Cell Phone Study Was Flawed, Say Some Experts

Roxanne Nelson

August 30, 2011

August 30, 2011 — The use of cell phones, and their possible detrimental effect on human health, is an issue that remains unresolved. A recent study, published in the Journal of the National Cancer Institute (2011;103:1264-1276), concluded that children and adolescents who use cell phones are not at an increased risk for brain tumors, but several experts are disputing these findings.

A report issued by L. Lloyd Morgan, BSc, senior research fellow at the Environmental Health Trust, and colleagues found that rather than showing no risk for brain tumors, the study's results indicate that an increased risk for brain cancer is a "major concern."

Mr. Morgan and coauthors Ronald Herberman, MD, director of the University of Pittsburgh Cancer Institute and the UPMC Cancer Center, in Pennsylvania, and Devra Davis, PhD, MPH, president of Environmental Health Trust, note that the study's results are flawed and mislead the public. They explain that these errors should have been picked up during the peer-review process and by the journal, because the results and conclusion sections of the paper contradict the actual reported results.

Several cell phone companies provided funding for this study, and some of the study's authors are known to be linked to industry and to other research that supports the interests of the industry, the report points out.

"The accompanying editorial was also written by 2 people who have worked for the cell phone industry," Dr. Davis told Medscape Medical News, "but there were no disclosures about that in their paper" (J Natl Cancer Inst. 2011;103:1211-1213).

"This study is not up to the usual quality seen in the Journal of the National Cancer Institute, and it seems that it was rushed into print," she said.

Not Likely in Young Children

The study in question, known as CEFALO, was an international case–control study that examined the association between cell phone use and the risk of developing brain tumors in children and adolescents. The participants were children 7 to 19 years of age in Denmark, Sweden, Norway, and Switzerland who were diagnosed with a brain tumor from 2004 to 2008.

The authors conducted face-to-face interviews with 352 case patients, 646 control subjects, and their parents. Cell phone use was determined from interviews and, when available, from operator records.

In the paper, the authors conclude that their primary analysis "does not point to a statistically significantly increased risk for brain tumors in children that is associated with the use of [cell] phones." They add that there was also no consistent exposure–response relation, either in terms of the amount of cell phone use or by the location of the tumor.

It is ridiculous to look at cell phone use in a 7 year old.

However, Dr. Davis pointed out that they weren't likely to find an association between cell phone use and brain cancer in young children because brain tumors can take 10 or more years to form. "Young children have not been heavy users of cell phones for that long," she said. "It is ridiculous to look at cell phone use in a 7 year old. What were they, 2 years old when they started using their cell phones?"

The use of cell phones has also quadrupled in the past few years, and this study could not possibly capture that, Dr. Davis explained.

"The authors were careful to note limitations," she added, but it was the invited editorial that stated that children face no risks from cell phones. "Public health is not well served by this."

Dr. Davis also emphasized that aside from brain tumors, other studies have linked cell phones to serious health risks in children, including learning problems, autism, behavioral impacts, insomnia, attention disorders, and a broad array of disturbances to the developing nervous system.

Problems in the Report

Another expert, Joel Moskowitz, PhD, director of the Center for Family and Community Health at the University of California, Berkeley, agrees that the findings of the study have been downplayed. "This report and the editorial are another example of biased reporting," he said in an interview. "The results actually verify higher tumor risks for children but the findings are downplayed. They dismiss any evidence or prior evidence of increased risk and harm, and then the media plays it out as either being not conclusive evidence or no evidence."

Dr. Moskowitz noted several glaring problems with the study, which are in line with the findings of Mr. Morgan and colleagues.

"In a subset of patients who had cell phone records available, there was more than a doubling of risk," he said.

That finding was also noted by the study authors, who reported that they found a "statistically significantly increased risk among users with the longest period since first subscription (odds ratio [OR], 2.15) among 24 case patients and 25 control subjects who had subscriptions for more than 2.8 years."

The authors dismissed this finding because it was based on a subset of the data, explained Dr. Moskowitz, and they had other data that were not statistically significant so they chalked this up to a chance finding.

The follow up to the study was also too short to expect much of an increase in brain tumor risk, he continued. "Only 13% of the sample had used cell phones for more than 5 years, yet brain tumors generally take decades to develop," Dr. Moskowitz said. "The study should have probably just focused on teens and not included young children."

Another problem was the size of the study. The sample size was too small, limiting the study's ability to detect an association between cell phone use and brain tumors, he pointed out. The study was originally designed to have 550 cases, but ultimately only 356 cases were included. As designed, the study would have been able to detect a 35% increased brain tumor risk; as implemented, it could only detect a 45% increased risk as statistically significant.

"This means if the investigators had 550 cases in the study and the overall result was similar, then the 36% increase in brain tumors reported for regular cell phone users would most likely have been statistically significant," said Dr. Moskowitz.

Also, the definition of "regular user" was problematic. The study authors defined "regular" cell phone use as using the device once a week for 6 months. "That's a very low threshold of use," he said. "These kids are not regular cell phone users. Most of the sample used less than 100 hours in their lifetimes, but there was still a high risk."

Only 13% of the cohort had used cell phones more than 144 hours in their lifetime, Dr. Moskowitz added. "The typical child in the United States reaches that amount in less than a year, so this study really isn't applicable to children and teens in the United States."

Text and Tables Not Matching

The report by Mr. Morgan and colleagues points out that the abstract and conclusion section of the paper state there was no "exposure–response" relation, and that "no increased risk of brain tumors was observed for brain areas receiving the highest amount of exposure," but this is contradicted by data reported in the paper.

In Table 5, for example, data showed that when the duration of the cell phone subscription was more than 4 years (the highest exposure), there was greater than a 3-fold risk for brain cancer (OR, 3.74). There was also close to a 3-fold risk when the number of cell phone calls was greater than 2638, the highest exposure (OR, 2.91).

Of importance, there are discrepancies between the tables in the paper and the text, which suggests that "the results may be misleading," Mr. Morgan and colleagues write.

As an example, the text reports that 35% of cases and 34% of control subjects had operator data since the start of their subscription; if calculated out, that would mean that there would be 123 cases and 200 control subjects with this information, they note.

However, Table 4 reports 196 cases and 360 control subjects. The reason for this discrepancy is unclear, "but it would likely make a large difference in the associated risks as there would be 37% fewer cases and 39% fewer controls," they write.

An odd finding is that ipsilateral risk was smaller than contralateral risk, which contradicts the findings of other studies, the authors write. The most likely explanation is that the reported laterality was not consistent with use. Nevertheless, these results are of major concern, Mr. Morgan and colleagues write.

More Research and Precautions Needed

"There is a great need for research, and we need more precautionary messages," said Dr. Moskowitz. "I don't know if we have enough science to come up with standards at this point, but we have a fair amount of evidence for increased brain tumor risk, as well as to the reproductive organs."

About a dozen countries have taken precautionary steps, so it would "not be radical for the United States government to do so as well," he said. No one is suggesting that we give up cell phones, "but there are ways to reduce the risk."


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.