Mastering IDMP — what’s involved in master data management and how can it pay off?
Master data management promises an efficient, definitive way to address the demands of the forthcoming ISO IDMP standards for product information recording. What is involved in practice though, and how can companies ensure the investment pays off? AMPLEXOR International’s Sonia Monahan explains more…
As the life sciences industry appears to have grasped, ISO IDMP compliance is more than just another regulatory hurdle to straddle. It also promotes the kind of structure and discipline needed if organisations want to break new ground and take their businesses forward. It is an essential facilitator for digital transformation and for exploiting new technologies such as artificial intelligence. All the surrounding ambitions begin with data, and with assumptions about its quality, completeness and reliability as an accurate record of operations.
As EMA’s IDMP requirements near finalisation, the scale of the standards’ impact on managing regulatory data becomes more apparent. The transition from the current xEVMPD submission requirement to the more extensive and rounded demands of IDMP will involve extensive work. Data is extracted from a wider range of sources than just regulatory affairs (i.e., also across chemistry, manufacturing, and controls; clinical; pharmacovigilance; and manufacturing). The same high levels of data integrity and data quality need to extend across all of these business areas, so that the combined data can be relied upon as a definitive reflection of product reality.
The IDMP journey should have already started, or should have at least entered an assessment phase, where the data preparation task is scoped. Companies who leave their preparations too late in the process are likely to find that, in their haste to meet the IDMP compliance deadline, they sacrifice broader potential wins. Taking steps now to get the business’ master product data in order will pay dividends later.
Prepare once, re-use ad infinitum
The next aim should be to progress from the initial IDMP data analysis to a broader plan for ‘master data management’ (MDM), that will set the company in good stead for wider transformation, not least by strengthening transparency across the different business operations.
EMA’s own ambition for ISO IDMP supports this effort to improve the data’s quality and integrity, thereby increasing its value. This requires getting the underlying data (the master data) in order, using agreed upon standards. The reason ISO IDMP has taken so long to materialise as a set of confirmed definitions is that so much groundwork has gone into getting the detail rights; it’s also why there are five standards in total, rather than just one.1 This is intended to be a comprehensive definitive structure for data management.
Companies can enhance and add to this source data for their own internal purposes. The idea is that building on the right foundations and using agreed upon terminology will make the complete data set more meaningful and easier to repurpose. This applies consistently, whether for publishing, pharmacovigilance, resource planning, or artwork preparation. If the underlying data is trusted, the subsequent stages can happen much more efficiently and effectively.
Beyond compliance requirements, companies should be striving for a 360-degree view of product data. This includes a global, integrated view of product information, which supports business processes throughout the product lifecycle and provides a definitive master data set servicing multiple applications. The data doesn’t have to be confined to product information either. Agreed upon identifiers can be used to define other core business entities such as customers, patients, partners, suppliers, locations and employees. Ultimately, controlling master data will change the standard operating practices for life sciences firms.
Defining MDM best practice
What might MDM best practice look like? Jens Olaf Vanggaard, a senior life sciences R&D consultant at HighPoint Solutions and a member of the ISO IDMP SPOR Task Force Referentials sub-group, provided a useful analysis at AMPLEXOR’s recent annual customer conference.
He noted that a single, finished product takes three forms from an IDMP perspective, including: the pharmaceutical product as administered; the authorised medical product; and the packaged product that ships to market. This is just one indication of the complexity that systems need to be able to cope with to keep data correct and in sync. Below these higher-level definitions are the more intricate product details.
“The set of processes and solutions used to acquire, enhance and share product data across the enterprise are the key tenets of master data management,” stated Vanggaard.
Further parameters include ‘reference data’, the set of permissible values to be used by other (master or transaction) data fields. This data is typically non-volatile (slow to change). However, managing the processes and solutions used to acquire, manage and share this reference data across the enterprise will become increasingly important with the introduction of IDMP.
Vanggaard believes the journey to MDM should be viewed as an evolutionary one, though the scale of the transition could be construed as daunting. The important thing, he notes, is that companies start somewhere and treat developments as a continuum — with people, processes and technology brought on in parallel.
Without good governance, chaos will ensue
The starting point should be data governance, Vanggaard says, warning that “without good data governance, [companies] are likely to fail — no matter what technology [they] implement.” As long as there is inconsistency in data quality and definitions, the value of the system and its potential ROI is eroded, because the data simply isn’t sufficiently dependable.
With this in mind, it is important to establish early on how quality and consistency will be managed, who owns the data, and who is accountable for its quality and integrity. “Establishing clear communication channels will enable stakeholders to have a say in the data management process, increasing stakeholder acceptance and ownership of the data across the different functions,” Vanggaard advises.
Another early priority must be to author a data strategy, which defines how the company will increase the value, timeliness and reliability of data assets, perhaps by including external data sources which can augment and improve data quality and completeness.
Data policies and processes should then provide the documented guidelines, procedures and tasks to direct data stewards and other stakeholders, enabling them to ensure the integrity, consistency and sharing of enterprise data resources.
Data stewardship will be critical in extracting value from MDM and IDMP investments. This involves proactive management and oversight of an organisation’s data assets. Operationally, the remit can be broken down into a number of clear steps, from initial data profiling/discovery/scoping, and data modelling, to data cleansing, profiling, enriching, matching, consolidating and relating.
Data matching and consolidation stages involve comparing overlapping data across the company to arrive at the best version of the truth, keeping full cross-references to enable un-linking if needed. Data relating allows records to be grouped logically for management and analysis.
MDM checklist
A checklist of stages companies can expect to go through on the transition to MDM will look something like this, according to Vanggaard:
Given that life sciences companies will have to do much of this groundwork anyway to fulfil the needs of ISO IDMP, it is strongly in their interests to invest the time in getting this right and deriving the maximum business benefit, while future-proofing any investment. Other new regulatory demands will be much easier to meet once the core data structure is in place. Research by Gens & Associates suggests that companies using a common model for regulatory information management are 3.5 times more likely to realise business benefits, are 18% more efficient, and have 2.5 times more confidence in data quality levels.2 The potential gains grow exponentially when we broaden this approach to more comprehensive product and operational information.
IDMP compliance will require solid data governance and use of MDM principles and processes for data stewardship, irrespective of whether an organisation plans to implement MDM technology to support IDMP or not. Ultimately, ISO IDMP’s main focus is master data. As such, it makes business sense to harness this master data for maximum effect.
References