This is a cross-post from Ezra Klein’s blog, where Austin and I are guest-blogging this week.
I can’t speak for you, but all of my friends seem to be in a tizzy about the new WHO classification of cellphones as “possibly carcinogenic”:
An international panel of experts has weighed in on the controversy about cellphone safety — and come to a conclusion that falls far short of recommending that consumers put down the devices, although it may make them more anxious.
Cellphones are “possibly carcinogenic” to humans, according to the panel organized by the World Health Organization … [this] classification marks a departure for the WHO, which previously said there were no risks from exposure to radio frequency electromagnetic fields emitted by the devices.
“The conclusion means that there could be some risk, and, therefore, we need to keep a close watch for a link between cellphones and cancer risk,” said Jonathan M. Samet of the University of Southern California, who chaired the WHO’s International Agency for Research on Cancer panel.
I should start by making sure everyone understands that this step doesn’t mean that there’s a clear link. This is more akin to saying that there could be some risk, and that more research is needed. Putting cellphones in this category means that they now join picked vegetables and coffee as “possibly carcinogenic.”
So I wouldn’t panic yet. But what should we make of all this?
Let’s start with some facts. Cellphones are hugely popular. Worldwide estimates put cellphone usage at more than 3 billion users. In the United States, more than 270 million people use cellphones, including about half of children ages 8-12. And yes, cellphones do emit radiation, specifically in the 800 to 2000 mHz range. So it shouldn’t come as a surprise that there is an ongoing debate as to whether the widespread use of a device that emits radiation causes cancer.
Many, many studies have been published in this area. The vast majority of them are what we call case-control studies. For a case-control study in this domain, you gather a group of people with brain tumors. Then, you gather a group of people without brain tumors. Then, you ask them various questions (like “do you use a cellphone?”) to see whether there are differences between the two groups.
Although case-control studies can be very important as a study design, it’s essential to recognize that they are among the weaker types of studies you can do. Unlike randomized controlled trials, you can’t prove causation with this type of study. And, unlike in prospective cohort studies, there is a real problem with recall bias. This is a type of bias that can alter your results because people with an issue (like brain tumors) are more likely to recall things that might have caused their brain tumors than those who don’t have that same problem. So the danger with case-control studies is that people with brain tumors might have heard about the theory that cellphones cause brain tumors and, therefore, might remember cellphone use more than people without brain tumors.
In 2008, a scientist summarized data from 33 studies in the peer-reviewed literature that look at cellphones and brain tumors. A surprising number come from one group in Sweden. This summary found major flaws in the research that would make it difficult to draw a convincing conclusion about cellphones and brain tumors.
Another group attempted a meta-analysis of the literature. They felt that 23 articles were good enough to be included in the review. All of them were case-control studies. They found overall that, compared with rarely or never using a cellphone, regularly using a cellphone was not significantly related to the development of malignant and benign brain tumors.
Even the most recent large study (Interphone) found that “cell phone users have no increased risk for two of the most common types of brain tumor — glioma and meningioma.”
As human beings, we make decisions every day that trade off benefits with harms. For instance, the No. 1 killer of children in the U.S. is car accidents. Yet no one thinks we should ban cars in order to reduce that to zero. We, as a society, accept that the quality of life derived from driving outweighs the increase in deaths of children from car accidents. If you disagree, don’t drive; almost no one makes that choice. Similarly, cellphones provide a lot of happiness to many people. Given the controversy, even if there was an association between cellphones and brain tumors, it would have to be a small one. If such a small association exists, few would likely forgo their cellphone to eliminate it completely.
Moreover, we know that at a population level, there has been no explosion of brain tumors recently. A study published in the Journal of the National Cancer Institute in 2009 examined data for cancers in four countries with registries from 1974 to 2003. Over that time, almost 60,000 people in a population of 16 million adults between 10 and 79 years were diagnosed with brain tumors. While a slow increase in rates of gliomas and meningiomas were seen throughout the time period, no changes in incidence trends were seen from 1998 to 2003, when you would expect to see an increase because of cellphone use after five to 10 years.
Case control studies are necessary for rare diseases. But cellphones are so common that if they were dangerous right now, we’d be able to see the effect easily. That’s not happening. Future work will be needed to see whether they cause long-term harm, but as for now, there’s very little evidence that they do.
Parts of this post have been adapted from my upcoming book, “Don’t Cross Your Eyes … They’ll Get Stuck That Way!,” which will be published by St. Martin’s Press on July 5.