In CRO industry, statistics work with very large data sets that can be inaccurate with smaller data and it works better with larger dataset. For the clinical trials, sites require certain number of patients to be able to give a statement that A caused B.
Screen failure ratio:
Site performing phase three trial and are enrolling three thousand patients in trial, it’s very possible to have a fifty percent screen failure ratio over the course of three thousand patients.
On the contrary, it’s also very possible for the sites to get one hundred percent screen failure ratio when they are enrolling very few number of patients such as four patients for the trials and they all are screen failures which is the perfectly normal random behavior. These sites enroll four patients that can easily affect exclusion criteria due to their anemic condition.
Ø This shows the reality of statistics on sample size that can be different for both small and larger sample size.
Health of patient population:
If the sites have sicker patient population for their study trials, they will have more effects on their results and data and sicker patient population on a small sample size will have more possibility of triggering inclusion-exclusion criteria.
Ø The healthier the patient population, more this criteria is muted.
Sites doing vaccine research on healthy people will get small screen failure ratio whether it’s on small number of patients. On small sample size, sites will have more predictable enrollment, randomization and revenue.
Ø Sites will have more volatility on their enrollment, randomization and revenues in case of sicker patient population.
Sites should recognize the sources of volatility to their enrollment predictability and revenue predictability in order to take necessary measure against them.
Sites can diversify when they are in a larger number of medical specialties that can result in screen failure ratio from 8% to 80% across all trials, across all specialties. An eight percent screen failure ratio in certain medical specialties, certain types of trials, and eighty percent plus screen failure ratios in others.
Diversification can help sites to balance their screen failure ratios across all of their trials and the entire company for average of fifty percent. There can be volatility on a site by site basis. Sites should also be able to counter when Sponsors accuse them for screening inappropriate patients because site’s screen failure ratio is so high.
Sites can also make a mistake for screen failing inappropriate patients where there is a need for them to learn and change their behavior. They should also educate their CRA and the project manager on what they have been doing to make screening completely appropriate.
Thinking in Terms of Statistics:
Sponsors and CROs should think in terms of statistics on small sample sizes. The industry can play better on the same team by recognizing the effect of the small sample sizes on sites that is getting smaller with time.
As the sample sizes are getting smaller, the industry is running protocols with smaller numbers of patients and is opening up more sites per protocol to get fewer patients per site which means that the sample size per site is going down.
For the healthy research site and healthy relationship between CRO industries, it’s vital to recognize the impact of these small sample sizes.