International Journal of Epidemiology 2001;30:403-405
© International Epidemiological Association 2001
Book Review |
Finding the Silver Lining: Review of ‘False Premises False Promises: Selected Writings of Petr Skrabanek’.
Dublin: The Skrabanek Foundation. Scotland: Tarragon Press Whithorn, 2000, pp.236, £15. ISBN: 1 870781 112.
It is rare to read a book of essays by an academic author and have an emotional, even visceral response, in addition to an intellectual one, but clearly Petr Skrabanek was no ordinary thinker or writer. A collection of his essays, published posthumously, begins with his self-description on the inside jacket of his work in Dublin as a ‘natural scientist, forensic toxicologist, doctor of medicine and connoisseur of the absurd’. This is not a job description typical of university postings, but if I had been searching for someone to fill such a role, Dr Skrabanek would have been perfectly suited to the job.
Before commenting on any of the 16 essays included in the book, particularly those that pertain to epidemiology and public health, it is necessary to consider the perspective he brings to all the topics he considers. The author deserves great credit for having found a plane of thinking about science, medicine, and public health that is distinctly different, even skewed, compared to the usual dissection of technical details of research or pragmatic questions of how to deliver medical care or set public health policy. His comments bear on those and many other aspects of the fields he considers, but he is looking at the work, the people who do the work and the implications of the work from a great distance. Although the attacks on epidemiology are much less enjoyable to read than the attacks on other fields, one is tempted to cheer for the Skrabanek as an outsider, free of the conventions and constraints of the field, able to cut through the layers of jargon and obfuscation to the essence of the matter and reveal the lack of substance underneath.
Dr Skrabanek approaches his subjects as a social observer, examining the whole system of modern science, medicine and public health. It is not a coincidence that he finds clever, apt quotes from historical literary figures and philosophers that are directly applicable to topics that range from the use of electroshock therapy and acupuncture to antivivisectionism, as well as risk factor epidemiology. He shares with those scholars an appreciation of universal themes and insight into human nature, and is unusual in recognizing that the worlds of health research and practice are in fact a part of civilization more generally. He also would appear to hold the view that science, done properly, can overcome human limitations and provide valid insights that have practical benefits. Despite some brutal criticism of epidemiologists, among others, there is a thread of optimism in the work, belief that by exposing errors we can and must do better. A constructive undercurrent is there, despite the title of the first essay in the volume, ‘In Defence of Destructive Criticism.’
Where Dr. Skrabanek's clear, witty articulation of ideas is compatible with my own views, the essays are truly delightful to read. He is irreverent, politically incorrect, and erudite in his attacks. I suspect that most readers could define a spectrum of essays in this volume scaled by level of agreement, and their appreciation of the essays would correlate closely with compatibility. His scathing attack on the logic and evidence pertaining to acupuncture was fascinating. His defense of the necessity of research with animals as a rational choice of tradeoffs in suffering, namely people with diseases versus laboratory animals, helped to organize and support views I had already vaguely held. Skrabanek eloquently considers the tension in psychiatry between looking for the organic bases for ‘diseased minds’, reflecting an attempt to become more ‘scientific’, despite our ultimate inability to ever explain the mind by studying the brain. Such essays are far more likely to solidify previously held beliefs than to change opinions. They are self-indulgent in a sense, brilliant rants, rather than a systematic, technical step-by-step reasoning that might lead the reader to abandon previously held ideas. If you agreed before reading the essay, you would be much more confident afterward, but if you disagreed, you might well become more entrenched in that contrary view.
His forays into public health policy include discussion of the history of the anti-tobacco movement. He acknowledges that: ‘Pleasurable vices, such as smoking, are, undoubtedly, at times rewarded by disease and death. But a cynic may observe that life itself is an incurable, sexually transmitted disease’. The epidemiologist tends to focus on tobacco's aetiologic role in a wide range of diseases, and considers the debate concerning health effects of environmental tobacco smoke (ETS) in such terms as measurement error and biased recall. Public health advocates focus on strategies to reduce tobacco use to achieve public health benefit, given the known devastating consequences. Skrabanek does not argue directly on either of those levels, but steps back to consider the historical tension between a seductive and enjoyable behaviour and the societal forces that have long opposed it for various reasons, including concerns with health. His view therefore cannot be countered with epidemiological evidence because he is examining the history, culture, and politics that bear on tobacco use and the current manifestations of those themes in the form of epidemiological evidence. He notes the forceful anti-tobacco stance taken in Nazi Germany and more recently in Iraq in the 1980s. His argument that the debate about ETS is a veiled attack on smoking more generally has merit, but he unfortunately extends that challenge to the next step to argue without much insight about the empirical evidence bearing on ETS and health. To make a truly engaging story built on his perception that the current controversy is largely a continuation of longstanding moral crusades against tobacco, it is essential that the evidence on ETS be unconvincing. To have highly biologically plausible, empirical evidence indicative of adverse effects of ETS (which many argue is present) would do damage to a much more interesting if somewhat misleading story. It would be a distraction from his essay to debate the evidence for or against confounding or the magnitude of measurement error in studies of ETS. Such technical fine points are off his radar screen, because his mission is so much bigger. To cogently argue his point of view, one cannot have much patience with ambiguous evidence and the subtleties in study design and interpretation that underlie that ambiguity.
An essay published in The Lancet in 1990, ‘Nonsensus Consensus’, points out with cruel humour the artificiality and arbitrariness of consensus in the face of incomplete, flawed information. It is a valuable reminder that despite the numbers of experts on a committee and the authority under which they operate, the evidence does not improve as a result of their deliberations. However, what Skrabanek fails to consider is that medical care and setting of policy must continue despite uncertainty, and decisions will be made on many issues for which the scientific evidence is severely limited. This leaves us with a menu of imperfect options, one of which is reliance on consensus among experts. One could argue in fact that the integrated, subjective judgement of a collection of experts is most needed when the evidence is most uncertain. Skrabanek shows little interest in the pragmatism required to act in the face of uncertainty, and offers no alternative, superior approaches to making wise decisions in such circumstances.
The essays specially focused on epidemiology include an historical excursion through the evidence that cervical cancer is rare among nuns and common among prostitutes, as many of us are taught in introductory epidemiology courses. Skrabanek wades through the citations of that classic epidemiological observation and finds a lack of empirical support. He contends, with some persuasiveness, that this ‘fact’ has become part of epidemiological folklore because it makes such a nice story, both as a clever inference from a simple observation and one that is consistent with the morality of sexual behaviour. A subtext of Skrabanek's essay is that there are reasons we have so uncritically embraced this ‘fact’, perhaps because it illustrates the way we would like observational epidemiology to function.
The direct attacks on the discipline of epidemiology include ‘Risk Factor Epidemiology: Science or Non-science?’, ‘The Poverty of Epidemiology’ and ‘The Emptiness of the Black Box’. Taken in the context of the other essays, the reason that current epidemiology is so annoying to Skrabanek becomes clearer and even logical. He has strong views about public health moralism intruding on the pleasures of life, so those who generate the so-called evidence to condemn those pleasures incur his wrath. He has high, perhaps unattainable standards of scientific purity, with little sympathy for generating imperfect evidence that may provide improved but still incomplete understanding. At its worst, he sees a conspiracy of incompetent, self-serving investigators generating meaningless statistical correlations that are applied by their co-conspirators in public health practice to usurp individual freedom and drain all the pleasures out of life. As a participant in this conspiracy, I cannot claim to have the neutrality required for objective examination of his attack, but I confess that I have some sympathy for that point of view. Try to step outside the world of public health and imagine the constant bombardment of information on a growing list of pleasurable behaviours now found to be bad for you, limitless contradictory advice from experts on how to live longer, more healthful lives, quotes from the latest findings of the epidemiological experts laden with incomprehensible jargon, and some glimpse of the scientific reasons those findings are often quite fallible. Add to that an enjoyment of being a contrarian, and Skrabanek's perspective becomes much more comprehensible and in fleeting moments enjoyable. Surely he was having more fun criticizing the alleged conspiracy than we are in perpetuating it.
It is difficult not to react defensively to an attack from the outside, and clearly Skrabanek is outside epidemiology. However, we would do well to expend some effort to understand his point of view because he is surely not alone. If we wish to influence the real world, we need to step back and consider how this whole enterprise is perceived and Skrabanek has given us a glimpse. One part that I believe to be especially important to examine is the willingness of some epidemiologists to leap from fallible findings to advise the public on how to fine-tune their lives for health benefit. It is not at all clear to me that we have done much of a service with such advice and may well have done a disservice in drawing attention to minor, perhaps non-existent effects of diet, pesticides, and electromagnetic fields and distracted the public from such major issues as abstinence from tobacco and use of seat belts. I tend to blame media exploitation and public irrationality, but certainly those who generate and promote their findings have a part in the system. I also share some of his libertarian views, troubled by forceful intrusions on some individual freedoms ‘for our own good.’ Those are political debates for which epidemiologists and public health workers can provide pertinent information, but considerations well beyond those fields must be incorporated and ultimately should dominate.
I strongly oppose his lack of patience, interest, and I believe knowledge of the mechanics of how epidemiology works. He might argue that the differences are slight, but observational epidemiology done well generates a profoundly different quality of evidence than epidemiology done poorly. You simply have to engage in the methodologic subtleties to make intelligent use of epidemiology or to criticize it constructively. Of course we generate conflicting findings, debate the quality of evidence, and design studies to challenge those that came before. All those are the essence of science and scientific progress.
What lessons can we learn from Skrabanek in spite of the limitations in his point of view? Any message that counters complacency has value. In addition to incorporating some insights regarding how epidemiology is sometimes perceived, we would do well to take a more systematic look at where we invest our energy and resources as researchers. In ‘The Poverty of Epidemiology’, Skrabanek examines the sea of fallible, mostly erroneous findings of epidemiology presented at the 1990 meeting of the Society for Epidemiologic Research and ridicules them individually and collectively. No doubt there is some bias in the selection of positive findings, many false positive findings, for abstract submission and in the selection for presentation. Furthermore, each fallible finding opens up other research that might be done to follow up and extend knowledge on the topic. The challenge we face is to do a better job of identifying which of those many threads is worthy of further effort and what research would make a material difference to the body of evidence. This is a different, higher standard than asking what would get the next paper published or the next project funded. Another logical consequence of his criticism is an examination of when observational epidemiology will and will not be helpful. He focuses solely on tight linkages to evidence from biology in ‘The Emptiness of the Black Box’, but some exposures and outcomes are more readily measured than others and some causal pathways are simply more amenable than others to fruitful epidemiological investigation. Skrabanek provides a forceful reminder in these essays that all is not well with epidemiology, sustaining the status quo will not serve us or the public well—we need to do much better.
Notes
NB: Copies can be obtained from: The Skrabanek Foundation, 199 Pearse St, Dublin 2, Ireland.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||