Extract of response options

Hi all,
When extracting data from OpenClinica 3.0.4.2, I get the database values of the selected items from radio buttons etc. E.g. I get a "1" when the "Yes" from a "Yes, No" / "1, 0" reponse options text / response values set has been selected.
In this example, the value "1" is rather obvious, but it may not be so in all cases: "No, absolutely not", "Nah, not really", "Well ...", "Perhaps a bit", "Sure thing!" coded as "NAN, NNR, W, PB, ST" isn't that obvious. The CRF itself - when viewed in OpenClinica - doesn't necessarily provide enough information to shed light on the mystic "NAN" and "W" values in the data set, as these may not at all correspond with the textual representation of the CRF.
I have found a place where these option / value sets are visible, deep down in the Build Study, but this isn't available when you're a data specialist or an investigator, so how is the data specialist informed about these option / value sets? I cannot find an option to add this of the data set when extracting data?
Best regards,
Janus

Comments

  • Dear Janus,

    These codelists are documented in the spreadsheet used to set up the OpenClinica CRF and in any ODM XML export from the study (in XML, rather than list format). The data specialist may understand OpenClinica or prefer the XML.

    Alternatively, you may want to document the codelists elsewhere. For example, you could document the codelist on an electronic copy of the paper CRF, if you have one.

    Yours,

    Michael
    Sent: 27 December 2011 14:13
    To: [email protected]
    Subject: [Users] Extract of response options

    Hi all,
    When extracting data from OpenClinica 3.0.4.2, I get the database values of the selected items from radio buttons etc. E.g. I get a "1" when the "Yes" from a "Yes, No" / "1, 0" reponse options text / response values set has been selected.
    In this example, the value "1" is rather obvious, but it may not be so in all cases: "No, absolutely not", "Nah, not really", "Well ...", "Perhaps a bit", "Sure thing!" coded as "NAN, NNR, W, PB, ST" isn't that obvious. The CRF itself - when viewed in OpenClinica - doesn't necessarily provide enough information to shed light on the mystic "NAN" and "W" values in the data set, as these may not at all correspond with the textual representation of the CRF.
    I have found a place where these option / value sets are visible, deep down in the Build Study, but this isn't available when you're a data specialist or an investigator, so how is the data specialist informed about these option / value sets? I cannot find an option to add this of the data set when extracting data?
    Best regards,
    Janus
  • kristiakkristiak Posts: 1,301 ✭✭✭
    Hi Janus,

    What we usually do is to provide the code lists manually. The reason that the numbers are displayed and not the decoded values is simply for statistical purposes. A statistician could not do much with the decoded information. If you need both you can find this information also in the XML files.

    Regards

    Krister
    Sent: 27 December 2011 15:13
    To: [email protected]
    Subject: [Users] Extract of response options

    Hi all,
    When extracting data from OpenClinica 3.0.4.2, I get the database values of the selected items from radio buttons etc. E.g. I get a "1" when the "Yes" from a "Yes, No" / "1, 0" reponse options text / response values set has been selected.
    In this example, the value "1" is rather obvious, but it may not be so in all cases: "No, absolutely not", "Nah, not really", "Well ...", "Perhaps a bit", "Sure thing!" coded as "NAN, NNR, W, PB, ST" isn't that obvious. The CRF itself - when viewed in OpenClinica - doesn't necessarily provide enough information to shed light on the mystic "NAN" and "W" values in the data set, as these may not at all correspond with the textual representation of the CRF.
    I have found a place where these option / value sets are visible, deep down in the Build Study, but this isn't available when you're a data specialist or an investigator, so how is the data specialist informed about these option / value sets? I cannot find an option to add this of the data set when extracting data?
    Best regards,
    Janus
  • JanusJanus Posts: 260
    Hi Krister,
    Thank you for your reply.
    I fully understand why it's the coded values being exported, but I'm just curious as to why these response option sets aren't exported as well in the .csv export? The response option sets could just be exported in a separate file? The statistician can't do much with coded values if she hasn't a clue to what information the coded value represents.
    Best regards,
    Janus
    "Krister Kristianson"
    Sent by: [email protected]
    04-01-2012 14:29
    Please respond to
    [email protected]

    To

    cc

    Subject
    Re: [Users] Extract of response options

    Hi Janus,

    What we usually do is to provide the code lists manually. The reason that the numbers are displayed and not the decoded values is simply for statistical purposes. A statistician could not do much with the decoded information. If you need both you can find this information also in the XML files.

    Regards

    Krister
    Sent: 27 December 2011 15:13
    To: [email protected]
    Subject: [Users] Extract of response options

    Hi all,
    When extracting data from OpenClinica 3.0.4.2, I get the database values of the selected items from radio buttons etc. E.g. I get a "1" when the "Yes" from a "Yes, No" / "1, 0" reponse options text / response values set has been selected.
    In this example, the value "1" is rather obvious, but it may not be so in all cases: "No, absolutely not", "Nah, not really", "Well ...", "Perhaps a bit", "Sure thing!" coded as "NAN, NNR, W, PB, ST" isn't that obvious. The CRF itself - when viewed in OpenClinica - doesn't necessarily provide enough information to shed light on the mystic "NAN" and "W" values in the data set, as these may not at all correspond with the textual representation of the CRF.
    I have found a place where these option / value sets are visible, deep down in the Build Study, but this isn't available when you're a data specialist or an investigator, so how is the data specialist informed about these option / value sets? I cannot find an option to add this of the data set when extracting data?
    Best regards,
    Janus
  • kristiakkristiak Posts: 1,301 ✭✭✭
    Hi Janus,

    Exactly, that’s why we provide decoding lists that we generate from the Excel sheets. There are obviously many different ways go get to these values, and I agree that it would have been nice to have a simple way to get them in the pre set export files. Maybe this can be added as future request!

    Regards

    Krister
    Sent: 05 January 2012 09:52
    To: [email protected]
    Subject: Re: [Users] Extract of response options

    Hi Krister,
    Thank you for your reply.
    I fully understand why it's the coded values being exported, but I'm just curious as to why these response option sets aren't exported as well in the .csv export? The response option sets could just be exported in a separate file? The statistician can't do much with coded values if she hasn't a clue to what information the coded value represents.
    Best regards,
    Janus
    "Krister Kristianson"
    Sent by: [email protected]
    04-01-2012 14:29
    Please respond to
    [email protected]

    To


    cc

    Subject

    Re: [Users] Extract of response options


    Hi Janus,

    What we usually do is to provide the code lists manually. The reason that the numbers are displayed and not the decoded values is simply for statistical purposes. A statistician could not do much with the decoded information. If you need both you can find this information also in the XML files.

    Regards

    Krister
    Sent: 27 December 2011 15:13
    To: [email protected]
    Subject: [Users] Extract of response options

    Hi all,
    When extracting data from OpenClinica 3.0.4.2, I get the database values of the selected items from radio buttons etc. E.g. I get a "1" when the "Yes" from a "Yes, No" / "1, 0" reponse options text / response values set has been selected.
    In this example, the value "1" is rather obvious, but it may not be so in all cases: "No, absolutely not", "Nah, not really", "Well ...", "Perhaps a bit", "Sure thing!" coded as "NAN, NNR, W, PB, ST" isn't that obvious. The CRF itself - when viewed in OpenClinica - doesn't necessarily provide enough information to shed light on the mystic "NAN" and "W" values in the data set, as these may not at all correspond with the textual representation of the CRF.
    I have found a place where these option / value sets are visible, deep down in the Build Study, but this isn't available when you're a data specialist or an investigator, so how is the data specialist informed about these option / value sets? I cannot find an option to add this of the data set when extracting data?
    Best regards,
    Janus
This discussion has been closed.