Jessica Flanigan, Pharmaceutical freedom, Oxford University Press, 2017

Most academic philosophers live in a world of pharmaceutical paternalism: new treatments are subject to premarket approval by State regulatory agencies, and patients will only be able to access them through medical prescription. Outside academic philosophy, there has been a long tradition of pharmaceutical libertarianism, advocating for allowing patients to take their risks with experimental treatments without State or medical interference. Precisely because this tradition has had almost no philosophical echo, Jessica Flanigan’s Pharmaceutical Freedom was a necessary book. Its broad claims and intuitions are not new. The real originality lies in the thorough examination and defence of the right to self-medicate from the standpoint of contemporary moral philosophy.

  

Flanigan grounds her arguments on the patient’s authority in knowing what will promote her overall well-being and in deciding what to do, with an original twist: she shows that supporters of informed consent should also endorse self-medication, as it is based on the same foundations. She criticizes regulatory paternalism for failing to adequately protect public health: informed self-medication could do better. Moreover, pharmaceutical regulators “wrongfully kill patients by withholding access to investigational drugs”, until they are approved. Drug prescription is equally objectionable for restricting the right to die, the right to use recreational drugs, the right to self-enhancement and the right to use pharmaceuticals for non-medical reasons. Flanigan defends that patients should “develop drugs outside the formal mechanisms of the approval process, disobey pharmaceutical regulations, and protest existing policies” (p. 165), even if they do not have the support of a majority of voters (“biased against rights of self-medication”). If, at this point, you are wondering whether pharmaceutical companies do have any particular moral obligation towards patients, Flanigan defends that it is an industry like any other: there is no reason to expect that they provide affordable products. Pharmaceutical marketing (e.g., direct to consumer advertising) should be deregulated, such as any other instance of paternalistic health regulation. Tort law provides enough defence for pharmaceutical consumers to stand against any industry misdeed.

These are the claims that Flanigan makes in her book, and I think the author deserves all the praise to make them as boldly and straightforwardly as possible. Most of these claims, although with different justifications, have been around since the passage of the 1962 Food and Drug Administration Act. It established a pre-market approval system for drugs on the basis of the safety and efficacy they showed in randomized clinical trials. Flanigan surely draws on these libertarian/pro-industry sources (e.g., W. Wardell and L. Lasagna are cited; T. Stossel is mentioned in the acknowledgements), but she constructs arguments that rely, to an unprecedented scale, on principled philosophical considerations (after the steps of Allen Buchanan, Robert Veatch, etc.). Nonetheless, Flanigan has also a policy proposal: a supervision system for pharmaceutical markets. Instead of pure market control (customers select the good treatments buying them), in the best libertarian tradition, or ex post withdrawal of the dangerous compounds (Wardell and Lasagna’s approach), Flanigan defends a certification system, in which a public agency tests the treatments already sold in the market and assesses their safety without gatekeeping powers.

I leave to moral philosophers the discussion of her more principled arguments and, as a philosopher of science, let me take issue instead with this latest proposal. My general contention is that Flanigan tends to overstate her moral claims at the expenses of every other consideration. A case in point is precisely the certification of drug safety.  For Flanigan,  “Drug safety is a normative judgment that requires knowledge about how the risks and side effects of a drug fit into a patient’s life as a whole” (p.10). The 1962 FDA Act definition of safety linked it to efficacy as shown in one particular experimental design: randomized clinical trials should establish whether a treatment for a given condition would reliably have certain effects. This is the positive concept of safety: the choice of a treatment outcome might be normative, but regulators care about it being predictable. Flanigan claims there is something more to safety: a drug may be unsafe for the FDA (e.g., having known adverse effects) and yet a patient may think it is safe for her to take it and, according to Flanigan, she should have the last word about it.  Flanigan provides three arguments about why people can consent to drugs that carry known (and unknown) risks (p.24), showing how consent to risky alternatives (rejecting treatments, participation in phase I trials) is considered acceptable, and yet it seems unjustified to consent to risk thresholds below FDA-safety. So why not leave the ultimate assessment of safety up to the patients?

This is Flanigan’s certification proposal: let regulatory agencies certify the predictable risks of a treatment and leave to patients the informed decision to take them. On purely normative grounds, I think this is a very persuasive idea for anyone with even a minimal anti-paternalist inclination. The problem is, of course, the positive implementation, about which Flanigan provides almost no detail. Her intuition is that for the majority of patients who value their health, not much would change: they will rely on certification, while the most desperate/daring patients will trust their own judgment (p. 62). But I disagree here: the devil is in the details. Randomized clinical trials are long and complicated experiments, often involving thousands of patients, and completing them usually takes years. Assuming that Flanigan’s system adopts the current evidential yardstick, certification will be equally slow. There will be many non-certified systems in the market for which patients will only have the non-certified information provided by the manufacturer. Patients in need of these treatments will have to trust the honesty of the manufacturing company in reporting risks (p.26), and litigate if, after trying the drug, she feels cheated: according to Flanigan, tort law provides a better protection for her rights than any pharmaceutical regulator (p. 221).

I doubt it. For a start, there is an extensive academic literature on the misdeeds of pharmaceutical marketing that should give us some pause about its honesty. This literature is simply absent from the two sections in which Flanigan discusses the topic (pp. 209-217). Instead, pro-industry views are presented on a positive note, giving them, I fear, more epistemic weight than they deserve. E.g., I was intrigued by an unusual reference on the positive effects of direct-to-consumer advertising (p. 211, footnote 13) for which just the title was given and a long Proquest link. Searching on Proquest, I found that the paper was published in the “Web Exclusives” supplement of a journal, with some very serious scholars as lead authors and some co-authors from a marketing agency. After reading Sergio Sismondo’s work on pharmaceutical publication planning, I would have expected Flanigan to add some caveats on the reliability of such a source (and contrast it with more standard sources on the topic).

But even if the libertarian pharmaceutical consumer cannot trust the manufacturer, if she feels cheated after receiving the non-certified treatment, she can always sue for dishonest reporting about risk. My question is: on what grounds can she contest the honesty of the non-certified brochure? On the one hand, there are many evidential standards that can be used to argue that a drug is safe/effective, ranging from clinical trials with different designs to observational studies or simple expert judgment. If the industry is self-serving in choosing the evidence reported in the brochure, how can a court adjudicate whether the information was accurate? On the other hand, given that testing treatments involve long and expensive studies, how can the dissatisfied libertarian patient ground her legal claims? Judges and patients may both defer, of course, on the certification agency test as a benchmark, but then we would be almost back at the status quo: which company will dare to launch a product without conducting long clinical trials, for fear of litigation? 

This are not mere counterfactual scenarios: this is the history of pharmaceutical markets. For example, legitimate British manufacturers in the 1930s were demanding a regulatory intervention in order to protect themselves against the rogue competitors who sold fake drugs based on pretence tests, undermining their own sales. Fraud prosecution offered little protection (against Flanigan’s optimism), because it was difficult to prove fraud in court without an evidential benchmark for authenticity. Again, I cannot help feeling that Flanigan cherry-picks the evidence that supports her case. She cites uncritically the usual bunch of old econometric studies that provide ammunition against our current regulatory system (Peltzman, Temin, etc). Yet, the key concept of Dan Carpenter’s landmark monograph on the FDA, a thick volume of analytical social science published less than a decade ago, is dispatched in just one sentence (p.127).

Summing up, there is a significant gap to bridge between Flanigan’s normative understanding of drug safety and the practical implementation of a certification system for “positive” safety. Flanigan is at her best in the discussion of moral principles (which is 2/3 of her book), but really does a poor job when it comes to policy recommendations (the reminding 1/3). I have been clearly unfair to her work in focusing on this latter aspect, but, I think, it is for a good reason. If I were a libertarian patient willing to dismantle the FDA, I would want some reassurance that we won’t see a regression to the world of unpunished pharmaceutical frauds of the early twentieth century. Moral argument, in other words, is not enough for policy action. But we should be grateful to Flanigan for articulating the libertarian case so well and making it part of our philosophical conversation on the pharmaceutical world.

{March, 2018}

{Metapsychology}