Author Topic: Data validation procedures  (Read 2714 times)

0 Members and 1 Guest are viewing this topic.

Offline Rock4mcal

  • Forum Member
  • Posts: 14
  • Respect: 0
Data validation procedures
« on: Sun, 31 Aug 2008 »
0

CRS

  • *
  • Newbie
  • Posts: 377

  • Thank You
  • -Given: 34
  • -Receive: 167
    • View Profile
    • Email
    • Personal Message (Offline)

Sponsored Links

Hi All,
Can someone throw more light on Data Validation procedures?
What is a validation?
What is a standard procedure?
How does a validation work?
What is a project specific procedure or Validation?
What is the backend process? how is a validation programmed?

Thanks a lot in advance for your help...

Offline ilikecdm

  • Global Moderator
  • Loyal Member
  • *
  • Posts: 1793
  • Gender: Male
  • Grown men do not need leaders!!
  • Respect: +3
Re: Data validation procedures
« Reply #1 on: Mon, 1 Sep 2008 »
0
Dear Rock,

Are you doing any project on this topic?

I Like CDM  :)
"The postings on this site are my own and do not necessarily represent my organization's positions, strategies or opinions."

Offline Rock4mcal

  • Forum Member
  • Posts: 14
  • Respect: 0
Re: Data validation procedures
« Reply #2 on: Wed, 3 Sep 2008 »
0
Hi I Like CDM,

I thought it would provide more of understanding of CDM as a whole.

As this would be one of the important aspects in cleaning of the data...

Offline ilikecdm

  • Global Moderator
  • Loyal Member
  • *
  • Posts: 1793
  • Gender: Male
  • Grown men do not need leaders!!
  • Respect: +3
Re: Data validation procedures
« Reply #3 on: Wed, 3 Sep 2008 »
0
Dear Rock,

Check out this presentation. I got this from net. Thanks to original author of the presentation. Hope you will like it.

Regards,
I Like CDM

Offline Rock4mcal

  • Forum Member
  • Posts: 14
  • Respect: 0
Re: Data validation procedures
« Reply #4 on: Sat, 6 Sep 2008 »
0
Hi I Like CDM,

I checked the presentation. It is very useful...to understand the CDM process as a whole...

Thanks again for your inputs


Offline vetaccent

  • Forum Member
  • Posts: 88
  • Respect: +1
Re: Data validation procedures
« Reply #5 on: Sat, 6 Sep 2008 »
0
Hi Rock,

I just wanted to share what i knew and understood about Data Validation Procedures.

Data Validation Procedures: Validations are set of programs to validate the data entered into the database (both paper and RDC). It checks the data points (questions defined in the body of Case Report Form) whether each data point meets the criteria as defined in the program.

Validations can be of three types.
i)Procedures checking data points within the form (Age violates protocol range).
ii)Procedures checking data points between the same form (AE form Vs AE forms)
iii)Procedures checking data points between forms (AE reported with Hospitalization, validation should check is there any hospitalization reported for the specified AE in the Hospitalization form).

Standard Procedure: Procedures are of three types (as per my knowledge), standard, indication specific and study specific. Standard procedures are the ones which can be used across therapeutic areas - Example age is validated against protocol specification.
Indication specific procedure is the one which can be used for a specific therapeutic area - example: validations related to pain assessment scale in arthritis studies, i.e.  entered values should be within the range of pain assessment scale.
Study specific: Within the therapeutic area, for a particular study / for a particular drug, program has been changed to check the range specified in the protocol, which is a deviation from the Therapeutic are level. Example: Pain score scale range is 0 to 10 for Musculo skeletal studies, which has been defined by MS research scientists. If we use the value as 0 and 10 then it is Indication specific, if we change it to 0 to 8 (because for this particular study, protocol changed the range from 0 - 10 to 0 - 8), then it is study specific.

Validation works: After data entry, SAS programs took values stored in the questions (Example: Age) into their variables (Example: Agex), and validate it against the values defined for them in the data base. For example, in this case the min and max of agex has been defined in the study start up as 18 and 60, if the age is entered as 61, the validation will fail.

Project specific Procedure: This is otherwise called as project specific procedure (Explained above).

Backend process (I explained the little what i know above topic "Validation works").

Thanks a lot in advance: You are always welcome and its my pleasure to share the things what i knew.

Request: Please correct me if any of my understanding about the above topics is wrong and do share more about this.

Best Regards,
Arun  :)

Offline Rock4mcal

  • Forum Member
  • Posts: 14
  • Respect: 0
Re: Data validation procedures
« Reply #6 on: Sat, 13 Sep 2008 »
0
Hi Arun,

Thanks  a lot for your updates...

Your right about the validations...

I just want to added a few points... Validations will fire at two times...Once as soon as a discrepant data is entered called as online validation...this is generally seen in case of EDC studies. In paper based studies it is the Univariate discrepancies.
The second type is offline validations...these fire only after a succeful batch run. Batch run is the process where all the derived values get populated and the system checks all the data points both that are entered and those that are derived... Here again the validaition procedures are of two types VH and DH, the VH validations fire form the data points that are manually entered... whereas DH validations are from the data points that are generated by the system derived on the values that are manually eg. Lab ranges...

Please do correct me if I am wrong.


 :)

Offline vetaccent

  • Forum Member
  • Posts: 88
  • Respect: +1
Re: Data validation procedures
« Reply #7 on: Sat, 13 Sep 2008 »
0
Hi Wonderful, i really missed that!

Yes there are univariates and multivariates!

Also there are VH and DH, along with there is another set of valdiations are there, VS: 'S' stands for standard validations.

Thanks.

Best Regards,
Dr. Arun.

CRS

  • *
  • Newbie
  • Posts: 377

  • Thank You
  • -Given: 34
  • -Receive: 167
    • View Profile
    • Email
    • Personal Message (Offline)

Sponsored Links


 

Related Topics

  Subject / Started by Replies Last post
10 Replies
2815 Views
Last post Fri, 5 Mar 2010
by pnreddy
46 Replies
11427 Views
Last post Tue, 29 Dec 2009
by pitu.v.7
0 Replies
578 Views
Last post Fri, 2 Aug 2013
by shridhar




SMF 2.0.3 | SMF © 2011, Simple Machines

SimplePortal 2.3.3 © 2008-2010, SimplePortal