

Revision History

V1.0 June 2008 Initial Release

V1.1 June 2008 Amended some typographical errors, added extra detail to the serial number

constraints and added information regarding the randomisation requirements.

V1.2 Aug 2008 Removed the background and abstract sections. Removed the SSCC section

(potentially applicable later in the project) and amended the serial number

constraints. Added section to define the human readable text requirements.

V1.3 Sept 2008 Removed a small amount of duplication. Cross referenced the code size

section to the GS1 specifications. Clarified the human readable text size

requirements. Corrected a small number of typos. Awaiting incorporation of

further clarification on the randomisation.

V1.4 Dec 2008 Provided information regarding the serial number randomisation criteria.

V1.5 Dec 2008 Amended the serial number criteria. Amended the human readable text

specification and scope. Amended the pack code size criteria.

V1.6 Jan 2009 Removed obsolete text from section 1. Removed paragraph from Product

code section referencing the use of the package level indicator initial setting.

Included some suggested clarifications from GS1. Added definition for

alphanumeric character range, comments about permissible code quality

standards. Amended randomisation section. Removed the requirement for

the batch/lot code to be placed at the end of the data set (due to the use of

ASCII encoding)

V1.7 Jan 2009 Amended text in section 1 regarding the minimum human readable content,

randomisation section definition of data retention period. Added clarification

regarding the human readable text requirements. Replaced the AI references

for Expiry and Lot mistakenly deleted from V1.6. Defined the use of, and

value of, the package level indicator digit in the GTIN/pseudo GTIN. Added

the reference to the GS character within the lot code definition. Amended

both GS character usage statements to ensure that GS is not appended to

the end of the code after the final field. Removed picture from the front page.

V2.0 Feb 2009 First official general release.

V2.1 May 2009 Added footnote to define a Pseudo GTIN to aide clarity.

V3.0 April 2011 Added section to define the additional element for national reimbursement

number inclusion

V3.1 Jan 2013 Added a) new GS1 NTIN rules; b) reference to other schemas such as PPN;

c) suggestions regarding placement of codes

V3.2 Jan 2013 Internal review

V3.3 Feb 2013 Added example of SN, minor changes throughout

V3.4 March 2013 Minor changes

V3.5 April 2013 Minor changes, WS-3 Approved

V3.6 June 2013 Wording on randomisation requirements in sec. 2.2 updated to provide

additional clarity

V3.7 August 2015 Wording revised to remove all but the technical detail. Randomisation

wording clarified. Serial number character set recommendation revised the

text. Revised the code quality level to be in-line with expected Delegated

Regulation requirements.

V3.8 April 2016 Amended serial/batch number character set slightly, added multi-market

section.

V3.9 March 2017 Added amends from Medicines for Europe, corrected and update the NHRN

content, de-branded from EFPIA to become a joint industry publication.

V3.91 June 2017 Added amends from EFPIA (Bayer)

V4.0 July 2017 Added further amends to section 2.3 and 2.9 from MfE and released.

European Pack Coding Guideline Page 3 of 11

Version 4

Table of Contents

Revision History .................................................................................................................. 1

1. The European Medicinal Pack Coding Specification ................................................ 4

1.1 Recommendation of GS1 Standard / 2D Data Matrix ............................................. 4

1.2 Accommodation of National Numbers .................................................................... 4

2. Content of the 2D Code ............................................................................................... 5

2.1 The Product Code .................................................................................................. 5

2.2 The Serial Number ................................................................................................. 6

2.3 The Expiry Date ..................................................................................................... 8

2.4 The Lot Number (Batch Code) ............................................................................... 8

2.5 The National Healthcare Reimbursement Number (NHRN) ................................... 8

3. Pack Code Attributes .................................................................................................. 9

3.1 Multi-Market Packs ................................................................................................ 9

3.2 Pack Code Size ................................................................................................... 10

3.3 Pack Code Quality ............................................................................................... 10

3.4 Human Readable Representation ........................................................................ 11

3.5 Code Placement .................................................................................................. 11

European Pack Coding Guideline Page 4 of 11

Version 4

Industry Coding and Serialisation Team

1. The European Medicinal Pack Coding Specification

1.1 Recommendation of GS1 Standard / 2D Data Matrix

In February 2006, the EFPIA board recommended the adoption of a unique standard for the

coding of pharmaceutical product across Europe based on the Data Matrix ECC-200 to be

introduced on all secondary packaging of prescription products sold in Europe.

The pack/item code will be accompanied by human readable text. The human readable text

will be in a font and size that are in accordance with country

specific requirements or GS1 recommendations depending on

local requirement.

As new members joined the EMVO alliance, EAEPC and

Medicines for Europe have helped to make this document a

common industry guideline.

1.2 Accommodation of National Numbers

National numbers may be required within the code in specific situations. The GS1 General

Specifications have been extended to accommodate the inclusion of national product

identifiers

. This capability is provided by the use of NHRN (National Health Reimbursement

Numbers) which can be added to the data content of the code using the appropriate

Application Identifiers (AI’s). This is described below in section 2.5.

Alternatively, in some countries, GS1 has allocated a range of GTIN’s (Global Trade Item

Numbers) specifically for use with that country. These are collectively referred to as NTIN’s

(National Trade Item Numbers).

See “EFPIA & GS1: a shared vision for product identification in the context of the EU Directive on

Falsified Medicines”, issued January 2012

Data Matrix ECC-200

European Pack Coding Guideline Page 5 of 11

Version 4

2. Content of the 2D Code

The recommendation for coding of pharmaceutical products is to encode in a Data Matrix

code a minimum of four items of data. The preferred code structure would utilise the GS1

standards and in particular would make use of the AI’s already in use throughout the supply

chain today. The preferred structure uses only four of the many AIs available to define

1. the product code,

2. the serial number,

3. the expiry date and

4. the lot number (batch code).

Some markets may require the addition of a national number to the code. This is described

in section 2.5.

2.1 The Product Code

The preferred (by manufacturers) implementation for the product Code is to use the GS1

GTIN (Global Trade Item Number) however the European Hub and thus EMVS supports

both GTIN/NTIN and PPN coding schemes.

The four critical aspects for the GTIN element are:

1) It must be preceded by the application identifier 01.

2) It must have a code length of 14 digits

3) It must use a GS1 standard checksum in the 14

digit

4) It must adhere to the GS1 use for the first digit:

a. For a sales unit a 0 is required.

b. For identification of higher package levels the range is 1-8. (the value 9 has a

special defined meaning – ref GS1 specifications.)

Where the GTIN is held by the organisation as a GTIN-13, to use this value in the 2D code it

must first be converted to a GTIN-14. This can be achieved by simply prefixing the GTIN-13

value with a zero ‘0’. This addition does not affect the value of the check-digit, any other

value of prefix would require the re-calculation of the check digit.

Note: Special Consideration NTIN (National Trade Item Number):

The GTIN number pool allows for the inclusion of national coding schemes (for example the

CIP-13 code in France, the post 2009 PZN in Austria, the Nordic Vnr or the extended

German PZN.) These codes are generally referred to as NTINs (National Trade Item

Numbers). Under special agreements with GS1, NTINs do not start with the company prefix

but with a GS1-approved country-specific prefix followed by the national license or

registration code, i.e. NTIN codes are not determined by the MAH but instead assigned by a

Number Issuer, Industry body etc.

The NTIN differs to the national identifier NHRN described below in section 2.5, The

National Healthcare Reimbursement Number (NHRN), in that NTINs are taken from the

GTIN number pool and their global uniqueness is therefore guaranteed. They formally

present as a GTIN and can generally be used as a GTIN

There are only some areas where the use of an NTIN instead of a pure GS1-GTIN requires special

consideration: Similarly, deriving GTINs for higher pack levels using the indicator digit is problematic: In the case

of a GTIN, this can be derived by the pharma company who owns the company prefix but in the case of NTINs,

companies do NOT own the prefix and would therefore be modifying a GTIN/NTIN that is not theirs to modify.

European Pack Coding Guideline Page 6 of 11

Version 4

Examples of NTINs in use in Europe and formation rules:

Market

NTIN formation rules

Austria

908888 + PZN + check digit

France

3400 + CIP/ACL Code + check digit

Germany

4150 + 8-digit PZN + check digit

Spain

847000 + Codigo Nacional

Sweden, Finland, Denmark, Iceland, Norway

704626 + Nordic Drug Code issued by Nordic

Number office + check digit

Switzerland

7680 + Code assigned by Swissmedic (consists

of 5 digits Product License number + 3 digits

Pack Size indicator) + check digit

2.2 The Serial Number

The serial number is preceded by the AI 21 and adheres to the GS1 specification where this

field is a variable length (up to 20) alphanumeric field followed by a Group Separator (GS)

character (to delimit it from the next field unless it is the last field).

It is recommended that three further modifications should be considered to aide consumer

readability:

 The alphanumeric range shall include the digits 0-9 and the letters of the western

alphabet but exclusion of the following letters: i, j, l, o, q and u. (I J L O Q U) might

help avoid confusion with similarly shaped characters/numerics.

 The serial number character string should only contain either lower case or upper

case letters, not a mixture.

 Use of the extended symbols, as defined by the complete GS1 specification and

documented below, should ideally be avoided.

The serial number will be unique per product code (i.e. not per batch or per product code-

batch code pair).

If these recommendations are fully adopted, this gives a range of 30 different alphanumeric

character options, 50 combinations if only the letter confusion aspects is adopted. 62

combinations if the case limitation is ignored and 82 combinations if the full GS1

specification is utilised. All of these options provide an essentially limitless provision of serial

numbers per product SKU

Using the GS1 specification (which is the industry norm and also represents the most

mature standard available in the supply chain today), the permissible characters include the

digit value 0 to 9 inclusive. Letters A to Z inclusive in both upper and lower case and the

non-numeric, non-letter based extended characters:

/.,-+*)(‘&%”!:;<?=>_

Thus the GS1 scheme permits 82 character combinations. The full GS1 supported character

set is documented in the GS1 Standard Specifications section 7.11 figure 7/11-1.

http://www.gs1.org/sites/default/files/docs/barcodes/GS1_General_Specifications.pdf

European Pack Coding Guideline Page 7 of 11

Version 4

2.2.1

Randomisation

In order to provide a reasonable level of complexity within the serial number, the probability

that a valid serial number can be guessed should be less than 1 in 10,000 (i.e. < 0.0001).

The rationale being that a potential counterfeiter would have to make and distribute 10,000

uniquely coded packs and only have one of those packs get through the system as valid.

Also, in order to minimise the opportunity for a counterfeiter to estimate the randomisation

pattern from two or more samples, the following randomisation rules or equivalent apply:

Given a sufficiently large set of (randomised) serial numbers for a product, the

randomisation substrings of the serial numbers

have to fulfil the following randomisation

criteria:

1. The randomisation substrings must be equally distributed. e.g. the serial number

substring should not contain fixed blocks of fixed digits.

2. Any randomisation substring must be independent of other substrings.

3. The randomisation substrings must not be built using an algorithm that is easy to find

out when knowing the given set of serials or a subset thereof.

Serial IDs shall not be reused within the longer of a) Exp Date +1 year or b) five years.

Examples that do not fulfil the randomisation rules:

Sequence

Sorted ascending

Comment

1,090,580

2,076,648

351,029

2,323,165

1,337,097

597,546

3,062,716

3,309,233

2,816,199

1,583,614

844,063

104,512

2,569,682

1,830,131

104,512

351,029

597,546

844,063

1,090,580

1,337,097

1,583,614

1,830,131

2,076,648

2,323,165

2,569,682

2,816,199

3,062,716

3,309,233

Not random: generated

using incremental value of

246,517

023010724

423000710

723020723

623030711

723000779

823030764

523090785

023050702

623050782

023010724

023050702

423000710

523090785

623030711

623050782

723000779

723020723

823030764

Poor randomness: fixed

substrings „230“ and „07“

Manufacturers might want to compose the serial numbers e.g. of a header substring with fixed contents (e.g.

determining the production site) followed by a substring with variable contents containing the randomisation part

of the serial id.

European Pack Coding Guideline Page 8 of 11

Version 4

2.3 The Expiry Date

This field will conform to GS1 standards and is preceded with the AI 17. It is a fixed 6-digit

field with the digits representing YYMMDD. Organisations more used to a four-digit year

code will be required to truncate the code to the two least significant digits. This field being

of fixed length does not require to be followed by a GS character and in the interests of

minimum printed code size should not be followed by a GS. It is required to enter ‘00’ for the

day when the day code is not specified/used and only a year and month identifier is used –

e.g. YYMM00 (ref GS1 specifications section 3.6.11) and setting the day code to 00 implies

the last day of the specified month.

Examples:

“171215” = 15 Dec 2017

“171200” = Dec 2017.

2.4 The Lot Number (Batch Code)

The lot number (batch code) conforms to GS1 standards and is preceded with the AI of 10

and is a variable length field of up to 20 alphanumeric digits followed by a Group Separator

(GS) character (to delimit it from the next field unless it is the last field). It is recommended

that the range of characters defined for the GS1 specification ( see section 2.2) is permitted

within the batch code as is already common industry practice, however where possible we

recommend to reduce the characters used according to the same guidelines described

above in section 2.2 for the serial numbers.

2.5 The National Healthcare Reimbursement Number (NHRN)

The updated GS1 General Specifications now makes provision for national numbers to be

accommodated within the GS1 standards. When a national number is held in this format in

accordance with GS1 standards, it will be called an NHRN (National Healthcare

Reimbursement Number).

It is not the intention for the NHRN to replace the GTIN, rather to allow national numbers to

be held and stored within systems and data carriers in a GS1 compliant way. This will allow,

for example, a GTIN and national number to be held in the same Data Matrix code so that

both can be captured with a single scan. In this way any stakeholder wishing to obtain the

national number for a product will be able to derive it based on the NHRN field.

The following Application Identifiers have so far been defined

(note that the NHRN part is

defined and controlled by the associated regulatory body or assigning organisation indicated

below):

Application Identifier

National Healthcare

Reimbursement Number

Organisation

710

variable length X

Germany (IFA)

711

712

713

variable length X

Brazil (Anvisa)

Additional NHRN AI’s can be requested through the GS1 GSMP.

The NHRN is usually assigned by a national authority to healthcare brand owners for

specific trade items and shall only be used for compliance to regulatory requirements where

the GTIN alone in a bar code symbol will not meet the requirements. Use of NHRN on the

item is controlled by and subject to the rules and regulations of national/regional agencies.

ref GS1 General Specifications Release 17.0.1 Jan 2017 Figure 3.8.17-2

European Pack Coding Guideline Page 9 of 11

Version 4

When a regionally specific NHRN AI is approved (as above), the overall variable length (i.e.

allowable number of characters) is specified by the national authority, with a twenty (20)

alpha-numeric characters maximum as noted in the general format above.

Where the NHRN is required, the data set will need to include more than four elements;

1. The GTIN (which will be provided by the manufacturing organisation)

2. The Expiry Date

3. The Lot Number

4. The Serial ID and

5. The NHRN.

In fact, it is possible to associate more than one NHRN with a given GTIN, dependent upon

regional and market needs, though in practical terms (size of code and associated issues

around printing and placement of larger Data Matrix codes) manufacturers will probably try

to avoid this.

3. Pack Code Attributes

3.1 Multi-Market Packs

Multi-market packs are those that can be supplied by the original manufacturer (or re-

packer) in an unaltered form to more than a single market. The simplest way to achieve a

common pack is where each market of intended sale is able to use the product code as the

identifier for the product. This is where a globally accepted coding scheme such as the GS1

GTIN provides great benefit. In this ideal scenario, each pack will be provided with a Data

Matrix code containing the four basic data elements only. In addition, the EMVS system

allows the product code (GTIN) and local reimbursement number to be linked at database

level further enhancing the advantages of a four element, GTIN only based code.

GS1 have published a very detailed guidance document covering multi-market packs

entitled “Recommendations on a harmonised implementation of the EU Falsified Medicines

Directive using GS1 standards”

There will be circumstances where local market conditions require that the local

reimbursement number is included within the code.

As an example, a two market pack is supplied where one market is able to utilise the GTIN

directly and the other requires that a local reimbursement number is provided within the

code. If we take the market requiring the local reimbursement number in this example as

Germany, the Data Matrix code might contain (using the GS1 scheme):

Data Element Description

Application Identifier

Example Data

Product Code

05060141900015

Expiry Date

190200

Batch Number/Lot Code

ABC123992

Serial Number

28574abczz3456

NHRN

710

45678912

The data encoded into the Data Matrix code for this example would be:

01050601419000151719020010ABC123992<GS>2128574abczz3456<GS>71045678912

The local reimbursement number, PZN in this example, is shown as the last element in the

code following the AI (Application Identifier) 710.

European Pack Coding Guideline Page 10 of 11

Version 4

The <GS> is a symbolic representation for the ASCII Group Separator character (ASCII

value 29).

3.2 Pack Code Size

The use of either rectangular codes or square codes is left to manufacturer preference and

the limitations of the printing technology available

The GS1 numbering specification for very small

healthcare items specifies the dimensions

shown in the table adjacent. It is recommended

that these targets are also employed on larger

healthcare items such as folding boxes and

bottles.

Note: Small module sizes should be avoided

where possible to increase the readability of the

code and avoid the need to use specialised code marking equipment.

The pack code may be printed in either normal orientation or inverse to allow for coding

techniques such as laser ablation.

3.3 Pack Code Quality

The quality of the Data Matrix code printing applied to the pack should be 1.5 (C) or better in

accordance with ISO/TEC 15415:2011, and printed using ECC200 error correction and will

utilise ASCII encoding according to ISO 16022. The use of ASCII encoding is to ensure

maximal interoperability with code reading devices likely to be used in the field.

European Pack Coding Guideline Page 11 of 11

Version 4

3.4 Human Readable Representation

The requirement is that the fields of product code, expiry date, lot number (batch code) and

serial number shall be shown on the pack in human readable format.

If a national Code (national reimbursement number/NHRN) is also included, the human

readable representation of that value shall also appear on the pack.

Where the dimensions of the pack allow, the human readable elements shall be adjacent to

the Data Matrix code.

In the event that the sum of the two longest dimensions of the pack are equal to, or less than

10 centimetres, the human readable fields do not have to be applied.

In most cases, country specific requirements will prevail both in terms of character size and

format e.g. DD/MM/YYYY for expiry date etc.

Each field printed on the pack should be preferably prefixed according to the following

guideline:

Field Name

Preferred Prefix

Product Code

PC:

Expiry Date

EXP

Batch Number/Lot Code

Lot

Serial Number

SN:

NHRN

NN:

3.5 Code Placement

Learning from the EFPIA Sweden Pilot and from manufacturers who are already serialising

packs is that where to place the code on the pack is an important consideration:

Experience has shown that printing high-quality codes is one of the decisive elements

between successful implementation and failure. In addition to selecting the most appropriate

printing technology this requires manufacturers to ensure that the pack carton is suitable, i.e.

that the appropriate substrate is selected, that varnish-free areas are available, and that

there is enough free space on the artwork (including possible interaction with the chosen

tamper-evident technology).

Establishing good control over the packs as they are presented to the printer is a key

measure in achieving high and consistent print quality and enables successful operation of

the vision systems.

Also consider the implications of placing the Data Matrix in relation to other barcodes, as the

Sweden Pilot has shown that carrying codes in close proximity will lead to read errors.