This article, published in this month’s Regulatory Rapporteur, presents a case for evaluation of medical device product information and argues against transferring accepted PIL (Patient information leaflet) user testing methodology to an IFU (Instructions for use) context. The article was co-authored with Maria Mehlin of Spectrum Regulatory Solutions.
Consumer involvement in IFU and device labelling development
This article looks at clinical trials consent forms and information sheets intended for patients. It compares the nature of clinical trial patient-oriented documentation in the USA and the EU and touches on readability issues.
Up until recently, reviewing translations was not accorded much, or any, importance. This is the sole reason why we often receive manuals translated into English for goods produced in East Asia that are nonsensical and useless. The reason for this is primarily cost.
Where reviews were undertaken, the back-translation method was widely practiced. While there are advantages to the back-translation approach, the disadvantages are evident: it is time-consuming and the costs of the back-translation equal those of the forward translation. However, there are areas of the pharmaceutical industry where back-translation is a Standard Operating Procedure, such as linguistic validation of measurement instruments, such as Quality of Life or Patient-Reported Outcome tools.
Nowadays, some kind of review is an industry expectation. However, because translations usually involve a rapid turn-around to the client, it has become standard practice amongst translation companies worldwide to involve an independent translator in the review process. This is primarily a proofreading exercise for spelling, grammar, syntax, etc.
Contemporary international experts in translation theory and practice now recommend a third party, or even a community review. Typically, if the translated document was intended for a healthcare professional audience, then the third party reviewer would be an expert in that field, native in the target language. If the translated document was intended for a lay audience, such as a Patient Information Leaflet, then the third party review would be one or two lay native speaker of the target language.
The review procedures differ as follows:
· Back-translation: For specific activities where this is an expectation (e.g. PRO tools)· Internal review: proofreading step for spelling, grammar, syntax, etc.
· Community review: review step for general sense, understandability, readability, cultural appropriateness.
The community review could be developed as a mini-readability test, whereby the reviewer, whether a healthcare professional or lay native speaker, could be asked to find an explain items of information in the translated document. Alternatively, the community review could entail open-ended questions for discussion about problematic sections (in an understandability context) of the document.
Some companies might value the review processes and be content to pay for quality, while others may be reluctant to fund the community review in particular, perhaps viewing it as excessive. The solution to these objections is simply to educate about the importance of the review processes, but, above all, listen to the client and supply their needs.
For more information, please contact mark@gibsonrs.eu
A number of my clients have pointed out to me that the translation process is the one of the only activities that goes unaudited in Big Pharma. Throughout the drug development lifecycle, employees can barely move without having to comply with some Standard Operating Procedure or Best Practice Guide. Years ago, when I started providing services to both the pharma industry, the notion of ‘audit’ was quite alien to me, having spent so much of my early career in academia where nothing was audited. In fact, I think a lot of the academics across Europe I know would probably go prolonged sick leave if an audit was ever suggested. I remember being terrified during my first ISO 9001 audit and suitably traumatised during my first client audit – both of which my team passed. Then, as I became more experienced, working to SOPs and preparing for audits became routine and easier , although the auditing process should never be taken lightly!
The fact that the translation process for much of Big Pharma remains unaudited is quite a revelation. In addition, that Big Pharma companies are starting to address this loose end may pose significant changes to the way they manage and outsource translation projects. For example, I know that the following is already happening in Big Pharma:
· Affiliates of Big Pharma around the world will no longer operate in a practice of laissez-faire, where, for example, the Jordanian affiliate looks after documentation translations in Arabic, making decisions about outsourcing independently of the parent organisation. Instead, many large multinationals are starting to think about centralising their translation outsourcing practices
· While the plug was officially pulled on PIM by the EMA in early 2011, some Blue Chip companies are forging ahead with their own Product Information Management systems and are conducting their own audits to identify how they can most efficiently manage their translation flow
· As a consequence of the above points, some companies are looking at translation strategies, according to their own business models, thus creating vacancies for senior expert consultants to help develop and implement these strategies.
Above all, the biggest realisation felt by Big Pharma is that they do not have a great deal of understanding and visibility of how the translation process works, even when they have partnered with translation suppliers for a number of years.
In earlier articles, I have been critical of the conduct of translation agencies in general. What is patently clear to me is that agencies will have to alter their working practices if they wish to maintain working relationships with Big Pharma. For instance, over the past 7 years or so, I have partnered with 8 or 9 translation agencies and I never had any visibility of a) who was undertaking the forward translation into the target language, b) what validation / reviewing processes were involved (this was particularly a mystery, given the errors I was receiving) and c) who else has had an input into the translation and what Translation Memory systems were involved. Yet these are precisely the kind of details that any auditor would need to know – for every single piece of translation documentation.
The reason why agencies tend to hide away these details is out of self-preservation: translations are typically engaged by agencies at a lower price than they quote the client. Supposing translator and reviewer details are disclosed to the client as part of a translation certificate, there is nothing to stop client and translator pursuing a direct working relationship with each other, thus making the agency redundant, particularly when the agency’s ‘cut’ in the translation transaction is exposed and put into question.
SPC harmonisation is notoriously regarded as one aspect of regulatory affairs that is seldom fully outsourced. This is for obvious reasons: a bit like the uncertainty I feel when I hand over the keys to my house to people working on my house who I don’t really know while I’m not at home. However, on occasions, SPC harmonisation can be such a tough job, particularly when this involves multilingual input, that there is no other choice but to outsource at least part of the harmonisation process. It is precisely the ‘how to’ aspects of this process that the following article focuses on, both involving monolingual and multilingual SPC Harmonisation processes.
The following article is simply a write-up from a recent informal meeting I had with information design specialists working in other industries, a special interest group, I suppose. My role in the meeting was to describe how the PIL Testing industry started off as one thing, developed a life of its own and then went on a Frankensteinian rampage across Europe!
In the context of the above-mentioned meeting with delegates from many other, unrelated sectors where often no legislation is in place for the development and iterative design of instructive texts, it is absolutely clear to me that pan-EU PIL design and testing is the poor relation of the information design industry, despite a legislation being in place. What a great opportunity the pharmaceutical industry had back in 2005 to reach out to their patient audiences in a clear and concise way, yet what a paltry set of circumstances we now find ourselves in. The reason why this is so is clear: the legislation created a business opportunity, a bandwagon and too many under-qualified people carried out the testing. Arguably, the quality of Patient Information Leaflet is the same as pre-legislation: some very good PILs that were co-produced by members of the public and quality testers and some very bad ones that were forced through the testing process in record time and where there is little evidence of the flair, creativity and expertise that the pharmaceutical industry should expect from their PIL Testing providers.
The following document was included in a market analysis report compiled in July 2009. As such, some recent developments, such as the Lithuanian readability testing guidance and the Romanian accreditation guidance are not included. Neither are any of the guidance published by countries, such as Croatia and Ukraine. All of the above are subjects currently being written up as separate blog articles…so, watch this space! Nevertheless, a lot of the references included in this article might still be useful.
mg_key-references-and-local-legislation-for-pil-user-testing_09jul09.pdf
This article presents a discussion of the processes involved in hypothetical projects involving linguistic validation of Patient-Reported Outcomes (PRO) and Quality of Life (QoL) instruments for use in clinical trials. For more information, please contact mark@gibsonrs.eu.
This article presents a discussion of the need to localise medical device software applications. For more information, please contact mark@gibsonrs.eu
A lot of companies I’ve spoken to who sell products into Central and South America state that they do not invest in further localisation of labelling text and product information into Canadian French, Brazilian Portuguese or Latin American Spanish varieties. Meanwhile, I work with a good number of regulatory and linguistic specialists throughout the Americas who do not think this approach is a good idea. This is a short thought-piece that explains why it is worth doing that extra bit and localising into the relevant language varieties and why soon companies may not have any choice in the matter!
Translating for Health: localising product information for the Americas