Making Lemonade out of Lemons

The “OMG” call came on Monday afternoon. On Thursday that week we were at our client trying to make sense of a documentation mess that they had regarding two consecutive releases of an Interactive Web Response (IWR) system that supports a prominent client (the sponsor). In a nutshell, 1) there was no supporting documentation regarding the testing activities that were performed, 2) there were no document sign-offs, 3) there were no plans or reports for either release and, 4) there was no comprehensive summary of what information was used for testing. But there was one impatient sponsor who needed reliable assurance that the IWR system used for their clinical studies operates as intended.

In less than five days, we 1) recreated the profiles of the test patients, 2) organized available evidence into a suitable validation package to show that, while informal, testing did occur in advance of placing each release into production and 3) identified and addressed, in writing, all of the existing documentation gaps. No document created during this effort was backdated and all of the documentation-related issues were fully disclosed to the sponsor. The final deliverables for each release were well received by the sponsor and accepted with no additional questions or comments.

How did we manage to get such positive feedback from the sponsor when the starting point was so far from being regulatory compliant? The secret ingredient was reconstructing the testing that was performed. We did this by compiling detailed, system-generated test evidence, stored on the server in electronic form, down to dates and times, which allowed us to prove that 1) testing was performed contemporaneously and against the sponsor-approved requirements, 2) documentation gaps did not impact the integrity of data collection, processing, etc., and 3) the randomization feature was working accurately.

Not only was our unconventional approach suitable and practical for providing confidence that the system’s validation effort can be defended in the eyes of the regulators, but it was also creative and efficient with respect to using the available means to assemble the validation documentation suite. So who said it is not possible to make “lemonade out of lemons” in the regulated environment?