I myself have been busy travelling, visiting the ISOQOL conference in Denver and the ISPOR conference in Madrid. The conferences are very different with ISOQOL being focused on quality of life research and outcomes research, while ISPOR is now primarily focused on health economics and market access. We had two posters on methodological issues associated with PRO translations at ISPOR (see publications link) while I gave a podium presentation at ISOQOL’s translation and cultural adaptation special interest group meeting. The presentation was well received and generated some lively discussion on the difficulties in pilot testing PRO translations with particular populations such as young children, severely ill patients and patients with mental and cognitive impairments. It will also now form the basis of a manuscript which should be completed by this time next year.

Since our last blog two new project coordinators have started. Hayley Simpson has a BA in French and Italian, while Gemma Wilkins has a BA in Modern Languages (French and Spanish) and an MA in Translation and Professional Language Skills. Both Hayley and Gemma are doing a great job and are a welcome addition to our team.

My argument above contradicts arguments put forward in a recent post on Lionbridge’s blog entitled “Medical Translation-Why Cognitive Debriefing Subjects Don’t Need to Be in Trial Disease State”. In the blog the author states the following:
“When we perform CD, we are looking to see whether the translation is culturally correct, linguistically correct, and clear about the information the instrument is trying to elicit from the patient.”

Unfortunately the author fails to also consider that patients have a completely different insight into their condition and its terminology than a healthy person. To give a clear example from a recent study we worked on in the area of bronchiectasis, one question asked patients about their sputum production. In the UK the term sputum would not be understood by the general public and therefore the wording was changed to phlegm during the in-country review, however all five patients with the condition suggested during pilot testing that in fact they refer to the symptom as sputum and that the wording should be changed back. If we had pilot tested the translation on the general population we would never have gained this insight and would have produced a translation which was inaccurate for the target population and the way they express their symptoms.

Lionbridge’s other argument for not recruiting patients appears to be a practical one, which is simply not justified. They state the following in their blog:

“Recruiting people in certain disease states is almost impossible. Chronic Obstructive Pulmonary Disease (COPD) is a good example.”

It is certainly not impossible to recruit patients with COPD (one of the world’s most prevalent diseases) or many other disease states. However, I do agree with their blog when they say that in certain cases creative thought does need to go into pilot testing with populations for which pilot testing is very difficult or unethical. The justification for this should not be a lack of effective recruitment procedures but should be based on whether the patient population is capable of completing cognitive debriefing which can take far longer and require greater concentration that completing the actual PRO measure, and whether the patient population can make a reasonable judgement about their involvement and its implications. Patient populations that fall into these groups include children, mental health patients and severely ill patients particularly those with communication problems. This is an area that needs greater thought and discussion and there are a number of methodologies available that could be combined to produce the best alternative to pilot testing with the patients directly.

I will be speaking and hosting a discussion on this specific issue at the Translation & Cultural Adaptation Special Interest Group meeting at the ISOQOL conference in Denver in October, and input from other industry experts, sponsors and PRO researchers will be most welcome.

PharmaQuest will be exhibiting at the ISPOR conference in Baltimore in May where we look forward to meeting many of our existing clients and hopefully some new ones! We have two poster presentations at the conference discussing two methodology issues that to date have been somewhat neglected in the literature. The first is how to translate the small percentage of PRO measures that are developed in a language other than English. The second is how to conduct linguistic validation with sensitive target populations such as severely ill patients and children. Copies of the posters will be available at our stand throughout the conference and we will be holding another translation prize draw to win a fancy gadget which is a must for any iPhone user!

“The management of medical translation services, specialising in the translation and linguistic validation of patient reported outcome measures conforming to BSEN15038:2006”

We are now one of the few companies in the world to be certified specifically for the translation and linguistic validation of PRO measures. Our management system and quality manual reflect the rigorous methodology and high standards we have been adhering to since our formation. Our certification conforms to BSEN15038:2006 which is a European Standard specifically designed to assess translation services. The inclusion of BSEN15038:2006 highlights the quality of our translation services, methodology and personnel in delivering the highest quality translations for our clients.

We would like to thank all our clients for their feedback and support, which has helped us reach this stage in the company’s development. The directors would also like to publicly thank all our staff for their hard work in attaining this valuable and important certification.

We presented a poster at ISPOR on Concept Elaboration highlighting that it is an essential stage in the translation of PRO measures. The purpose of the poster was to define what is meant by concept elaboration, why it is a critical step in the translation process, how to create a detailed and specifically-targeted concept elaboration document, and how it can be used to its full potential by being distributed to anyone who translates the measure. The result of a well developed and applied concept elaboration document will be that it eliminates misinterpretations and encourages standardised decision making across all translation providers, thereby reducing unnecessary variation between translations. If you would like us to send you a copy of our poster, please feel free to get in touch via our contact page.

These positive changes reflect our drive to improve our services and continue to offer our clients translations of the highest quality.  To meet these aims we have recently employed Maddelena Giovanaz as our translation quality manager.  Maddi has worked in the translation industry for several years and will be responsible for implementing a number of strategic changes at PharmaQuest to enable us to continually improve our translation quality and be responsive to the needs of our clients.  We are now working towards both ISO 9001:2008 and BS EN 15036:2006 certification and hope to achieve these goals before our 6^th anniversary.

PharmaQuest will be exhibiting at both the ISPOR and ISOQOL conferences this coming autumn in Prague and London respectively.  If you are attending either conference, we would love to meet you and discuss how we can assist you with your translation requirements. 

Firstly, linguistic validation should not be confused with psychometric validation. The latter refers to statistically analysing whether a patient reported outcome (PRO) measure demonstrates acceptable levels of reliability, validity and sensitivity to change. Linguistic validation is not an assessment of the PRO measure’s statistical performance but is instead an assessment of the validity of the language used in the translation.

The key way to assess whether a PRO measure is linguistically valid is by pilot testing the translated measure in the target country and language with patients from the target population. This usually involves administering the translated PRO measure to between five and ten patients and asking them a series of probing questions on their perception of the measure. These often include asking the patients if they understood the questions, whether any wording confused them, whether they would word any questions differently etc. Pilot testing may or may not include cognitive debriefing; terminology which is often incorrectly used inter-changeably with linguistic validation. Cognitive debriefing refers to the process of asking patients to verbally describe to the interviewer what each question means to them in their own words. This process allows experienced project managers at PharmaQuest to then determine whether the patient has correctly understood the conceptual meaning of the question and therefore, by implication, whether the translation is acceptable. Although we recommend including cognitive debriefing when pilot testing PRO measure translations, some very widely used PRO measures do not include cognitive debriefing in their pilot testing procedure. Therefore, cognitive debriefing should not be used to describe linguistic validation as it is one part of the process rather than a definition of the process. 

In summary, linguistic validation refers to the process of pilot testing a translated PRO measure in the target country and language with patients from the target population, and in doing so assessing whether the patients have fully understood the instructions, questions and response options. 

We hope you enjoy reading PharmaQuest’s new blog and exploring our website. We have also recently moved to bigger and brighter offices and will update this blog on a regular basis.

Publications

Clayson D, Verjee-Lorenz A, Two R, Gerber R, Beaudreuil J. (2011). Translation and linguistic validation – Methodological implications when the source measure is not English. Value in Health 14(3): A154.

Clayson D, Verjee-Lorenz A, Miller F, Two R. (2011). The role of the instrument developer in the translation of patient reported outcome measures. Value in Health 14(7): A432.

Two R., Verjee-Lorenz A., Clayson D., Dalal M., Grotzinger K., Younossi ZM. (2010). A Methodology for Successfully Producing Global Translations of Patient Reported Outcome Measures for Use in Multiple Countries. Value in Health 13(1): 128-131.

Two R., Verjee-Lorenz A., Clayson D., Dalal M., Grotzinger K., Younossi ZM. (2010). Response to the Letter from Ms. Tamzin Furtado. Value in Health 13(4): 508-508.

Verjee-Lorenz A, Clayson D, Two R, Giovanaz M. (2010). Concept Elaboration—an essential stage in the translation of PRO measures. Value in Health 13(7): A338.

Verjee-Lorenz A, Clayson D, Two R. (2011). Pilot Testing Translations of PRO Measures with Sensitive Populations. Value in Health 14(3): A153.

Verjee-Lorenz A, Two R, Clayson D, Miller F. (2011). Comparison of reconciliation and review methodologies for the translation of patient reported outcome (PRO) measures. Value in Health 14(7): A432.

Wild D., Grove A., Martin M., Eremenco S., Ford S., Verjee Lorenz A., Eriksen P. (2005). Principles of Good Practice for the Translation and Cultural Adaptation Process for Patient Reported Outcomes (PRO) Measures: Report of the ISPOR Task Force for Translation and Cultural Adaptation. Value in Health 8(2): 94-104.