There seems to be a lot of fraud these days, but perhaps there always was; maybe it was just that we more naïve in those days. As soon as the Volkswagen scandal broke, the personal-injury lawyers were out in force — a group whose activities are usually morally fraudulent if not illegal.
Research fraud in medicine is quite common and falls into two main categories:
- Fraud perpetrated by individuals taking short cuts to a brilliant reputation, or rather a reputation for brilliance.
- Fraud perpetrated by drug companies attempting to prove that their product developed at such huge expense is better, safer and more therapeutic than any other product on the market.
Although I have spent much of my career exploring the less meritorious aspects of human conduct, there is a type of research fraud that I had not suspected to exist until I recently read an article in the New England Journal of Medicine. Once you know that volunteers for pharmacological experiments are paid, it becomes obvious, and I feel slightly foolish for not having realized it before: the volunteers also commit fraud.
Some exaggerate the severity of their symptoms so that they are included in a study, and some do not reveal that they are taking prescription drugs, allowing chemical interactions which could alter the results of the experimental drug by more than one possible mechanism and in more than one direction. Others conceal (not surprisingly) that they are taking controlled substances. Generally speaking, the word of potential subjects for experiments is taken at face value; they are not tested because it would be too expensive to do so. Research is quite expensive enough to conduct without this added burden.
One study found that of 100 research subjects who had enrolled in two trials in the last year, or three in the last four years, many had lied about themselves. A quarter had exaggerated their symptoms in order to be included and a seventh had claimed to have health problems they did not have. Nearly a third concealed health problems they did have; more than a quarter concealed the use of prescription drugs and a fifth the use of illicit drugs. And 43 of the 100 people failed to mention that they were currently enrolled in another trial at the same time.
One cannot conclude from this that the general population is appallingly dishonest because people who seek to enter trials for pay are not representative of the population as a whole. Nevertheless, where scientific results are founded on the use of paid subjects, the results are likely to be skewed.
For example, if a person has exaggerated his symptoms at the outset of the trial, it is likely that he will also exaggerate the benefit of both the placebo and the active drug. Those who lie in this fashion may not be equally distributed between the placebo and the active treatment, so the benefits of the drug might be either inflated or underestimated. At any rate, the results would not be trustworthy. This is especially so, of course, in the case of diseases whose symptomatology correlates poorly or not at all with any objectively measurable biochemical marker.
One way to lessen the problem would be to make payment contingent on the truthfulness of the experimental subject. In combination with random tests, this might be enough to deter such fraud. However, it would not eliminate it entirely, for some fraud is probably disinterested, so to speak: committed for the sheer pleasure of committing it, as an end in itself or to fool people of supposedly superior intelligence. One should never underestimate the perversity of the human soul. Where, indeed, would I be without it?