Query Process - Generating, Research Informatics …

1 Research Informatics1 Query Process Query Process --Generating, Generating, reviewing , tracking , and reviewing , tracking , and resolving queries resolving queries Erik Jolles, Research Informatics ,Family Health InternationalResearch Informatics2 Data Query Process Data Query Process --An OverviewAn OverviewGenerate data discrepancies/queriesAutomatic vs. Manual queriesReview queriesInternal resolutionExternal resolution - Query processingTrack queriesResolve and receive queriesReview Query responsesApply resolutionDocumentationResearch Informatics3 Query Process Query Process Generating DiscrepanciesGenerating DiscrepanciesQueries or data discrepancies present potential data problems and propose resolutions to those queries Manual QueriesAutomatic queries through data cleaning / validation processManual discrepancies from flags during data entry.

2 QA checks Informatics4 Examples of Automatically Examples of Automatically GeneratedGeneratedQueriesQueriesMissing item valuesVariations from an expected rangeInconsistencies between values within forms and/or between multiple formsMissing or late CRFsDuplicate CRFsMissing Informatics5 Automatically Generated Query ExampleExample of a missing item number 2aAutomatically GeneratedResearch Informatics6 Examples of Manual QueriesExamples of Manual QueriesMissing Header InformationIllegible DataMissing Pages Blank CRFsInvalid Patient NumbersInvalid Center Numbers Invalid DatesForm Sequence Numbers (FSN)

3 That differ on multiple pages of the CRFsResearch Informatics7 Manual queries contManual queries messages written in the body of the CRF by the siteInformation incorrectly crossed out on CRFData that cannot be entered as shown on the CRFI tems that have two written answers but only one is requiredQueries that were returned without a signature and date, or answered incorrectly at the siteResearch Informatics8 Manual Query ExampleManually generatedTwo written answersResearch Informatics9 Automatic Query Automatic Query vsvsManual QueryManual QuerySite manual queryAutomaticManualResearch Informatics10 reviewing QueriesReviewing QueriesReviewing is the Process of comparing data on CRFs to data in the database If the problem reported on the Query is valid, then the Query is sent to the site for resolutionIf the problem reported on the Query is a result of a data entry error or another valid reason.

4 Then the Query will not be sent to the Informatics11 reviewing QueriesReviewing QueriesBefore reviewing individual queriesConfirm that the program or error spec that generated the queries is correct Review the frequency distribution of new querieszIf the same Query was generated for each participant, perhaps there is a problem with the Query specification or the program code that generated the queryResearch Informatics12 Review Each Individual QueryReview Each Individual QueryDetermine if Query is correct Are there handwritten notes on the CRF that will help resolve the Query ? Internal resolutionDo multiple queries relate to one problem?

5 Link queriesResearch Informatics13 Report ExamplesReport ExamplesNumber of New (N) QueriesNumber of New (N) QueriesReport: GSS_9726/ Frequency Distribution by All Query StatusSTATUSCOUNTACD 10CU 2N 1898 Research Informatics14 Report ExamplesReport ExamplesFrequency Distribution of New Frequency Distribution of New QueriesQueriesReport: GSS_9726/ Frequency Distribution by QueryData TableNameFrequencyFollowUpReport1 10 Report2101111 FinalReport53 Report758 Research Informatics15 Multiple queries (3) related to one problemResearch Informatics16 tracking QueriesTracking QueriesAllqueries are tracked through the system database regardless whether they are automatic or manual Informatics17 tracking QueriesTracking QueriesEach Query has a unique identifier (number)Track the following information.

6 Site Query is sent toDate sent to siteDate received from siteDate Query is resolvedResearch Informatics18 Example of tracking QueriesDate sent to SiteDate received from siteDate Query is resolvedSite Query is sent toResearch Informatics19 Received queries From SiteReceived queries From SiteWhen queries are received back from the site they are processed and date stamped. They are then forwarded to the Query Manager for resolution through the system. Each Query has a unique Discrepancy Id number that is used to locate that discrepancy in the Query responses received from the data collection site are no further clarification is required, the database and the CRFs are updated based on the responses on the DPSFs.

7 Research Informatics20 Example of Received QueriesDate StampedResearch Informatics21 GCP ConnectionGCP Any change or correction to a CRF should be dated, initialed, and explained and should not obscure the original entry. That is, an audit trail should be maintained. This applies to both written and electronic changes or Informatics22 resolving QueriesResolving QueriesQueries are resolved by updating the database with the information recorded on the DPSF from the site. The Query status is DPSF from the site, initialed and dated by the Query Manager, is attached to the Informatics23 resolving queries resolving queries DPSFDPSFI nitialed and dated by Query managerResearch Informatics24 GCP ConnectionGCP (c) When using electronic trial data handling and/or remote electronic trial systems, the sponsor should ensure that the systems are designed to permit data changes in such a way that the data changes are documented and that there is no deletion of entered data ( , maintain an audit trail, data trail, edit trail).