Most of the articles I write concern unconventional or alternative therapies. The primary reason for this is that the overwhelming majority of the information available to pet owners about such therapies comes from practitioners and true believers, often with a commercial interest in these therapies. People deserve an independent, skeptical, science-based perspective they can consider when deciding what’s best for their pets, and since there are almost none of these, I have done what I can to provide one.
One of the more superficially reasonable objections to this focus on alternative therapies is that there are many problems with the evidence concerning conventional medicine as well, and there is a need for skeptical and science-based evaluation of all our practices. This is true, and I do engage in such critical appraisal in my daily work, through organizations I belong to, such as the Evidence-Based Veterinary Medicine Association, and sometimes here in this blog. The pharmaceutical reps that come to our practice are no happier to see me than are the chiropractors!
However, the existence of unsolved problems in science-based medicine does not reduce at all the need for critical evaluation of alternative therapies. The limitations of the evidence for conventional practices does not make the implausible plausible, the unproven proven, or the outright ridiculous reasonable. To suggest that no one should critique alternative medicine without giving equal time to criticism of conventional therapies is a fallacy (tu quoque) and a distraction from the weaknesses of the claims made for alternative therapies.
The other problem with the “glass houses” argument is that there is, in fact, much criticism of conventional, science-based medicine coming from within mainstream medicine. Plenty of people are working on identifying and fixing many of the problems with the development and delivery of science-based medical care, and while perfection is nowhere in sight, it is possible to track improvement over time (albeit often frustratingly long periods of time). The folks promoting alternative medicine seem, at least in public, far less willing to criticize their own colleagues or take on the dramatic and egregious problems with claims not based on good evidence within their fields.
A shining example of critics of alternative medicine and proponents of science-based medicine taking on the task of setting our own house in order is the recently published Bad Pharma by Ben Goldacre (recently published here in the U.S., at least, though it’s been out in the U.K. for a while).
Dr. Goldacre, a British academic physician, has been a long-time critic of nonsense of all kinds in all areas of medicine. He is a prominent, articulate skeptic who has given much grief to promoters of homeopathy and other forms of alternative medicine through his well-researched and well-reasoned columns in the Guardian newspaper, collected in his previous book Bad Science. But he has never spared the pharmaceutical industry, government regulators, or his colleagues in journalism and medicine from direct criticism when they fail to meet his standards for a scientific, evidence-based approach to medicine.
This book provides a set of cogent arguments and a good collection of empirical evidence to illustrate the deep flaws in our system for establishing safety and efficacy of pharmaceutical therapies and for making clinical decisions about the best use of these medicines. It is not without flaws, which I will discuss, but overall it is a readable and important contribution to the betterment of science-based medicine, and I encourage everyone involved in any aspect of medicine, as a researcher, doctor, or as a patient, to read Bad Pharma. And since we are all patients at some point in our lives, that means everyone!
The book has been reviewed in fair detail at Science-Based Medicine, so I will focus on making some general comments about the strengths and weaknesses of the case Dr. Goldacre makes.
Strengths & Weaknesses
The strongest part of the book by far is the analysis of how incomplete and potentially misleading the evidence is from clinical trials sponsored and conducted by the pharmaceutical industry. Dr. Goldacre does a brilliant job of illustrating how a systematic, comprehensive, devastatingly effective system for misleading regulators, doctors, and the public about pharmaceutical therapies can exist without widespread, deliberate lying or an omnipotent conspiracy involving industry, government, and the medical professions. In the sneaky way so common to corporations generally, the pharmaceutical industry manages to rig the game through small, often seemingly innocent steps that aggregate themselves into a deeply flawed system without much if any need for shadowy conspiracies and puppet masters.
Oh, there’s undoubtedly individual malfeasance involved, of course. Much of what we know about how the system works comes from internal documents made available as part of legal cases against the big drug companies. However, the vast majority of individuals working in this industry are smart, well-intentioned, honest people trying to develop new and better medicines. The importance of recognizing this lies in understanding that the way to fix the problems of this industry involve much more than simply finding and cleaning out the bad apples. The system for generating and distributing research information about drugs has to be designed to compensate for the often innocent and unconscious biases that tilt the findings in a direction favorable to the companies even when there is no deliberate program to do so.
Dr. Goldacre does offer some reasonable proposals for achieving this, though he also illustrates how previous reasonable proposals have failed, through lack of enforcement, lack of interest, or unexpected ways the system can adapt to achieve the same misdirection in new ways when some problems are fixed. Overall, Dr. Goldacre does such an effective job illustrating the problem, it is hard to come away with much hope that it can be fixed, despite his assertions that he believes it can. One of the weaknesses of the book is that his efforts at driving home the complexity and severity of the problem are often more successful and convincing than his proposals to solve it.
Bad Pharma also reviews the process by which drugs are developed and approved for human use. Since this is a lengthy and complex process, Dr. Goldacre necessarily presents a brief, simplified version with a frame intended to illustrate primarily the opportunities for trouble and erroneous or unreliable data. He does not, perhaps, offer an entirely fair view of the difficulties both inherent in the process for biological reasons and imposed by the commercial nature of the endeavor and the often irrational and inconsistent regulatory conditions under which it operates. There are many reasons why developing safe and effective medicines is tremendously challenging apart from the production of misleading and unreliable research data. However, the book was not intended as a “fair and balanced” review of the process but as an expose of its flaws and a call to action. In keeping with Dr. Goldacre’s own suggestion later in the book, it is appropriate to view this section, and the others, with this agenda in mind. The specific points he does make are, as best I can tell, accurate.
Dr. Goldacre also spends a fair bit of time discussing the gross inadequacies of the system for regulating the pharmaceutical industry. He presents something of a conflicted view, both making a compelling case for the failure of government regulation and yet repeatedly suggesting regulation as a solution to many of the problems in the industry. I appreciate this conflict as I tend to share it. I have written often about the regulation of medicine in general (e.g. 1,2,3,4) and why I feel it is frequently inadequate at protecting patients from even the most egregious pseudoscience and quackery (e.g. 5).
However, I also believe that consumer choice and market mechanisms cannot do the job, and that the solution to ineffective regulation is more and better regulation properly enforced. Again, I think Dr. Goldacre lays out the problem clearly and effectively, but it does somewhat undermine his own case when he suggests further regulation as an important part of dealing with the problems he identifies. It is important to recognize that the quality of medicine, and the usefulness of research evidence, was far worse in the era of “medical anarchism” before widespread regulation, and it continues to be far worse today in unregulated or under-regulated domains such as dietary supplements and herbal remedies than it is in the domain of pharmaceuticals. The legitimate failings Dr. Goldacre identifies are important and urgent problems, but not a reason to abandon ship or forget the value we get from the regulatory mechanisms we have.
As an epidemiologist-in-training, one of my favorite parts of the book were the sections dealing with ways in which clinical trials can be designed to generate misleading findings favorable to the sponsor. The whole point of clinical trials, of course, is to compensate for chance, bias, confounding, and sources of error that lead us so easily astray in our day-to-day evaluation of how the world works. The success of science has been due primarily to this ability to compensate for the collective and several weaknesses of individual scientists.
However, the system is not perfect, and there is good reason to believe residual bias and improper design and execution bedevil much medical research. This is part of what makes the processes of evidence-based medicine, which critically evaluate all the research and draw conclusions based on the entire body of evidence in a given area, so crucial.
Dr. Goldacre gives many specific examples of how trial results can be shaped to be misleading, from the selection of unrepresentative patient populations to inappropriate or shifting outcome measures, inappropriate controls, inadequate randomization and blinding, early termination, and dodgy statistics. All of these are flaws which can be easily identified if the full details of the process are publically available from start to finish, which is one of the most important recommendations made in the book for dealing with the problems identified.
Finally, Dr. Godlacre addresses the abuses and excesses of the pharmaceutical industry marketing machine, which exists entirely to increase the revenue generated by the industry’s products and yet which masquerades as a system for disseminating useful information to doctors, patients, and politicians. This subject has been covered extensively by others, so there is not much new in this review of the problem.
I also think this section contains some of the weakest parts of Dr. Goldacre’s argument. When he discusses potential conflict of interest, he acknowledges that it is difficult to argue that anyone doesn’t have one. Ideological or philosophical biases and personal relationships can shape our conclusions as much as financial incentives, so no one can claim to have no bias at all. That is, after all, why we need the methodology of science.
But there likely is a reasonable, if somewhat arbitrary point, at which one can distinguish between the ordinary bias that comes from having a point of view and recessive, inappropriate corporate influence on doctors and political decision makers. Dr. Goldacre’s primary suggestion for dealing with conflicts of interests is extreme transparency, suggesting, for example, that any and all affiliations with industry a doctor may have be disclosed in the waiting room of their office or on the Internet. Dr. Goldacre seems to feel that this disclosure will convey the impression to the public of possible conflict of interest and that subsequent public pressure will discourage inappropriate relationships between doctors and industry.
My concern with this line of reasoning is that it feeds all kinds of cognitive bad habits that can lead us to judge people’s claims inaccurately and inappropriately. If any association with industry is seen as suggestive of possible bias, and if almost everyone has some such association, however tenuous, than disclosing these alone will simply convey the impression that all doctors, and medicine as a whole, is merely a tool of the pharmaceutical industry and that recommendations made by our doctors can be dismissed on this basis alone, regardless of the evidence for or against them. I realize, of course, most people are unlikely to be this unreasonable, but unfortunately in the context of this blog I meet a fair number of people who are not and who are looking for any opportunity to play the “pharma shill” card to dismiss science-based medicine in favor of some alternative that is even less evidence-based and reliable.
I do support disclosure of commercial interests that might plausibly introduce bias into research or clinical practice, but I think it would be appropriate to make some reasonable distinctions about what kind of relationships matter or are clearly inappropriate and what kind are less concerning. And ultimately, we cannot eliminate individual bias, so our efforts are likely to be most productive if we focus on ensuring transparent and effective research methods to control for it.
In any case, there is no question that the shift in public political values and priorities, especially in the U.S., has led to a situation in which the lion’s share of medical research is supported by commercial interests rather than government. This has had the consequence of raising the threat financial bias poses to the integrity of research data. And the marketing mechanisms employed by the pharmaceutical industry are often a paragon of cynical, blindly self-interested manipulation of opinion not conducive to the best interests of patients.
Bad Pharma is a compelling and important exposé of the well-known but still unresolved problems with the influence of pharmaceutical companies on the generation and distribution of research evidence concerning drug therapies. It skillfully and convincingly identifies many specific problems. It also offers many reasonable and practical suggestions for improving the situation, though the strength of the solutions offered is noticeably weaker than the strength of the case made for the problems themselves, leaving one struggling with a bit of pessimism about the chances of reforming the system in significant and effective ways.
In addition to illustrating the problem and discussing solutions, Dr. Goldacre has partnered with the British Medical Journal, The Centre for Evidence-Based Medicine, Sense About Science, and the James Lind Initiative to form the All Trials Campaign. This effort is focused primarily on pressure the pharmaceutical industry, governments, medical professionals, researchers, and patient organizations to take steps to ensure all research data related to pharmaceuticals is freely, efficiently and widely available so that the evidence-based analysis necessary to guide medical practice has access to the evidence it needs to draw correct conclusions. I encourage everyone interested in improving the state of medical science and healthcare to support this initiative and to read Dr. Goldacre’s book.