|
The implementation of Clinical Data Interchange Standards Consortium (CDISC) standards is proceeding at various rates across the pharmaceutical industry, and there seem to be nearly as many approaches as there are organisations. This article will discuss one of the practical aspects of implementation, namely the mapping of source data structures to study data tabulation model (SDTM) and analysis dataset model (ADaM) standards, and examine why some contract research organisations (CROs) have been slow to invest in technical solutions to expedite this process.
According to a study conducted by CDISC, Tufts and Gartner, the lack of standards implementation is costing, on average, an estimated $9 million per trial in direct costs, and reducing the potential to shorten the time taken for clinical trials by 60 per cent (1). Of most interest to CROs is the second figure; a great deal of the direct costs to a CRO’s bottom line is taken up with the expensive process of converting source data to CDISC SDTM and ADaM standards.
This process generally involves a complex restructuring of data in a way that must be transparent and validated; a reasonably sized Phase III study can take around six months to convert if done through the traditional handcranked code method. This would be a validated process, but the provenance of each item of data is not immediately clear, and clarity around derived items is something the FDA is particularly keen to ensure. |
