The most appropriate action to take in response to monitoring findings depends on what issues are found. It could include more training or feedback for trial sites or CTU staff, or amendments to the protocol, statistical analysis plan or data collection forms.
If monitoring shows that the originally expected frequency of participation changes is different to the observed frequency, action may be required to amend the protocol or analysis plan content about handling participation changes, or even to amend the trial’s sample size.
Trial statisticians can advise on prospective thresholds to indicate where the frequency of participation changes might be high enough to impact on trial integrity.
Site-level thresholds can be used to target site-specific monitoring, e.g. via a ‘triggered monitoring’ approach.