Your complimentary articles

You’ve read one of your four complimentary articles for this month.

You can read four articles free per month. To have complete access to the thousands of philosophy articles on this site, please

SUBSCRIBE NOW

Bioethics & Medical Ethics

Can We Trust Medical Science?

Simon Kolstoe says that all is not well with medicine.

It is said that philosophy is driven by concerns about mortality. If there is even a small grain of truth in this, then from a purely self-interested perspective the philosopher has good reason to be supportive of the work of doctors and health scientists. After all, it is rather useful to have access to headache pills and other over-the-counter medicines to help clear the mind to think when sitting huddled in an armchair and suffering from a cold; and having paramedics on the end of a phone, and a well-equipped hospital at the end of the ambulance ride are particularly useful when more serious ailments strike. Philosophers are reliant upon modern medicine to stave off their demise at least long enough to have the chance to write about it.

We all rely on our doctors remaining up-to-date with new advances and prescribing them when our need arises. To grossly oversimplify, modern medicine is developed in the following way. First, an often state-funded academic working in a laboratory makes a discovery about how some small part of the body or the immune system works. He or she then talks to clinical colleagues, who help develop the discovery into a potential medical application that can be patented. Following the patent, a flurry of papers are published describing the invention to the world, hopefully drawing the interest of a pharmaceutical company, who license the invention for clinical investigation. Next, a ten-year period of silence ensues, whilst the company spends £100 million or so developing the idea into a useable product and proving that it works through clinical trials. Finally, for the one-in-ten discoveries that make it through this process, the new drug or treatment is launched in a torrent of publicity as the company endeavours to convince doctors to prescribe, and health systems to pay for, the new miracle cure. If everything works as planned, patients benefit, the medical profession gets a new tool, the pharmaceutical company makes a hefty profit, and the original academic gets a pat on the head, and if in the UK, a REF impact case study (if you don’t know what this is, feel pleased that you live in the real world!).

The entire process of developing medicine is a triumph of enlightenment philosophy dating back to William Harvey (1578-1657), who was one of the first people to apply the new mechanical philosophy to the human body, and so give birth to the modern understanding of medicine. However, the entire medical revolution is critically dependent upon one thing – the accurate communication of scientific discoveries to the clinicians who are treating patients. After all, it seems hardly worthwhile finding a cure for cancer if you do not tell any doctors about the discovery.

A number of methods have been developed to aid this communication in modern times, starting principally with peer-reviewed scientific literature. This is then translated into educational courses for clinicians, summarised into pharmacopoeias, information bulletins, news articles, and finally popularised on websites, blogs, and tweets. Whilst there are certainly critics of ‘evidence-based medicine’, as this approach is called, and indeed there are quacks and charlatans who try to subvert the system for their own ends, the successes of the system for advancing medicine are fairly obvious, at least in terms of ever-decreasing death rates (that is to say, an increasing average age of death), and improved strategies for people to live more comfortable lives despite their ailments.

The old-school marketing of pharmaceuticals
Emili Casals 1882

Doctoring The Evidence

Yet, despite the astonishing success of evidence-based medicine, there are rumblings that all might not be right within the system. Let me tell you about three examples.

Paroxetine (trade names Paxil and Seroxat) is a type of anti-depressant drug commonly referred to as an SSRI (Selective Serotonin Reuptake Inhibitor). It has been used extensively since its introduction in 1992, resulting in tens of billions of dollars in profit for its makers, GSK (GlaxoSmithKline). However, ten years after its introduction, an internal whistleblower triggered an investigation that resulted in GSK having to pay a three billion dollar fine and answer criminal charges (see ‘GlaxoSmithKline to Plead Guilty and Pay $3 Billion to Resolve Fraud Allegations and Failure to Report Safety Data’, United States Dept of Justice, 2012). The charges concerned illegal marketing of several drugs. In the case of paroxetine, it had been promoted for use by depressed under-eighteens, despite three studies suppressed by the company having shown that the drug was no better and in some cases was notably worse than a placebo. Perhaps most infamously, ‘Study 329’, conducted in North America between 1994 and 1998, not only failed to show any benefit to the 188 teenagers taking the study drug, but recorded five times more serious adverse events in participants on the study drug compared to the placebo arm, including an increase in suicidal behaviour, and seven hospitalisations for that (‘Restoring Study 329: Efficacy and harms of paroxetine and imipramine in treatment of major depression in adolescence’, British Medical Journal 351:h4320, 2015). Given the huge profits being made from this drug, GSK hired a PR company and paid them $17,250 to ghost-write a scientific paper describing Study 329 (‘Adolescent Depression Study 329: Proposal for a Journal Article’, Scientific Therapeutics Information, Inc, 3 April 1998). The PR company played down the findings, not reporting many of the outcomes or the serious adverse events, and concluded: “Paroxetine is generally well tolerated and effective for major depression in adolescents.” This dishonest report, once published, was used by the drug company in a marketing campaign and provided the evidence base for clinicians to confidently prescribe this drug to their vulnerable young patients millions of times before GSK were finally brought to account (see ‘Drug company experts advised staff to withhold data about SSRI use in children’, Canadian Medical Association Journal, 170(5):783, 2004).

Lorcainide was an anti-arrhythmic drug developed to treat abnormal heart rhythms after a heart attack. In 1993 a study was published describing how nine men had died on the drug arm of the original clinical trial, compared to only one on the placebo arm. Unfortunately, this 1993 paper was reporting the results of a study that had actually been conducted way back in 1980. The authors of that study had tried to publish it at the time, but as a negative result of a trial on a drug whose development had, in any case, been halted, it was rejected by the major medical journals. (‘The True Lorcainide Story Revealed’, BMJ, 350:g7717, 2015.) Unfortunately, over the next decade a number of other drug companies came up with the same idea. The idea of an anti-arrhythmic given after a heart attack apparently seemed quite sound to all involved, and since there were no specific warnings about this approach in the literature, these companies rushed their drugs to market. It has been estimated that about a hundred thousand people in the US alone died unnecessarily (Ben Goldacre, ‘What Doctors Don’t Know About The Drugs They Prescribe’, TED, 2012). By 1993 the dangers of this approach were all too clear, and the authors of the original 1980 clinical trial finally managed to publish their results.

In 2009 fear gripped the Western world due to an outbreak of H1N1 – ‘Swine Flu’. As no vaccine was immediately available, governments stockpiled a general anti-viral drug called Tamiflu (Oseltamivir in the US). As of June 2009, the British government estimated it had enough Tamiflu to cover 50% of the population, and it continued buying it until 2013 in order to cover 80% of the population, at a cost to taxpayers of £424 million. It was unfortunate that £74 million of this stock had to be destroyed because of poor record-keeping (‘Access to clinical trial information and the stockpiling of Tamiflu’, National Audit Office, HC 125, 2013). It was even more unfortunate that a systematic review of the evidence behind the clinical efficacy of Tamiflu found that eight of the ten trials used by the company to show that the drug was useful in preventing complications such as pneumonia had never actually been peer-reviewed or published. Instead clinicians, and eventually governments, were relying on a marketing spiel claiming successful trials of this drug, rather than considering the actual evidence concerning its efficacy. Furthermore, when systematic reviewers wrote a report considering the two published trials alongside actual clinical experience during the Swine Flu outbreak, they reported that Tamiflu showed very few clinical benefits at all (‘Neuraminidase inhibitors for preventing and treating influenza in adults and children’, Cochrane Database Systems Review, 4: CD 008965, 2014). This was somewhat more than unfortunate given the amount of taxpayers’ money that had been spent stockpiling the drug.

It would be nice to think that these were three isolated examples of what has rather euphemistically become known as ‘reporting bias’ or ‘publication bias’. These are the names given to the perhaps understandable tendency to report successful drug trials more widely and more fully than unsuccessful ones. However, a recent review by German researchers shows evidence of reporting bias in treatments for “depression, bipolar disorder, schizophrenia, anxiety disorder, attention-deficit hyperactivity disorder, Alzheimer’s disease, pain, migraine, cardiovascular disease, gastric ulcers, irritable bowel syndrome, urinary incontinence, atopic dermatitis, diabetes mellitus type 2, hypercholesterolemia, thyroid disorders, menopausal symptoms, various types of cancer (e.g. ovarian cancer and melanoma), various types of infections (e.g. HIV, influenza and Hepatitis B), and acute trauma” (‘Reporting bias in medical research – a narrative review’, Trials, 11:37, 2010). In terms of drug compounds, as well as the SSRIs, lorcainide and Tamiflu discussed above, reviews of reporting bias have mentioned many others, even including some well-established drugs such as quinine. Not that any of these treatments are necessarily dangerous or ineffective, rather, that trials have been conducted that are either not in the literature at all, or not fully reported. This suggests that there is information about these treatments that is deliberately not available to your doctor when he or she fills out their prescription for you.

Philosophical Issues

When people find out about this huge but relatively unknown problem, the common response is outrage. Indeed, the doctor and journalist Ben Goldacre has become a mini celebrity after his TED talk on this topic and his book Bad Pharma (2012). There is also a growing grass-roots movement spearheaded by the organisation Sense About Science with their ‘Alltrials’ campaign that seems to be making progress in getting research funders, journals, and even pharmaceutical companies, to commit to making all their research findings available. In this regard it is perhaps notable that GSK have gone beyond almost all their peers in both effort and expense to ensure that they are not again culpable. Likewise, stung hard by criticism that they were not handling results appropriately, medical journal editors have created the CONSORT Statement binding them to stricter standards, and have even started journals specifically in order to publish negative results. Regulators have also begun to act in some countries by changing policy to now require all clinical trials to be registered on publically accessible databases prior to recruiting a single patient (although notably this is only for new trials, not historical ones, on which the rationale for the majority of medicines in present-day use are based). In terms of action on this issue, the Cochrane Library is the real hero, in its support of systematic reviews into drug treatments, gathering much of the evidence that has now exposed the true scale of the problem at the heart of modern medicine.

Yet from a philosophical perspective, the righteous anger and outrage felt by many as they discover the issue and the extent of reporting bias is rather interesting.

Almost everyone reporting on the issue, including me, immediately considered the problem to be a matter of research ethics, and started to question how or why the system had gone wrong. Indeed, improving the review and monitoring processes by research ethics committees has been suggested as one potential long-term solution to the problem. However, it’s philosophically interesting that most people automatically consider this to be a problem of ethics, because it implies that there’s generally assumed to be something different about the way drugs or medical treatments should be reported and marketed in comparison to other consumer products such as shampoo or computers. But what creates this difference?

One explanation might be that since medical products directly affect people’s lives and wellbeing, a higher ethical standard is expected from those who make and sell them. This assumed obligation is perhaps manifest in an expectation that such companies will investigate their products more thoroughly, communicate their findings more clearly, and perhaps market them more honestly than manufacturers of other types of products. But after reflecting on this issue, I am not sure that such a neat argument is possible. If the moral obligation arises out of the effect on people’s lives and wellbeing in terms of health benefits to a great many people, then surely the manufacturers of cigarettes, alcohol, sugary drinks and pastries have a similar, if not greater, moral responsibility? After all the negative health effects of these products impact far more people than have ever taken, or are likely to need, SSRIs, lorcainide, and Tamiflu. Likewise, the oil industry, car industry, airlines, and shipping companies, create a great deal of atmospheric pollution that both cause and significantly exacerbate life-limiting health problems. What exactly is it then about medicinal products that make them any different, from an ethical perspective?

Perhaps, rather than their actual effect on people’s lives, we expect medicinal products to be better investigated and reported on than other products because one has an expectation that medicines will make you better or healthier; and therefore any publication bias or spin goes directly against the ethos of the product itself. But once again, medicines are not unique in this regard. There are many food or lifestyle products that make exactly the same claims and employ the same sort of spin without this activity receiving the same moral and ethical disapproval. Maybe then it is just back to pure self-interest, in that we would rather not be in the position of being sick and taking a drug that does not work. But again, how many times have we bought important products that promise much and deliver little – for instance, children’s car seats that do not meet safety standards?

So, although there’s a significant reporting bias problem in medicine that leads to ethical outrage, it’s not immediately clear why medicines occupy a different ethical sphere to many other consumer products. There certainly does seem to be an instinct that makes us want to hold pharmaceutical companies to higher ethical standards; but I’m not sure a philosophical justification of our fury about reporting bias in medicine is quite as easy as many might think.

A Hard Pill To Swallow

Perhaps the core problem with reporting bias is that we live in a consumerist and relatively free market society where products have to compete against each other, be they fizzy drinks or heart medication. As a result we have a cultural and economic system that encourages marketing spin. If we think that medicines are in a different ethical domain, perhaps then we need to lobby for health research to be taken outside the market environment altogether. But, lest we get carried away, is society really prepared to pay the significant cost of developing new medicines through tax revenues rather than company accounts? Although it is unfortunate, and certainly something to be minimised where possible, perhaps we just have to accept that reporting bias in medicine is an inevitable product of our wider cultural system.

Simon Kolstoe is an academic biochemist whose work in drug development has inspired thoughts on ethics.