Ethics & Engagement across the Wellcome Trust Major Overseas Programmes

Update on Informed consent project in Vietnam.

The prọject to explore options of clinical research stakeholders (patient, researchér, doctor and ethical committee members) in order to create guidance and informed consent sheet template valid in Vietnamese.
interviews, I found the term "Nonclinical trials" is not easy to explain to the patient while it is not a problem to the researcher. Interviews from the researchers are more informative than the one from the patient. Transcription and translation are more time consuming than I thought. I m fighting with coding and don't know how to pick up the theme since every time I read the scripts or listen to the audio I find differentừ things. How to pick up the theme? How to relate them to the context? What does it mean, how to demonstrate it etc...
Thanks for reading my blog.


Views: 179

Comment by Nguyen Thi Thanh Thuy on November 15, 2014 at 15:30
I m sorry, i would like to add, 42 interviews have been conducted.
Comment by Dina Rippon on November 19, 2014 at 9:56

Dear Thuy,

Thanks for posting your update! It's interesting to hear about the successes and challenges you've faced. The question of 'how to pick a theme' when reading through transcripts is a good one, and one I will pass onto the other bursary holders to see if they can provide you with some insight! Thanks again, and I look forward to hearing more!

Comment by Maureen Njue on November 19, 2014 at 11:48

Hi Thuy,

Great hearing from you, sounds like you are having an interesting time and you have done so much work...kudos. Transcription and translations are really challenging and I empathize with you. We have, what we call a transcription SOP, I will forward that to you, hope it will be helpful. When we have non-English audios, we first transcribe them in the original language and then translate these into English being careful to translate the meaning and not only the words...often the meaning can be lost in translation if you are not careful. In transcribing, its best to type up what is said by the participant (verbatim).

There is a free software, Express scribe ( http://www.nch.com.au/scribe/) that we use when doing transcriptions, I have used it and I love it....much easier to use than windows media especially when it comes to forwarding and rewinding. try it.

You mention that you have some transcripts that you find quite informative, you could read a few (2 or 3) that you think have loads of information and find out what are the common things that people are talking about in regards to a specific question that you asked....this common things that people say ....become your codes or themes...e.g. Looking at informed consent maybe most of the researchers think that the consent process is too legalistic or its there to tick a box....you could have as one theme....Perceptions of researchers about consent....under this you could have....Legalistic process etc....sorry if these are not true in your case. - this is mainly used if you are using a thematic content analysis where your themes or codes emanate from your data.

Thuy, even those patient transcripts that don't seem to provide much information and still useful...no information is still information...it might be that the patients know nothing about the consent processes that go on....since you are planning on guidelines, that is interesting information on how the consent process can be strengthened not only to make researcher's work easier but to also ensure that the process does what its supposed to do....inform participants about their involvement in research and for them to know that they have a choice in this process

There is a book...Qualitative Methods for Health Research by Judith Green and Nicki Thorogood....you might find it helpful if you can lay your hands on one. Will try look for other material that you can access online.

Hope this is helpful in a small way, happy to chat more about this on mail.

Best wishes


Comment by Maureen Njue on November 19, 2014 at 12:01

Oh… I forgot to mention….when developing your themes, ask another two or three people in your team  to do the same using the same (2-3) transcripts and you can discuss the different versions of the developed themes and come to an agreement which are the most appropriate themes and sub themes to use. Once you’ve agreed on these….you can apply this thematic framework on all the other transcripts ….you can also have themes that are slightly different when looking at patients in comparison to the researchers since their issues might be very different.

Comment by Michael Parker on November 22, 2014 at 10:28

Wow. 42 interviews. That's impressive.

I think that you identify a problem that is very common in these kinds of projects i.e. the problem that many of the interviewees do not understand or have had no experience of the key concepts or ideas. In your case, they struggle with the term 'clinical trials'. The key in interviewing people in these kinds of situations is to try to find ways of helping them to talk about their own experience of research in a way that is related but without using the terms e.g. to ask them about their personal experiences relating to their illness in the first instance and then if it seems possible to gradually focus in on the experience of the research project. Perhaps we could talk about this kind of problem at the summer school? I think everyone would benefit from hearing about your experiences.

With regard to the coding. I think that what Maureen has said above is very helpful. One way of thinking about this would be to start out by reading a small selection of the interviews a few times and mapping out all of the issues that you think are important as you go through. This will probably create a long list of 'codes'. Key meaningful elements of what is being said. Then you could look at this list and see whether based on your experience, your reading of the interviews, and common sense, any of these issue can be grouped together under higher level headings and go on doing this until you have a more manageable number of codes. For example if you have a lot of codes that are to do with difficulties understanding key scientific or medical terms at the time of consent you might put all of these together under the concept of 'challenges with informed consent'. When you have generated this shorter, more manageable list of codes you could then go on to read a few more interviews to see whether the codes you have generated also cover all of the issues in these new additional interviews or whether you need to add new codes etc. 

Comment by Lindsey Reynolds on November 23, 2014 at 18:20

Hi Thuy,
I will add a few comments to Maureen's very useful feedback.
In terms of translation, if there is a particular word, term or phrase that is difficult to translate or important for the project, I would suggest you include the original term, and the literal translation in parentheses within in the english transcript. It may be important to use it when you write up the article or to point out that the term is confusing to translate. Sometimes questions of how words get translated are essential to understanding meaning and local context. For example, in my research on programmes for orphans and vulnerable children in South Africa, we found that the way people translated the term orphan was to use a Zulu word 'intandane.' However, people explained that this term carried a lot of stigma, and actually didn't only mean a child whose parent(s) had died, but also meant that the child had been abandoned by their whole family. This has to do with the social reality that in this area people often don't think about children as belonging only to their mother and father, but rather to the entire extended family.. So to use this term 'intandane' in a research context was very problematic, and something we needed to document and analyse.
In terms of your question about coding, my general advice is to remember that qualitative data analysis is iterative and fluid, meaning that codes and themes should continue to change as you move through the analysis. This does make the process more difficult, as it means you often have to go back and re-code multiple times, but I think imposing too strict of a set of themes up front often makes your findings less complex and nuanced. Remember that coding in qualitative analysis is a tool for thinking. It does not have right or wronge answers (like statistical analysis), and you really only come to understand themes through analytic writing.

Hope that helps!


Comment by Nguyen Thi Thanh Thuy on November 24, 2014 at 2:09

Dear Mike, Maureen and Lindsey,

Thank you very much for all of your comments and advices, they are very helpful. I download the book and read through it, yes it is very practical and adaptive book for health care related field. Yes, transcribing and translation are time consuming and it can be delegate to someone but there is a risk of loosing the important meaning while transfering from local language to English; that's why I have to do the proof reading and this process help me somehow "analyse". Maybe the difficult thing I feel is that I make many questions like "why this person say this, what is the best way to interprete this expression, where is the proposition the person stand??? (sometime he/she represent for the patient as individual, sometime as a whole, sometime for the researchers...), the analysing process is challenging but interesting. Yes, as you mentioned I will try to focus on the research question at first and use the coding as a tool for analysing. Mike has made clear to me the good way is to make the coding into manageable list of code.



Add a Comment

You need to be a member of e-MOPs to add comments!

Join e-MOPs

© 2022   Created by Dina Rippon.   Powered by

Badges  |  Report an Issue  |  Terms of Service