This just really doesn't make sense for a lot of research. Most science isn't this kind of "Does chemical X do Y?"- these kind of clinical trials and safety reviews are not really what most scientists are engaged in. There could be some centralized system (do you know that clinicaltrials.gov already exists and does this for pharmaceutical research?) for safety and efficacy experiments, but most science goes more like "let's figure out how the heck this thing works!" and are an investigation involving hundreds of constantly changing mini-experiments that'd be impossible and counterproductive to predict beforehand.