Ian Barrow, Head of IT and Technology at UK-based LSP Conversis joined us at XTRF Summit #online 2020 to discuss the challenges inherent in healthcare translation and the role that technology plays in overcoming them. This article is based on his talk.
Many areas of healthcare pose unique problems for translation. Clinical trials are an excellent example. Clinical trials need to be thorough and reach as many people as possible. For us, this means large numbers of languages. A typical trial might include over 50 language combinations, but some projects can go up to as many as 80, often including some uncommon languages. What’s more, trial documentation is sometimes also localized into a range of language variants. Naturally, all this means you need a large number of linguists. And an efficient way of selecting the right linguists.
Another challenge we face is that healthcare clients’ systems tend to be quite closed and they’re tight on time, meaning we need a way of mapping the client’s list of languages into our tools to make sure we don’t miss anything. Plus, the source content often varies based on locale. For instance, various files may need to be adapted to bring them in line with the regulatory requirements, healthcare system, or terminology of specific countries.
As for the content itself, it’s highly specialized and complex. This means subject matter experts are essential. Not to mention extremely thorough QA processes. Accuracy is crucial. After all, in healthcare, it really can be a matter of life and death.
When it comes to deliverables, clients often have specific requirements in terms of file names. They often do their own additional rounds of linguistic and QA checks, or we may provide a back translation for them so they can check the translation aligns with the source. We also deliver certificates. These are documents needed to fulfill the requirements of the clients’ quality management systems. There could be one certificate per file or one per language.
After delivery, our clients often make changes to both source and target texts to account for a country-specific rule. Consistency is key here, and we need to make sure that this change is also updated in the TM and TB for future translations. Then there are protocol amends. This is where feedback from trial participants leads to changes to the documentation. For instance, maybe the dosage is amended based on patient feedback. In this case, it’s essential that we only amend the specific part needed. Otherwise, the ethics committee will have to redo the same steps over and over.
Time is another challenge with clinical trials. Not only are these projects time-sensitive, requiring fast turnarounds, but they also tend to last for a long time. A standard project lifespan is around three years, but they can run on much longer. This means you need to have good processes in place for storing the content all together in one place, so it can be drawn on when needed.
Let’s look at an example project to put things into perspective. Let’s imagine it looks a little like this:
- Translation into 80 languages
- 240 native linguists
- Availability request sent to around 480 – 720 linguists
- 6 or more linguistic and QA steps, plus PM, technical, and DTP steps
- Up to 1,000 steps (otherwise known as tasks or jobs)
With a project like this, any five-minute task that needs to be applied to all 80 language combinations will end up taking around 6.5 hours of PM time. So it’s essential you manage all these processes as efficiently as possible. And that you make use of automation where you can. If we think of the certificates we mentioned above, if we’re providing one per file or one per language combination… that adds up to a lot of certificates! Naturally, it makes sense to automate as much of that as we can.
As you can see, technological solutions are an absolute must for this type of work. Without XTRF and all the systems we have in place, we simply wouldn’t be able to support our clients as we do. In fact, we have a motto: “If you’re not adding value, you’re probably wasting time”. For us, this means making our systems run as smoothly and efficiently as possible, and using automation so that our PMs and engineers aren’t wasting their time on manual tasks. Instead, they’re only working on the areas where they bring a genuine added value, where they need to make a decision. Everything else is automated.
Head of IT & Technology