ODM Import into CRF with 2 repeating question groups

Hello,
I encountered this problem the other day and would welcome any thoughts. I want to import data into an OC CRF which contains two repeating question groups; it is a diabetes patient diary, one group is for recording insulin, the other glucose. The number of records in the repeating question group will equal the number of recordings for a given day, but the number of glucose records is usually more than for insulin. The CRF/event is repeated for each day.

I suceeded in generating the ODM xml file and imported it into my test environment without any grumbling. The problem was that the number of repeats in each of the two repeating groups was the same and as many as the larger of the two groups. This usually lead to more insulin records than were required.

I got around this by importing the two groups separately from separate files. I don't know if this is just the way to do it, or this represents a design flaw in my CRF (i.e. I might have held insulin and glucose in separate CRFs, bundled together in the same event), or some other. If anyone has any thoughts/advice, I'd be very grateful.

Yours,
James Heywood
Tagged:

Comments

  • lindsay.stevenslindsay.stevens Posts: 388 ✭✭
    via Email
    Can you give an example? I don't understand the study design, or what the
    problem is.
  • jjnh2jjnh2 Posts: 5
    I will try, and have attached a pair of screenshots to (hopefully) help.

    I have a study which has a single, repeating event (repeats are not fixed in number). This event has a single CRF, with 2 pages (or 2 that concern us here). Page 1 contains a repeating question group for insulin, page 2 has a repeating question group for glucose readings - I am collecting data from insulin diaries for diabetics. In an ideal world, the data would look like it does here: (image 'page1-page2').

    However, when I make the xml file with both glucose and insulin data present, the result is as follows

    i.e., there are more insulin records than their should be. In all cases, the number of records in pages 1 and 2 repeating groups for a given event (I.e. a day) is the maximum for either or insulin or glucose for that day. In this case, it is 8 (instead of 5 and 8. Over each separate event, that number will vary.

    I got around this by importing the insulin (page 1) and glucose (page 2) data separately. Consequently, there is no problem, as such, but I wondered whether the fact that I had to look for a workaround was symptomatic of something (and that something could very well be that my design is inappropriate) and wondered if anyone had any thoughts on the matter. For example, why does it happen like this?

    Thanks again.

    James


  • lindsay.stevenslindsay.stevens Posts: 388 ✭✭
    via Email
    OK, thanks. Assuming your XML is written correctly, that shouldn't happen.
    Can you post an example XML that causes the problem?
  • jjnh2jjnh2 Posts: 5
    Many thanks for your time. Here is what the result looks like in the GUI (for 1 of 2 days, i.e. 2x repeating events). ,

    and the file is here:
    https://drive.google.com/file/d/0B2j7VPr8NZgLVFYzMzVpb3Zfd2c/view?usp=sharing

    (The attach file function didn't work for this, though it did for the images. I wasn't sure what the best way to post it instead, I hope this is sufficient/appropriate).
    Thanks again,
    James

Sign In or Register to comment.