Skip Ribbon Commands
Skip to main content

Reflections on a dramatic month for IDMP

NNIT_LifeSciences (9)_jpg2.jpg

By Rune Bergendorff

June 2017 turned out to be a dramatic month for IDMP in Europe. With the final ISO implementation guidelines in progress of publication and a commitment from EMA, which was stressed by CIO Alexis Nolte in March, on delivering a reduced scope of data in iteration 1 but still according to plan, along with BREXIT as a burning platform, everything looked positive.

However, things changed significantly at the online EMA SPOR/IDMP taskforce meeting and the following subgroup working days at EMA. Everyone awaited the outcome from the gate meeting that was held at EMA during the end of May with great excitement. The expected outcome was a green light to proceed at full speed and bring IDMP closer to its implementation in Europe.

It turned out quite differently. Now, there is no official deadline for the life sciences industry to submit product and substance data in Europe. The data scope was deemed too large (even with substances de-scoped), the target operating model too ambitious, and as a result a 1:1 technical data migration is the only thing budgeted until EMA's relocation in 2019.

What happened?

This question seems to be on everyone’s lips these days, and there may not be a single exhaustive answer. However, two themes recur in all of the answers.

The first one being that the entire foundation of IDMP in Europe was fragile from the beginning. With the issuing of Implementing Regulation 520/2012, specifically Articles 25, 26, and 40, ISO IDMP is mandated. However, this regulation was established in 2012, the same year as the standards are published, not recognizing the need for ISO Implementation Guidelines in order for the standards to be actionable.

The dependency on the implementation guidelines serves for a volatile foundation during the early days of IDMP, and is the direct consequence for EMA getting approval to interpret the 1 July 2016 as a starting date, thereby changing the timelines that, since 2014, have changed 9 times so far.

The other theme has to do with ambitions. IDMP is by nature ambitious and requires a variety of skills to succeed and spans a large stakeholder group within the life sciences, including industry, national competent authorities (NCA), and EMA. This complexity may have been taken too lightly. In comparison, CTD took 14 years to implement in Europe, and IDMP was expected to be implemented within three years. At the end of the day, IDMP is the digitalization and logical next step after CTD to leaving documents.

Integrating IDMP data process-wise with the CTD from the beginning may have been too much to handle. Furthermore, the involvement of NCAs has also been taken too lightly. NCAs were not part of xEVMPD and are, therefore, one step behind industry and EMA.

What could have been done differently?

The focus on data fields was intense and a lot of time and effort has been spent on defining the data elements for the different iterations. Compared to other industries, e.g. retail, microscopic amounts of data are being gathered. The discussions on the inclusion of an additional 60 data elements on top of xEVMPD have been debated for three years.

Retail is gathering and reporting thousands of data points on their products, not because they have to but because they want to optimize their business, and to optimize business you need reliable big data to support your decisions. And while retail is not subject to similar regulation and, therefore, defines the scope themselves, banking is in a regulated setup similar to pharma, and still they have managed to report a serious amount of data elements.

When looked at retrospectively, the roll-out plan could and should have been different. Two of three parties were somewhat ready and mobilized: EMA and industry. The focus should have been on expanding the dataset and building on the existing xEVMPD reporting process.

In doing so, the NCAs could start to get involved in the validation of the data, get a feel for it, and once it is settled, the next step could be to integrate with the submission process. A change that will take years and will need to be at a different tempo based on when the NCAs are ready.

The last thing is the taskforce and subgroup design. It has been proven that on-site workshops for subgroups are the right way to move ahead with activities. However, for the task force, it seems that the number of members is too large and the modus operandi doesn't support the program in moving ahead in an agile way. The consensus is difficult to reach and, as the group meets rarely, it takes time to move proposals ahead.

An alternative could be to have a core group working more or less on a daily basis as part of the program with representatives from each group (industry, NCA, vendor). This group could then liaise with their counterparts and ensure commitment. Reporting sessions, or DIA info days, could be conducted on a more regular basis to inform the general public on the progress and important decisions. This could ensure the needed agility and drive.

The unanswered questions

The focus will now be on implementing organizations and referential data in our current processes via Organizational Management System (OMS) and Referentials Management System (RMS), thereby starting the adoption of SPOR in our regulatory processes.

However, some questions remain unanswered with the new approach:

What will happen to the Falsified Medicines Directive (FMD)?
During the March 2017 task force meeting, European Medicines Verification Organisation (EMVO) was accepted as official members, which is due to the link between the two initiatives where FMD will require product master data, which was expected to come from Product Management System (PMS). Now, this data will not be available and a new solution must be found.

What about ISO ICSR?
While ISO ICSR and E2B R3 is designed to be able to run without IDMP, it is not really getting the true benefits before the decision trees can be supported by IDMP data.

What about xEVMPD?
At the latest task force meeting, it was stressed that the development and maintenance budget for xEVMPD has been frozen to prioritize resources for SPOR. For EMA, the SPOR roll-out will continue, although at a slower tempo, but if xEVMPD needs to run until 2021, 2022, or whenever PMS will be live and available, the platform will need further maintenance in order to keep it alive.

Further questions arise around the link to the clinical trial portal, the global introduction of Substance Registration System (SRS) etc.

Where do we go from here?
The overarching question is where do we go from here and what happens to IDMP? Is it dead?

No, IDMP is not dead. IDMP is not a European project, but rather a set of ISO standards to implement globally; other regulators are still following the plan and are awaiting the publication of the ISO implementation guidelines that are currently in process of publication.

The first interesting regulator is, of course, the FDA. They have shown a huge commitment through their involvement in ISO and supporting EMA. They have put in tremendous effort in developing the SRS application together with NCATS, and have already demonstrated the application with integrations to adverse event reporting and products, making substances the centerpiece. Furthermore, the draft guidance notice on the submission of manufacturing information shows that FDA is thinking about IDMP in their continuous effort of improving data. If all goes well, we may see more from the FDA early next year.

While waiting for other regulators, a more indirect implementation of IDMP within the industry may be seen. As applications such as RIMS, EDMS, and ERP are updated or replaced, it should be required that IDMP compliance is built into the new applications. In this way, the foundation for a future IDMP submission is created while focusing on added benefit for the organization. The downside will be that the cross-organizational effort and driver will diminish and we risk falling back into solving the challenge from within our separate silos.

No matter what happens, IDMP will remain, but the way of implementing and the time frame has changed significantly. However, perhaps it will be beneficial to see the implementation first in the US.


Article written: July 2017




Rune Bergendorff+45 3075 1747 rrb@nnit.comManaging Consultant - Life Sciences Advisory Bergendorff



Algorithm sorts 18,000 documents at LEO Pharma,000-documents-at-LEO-Pharma.aspxAlgorithm sorts 18,000 documents at LEO Pharma
Scale Up 360° Pharma MES Europe°-Pharma-MES-Europe.aspxScale Up 360° Pharma MES Europe
Pharma MES Asia 2020 MES Asia 2020
The Pharma Factory of the Future is Digital - and it needs MES Pharma Factory of the Future is Digital - and it needs MES
Accelerated Implementation Methodology Implementation Methodology
6th Data Integrity Roundtable Data Integrity Roundtable
Why SMBs often miss the full potential of Veeva Vault SMBs often miss the full potential of Veeva Vault
Future of Regulatory Affairs of Regulatory Affairs
Contact Us Us
Webinar: IDMP Revised - The EMA implementation guide is out, but what is your strategy?,-but-what-is-your-strategy.aspxWebinar: IDMP Revised - The EMA implementation guide is out, but what is your strategy?