Thursday, October 18, 2007

Memory Hole (18 October): Like deja vu all over again

6 years ago today: Jan Hendrik Schoen paper published - a scientific fraud

In the 18 October 2001 issue of Nature, Jan Hendrik Schoen of Bell Laboratories published an astonishing paper describing a transistor in which a single molecule acted as a "switch".

Schoen was named as one of science's top young innovators. During 2001, Schoen churned out a new paper on average every 8 days. However science star soon turned to Icarus.

In April 2002 other researchers noticed a strange similarity between a figure in the Nature paper and another figure Science had published on a different device. Other apparent problems were soon found, and then many more. Ultimately 24 critical problems involving 25 papers and 20 co-authors were examined. In at least 16 of these cases Schoen had falsified or fabricated data. Crucially he had also deleted his data files, making it impossible to check.

All nice and tidy and easy to define as misconduct within the restrictive definitions.

Schoen said, "I have to admit that I made various mistakes in my scientific work, which I deeply regret,... However, I would like to state that all of the scientific publications that I prepared were based on experimental observations". Schoen apologized for his "mistakes". The words are familiar to me.

But it gets more difficult when the culpability of coauthors is raised. Although it was determined that there was "no evidence of misconduct by any of Schoen's co-authors" it is not clear that any of them bothered to check anything. That seems to be culpable negligence.

The coauthors were "certainly happy to bask in the glow when things looked wonderful. But you can't have it both ways and not accept some level of the burden of responsibility".

Source: "Bell Labs Fires Star Physicist Found Guilty of Forging Data" Science 298, 4 Oct 2002, 30-31

45 years ago today: Nobel Prize to Watson, Crick and Wilkins

On 18 October 1962 Dr. James Watson Dr. Francis Crick and Dr. Maurice Wilkins were awarded the Nobel Prize for their determination of the molecular structure of DNA. Watson and Crick had received crucial unpublished data from Rosalind Franklin's lab, perhaps without her knowledge.

22 years ago today: Dispute Over Access to Reye's Study Data

On 18 October 1985 Science reported on a dispute over raw data. Many epidemiological studies have shown that aspirin intake in children is a trigger for Reye's syndrome. Following advice about aspirin use in children, the incidence of Reye's syndrome decreased about 10 fold from a high of about 1/100,000 children per year (with mortality around 50%).

Plough (maker of St. Joseph's aspirin) were sued by a family. In their response to the case, Plough demanded that the US government release to it the raw data it has on aspirin. This was refused.

I think the government was entirely wrong to have refused (within the limits of patient confidentiality). Science which cannot be scrutinized is not science at all.

Schering-Plough is of course the company that writes £10,000 checks to doctors in exchange for prescribing their drugs. They also manufacture:
  • Remicade (infliximab)
  • Intron A & PEG-Intron(Interferon Alfa-2b)
  • Levitra (vardenafil)
  • Zetia (ezetimibe)
  • Vytorin (ezetimibe/simvastatin)
Schering-Plough sell these products under the banner of science. They will of course supply raw data to scientists, prescribers and patients who wish to check the scientific and ethical basis of their prescribing. Such is the way of proper science and the meaning of that banner.

Source: "Dispute Over Access to Reye's Study Data," Science 230 (18 October 1985), pp. 297 298.

8 years ago today: Probe into deaths of 25 experimental subjects at Porton Down

On 18 October 1999 the BBC reported that police were investigating the deaths of 25 ex-servicemen in the UK who were used as guinea pigs in germ warfare experiments at Porton Down (the defence research establishment in Wiltshire). The investigation was sparked by a probe into the death of airman Ronald Maddison in 1953.

Maddison, then 20, died after 200mg of deadly nerve agent Sarin B was dripped onto a piece of uniform stuck to his arm.

"The investigation is hugely sensitive because it could lead to the prosecution of former Ministry of Defence staff"

Solicitor for relatives of the Porton Down victims, said the "guinea pigs" were tricked into believing they were taking part in research into the "common cold". Gordon Bell, 61, says he was paid two shillings. "In one test we had to stand in front of a stream of gas which I could not stand for more than a minute," he said. "It was a dirty trick, plain and simple."

Source: BBC

Sheep or Cow

6 years ago today: UK Government science project confuses cows and sheep

On 18 October 2001 the UK government announced that a critical four-year government study looking at infection of sheep with BSE (mad cow disease) was dropped. The government claimed the scientist involved had been testing cow brains thinking they were sheep brains.

They admitted that testing on the wrong tissue had been carried out for the past five years. The supposed discovery of the mixup took place just two days before the research was due to be presented to the Spongiform Encephalopathy Advisory Committee. A positive result without the supposed mix-up would have led to millions of UK sheep being culled.

The UK's Rural Affairs Secretary, Margaret Beckett was accused of trying to hide news of the supposed confusion.

5 years ago today: Throwing science down the toilet - Judge Henry Kennedy and pediatric rule for regulators

On 18 October 2002 a federal District Court struck down the Food and Drug Administration's Pediatric Rule, saying that the agency did not have the authority to require drug makers to test some of their products for childhood use.

Several groups, including the "Competitive Enterprise Institute" sued the FDA to overturn the Pediatric Rule, arguing that it improperly expanded the agency's authority because it opened the door to greater oversight of "off label" uses of drugs.

Amazing stuff. Perhaps all drugs should be tested on 80 year old Mongolian men so that "off label" (read non scientific) use could be more widespread. The level of evidence for consumer products will then be similar to any other type of quack medicine, but mis-sold under the banner of "science".

Judge Henry H. Kennedy Jr., you have behaved like an impetiginous inquinate orosscrolest and have approached a problem of science with mental discalceation.

Sources: Washington Post, "Court Strikes Down FDA Rule", Oct 18 2002; Vera Sharav, Children in Clinical Research: A Conflict of Moral Values, American Journal of Bioethics 3(1) 2003

2 years ago today: FDA suggests that psychotropic manufacturers test their drugs

On 18 October 2005 the Wall Street Journal reports that the FDA has sent out feelers to drug manufacturers to provide "longer-term efficacy data" for psychotropic drugs.

So what happened to the feelers?

Are patients are told that most clinical trials of psychotropic drugs last only a few weeks, and that a majority have shown that the drugs work no better than placebo, and have non-efficacy to rival homeopathic medicines (which in turn work as well as placebo)?

High science.

Source: Wall Street Journal, "FDA May Require Longer Studies Before Clearing Psychiatric Drugs" Oct 18, 2005

2 years ago today: Serono plead guilty to scam pseudoscience

On 18 October 2005 the company's Serono Labs agreed to plead guilty to conspiracy to market the drug Serostim by "supplying doctors with scam "diagnostic software" that was about as science-based as colonic irrigation. Serostim is an injectable form of growth hormone marketed for AIDS wasting.

Basically Serono created some crappy software that sought to "redefine AIDS wasting". This computerized medical test ostensibly determined "body cell mass." The results signaled that patients had lost "body cell mass" and were "wasting", even if they had lost no weight or had actually gained weight. With these test results in hand, doctors would justify prescribing Serostim to treat the supposed problem - at $21,000 a treatment.

The company also agreed to plead guilty to the other part of the scam - offering doctors all-expense-paid trips to a medical "conference" in Cannes, France, in exchange for writing more prescriptions for Serostim.

85 percent of the prescriptions written for Serono's growth hormone Serostim weren't necessary.

$704 million penalty for drug maker.

But nobody went to jail (and what of the doctors who colluded with the scam?). I wonder if any were in the UK?

Sources: "Settlement in Marketing of a Drug for AIDS", New York Times, Oct 18, 2005; also Health care Renewal

1 year ago today: Cochrane verus industry influence on haemoglobin targets

On 18 October 2006 a Cochrane review reported that higher haemoglobin targets in dialysis and pre-dialysis patients is not associated with lower mortality. This clinical problem assumed some importance as one of inappropriate commercial influence over rational therapeutic strategy ([Link], [Link]).

Source: Strippoli GFM, Navaneethan SD, Craig JC. Haemoglobin and haematocrit targets for the anaemia of chronic kidney disease. Cochrane Database of Systematic Reviews 2006, Issue 4. Art. No.: CD003967.

Earlier|Later|Main Page

No comments: