Trust me, I’m a Scientist!

According to a MORI poll, nine out of ten people in England trust scientists. That’s a higher confidence level than for most professions. Apart from a few bad guys in history and fiction, the public often caricatures scientists as benign boffins with wild hair and white coats who diligently study nature’s most vexing mysteries. Sorry to quash this rosy view.

A Dr. Han of Iowa State University was recently sentenced to nearly five years in prison and a $7 million fine for scientific misconduct. He admitted spiking blood samples with antibodies to make an HIV vaccine look effective. In Japan, a researcher was so ashamed that he hanged himself after a junior researcher was accused of falsified data. The study had claimed to find a short-cut for making pluripotent stem cells in regenerative medicine and was published in a highly reputable journal.

These cases came to light because they were making large and surprising claims. Falsification of esoteric research is more likely to go unnoticed, and years of effort and funding can be wasted when researchers follow a false lead.

A century ago in Britain, the Piltdown Man was alleged to be the “missing link” in human evolution, but it was forty years before the hoax was uncovered and paleontology could reset its course. At the same time in Vienna, Paul Kammerer claimed to support the discredited Lamarckian theory of evolution based on forged evidence from mid-wife toads. He committed suicide. Years later, Arthur Koestler wrote a spirited defense suggesting that a Nazi competitor may have fiddled with the evidence to discredit Kammerer as a Marxist sympathizer.

Most researchers accused of wrongdoing nowadays are young, highly talented and ambitious people striving for recognition. Their watchwords are “publish or perish.” They often work in competitive fields funded by an insecure succession of grants; they yearn for a tenured job with the other rewards that a big breakthrough could give them. Cheating sometimes starts when an unintentional slip-up is followed by a cover-up; sometimes the researchers think they know the answers before the data are in and succumb to a temptation to short-cut by fabricating results. Being first is everything in science, and there are no prizes for being runner-up.

Perhaps there are more cases of fraud in biology and biomedical science than in physics because the research teams are smaller, have less oversight and more liberty for individual endeavor. Besides, modern biology depends heavily on image data which can easily be altered with Photoshop. And lastly there are pressures and temptations when research offers commercial applications and industrial sponsorships.

Once trust is lost, it is very hard to regain; the scientific community treats offenders like lepers, at arm’s length.  No matter how talented the guilty party, no one wants to hire a scientist barred from research grants, which are life-blood for scientists. Unfortunately for those connected in any way with the guilty there is collateral harm even if they are completely innocent.

Trust a scientist!
Trust a scientist!

Sensational cases are much rarer than the tip of an iceberg. Most doubtful science goes unreported, hardly crimes at all but flaws of execution and fallacies of interpretation. Yet, the scale is humungous. A US study in PLoS Biology estimated that $28 billion is wasted annually on biomedical research that cannot be replicated. In an influential paper from Stanford University, John Ioannidis (2005) concluded that “in modern research, false findings may be the majority or even the vast majority of published research claims.” Financial interests, prejudices and fields that are “hot” and “competitive” are common factors in the stock of bad science.

The US Office of Research integrity, the NIH, journal editors and universities have all started to wonder what it means and what should be done. Only the most important findings attract the funding needed for confirmatory studies. Notable papers that can’t be replicated for any reason are retracted, meaning that the journals in which they were originally published disown them and purge them from the electronic record.

To bring these cases quickly to the notice of researchers, a journal editor and a medical journalist have created a blog called Retraction Watch. It is shocking to see 50-60 retractions recorded every month, including some from prestigious journals like Nature, Science, Cell, The Lancet and The New England Journal of Medicine. Neither does research originating in an ivy-clad laboratory guarantee authenticity. Editors who contemplate lowering the critical bar for flashy papers and famous names beware!

I dare say that most retracted papers and others that deserve to be printed with disappearing ink were authored by people who never intended ay deception. Their fault was in the design and execution of the experiment or field trial. The underlying reasons are legion: sloppy methods, data massage, flawed design, small numbers of observations, weak statistics, poor controls, unconscious bias, and attention is now being given to poor quality research materials.

Sometimes a study is hard to replicate because it requires a rare skill, but the excuse is getting weaker by the year as biology depends more on technology to eliminate the human factor. Koestler wondered if this explained why the toad experiments could not be repeated in other laboratories. Or perhaps Kammerer was ahead of his time with inklings of the yet unnamed field of epigenetics, but his story is usually given as a warning against forfeiting scientific trust.

Before admission to most professions, novices are required to swear adherence to an ethical code. Doctors swear “first do no harm;” lawyers recite a long oath upholding the laws of the state; and Roman Catholic bishops vow to be faithful to “the Church and Our Lord.” Unlike those ancient professions, scientists are not required to make any oath, and few attend classes in ethics or the philosophy of science (except perhaps in France). Perhaps they should.  Swearing to uphold an ideal does not guarantee integrity, and there are plenty of malpractice doctors, crooked lawyers and agnostic priests, but it might focus young minds on what science stands for and the trust that the public expects. Surely it is truth in endeavor.

Next Post: Virginia Nature Journal for July