Almost all of us, at some point or another have had some kind of physical ailment and taken regulation-body approved chemical therapies: drugs. These drugs don’t just spring out of the ground, however, but are the result of a long and arduous research and development process undertaken by thousands of scientists. How this research goes down can have major effects on the quality of the pills we end up popping.
Drug research happens in three major stages: basic research (looking into a biological process), preclinical drug studies (studying the effect of a compound on a condition but not in humans), and clinical trials (studies in people to assess safety and efficacy). The cost of drug research is undertaken by both public funding like government research grants; and private funding, like that from drug companies or “big pharma.” There are, of course exceptions, but for the most part public funding is concentrated in early research with pharmaceutical companies taking much of the responsibility for funding later in the process, including clinical trials.
Most of the clinical drug trials that are taking place are directly or indirectly funded by drug companies. Direct funding includes, for example, a grant specifically funding a drug trial granted by a drug company. Indirect funding includes financial conflicts of interest involving the company and one or more of the principal investigators. I should mention that “conflicts of interest” do not preclude researchers from doing the work, but rather must be declared when the research is being reported to ensure transparency.
What we have to consider as the drug-consuming public is the consequences of large pharmaceutical companies funding drug clinical trials. A meta-study looking at literature examining this very topic was recently published in two parts in the Deutsches Ärzteblatt International.
This group found that the consequences of Big Pharma funding were, in fact, discernible, and that they fell into several categories. The first category is the actual outcome of the study, or whether or not the drug passes the clinical trial stage and is licensed for sale. As it turns out, studies funded by “big pharma” tend to have more favourable outcomes for the company. Statistically, more drugs in privately funded trials get approved for sale than drugs in publicly funded trials. It should be mentioned that this particular finding does have a couple of confounding factors. Public funding is more common for trials of “priority review” compounds which tend to differ more in action and form from previously approved drugs. It’s easy to imagine, therefore, how the “standard review” compounds more often funded privately might have an easier time getting through as there is precedent for modes of action, etc. However, it’s potentially significant that privately funded drug trials tend to go better for the patent holder than those funded publicly.
This meta-review also found that the funding companies were able to influence study protocols. This includes such paramaters as what placebos were included, etc. They also found, however, that this did not affect the protocol quality. All of the methodologies themselves were up to snuff with current standards, but crucial decisions which can have an effect on study outcome could be influenced by the funding bodies.
Another major problem this study found was with clinical trials in general, and not necessarily specific to those funded by pharmaceutical companies: selective publication. This is sometimes called publication bias. Basically, you report (or publish) results where the study showed positive results, but fail to publish either inconclusive studies or studies that turned out negatively for the drug. This can also include multiple publications of positive results. This is not really allowed in any scientific arena but people do get away with repackaging old data and not making it completely explicit that it has been published elsewhere. Publication bias is not a problem exclusive to drug trials either, but is pervasive in all research science. There are many reasons for this, causes of it, and dimensions to it. Suffice it to say that in the context of a drug trial, if negative results are not published, new trials of the same drugs can be designed that are slightly tweaked to come up with a more positive result. This is obviously questionable in the context of medicines.
In order to combat publication bias in clinical trials, all trials are supposed to be publicly registered PRIOR to the study reaching its conclusion. Researchers are supposed to report the study methodology, and importantly, the criteria they are using to evaluate study outcomes. The literature study we’re discussing also found that privately funded trials were often not registered, or not completely registered. This opens the door for publication bias through not publishing a study with negative results, as well as provides the opportunity for changing study criteria after the data is in to alter the overall study outcome. Again, this is pretty questionable when we’re talking about evaluating drugs.
In the same sort of vein, it was also found that often privately funded clinical trials with-held knowledge of adverse drug reactions. This can potentially slip through the cracks even if the trial is properly registered as registration talks about expected outcomes and study criteria but not necessarily unexpected side effects. Obviously its a bad thing if side effects coming out of a clinical trial aren’t reported as it hinders the ability for licensing boards to determine whether a drug is safe, or more likely to evaluate possible contraindications or situations in which a certain drug is not appropriate.
The final findings of this study unearthed more issues regarding publication. Firstly, there are often publication constraints put on study authors severely limiting their ability to publish study-related data independently. This of course gives extra control of information dissemination to the funding body. Finally, the company can also introduce an element of “spin” by using ghost writers or guest authors. Ghost writers include writers that did contribute to the work (often company statisticians) who don’t get listed as authors and are not acknowledged to have worked on the project. Guest authors are sort of the opposite, big name scientists listed as authors who really did not contribute significantly to the work. Either case is deceptive and can help skew perspective with respect to how the work is received.
In short, there are some serious issues innate in private bodies funding clinical drug trials. These are not always happening on purpose or with some nefarious intent, but statistically these problems do exist. The funding burden for drug development is split between private and public sectors, and I for one am not convinced that either could manage to fund these necessary innovations on their own. It is, however, vital that consumers and regulators are aware of the pitfalls and insist upon measures designed to minimize them.
