How to Run a GLP Multi-Site Study
Split a study across sites without splitting responsibility for it.
What a how-to is not
A how-to is SPEQ’s practitioner method, not a procedure. It does not replace your own SOP, it is not a validated approach, and the judgement calls in it belong to your quality unit.
Modern nonclinical studies routinely run phases at different facilities, sometimes in different countries under different monitoring authorities. Good laboratory practice accommodates this through a specific arrangement: phases are delegated to principal investigators at test sites, while the study director retains overall responsibility for the study and its report. The arrangement fails when delegation is read as transfer.
- 1
Name the phases and the sites in the study plan
The plan identifies which phases run where and who the principal investigator is at each. Sites added by correspondence after approval, without amending the plan, leave a study whose documented scope does not match its conduct.
- 2
Appoint a principal investigator for each delegated phase
The principal investigator acts for the study director for that phase and is responsible for its conduct at that site. The director does not become a coordinator: overall responsibility for the study and for the report does not move.
- 3
Agree how quality assurance will cover each site
Each site’s unit inspects its own phase and issues a statement covering it; the lead unit needs enough visibility to assure the study as a whole. Assuming the other site’s unit has it covered, without an agreement saying so, is how a phase ends up uninspected.
- 4
Define raw data ownership and transfer for each phase
State what stays at the test site, what transfers, and what the lead facility archives. Raw data scattered across sites with no map is retrievable in principle and unfindable in practice when a reviewer asks.
- 5
Route deviations to the study director without delay
A deviation at a test site is a deviation in the study. Local resolution followed by a summary at the end deprives the director of the ability to act, and it is the most common practical failure of the arrangement.
- 6
Consolidate into one report with contributing reports attached
The principal investigators sign for their phases; the study director signs the study. Contributing scientists’ reports are attached rather than paraphrased, so the reader can see the phase evidence as its author wrote it.
- !Delegation read as transfer, leaving the study without a single point of overall responsibility.
- !Sites added after plan approval by correspondence rather than by amendment.
- !Quality assurance coverage of a delegated phase assumed rather than agreed, so no unit inspects it.
- !Test-site deviations resolved locally and reported in summary at the end.
How to Run a GLP Multi-Site Study: frequently asked questions
Common questions on run a glp multi-site study.
Does the study director stop being responsible for a delegated phase?
No. The principal investigator is responsible for the conduct of that phase at that site and acts for the director there, but overall responsibility for the study and for the final report stays with the director. That is the whole design — delegation of conduct, not of accountability.
Who inspects a phase run at another facility?
That facility’s quality assurance unit inspects its own phase and issues a statement covering it, and the arrangement is agreed in advance so the lead unit can assure the study as a whole. Coverage that is assumed rather than agreed is how a phase goes uninspected.
Where do the raw data live?
Wherever the plan says, provided the plan says. Some raw data properly stay at the test site under its archiving arrangements. What is not acceptable is an undocumented distribution, because a study whose raw data cannot be located on request is functionally unreconstructible.