Info

Laboratory

Up to OpenClinica Building Blocks

Laboratory

Posted by Jolanda Strubel at December 02. 2016

Special thanks to Marlien Beusink (VU University Medical Center) and Sander de Ridder (Netherlands Cancer Institute) for setting up this CRF.

 

An example of the CRF is present in the sandbox environment ("TraIT_BB_General_Laboratory"). You need to be logged in, to open the link.

If you have any questions or suggestions for this particular CRF, please add a comment below.

Please find attached here:

- the actual CRF in Excel format (NOTE: please change the name of the CRF if you use it in your study)

- an overview of the conventions used for the setup of the building blocks

Attachments

Re: Laboratory

Posted by Sander de Ridder at December 21. 2016

Hi, I have one general comment: Currently the CRF contains e.g. RESPONSE_OPTIONS_TEXT: (please select), -unknown, -temporarily unavailable RESPONSE_VALUES_OR_CALCULATIONS: ,-3,-6 DEFAULT_VALUE: (please select) which used to be the way to go. However, currently this is the better way: RESPONSE_OPTIONS_TEXT: -unknown, -temporarily unavailable RESPONSE_VALUES_OR_CALCULATIONS: -3,-6 DEFAULT_VALUE: (please select) So basically: remove the "(please select)" from the RESPONSE_OPTIONS_TEXT and the "," from the RESPONSE_VALUES_OR_CALCULATIONS. This prevents errors with the SPSS syntax. Groet! Sander

Re: Laboratory

Posted by Jolanda Strubel at December 22. 2016

hi Sander,

thanks for pointing this out to us. I will update all BBs to include this changed setup (see attached image).

Attachments

Re: Laboratory

Posted by Jolanda Strubel at December 29. 2016

Version 1.1 has been uploaded (here in the forum and in OpenClinica). This version includes the "please select" change and the "undo radio button" option.

Re: Laboratory

Posted by Frank Zweerus at January 05. 2017
Jolanda, een paar kleine puntjes die mij opgevallen waren: 1) In het Laboratory form zijn er veel velden (waaronder: Hema_not_perform_reas) waarbij er een SIMPLE_CONDITIONAL_DISPLAY is gespecificeerd waar een komma in gebruikt wordt. Het viel mij op dat dit in een discrepancy note als ‘##’ wordt weergegeven. (“Question whether hematology was done was answered yes## but values were entered. Please remove these values]”) Bij andere formulieren (bijvoorbeeld (Cardiac_Ego, Echo_Available ) heb je in een vergelijkbare melding zonder komma. Dat ziet er prima uit. 2) Het valt mij in het Laboratory form trouwens ook op dat de hearder van de kolom SIMPLE_CONDITIONAL_DISPLAY’, verminkt lijkt (SIMPLE_CO_NullITIONAL_DISPLAY), en dat is trouwens ook het geval op het tabblad ‘Instructions’ 3) Bij het veld Hema_not_perform_reas zijn er meerdere opties (‘-hemolized sample,-no sample available, -other, -unknown, -temporarily unavailable’ met stored values: ‘1,-101,-7,-3,-6’)> Heb je hier bewust ‘-hemolized’gebruikt en niet ‘hemolized’? Omdat de stored value niet negatief is, had ik verwacht dat je er geen minteken voor zou willen zetten. Frank

Re: Laboratory

Posted by Jolanda Strubel at January 05. 2017

Thank you Frank for your feedback. The first 2 points will be changed in the next version.

Regarding the "-hemolized" text: in principle this is a kind of "null value" (as no values are to be expected), but as it is not a standard null value it was given positive response value. Does that make sense? Let me know what you think (or anyone else).

Re: Laboratory

Posted by Jolanda Strubel at January 06. 2017

A new version (1.2) has been uploaded here and on the OC sandbox environment. After internal discussion we decided to give a negative response (-103) value to the response option "-hematolized sample".

Re: Laboratory

Posted by Ruud Houben at July 24. 2017
Nice work with the building blocks CRF's! I will definately look at them for my next study. One thing I noticed on first glance is that non-repeating items do not have a Group label attached. In OpenClinica they are automatically assigned the UNGROUPED label. This label is important for writing rules, since all items are uniquely identified using the combination of Event object ID (OID), CRF OID, Group OID and item OID. The section where an item is displayed doesn't play a part in writing rules. When you have a large CRF with a lot of UNGROUPED items (like the Laboratory CRF), it can be a bit cumbersome to identify the correct item for your rules, or translate a specific rule back to the appropriate item. To prevent this, I have made a habit of providing a Group label for all items. For non-repeating items, this is usually the same as the Section label.

Re: Laboratory

Posted by Jolanda Strubel at August 04. 2017

Thanks Ruud for your reaction.

Good point you make regarding the group labels. We did not include these as in the past there was some trouble with non-repeating group items in the CRF layout and with the HIDE function. I understand that that is not the case anymore though.

We will keep your suggestion in mind for a next revision round, but for now I fear you will have to add these yourself to the CRFs.

Reply to this
You are an anonymous user. If you want, you can insert your name in this comment.
URLs will be automatically hyperlinked. Basic HTML tags are OK.
(Required)
Provide the text in the image. Just to avoid spambots
Powered by Ploneboard