FDA validator rule ID SD1331

Sergiy

on June 18, 2020

P21

Is anything wrong with logic of SD1331 rule?

Hi Serhii, 

Is anything wrong with logic of SD1331 rule?

"Start Date/Time of Event, Exposure (--STDTC) must be less or equal to Date/Time of Collection(--DTC)"

Thank you,
Sergiy

What was FDA's motivation for took this rule off from the list.

Hi Sergiy.

I am just wondered why this inconsistency appeared. And what was FDA's motivation for took this rule off from the list.

Personally me faced with a specific and particular problem. In simple words:

Our data collection system and study design built in a way that, for example, the folder and folder creation date will be the same for all Inform Consents ever signed or withdrawn by a subject. Also a collecting of the folder creation date to DSDTC is a sponsor requirements. And we have the situation when date of INFORMED CONSENT OBTAINED (DSSTDTC) is less or equal to the folder creation date (DSDTC) but a date of when informed consent was WITHDRAWAL BY SUBJECT (DSSTCD) is more than the folder creation date (DSDTC).

But my example is particular. I am more interested if Pinnacle 21 uses a rules database created by using accumulating of all existing rules ever submitted or it uses the most recent edition.

Thank you in advance,

Serhii

Sergiy

on June 25, 2020

P21

You have incorrect SDTM mapping

Hi Serhii, 

You have incorrect SDTM mapping. Creation of CRF is not the same as data collection event. So, P21 validation identified a real mapping problem in your study data.

Management, maintenance, and publishing of validation rules is quite complex process. There are three major concepts of Data Quality:

• An issue is still an issue regardless if it is reported or not
• Data quality is defined by intend use
• All users are different

The last item is the most important. Each user has unique interpretation of quality data depends on how study data is utilized by this user. For example, data managers focus on data collection problems; data programmers focus on mapping issues they can handle; data warehouse specialists focus on specific standard requirements and structural consistency; safety reviewers focus on subset of safety data and if those data can support meaningful analysis; etc. Different regulatory agencies have different internal processes with different requirements for study data. A scope of Pinnacle 21 Community is to help users to prepare study data for regulatory submissions to FDA and PMDA. Note that submission processes to these agencies are different. PMDA has very formalized and straightforward requirements. Sponsor must use pre-specified versions of Pinnacale 21 validation engine, fix all Reject issues and explain all non-fixed Error issues. PMDA performs independent validation and reconciles their results with results reported by sponsor. In contrast, FDA utilizes the latest version of available rules in FDA DataFit (custom version of Pinnacle 21 Enterprise), which also includes additional analysis-specific rules. Often, FDA reviewers may execute additional study validation checks to identify data issues important to their specific review process. FDA study data quality evaluation process is less formalized and depends on reviewers assigned to your submission. “Less formalized” means that they focus on actual data issues rather than on some pre-specified set of business rules.

There are many stakeholders in validation rules development. The major players are CDISC (standard compliance), FDA/PMDA (regulatory review) and Sponsors/Vendors (custom). As I said before, rules in Community are about the CDISC and FDA/PMDA. Rules publishing process is very different across different organization. Official rules from FDA are quite outdated. There are many new and formal data specific requirements introduced in new versions of FDA Study Data Technical Conformance Guide, eCTD and other guidance documentation. We update Community validation rules with formal FDA requirements. We constantly work with both FDA and PMDA to refine validation rules (for example, see our recent webinar ) and provide them to our users.

In summary, Pinnacle 21 works with all our clients for creation of validation rule to support their internal processes. We work with FDA and PMDA teams to select and refine a set of rules to help the industry to prepare study data for regulatory submissions. Note that any set of validation rules is never final. An issue is still an issue regardless if it is reported or not.

Kind Regards,
Sergiy

Thank you for explanation

Hi Sergiy,

I understand that creation of CRF record creation date is not the same as the date of event occurrence. Incorrect mapping appears as due to sponsor specific rule that require to map a CRF record date into the DSDTC as  well as specific of data collection system used. It seems like this oddity could be explained in SDRG.

Thank you for explaining the principles so widely. Also thank you for the webinar referring.

Best regards,

Serhii

Sergiy

on July 6, 2020

P21

Hi Serhii, Date of CRF…

Hi Serhii, 

Date of CRF creation is not the same as date of CRF completion or data collection date. So, you have incorrect SDTM mapping. It would be difficult to explain in Reviewer's Guide, that you are intentionally not SDTM compliant. In general, a date of CRF creation is operational information which is not relevant to regulatory review process and analysis of study data. So, this information should not be included into SDTM submission-ready data.

Your case of invalid utilization of DSDTC variable is a good example of potential risk of new guidance from CDISC. According to CDISC new validation rules for SDTM-IG 3.3, users are not required anymore to follow standard variable labels. I believe that utilization of custom labels for SDTM-IG variables will increase the misuse of CDISC stadnards.    

Kind Regards,
Sergiy