Research continues into the question of whether or not cell phone use is associated with an increased risk of brain cancer. The latest epidemiological study, a Danish study of 358 thousand cell phone subscribers, reassuringly shows no correlation.

Researchers compared subscribers to non-subscribers for the incidence of central nervous system (CNS) tumors, such as glioma or meningioma. They found no difference in the incidence of any such tumor relating to use of cell phones. There was also no correlation to duration of use – i.e. no dose-response. And there was no correlation with location of tumor – users were no more likely to have a tumor in the part of the brain next to where a cell phone is typically held.

The study, in other words, was completely negative. This study was an extension of previous studies using the same databases, but extending the time of follow up. Of course, we can only study a duration of exposure that is less than the time cell phones have been in widespread use. Researchers will therefore have to continue to monitor this data to see if long term exposure is beginning to increase the risk of CNS tumors.

So far health organizations around the world have taken a cautious approach to this question. On the whole they have concluded that the current evidence neither demonstrates or rules out a connection between cell phone use and cancer. They therefore recommend avoiding excessive use of cell phones, holding the phone away from the head when possible, and that children under age 16 either do not use a cell phone or only very briefly. These are reasonable recommendations given the uncertainty.

There always will be a certain amount of uncertainty, because we can only use observational data to answer this question. We cannot do a controlled experiment – randomize people to either use or not use a cell phone, or have a control group that is blinded to using a sham cell phone. So we have to make do with observational studies, like the one described here. This always allows for the interference of confounding factors – factors that are not controlled for or measured that affect the outcome in some way.

Given that cell phone use is so common, and increasingly so, there are a few facts that are reassuring. The first is that epidemiological studies have not found any clear correlation, and this study adds greatly to that conclusion. Some studies have shown a correlation, and overall the results have been mixed. But there has been no clear signal in the data of a correlation, so either there isn’t one, or if there is, it must be very small and inconsistent.

Further,while there has been an increase in the overall incidence of CNS tumors over the last 20 years, the increase is small and similar to other cancers. This increase is likely explained by increased use of MRI scans and other technologies that are catching benign tumors, like meningiomas, in the population. If cell phone use caused brain cancer, we would expect a more significant increase, greater than can be explained by improved diagnosis.

Finally, the biological plausibility of a causal relationship between cell phones and cancer is small (if non-zero). Cell phone use non-ionizing radiation, which is therefore not energetic enough to break chemical bonds and cause mutations in DNA or damage to proteins. There can still be a biological mechanism, but the likelihood is low.

This latest study is reassuring. It is a large data set followed for a long time and is completely negative. This kind of data cannot completely rule out a causal connection between cell phone use and cancer, but it does set limits on how strong the connection can be. At this point we can say that for cell phone use up to about 16 years there is either a very small or no causal connection to CNS tumors.