PrevPrev Go to previous topic
NextNext Go to next topic
Last Post 31 Jul 2013 12:00 AM by  SuperUser Account
tblBAS SEROHOW codes
 6 Replies
Sort:
You are not authorized to post a reply.
Author Messages

SuperUser Account



Basic Member


Posts:289
Basic Member


--
03 Jan 2013 12:00 AM

    In the DAD study we have a fourth option for the SEROHOW variable in tblBAS


    4 = First pos HIV-1 test


    I would very much like the "First pos HIV-1 test" option as part of the SEROHOW variable. If this could be added to HICDEP 1.70 I would appreciate it.


    I would suggest placing the "Other" option just before "Unknown" - In this case 8:

    1 = Midpoint between last neg. and first pos. HIV-1 test

    2 = Lab evidence of seroconversion

    3 = Seroconversion illness

    4 = First pos HIV-1 test

    8 = Other

    9 = Unknown


    As a general note I think we should consider making a change to how 'other' is coded in various tables. For example in cases where the code is "99=Unknown", I would suggest making "98=Other". This would preclude having to keep giving the 'other' option a new number based on actual codes already in the list. Perhaps we can consider this if and when new codes are being added to lists rather than run through everything at the same time.


    SuperUser Account



    Basic Member


    Posts:289
    Basic Member


    --
    08 Jan 2013 12:00 AM


    I have updated HICDEP 1.70 with the suggested changes to SEROHOW: 4 = first positive HIV-1 test, 8 = other (previously 4 = other).

    SuperUser Account



    Basic Member


    Posts:289
    Basic Member


    --
    09 Jan 2013 12:00 AM


    Thank you

    SuperUser Account



    Basic Member


    Posts:289
    Basic Member


    --
    18 Mar 2013 12:00 AM


    Since we last had this discussion there was a TC that removed the "4= first positive HIV-1 test" as a possible coding.

    subsequent to this change there has been a COHERE merger where some questions arose from cohorts how to code the SEROHOW variable in different circumstances.
    One of the queries was how to code it if indeed the seroconversion used the first positive HIV-1 test as a proxy.
    The reply given to the question was :

    "If date of seroconversion is not known/cannot be estimated (the majority of clinic patients), then SEROCO_D should be the first positive HIV test (as in previous COHERE mergers), and SEROHOW should be coded as 9 (unknown)."


    I disagree that this should be coded 'unknown' as we DO know the proxy used. Instead I would like the "First positive HIV test" to be re-introduced as a possible coding.

    SuperUser Account



    Basic Member


    Posts:289
    Basic Member


    --
    25 Jul 2013 12:00 AM
    Why don't we use the variable SEROCO_D for collecting only the date of seroconversion(if HIV positive individual is a seroconverter), and introduce a new variable in HICDEP 1.70 HIV1_POS_D for collecting the date of first positive HIV test?

    As a consequence the variable SEROHOW is only requested if the patient is a seroconverter. The codings for SEROHOW could then be limited to

    SEROHOW (How was seroconversion date determined?)

    1=midpoint between last neg/first pos test

    2=lab evidence of seroconversion

    3=seroconversion illness

    4=other

    9=unknown

    SuperUser Account



    Basic Member


    Posts:289
    Basic Member


    --
    30 Jul 2013 12:00 AM


    I would suggest to modify the variable name HIV1_POS_D into HIV_POS_D to capture either the first HIV1 or HIV2 positive test date.

    SuperUser Account



    Basic Member


    Posts:289
    Basic Member


    --
    31 Jul 2013 12:00 AM


    I have modified the wiki page of http://www.hicdep.org/wiki/Hicdep_1.70/TableBas to include HIV_POS_D and reverted the SEROHOW coding to what we had in hicdep 1.60.
    You are not authorized to post a reply.