In the United States people who need medications to treat illness are dependent on the mercy of Food and Drug Administration (FDA). It serves as a gatekeeper to drug services, deciding which medications will and will not be available to consumers. The FDA demands that developers and manufacturers of drugs furnish data on a drug’s efficacy and side effects, which it uses to make its determinations.
Some organizations are demanding a change in the process to permit terminally ill people to gain speedier access to possible cures. Their demands have so far been rebuffed by regulators. Below I argue the case for liberalizing the approval process.
Drug testing for FDA approval usually consists of three phases of trials, with each phase involving larger numbers of patients. In Phase 1, a medication is given to small numbers of healthy individuals who are observed for the development of acute side effects. Phase 2 expands to include the population of patients for which the drug is intended, and may consist of several hundred people. Phase 3, which can involve a thousand or more patients, includes double-blind, placebo-controlled trials which both measure efficacy and monitor for adverse effects. [1]
After these studies, the FDA still must evaluate the data and deliberate over approval. The entire process can take years. For chronically ill patients, the wait may mean a loss of quality of life or a progression of disability. But for the terminally ill, the time it takes for drugs to wind through this process can mean the difference between life and death.
The Abigail Alliance for Better Access to Developmental Drugs has sought to change this. The Alliance recently won a court victory in a lawsuit seeking to gain access to experimental drugs for terminally ill patients. [2] In this context, “experimental drugs” means medications that may have only undergone Phase 1 testing. Essentially, this suit seeks permission for terminally ill people to bypass the normal drug approval process of clinical trials and evaluation by the FDA bureaucracy, and the protection this process is supposed to provide.
Bypassing the process will allow them to quickly receive drugs that have only been tested on very small numbers of people. This idea has met with opposition from bioethicists and others who fear that it will have only negative consequences for all patients. However, the arguments they advance are faulty. Allowing terminally ill patients access to experimental drugs is not only beneficial and just to them, but will not have the negative consequences that opponents of the idea fear.
Opposition to allowing terminally ill people to bypass the drug testing and approval process involves three criticisms.
First, that the court decision creates a “right to experimental drugs”; [3] that is, it creates a positive right that will impose on certain people an obligation to provide other people with experimental medications. [4]
Second, that giving people access to experimental drugs outside of clinical trials will lessen the incentives for patients to participate in randomized, controlled trials that the FDA uses to determine whether drugs are “safe and effective,” its criteria for drug approval. Thus, giving access to drugs outside the clinical trials process will undermine the FDA’s ability to evaluate the risks and benefits of drugs and decide which ones deserve approval and which ones do not. [4] [5] [6] [7]
Third, that very ill patients are desperate, may not be able to rationally evaluate the choices available to them and will tend to pursue any remedy anyone claims will help them regardless of its risks or effectiveness. [4] [8] It follows, therefore, that such patients need to be protected from the possible adverse consequences of non-rational decisions they might make as a result. [8]
All of these criticisms are flawed. For purposes of this argument, I will leave aside the more fundamental question of whether the FDA should even have a role in deciding which drugs are “safe and effective” and should be available to consumers. Although that is an important issue, which has been well argued by others, [9] [10] the changes sought in this court case could represent an improvement in the system we have to live under now. If objections to it can be addressed, this kind of incremental change could benefit at least some very ill patients who are now being harmed by the current system, and the change sought by the Abigail Alliance can result in a system that is both more just and more beneficial for all patients.
The “right” to experimental drugs
The first argument against access to experimental drugs for the terminally ill springs from the idea that the court victory by the Abigail Alliance creates a positive right to certain drugs. [3] [4]
This is erroneous. The right at issue in the court case, and in this article, is a negative right: the right of a willing drug manufacturer and a willing drug consumer to engage in a voluntary transaction for a medication without interference from a third party, such as the FDA.
It has been suggested that requiring the FDA to allow consumers access to certain untested drugs may create a legal avenue to force insurers to pay for them. [4] However, this is a defect created by government intervention in the insurance market, not a logical consequence of the right under consideration.
The drug testing process: patients in conflict?
The second major argument is that the drug testing process needs to be a balancing of opposing interests, with the interests of terminally ill patients who immediately want untested drugs balanced against those of less critically ill patients who must wait for drugs to receive FDA approval. [4] Such approval will not be forthcoming until clinical trials have been completed and the drugs’ risks and benefits are defined.
By allowing free access to drugs that have not gone through trials, the argument goes, we undercut the incentives for individuals who want the drugs to participate in trials. This ends up sacrificing the interests of those who will not get access to a drug until it has been further tested. Thus, individuals who might benefit from access to experimental drugs outside of trials benefit at the expense of other patients and society in general, and “The benefit of a few desperate patients would come at a steep cost to the rest of us.” [4] But the assumptions underlying these objections are not necessarily true, and basing the drug approval system on them is unjust.
First, it is doubtful that giving patients access to drugs outside the clinical trial system will undercut the incentives to participate in trials. People have different preferences and tolerances for risk. [9] Based on risk preferences, patients seeking treatment for a disease can roughly be divided into two groups. One group consists of those who are willing to try untested or barely tested drugs. This may be because they have run out of meaningful therapeutic options (their disease is refractory to treatment with available therapies), or existing therapies don’t do enough for their disease to give them the quality of life they desire.
They may also feel that having people who are willing to try untested drugs first is an important part of advancing the knowledge and science of disease therapy. The second group consists of patients who are more risk-averse. These patients do not want to take drugs unless they have been tested and their risks and benefits are defined to some extent (i.e., the drugs have been tested on other human beings first and the outcomes of this testing are known). Although it tends to be overlooked, there may be seriously or terminally ill patients who fall into this second group.
Some people in the first group of patients might want access to drugs that have not gone through the clinical trial system. But since they are willing to tolerate risk from drugs that have unknown effects, they also provide a pool of people who are willing to participate in clinical trials, for their own benefit and the greater benefit of others with the same illnesses, because they are willing to go first.
The only way the clinical trial system breaks down is if every single patient with a given disease wants immediate access to drugs before they have gone through any trials. Unless every patient with a given disease has the exact same level of risk tolerance, this is an unlikely scenario. Instead, the dynamics of the market can meet the needs both of patients who want immediate access to drugs and those who want better-tested drugs.
It’s vital to keep in mind that manufacturers of drugs develop and market them because they see an unmet demand for the drug, and thus a possible profit opportunity. Some of that unmet demand and profit opportunity will be among patients who want drugs immediately, and are willing to take the risks involved in being the first to try them. But meeting the demand only of these patients ignores another profit opportunity: the demand created by individuals who want drugs whose risks and benefits have been defined to a greater extent by testing on other individuals. The more risk-averse second group — the individuals hesitant to try drugs that haven’t been tested on others first — will still drive the demand for medications that have been rigorously tested in the clinical trial setting.
Even if drug manufacturers were permitted to sell untested medications to patients who wanted them, they would be marketing to the few people willing to take an untested or barely tested drug. It is doubtful that these people will create a market so profitable that drug companies will abandon the profit opportunities presented by drugs that have gone through the clinical trials process, and simply start selling untested drugs directly.
The larger drug market, and hence unexploited profit opportunities, will still be people who want drugs that have been through the clinical trial process. The market will adapt to meet this demand for testing and information, sustaining a system of clinical trials done to meet FDA approval standards. In this way, giving access to experimental drugs outside of trials does not actually undercut the clinical trial-based FDA approval system.
Different populations of patients with different needs and different risk tolerance levels can have both access to untested drugs and tested drugs that come with information gained through clinical trials, without sacrificing one for the other.
There is also a concern that since double blinded, controlled clinical trials are the gold standard for learning about the risks and benefits of new drugs, allowing self-selected individuals to use medications on an individual basis, not as part of any controlled research study, does not provide useful data on the risks or therapeutic effectiveness of a particular medication. [6]
As explained by Norman C. Fost, a professor of pediatrics and the chair of the ethics committee at the University of Wisconsin Hospital, “There’s a strong societal interest in advancing knowledge and having new drugs done in trials first rather than in the marketplace, because otherwise we don’t learn anything and we waste not only lots of money but sometimes lives, and there’s no advancement so we remain in a state of perpetual ignorance.” [6]
Allowing a group of individuals who self-select to have a drug outside of a research setting may not provide data that is of use to others who desire rigorous, scientific information about risks and therapeutic effectiveness, but calling this a “waste” because we “don’t learn anything” treats some individuals as though they are nothing but means to the ends of others.
Essentially, this argument says that a particular individual should not be able to receive a disease therapy that might possibly benefit them, unless their doing so produces data that is of use to other individuals or groups of people. If no data that is useful to other people is produced from an individual’s use of the therapy, there is no justification for allowing them to have it.
The consequence of this criterion is that patients who are willing to try a new drug are treated as though they are nothing more than human guinea pigs who only deserve access to disease therapies because they produce information that other people wish to have, and the benefit to the patients themselves is of secondary concern. This use of some individuals as means (research subjects) to ends (risk and effectiveness data) that other individuals choose is an injustice, and treats some people as means to serve the ends of others, rather than as individuals entitled to pursue their own ends, including the end of survival itself.
It’s also vital to remember that the FDA does not possess some sort of crystal ball that tells them anything about a drug’s safety or effectiveness. There is only one way to obtain this information: to have someone try it first. Saying that a drug “needs more testing” or must be FDA-approved before you’re wiling to take it is the same as saying that other people should try it first so you can watch what happens to them (this is essentially what a clinical trial is), and then you will decide if you wish to take it yourself.
There is nothing inherently wrong with that; people have different preferences for information and different levels of risk tolerance. But trying to compel others who want access to experimental drugs to participate in clinical trials, by demanding that a drug be withheld from them until enough data have been produced to satisfy your level of risk tolerance, allows you to use others to satisfy your preference for information — again, to use others as a means to your ends.
Individuals who are not willing to participate in clinical trials do not possess a right to structure the system such that other people can be compelled to do so. Reducing patients’ options only to drugs that have gone through the full clinical trial and approval process leaves us with a system where patients are adversaries pitted against one another, where seriously ill patients and possible future patients are in competition with one another, and some individuals’ health, well-being, and even their lives are sacrificed for the sake of others. This is unjust and unnecessary when small changes in the system could serve the interests of all patients without denying access to the terminally ill or undermining the clinical trial system.
Paternalism for the Terminally Ill
The third major argument against access to experimental drugs — that of protecting the very ill from their own choices — is a thornier problem because it rests on the same attitude of paternalism that is the entire underlying justification for the FDA. It assumes that certain individuals possess the ability to decide when other individuals are incapable of making good or rational decisions on a subject, and should be able to substitute their own judgment accordingly, forcing the original person to live with the substitute decision-maker’s choice.
Note that by “ability to decide” I mean the supposed capability of one person to make a better decision for another person than that person would have made for themselves, not simply to the power Congress has conferred on some people to make other people’s decisions for them. The government can give you the ability to make choices for other people. That does not mean that your choices are automatically superior; it simply imposes your own risk preferences on another person. [9] But the argument for substituting the “more rational” judgment of regulators for the “less rational” judgment of terminally ill patients is flawed.
If we are using the standard of rationality to justify paternalism, we must immediately confront the question of how we judge a patient’s ability to decide rationally. Bioethicist Ezekiel Emanuel defines “the desire for unproven treatments by a few patients” as “the height of irrationality born of desperation.” [4] The first major problem with this is that individuals make real choices to try untested drugs in the context of clinical trials all the time. If they did not, these trials would not exist!
In the case of Phase I trials, completely healthy people choose to take completely untested drugs with unknown risks. In the case of Phase 2 and 3 trials, the risks may still be poorly understood. Why are these decisions rational and laudable, while demanding to try these same drugs outside a trial is suddenly irrational?
Even if participating in a clinical trial benefits others and taking a drug outside a trial does not, the fact that a particular choice serves the greater good is not in itself evidence that the choice is rational and alternative choices are not. Emanuel’s position is also circular: A person who has run out of tested therapeutic options has only high-risk, untested drugs remaining. They are willing to try these high-risk, untested drugs to extend their lives.
This is evidence that they are desperate, which makes them irrational. Therefore, people who have run out of tested drugs should not be allowed to choose risky new drugs, because they are too irrational to appreciate the risk. We know they are too irrational to appreciate the risk because, given the choice, they choose from among the risky drugs that happen to be the only options remaining to them. And so on.
But this assumes that terminal illness and the willingness to assume great risks are themselves prima facie evidence of irrationality. As Alex Tabarrok points out, this is not true. Terminally ill patients are still capable of rational assessment of their alternative choices: “When best practices fail, patients demand innovation. Patients with terminal diseases rationally demand new approaches, despite possible dangers and low odds of success, because they face low costs of experimenting with new therapies.” [10]
So is it possible to assess terminally ill patients’ choices at all? Here, we can borrow the central economic insight that all value is subjective. This is usually applied in the context of individuals evaluating economic goods. But we can also apply this insight when we consider how people form valuations of the risks and benefits of disease treatment options: “…both the expected benefit of using a product and the burden of risk bearing are subjectively experienced and knowable only to the individual actor.” [9]
Thus, even the way that different people value the quality of their own lives, the value of a life with or without disability, the value of a risk to their life or health, and the value of a tradeoff between the two, is subjective. The value of those kinds of tradeoffs is not the same for every disease, or even for every individual with the same disease. [10]
Quoting Tabarrok again, “In reality, all drugs have side effects, and what is safe enough depends on alternatives and personal preferences.” [10] The issue of alternatives is crucial. How much value someone places on the benefits of a drug when balanced against its risks may depend in part on what other therapeutic alternatives are available to them. If someone has a debilitating disease for which no treatments exist, they may choose to take an experimental drug that carries substantial risks of disabling or even fatal side effects.
On the other hand, someone who has an equally debilitating disease for which several treatments are available may deem that same experimental drug too risky. Assume for a moment that the willingness to assume greater risks with less information in the absence of viable treatment alternatives results in choices that are by definition irrational. This suggests that the level of rationality a patient with a given disease can be said to possess depends on how many drugs the FDA has previously approved for that particular disease! Clearly, this is a nonsensical way of assessing rationality, but it follows logically from the way the current drug approval and access system is designed.
If the value of competing therapeutic alternatives for individual patients is subjective, then there is no objective standard an expert can use to determine how other individuals do value or should value their lives and their health. Yet we allow regulators to carry out this kind of valuation for patients all the time, without asking them if the patients themselves want it done for them.
Patient testimony at FDA advisory committee hearings is a good indication of the fact that they do not. FDA advisory panels have heard from patients who want access to drugs with dangerous, even fatal side effects, effects that are well-publicized and that these patients are aware of. [11] [12] Patients come to such meetings to tell the FDA that they understand not only the risks of drugs, but the damage that will be done to their health and quality of life if they cannot have access to new therapeutic options.
During the public comment period of such a meeting, patient after patient can be heard making the same point: that until you have personally experienced the life-altering force of disabling, chronic, or incurable illness, you cannot understand why people who have would choose the risk-benefit tradeoffs that they do.
There is no risk-benefit calculus that is independent of the individuals making the choices, and no outsider can experience the lives of ill individuals in a way that would allow them to make that decision for someone else. There is no basis for the claim that a decision about untested, possibly risky treatment made by someone other than the patient is better or more rational than the choice made by the patient himself, because the value the patient places on the risks and benefits is subjective. This includes the risk that the drug will be totally ineffective; just because ex post a drug does not work does not mean that ex ante the choice to try it was not rational. [10]
The value attached to different risks and benefits cannot be measured or defined in any way that would make it possible for one person to claim that they can make a more rational decision about them than another person. How, then, can we base people’s access to drugs on the idea that some people can, and will, make better decisions for patients than those patients would make for themselves? We cannot, and should not, allow access to medical innovation for the sickest people to rest on such a weak and unjust foundation.
Conclusion
Although it may seem like a small victory, reforming the system for the most seriously ill is progress not only for them, but for every patient. As fundamentally flawed as the FDA drug approval system is, it is entrenched, and the majority of people are convinced that we cannot survive without it. Even so, the situation is not hopeless. We can take on the arguments justifying the FDA one at a time. We can argue convincingly that incremental reforms making drugs more widely accessible to more patients will not end in disaster, bring medical progress to a halt, or leave a trail of dead and dying people in their wake.
Allowing terminally ill people to bypass the drug testing and approval process will not create a “right to experimental drugs.” It will not destroy the incentives for patients to participate in the clinical trial system. It will not make it impossible to gather scientific data on how drugs work. What it will do is allow individuals who are fully capable of rational choice to make the most important choices of all according to their own values. For the sickest people, it is not only beneficial and just, it is a matter of life itself.
Notes
[1] Food and Drug Administration Center for Drug Evaluation and Research. 2005. New Drug Application (NDA) Process.
[2] Associated Press. 2006. Court: Terminally Ill Patients Can Sue FDA. May 2.
[3] Washington Post. 2006. A Court Makes Up a Right.
[4] Emanuel, Ezekiel J. July 3, 2006. Cancer in the Courts. The New Republic 235(4772): 9-12.
[5] Adler, Jonathan. 2006. A Constitutional Right to Drugs. The Volokh Conspiracy, May 2.
[6] Feingold, Mary. 2004. Don’t Your Breath: Why We’re Not Exactly Racing To a Cure for Lung Cancer. Madison Magazine. November 1.
[7] Kerr, Orin. 2006. New Right to Take Drugs Announced Today by D.C. Circuit. May 2.
[8] Addicott, D. 1999. Regulating Research on the Terminally Ill: A Proposal for Heightened Safeguards. Journal of Contemporary Health Law and Policy 15:482-524.
[9] Higgs, Robert. 1994. Banning a Risky Product Cannot Improve Any Consumer’s Welfare (Properly Understood), with Applications to FDA Testing Requirements. Review of Austrian Economics 7(2):3-20.
[10] Tabarrok, Alex. 2000. Assessing the FDA via the Phenomenon of Off-Label Drug Prescribing. The Independent Review 5(1):25-53.
[11] Henderson, D. 2006. Users Ask FDA Panel for Return of Tysabri. March 8. Boston Globe.
[12] Arnst, Catherine. 2005. When Patients Say: Don’t Ban My Drug. Business Week. April 25.
