Project Management in Clinical Trials. Alexey Levashov

and start planning. Those who will fulfill the plan should be involved in planning. It is not only a matter of respect, but also a matter of validation of assumptions and engagement – people prefer to realize plans that were created with their participation.

      There are a lot of excellent tools for scheduling including MS Project and many others with detailed guidance on how to use them, so let’s not waste time on this topic.

      I recollect my conversation with a professional leading start-up activities in a country. His boss promised to the client without prior consultation with him that a submission will be performed in a much shorter period of time than it could be done at all. As a result the promise was not kept and the start-up lead was somewhat demotivated. But what is more important, he did not even try to shorten the timelines for this submission because of his disengagement.

      But even if a plan in a clinical trial is written, it is only used as a collection of rules and norms in the vast majority of cases. It does not serve its main purpose – setting goals.

      Let’s take the essential part of an enrollment plan for a mock study as an example (as per scenario, the very last sites are activated in February and March):

      Table 1. Enrollment plan for a mock study

      What is so good about this enrollment plan? The numbers differ from month to month. If your tool shows the same enrollment in August as in April (given one and the same number of active sites), there is something wrong about it. The list of reasons for different intensity of enrollment includes but is not limited to:

      1. Seasonality of the disease: it is easier to enroll patients with community acquired pneumonia during winter period, with the flue – during epidemics of this disease (February-March normally) and so on.

      2. Holiday seasons:

      – Christmas and New Year – end of December – very beginning of January in the vast majority of countries, beginning of January (about 10 days in a row) in some countries (Russia, Ukraine);

      – Easter;

      – Periods of long national holidays like at the beginning of May in Russia, Ukraine, and Bulgaria;

      – High holiday season (July and August) in Europe.

      3. Site tiredness: centers that enroll for more than a year tend to become tired of a study and loose or shift their focus. At the same time, opposite examples are also possible, when while enrolling sites become more trained, proficient in a study and enroll better and better month by month with no signs of tiredness.

      But we need to return to our enrollment plan itself. We might have a good plan, created lege artis, with involvement of all stakeholders, using the latest predictive technologies, but not working. Where is the disconnect?

      The problem is that a plan does not translate into concrete goals for country teams. To do so we need to take our projections for the next month, for February, for example:

      Table 2. Original enrollment projections for February in a mock study

      and ask relevant country team members (Clinical Team Managers, CRAs, and so on) to provide their feedback before the end of the previous month – January. As a result we may learn that personnel of most sites in Germany will not be available for one week in February because of a congress and we need to reconsider the projections and to set the country goal for the month as 3 randomized patients. At the same time, in Hungary there may be a big pool of prescreened patients and we may project more: 9 patients rather than 7. In Russia there was a delay in site activation, and in Ukraine, on the opposite, all sites were initiated earlier than expected, so for these countries we should change the targets to 12 and 15. Once we have all necessary feedback, we set the monthly target for all countries – they are quite different from our original plan on the country level:

      Table 3. Corrected enrollment projections for February in a mock study

      Please note that if screening duration exceeds one month, country-specific targets should be expressed in the number of patients screened rather than randomized: the events counting toward the goal (screening or randomization) should be the result of activities performed during the month in question, for which the target is set.