33% of clinical trials have problems with randomization, statistical analysis and patient recruitment. AI assists in several bottlenecks.
A clinical study is a scientific investigation designed to evaluate the safety and efficacy of a medical treatment or intervention in humans. Clinical studies may involve patients, healthy volunteers, or both, and are carried out to determine if a medical intervention is safe, what side effects it may have, and if it is effective in treating a particular disease or condition (1). Clinical studies can have different designs, including randomized and controlled studies, in which the results of a group receiving the treatment are compared with the results of a group receiving a placebo or a different treatment (2). They can also be phase I, II, III, or IV, depending on the objective and developmental stage of the treatment or intervention (3). A study conducted by researchers at the University of Toronto found that around 33% of randomized clinical trials published in major medical journals had problems with randomization, blinding, or statistical analysis, which can affect the validity of the results (4). And the NIH showed that 33% of clinical trials registered on the platform failed to recruit enough participants, meaning that many studies were either not completed or significantly delayed (5).
There are several bottlenecks that can affect the development and conduct of clinical studies. Some of the main ones include: