Clinical trials are designed to make sure that new therapeutics are both safe and effective. They can also be used to identify side effects, to compare how well different drugs work relative to each other and to see if certain populations react differently to different treatments. In order for doctors to prescribe the most appropriate drugs to their patients, they need to know the results of such clinical trials. Unfortunately, that information is not always so easy to come by.
Publication bias means that negative results generally do not get published. This is problematic because it skews the publication record. If only positive results get published, showing that a given drug is effective in assuaging a certain condition, people assume that that is the full story. Even if ten studies have been done showing that that same drug is useless, since negative data does not usually see the light of day no one knows about them and people think the published positive results are "fact." Approximately half of all clinical trials performed globally have never been published in academic journals, and trials with positive results are twice as likely to be published as those with negative results. No one wants to publicize that their drug doesn't work. Because if doctors don't know that a drug doesn't work - or a more realistic scenario, that a new, expensive drug doesn't work better than the old generic - then why on earth wouldn't they prescribe that drug to their patients? Moreover, it has been perfectly legal for pharmaceutical companies and universities to withhold the results of clinical trials as proprietary information.
To mitigate the misperceptions caused by publication bias and the withholding of trial data by the pharma industry, the Food and Drug Administration Modernization Act of 1997 created ClinicalTrials.gov. All clinical trials with at least one testing site in the US are supposed to register there before the trial starts. It went online in 2000 but only really became a force in 2005, when the International Committee of Medical Journal Editors made registration a prerequisite to having a trial published in a journal. Since researchers must register before the trial begins, they must lay out their initial hypothesis and thus cannot "move their goalposts" - claim to have always been looking for whatever it was they found. In 2007, the FDA added the requirement that results must be published on the site within a year after a trial is completed. Thus even if results are not published in journals doctors and patients have another place to search for them, and it should, in theory, be more difficult for researchers to hide negative results, since there is a record of the trial having taken place. However, neither the requirement to register trials nor the requirement to report results have been rigorously enforced or followed. So often not only do doctors still not know the results of trials - they might not even know that a trial has been done.
On April 2, 2014, the Members of the European Parliament voted to adopt the Clinical Trials Regulation. This regulation makes it law in the European Union that clinical trials be registered before they begin, that results be published somewhere within a year after the trial ends, and that a summary of results written in lay terms be published on the publicly accessible register. Failure to comply with these new requirements will be punishable by a fine. It also dictates that information contained in Clinical Study Reports will no longer be considered commercially confidential. These reports contain many details that are often omitted in academic papers but are nonetheless important, like research methodologies.
This new European law is expected to come into effect in mid-2016 at the earliest. It is an enormous stride forward, but most of the medicines currently in use went through trials that have already been done. Results of these trials can still legally be withheld, so doctors must still make prescribing decisions without complete, accurate, and up-to-date information about which drugs now available are best for which patients.
Those with rare diseases can be particularly impacted by the transparency, or lack of it, in clinical trials. Pooling results from different studies into meta-analyses can often reveal the most telling effects of a drug; since fewer people have these disorders fewer studies can be done, and thus withholding data from any one of them can thus have an outsize effect. Moreover, subjects who participate in such trials often do so to benefit their fellow patients in addition to themselves, and withholding the results that they helped provide is a betrayal of their trust.