Every sponsor has different things that are important to them with regards to clinical monitoring and site visits.
In some cases, cost is a driving factor for sponsors when engaging clinical site monitors, as visits are one of the most expensive parts of a trial. We are often asked whether our trial monitors are regionally based, as this may be less expensive. In other cases, the number of site visits is critical to a sponsor. Generally, clients just want to be sure that we conduct visits as they are needed. Even if we are contracted to go every 6 weeks, if we determine we need to go more often due to training issues or higher enrollment, the client wants to be sure we are flexible and can do that.
When developing a monitoring plan, there are several considerations when determining the number of site visits. First, what did the client pay us to do? If the client feels that we need to go out to a site every 2 weeks and we have explained our philosophy and they disagree and are paying us to go out every 2 weeks, we absolutely will do just that. Secondly, this is a tough economy and we need to find the right balance between cost and quality for our clients. If we can still deliver the same level of quality (which I believe we can) through remote monitoring and less visits, we should certainly demonstrate that to the client. Finally, we also consider the complexity of a trial and the enrollment at each particular site.
My opinion is that visiting sites based on time elapsed is an outdated methodology. This model tends to over-monitor sites that don’t have the enrollment and under-monitor the sites that do have the enrollment.
To what should you time site visits, instead? This truly depends on the study. For all of our studies we make an initial visit very soon after the first or second patient is enrolled to establish a baseline for the site. After this initial visit, the trigger could be a number of things. When remotely reviewing the data, does it indicate that the site needs more training either in the EDC system or in the protocol? If so get out to that site quickly; the errors you correct today pay off exponentially down the road. Does the site seem to have a higher number of adverse events than expected? You may want to go on a visit. Is the site a quick enroller? Enrollment rates obviously should be taken into consideration. If they are a quick enroller you don’t want to get behind in the source data verification and you want to really be on top of the sites processes because they may be overwhelmed with the fast enrollment.