Given the confusion about terminology about this issue one more a more frequent approach is to just say what you actually did/assumed and avoid labels. 

See attachment.

Pablo


Pablo Alonso Coello

Centro Cochrane Iberoamericano

Instituto de Investigación Biomédica (IIB Sant Pau)

CIBER Epidemiología y Salud Pública (CIBERESP)

C/ Sant Antoni Maria Claret 167 (Pabellón 18, planta cero)

08025 Barcelona |  España

[log in to unmask] | Skype: tartanedo

T: (+34) 93 553 78 14 | F: (+34) 93 553 78 09

Cómo encontrarnos? How to find us? http://es.cochrane.org/sites/es.cochrane.org/files/uploads/Plano_Cast.pdf


De: Evidence based health (EBH) [[log in to unmask]] en nom de Ahmed Abou-Setta [[log in to unmask]]
Enviat el: dimarts, 10 / maig / 2016 20:58
Per a: [log in to unmask]
Tema: Re: Intention to Treat Question

Hi Anoop,
 
I think there is a misconception/ misnomer here. A study is not per protocol, ITT, but rather the analysis. Therefore, for example, you may analyze efficacy outcomes using the per-protocol principle, but the safety outcomes using the ITT principle. You'll see this done in a lot of drug trials. In all cases, be transparent on what you did.
 
Since you only have 36 participants (if I understood correctly), I highly doubt that any imputation method will markedly affect the results unless it was already borderline significant without the missing data. Set up your analysis plan a priori so that you are not biased by the results and report what you did (especially the sensitivity analyses to test the robustness of the results).
 
Good luck.
 
Ahmed
 
Date: Tue, 10 May 2016 14:17:53 -0400
From: [log in to unmask]
Subject: Re: Intention to Treat Question
To: [log in to unmask]

Thank you!

My question is that I have the person come back and do the primary outcome. But the person  didn't do the secondary outcomes. Since we are mainly concerned about the primary outcome, Can I still call my study ITT without doing any imputation or drooping the subject. 

And I will do a per protocol analysis too. Thanks!

On Tue, May 10, 2016 at 11:55 AM, Mayer, Dan <[log in to unmask]> wrote:
Of course, the BIG problem is what outcome to attribute to the subject who dropped out and could not be followed.  There are several ways of dealing with this.
1. Leave the missing patients out of the analysis since you don't know how they turned out.
2. Create a statistical imputation of how they are "likely to have responded", which is another form of multivariate analysis, which has its own problems.
3. Best case and worst case scenario gives you a "sensitivity" analysis of the possibilities of outcomes
It is also a good idea to compare those subjects who completed the study in their originally assigned group with those who completed the study protocol, the per protocol group compared with the intention to treat group.

Hope that this helps

Best wishes,

Dan 


Dan Mayer, MD, FAAEM, FACEP
Professor Emeritus of Emergency Medicine
Albany Medical College
Albany, NY, USA
From: Evidence based health (EBH) [[log in to unmask]] on behalf of ya dev [[log in to unmask]]
Sent: Tuesday, May 10, 2016 7:47 AM
To: [log in to unmask]
Subject: Re: Intention to Treat Question

If you have included all in the analysis,  those with missing values,  then it is ITT. Participant should be analysed in the group he was supposed at the start of the study
Regards
Ya dev


On Tue 10 May, 2016 4:37 pm Anoop Balachandran, <[log in to unmask]> wrote:
I had a question regarding intention to treat (ITT): I have all my subjects come back for testing (36). But one subject could only do the primary outcome and couldn't do the post testing for secondary outcomes.

Can I still call my study Intention to Treat?

Thank you
Anoop

----------------------------------------- CONFIDENTIALITY NOTICE: This email and any attachments may contain confidential information that is protected by law and is for the sole use of the individuals or entities to which it is addressed. If you are not the intended recipient, please notify the sender by replying to this email and destroying all copies of the communication and attachments. Further use, disclosure, copying, distribution of, or reliance upon the contents of this email and attachments is strictly prohibited. To contact Albany Medical Center, or for a copy of our privacy practices, please visit us on the Internet at www.amc.edu.