Reconciling the data between clinical and safety databases is a tedious process requiring coordination between multiple departments such as clinical, safety, programmers etc. In this study we investigate the feasibility of building a framework to reconcile data with minimal human intervention.
Method
The experiment involved:
Identification of elements that need to be reconciled between clinical and safety databases
Extraction of elements from both clinical and safety systems into a single repository
Surfacing the datasets and highlighting mismatches which can be exported into excel and PDF
Results
A sample of over 100 case records were created for two studies in the clinical database, some of these cases were pushed to safety database. In the safety database the details such as patient initials, age, gender, suspect product, event verbatim, event serious, event onset date, event cessation date, preferred term, outcome, causality etc were updated for these cases. All of this data was imported into single repository (Safety Datamart). Subsequently this data was surfaced to users to perform streamlined reconciliation of mismatches. The safety gateway reconciliation tool (SGR) was not only able to highlight and but also enumerate the mismatches for all the elements individually with separate color coding to facilitate ease of review. In addition the SGR tool was also able to export the results into two of commonly accepted formats of Microsoft excel and PDF.
Conclusion
Based on the review of multiple contracts between our company and various clinical research organizations on an average the serious adverse event reconciliation process for a single study requires about 24 excel spreadsheets, 200 man hours and costs $25000 per year. The Safety gateway reconciliation tool (SGR) was able to eliminate about 1320 excel spreadsheets, 11000 man hours and realized $1.3 M in savings per year. The SGR Tool was able to transform the reconciliation process into an end-to-end solution in building the framework which is feasible, enabling easy assistance and ensuring a single “source of truth”. The efficacy of the tool strengthened our clinical and safety practices providing listings rationalizing the process optimization.
References:
Barry Burnstead, Giovanni Furlan. Unifying drug safety and clinical databases Pub Med:2013, PMID:23656448
