My Comments on HHS’s Proposed Rule Criminalizing Trade in Blood Stem Cells

Professor Nalini Ambady (1959-2013)

Professor Nalini Ambady (1959-2013)

By Michelle Meyer

Bonjour de Toulouse, where I’m visiting this month at the Institute for Advanced Study (IAST), which is hosted by the Université de Toulouse Capitole and physically (and in many senses conceptually) situated inside the Toulouse School of Economics. That setting is a particularly good one for this post on markets in human tissue.

In early October, I wrote about a rule that the Department of Health and Human Services has proposed that would amend the National Organ Transplant Act (NOTA) to criminalize compensation for providing peripheral blood stem cells (PBSCs) given through apheresis. Compensating those who allow their bone marrow to be aspirated — the other major source of hematopoietic stem cells (HSCs) — is already explicitly criminalized under the statute. At the end of the post, I promised to follow up with a second post on why I’m skeptical about this rule, as a matter of policy. This is that post.

But before discussing appropriate policy for addressing the problem of the 3,000 Americans who die each year for want of a bone marrow transplant, and for reasons I’ll explain near the end of my post, I want to say something about one particular person: Professor Nalini Ambady, a Stanford social psychologist who, about one year ago, began an international search for a compatible, willing HSC donor to treat her leukemia.

Because she is South Asian, and because few Asians and other racial and ethnic minorities are listed in the U.S. bone marrow registry — South Asians comprise only 2% of the U.S. registry — she had a difficult time finding a match. She nevertheless found 12 potential matches in the U.S. bone marrow registry, and apparently at one point had scheduled the transplant. In the end, however, half were ruled out as imperfect matches — and each of the remaining six matches declined to donate.

Having exhausted the U.S. registry, Ambady and her supporters sought a match in South Asia registries — but even fewer people are registered in that region than in the U.S. Of India’s 1.2 billion people, for example, only about 45,000 are registered. Nevertheless, a potential thirteenth match was found in an Indian registry. He matched at least 6 of 10 HLA markers; he would need a blood test to determine whether he was a complete match. Unfortunately, after a worrisome week of silence about whether he was indeed a complete match, Ambady and her family learned that the potential donor had backed out.

And then a potential fourteenth match:

Weeks later, it happened with another potential match for Nalini, this time an 8/10 and so an even likelier match. This man wanted to watch someone else donate their bone marrow before making up his mind. A registry chief actually arranged this for him in early May. Somehow, it wasn’t persuasive enough. He, too, decided not to donate.

Professor Ambady died on October 28 after eight compatible or potentially compatible registered donors ultimately declined to donate.

Apparently, declining to donate after agreeing to be listed in a registry and being found to be a match is quite common. According to one registry chief, the attrition rate is 50% in the U.S., Europe, and India. Would a financial incentive have tipped any of these vaccillating prospective donors towards a decision to donate? Although we can’t say, this is an eminently answerable question. Unfortunately, the Department seems uninterested in learning whether a compensation scheme would save lives — or have ill or counterproductive effects — or not.

In its Notice, the Department lists the following policy reasons for expanding NOTA’s list of “organs,” the exchange of which for valuable consideration is prohibited, to include PBSCs: (1) preventing commodification; (2) curbing opportunities for coercion of PBSC providers and (3) exploitation of PBSC recipients; (4) encouraging altruistic donations, and (5) decreasing the likelihood of disease transmission resulting from paid donations. My responses to these arguments are after the jump. I end with some thoughts about alternative policy tools for increasing the supply of PBSCs in the absence of a legal compensation scheme.

Preventing Commodification

At the outset, I’ll confess that, as a category of argument, I find commodification concerns about as persuasive as Leon Kass’s (in)famous argument from the “wisdom of repugnance.” There are, of course, some things that I don’t think should be bought and sold. Votes and grades, for instance. But that’s because buying and selling these items defeats their purpose (at least in a democracy and a meritocracy, respectively). HSCs, by contrast, fight cancer whether they’re donated or sold.

And of course commodification can have bad effects (such as those reflected in the Department’s next four policy arguments). But then the debate should focus on those purported bad effects (and whether they can be mitigated or are outweighed by good effects) rather than on commodification, per se.

But even if I did find anti-commodification principles attractive, I would find the Department’s proposed rule with regard to PBSCs difficult to square with the fact that NOTA does not prohibit the buying and selling of either blood and plasma or sperm and ova, where any dangers of commodification should be heightened. After all, gametes are not only human body parts; they also have the potential to create new persons. Buying and selling them should, then, be doubly commodifying. And yet a robust, legal trade in gametes exists.

The Department’s anti-commodification argument is also at odds with the fact that every other party in the transfer of PBSCs from provider to recipient is compensated financially for her role in that transaction — except the one who bears the physical costs. Applied selectively, anti-commodification arguments can wind up being vehicles for unequal treatment.

“Coercion” (Undue Inducement) of PBSC Providers

Although the Department’s Notice refers to “coercion,” under most accepted philosophical definitions of coercion, none would be involved in a PBSC market. Coercion proper involves a threat that reduces a person’s choice set. For example, my doctor says to me: “Donate” PBSCs to my other patient, or I won’t treat you anymore. Before the threat, I had medical care from my doctor and was independently free to provide PBSCs to others, including my doctor’s other patient. After the threat, my ability to continue enjoying the former benefit is contingent on my “choice” to perform the latter action. By contrast, an offer of payment in exchange for providing PBSCs, which I am free to accept or refuse without penalty or loss of benefits to which I am otherwise entitled, expands my choice set.

Still, even offers that expand the choice set can be problematic. Sufficiently attractive offers induce people to do something they wouldn’t otherwise do. Compensating PBSC providers may be partially motivated by a desire to express gratitude to providers in a tangible way or to make providers whole by compensating them for their time, pain, and suffering. But honesty requires acknowledging that a big part of the motivation behind a compensation scheme is surely the hope that it will induce some to provide PBSCs who otherwise would not have done so.

Yet people do things for money (and other “valuable consideration”) every day that they wouldn’t otherwise do. Many of you are doing them (or supposed to be doing them!) right now; they’re called jobs. What we’re really worried about is undue inducement, which occurs, I suggest, when each of two conditions obtain: (1) the actor is relatively poor or otherwise desperate for money, such that the offer of compensation blinds her to an honest assessment of the action’s risks and expected benefits; and (2) what an actor is paid to do is more than minimal risk.

Note, with respect to the first condition, that it excludes cases involving an accurately informed, sufficiently comprehended and considered decision to assume risk or cost in exchange for money because the actor rationally judges that choice will improve her situation. Although it is bad (usually) that some people have fewer and/or worse options than others, removing from a disadvantaged person’s choice set (or prohibiting others from adding to it) the very option that she rationally prefers to her meager alternatives does nothing to improve those alternatives. As for the second condition, the idea is that we shouldn’t much care if someone is blind to the costs of an action she’s paid to perform if that action doesn’t actually have any serious costs.

Whether an offer of payment to provide PBSCs is sufficiently attractive to distort the actor’s decision-making judgment is likely depend, among other things, on the amount offered, the offeree’s perceived need or desire for money, and her alternatives for procuring money. These factors, obviously, will differ quite a lot among individuals. We could set the price of PBSCs very low to avoid undue influence on decision-making, but such a price would likely induce very few people to shift from non-providers to providers (this is perhaps especially true given that, unlike a market in, say, kidneys, where virtually everyone would be a match for someone in need, HLA typing for HSCs is much more exacting, and will need to draw from providers across the socio-economic spectrum).

Notably, a relatively low price would also arguably assuage worry about undue inducement at the expense of creating a new worry about exploitation of providers paid offensively low wages. In short, any price that is likely to induce a significant number of providers (and provides sufficient numbers of them with compensation that meaningfully reflects the value of their time and the physical symptoms) is very likely to sweep in at least some people for whom that amount of money would be unduly attractive. We could address some of the problem of undue inducement by setting the price for providing PBSCs as some portion of the provider’s annual income or net worth, thought that would be logistically cumbersome and would likely raise fairness issues. Even as to providers for whom compensation is clearly very significant, we could address some of the worries about decision-making distortion (which, recall, is what we’re really worried about) through comprehension tests, cooling off periods, and so on.

But let’s stipulate that the first condition is met. What about the second condition, concerning the costs of distorted decision-making? The Department notes:

Though safer and less invasive than aspiration, apheresis still carries risks to the donor. Side-effects of the HSC stimulating drug may include rupture of the spleen or a low platelet count (thrombocytopenia). There may also be serious risks related to the placement of a central venous line in larger veins (jugular, subclavian, or femoral) in donors without adequate peripheral vein access. More importantly, aspiration is the medically indicated method of donation for a substantial number of transplants.

But crossing the street “carries risks” that are even more serious — like paraplegia, brain damage, and death. What are the probabilities of the risks that the Department lists?

Before getting to that, I think proponents of a market in PBSCs would do well to acknowledge a version of what the Department is saying in the above excerpt — namely, that although there are significant differentces between providing HSCs through bone marrow aspiration, on the one hand, and procuring PBSCs through stimulating drugs and apheresis, on the other hand, the latter is not, for the vast majority of people, “painless” or “as painless as donating blood.” To the contrary, giving HSCs through aspiration and through apheresis have remarakably similar overall risk-benefit profiles (see also here). Virtually all providers, whatever the method by which they give HSCs, will experience pain, and many will experience other symptoms such as fatigue, headaches, and nausea. The main differences are that with apheresis, the frequency of serious and unexpected adverse events is cut in half, average recovery time is cut by two-thirds, and pain is experienced as bone pain spread througout the body during the five-day drug course leading up to the procedure (whereas those who donate through aspiration primarily experience pain at the aspiration and anesthesia sites immediately following the procedure and thereafter).

That said, the risks are not especially great with either procedure. Focusing now only on PBSC provision through apheresis (since that’s the only procedure compensation for which is an open question), the overwhelming majority of PBSC providers experience only mild or moderate symptoms (with most symptoms mild) — primarily pain, fatigue and insomnia. In a frequently-cited study of 7,850 PBSC donors over a ten-year period, only 39 (0.5%) experienced any serious adverse event related to the procedure, and all such adverse events were transient and resolved completely. Only four (0.05%) experienced the thrombocytopenia that the Department mentions, and the researchers suggest that in three of those four cases, the donor had preexisting low platelet counts that were simply exacerbated by the drugs. As for central lines, these are required in only a minority of PBSC providers (4% of male donors and 17% of female donors) and in the aforementioned study, only 5 donors (0.06%) experienced central line complications. These aggregate risks go down still more if we discourage donations from females, those who are older, and those who are overweight and obese, all of whom, for unknown reasons, have significantly higher risks of pain and/or adverse events.

In short, every study of PBSC transplantation has found it mildly or moderately unpleasant but “safe,” in both the short and long terms, for providers and “effective” for recipients. People pay to assume greater risks pursuing far less noble ends. Closer to the case at bar, women are permitted to receive money to assume greater risks and undergo a more invasive procedure in order to provide ova for others’ reproductive or research purposes. We don’t criminalize paid logging, mining, firefighting, military service, or law enforcement, all of which remain risky even after regulations. And so on.

In many ways, in fact, the voluntariness of decisions to provide PBSCs is far more uncertain when the exchange is “altruistic.” We have long known that related donors (by far the majority of “altruistic” donors) feel considerable family and moral pressure to donate. It’s unclear why financial pressure poses a unique problem that merits a ban on using that form of pressure. Indeed, it is these family pressures — and not offers of money — that can easily become coercive explicit or implicit threats (e.g., “donate” HSCs to your sibling/parent/cousin, or your relationship with both her and the rest of your family will be forever altered and perhaps broken).

The absence of a market to which patients can turn to recruit willing but unrelated PBSC providers likely exacerbates this pressure on related donors. Indeed, by permitting transplants that occur as a result of family and moral pressure but prohibiting those that would owe their occurrence to financial motivation, we concentrate the burden of serving as PBSC providers on those who happen to be related to those in need and those among us who, for various reasons, tend to feel a stronger sense of moral duty. Critics of organ markets often focus on the distributive injustice of a policy that concentrates the burdens of provision on those for whom financial compensation is especially attractive. But our current policy, which concentrates those burdens on those for whom family pressure or a sense of moral obligation are especially powerful incentives, has its own distributive justice problems.

Finally, although it is true that in some 20-30% of transplant cases, donation of blood stem cells through aspiration, rather than apheresis, is medically indicated, it is unclear how this fact is relevant to the question of whether compensating providers of PBSCs through apheresis is appropriate. Is the Department’s concern that people who, prior to a market in PBSCs, would have freely donated stem cells through aspiration will instead sign up to be paid providers of PBSCs through apheresis? If so, this speculation should be tested before basing policy on it. Even if the number of aspiration donors declines after a market for PBSCs through apheresis is introduced, so long as those numbers are more than made up on the paid apheresis side (by new providers who either would not have donated through aspiration under any circumstances or who are motivated at least partially by money), it seems probable that we would prefer the policy that yields the greater total number of transplants.

Exploitation of PBSC Recipients

The Department argues that, “in light of the much more stringent matching required between donor and recipient for HSC transplants to be successful, the opportunities for exploitation of those in medical need of HSC transplantation are much greater than for solid organ transplantation.” In other words, the Department is worried about the fact that in many cases, a matching PBSC provider holds a monopoly on that life-saving resource and engage in price gouging. Whatever the merits in other contexts may be of protecting people from their own decisions to give away their labor for a “too low” price — or, in this case, to pay a “too high” price for someone’s else’s labor — those merits do not seem to apply here.

First, it’s not clear that PBSC recipients would be on the hook for the costs of compensating providers; some or all of those costs might be borne by government, nonprofits, or insurance. Moreover, there are less blunt instruments for addressing this concern than a ban — such as price controls. The Department notes, of the Ninth Circuit opinion in Flynn v. Holder, that “[w]hile [it is] focused on the proposal of the plaintiffs before it, the Court’s holding does not limit the compensation donors can demand to scholarships, housing allowances, or charitable gifts.” That’s true. But although I haven’t researched this question, it is not clear to me that the Department lacks all authority to regulate a compensation scheme to curb prices and avoid other features that the Department, in its wisdom, might find objectionable (and of course Congress has this authority). The Department should give serious consideration to a regulated market before pressing ahead with a ban of all compensation schemes in an effort to avoid those it deems exploitative or coercive.

But let’s assume that the Department lacks the authority to regulate the price of PBSCs and that the money to compensate PBSC providers comes from recipients’ own pockets. Unlike some other desperately ill patients, these patients are not asking to spend their money on something that is known (by others) to constitute proverbial snake oil or whose efficacy has yet to be proven. What they want (by hypothesis) is to be able to purchase a treatment that is known to extend life. Although demanding very large sums of money from desperate patients is worrisome, or at least distasteful, we should remember that prospective PBSC recipients are, in fact, dying. It’s not unreasonable to ask the Department to state exactly what price is “too high” to pay to save one’s own life. And it’s more than a little to odd to “protect” people from being exploited, only to allow them to die for want of a sufficiently incentivized PBSC provider. (For a current film that explores these issues, I recommend the excellent Dallas Buyer’s Club; look for Matthew McConaughey and Jared Leto to clean up come Oscar time.)

Finally, if the Department is going to weigh the purported costs to recipients of compensating PBSC providers, it should also consider the potential benefits to the recipient of a compensation scheme. Sociologists have found that gifts are normally governed by strong norms of reciprocity. When the gift is the gift of life, however, it cannot be reciprocated, often leading to what has been called the “tyranny of the gift.” Compensating PBSC providers may mitigate this burden on recipients by allowing them to tangibly pay back providers and be free of their perceived debt.

Encouraging Altruistic Donations

Two versions of this argument are often invoked in debates about moving from an altruism-based to a market-based exchange system. Since the Department merely lists all of its policy arguments in a single sentence, it is unclear which version it means to invoke. I’ll address both.

The first version simply notes that altruism is good and laments that opportunities to express it will be eliminated. The weaknesses of this argument should be readily apparent. First, no provider of PBSCs is forced to accept monetary compensation; she can decline the compensation or, as the pilot program at issue in Flynn v. Holder had proposed, elect to donate the money to a charity of her choice. Thus, an opportunity for altruism is not really being eliminated. Second, even if providers do accept compensation when they otherwise might have donated PBSCs altruistically, there is — alas — a tremendous amount of need in the world, such that there are ample other opportunities to give freely of oneself. If you’re looking to get your mitzvah on, not to worry; you’ll have plenty of other opportunities. Finally, however important a good altruism is, surely it is not more important than saving lives.

The more serious version of the argument rests precisely on the importance of saving lives. According to this argument, markets sometimes “crowd out” altruism: some who would have been willing to perform an action for free may be less likely, for various psychological reasons, to perform that action for payment (although they are often again willing to perform it if the compensation is high enough). If the crowding out phenomenon obtains in the context of PBSCs, it is possible that offering compensation would have a net decrease on the supply of PBSCs — the opposite of what organ and tissue compensation policies intend. That would indeed be a strong policy reason not to compensate PBSC providers.

But whether a crowding out effect exists in the very specific PBSC context is an empirical question to which we do not have the answer. Markets do not always crowd out altruism, and the extent of this effect in general is subject to considerable scholarly debate. Even in the context of blood donations, where the crowding out hypothesis has held the most sway, one recent study of nearly 100,000 people found that those who were offered an economic reward for providing blood were more likely to do so. And in any case, the PBSC context is quite different from blood donation, and so presents a separate empirical question.

Happily, rather than basing policy on speculation, the Department has a unique opportunity to engage in evidence-based policy. If the Department feels that it lacks the authority to oversee a randomized, controlled trial of paid and unpaid PBSC providers, it should permit some form of compensation and collect data about rates of donation/vending before and after its policy goes into effect and also compare paid PBSC provision to altruistic bone marrow provision (admittedly, the two procedures are not identical — and, more importantly for testing the effects of incentives, they are likely not viewed as identical by prospective providers — but some data is better than nothing). Indeed, the research team behind the study that found that compensating people made them more likely to give blood was finalizing a research protocol that would have investigated the effects of compensating PBSC providers when they learned of the Department’s plans to criminalize such compensation.

Disease Transmission Resulting from Paid Donations

The claim that the quality of PBSCs will decline under a market scheme because providers motivated by money will have incentives to conceal health issues is also an empirical claim that can and should be tested. Moreover, any reduction in the quality of PBSCs due to a compensation program can be mitigated by extra screening, the costs of which recipients and bone marrow advocates would likely be only too happy to absorb, if it means more lives saved.

Policy Alternatives?

I hope that HHS reverses course, or else that the Institute for Justice wins the lawsuit that it vows to revive if HHS can’t be reasoned with. But in the absence of either development — and I think it is a distinct possibility that neither will occur — we need to think about other ways of increasing the supply of hematopoietic stem cells. I suspect — the apparent 50% attrition rate notwithstanding — that the hematopoietic stem cell supply problem has as much to do with a failure to get sufficiently large numbers of people to sign up for the bone marrow registry in the first place as it does with converting registrants to donors. If so, I can think of two things off the top of my head that we could be doing.

Mandated Choice

First, we should routinely offer people the chance to register. The process is easy: it only requires a quick cheek swab. (You can sign up here.) States offer drivers the chance to register as a (posthumous) organ donor when they first receive or renew their driver’s license. Blood drives, too, involve vans that park on college campuses and other locations with lots of healthy potential volunteers. Apparently, many current registrants are on the list because they were asked to register while they were donating blood. Are all blood donors asked to be listed on the bone marrow registry? If not, they should be, and similar opportunities should be exploited. At the extreme end of this policy approach is “mandated choice,” in which every individual at some point is required to say yes or no to being tested and listed in the registry. And of course there are probably lots of opportunities to tweak the choice architecture surrounding the decisions to register and donate.

Paying People to Register, Not Donate

If you go to the bone marrow recruitment site, you’ll be asked to check off a few eligibility boxes attesting, for instance, that you’re in good health and that you haven’t already registered. One of the eligibility boxes, however, asks you to pledge that “I am willing to donate to any patient in need and I understand the donation process — for more information, see Understanding Your Commitment.” Following that last link, you’re told:

Before you join . . . When you register to be a bone marrow donor, patients are counting on your commitment. Their lives depend on it. If you match a patient, you have the right to change your mind. However, a late decision to not donate can be life-threatening to a patient. Please think seriously about your commitment before joining the registry.

This will be controversial, but I wonder if all this foreboding talk of commitment at the beginning of the process might be exactly the wrong way to go about getting large numbers in the registry. Encouraging more people to enter the registry — rather than discouraging them with stern warnings that they must commit now and forever more to being a donor — has the potential to increase the supply of stem cells by exploiting a sort of cognitive bias in how we assess our moral duties: We tend to think that we have stronger obligations to help identified individuals than “statistical” lives. I suspect that many people (myself included) who do not go out of our way to register as a donor, despite knowledge that many people are awaiting life-saving transplants, would, if faced with the knowledge that they are the only match for a specific person — a Nalini Ambady — would not hesitate to donate, compensation or not. I myself would not look forward to the procedure, but I cannot fathom allowing someone (even a stranger) to die simply in order to avoid an unpleasant — but safe — week-long process. Of course not everyone will share this intuition, but I don’t think I’m at all alone in holding it.

This leads me to my second policy alternative. If a major stumbling block is getting people into the registry, which is a stage at which dying patients are for many prospective registrants merely “statistical lives,” then maybe what we should be doing — all without violating NOTA, even as interpreted by the Department — is to pay people to join the registry.

It’s true that at some point, a refusal can put the would-be recipient in a worse position than she would have been without any offer to donate. Prospective donors should be made aware of the seriousness of their commitment well in advance of that point of no return. And of course registrants should be informed about what they’re signing up for. But there’s  no need to front-load stern warnings about the seriousness of the commitment at the beginning of the process in ways that might discourage them registered.

And it’s also true that patients and their loved ones find it extremely frustrating when someone who had agreed to be added to the registry later declines to donate. For instance, a friend of Ambady’s writes, with understandable bitterness, of the Indian man who declined to be tested for full HLA compatibility with Ambady:

Why then did he insinuate himself into the registry in the first place? . . . I imagine him signing up in a fit of enthusiasm that drowns out advice about what the donation itself involves. I imagine him going home in a warm fuzzy cloud, feeling good about himself. Months or years later, he’s called to donate. Reality bites now. . . . Indeed, why this devolved the way it did was a toss-up between his ambivalence and the failure to adequately counsel him when he signed up.

But surely these disappointments would be worth the pain if “overregistering” people led to more total donations. (Whether it would have this effect or not is, of course, an empirical question.)

Finally, I realize that this policy proposal is in some tension with the suggestion I made above, that our current system may unfairly burden those who are motivated by altruism. Doesn’t paying people to register, in the hope that, when confronted with a particular person in need of their unique help, they will agree to donate out of a sense of moral obligation, amount to the same thing? Probably, yes. Clearly, the 50% attribution rate tells us that not everyone shares that sense of moral obligation to donate, even when confronted with a specific person in need; the burden will fall on those who do. And I would prefer a system in which a more diverse range of people who are differently motivated end up providing PBSCs. But if federal law prohibits compensation, then I’m willing to try a policy that harnesses the “coercive” pull of conscience.

[Cross-posted at The Faculty Lounge]