An Examination of Poor Study Design
Resulting in an Underestimates of the Risk of Brain Tumors
Lloyd Morgan, January 23, 2008
INTERPHONE is a 13-country case-control study examining the risk of
acoustic neuroma, glioma, meningioma and parotid gland tumors from
cellphone use. Eleven studies [1-11] have been published on the risk of
brain tumors from cellphone use and two studies on the risk of parotid
gland tumors (a salivary gland located near to the ear). [12-13]. Two
of the eleven Interphones studies were studies of 5 countries [1,2],
which partially overlap other studies. Therefore this examination
focuses on 9 studies for risk of a brain tumor from cellphone use.
That all 11 of the studies showed use of a cellphone provided
?protection? from a brain tumor and reported there was "no risk" for
brain tumors from "regular" cellphone use, [1-11] is a fraud
perpetrated on the public. This should not be a surprise, given that a
substantial proportion of the cost of these studies has been paid for
by the cellphone industry.
As I will show, there are six major flaws in these studies, five of
which underestimate the risk of tumors. Yet, as we will see, and this
is very alarming, in spite of these flaws, where the tumors are on the
same side of the head where the cellphone was held and the phone was
used for 10 year or more, all of the results show a risk of tumors.
Before we examine the flaws I will discuss how you can detect flaws
similar to what exists within the Interphone Protocol. First, examine the
full set of odds ratios (ORs) within any single study, or better yet,
combinations of Interphone studies.
An odds ratio (OR) is a ratio that measures the odds of cellphone
radiation exposure for cases compared to controls. In effect, it is the
risk of tumors from cellphone use. Cases are subjects with a disease
(in this discussion, brain tumors). Controls are subjects without the
disease. Controls are chosen at random, and then matched to the cases
by various factors. Typically these factors are age, gender,
residential region, education and social economic status (SES).
Reported odds ratios are adjusted for each of the matched factors in
order to minimize confounding.
Studies report both the OR and the 95% Confidence Interval (CI). 95%
confidence means there is >95% confidence that the result is not due to
chance. If the CI spans 1.0 it is said to be non-significant and if it
does not span 1.0 it is said to be significant. The reported OR is the
most likely value within the CI.
Let's suppose that there is no risk of tumors from cellphone use. What
would you expect to find by examining all the ORs reported in a study,
or a set of studies? If there were no risk, then regardless of the
significance or non-significance, roughly half of the ORs would be
elevated (OR>1.0) and roughly half would not be elevated (OR<1.0).
Further about 95% of the ORs will be non-significant and about 5% of
the ORs will be significant. Think about tossing a coin. About half the
time heads comes up and about half the time tails comes up. The ratio
of heads to tails approaches 1.0 as the number of coin tosses increase.
Findings with OR=1.0 are not counted.
Risk of Brain Tumors from Cellphone Use
An examination of the ORs for all brain tumors reported by the 9
Interphone studies found that 65 of the results had an OR>1.0 and 308
had an OR<1.0 (observed ratio=4.7; expected ratio ~1.0).
For significant findings, there were 3 results with OR>1.0 and 50
results with OR<1.0 (observed ratio=16.7; expected ratio ~1.0). For
the 3 results reporting OR>1.0, all had >10 years of exposure with the
tumor on the same side of the head where the cellphone was held.
It would appear that when a cellphone is used for >10 years and the
tumor is on the same side of the head where the cellphone was used,
there is a significant risk of a brain tumor.
As I said in the introduction, to see if a study, or a set of studies,
are flawed, all you have to do is to examine the odds ratios. Out of
373 published ORs, 65 are >1.0 (20%) and 308 are <1.0 (80%). The
ratio should be close to 1.0, but it is 4.8 (309/65=4.8). As previously
noted, of the 271 findings we would expect, about 5% to be significant
findings. Yet there are 53 significant findings (14%) when the expected
would be about 14 (5%).
A closer examination of all the significant findings for <10 is
incredulous. There are 50 significant findings with OR<1.0 and zero
significant findings with OR>1.0. A significant OR<1.0 indicates
that cellphone use protects the user from brain tumors! There are two
possible conclusions to explain this incredulity: either using a
cellphone provides protection from brain tumors, or there are major
flaws in the studies.
Now let's examine the 6 design flaws in the Interphone studies. Five of
these flaws result in an underestimation of risk. Could this be the
reason that there are 50 significant finding showing a protection and
none showing a risk of brain tumors from cellphone use?
Flaw 1: Selection Bias
The first flaw is called selection bias. It is likely the result of the
low percentage of controls that participated in the studies (weighted
average of 59%). Think about being randomly selected for a cellphone
study. You are told you will be asked to answer a long questionnaire.
If you use a cellphone you are more likely to agree to participate than
if you do not use a cellphone. If this happens it is called selection
bias. Selection bias will result in an underestimation of risk.
Flaw 2: Inclusion of Tumors Outside the Cellphone's Radiation Plume
The second flaw is the inclusion of all brain tumors without regard to
their location. Because the cellphone's radiation plume only penetrates
a short distance into the head, nearly all of this radiation is
absorbed by the temporal lobe, the acoustic nerve, or the parotid gland
(not discussed in this column).
Even when cellphone exposure of one side of the head is considered on
the side where the cellphone was held, a substantial portion of half
the brain is unexposed (the opposite side is completely unexposed).
The colors in Figure 2 indicate The Specific Absorption Rate (SAR) of
is the amount of power absorbed by brain tissue, in
this discussion, brain tissue, in Watts per kilogram.
The temporal lobe and the acoustic nerve, the nerve from the ear to the
brain (not shown), and the parotid gland, absorb almost all of the
cellphone's radiation plume. The plume is only a small portion of the
brain and is entirely on the side of the head where the cellphone is
used. The depth of the cellphone's radiation plume's
penetration is quite shallow. More that 67% of the radiation plume's
power is absorbed within an inch (2.54 cm) of the surface of the skull.
Studies that include brain tumors outside of the cellphone radiation
plume underestimate the risk of brain tumors.
Flaw 3: Latency Time and Definition of Regular User
The third flaw is the definition of "regular" cellphone use in relation
to a reasonable latency time. "Regular" cellphone use is defined as use
of a cellphone on average once per week for at least 6 months. Exposure
within 1 year of the diagnosis date is not considered. The result of
this definition, combined with the incredibly fast rate of new
cellphone users, is to overweight "regular" users with, an incredibly
large group of short-term users, far too short a time to expect a tumor
to be diagnosed.
Latency time is the time from an exposure to the diagnosis of a tumor.
What we know about the latency time for brain tumors comes from
ionizing radiation exposures. Based on ionizing radiation, the latency
time for brain tumors is between 25 and 40 years, similar to the
latency time of lung cancer from tobacco exposure.
For the 9 Interphone studies, using weighted averages for cases or
controls, we see that 0.61% of cases and 10% of controls have used a
cellphone for 10 years or more, and 18% of cases and 21% of controls
have used a cellphone for 5 years or more.
The result: for a reasonable latency time, it would be unlikely to find
any risk of tumors, given the percentage of cases and controls. Yet, as
we saw in the Studies on the Risk of Brain Tumors from Cellphone Use
section, there is a risk. "It would appear that when a cellphone is
used for >10 years and the tumor is on the same side of the head where
the cellphone is used, there is a significant risk of a brain tumor."
Because such a large percentage of "regular" users have used a
cellphone for an unreasonably short latency time the reported results
for <10 years as well as for >10 years (6.3% of cases) are an
underestimation of risk.
Flaw 4: Children and Young Adults Are Not Included in Interphone Studies
The Interphone Protocol states that cases be between 30 and 59 years of
age. While a few studies have included cases as young as 20, the
non-inclusion of <20 year olds results in an underestimation of
risk. Why? Because children, with their high rate of cell division, are
at higher risk of tumors than adults. As we know there a considerable
proportion of cellphone use by children. And, we know that children,
especially teenagers, spend more time on cellphones that do adults.
Flaw 5: Cellphone's Radiated Power
It is reasonable to expect that risk of a tumor from a cellphone, after
a reasonable latency time, would be the cellphone's power multiplied by
cumulative time of use. In the early days of cellphone use all
cellphones used analog technology. These always radiated a fixed amount
of power (~2 Watts). Analog cellphones use has been totally displaced
by digital cellphones. Digital cellphones have a feature called
Automatic Power Control or APC. At the beginning of a call the
cellphone radiates maximum power (~2 Watts) but quickly reduces the
power so the radiated power is sufficient to have a reliable link to
the cell tower (AKA masks or base stations). The result is that
cellphones radiate far less power in urban areas compared to rural
areas. This is because cell phone towers are much closer in urban areas
compared to rural areas so the cellphone radiates less power in urban
areas and more power in rural areas. When rural and urban cellphones
are not reported separately the result is an underestimation of risk.
Flaw 6: Number of Cases Included in a Study
The weighted average time in these 9 studies for a case to be eligible
for inclusion in the study was only 2.6 years. When one considers 4 of
the 5 previous flaws, it becomes obvious that such a short period of
time for eligibility will result in too few cases to resolve these
flaws. For example, if tumors were limited only to the exposed region
of the brain then there would be far fewer cases; if a reasonably long
latency time was included, again there would be far fewer cases; if
children had been included there would have been more cases; and, if
rural users were to be compared to the far larger number of urban users
a much larger number of cases would need to be eligible to participate
in the Interphone Study.
Conclusion and Discussion
With five flaws, each independently underestimating the risk of tumors,
it is no wonder why the Interphone studies report a large number of
results suggesting cellphone use protects the user from having a brain
The Interphone Study has substantial funding from the cellphone
industry. The additional cost to resolve these flaws could have been
accomplished if the industry provided more funds. In addition if the
participating countries had anticipated the potential cost of a
pandemic of brain tumors, the cost effectiveness of contributing
substantially more funds, would have been obvious. Lastly, relying on
the cellphone industry funding is equivalent to having the fox guard
the hen house.
The cellphone industry will state that there is a "firewall" between
their funds and the research teams who do the studies. While it is true
that the cellphone industry provides the funds to another organization
(UICC) which then decides on the teams that will do each study, the
researchers are aware that most of their funds are coming from the
cellphone industry. While I do not doubt the integrity of the
researchers, I also believe there in an inherent conflict-of-interest,
best described by the saying, "Don't bite the and that feeds you."
The fundamental problem is not conflict-of-interest. The fundamental
problem is the Interphone Protocol. While I have no evidence, it would
appear that the cellphone industry influenced the Protocol, if not
actively participating in its creation. The end result is the Protocol
is designed to not find any risk. That it has found a risk is sobering!
Tragically, the window of time to do a large, well-designed
case-control study is closed. Case-control studies require exposed and
unexposed subjects. It is no longer possible to find unexposed
1. Schoemaker et al. Mobile phone use and risk of acoustic neuroma:
results of the Interphone case"control study in five North European
countries. British Journal of Cancer. 2005 Oct 3; 93 (7): 842-8. A
Scandinavian and United Kingdom Interphone Study reporting, "The risk
of acoustic neuroma in relation to regular mobile phone use in the
pooled data set was not raised . . ."
2. Lahkola et al. Mobile phone use and risk of glioma in 5 North
European countries. International Journal of Cancer: 120, 000"000
(2006). A Scandinavian and UK Interphone Study reporting, "We found no
evidence of increased risk of glioma related to regular mobile phone
use . . ."
3. Lonn et al. Mobile Phone Use and the Risk of Acoustic Neuroma.
American Journal of Epidemiology 2004; 159:277?283. A Swedish
Interphone Study reports, "The overall odds ratio for acoustic neuroma
associated with regular mobile phone use was 1.0." [i.e., no risk]
4. Christensen et al. 2004, Cellular Telephone Use and Risk of Acoustic
Neuroma. American Journal of Epidemiology, Vol. 159, No. 3, 2004;159:
277?283. A Danish Interphone Study reporting, "We did not observe
increased risk of acoustic neuroma among regular cell phone users . .
5. Lonn et al. Long-Term Mobile Phone Use and Brain Tumor Risk,
American Journal of Epidemiology 2005; 161: 526?535. A Swedish
Interphone Study reporting, "For regular mobile phone use [there was no
risk) for glioma . . .and for meningioma."
6. Christensen et al. Cellular telephones and risk for brain tumors: A
population-based, incident case-control study. Neurology. 2005 Apr 12
;64 (7): 1189-95. Erratum in: Neurology. 2005 Oct 25; 65 (8):1324. A
Danish Interphone Study of glioma and meningioma reported, "The overall
analyses did not reveal an increased risk for . . . regular use of a
cellular telephone . . ."
7. Schüz et al. Cellular Phones, Cordless Phones, and the Risks of
Glioma and Meningioma (Interphone Study Group, Germany). American
Journal of Epidemiology 2006 Mar 15; 163 (6): 512-20. Epub 2006 Jan 27.
A German Interphone Study reporting, "For regular cellular phone use
[there was no risk] . . . for glioma and . . . for meningioma."
8. Takebashi et al. Mobile phone use and acoustic neuroma risk in
Japan. Occupational Environmental Medicine, 2006; 63; 802-807. A
Japanese Interphone Study reporting, "No significant increase of
acoustic neuroma risk was observed . . . when regular mobile phone . .
. was [used]."
9. Klaeboe et al. Use of mobile phones in Norway and risk of
intracranial tumours. European Journal of Cancer Prevention 2007,
16:158?164. A Norwegian Interphone Study reporting, "No increased risk
was observed for gliomas, meningiomas . . . and acoustic neuromas . . .
among regular mobile phone users . . ."
10. Hours et al. Cell Phones and Risk of brain and acoustic nerve
tumours: the French INTERPHONE case-control study. Revue
d?Épidémiologie et de Santé Publique 2007 Oct, 55 (5): 321-32. Epub
2007 Sep 11. French. A French Interphone Study reporting, "Regular cell
phone use was not associated with an increased risk of neuroma . . ."
11. Hepworth et al. Mobile phone use and risk of glioma in adults:
case-control study. BMJ Online First bmj.com. A United Kingdom
Interphone Study reporting, "Overall, we found no raised risk of glioma
associated with regular mobile phone use . . ."
12. Sadetzki et al. Cellular Phone Use and Risk of Benign and Malignant
Parotid Gland Tumors?A Nationwide Case-Control Study. American Journal
of Epidemiology Advance Access published December 6, 2007. An Israeli
13. Lönn et al. Mobile phone use and risk of parotid gland tumor.
American Journal of Epidemiology 2006 Oct 1; 164 (7): 637-43. Epub 2006
Jul 3. A Swedish Interphone Study.