Artefact handling matters. But what they prove matters more.
[see also part 3 of this series]
The clinical-trial industry spends enormous effort moving documents between participants. Yet sooner or later a human reviewer still has to sit down, read the document, determine what it says, determine whether it says the right things, and determine whether those things are said correctly.
The previous articles argued that clinical research may be facing a challenge similar to the one international banking encountered before SWIFT. The industry has successfully digitised large parts of the Trial Master File and has established a common framework for organising documentation. At the same time, significant effort continues to be invested in review, reconciliation, interpretation, and quality control. I came to the conclusion that the industry’s primary challenge may no longer be document management, but rather achieving a common understanding of the information contained within those documents.
If that observation is correct, an obvious question emerges.
What exactly is being interpreted?
The instinctive answer is “documents.” Sponsors review documents. CROs review documents. Auditors review documents. Inspectors review documents. Much of the industry’s operational language is built around the management, exchange, filing, and review of documents.
Yet compliance decisions are rarely based on documents themselves.
They are based on the information contained within those documents.
A delegation log is not reviewed because it is a delegation log. It is reviewed because it contains information about who was authorised to perform specific study activities. A training record is not reviewed because it exists. It is reviewed because it contains information about qualifications and competency. An informed consent form is not reviewed because it occupies a particular section of the Trial Master File. It is reviewed because it contains evidence that consent was obtained appropriately.
In each case, the document acts as a container. The actual review activity focuses on the information contained within that container.
This distinction becomes important because documents are inherently heterogeneous. The same information can appear in different formats, different layouts, different templates, and different systems. A sponsor may use one template, a CRO another, and a site a third. The artefacts differ, yet the underlying information often remains remarkably similar.
Banking encountered a comparable challenge many years ago. Financial institutions exchanged confirmations, payment instructions, account statements, and numerous other operational documents. Each institution maintained its own internal records and formats. The industry gradually discovered that interoperability could not be achieved by standardising every document. Instead, it standardised the information that needed to be exchanged between participants.
The resulting SWIFT messages were not replacements for the underlying records maintained by banks. They were standardised representations of the information required for communication and processing. Each bank remained responsible for its own systems of record, while the network operated on a common representation that all participants could understand.
Clinical research may be approaching a similar point.
A delegation log maintained by a site is not fundamentally interesting because of its layout or formatting. What matters is what it demonstrates. Which individual performed which role? During which period? Under which authority? An informed consent form is not interesting because it occupies a particular section of the Trial Master File. It is interesting because it demonstrates that consent was obtained appropriately.
Similarly, an informed consent form is ultimately evaluated based on facts such as consent date, version, signatories, and study context. The document remains important, but the compliance assessment focuses on the information extracted from it.
Reviewers do not ultimately evaluate artefacts. They evaluate what those artefacts demonstrate.
Viewed from this perspective, many quality-control activities begin to look different. A reviewer rarely asks whether a delegation log exists. The real question is whether the delegation log demonstrates appropriate authority. A reviewer rarely asks whether an informed consent form is present. The real question is whether it demonstrates valid consent. The artefact provides provenance. The reviewer determines whether it proves the required assertion or obligation.
This may also explain why document-centric approaches struggle to scale. Every day, reviewers across sponsors, CROs, sites, auditors, and regulators read similar artefacts and derive similar conclusions. The artefacts may differ. The conclusions are often remarkably similar. Yet the industry continues to invest significant effort repeatedly deriving the same understanding from different representations of largely the same evidence.
The answer is not to abandon documents. Documents remain essential. They provide provenance, accountability, legal standing, and regulatory evidence. Systems of record will continue to store them. Inspections will continue to reference them.
The more interesting question is whether we have been treating two different activities as one. Managing artefacts is one activity. Determining what those artefacts prove is another.
Anyone who has reviewed a delegation log has experienced this distinction. The document itself is rarely the challenge. The challenge is determining what it is supposed to prove, and whether it actually proves it.
Such a distinction would not eliminate the need for human judgement. Nor would it eliminate the need for systems of record. It would, however, create the possibility that independent participants could operate from a more consistent understanding of the information they exchange.
That possibility is worth considering because it shifts the conversation entirely.
The next generation of clinical-trial governance may not be about managing documents more efficiently. It may be about evaluating, more consistently, what those documents are supposed to prove.
One Reply to “”