Do no harm. Nearly all US MDs and PhDs recite this three-word oath instinctively when asked why they object to helping with executions. “It violates our ethical oath and erodes public trust,” says Priscilla Ray, head of ethics at the American Medical Association (AMA), which prohibits members from participating in executions. Even helping to design a more humane protocol would disregard the AMA code, Ray says. “Formulating a way to kill somebody would violate the spirit of the policy.”

As a result of that stance, lethal injections — the dominant method of execution in the United States — are generally carried out by technicians without scientific or medical training, and the protocol does not seem to have been reviewed in 30 years. Typically, prisoners are injected with three drugs in sequence meant to knock out, paralyze and then kill (see graphic, below). But many experts are concerned that because of that lack of training, the first drug does not always knock the condemned out properly, leaving him or her paralyzed but excruciatingly conscious as the lungs stop moving, and burning potassium chloride races towards the heart.

Evidence that lethal injections are inhumane are being used to delay executions. Credit: AP/ARIZONA DEPARTMENT OF CORRECTIONS

None of this seems until recently to have hindered the popularity of the technique. It was devised in 1977 by Oklahoma medical examiner Jay Chapman, despite his admitted lack of experience. It was meant to be more humane than the electric chair, which was occasionally setting inmates on fire, and the method has since been copied by 36 other states, apparently without any modification. It has killed more than 800 prisoners so far in the United States, and other countries, including China, Guatemala and the Philippines, have adopted the method in recent years. US prisoners have brought cases against their states for years, arguing that lethal injection is a “cruel and unusual punishment”, and therefore prohibited by the constitution. But they have not made significant progress in the courts.

Now, however, the tide may be turning. In recent months, several death-row inmates and their lawyers have had some successes, helped by evidence that as many as 43% of prisoners may be conscious until the point of death (L. G. Koniaris, T. A. Zimmers, D. A. Lubarsky and J. P. Sheldon Lancet 365, 1412–1414; 2005).

Doesn't the state have some minimal obligation to do research to find a method as humane as possible?

Last week for example, the Supreme Court heard arguments on whether condemned murderer Clarence Edward Hill can use a particular legal avenue to object to the way the state of Florida plans to execute him. Of the judges involved in the case, Justice Anthony Kennedy seemed upset by Florida's insistence on the standard protocol, asking: “Doesn't the state have some minimal obligation to do research to find a method as humane as possible?” If the Supreme Court lets Hill try this method of appeal in the district court, they may see it on their docket in the near future.

In California, executions have been suspended until a September hearing rules on the constitutionality of the method used there. The hearing could have been avoided had the state been able to fulfill a judge's order that murderer and rapist Michael Morales be executed by a single dose of anaesthetic administered by a licensed practitioner, or that a trained anaesthesiologist be present. The state couldn't find any medical personnel willing to participate.

And in North Carolina last month, a judge ruled that the execution of Willie Brown Jr, a convicted murderer, could proceed only if a medical expert ensured Brown was fully anaesthetized when the final two drugs in the sequence were injected; anonymous medical personnel were persuaded to take part.

Ethical dilemma

These latest cases place doctors and scientists, who might be able to come up with an alternative method, centre-stage in the debate. “There is a real problem here,” says Richard Dieter, executive director of the Death Penalty Information Center, a clearinghouse, an agency based in Washington DC with no official position on the death penalty itself. “Lethal injection is a medical procedure. I think that it almost requires the presence of the medical community.”

Nathan Barankin, spokesman for the California attorney-general's office believes that arguing this need for trained overseers is the operational strategy of campaigns against the death penalty. “The goal of death-penalty opponents is to get a court order that says that lethal injections can only be administered by licensed professionals, because the ethics of medical professionals prohibit them from participating.”

Opinion among medics about the stance they should take — violate ‘do no harm’, or allow prisoners to die in pain? — is divided, however, and often not along predictable lines. Atul Gawande, an assistant professor of surgery at Harvard Medical School in Boston, Massachusetts, interviewed several doctors who help with executions for a March article in the New England Journal of Medicine (A. Gawande, N. Engl. J. Med. 354, 1221–1229; 2006).

Gawande supports the death penalty, but argues that doctors ought not use their skills to do anything but heal. “If I have to choose between core principles of medical ethics and the death penalty, I will choose medical ethics,” he says. “If the only way we can do an execution and have it stand up to constitutional scrutiny is with physician involvement, then it is more important to have physicians not be involved and to lose the death penalty.”

David Lubarsky, chair of the anaesthesiology department at the University of Miami in Florida, opposes capital punishment. He is a co-author of the Lancet study published last year, which analysed the amount of anaesthetic in the blood of 49 executed prisoners. Some other anaesthesiologists, also opposed to the death penalty, vehemently disagree with Lubarsky and colleagues' finding that almost half of the executed prisoners had been awake for their deaths. They question the assumption that the level of anaesthetic post-mortem would be the same as at the time of death. The dispute continues.

The research, even though it only used post-mortem records, presented its own ethical challenges. Lubarsky says he worried that if the data showed lethal injection was knocking people out without fail, he would have two opposing ethical obligations: “Would we be scientifically ethically obligated to publish the results, which might go against physicians' ethics for aiding the death penalty?” He decided to publish whatever the outcome.

Punishment by pain

Lubarsky argues that more than voluntary non-participation is called for. “I believe that if death-penalty opponents were serious, they would seek the names of the physicians who participate under the freedom of information act,” he says, in the hope that those doctors would then lose their licenses.

Many physicians who help with executions do so despite being opposed to the death penalty, however. Carlo Musso, a Georgia physician, is one of the few willing to be named. He works in jails and feels both that the death penalty is wrong and that his patients deserve a good death. Musso was reportedly summoned to Washington by the AMA as a result of going public in Gawande's article, and could be kicked out of the association.

Sara Tofte, an author of a report on lethal injection released last week by New York-based Human Rights Watch. She is among those who hope that if physicians and other experts refuse to touch the death penalty, it will be impossible for states to continue the practice. But she says she is disheartened by the research she did for the report: “I think a number of individuals say ‘Who cares if the inmate feels a little bit of pain? If they murdered someone, maybe they deserve to feel pain.’”

The argument that lethal injection is inhumane “is coming from a scientific and legal perspective — it's not a popular groundswell coming from the public”, says Dieter. He does not believe the stalemate will damage the death penalty. “Even if 100% of doctors refuse to participate, it will not be enough,” he says. “The state will do it without them.”