[Loris-dev] populating Visit_Windows based on administration of baseline test battery, not age

[Carolina Makowski]

Hi Rida,
Thanks for your detailed and informative reply. If you could discuss this
with the LORIS team that would be excellent - I definitely see the value in
being able to configure more easily visit windows based on a particular
date rather than age. In particular, I think LORIS will likely be getting
this question more and more as more clinically and
pharmacologically-oriented studies house their data in LORIS, where many of
these studies have a time-critical component to assess treatment efficacy
for instance. Just my two cents :)

For our purposes, indeed we will have multiple visits (12 or more in some
cases!) but the visit window will always be based on the same date at
baseline, and these visit windows have already been clearly defined, so
given the link you have shared with me, I do believe it shouldn't be too
tricky for us to test this out.

Thanks again and have a great weekend!

Carolina

On Sat, May 6, 2017 at 8:27 AM, Rida Abou-Haidar <rida.loris at gmail.com>
wrote:

> Hi Carolina,
>
> The short answer with the current achitecture would be *NO*.
>
> However, that does not mean that it could not be added to our roadmap. I
> will add this item to our weekly discussion and see the response from the
> team. We actually have a similar case on the CCNA study that I am involved
> with and at this point in time I have the Visit_Windows table populated as
> suggested in the WIKI (i.e. 0 and 39999).
>
> On your end however, nothing says you can't have a prototype
> implementation of this. If you follow the link below, I am pointing to the
> code dealing with this functionality. The current code is calculating an
> age and comparing it to the otimum and max/min values; a few simple changes
> could change the age calculation to subtract the date of visit with the
> date of the first visit instead of the date of birth. Doing so, I would
> assume you could populate the Visit_Windows table as you wanted to and have
> it work.
>
> On our end, if the decision is made to implement this behaviour we need a
> much more generalised configurable approach to accommodate the needs of all
> different projects that would want this feature. Some things to keep in
> mind:
>
>    - When comparing dates, you might want to compare the date of visit,
>    date of creation of visit on loris, date a visit was actually started on
>    loris, date of completion of the initial data-entry, date of completion of
>    the entire visit, ... Depending on what you are trying to measure. Just
>    make sure you choose the date that is most relevant to use for you.
>    - If you have a third visit, do you want to compare it with V1 or V2 ?
>    if you have a V4 ?? It gains complexity rather quickly so make sure those
>    decisions are made early on.
>
> Finally regarding the NULL values in Visit_Windows, I have personally not
> made the attempt, however we used to use values (0,0) and this would cause
> the "Within permitted" field to constantly be red so we switched back to
> the recommended values (0 and 39999)
>
> Thank you for your patience and sorry for the delay in the reply,
> Rida Abou-Haidar
> CCNA Software Developer
> Montreal Neurological Institute
> McGill University
> rida.loris at gmail.com
>
> On May 3, 2017, at 5:49 PM, Carolina Makowski <carolina.mak3 at gmail.com>
> wrote:
>
> Hi,
> I was wondering what the correct syntax would be when populating the
> Visit_Windows table for a longitudinal study that bases optimal visit
> windows against the baseline timepoint, and not necessarily based on age.
> For example, if I have two visits (baseline [V1] and 6-month follow up
> [V2]) and the optimal time window for the 6-month follow-up is within 60
> days, would it be correct to populate Visit_Windows with the following for
> the 6-month follow up?
>
> INSERT INTO Visit_Windows (Visit_label,  WindowMinDays, WindowMaxDays,
> OptimumMinDays, OptimumMaxDays, WindowMidpointDays) VALUES (‘V2’, ‘120’,
> ’240’, ‘120’, ’240’, ’180’);
>
> I suspect something is missing here in terms of specifying the calculation
> based on the baseline assessment date.
> Also, is it possible to list the window for the baseline visit as NULL, or
> would you recommend going with what is outlined on the wiki page (i.e. 0
> and 39999)
>
> Thanks in advance for your help!
>
> Carolina
> _______________________________________________
> Loris-dev mailing list
> Loris-dev at bic.mni.mcgill.ca
> http://www.bic.mni.mcgill.ca/mailman/listinfo/loris-dev
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.bic.mni.mcgill.ca/pipermail/loris-dev/attachments/20170506/5fe51792/attachment.html>