Most readers of this blog probably live in countries where a pharmaceutical regulatory agency decides what medical treatments are made available. Following the paradigm of the US Food and Drug Administration, for the last fifty years these agencies have been testing new treatments for safety and efficacy with a standardized experimental design called randomized clinical trials (RCTs). Historians of medical experimentation (like the late Harry Marks) claim that RCTs were adopted because they implemented controls that warranted the impartiality of the experiment.
There is an open debate over how the experiments bear relevance to the world in which regulatory decisions play out. But regulatory agencies have trusted evidence from well-designed RCTs as the most impartial ground to decide on the safety and efficacy of a new treatment.
A clinical trial often generates major conflicts of interest because the sponsoring pharmaceutical company wants its treatment to succeed in the trial and patients often have preferences on treatments even before enrolling in the trial. Blinding all participating stakeholders is a device to control the experiment and prevent interferences. If neither physicians nor patients know whether they are receiving the experimental treatment or a placebo, their preferences will not influence the outcome of the trial.
Patients have rarely liked blinding, as documented in the study of Act Up activists against blinded AZT trials in the 1980s. The activists revolted against the trial protocol, taking their treatments to chemists for analysis and dropping out of the trial if they weren’t getting desired treatments. But this has been an unusual situation, because most trial participants do not have an easy way to find each other and connect. Trials often are run in different centers, sometimes scattered all over the world. Patients might know only a few or no other fellow patients.
Today, online digital communities provide an infrastructure to exchange information about all sorts of medical conditions and health situations. PatientsLikeMe is a well-known example. Originally an initiative by a group of ALS patients and activists, the website grew to incorporate standardized tests and tracking devices, which allowed patients to measure and report on the progression of their condition and share and discuss their data with peers.
Drawing on our ethnography at the parent company, our research documents how a group of participants in a trial of an ALS treatment shared data on its effects on PatientsLikeMe. Some patients who weren’t eligible instead used Google Sites to run their own test with a home-brewed substance that simulated the experimental treatment, and shared information on PatientsLikeMe and other websites. PatientsLikeMe researchers collected these data, ran their own analysis, and later published a report suggesting the home-brewed test was harmful and called for a wide discussion of the risks of these developments.
This initiative has been discussed as an example of patient-led experimentation. Rather than being passive actors in clinical trials, self-organized patients should have a say in the design of the experiment. Now that digital networks give patients the power to start a coordinated revolt against research protocol, reform would require no randomizing and blinding treatments without the participants’ effective consent.
Empowering patients seems indeed a good thing, since pharmaceutical companies have a long history of abusing patients in trials. In our research, we weigh its collateral effect, the risk of regulatory agencies inadvertently taking decisions on the basis of evidence biased by unblinded trial participants. Without additional methodological precautions, the interpretation of this evidence could lead to mistaken evaluations of the safety and efficacy of treatments.
Our research warns against taking a quick position that we could call pharmaceutical populism. It is tempting to present conflicts in clinical trials as a polar opposition between trial participants, victimised and bolstered by charismatic leaders, against an industrial system that fails them.
We show that there are many more interests at stake in regulatory trials, and the current regulatory regime is an attempt to balance between all of them. There may be new trial designs that better accommodate the demands of trial participants (against randomization and blinding). But it remains to be seen whether the evidence they yield will be enough to make solid regulatory decisions about new treatments. If these trials lead to more errors, the price to dispense with trial controls and accommodate the participants’ preferences will be paid by the patients suffering adverse treatment effects in the future.There are many other forms of production of evidence and clinical practice that are influenced by industry bias. Patients and activists can and do fight them. And other current developments also might turn into a poisoned apple such as, for instance, the ‘right to try’ laws. Patient activism might need to pick its battles carefully.
Niccolò Tempini and David Teira. “Is the genie out of the bottle? Digital platforms and the future of clinical trials.” Economy and Society 2019.
Image: pxfuel (CC0)