What Life Sciences Firms can Learn From Other Industries About Optimizing Routine Document Production
In fields with complex information needs, such as engineering, use of structured authoring is second nature – particularly in regulated industries. Organizations spanning automotive, aerospace manufacturers and power utilities have teams of people routinely coordinating structured content, populating complex documents with approved information components.
In life sciences, the set-up has traditionally been quite different. Take the teams responsible for preparing regulatory content. As medical doctors, pharmacists and chemists by background, they have tended to favor office-style content tools, such as Microsoft Word, for generating documents. The downside of this practice is that, beyond manually cutting and pasting from existing documents, teams must largely start afresh each time a new or updated submission is needed — finding the right data all over again, and compiling it under the necessary headings.
A process that is at best inefficient, and at worst laden with risk of error.
Old habits die hard
Departmental information silos and practices, and the difficulties of harmonizing these, have not made it easy to move forward with more efficient document creation opportunities. But that is now beginning to change, thanks to technology approaches built around the concept of a single, master data and content repository, and the ability to automatically populate Word-like document templates linked to approved content components.
There are a number of drivers for establishing more efficient, reliable and automated document-building processes. It is the way regulatory authorities are heading in their evolving standards and compliance requirements, for instance. In addition, cost pressures on life sciences firms are increasing. A further impetus is the growing pressure on the life sciences to be more transparent, with ready answers available on demand and more detailed reporting.
The challenge is how to get to a data-driven approach to broader content management — from the current monolithic way of creating regulatory documents to a much more structured and readily automated method. This all starts with a single, definitive source of “product truth.” This must be expressed in a standard way, in an agreed format. Once available “off the shelf,” this can be repurposed to readily support multiple use cases.
Practical considerations
It is important to realize that the transition to structured authoring and its fullest potential could take months or even years. So, it’s a good idea to plan the transformation across a series of stages.
Beyond working towards good master data, companies may decide to transform simpler document production first — the kinds of submissions that re-use a lot of the same data, expressed in a very similar way, time after time.
Other documents, such as clinical study reports, will have a higher proportion of unique content — flowing narratives, hypotheses and interpretations of findings. Although parts of these documents will lend themselves to automation, based on approved data combinations or content “fragments,” compiling the fuller report will require a hybrid approach — part structured authoring/automation, part manual assembly. The benefit of structured authoring here is that less time needs to be spent on ‘stock’ content, and more focus can be devoted to developing the narrative.
The aim should be to work towards a single, central platform for not only master data but also approved content fragments ready for structured authoring, maximizing the opportunity to quickly and reliably create new documents — or roll out data changes. So, if the ingredients of a drug or a manufacturer’s process change, this information can be reflected promptly and confidently across all regulator notifications and revised labelling in all affected markets. This is an application that has captured the interest of many life sciences companies already, for fairly obvious reasons.
Building blocks for success
What does early best practice look like? Pilots and early rollouts to date suggest the following:
- Transform routine monolithic documents into “smart” documents. Replace some of the data items and values with intelligent tags, linked to master data, so when information needs to be updated, this only needs to happen once, at source. While not strictly “structured authoring,” this brings intelligence and a level of efficient automation to existing processes.
- Use the table of contents to structure the outline of documents. Combined with tagging, this will result in content fragments that can be pulled into other documents, at speed.
- Break down those content fragments further so they can be used in more documents. Aim to get down to the level of names of medicinal products, for example. These should be expressed in a standard way, even a consistent word order – eg “Aspirin, 100mg tablets” every time, and not variations such as “Tablets, 100mg, Aspirin.” Look to create templates for these content fragments to aid this process, and to make re-use straightforward.
- Continue this exercise until the company has arrived at the level of granularity it needs – e.g. for high-volume use in CMC, labelling and forms which feature lots of inherent structure, so lending themselves to maximum automation.
The impact of developing a roadmap and applying these measures will be multi-faceted.
Exploiting structured authoring based on approved master content will accelerate submissions for new products or additional market entry, and improve ongoing compliance, maintaining market status. Once you have a structured, automated emphasis to document creation, quality and compliance are achieved by design. This in turn could lead to operational cost savings, freeing up professionals’ valuable time and allowing companies to think laterally about additional market opportunities which now begin to look more viable economically.
[javascriptSnippet]
