Folk Epidemiology Part I: Harvard, Erin Brockovich, and Godzilla

Folk Epidemiology Part I: Harvard, Erin Brockovich, and Godzilla
by Craig Nelson

The “hedonism” component of Enlightened Hedonism is the easy part. Enjoying good food and drink, working up a good sweat, avoiding the obvious hazards like cigarettes…what’s not to like? It’s the “enlightened” element that can be a bit challenging. The dictionary defines enlightened as: 1. freed from ignorance and misinformation; 2. based on full comprehension of the problems involved.

This is often not easy, particularly when it comes to our health. To be freed of misinformation and in full comprehension of the problem to be solved sounds like a lot of work. And it also may lead us in directions we’d rather not go. We all carry around with us cherished beliefs about our health based on misinformation and incomplete comprehension of the problem. And we really don’t want to abandon these beliefs. For some, this week’s post may be challenging in this regard.

In October 2005, the Harvard School of Public Health awarded its highest honor, the Julius Richmond award, to Erin Brockovich. The award was given for her efforts “on behalf of all of us, and especially the residents of Hinkley, California, whose health was adversely affected by a toxic substance dumped by a utility company.” You remember the story of Erin Brockovich (portrayed in the movie by Julia Roberts): against all odds, a single mother (who happens to be beautiful) takes on the evil corporation that has polluted the ground water and thereby caused untold health damage to the residents of the community.

The Harvard/Erin Brockovich story was reported in the media two different ways. The first was as a straightforward public interest story: The still lovely Ms. Brockovich receives more well-deserved accolades for her heroic actions. The second way the story was reported was as a scandal: Has Harvard lost its mind? The Erin Brockovich myth has been long since debunked. There never was and never will be any deleterious health effects from whatever violations the utility in question may have committed. Not one pimple, not one case of heartburn, and not one hang nail has ever been shown to have resulted from the incident in question. As much as it’s possible to prove a negative, it has been proven that the whole incident was an extreme example of junk science.

The latter interpretation is the correct one. The lawsuit, the movie and the subsequent fame of Ms. Brockovich notwithstanding, all scientific investigations into the episode have come to the same conclusion: There have been no measurable adverse health effects as a result of the toxic dumping. Why would Harvard do this? Surely, the scientists and administration at the School of Public Health knew the truth of this matter. The cynical (but realistic) answer is that the award ceremony is a gala event that affords a great opportunity for fundraising. The glitz and glamour of the Brockovich/Hollywood connection was probably too great an opportunity to pass up. Maybe Julia Roberts herself might show up!

I suppose there’s really no reason to be shocked by this—we don’t really think that institutions like Harvard are above such nonsense, do we? But it does call to mind the troubling phenomenon of what has become known as “folk epidemiology.” Folk epidemiology is the process whereby an alert citizen notices an unusual occurrence of diseases—a disease cluster—and associates that cluster to some environmental contaminant. Then, abetted by the popular press (which doesn’t know any better and likes the story line), by Hollywood (which also likes the David vs. Goliath aspect of these stories), and even by credulous scientists (who are either simply stupid or have an agenda of their own to advance), this correlation becomes enshrined in the public’s consciousness and even ratified by the courts.

There are many recent examples of folk epidemiology in action. Another Hollywood epic “A Civil Action,” starring John Travolta, documented the sins of the W.R. Grace & Company in polluting Woburn, Massachusetts and causing a series of plagues among the townsfolk. But just as with the Erin Brockovich story, it has the unfortunate attribute of not being true. No disease. No epidemics. No plagues. Nothing.

There is the immunization/autism connection—perhaps the most exhaustively studied of folk epidemiological events. No connection. There’s the silicone breast implant/connective tissue disease hypothesis. Well, more than a hypothesis—hundreds of millions of dollars have changed hands and a company (Dow Corning) has gone bankrupt based on this connection. Alas, it doesn’t really exist.

The scare over the chemical Alar cost the apple producers of Washington State tens of millions of dollars. The evidence upon which this scare was based was the flimsiest junk science and there has subsequently been no additional evidence of harm to humans.

For years, women have feared a Long Island breast cancer “epidemic” caused by some unknown environmental contaminant. A specially commissioned study was undertaken to examine that phenomenon with this conclusion:

“In summary, the LIBCSP [Long Island Breast Cancer Study Project] studies have not identified any environmental factors that could be responsible for the elevated incidence of breast cancer on Long Island or the other locations. There were a few suggestions of associations between certain exposures and increased risk of breast cancer, but these observations would need to be confirmed in other population studies. No association was found between exposure to organochlorine compounds and increased risk of breast cancer. The compounds examined included the pesticides DDT (and its metabolite DDE), dieldrin, and chlordane, and polychlorinated biphenyls (PCBs).”

There are countless other less well-known examples of folk epidemiology. In the last few decades, there have literally been hundreds (thousands?) of cases of disease clusters attributed to some environmental cause. Folk epidemiology is predicated on two false premises:

1. Disease clusters don’t just happen. If one observes within a small geographic area the occurrence of an unusual number of cases of condition X (or even the occurrence of a single X, a single Y, and a single Z—each of which is rare in its own right), then there must be an underlying cause uniting these disease events. Such an anomalous series of events can’t just happen.

2. Human health is exquisitely sensitive to minute quantities of chemical, radiological other types of contamination. Indeed, most of the ill health that we observe in industrialized countries (cancer, rheumatoid and other autoimmune diseases, metabolic disorders, degenerative diseases) can be attributed to some kind of environmental pollutant or perhaps some food additive. It is commonly observed that we are in the midst of a “cancer epidemic” that can be traced to the detritus of modern life.

Folk music? Yes. Folk epidemiology? No

The only problem with folk epidemiology is that it is almost never right. There are almost no examples in a non-industrial setting of such disease-cluster/environmental-contaminant relationships that have been shown to be causal. (A lone exception is lead-based paints, which have caused disease in residential settings. There are also many examples of disease-causing contaminants in an industrial setting where the concentrations of the contaminants are many orders of magnitude higher.) And it is not for a lack of investigation. Federal, state, and local departments of health have been chasing down these connections for several decades to no avail. This includes the most notorious of all these events: Love Canal.

Yes, Love Canal. Please consider the following:

• Were large amounts of toxic waste illegally dumped in Love Canal in the 1930s and ’40s? Yes.
• Was a residential neighborhood including a grade school built on this site? Yes.
• Were traces of the original contaminants found in the water and soil in this area? Yes.
• Were some of these contaminants found in the bloodstream and tissues of some of the residents? Yes.
• Were adverse health events found to result from all this?

Er . . . well . . . um . . . you see . . . um . . . no. Decades of research have failed to identify any significant health problems that have resulted from the Love Canal contamination. Epidemiologists have studied cancer rates, birth defects, chromosomal abnormalities, premature births, and low birth weight and have found no meaningful harm that has resulted. Just one of the many scientific studies of the Love Canal episode that found no evidence of harm concluded:

“Rates of liver cancer, lymphoma, and leukemia, which were selected for special attention, were not consistently elevated. Among the other cancers studied, a higher rate was noted only for respiratory cancer, but it was not consistent across age groups and appeared to be related to a high rate for the entire city of Niagara Falls. There was no evidence that the lung cancer rate was associated with the toxic wastes buried at the dump site.”

Note: At this point, I must take a step back to make the following disclaimer: None of this narrative suggests in any possible way that the laws and regulations enacted to ban the dumping of raw sewage into our rivers or to prohibit the dumping of industrial wastes into our air and water are in any way unnecessary or imprudent. It does not suggest that enforcement of these laws and regulations should be anything but rigorous. Human beings and other living creatures are better off in countless ways because we no longer engage in those practices. The “good old days” when companies could indiscriminately use our air and waterways as dumping grounds were not the good old days. They were the bad old days, and we don’t want to return to them.

So how does folk epidemiology continue to thrive in the face of such contrary evidence? We can identify four factors that drive the phenomenon:

1. Scientific Illiteracy. The central scientific fallacy of folk epidemiology is the belief that disease-clustering is likely to mean something. Our intuition tells us that if events (diseases) occur at random (that is, they are not connected causally), then we should see such event evenly distributed geographically, temporally, and in every other way. Thus, when we see uneven distribution, we are seeing a non-random event. An uneven distribution would be something like, say, two children in a neighborhood with leukemia; or several people in the same neighborhood with different rare diseases, e.g., lupus and ALS.

But that is a fallacy. Random events are not uniformly distributed. They tend to be very uneven, patchy, and irregular in occurrence. An obvious example of this is a random coin flip. Coin flips don’t produce a H-T-H-T-H-T-H-T-H-T pattern. They produce a H-H-T-H-T-T-T-H-T-T-H-H-H-H-H-T-H-H-H pattern, or something like that. Indeed, not only are random disease clusters possible, they’re inevitable. If you consider the tens of thousands of possible disease occurrences that might happen in your neighborhood, it is inevitable that some weird-looking clusters will occur. If you were able to perform a health census in, say, a ten-block radius around your own home, I guarantee that you would see some odd-looking disease clusters.

Regrettably, real epidemiologists are not always as critical toward folk epidemiology as they could be. There is abundant funding for the sort of studies that produce these dubious findings, and, as noted, they make very good news stories when published. We have all seen these scientists interviewed by the popular media and solemnly intoning something like this: While this study does not absolutely establish a causal link between X and Y, we think it’s important to raise public’s awareness about this potentially dangerous connection. Utter nonsense.

2. Populism. The storyline of folk epidemiology is nearly always the same. The victims are common citizens. The villains are the utility company, the chemical company, the pharmaceutical company, the oil company. In these match-ups, we always know right from wrong just by looking at the players. These stories are so good they just have to be true. A populist storyline will always have a sort of automatic credibility; anyway, it’s just a good story.

3. Money. Obviously there’s a buck to be made here. Both the supposed victims and their attorneys have a lot to gain financially if they can convince a jury (or threaten to convince a jury) that injury has resulted from some contamination. As a management consultant might say, the use of folk epidemiology in the courts represents a viable business model. Billions of dollars have changed hands on the basis of folk epidemiologic claims.

4. Culture. How did our culture—which once used radiographic fluoroscopes in shoe stores to look at our foot bones and was in love with a company whose slogan was “Better living through chemistry”—become so collectively phobic about toxins and contaminants? It all started with Godzilla.

Next in Part II: How Godzilla, Gamera and Rodan set our public health agenda.