A few days ago, I came across an article on Engadget that mentioned almost in passing some studies that seemed to indicate health problems or no health problems, depending on the specific study, due to the ubiquitous and maligned cellular telephone. Not having dealt with this issue much on my blog, I decided to take a look, mostly out of curiosity. The claims that cell phones somehow cause cancer have been circulating for many years, and the studies marshaled to show such a link have in general been not that impressive. However, even though radiofrequency radiation of the sort used in cell phones does not break chemical bonds or induce mutations, it’s not entirely implausible that prolonged exposure to cell phone radiation over a long period of time, although it does, on the surface, appear fairly unlikely using a prior probability argument.
It turned out that fellow ScienceBlogger Revere took on what has to be the weakest of the bunch, a unimpressive study1 in which normal volunteers were exposed to radiofrequency modulated electromagnetic fields similar to that of cell phones, after which punch biopsies from exposed areas of skin and from unexposed areas were taken. The researchers then examined the levels of 579 different proteins by a test known as 2D gelelectrophoresis to determine whether the levels of any of them rose or fell. They found that there were eight proteins whose levels varied by a statistically significant amount. Unfortunately, by random chance alone, we’d expect to find around 5% of the proteins (or 29) appearing to vary by a statistically “significant” amount. More importantly, for nearly all the “positives,” the variability of the ratios of EMF-exposed/control was so large as to make any biologically meaningful conclusion virtually impossible. Even worse, there was no verification of whether messenger RNA rose or fell synchronously or other confirmation of the result.
Examining the totality of the study, including design, statistics, and protein gels, I was not at all impressed with it as evidence for a specific biological effect due to cell phone radiation. That’s even cutting it some slack as a pilot study. Indeed, I’m a little surprised Revere even brought this study up as evidence for the possibility of a specific effect, given that, both statistically and from a protein expression point of view, it appears actually to be at least as consistent with no significant biological effect due to cell phone radiation on human skin as with a biological effect.
That’s why I spend so little time on it and instead want to move on to a different study, which purports to show that cell phone radiation contributes to male infertility, a concern for all men, to put it mildly.
The article2, published by a group from the Cleveland Clinic and an Indian group and entitled Effect of cell phone usage on semen analysis in men attending infertility clinic: an observational study concludes that cell phone use negatively affects a variety of parameters indicative of sperm health. But is the implied conclusion, namely that cell phone radiation harms male fertility, justified?
Probably not from this study.
Briefly, the study examined 361 men attending an infertility clinic from September 2004 to October 2005. The mean age of the study population was 31.81Â±6.12 years. The men were divided into four groups based on their self-reported cell phone usage: none; less than 2 hours a day; 2-4 hours a day; and greater than 4 hours a day. Semen samples were collected, and several standard parameters measured. It was found that there was no correlation between cell phone use and semen volume, liquefaction time, pH, and viscosity. However, a negative correlation adjusted for age was observed between increasing cell phone use and sperm count, motility, sperm viability, and percent normal morphology. The authors concluded that cell phone usage was associated with sperm and semen parameters associated with decreased male fertility.
This study has a number of shortcomings that make me take its conclusion with a huge grain of salt. Right off the bat, I note that it’s an observational study, and such studies are fraught with problems. Another huge problem with this study is that the men being studied are not a representative group by definition. After all, they are coming to an infertility clinic. As the article itself points out, infertility affects approximately 15% of reproductive age couples, with nearly half of these cases being due to male factor infertility. That means, right off the bat, without knowing anything more about the study we can predict that close to 50% of the men being examined will have problems with sperm count, motility, or other parameters indicative of low fertility. That’s much higher than the 7% or so that would be expected among the general population of men of reproductive age at large.
A second problem is that the four parameters that showed a decrease are not independent. They are related, and in this study showed significant positive correlation with each other. If one goes up, all tend to go up, and if one goes down, all tend to go down. However, together, they do correlate with the health (or lack thereof) of spermatozoa.
However, these problems pale in significance with the single biggest problem of the study: the utter failure even to try to control for confounding factors other than age. Not only did the authors not validate cell phone exposure independently, but they did not examine the occupational history of the subjects or other potential exposures of radiofrequency radiation. The reason this is significant is that there could be a factor common to these men with decreased fertility that is the real cause of their problems that is also correlated with cell phone use. Examples that come to mind might include number of airplane flights per year, given that airline travel exposes travelers to natural background radiation. I’m not saying that this is the case; I’m merely saying that there’s a strong possibility that there could be a confounding factor like that that only makes it appear that increasing cell phone use is correlated with decreased fertility and that it is a serious shortcoming of the study that such a factor was not more actively sought. Had I been a peer reviewer for this paper, I would have come down hard on the authors for this. Indeed, a similar critique was leveled at the same investigators for the same study when they presented it at the American Society for Reproductive Medicine meeting a year and a half ago:
However, Dr Allan Pacey, senior lecturer in andrology at the University of Sheffield, said “This is a good quality study but I don’t think it tackles the issue.
“If you’re using your phone for four hours a day, presumably it is out of your pocket for longer. That raises a big question: how is it that testicular damage is supposed to occur?”
He said mobile phone use may be a marker for other lifestyle factors known to affect sperm quality.
“Maybe people who use a phone for four hours a day spend more time sitting in cars, which could mean there’s a heat issue. It could be they are more stressed, or more sedentary and sit about eating junk food getting fat. Those seem to be better explanations than a phone causing the damage at such a great distance” he added.
The comment about “great distance” brings up an aspect of this study that bothers me as well: Most people do not use their cell phones close to their genitals. True, with the increasing use of Bluetooth headsets this may be changing, as men leave the phone in its belt holster and use those funny-looking headsets with the blue blinking lights, but for the most part people still hold the headset to their ear when using a cell phone. Radiation intensity drops off rapidly with distance, and there’s a huge difference between holding the phone to one’s ear and and using it in the holster.
The bottom line is that, although there is some evidence in mouse and rat studies suggesting that cell phone radiation might induce sperm abnormalities, the evidence that cell phone use is associated with decreased fertility in human men is far from convincing thus far. The animal studies render the possibility of a negative effect on fertility somewhat plausible from a biological standpoint, but most human studies, this one included, that I’ve seen concluding such a problem exists tend to be plagued by methodological difficulties and inadequate controls for confounding factors. Still, further research into this question is warranted. In any case, if, as Steve Novella suggests in his cautious examination of whether cell phone usage is a risk factor for brain or head and neck cancers, one way to mitigate any risk that might exist for such tumors is to use a headset in order to keep the cell phone antenna away from the head, the consequence might be more male infertility. Given that possibility, I’d guess that most men would tend to opt for an uncertain possibility of perhaps a slightly increased risk of brain cancer over a similar uncertain possibility of a slightly increased risk of infertility.
That’s assuming, of course, that there even is a problem. In the meantime, I’m not planning on getting rid of my cell phone.
1. Karinen, A., Heinavaara, S., Nylund, R., Leszczynski, D. (2008). Mobile phone radiation might alter protein expression in human skin. BMC Genomics, 9(1), 77. DOI: 10.1186/1471-2164-9-77
2. AGARWAL, A. (2008). Effect of cell phone usage on semen analysis in men attending infertility clinic: an observational study. Fertility and Sterility, 89(1), 124-128. DOI: 10.1016/j.fertnstert.2007.01.166