Here’s another story of politically-appointed FDA brass overruling the overwhelming consensus of its own scientific board to make a scientifically dubious decision on approval of a marketing application. The best-known – and previously unique – example of behavior of this kind was the bizarre saga of Plan B emergency contraception; now another case has surfaced involving experimental therapeutic usage of a surgical-implant device that FDA reviewers say – on the basis of the manufacturer’s own data showing no statistical difference between the experimental and control groups – is worthless. The division Director unilaterally approved the device anyway.

If “two data points is a trend”, this establishes a very puzzling and very disturbing trend at the FDA. It is fast squandering its reputation as “the world’s most reliable medical safety agency”.

A top federal medical official overruled the unanimous opinion of his scientific staff when he decided last year to approve a pacemaker-like device to treat persistent depression, a Senate committee reported Thursday.

The device, the surgically implanted vagus nerve stimulator, had not proved effective against depression in its only clinical trial for treatment of that illness. As a result, scientists at the Food and Drug Administration repeatedly and unanimously recommended rejecting the application of its maker, Cyberonics Inc., to sell it as such a treatment, said the report, written by the staff of the Senate Finance Committee.

But Dr. Daniel G. Schultz, director of the Center for Devices and Radiological Health at the agency, kept moving the application along and eventually decided to approve it, the report said.

That approval did follow the backing of a divided F.D.A. advisory committee. Still, the Senate committee, which for two years has been investigating the decision-making processes at the F.D.A., could find no previous instance in which the director of the center had approved a device in the face of unanimous opposition from staff scientists and administrators beneath him, the report said.

There is no explanation what the “divided advisory committee” refers to. It is clear that the actual scientific review committee was unanimously against the device, and it’s not hard to see why:

The vagus nerve stimulator is surgically implanted in the upper chest, and its wires are threaded into the neck. Batteries in the device stimulate a nerve leading to the brain.

The nerve stimulator has been approved since 1997 for the treatment of epilepsy in some patients. Common side effects include voice alteration, increased cough, shortness of breath, neck pain and difficulty swallowing. The device has also been linked to rare reports of death, heart problems and vocal cord paralysis.

When some epilepsy patients reported that their moods had changed after receiving the devices, Cyberonics, based in Houston, implanted them in 235 depressed patients and turned the machines on in half of them. After three months, the two groups were equally depressed. The trial had failed.

Cyberonics then turned the devices on in all 235 patients and determined that 30 percent showed significant improvement after six months or more. Without a control group, however, it was impossible to determine if the device had caused the improvement.

Cyberonics is now defending the device as “the only safe and effective treatment option ever specifically developed, studied, F.D.A.-approved and fully informatively labeled for the treatment of chronic or recurrent treatment-resistant depression.”

This is shocking. If the description above is correct, the company not only failed to demonstrate any therapeutic effect from their device, they actively manipulated the trial to obscure disconfirmatory data and then claimed a positive result based on an uncontrolled trial which they had purposefully initiated after the controlled trial failed. This goes beyond a merely questionable device: based on this news report (which may not be reliable, it is important to remember), this appears to be deliberate scientific fraud.

The company’s behavior is absurd: they deliberately eliminated the control group and then claimed a positive result that was only apparent in the trial group when no control group was available for comparison – doing so after the controlled trial demonstrated no distinct effectiveness in the experimental group! This is just outrageous. And their claimed “benefit” is laughable – famously, about one-third of untreated psychiatric patients show improvement over time, and at any rate you would expect some percentage of the experimental group to get at least somewhat better after 6-9 months; to openly embrace the post hoc fallacy by claiming it was due to your device simply because something happened (as if nothing would have happened without the device – when we know patients change naturally), is nonsense. (Based on the company’s reasoning, it would be equally valid to claim that the device prevents improvement in 70% of patients!) Again, this is indistinguishable from fraud.

Even ignoring the company’s outrageous behavior, however, that an FDA Director overruled the scientific panel, and in such an egregiously obvious case!, is worse yet. It’s hard to tell what motivation this Director would have had – it doesn’t appear to be a case of pandering to the right wing, as with Plan B – and there is yet no reported evidence of payoffs or conflicts of interest. But clearly something went very wrong on this case, and clearly, too, the FDA has become the kind of organization at which such things can and do happen.

Once that would have been unthinkable. Under the current administration – anti-science, pro-business, and with a conviction that truth is what you say it is – it appears to have become policy.

Kathryn Hinsch makes a good point at the Women’s Bioethics Project blog:

Much of the “assisted death” debate gets reduced to two questions: “Do people have a right to commit suicide?” and “Should we allow physicians to assist in hastening death?” But before we tangle with those tough public policy questions, it is important to ask “how might people’s different life circumstances impact the issue?”

Looking at these questions from a gender perspective will be imperative as we move forward in crafting new laws. There are some key facts that make a woman’s end-of-life decision quite different from a man’s—women on average live longer than men. Additionally, women are more likely to be impoverished, receive inferior health care, experience poorer pain relief, and are two times as likely to suffer from depression as men. Women, who have often lost their life partner by the time they face debilitating disease, may feel a stronger cultural pressure not to be a burden on their families. All these factors must be considered when crafting a policy to allow “physician assisted death.”

Looking at gender implications is just one step in a thorough public policy analysis of this issue. We also must look at the implications of physician assisted death for disabled, poor, and minority populations. From some groups, the fear that death with dignity could quickly lead to duty to die is not an unfounded fear and something we must be vigilant to prevent.

She’s right that it’s traditional to boil bioethics issues down into clashes of abstract principles. Feminist philosophers in particular have been invaluable in demanding “thick descriptions” of cases, to situate them in a fuller human context and identify the systematic inequities and pressures that impinge on the people affected. Just as career opportunities turned into the double burden of a full-time job plus sole responsibility for homemaking, for many women, the “empowerment” of autonomy may merely mean the reduction of support systems in the healthcare setting if the real burdens people face are not kept in mind.

As someone who is partial both to strong autonomy rights and to the “airy principlist” approach to ethical problems, this is a useful reminder that principles do not play out in real lives the same way for everyone. The practical impact of general principles must always be the touchstone for practical policymaking.