We are currently working on the forum. For the short-term, all forum content will be in read-only format. We apologize for the interruption and look forward to collaborating with you shortly. All the best in your research!
Best Of
Re: French translation and Unicode.
Hi Romain,
You should do a global search and replace of \\ into \
Somehow all unicode entries have an extra \
Hope that helps,
Gerben Rienk
Re: New documentation for the contains operator!
Also check out the TDS post here for a great (and in-depth) example of using the ct operator with checkboxes and multi-option-items: http://www.trialdatasolutions.com/tds/howto/ctoperator.jsp
Re: Data missing in export
Hi Jo,
I'd try creating different extracts with limited sets of items/CRFs to see
if the missing subjects ever show up. It will require trial and error but
may help you narrow down if there's a particular offending item or CRF.
If you upgraded from 3.0.x or earlier make sure you've run all the scripts
and have the uniqueness constraint enabled on the item_data table
(reference:
https://docs.openclinica.com/3.1/installation/upgrading-openclinica-windows#content-title-3772
You could also try upgrading to 3.2 on a backup copy of your data and see
if it helps, though I don't think much changed with Extracts in the more
recent releases.
Cal
I'd try creating different extracts with limited sets of items/CRFs to see
if the missing subjects ever show up. It will require trial and error but
may help you narrow down if there's a particular offending item or CRF.
If you upgraded from 3.0.x or earlier make sure you've run all the scripts
and have the uniqueness constraint enabled on the item_data table
(reference:
https://docs.openclinica.com/3.1/installation/upgrading-openclinica-windows#content-title-3772
You could also try upgrading to 3.2 on a backup copy of your data and see
if it helps, though I don't think much changed with Extracts in the more
recent releases.
Cal
ccollins
5
Re: unable to install
Hi Michelle,
Please find attached the OC_XML_to_SAS_R_CSV.zip file that Linas developed if you are having issues downloading it from the link I gave you.
Thanks,
Thasbiha
Please find attached the OC_XML_to_SAS_R_CSV.zip file that Linas developed if you are having issues downloading it from the link I gave you.
Thanks,
Thasbiha
tkhaja
5
Addition of indexes to enhance performance
Dear group,
To decrease the duration of saving CRF's we at CTMM-TraIT have found a number of missing indexes in the OpenClinica datamodel. For example after adding the indexes the duration of saving a CRF with a lot of dynamic items decreased from approx. 45 seconds to approx. 5 seconds. More details can be found in the OpenClinica call http://jira.openclinica.com/browse/OC-5087 .
If you experience performance issues, then you could apply the script to your DB now. Before applying it to a production instance, please do first test this extensively on a backup copy of your DB!
Regards,
Jacob
Jacob Rousseau
CTMM - TraIT / VU University Medical Center Dept. of Medical Oncology (Room 3A35) De Boelelaan 1117
1081 HV Amsterdam
Mob: +31 (0)6-22442292
Tel: +31 (0)20-4443257
Fax:+31 (0)20-4444970
-- ===============================================================================
--
-- Test script to add indices on the dyn_item_form_metadata mapping and other tables to
-- hopefully improve performance for the XXXXX and (probably) XXXXX studies.
Re: Calculated field with missing values
Hi Adrienne,
One way would be to make the field required and ask that a value be entered
if not done (or null); for example "999". Then, in the calculation you can
use decode. In the example below I'm using ITEM1, ITEM2, ITEM3 as the
ITEM_NAMEs. In the first part of the calculation I'm basically saying, look
up item1, if the response value is "999" return a 0, else retun its value.
In this case I'm adding the 3 values and if 999 it will be summed as "0".
In the second part of the calculation I am trying to do a count. I use
decode to say if the value is "999" then return a "0" else return a "1". I
first do a sum of the values and then divide by the count. I didn't test
this specific example but I've used this type of calculation before.
Example:
ITEM1
ITEM2
ITEM3
CALC_AVG
func:
((sum(decode(ITEM1,999,0,ITEM1),decode(ITEM2,999,0,ITEM2),decode(ITEM3,999,0,ITEM3))/
(sum(decode(ITEM1,999,0,1),decode(ITEM2,999,0,1),decode(ITEM3,999,0,1)))
I hope this helps.
Best Regards,
Alicia
One way would be to make the field required and ask that a value be entered
if not done (or null); for example "999". Then, in the calculation you can
use decode. In the example below I'm using ITEM1, ITEM2, ITEM3 as the
ITEM_NAMEs. In the first part of the calculation I'm basically saying, look
up item1, if the response value is "999" return a 0, else retun its value.
In this case I'm adding the 3 values and if 999 it will be summed as "0".
In the second part of the calculation I am trying to do a count. I use
decode to say if the value is "999" then return a "0" else return a "1". I
first do a sum of the values and then divide by the count. I didn't test
this specific example but I've used this type of calculation before.
Example:
ITEM1
ITEM2
ITEM3
CALC_AVG
func:
((sum(decode(ITEM1,999,0,ITEM1),decode(ITEM2,999,0,ITEM2),decode(ITEM3,999,0,ITEM3))/
(sum(decode(ITEM1,999,0,1),decode(ITEM2,999,0,1),decode(ITEM3,999,0,1)))
I hope this helps.
Best Regards,
Alicia
agoodwin
11
Re: Am I missing something here?
Hello Don,
This puzzling situation was also addressed at: https://forums.openclinica.com/discussion/4680/problems-concerning-recording-and-presentation-of-discrepancy-note-information/p1
Bottomline is: do not use discrepancy_note_type.
I think you will find what you're looking for in DiscrepancyNoteType.java of the source.
Hope this helps,
Gerben Rienk
http://www.trialdatasolutions.com