WARNING: This Blog Post is a Potential CarcinogenPosted: June 30, 2011
About a month ago, I read what is probably the best article of 2011, “Do Cellphones Cause Brain Cancer?” (link). The author, Siddhartha Mukherjee, does an incredibly thorough job detailing exactly why assessing cancer risk is so fraught with unknowns. A sample:
There are certainly methods in epidemiology to counteract the biases created by selective memory: Interphone researchers could have initially identified a cohort of high-volume cellphone users and of nonusers, and followed them over time to determine who developed or did not develop cancer. Such a study — called a “prospective trial” — would certainly erase the biases of memory. But it would be logistically impossible to perform. Since the rates of brain cancer are small, about 6.5 cases per 100,000 persons, a trial of this design would need to follow an enormous cohort of cellphone users — hundreds of thousands of participants — to record even a few cancers. And where on earth would you find the nonusers for the study? In most nations, cellphone usage is so common that finding 500,000 people who will not use phones for a decade is hard to imagine.
A large portion of research on the cancer dangers of using cell phones is to utilize survey data that asks responders to remember and note their past usage. As with dietary surveys, it isn’t difficult to imagine how many problems this approach faces as responders either have hazy memory or their responses are biased in a particular direction. Prompting a particularly heavy cell phone user who also happens to be embarrassed by their habit will typically result in a downplay of the actual usage (i.e. “How often do you use your cell phone?” “Oh…not that much”). Thus, the only truly reliable trial, as in medicine and elsewhere, is to utilize a randomized controlled trial (RCT) as elaborated upon above. If survey data is unreliable, and RCTs unrealistic, can animal studies tell us anything?
Nonetheless, biologists have exposed mice and rats to chronic nonionizing radiation (comparable to that emitted by phones) to determine whether it causes cancer. In rats prone to developing breast cancer, there was no acceleration of breast cancer. In another experiment, rats were treated with a chemical carcinogen in utero (to “prime” them to develop brain tumors) and then exposed to radiant energy comparable to cellphone radiation for two hours per day, four days a week, for 22 months. The experiment revealed no increased incidence of brain tumors in rats. Nor was there any accelerated growth in previously established brain tumors. From 1997 to 2004, six independent experiments on mice and rats studied the effects of chronic radiation on brain cancer. No experiment revealed an increased risk of brain cancer.
Seriously go read the whole thing, as it goes into much greater (and fascinating) detail about how creative researchers have to get to tease out such minute and rare effects.
I was very heartened to see such excellent commentary on the overall perils of epidemiology on such a mainstream venue. That feeling did not last very long however, because soon afterward, the World Health Organization changed their categorization of cell phone use as a “potential carcinogen” and the deluge of hyperbolic headlines commenced. Let’s take one example from the corresponding CNN article:
WHO: Cell phone use can increase possible cancer risk
(CNN) — Radiation from cell phones can possibly cause cancer, according to the World Health Organization. The agency now lists mobile phone use in the same “carcinogenic hazard” category as lead, engine exhaust and chloroform.Before its announcement Tuesday, WHO had assured consumers that no adverse health effects had been established.
A team of 31 scientists from 14 countries, including the United States, made the decision after reviewing peer-reviewed studies on cell phone safety. The team found enough evidence to categorize personal exposure as “possibly carcinogenic to humans.”
I can’t fault the WHO for the media frenzy, as they simply stated the truth: Cell use is possibly a carcinogen. The real problem is the misinterpretation of the categorization because most people do not realize that anything, literally anything, (including this blog post you are reading, ha!) is a possible carcinogen. Given that it is impossible to conclusively prove a negative, the scientific onus compels us to maintain whatever avenue of possibility we happened to be on. Hence, since it is impossible to prove conclusively that cell phone use will not give you cancer, the only way to remain good scientists is maintain that cell phone use can still potentially give you cancer. Props to Geoffrey Kabat for bringing more sanity to this discussion: (link)
We are faced with a paradox in our increasingly health-conscious society. It is simply a fact of life that research is going to be done on topics like cellphones. But we can never prove a negative or exclude the possibility of a miniscule risk, no matter how large the study. So even when expert bodies concede that there is no convincing evidence of a threat, we get impossibly vague advisories like the current one warning us of “possible carcinogenicity.”
In an echo of the Harvard incident, Donald Berry, a professor of biostatistics at M.D. Anderson Cancer Center at the University of Texas, said “anything is a possible carcinogen.” Speaking from his cellphone, he added, “This is not something I worry about and it will not in any way change how I use my cellphone.”
The conclusion? Cell phones are unlikely (but still potentially!) to be carcinogens. Even if they indeed were, how would you respond to the worst case scenario? Would you be willing to give up your cell phone for good in order to gain an additional one or two years of additional life later on? I know I wouldn’t, so why should I worry about an even much smaller risk?