Clinical outcome assessments validated into Hindi, Urdu, Marathi and Gujarati. Two independent forward translations through to cognitive debriefing with patients, run as one managed process by one accountable person.
Most freelance vendors stop at harmonisation, because the steps after it need a clinician and a room full of patients rather than another linguist. I run those steps too. That is the whole difference, and it is why this page exists.
Two translators work from the source instrument without seeing each other's version. Independence is the point: where they diverge is exactly where the source is ambiguous.
Native target-language linguist, blind to B.
Native target-language linguist, blind to A.
Both versions come to me together.
I compare the two forward versions item by item and produce a single reconciled version, recording why each choice was made. That record is what a sponsor reads when they want to know how a wording was arrived at.
A linguist who has not seen the source translates the reconciled version back into English. If the back translation says something the original did not, the reconciled version is wrong, not the back translation.
You compare the back translation against your source and raise anything that reads differently. Nothing proceeds until that is settled.
Findings from the review are worked back into the target version, and terminology is aligned across every item in the instrument and across languages where more than one is running.
A clinician who practises in the target language reads the instrument for clinical sense, not linguistic accuracy. Wording can be a faithful translation and still be something no clinician would say to a patient.
Participants are recruited in India through a local interviewer I brief and coordinate. Typically five participants per language, with the sample size set by you. Each is asked what the question means to them in their own words, and where an item is misread, the misreading is recorded rather than smoothed over.
This is linguistic debriefing only. It tests comprehension of the wording. No clinical or medical data is collected.
Debriefing findings are worked in, the instrument is proofread, and you receive the final version with a report covering every step, every decision and every change, in the format your submission needs.
All of it. I select the linguists and the clinician, brief the interviewer, run every step, and answer for the result. You are not managing a chain of subcontractors, and you are not chasing an account manager who will have to go and ask somebody.
There is a real trade-off here and it is worth stating. A large vendor has redundancy I do not: if I am ill, your project waits. What you get in exchange is that the person who reconciled item 14 is the person who answers the question about item 14, three months later, without reading a file first.
Partner linguists and clinicians are people I have worked with across three decades, not a marketplace. Every one of them signs a non-disclosure agreement before seeing any client material, and that obligation flows down from whatever I have signed with you. I will tell you who is involved in your project if you ask.
Full confidentiality and data handling statement.
Debriefing runs under your protocol and your approvals, using your consent form, with signed copies returned to you. I do not hold ethics approvals independently, and a vendor in my position is not expected to. If your protocol treats the debriefing as covered by the parent study, that is your call to make and document, not mine.
Participant contact details and consent records are held only as long as you need them and are disposed of on your instruction. No clinical data is collected at any point.
I am in Depew, New York. The debriefing panel is in India. That is deliberate rather than a compromise: for a Hindi, Urdu, Marathi or Gujarati instrument, in-country participants who use the language daily are the ones whose comprehension actually tells you something. Saying it plainly here beats you discovering it during vendor qualification.
Informed consent documentation for Boehringer Ingelheim and Novartis, in six directions for each sponsor: into Urdu, Hindi, Marathi and Gujarati, and back out of them again. Courses for the WHO Better Birth programme. Medical equipment documentation. Three point seven million words of medical work across fifteen lines of the record, the field I have been commissioned in most often.
References from past sponsors are available on request, specific to validation work rather than general translation.
Hindi, Urdu, Marathi and Gujarati, all four translated personally rather than passed on. Additional languages can be arranged through vetted partners on request, though I would rather tell you honestly where my own reach ends than take a fifteen-language study and hope.
Quoted per project, not per word. The number of items, the number of languages, the sample size you want for debriefing and the reporting format your submission needs all move it more than word count does. Send the instrument and the protocol and you will have a figure and a timeline.
Enquiries answered from Depew, New York. An NDA is available before anything is sent.