Beyond surface value — making label management less superficial

As long as a product’s packaging and labelling is treated separately from internal regulatory and product information management, the scope for inefficiency, error and risk remains significant — especially with the growing frequency of regulatory updates around the world. With this in mind, AMPLEXOR’s Romuald Braun explores the potential to make label management a less superficial activity.

Mislabelling is one of the leading causes of costly product recall,¹ and a particularly frustrating one given that it is so easily preventable. Yet, as long as product labelling is treated as a distinct, manually-driven process, life sciences organisations will continue to leave themselves open to the risk, not to mention a level of inefficiency they can ill afford.

Intensifying regulatory requirements and the evolution towards IDMP are causing regulatory and quality teams to look again at their processes and systems for managing content, however. In this context, labelling management as a topic subject was the subject of much discussion at AMPLEXOR’s recent annual conference. As organisations start to think more laterally about regulatory information and greater efficiency in how they respond to new demands, they are beginning to realise that the only real way to manage this in a sustainable way is by creating and drawing from a definitive master data repository that is capable of supporting multiple applications. Instead of starting from scratch each time there is new content (such as new labelling) to create, regulatory teams and those responsible, can simply call up approved content from a central resource.

Beyond single-purpose regulatory systems

This approach becomes a natural one as organisations move towards a more holistic, next-generation regulatory/product information management strategy. This is the vision advocated by Steve Gens, founder of Gens & Associates, one of the leading authorities on regulatory information management (RIM). He maintains that RIM should be as wide-ranging in scope as possible, encompassing dossier management, submission planning and tracking as well as manufacturing, change control, safety reporting, master data management as well as labelling and document management.

As long as such activities take place separately, each served by standalone systems and processes, companies risk repeatedly reinventing the wheel and introducing inconsistencies with each new content preparation task. If a company has procured systems for each distinct process from a series of different vendors, integration can be an issue and data’s dominion may be unclear: which is the authoritative, correct version of content and how is this determined and controlled?

Next-generation RIM needs to facilitate a seamless, reliable end-to-end process — from data/content collection to submission tracking and reporting. To further maximise the benefits, companies should be extending this same systematic process of content management to all important product data across a drug’s lifecycle. If this, rather than a targeted application of the data, becomes the core resource, it becomes possible to derive maximum efficiencies each time that content is repurposed for a particular use case.

So, ideally any master data management initiative should start with a product master data object model, of which regulatory intelligence is a part. The regulatory factors may not fit generic system data fields, being the proprietary IP of each company. However, if it the information is structured, it can still be reflected in the main product information system, contributing to that holistic, 360-degree resource, which caters for all information needs.

Right first time — every time

Combining product master data with regulatory intelligence makes it possible to automate more processes — including labelling management. Suddenly more becomes viable, and the need for heavy manual work is reduced each time there is a new content-related requirement. In the next-generation scenario, whether the result is new labelling or an IDMP-related submission, the output is ultimately just an expression of the company’s product data, in a particular format.

Taking a master data/complete product profile approach means all of the correct content for accurate, compliant labelling can be called up quickly and easily for the given use. In addition to ingredients and manufacturing information, it should be possible to call up detail for all authorised medicinal products alongside all the respective countries’ procedures, health authority organisation information and marketing authorisation programmes and processes. Labelling processes, change requests, sequences and templates should all be possible to manage in a clear and structured way.

Proper provision for labelling, to reduce risk and improve efficiency, should include the ability to select approved content elements as self-contained label ‘objects’ or assets. These might include the name of a medicinal product or its clinical particulars, pharmaceutical particulars or pharmaceutical form. Some elements might be market specific; others could be global/core content. However, using an object-based master data management approach, grouping fields becomes very easy to do, linking components as appropriate to the various applications — for example a particular artwork, or at a global level the company core data sheet or the core package insert. All of these fuller objects are referenced but the components are sourced from master data.

At-a-glance data trails

The clever part, where content is linked logically using metadata, is that by selecting one object it becomes possible to see at a glance the relationships to all the other objects, and therefore the knock-on effect of any changes to the content. So, the impact and risk assessment of any changes becomes a natural part of the design.

When content ‘fragments’ are used in several places in several documents, the master data and object-based approach means that if something fundamental needs to be changed, the ripple effect of those changes can be seen at a glance and the changes automatically reflected wherever this is needed, with full traceability. So, for instance the user might create the fragment name of a medicinal product once and then re-use it in multiple documents. As this is reflected in the master data, it can be referenced by other documents.

Each element is managed in the same way teams manage documents already, and creating the labelling structure is similar to the way teams create a submissions structure. It is a virtual document with all the respective sections: listed are the core company data sheet, a cover page, a table of contents and the name of the medicinal product, below which are the text fragments. The real leap is that those authorised to do so can now alter labelling elements fragments in a virtual document environment, and whole steps can be automated because the correct data will be applied automatically.

Once all of the necessary text elements have been input into the structure, it’s then just a case of deciding what to do. Similar to publishing, the submission structure is created independently from the published output. In this case, it’s possible to transform the content to the desired format according to rules that have been pre-set.

Previewing makes perfect

An ability to preview the document is desirable too, for instance as a complete PDF. The underlying enabler is the data-centric, object-oriented model, which allows labels and related documents to be created and amended using approved master data. Automated creation of labelling documents becomes something teams can do with confidence. And where regulatory intelligence is also reflected in the system, label creation can be done in accordance with the specific country requirements.

By seeing the bigger picture around data management, life sciences firms open themselves to a range of new possibilities — to reduce complexity, cost and risk, while improving productivity, accuracy and speed.

Reducing complexity begins with reducing the number of systems in use for different tasks, the number of interfaces and, as a consequence, the number of manual interactions (and associated risk factors, and inefficiencies — i.e., as it becomes easier to re-use approved content and automate the editing and creation of new materials). It goes without saying that there need to be important safeguards, controls and checks — especially where affiliates are part of the process. However, ideally these checks will be built in to the core system, ensuring compliance and building confidence and, as an extension of this, user acceptance.

It has taken time to get there but quantifiable simplification and de-risking of labelling management is now within reach in life sciences, a day many across the industry have been greatly anticipating because of the many practical challenges it can now overcome with a palpable ROI.

Reference:

1. Characteristics of FDA drug recalls: A 30-month analysis, US National Library of Medicine/National Institutes of Health, 2016: https://www.ncbi.nlm.nih.gov/pubmed/26843501