Geoffrey Kabat is a cancer epidemiologist at the Albert Einstein College of Medicine and the author of Hyping Health Risks: Environmental Hazards in Daily Life and the Science of Epidemiology (Columbia University Press, 2008).
Over the past thirty years we have been bombarded with a steady succession of putative health risks, including electromagnetic fields (EMF) from power lines and electric appliances, radon seeping into homes, and pesticides in our food and water, to name just a few. Today, radiation from cell phones and exposure to minute amounts of BPA (a chemical used in some plastics) are major concerns that are grabbing headlines.
When one looks back at many of the alleged health hazards that have received enormous attention (from scientists, regulators, the media, and the public), a striking pattern is discernible. Early studies seeming to indicate the existence of a novel threat received enormous publicity. But these studies tended to have small sample sizes and crude methodology. In some cases, their weaknesses were pointed out by commentators at the time, but, nevertheless, the results were given more credibility than they should have been. As larger and more methodologically rigorous studies were carried out, the elevated “relative risks” either diminished considerably, or evaporated completely. Interestingly, in a recent article in the New Yorker, Jonah Lehrer has pointed out a similar phenomenon in a number of areas of science, including psychology and clinical medicine (“The Truth Wears Off,” December 13, 2010).
How are we to explain this phenomenon? Undoubtedly, a variety of factors contribute to the manufacture of a hazard. First, as we have come to recognize (most famously in the case of hormone therapy), the findings from observational studies can be misleading, and we need to be extremely cautious in attributing causality to a correlation between an exposure, often measured at one point in time, and the development of disease many years later. Second, scientists are motivated to discover new causes of diseases and, understandably, want their results to be meaningful. It is a basic, if rarely acknowledged, fact of life that interesting findings are crucial currency in getting scientific papers published, obtaining funding, and advancing professionally. Third, the media, regulatory agencies, and the public are all hungry for novel information that might explain why people develop terrible diseases. Finally, positive results, even when they come from a study with many flaws and limitations, tend to get more attention than results showing no effect.
To take just one example, in 1979 researchers in Colorado carried out a study which appeared to show that children who died of cancer in childhood were 2-3 times more likely to live in close proximity to electric power lines compared to healthy children. In spite of its many acknowledged limitations – the researchers did not measure actual fields but rather made assumptions about exposure based in the thickness of the power lines — the provocative result captured the imagination of scientists and regulators, and stimulated epidemiologic and experimental research resulting in hundreds of publications over the ensuing thirty years. This program of research fed on weak and inconsistent results that, at the time, were interpreted as evidence of a possible hazard, and aroused fear in the public. Because exposure to electromagnetic fields (EMF) is virtually ubiquitous, the argument was made that even a slight increase in risk of disease due to exposure could have substantial public health implications.
At the same time, the arguments of physicists and biophysicists that the fields to which people are exposed in the course of daily life are 10,000 times lower than the normal thermal excitation of the molecules in our bodies and, therefore, were highly unlikely to have any effect on biology whatsoever, were widely ignored. After 30 years of intensive study, it is now generally accepted that ambient EMF are not a hazard.
What can we learn from our experience with EMF and other similar hazards? As consumers of news pertaining to such threats, we should be aware that science in this area takes place in a politically and professionally charged landscape and that we need to be more skeptical. Just because something is being studied does not mean that it will turn out to be important. Many things get studied, but there are many false paths, and few things turn out to be of earthshaking importance.
As a society, we need to take potential hazards seriously but, at the same time, but we have a right to expect scientists to be brutally critical in evaluating the evidence and putting findings in perspective, taking all of the relevant evidence into account. Where preliminary findings point to a potential hazard, it is important to convey to the public that, while a question may merit study, this does not mean that there is reason to be concerned or to take any action. We also need to remember that it takes time and painstaking work to discover new and important relationships, and that this work does not lend itself to catchy headlines.
We have been conditioned by forty years of research to be highly sensitive to potential threats to our health and to the environment. But we need to keep in mind that longevity and overall health have increased steadily and dramatically over the past hundred years, and that the burden of chronic diseases is largely a function of our living to a much older age.
There are real problems that need attention and resources, and we have to try to learn from past distractions and recognize that the problems that receive the most attention are often not the ones that turn out to be important.