Telecommunications, electromagnetic fields, and human health

Oct 30

ENVIRONMENTAL CLAIMS JOURNAL 2019, VOL. 31, NO. 2, 93–132 https://doi.org/10.1080/10406026.2019.1603442

Telecommunications, Electromagnetic Fields, and Human Health

Robert A. Michaels

RAM TRAC Corporation, Schenectady, New York

ABSTRACT Telecommunication generates electromagnetic fields (EMFs) at radio and microwave frequencies. Transmitters have proliferated with siting of wireless communication networks, often co-located among other transmitters. “Cell” phones also have proliferated, representing small transmitters used in contact with human heads, and stored on human bodies. Telecommunications equipment is ubiquitous, and EMF exposure prolonged, raising the issue of possible health risks. Such risks, if any, must be managed. For example, epidemiology studies reported higher exposure to analog cell phone EMFs among brain cancer patients than among controls, but those risks were “managed” via replacement of analog phones with today’s digital phones, which have not been associated with human cancer. Challenges remain, recently from rodent bioassays that show dose-related association of lifetime exposure to cell-phone-type EMFs with heart schwannomas (cancers of Schwann cells, which insulate nerve cells) in male rats, though not in females. Human cancer risk, if any, remains to be characterized and quantified, which partly will depend upon whether EMFs indeed are non-ionizing as has been assumed, and whether a threshold or non-threshold (genotoxic) mechanism caused the cancers in the male rats. Health concerns have motivated further exposure reduction suggestions, and sometimes opposition to siting transmitters. Credible, objective explication of technical information to primarily non-technical audiences is necessary to support informed public participation and dispassionate weighing of telecommunications risks and benefits in community decision-making. Ultimately, experts and non-experts should adhere to the “precautionary principle,” requiring adoption of reasonably (but not excessively) pessimistic exposure and risk assumptions, whether or not they are likely to materialize. Introduction This report assesses potential electromagnetic field (EMF) impacts, such as those arising from wireless communication networks of mobile phone and stationary emergency communication system transmitters and repeaters. Stationary radio frequency (RF) and microwave frequency (MW) transmitter CONTACT Robert A. Michaels ram@ramtrac.com; www.ramtrac.com RAM TRAC Corporation, 3100 Rosendale Road; Schenectady, NY 12309-1510. (cid:1) 2019 Taylor & Francis Group, LLC

94 R. A. MICHAELS

networks often are pole-mounted or tower-mounted, and may cover large areas, as in wireless statewide emergency communications systems. Indeed, emerging 5G “smartphone” technology utilizes extremely short (millimeter), high frequency waves (between 30 and 300 GHz). Such frequencies have short range compared with more common 4G technology. This means that trans-

mitters must be sited closer together, in exponentially denser arrays. Such transmitters are highly directional, protecting potential human receptors, but necessary 5G infrastructure will exert significant environmental impacts. Assessment of potential transmitter health impacts includes four primary components:

1. quantifying the EMF power density at a “critical receptor,” 2. evaluating compliance with U. S. Federal Communication Commission (FCC) standards, 3. assessing potential cancer risks, and 4. assessing potential non-cancer risks.

The issue of health effect causation by EMFs generated by wireless communications has been of concern over at least the past two decades, fueled by occasional though inconsistent epidemiological associations with cancer (see, for example, Alpert 2000). Emergency communication antenna networks include transmitters and receivers. Whereas receivers may exert impacts that are unassociated with generation of EMFs (for example, visual impacts), such impacts are addressed elsewhere. In this report, only transmitters are addressed, consistent with the current focus on adverse health effects potentially caused by exposure of people to EMFs generated by such sources. Concern about antennas may be exaggerated by lack of awareness that roughly half of vis-

ible antennas are receivers, not transmitters, and do not generate EMFs. Transmitters can be of many types and models, but they will be treated generically here, in discussion omitting specific information about manufacturers, models, and most other technical details. The more useful focus here will be upon consideration of a worst-case equipment scenario. A worst-case scenario is a scenario that can be assumed to overestimate actual health impacts. The premise of a worst-case assessment is that it preserves the maximum possible degree of flexibility, constraining a telecommunications (or other) project only to within acceptable impacts rather than to specific equipment. For practical reasons, this approach is

preferred over prescribing a unique configuration, because pre-defining a unique configuration is likely to undermine attempts to maximize system performance within a multitude of real-world constraints unrelated to potential risks to human health.

TELECOMMUNICATIONS, ELECTROMAGNETIC FIELDS, AND HUMAN HEALTH 95

The scope of this study includes preexisting transmitters that might be present on mounting structures, such as poles or towers, subject to limited availability of data describing such additional sources. The scope excludes risks of physical accidents, such as vehicular accidents, falling antennas, construction mishaps, and lightning strikes on poles or towers. Such risks are unassociated with the interaction of EMFs with human physiological

and biochemical systems.

Methods Methods utilized in this analysis include standard health risk assessment (HRA) procedures for quantifying worst-case risks to public health, and pro-

cedures specified by the OET Bulletin 65 of the U. S. Federal Communications Commission, Office of Engineering and Technology (US FCC 1997a, 1997b). A worst-case risk is a risk that is unlikely to be exceeded by the actual (but unknown) risk. To dispel a common misconception, a worst-case risk is not a maximum possible overestimation of the actual risk. Indeed, risks (including worst-case risks) always can be more extremely overestimated, until a risk value of unity (probability of one) is attained, corresponding to 100 percent certainty that an adverse health effect that is being assessed actually will materialize as a result of the exposure. Health effects potentially elicited by MW and RF EMFs are assessed via evaluation of medical, epidemiological, toxicological, and other scientific literature. Calculated worst-case exposure levels are compared with levels reported to cause health effects in animals or people. Health risk assess-

ments may report health effects associated with extreme exposure to an agent, but calculated worst-case exposure nonetheless may fail to attain levels capable of causing them actually. This assessment is based upon input engineering data provided in connection with a statewide emergency communication network, or otherwise originating from transmitter manufacturers’ product literature. Such information is subject to verification via standard and appropriate safeguards. These safeguards most notably include technical specifications that are set forth in transmitter manufacturers’ product literature, and that are publicly accessible. The public therefore is protected. Most notably, if data supporting a permit application are found to be false, permits supported by the data may be subject to invalidation, depending upon the seriousness of data discrepancies.

Findings Agents of Potential Concern

The agents of potential concern in this assessment are electromagnetic waves emitted by RF and MW transmitters such as those that might be

96 R. A. MICHAELS

deployed in configuring a wireless statewide emergency communications network. Radio transmitters would utilize an approximate frequency band of 700–900 megahertz, whereas directional point-to-point microwave transmitters would utilize an unspecified portion of the microwave band. Frequencies below 1,000 megahertz are classified as radio waves, whereas

those from 1,000 to 2,000 megahertz are classified as microwaves. Radio waves and microwaves are physical rather than chemical agents, but analysis of risks that they potentially might pose is generally similar to analysis of chemical agent risks. In the case of radio waves and microwaves, however, the physics rather than the chemistry of the agent is of primary concern. Radio waves and microwaves are electromagnetic radiations in the frequency band utilized by radios and microwave ovens, respectively, within the larger spectrum of electromagnetic radiation types. Radiation generated by radio and microwave transmitters is believed to be non-ionizing, like visible light, insufficiently energetic to break chemical bonds in

organic molecules such as DNA. Ionizing radiation, in contrast, can break chemical bonds in DNA, causing mutation, and also may cause cancer. Non-ionizing radiation is distinct from ionizing radiation, such as X-rays or nuclear fallout, which may be released by nuclear weapons or nuclear reactor accidents. Ionizing radiation is uninvolved with emergency telecommunications network facilities. The section below uses only elementary algebraic expressions to inter-relate selected parameters describing electromagnetic radiation.

Electromagnetic Radiation Spectrum The various forms of electromagnetic radiation are distinguished by their characteristic spectrum, that is, their range of wavelengths (expressed in

familiar units of length, such as meters, M). The Handbook of Chemistry and Physics (Weast and Lide 1989) defines “electromagnetic spectrum” as follows: The range of wavelengths over which electromagnetic radiation extends. The longest waves (105 to 10 (cid:1)3 M) are radio waves, the next longest (10 (cid:1)3 to 10 (cid:1)6 M) are infrared waves, then comes the narrow band (4 to 7 x 10 (cid:1)7 M) of visible light, followed by ultraviolet waves (10 (cid:1)7 to 10 (cid:1)9 M), x-rays (10 (cid:1)9 to 10 (cid:1)11 M), and gamma rays (10 (cid:1)11 to 10 (cid:1)14 M) (Weast and Lide 1989, page F-94). The types of electromagnetic radiation, listed above in order of decreasing wavelength, also exhibit different energy (E) and frequency [Greek nu (m) usually expressed in cycles or vibrations per second, in either case, hertz (Hz)]. Indeed, wavelength varies inversely with frequency; that is, the longer the wavelength [Greek lambda (k)], the lower the frequency [Greek nu

TELECOMMUNICATIONS, ELECTROMAGNETIC FIELDS, AND HUMAN HEALTH 97

(m)]. In a vacuum, through which electromagnetic radiation travels at the speed of light (2.99793 (cid:3) 108 M/s) regardless of wavelength or frequency, the following three equivalent equations apply:

c ¼ vk; m ¼ c=k; and k ¼ c=v:

Shorter wavelengths, and higher frequencies, of electromagnetic radiation exhibit higher energies, as expressed by Planck’s equation: E ¼ hv; where

E is energy (typically expressed in ergs1 or joules, where 1 joule ¼ 107 ergs), h is Planck’s constant, 6.62620 (cid:3) 10 (cid:1)27 erg sec (6.62620 (cid:3) 10 (cid:1)34 joule sec), and m is frequency (Greek nu, expressed in Hz). Exposure to EMFs may be expressed in terms of electric field strength, magnetic field strength, or electromagnetic power density (US FCC 1997), defined by the U. S. Federal Communications Commission (FCC) as follows:

Electric field strength (E): a field vector (directional) quantity that represents the force (F) on an infinitesimal unit positive test charge (q) at a point divided by that charge, expressed in units such as volts (V) per meter (M), via the equation: E ¼ V=M:

Magnetic field strength (H): a field vector that is equal to the magnetic flux density divided by the permeability of the medium, expressed in units such as amperes (A) per meter (M), via the equation:

H ¼ A=M: Electromagnetic power density (S): power per unit area normal (at right angles) to the direction of propagation, expressed in units such as watts (W) per square meter, via the equation:

S ¼ W=M2:

The parameters (and units) above are interconvertible, because electric and magnetic fields have been unified in physics within the inclusive category of EMFs, arising from the same physical source: quanta. In the visible spectrum, quanta of light are termed photons, but photons are simply

1 An erg is the primary unit of energy (and also of work) in the centimeter-gram-second (CGS) system of measurement, and is defined as the energy (or work) associated with a force of 1 dyne acting through a distance of 1 cm. A dyne is defined as the force required to accelerate a 1-gram mass by 1 cm/s, in accordance with the rule that force (F) equals mass (m) times acceleration (a), that is, F ¼ ma.

98 R. A. MICHAELS

a subclass of quanta. Electric and magnetic fields are related by the equation, where the units are as exemplified in the above definitions:

S ¼ E

Exposure Routes of Potential Co Exposure of concern in this a atmosphere. Inhalation, ingesti sures to chemicals are irrelevan encounter electromagnetic radi Indeed, EMFs are ubiquitous. EMFs, emitted by sources suc

power lines, and even the sun tower-mounted transmitters re EMFs associated with these sou

Exposure Assessment

Exposure to EMFs may be exp magnetic field strength (H), or U. S. FCC standards setting for radiation are expressed in term In this section, however, the po ment, as it is the most general vation that limits on EMF consideration of both the electr associated with such fields. Co

1997b), the exposure equation i

Table 1. U. S. Federal Communicatio Population Limits for Uncontrolled Expo Frequency range: Electric field from to strength (E) (MHz) (MHz) (V/M) 0.3 1.34 614 1.34 30 824/fb 30 300 27.5 300 1,500 … 1,500 100,000 … aAdapted from US FCC. Evaluating Complian Electromagnetic Fields. Washington, DC. U. S Technology, OET Bulletin 65 (Edition 97-01), bf ¼ frequency in megaherz (MHz). cPlane wave ¼ equivalent power density.

E2=3; 770 ¼ 37:7H2:

oncern assessment is whole-body exposure via the ion, and dermal exposure that typify expont in the case of exposure to EMFs. People iation when they are located within EMFs. People typically are located within various ch as household appliances, radio stations,

n. Human exposure to EMFs generated by equires people to come within the range of urces.

pressed in terms of electric field strength (E), r electromagnetic power density (S). Indeed, rth upper exposure limits to electromagnetic ms of all three of these parameters (Table 1). ower density is selected for exposure assessparameter. This is consistent with the obserstrength must be derived based upon ric and the magnetic field effects that may be onsistent with U. S. FCC (US FCC 1997a,

is as follows:

on Commission (FCC) Maximum Permissible General osure to Electromagnetic Fields.a Magnetic field strength (H) Power density (S) Averaging time (A/M) (mW/cm2) (minutes) 1.63 (100)c 30 2.19/f [2] (180/f)b,c 30 0.073 0.2 30 … f/(1,500)b 30 … 1.0 30 nce with FCC Guidelines for Human Exposure to Radiofrequency S. Federal Communications Commission Office of Engineering and Supplement B (Edition 97-01), p. 47, Nov. 1997.

TELECOMMUNICATIONS, ELECTROMAGNETIC FIELDS, AND HUMAN HEALTH 99

S ¼ 1; 000 n P G C ou S ¼ electromagnetic power density (

1 W ¼ 1,000 mW), n ¼ the numbe the power to the transmitter in watts of the antenna to the critical recept the mounting height is adjusted by assumed height of the person. G ¼ t in the principal transmission directi hypothetical critical receptor (person

G ¼ 10 ½ða (cid:1) bÞ dB

a ¼ decibels (dB) in the direction of in the main transmission direction, C

imum power output (default ¼ 1), upward signal reflection from the gro

Exposure Scenarios

Potential exposure of people to EM transmitters, including radio and M upon scenarios such as the following

1. a tower-mounted radio transmit 2. a pole-mounted radio transmitte 3. a tower-mounted microwave tra 4. a pole-mounted microwave tran

Based upon the equations set forth of a hypothetical critical receptor lo calculated for each of the four scena the worst-case scenario (pole-mount Table 2, provided as an exhibit. In Table 2, the distance from each person) is quantified as the transmitt assumed six-foot person height, rathe

ter from the ground. In some rea mounted above a structure of some s plicity, the assumption is made here and that its occupants still are at gro ing for prolonged periods on the roo

C =PR2; where : utput reflection (mW/cm2), 1,000 ¼ conversion factor:

er of transmitters (dimensionless), P ¼ s (W), R ¼ distance from the centerline tor. If the receptor is below the tower, y subtraction of six feet, which is the the ratio of antenna gain (decibels, dB) ion, to the gain in the direction of a n) near the source, where

(cid:4) B=10 dB ; and where : f the critical receptor; b ¼ decibels (dB) C ¼ correction factor: actual/maxoutput and C ¼ correction factor for reflection ound (default ¼ 1.6).

MFs produced by emergency network MW sources, may be quantified based g:

tter 180 feet above ground, er 60 feet above ground, ansmitter 180 feet above ground, and nsmitter 60 feet above ground.

h earlier, potential worst-case exposure ocated directly beneath each source is arios described above. Calculations for ted radio transmitter) are illustrated in

h transmitter to the critical receptor (a ter mounting height minus a nominally er than just the height of the transmit-

al-world situations, the transmitter is sort, such as a utility building. For simthat the building is just one story tall, ound level, as they would not be workof of the building. Other structures on

a gni mumix re )s 100 R. A. MICHAELS

itroppus

sannetna

gnittimsnart

oidar

elibom

dnal detnuom-elop

yb detareneg

sdleif

citengamortcele

ot erusopxe

dellortnocnu

laitnetoP

.2 elbaT .krowten snoitacinummoc ycnegreme ediwetats sseleriw lacitehtopyh :rotcafnoitcerroC rotcafnoitcerrocrewoP xamdetaR d]Bd01/Bd)b (cid:1) a([ ¼ mumixam/rewoplautca noitceridnirewoprof otrewoP rettimsnarT forebmuN 01 G c,a b a a a ewop tuptuorewop rotpecerlacitircfo P,)s(rettimsnart ycneuqerf srettimsnart annetnagnittimsnarT … (cid:1) b a 01/)b a( C G P f n tuptuo … … … sttaw( )Bd( )Bd( )slebiced/slebiced( )sttaw( )zHM( oiranecsnoitarugifnoclautcA (cid:1) (cid:1) 005 0 8 8.0 00.1 851.0 04 698ot428 1 YX-E09H448BD (cid:1) 005 0.21 0.4 8.0 00.1 851.0 001 698ot428 1 YX-09H448BD … (cid:1) 7.21 07.2 0.1 00.1 001.0 87 668ot158 1 08H448BD (cid:1) (cid:1) 001 8.11 2.31 5.2 00.1 300.0 01 198ot428 1 501H478BD … (cid:1) (cid:1) 0.9 0.41 3.2 00.1 500.0 001 068 1 CX3T-908BD … (cid:1) (cid:1) 0.9 0.12 0.3 00.1 100.0 008 058 1 inmOCX-985BD … (cid:1) 7.11 7.2 9.0 00.1 621.0 001 069ot098 1 2A_50-21-09VRSME … (cid:1) 7.11 7.2 9.0 00.1 621.0 001 069ot098 1 2A_50-21-09VRSME … (cid:1) 7.11 7.2 9.0 00.1 621.0 001 069ot098 1 GX00-21-09VRSME … (cid:1) (cid:1) 2.21 8.71 0.3 00.1 100.0 3.39 088ot078 1 2LAXX-21-501VRSME (cid:1) (cid:1) 07 0 02 2 00.1 010.0 52 668ot158 1 4/5097-DPLletnA … (cid:1) (cid:1)

…

9.71

1.21(cid:1)

0.3(cid:1)

00.1

100.0

498ot088

)nozireV(NID-2109CS (cid:1) (cid:1) 005 01 51 5.2 00.1 300.0 001 538 1 )pihw(CX-3T018BD (cid:1) (cid:1) 005 4.9 6.01 2 00.1 010.0 001 048 1 )pihw(CX-3U018BD (cid:1) (cid:1) 005 0.21 0.61 8.2 00.1 200.0 001 068 1 X-F218BD oiranecsnoitarugifnocesac-tsrow (cid:1) (cid:1) 005 0 8 8.0 00.1 851.0 005 698ot428 1 YX-E09H448BD (cid:1) 005 0.21 0.4 8.0 00.1 851.0 005 698ot428 1 YX-09H448BD … (cid:1) 7.21 07.2 0.1 00.1 001.0 005 668ot158 1 08H448BD (cid:1) (cid:1) 001 8.11 2.31 5.2 00.1 300.0 001 198ot428 1 501H478BD … (cid:1) (cid:1) 0.9 0.41 3.2 00.1 500.0 005 068 1 CX3T-908BD … (cid:1) (cid:1) 0.9 0.12 0.3 00.1 100.0 005 058 1 inmOCX-985BD … (cid:1) 7.11 7.2 9.0 00.1 621.0 005 069ot098 1 2A_50-21-09VRSME … (cid:1) 7.11 7.2 9.0 00.1 621.0 005 069ot098 1 2A_50-21-09VRSME … (cid:1) 7.11 7.2 9.0 00.1 621.0 005 069ot098 1 GX00-21-09VRSME … (cid:1) (cid:1) 2.21 8.71 0.3 00.1 100.0 005 088ot078 1 2LAXX-21-501VRSME (cid:1) (cid:1) 07 0 02 2 00.1 010.0 07 668ot158 1 4/5097-DPLletnA … (cid:1) (cid:1) 9.71 1.21 0.3 00.1 100.0 005 498ot088 1 )nozireV(NID-2109CS (cid:1) (cid:1) 005 01 51 5.2 00.1 300.0 005 538 1 )pihw(CX-3T018BD (cid:1) (cid:1) 005 4.9 6.01 2 00.1 010.0 005 048 1 )pihw(CX-3U018BD (cid:1) (cid:1) 005 0.21 0.61 8.2 00.1 200.0 005 068 1 X-F218BD

)deunitnoc(

2 elbaT fotnecreP elbawollamumixamCCF :rotcafnoitcerroC otecnatsiD otnoitceriD gnitnuoM mumixam taytisnedrewop rewoP otecnatsiD noitcelferdnuorg f a e rotpecerlacitirc rotpecerlacitirc thgieh elbawolla rotpecerlacitirc ytisned rotpecerlacitirc rotpecerotkcab annetnagnittimsnarT … … … … H S R C noitcelfer … 2 2 … … )teef( )seerged( )teef( ) ( )mc/Wm( )mc/Wm( )mc( oiranecsnoitarugifnoclautcA (cid:1) 45 09 06 312.0 75.0 22100.0 646,1 46.1 YX-E09H448BD (cid:1) 45 09 06 435.0 75.0 50300.0 646,1 46.1 YX-09H448BD (cid:1) 45 09 06 462.0 75.0 15100.0 646,1 46.1 08H448BD (cid:1) 45 09 06 100.0 75.0 10000.0 646,1 46.1 501H478BD (cid:1) 45 09 06 710.0 75.0 01000.0 646,1 46.1 CX3T-908BD (cid:1) 45 09 06 720.0 75.0 51000.0 646,1 46.1 inmOCX-985BD (cid:1) 45 09 06 393.0 26.0 34200.0 646,1 46.1 2A_50-21-09VRSME (cid:1) 45 09 06 393.0 26.0 34200.0 646,1 46.1 2A_50-21-09VRSME (cid:1) 45 09 06 393.0 26.0 34200.0 646,1 46.1 GX00-21-09VRSME (cid:1) 45 09 06 300.0 85.0 20000.0 646,1 46.1 2LAXX-21-501VRSME (cid:1) 45 09 06 900.0 65.0 50000.0 646,1 46.1 4/5097-DPLletnA (cid:1) 45 09 06 010.0 65.0 60000.0 646,1 46.1 )nozireV(NID-2109CS (cid:1) 45 09 06 110.0 75.0 60000.0 646,1 46.1 )pihw(CX-3T018BD (cid:1) 45 09 06 430.0 75.0 91000.0 646,1 46.1 )pihw(CX-3U018BD (cid:1) 45 09 06 500.0 75.0 30000.0 646,1 46.1 X-F218BD oiranecsnoitarugifnocesac-tsroW (cid:1) 45 09 06 366.2 75.0 72510.0 646,1 46.1 YX-E09H448BD (cid:1) 45 09 06 866.2 75.0 72510.0 646,1 46.1 YX-09H448BD (cid:1) 45 09 06 386.1 75.0 36900.0 646,1 46.1 08H448BD (cid:1) 45 09 06 110.0 75.0 60000.0 646,1 46.1 501H478BD (cid:1) 45 09 06 480.0 75.0 84000.0 646,1 46.1 CX3T-908BD (cid:1) 45 09 06 710.0 75.0 01000.0 646,1 46.1 inmOCX-985BD (cid:1) 45 09 06 769.1 26.0 31210.0 646,1 46.1 2A_50-21-09VRSME (cid:1) 45 09 06 769.1 26.0 31210.0 646,1 46.1 2A_50-21-09VRSME (cid:1) 45 09 06 769.1 26.0 31210.0 646,1 46.1 GX00-21-09VRSME (cid:1) 45 09 06 710.0 85.0 01000.0 646,1 46.1 2LAXX-21-501VRSME (cid:1) 45 09 06 420.0 65.0 31000.0 646,1 46.1 4/5097-DPLletnA (cid:1) 45 09 06 710.0 65.0 01000.0 646,1 46.1 )nozireV(NID-2109CS )deunitnoc( TELECOMMUNICATIONS, ELECTROMAGNETIC FIELDS, AND HUMAN HEALTH 101

.deunitnoC

TELECOMMUNICATIONS, ELECTROMAGNETIC FIELDS, AND HUMAN HEALTH 103

which transmitters may be mounted, such as water towers, typically would not be constructed over a utility building.

Pole-Mounted Radio Transmitters As Table 2 describes in quantitative detail, numerous pole-mounted radio transmitters were considered to exemplify equipment that might be deployed in configuring a statewide emergency communication network. Calculations based upon these specific transmitters revealed power densities at the critical receptor ranging from 0.00001 to 0.003 milliwatts per square centimeter (mW/cm2) under a realistic configuration scenario that assumes a 60-foot pole-mounting height and realistic power flows to the transmit-

ters. The resulting power density range compares favorably with an FCCallowable power density of 0.6 mW/cm2. Thus, the FCC-allowable power density ranges from 200 to 60,000 times that expected at the location of the critical receptor. When a worst-case equipment configuration scenario was assumed, in which each transmitter received the maximum power allowable by law (500 watts) or by engineering specifications, power densities at the critical receptor were found to range from 0.00006 to 0.01 mW/cm2. Thus, under maximum power to each transmitter, the allowable power density still ranged from 60 to 10,000 times that expected at the location of the crit-

ical receptor. In practice, transmitters often are clustered (“co-located”) on a common structure, such as a pole or tower. To consider the possible impact of this practice, all 15 transmitters were assumed to be operating from a single pole (notwithstanding that this might be physically impossible). Further, a second set of the same 15 transmitters was assumed to be co-located on the same pole, but operating under the worst-case (maximum-power) scenario. This effectively produced a scenario in which 30 transmitters were operating simultaneously. The resulting power density at the critical receptor was 0.09 mW/cm2, which is under 16 percent of the FCC-allowable

power density.

Tower-Mounted Radio Transmitters Numerous tower-mounted radio transmitters were likewise considered to exemplify equipment that might be deployed in configuring a statewide emergency communication network. Calculations based upon these numer-

ous specific transmitters revealed power densities at the critical receptor ranging from less than 0.00001 to 0.0003 mW/cm2 under a realistic configuration scenario that assumes a 180-foot tower-mounting height and realistic power flows to the transmitters. The resulting power density range

104 R. A. MICHAELS

compares favorably with an FCC-allowable power density of 0.6 mW/cm2. Thus, the FCC-allowable power density ranges from 2,000 to 60,000 times that expected at the location of the critical receptor. When a worst-case equipment configuration scenario was assumed, in which each transmitter received the maximum power allowable by law (500 watts) or by engineering specifications, power densities at the critical recep-

tor were found to range from 0.00001 to 0.001 mW/cm2. Thus, under maximum power to each transmitter, the allowable power density still ranged from 600 to 60,000 times that expected at the location of the critical receptor. In practice, transmitters often are clustered (“co-located”) on a common structure, such as a pole or tower. To consider the possible impact of this practice, all 15 transmitters were assumed to be operating from a single tower (notwithstanding that this might be physically impossible). Further, a second set of the same 15 transmitters was assumed to be co-located on the same tower, but operating under the worst-case (maximum-power) scenario. This effectively produced a scenario in which 30 transmitters were operating simultaneously. The resulting power density at the critical

receptor was 0.009 mW/cm2, which is less than 2 percent of the FCC-allowable power density.

Pole-Mounted Microwave Transmitters Numerous pole-mounted radio microwave transmitters likewise were considered to exemplify equipment that might be deployed in configuring a statewide emergency communication network. Calculations based upon

these specific transmitters revealed power densities at the critical receptor ranging from 0.0000004 to 0.0002 mW/cm2 under a realistic configuration scenario that assumes a 60-foot pole-mounting height and realistic power flows to the transmitters. The resulting power density range compares favorably with an FCC-allowable power density of 1.0 mW/cm2. Thus, the FCC-allowable power density ranges from 5,000 to 2,500,000 times that expected at the location of the critical receptor. When a worst-case equipment configuration scenario was assumed, in which each transmitter received the maximum power allowable by law (500 watts) or by engineering specifications, power densities at the critical receptor were found to range from 0.000006 to 0.02 mW/cm2. Thus, under maximum power to each transmitter, the allowable power density still ranged

from 50 to more than 150,000 times that expected at the location of the critical receptor. In practice, transmitters often are clustered (“co-located”) on a common structure, such as a pole or tower. To consider the possible impact of this practice, all nine transmitters were assumed to be operating from a single

TELECOMMUNICATIONS, ELECTROMAGNETIC FIELDS, AND HUMAN HEALTH 105

pole (notwithstanding that this might be physically impossible). Further, a second set of the same nine transmitters was assumed to be co-located on the same pole, but operating under the worst-case (maximum-power) scenario. This effectively produced a scenario in which 18 transmitters were operating simultaneously. The resulting power density at the critical receptor was 0.03 mW/cm2, which is under three percent of the FCC-

allowable power density.

Tower-Mounted Microwave Transmitters Numerous tower-mounted LMR microwave transmitters were likewise considered to exemplify equipment that might be deployed in configuring a statewide emergency communication network. Calculations based upon these specific transmitters revealed power densities at the critical

receptor ranging from less than 0.0000001 to 0.00002 mW/cm2 under a realistic configuration scenario that assumes a 180-foot tower-mounting height and realistic power flows to the transmitters. The resulting power density range compares favorably with an FCC-allowable power density of 1.0 mW/cm2. Thus, the FCC-allowable power density ranges from 50,000 to 10,000,000 times that expected at the location of the critical receptor. When a worst-case equipment configuration scenario was assumed, in which each transmitter received the maximum power allowable by law (500 watts) or by engineering specifications, power densities at the critical receptor were found to range from 0.0000006 to 0.001 mW/cm2. Thus, under

maximum power to each transmitter, the allowable power density still ranged from 1,000 to more than 1,500,000 times that expected at the location of the critical receptor. In practice, transmitters often are clustered (“co-located”) on a common structure, such as a pole or tower. To consider the possible impact of this practice, all nine transmitters were assumed to be operating from a single tower. Further, a second set of the same nine transmitters was assumed to be co-located on the same tower (notwithstanding that this might be physically impossible), but operating under the worst-case (maximum-power) scenario. This effectively produced a scenario in which 18 transmitters were operating simultaneously. The resulting power density at the critical

receptor was 0.003 mW/cm2, which is under one percent of the FCC-allowable power density.

Hazard Assessment

The hazard assessment component of the overall analysis was conducted based upon the worst-case assumption that FCC standards might be

106 R. A. MICHAELS

outdated, or otherwise might fail to protect the critical receptor adequately. The possibility that standards may be inadequately protective occasionally arises when emerging research reveals adverse health effects that were not accounted for in derivation of standards promulgated that predate publication of the research. To accomplish this analysis, recent reports in the peer-reviewed literature, including reports of adverse EMF effects on ani-

mals, were examined and evaluated critically. Data relating health effects to LMR-specific frequencies (say, in the 700to 900-megahertz range) were unavailable. The issue of risk acceptability of tower-mounted or pole-mounted radio or microwave transmitters to be utilized in configuring a statewide emergency communications network instead was elucidated via a broader database. That is, data pertaining more broadly to the wider non-ionizing portion of the electromagnetic spectrum, including LMR and microwave components, were utilized. Specific findings are documented in the subsections that follow.

Potential Non-cancer Effects Tissue Heating by EMFs Microwaves can heat tissue, as in the case of microwave cooking

(Goldsmith 1997). This effect, though significant in the case of microwave ovens, is vanishingly small in the case of wireless telecommunications (Alpert 2000). For example, the maximum temperature increase that a cell phone placed directly against a user’s head can cause has been reported to (cid:5) be approximately 0.1 C, whereas brain temperature normally fluctuates by more than 10 times this amount. The heating effects of electromagnetic radiation will receive no further attention herein.

Ambient (Background) EMFs The ability of ambient EMFs to influence organisms has been known long, and long acknowledged. Earth’s dipole magnetic field has been reported to influence the behavior and orientation of numerous types of organisms (Guyton 1971). These include magnetotactic bacteria, algae, marine mollusks, honeybees, hornets, salmon, tuna, salamanders, turtles, pigeons, cetaceans and, indeed, humans (see, for example, Able and Able 1993; Baker 1980; Hsu and Li 1994; Lohmann and Willows 1987; Wiltschko et al. 1993). Such effects presumably lack clinical significance. Rather, they would appear to be adaptive effects improving the function-

ing of various organisms in their environment. At worst, natural EMFs would appear to be a neutral feature of the environmental landscape, to which millions of years of biological evolution have adapted all presentday organisms.

TELECOMMUNICATIONS, ELECTROMAGNETIC FIELDS, AND HUMAN HEALTH 107

As early as 1980, direct evidence for a magnetic-field-based directional sense in humans was obtained by displacing blindfolded volunteers to a significant distance from their university campus, and asking them to identify the compass direction to the campus (Baker 1980). This phase of the experiment convincingly demonstrated a human directional sense, suggesting a human ability to sense and respond to Earth’s magnetic field. The

investigator then repeated the experiment, but placed a bar magnet behind the heads of randomly selected volunteers, which was held in place by the blindfold headband. This treatment extinguished the directionality sense of the volunteers, preventing them from identifying the compass direction to their campus. This finding supported and reinforced the suggestion that people can sense and respond to Earth’s magnetic field. The study described above, if correct, would firmly place humans among organisms adapted by virtue of evolution to naturally occurring EMFs. For organisms thus adapted to ubiquitous natural EMFs, adverse clinical reaction to EMFs of technological origin generally would be unexpected. Adverse response to ambient EMFs in the environment, if it occurs at all, presumably would be restricted only to such EMFs that dif-

fer significantly in strength or other attributes from the ubiquitous natural kind.

Technologically Generated EMFs Technologically generated electric and magnetic fields appear to be less able than the natural kind to interact (and interfere) with organisms. This is because technologically generated fields typically oscillate, for example, at

a frequency of 60 Hz. Such fluctuation introduces the problem that any field-induced displacement of a biological structure or molecule would be expected to be reversed by field reversal, with a net displacement of zero. Thus, one might imagine that a weak but persistent force (such as Earth’s natural magnetic field) might elicit an effect as a result of its persistence. In contrast, one might imagine that an equally strong, but rapidly oscillating, field (such as EMFs of technological origin) might fail to elicit the same effect because it might reverse itself before it has acted long enough to elicit the effect. Notwithstanding the presumed reversal of displacements expected in an oscillating, technology-generated EMF field, one also might imagine biological effects occurring that might be associated with the oscillation itself,

rather than with the net displacement, which would be zero on average. An analogy might be helpfully explanatory: if one hits a Ping-Pong ball back and forth, its “net” displacement is zero, on average: that is, its average location is right over the Ping-Pong net. Still, one might imagine the ball eventually being destroyed, and the Ping-Pong rackets worn, by the beating

108 R. A. MICHAELS

they take in volleying over a long period of time, even with no “net” displacement. This paradigm must be regarded only as a speculation though (at least conceptually) it could give plausibility to the hypothesis that biological effects, possibly including clinically significant adverse health effects, might be caused by oscillating, technology-generated EMFs. It is a specula-

tion only, because no biological process susceptible to such sensitivity to field oscillation was found in available peer-reviewed technical literature. Despite the considerations explicated above, electric and magnetic fields (in most cases, relatively strong) have been associated, both experimentally and epidemiologically, with a variety of adverse health effects. Indeed, laboratory and epidemiological investigations to elucidate possible associations of EMFs with a variety of health effects have become extensive. For example, reports of such investigations have addressed possible field strength associations with immunological effects (Lyle et al. 1983), neurological effects (Adey, Bawin, and Lawrence 1982; Bawin, Kaczmarek, and Adey 1975; Blackman et al. 1980; Dutta et al. 1984; Graham et al. 2000;

Lai, Horita, and Guy 1994), adverse reproductive outcomes (Meyer, Aldrich, and Easterly 1989; Shaw and Croen 1993), adverse developmental effects (Berman et al. 1990; Hsu and Li 1994; Nelson et al. 1994; Nelson, Snyder, and Shaw 2001), cancer (Chen et al. 2000; Coleman et al. 1989; Dolk et al. 1997a, b; Feychting and Ahlbom 1993; 1995; Fulton et al. 1980; Goldsmith 1997; Hocking et al. 1996; Kaune et al. 1987; London et al. 1991; Myers et al. 1990; Repacholi 1997; Repacholi et al. 1997; Salford et al. 1993; Savitz et al. 1988; Severson et al. 1988; Tomenius 1986; Wertheimer and Leeper 1979, 1982, 1987; Youngston et al. 1991), and cancer promotion (Adey 1990; Anderson 2000; Balcer-Kubiczek and Harrison 1991; Beniashvili, Bilanishvili, and Menabde 1991; Cleary, Liu, and Merchant € 1990; Loscher et al. 1993; McLean 1993; Mevissen et al. 1993; Stuchly et al. 1992). Attempts have been made to undermine the biological plausibility (and hence the causality) of such associations, especially in the case of relatively weak EMFs (see, e.g., Adair 1991; Bennett 1994). Indeed, relatively weak EMFs prevail, even in the vicinity of potentially strong EMF sources, such as power lines. This is because field strength typically tends to diminish rapidly with distance from the source (diminishing with the square or even the cube of the distance), and few people have high-tension power lines running through their homes. In response to the above, mechanisms also have been proposed to cir-

cumvent such objections (see, e.g., Blank 1995; Liboff 1985; Smith 1996; Weaver and Astumian 1990). One study, in which human cancer cells were reported to be more susceptible than normal cells to ultrasonic radiation (sound waves; Lejbkowicz and Salzberg 1997), might seem to suggest that

TELECOMMUNICATIONS, ELECTROMAGNETIC FIELDS, AND HUMAN HEALTH 109

such attempts to circumvent objections have succeeded. The possible applicability of this study to EMFs is via analogy only, however, as sound waves do not constitute a form of EMF. The clearest and most succinct postulation of the EMF/health effect link (Smith 1996) likewise acknowledges its hypothetical status:

Although numerous research groups have documented subtle changes in plasma membrane properties, intracellular calcium levels, and the synthesis of DNA, RNA and proteins in cells exposed to EMFs, it has not been possible to link together these many observations into a single unified hypothesis that explains how cells detect and respond to EMFs, and how such responses translate into possible health risks for exposed individuals. In a recent flurry of reports, Reba Goodman, Martin Blank, and colleagues at Columbia University in New York have furthered our understanding of this complex subject by demonstrating that cells exposed to EMFs exhibit a heat shock protein response typical of cells exposed to other environmental insults" (Smith 1996, page 23). Blank (1995) suggested that electric and magnetic fields associated with some sources of technology-generated EMFs oscillate at a frequency (50–60 Hz) close to the frequency of some biological processes with which, he asserts, they might interfere. He cites physiological processes involving ionic currents, as well as redox reactions involving electron transfer between molecules. To exemplify the former, Blank cites muscle and nerve action potentials lasting one to 10 milliseconds or assuming rapid firing, 100–1,000 Hz. To exemplify the latter he cites turnover numbers of þ enzymes in the range of 50 Hz. He reports that EMFs influence the Na , K þ -ATPase cell membrane enzyme system. However, he reports, “electric and magnetic fields have opposite effects on enzyme activity” (page 177). The influence of electric and magnetic fields, notwithstanding their tendency to interfere rather than enhance one another, actually might underlie associations of low-frequency EMFs with biological effects, such as inducing lateral concentration gradients in lipid monolayers (Lee et al 1994), changes in biosynthesis, G-protein linked membrane receptors, and melatonin-linked processes Blank (1995).2 Blank likened EMF-associated changes in protein synthesis to changes in protein synthesis known to occur in "the cellular stress called heat stress" (page 176). He concluded that "the evidence is strong that weak magnetic fields stimulate the stress response pathway in cells" (pages 176–177), but also concluded that, nonetheless:

2 G-protein: any of a class of cell membrane proteins that transmit signals from certain receptors to effector enzymes, enabling the cell to regulate its metabolism in response to hormones and other extracellular stimuli (American Heritage Medical Dictionary, 2007); Melatonin: the mammalian pineal gland hormone N-acetyl-5methoxytryptamine; formed by the pineal gland, it appears to depress mammalian gonadal function; in amphibians, it causes melanopore contraction (Stedman’s Medical Dictionary, 24th Edition, 1982).

110 R. A. MICHAELS

we do not know if there is a health risk associated with exposure … The research findings justify caution regarding exposure to EMFs, but no clear links to disease (Blank 1995, page 177; emphasis added). The association between low-energy (that is, low-frequency) EMFs and health effects proposed as described above is problematic. Although the energy level of the EMF source is given in terms of the frequency, import-

ant information is unreported, specifically, the resulting field strength (except in one example given by Blank 1995, involving leukemia risk vs. magnetic field strength). The leukemia studies are discussed elsewhere in this report. One might think of the source frequency as a measure of the “color” of the source, whereas the field strength might be visualized as the “brightness” or intensity. To visualize the importance of intensity, consider that light in the ultraviolet range of wavelengths can burn people’s skin, but more so on days when the atmosphere is clear and, therefore, the sun is strong. In the absence of field strength data, the possible relevance of the Smith (1996) and Blank (1995) reports to wireless communications networks is indeterminate. The reports (considered above) are silent on the issue of

whether and to what degree field strengths might have been in the same or a different frequency range. An a priori assumption is reasonable: that if field strengths at which biological effects were reported had been “newsworthy” (as in the case of the leukemia data), they would have been reported (as in the case of the leukemia data), and emphasized (as in the case of the leukemia data). The leukemia data are discussed below in this report, in the subsection titled “Associations of power line EMFs with cancer.” They were inconclusive.

Adverse EMF Effects on Farm Animals Potential adverse effects on humans may be suggested by evidence from farm animals. Adverse EMF effects on farm animals and production by dairy and beef cattle have been reported. Effects studied have included reproductive effects on cows (Algers and Hennichs 1986; Algers and Hultgren 1986, 1987; Amstutz and Miller 1980; Martin et al. 1986), genotoxic effects on cows (albeit at unreported, possibly high field strengths in Latvia; Balode 1996), effects on milk production (Blank 1995; Burchard et al. 1996), cow milk composition (Amstutz and Miller 1980; Blank 1995), and effects on sheep (Lee, Stormshak, and Thompson 1997). Many of the

studies involved electricity transmission lines (Algers and Hennichs 1986; Algers and Hultgren 1986, 1987; Amstutz and Miller 1980; Angell et al. 1990; Lee, Stormshak, and Thompson 1997; Martin et al. 1986). Many if not most studies yielded negative results, such as failure to affect beef cattle production adversely (Angell et al. 1990) and failure to adversely affect

TELECOMMUNICATIONS, ELECTROMAGNETIC FIELDS, AND HUMAN HEALTH 111

milk production or reproductive capacity in cows near a high-voltage direct-current power line (Martin et al. 1986). Also in cows, long-term exposure (4 months) to a 400-kilovolt, 50-Hz transmission line generating electric fields measured at 4 kV/M and magnetic fields measured at 2 mT, failed to affect reproductive success parameters (Algers and Hennichs 1986; Algers and Hultgren 1986, 1987; Amstutz and Miller 1980). Unaffected

parameters included length and regularity of the estrous cycle, mid-cycle plasma progesterone level, intensity of estrous, number of inseminations per pregnancy, proportion of cows conceiving, and proportion of fetuses showing heart activity shortly after slaughter of the dam. € € Summarizing these and other studies, Loscher and Kas (1998) reported that, even in the vicinity of such high-power sources: Examinations of dairy cows that had been exposed to magnetic fields resulted in inconsistent findings which ranged from no influence at all to a reduction in milk yield, changed milk composition and fertility problems. € € Of greater possible relevance, Loscher and Kas (1998) summarized evidence of behavioral abnormalities observed among cows in a dairy herd situated near a television and radio transmitting antenna. The authors were ambiguous, however, about where (and whether) this study was published € (and therefore, whether it was peer reviewed). According to Loscher and € Kas, transmitting antennas in close proximity to the farm (all installed on the TV tower) included "39 directional radio transmitters" with a mean power of 124 watts per transmitter. The mounting heights of the antennas, however, were unreported. As with the studies involving high-tension power lines, according to € € Loscher and Kas, no consistent pattern of response was discernible among "farm animals who are exposed … " because: … every cow reacts to the electromagnetic tension in an individual manner. Likewise, as with the studies involving high-tension power lines, € € according to Loscher and Kas, power levels evidently were extraordinarily high for a telecommunications tower. With 39 radio transmitters and multiple television transmitters, the total power to the antennas amounted to approximately 5,000 watts. That power is roughly 10 times the worst-case, maximum-power scenario for the hypothetical wireless network considered here. It is approximately 100 times the worst-case, actual-power scenario for transmitters in the network considered. Thus, the relevance of the 5,000-watt aggregate transmitting power figure must be deemed questionable. Adverse effects reported in the study evidently occurred at power densities at least an order of magnitude and possibly two orders of magnitude (10–100 times) higher than those that reason- € ably would be expected from the hypothetical network. The Loscher and

112 R. A. MICHAELS

€ Kas report is of questionable import also for another reason, as described below. € € In the study of EMF-exposed dairy cows described by Loscher and Kas, power densities were measured at various locations. Power densities were reported to range from 0.002 to 0.08 mW/cm2 (microwatts per square centimeter) in the stable, 0.004 to 0.46 mW/cm2 alongside the stable, and 0.06 to 0.7 mW/cm2 in the surrounding area. These are very low power densities. The interpretation of these low power densities relative to causation of adverse effects on dairy cows, however, is problematic. It is problematic, in part because numerous, relatively high-powered transmitters (albeit of unreported mounting heights) appear to have generated anomalously low power densities at ground (cow) level, with which adverse effects in the dairy herd were associated by the authors. € € Closer examination of the on-line report by Loscher and Kas (1998) reveals the following startling statement about the measurement of power density: the measurements were predominantly conducted at one frequency (512 MHz). This statement is ambiguous. Yet, it most reasonably can be construed to mean that the power density contributions generated by the preponderance of transmitters at the site simply were not counted in estimating the EMF exposure of dairy cows. These predominant power density contributions seem to have been uncounted because the 39 directional radio transmitters were broadcasting in a frequency band of 2,200–18,700 MHz. That frequency band is nowhere near the narrow frequency band around 512 MHz that was used to quantify power densities to which the dairy herd was exposed. Perhaps the authors’ justification for the omitting the preponderance of

transmitters in quantifying the power density to which cows were exposed is embodied in their description of the radio transmitters as "directional," that is, not pointed toward the cows. If so, however, that would constitute weak justification. Similar antennas typically are pointed in the horizontal direction, yet compliance with FCC power density limits (US FCC 1997a, 1997b) still must be assessed via inclusion of downwardly directed sidelobes in the transmission pattern. Presumably this also is the case in Germany, the authors’ (and cows’) home country. The authors’ reason for the apparent omission can be only a matter of speculation. A more relevant conclusion is not a matter of speculation: that

the report as written appears to associate adverse effects on cows with exposure to very much lower power densities than the cows actually must have experienced. This conclusion also casts doubt on the credibility of the entire report, beginning with the authors’ interpretation of what constitutes

TELECOMMUNICATIONS, ELECTROMAGNETIC FIELDS, AND HUMAN HEALTH 113

“abnormal behavior” in a cow, and including the power levels with which such “abnormal behaviors” were associated. A further indication that relatively high EMF power levels have been required to elicit biological effects comes from a study by Burchard et al. (1996; see also Blank 1995). They confined 16 cows to wooden metabolic

cages, and exposed them to a vertical electric field of 10 kilovolts per meter (kV/M) and a uniform horizontal magnetic field of 30 microteslas (mT). Only modest EMF effects were noted in milk production (no change in “uncorrected milk yield”) and composition ("no change in milk composition, except the percentage of milk fat"). The exposure level required to produce such modest changes were about 300 times the background magnetic field, which would be approximately one microtesla.3 Burchard et al (1996) concluded the following:

No indications of health hazards to dairy cattle were found by the exposure to 10kV/M and 30-mT EMF (page 1553). As a final example, Marks, Ratke, and English (1995) reported on the association of “stray voltage” with developmental, reproductive and other toxicology problems in dogs, cats, and cows. The authors viewed stray volt-

age as a likely unconsidered variable that might explain variability and inconsistency in studies of EMF effects on health. They criticized the preponderant literature focus upon EMF effects on health: while ignoring ground currents. We feel that this is a major error, which could, at least partially, explain the inconsistent results … especially in cancer studies … (page 168).

This is a startling statement, because the ground currents to which the author refers may be quite powerful. In one case: Farmers with stray voltage problems have reported to us that they and members of their families have experienced severe physical trauma on their farms, including the near electrocution of a son while taking a bath … (Marks, Ratke, and English 1995, page 167).

Field Measurement and Data Quality Issues EMFs have been quantified narrowly, and have produced inconsistent results in studies of their possible association with adverse health effects. As discussed above, for example, stray voltage has not been quantified at all as a possible dose metric, yet it might explain or help toward explaining EMF associations with adverse health effects, and the variability and

3 Magnetic field strength (’flux density’) units include gauss (G) and tesla (T), where gauss is the centimetergram-second (CGS) metric system unit of magnetic flux density and the following relationships apply: 1 G ¼ 1,000 mG ¼ 10(cid:1)4 T ¼ 100 mT; 10(cid:1)6 T ¼ 1 mT; background (see 23) ¼ 1 mG ¼ 10(cid:1)7 T ¼ 0.1 mT.

114 R. A. MICHAELS

inconsistency of such associations. Considering the potential significance for human health; Marks, Ratke, and English (1995) reported as follows:

Attempts to associate illnesses such as cancer … and reproductive and developmental toxicity … with electric and magnetic fields have failed to show consistent results. Although leukemia, especially acute myeloid leukemia, and brain tumors have been implicated in some studies that included electrical workers and/or proximity to sources of electricity … , clearly such studies are hampered by problems in measuring exposure, the ubiquity of exposure in the community, and a lack of understanding of other variables that probably are involved. These problems could be overcome to at least some extent by utilizing dairy farmers living on sites shown to be experiencing consistent exposure to stray voltage. This would require that one also accept the premise that EMFs are not the only electrical components that have to be included as variables (Marks, Ratke, and English 1995, page 170; emphasis added). Epidemiological studies typically are incapable of quantifying exposure levels reliably. Epidemiological studies of EMF effects are no exception, though efforts have been exerted to infer lifetime field exposure levels from wire codes and point field strength measurements (see, e.g. Kaune 1993; Miller et al. 1995; Sagan 1992; Savitz 1993, 1995; Siemiatycki, 1993; Stevens 1993; Wertheimer and Leeper 1979; US EPA 1992). Further, the adequacy of the field strength dose metric itself also has been questioned (see, e.g., Adams et al. 1995; Blackman et al. 1989; Holmberg 1995; Kaune 1993;

Miller et al. 1995; Sagan 1992; Savitz 1993, 1995; Severson et al. 1988; Shaw and Croen 1993; Siemiatycki 1993; Stevens 1993; Wertheimer and Leeper 1987; US EPA 1992). Although a theoretical minimum effective field strength may exist (Adair 1991; Adams et al. 1995; Bennett 1994; Holmberg 1995; Kaune 1993; Weaver and Astumian 1990), alternative dose metrics also may require elucidation, as discussed below. Field strength windows might be useful as a dose metric for characterizing EMF effects. To understand this concept, consider automobile speed windows that are known to apply in the case of damage to wheel bearings.

The vibrations caused by “bad bearings” often begins when a car attains a particular wheel speed, such as 25 mph, and then ceases when the wheel speed increases beyond a higher speed, such as 35 mph, leaving a narrow window of effect. This “window” paradigm, however, must be regarded as speculation only. The existence of an analogy in the context of automotive mechanics does not elevate the hypothesis to a significant toxicological finding. In the case of EMF, no analogous biological process contravening the well-established dose-response paradigm seems readily evident. Proposed alternatives to the field strength dose metric are numerous.

They have included average field strength, cumulative field exposure, threshold effects, switching transients, field strength windows, frequency windows, and the presence of other agents, such as direct current fields

TELECOMMUNICATIONS, ELECTROMAGNETIC FIELDS, AND HUMAN HEALTH 115

(Adams et al. 1995). This Hazard Assessment, however, must be constrained by the extent of scientific knowledge. Given the imprecision of epidemiological studies in estimating exposure, derivation of safe exposure levels might proceed from laboratory bioassay studies. This strategy is likewise problematic, however, for two reasons. First laboratory studies must select the appropriate dose metric for quanti-

fying and predicting EMF effects. Second, laboratory studies must employ relatively high exposure levels to observe effects in a small bioassay population. As noted by Wertheimer and Leeper (1987): The essential question is whether the epidemiologic association seen between AMF (alternating magnetic field) exposure and cancer is a direct one, or whether it is an artifact of some unassessed third variable. One would like to see this question answered by experiment, but the investigator is hampered because so little is yet known about electric- and magnetic-field dosimetry. Experiments have shown that exposure to fields or currents at low intensity sometimes produce effects that disappear as the intensity is increased. This finding of intensity ‘windows’ must cause us to question the traditional experimental assumption that, by applying large doses to a small experimental population, we can detect any effects that might be produced by low doses applied to a large population (Wertheimer and Leeper 1987, page 50).

Potential Cancer Effects Associations of Power Line EMFs with Cancer As early as 1979, epidemiological associations between cancer incidence and EMFs arising from high-current electric power lines running adjacent to residences had been educed (Wertheimer and Leeper 1979, 1982, 1987, 1994) and disputed (e.g., Trichopoulos 1994). According to Wertheimer

and Leeper (1987), such power lines can expose residents “to fairly continuous 60-Hz alternating magnetic fields ranging from about 0.001 to 0.01 gauss” (1–10 milligauss, mG). More recent studies have linked elevations in cancer incidence to the bottom of this exposure range. For example, largescale studies by Feychting and Ahlbom (1993, 1995) examined cancer incidence, by type, among over 400,000 people living within 300 meters of high-voltage power lines between 1960 and 1985. The group included 142 children with cancer, and 548 adults with leukemia or brain tumors. Exposure records of these case populations were compared with exposure records of individuals not contracting either of these cancers as a control population. Children experiencing chronic exposure to 1-mG incremental fields

(background being about 1 mG) were found to be twice as likely as children in the control group to contract cancer. Exposure to 3 mG (about three times background) was found to be associated with cancer incidence approximately four times that of controls. Adults exposed to relatively

116 R. A. MICHAELS

strong fields were found to be 1.7 times more likely to contract leukemia, but not more likely to contract brain cancer, than controls. Information such as the above might suggest that a safe level of chronic exposure to magnetic fields might be set significantly below 1 mG. This conclusion, however, is untenable and premature, in part because house-

hold appliances routinely and chronically expose people to fields stronger than 1 mG. Magnetic field strength one foot from a digital clock, for example, might be approximately 1 mG. No one seems to have figured out why low field strengths generated by power lines might be associated with any form(s) of cancer, whereas the same, more ubiquitous field strengths generated by common household appliances apparently are not. A metaanalysis of multiple studies addressing the issue of EMFs vs. childhood cancers (Meinert and Michaelis 1996) concluded the following:

From the studies published so far, there are several hints of an association between EMF and childhood cancer … The most noteworthy finding … is the increase in ORs [odds ratios] … for all cancers combined … As regards the specific diagnostic subgroups, results are not so clear. Although almost all estimated ORs are above the null value, the evidence of an association remains doubtful, since data are sparse, and dose-response relationships are not always recognizable (Meinert and Michaelis 1996, page 17). The problem of background EMFs being higher than power line fields associated with elevated cancer incidence significantly undermines the possible causality of the EMF/cancer association described above. Two factors have been posited to preserve the causation hypothesis. First, background cancer incidence is relatively high, in the range of 250,000 per million as a lifetime risk so that, perhaps, attribution of a small portion of this background cancer to EMFs cannot be precluded completely. Second, power line EMFs might differ in some unrecognized way(s) from background

fields. Nonetheless, the predominance of background exposure to EMFs, and the apparent absence of affirmative evidence of differences between power line fields vs. predominant background fields, together render the causality hypothesis questionable. A further difficulty with epidemiological associations of health effects with power lines has been reported. Specifically, possibly confounding use of power lines for data transmission has been recognized in recent years (Vignati and Giuliani 1997). RF magnetic fields have been detected emanating from power lines used to convey information. Frequencies in the range of 30–450 kHz were reported, arising from “conveyed waves” produced

when a RF generator feeds a transmission line out of tune, with a power of approximately 10 watts per channel. Accordingly, the authors suggested “that the presence of radio frequency magnetic fields must be considered in the context of epidemiologic studies” of power lines (Vignati and Giuliani

TELECOMMUNICATIONS, ELECTROMAGNETIC FIELDS, AND HUMAN HEALTH 117

1997, p. 1569). This possible complication increases the already-great distance between the hypothesis vs. the demonstration that ambient EMFs might be associated with clinically significant adverse effects on animals and/or humans.

The Melatonin Hypothesis The melatonin hypothesis most essentially suggests that EMF might suppress the pineal gland hormone melatonin and, in turn, melatonin reduction might facilitate the onset, spread, and/or growth of breast cancer. Its status as a “hypothesis” relegates it, a priori, to less credibility than a principle, rule, or theory. Indeed, in science, a hypothesis is most essentially an unproved concept or speculation that someone has put forward for further

consideration. This is illustrated with respect to the melatonin hypothesis in the following paragraphs. The melatonin hypothesis has received a significant amount of further consideration, as exemplified Stevens, Wilson, and Andersen (1997). In it, € Loscher and Mevissen (1997) describe direct evidence for the hypothesis. They show that rats exposed to magnetic fields of strength 0.3 to 1 microtesla, and to 10 microtesla, exhibited reduced melatonin in blood serum samples drawn at night. Whether this or a similar effect occurs in humans cannot be inferred from the rat data, in part because a report by Lee, Stormshak, and Thompson (1997) involving sheep reported that "we saw no effects of EMF on serum melatonin concentrations or on patterns of secretion in our studies … " (page 420). € In the rat study, Loscher and Mevisson went on to determine whether reduced nocturnal serum melatonin would be associated with increased cancer incidence. They did this by exposing the rats to a chemical carcinogen [DMBA, or 7, 12-dimethylbenz(a)anthracene]. They documented EMF “co-promotion” of DMBA-induced carcinogenesis. They acknowledge, however, that: the flux densities at which we found significant tumor co-promoting effects (50 and 100 microtesla) of magnetic field exposure greatly exceed the range of flux densities occurring in human occupational or residential exposures (L€oscher and Mevissen 1997, page 575). In a second report appearing in "The Melatonin Hypothesis" (Stevens, € Wilson, and Andersen (1997), Liburdy and Loscher (1997) reviewed laboratory studies aimed at elucidating any possible association of extremely low frequency (50/60 Hz) magnetic fields with cancer. Their review is lengthy (over 80 pages). Without going into details, they were unable to associate low-power fields with cancer, instead concluding as follows:

118 R. A. MICHAELS

In view of the high tumor incidence found in some animal studies on life-span magnetic field exposure … , the possibility of tumor initiating or co-initiating effects at high flux densities cannot be ruled out. Although most of the available evidence does not suggest that magnetic fields cause genotoxic effects … , nongenotoxic carcinogens do exist … ; magnetic fields may belong to this category of carcinogens (Liburdy and Lo€scher 1997, page 646).

NIEHS The National Institute for Environmental Health Sciences (NIEHS), in the U. S. Department of Health and Human Services, National Institutes of Health (NIH), conducted one of the earlier evaluations of evidence of possible adverse health effects exerted by extremely low frequency (50–60 Hz) EMFs associated with power lines (NIEHS 1999). NIEHS found the evidence for ELF EMF effects not to rise to the level of regulatory concern, specifically:

The scientific evidence suggesting that ELF-EMF exposures pose any health risk is weak (NIEHS 1999, page ii) …

The NIEHS concludes that ELF-EMF exposure cannot be recognized as entirely safe because of weak scientific evidence that exposure may pose a leukemia hazard. In our opinion, this finding is insufficient to warrant aggressive regulatory concern … The NIEHS does not believe that other cancers or non-cancer health outcomes provide sufficient evidence of a risk to currently warrant concern (page iii).

Associations of Broadcasting EMFs with Cancer EMFs arising from television, radio, and cell tower broadcast transmitters have been associated with elevated incidence of cancer in animal and/or epidemiological studies (reviewed, e.g., in Goldsmith 1997; Repacholi 1997).

Results of studies involving animals in bioassays have been inconsistent (Alpert 2000). In Australia, Repacholi et al. (1997; see also Alpert 2000; Goldsmith 1997; Repacholi 1997) exposed Em-Pim1 transgenic mice to pulsed 900-megahertz EMFs. The mice, genetically engineered to be highly susceptible to lymphoma (a type of cancer), were exposed for 1 hour/day for 18 months to low-power electromagnetic waves similar to cell phone emissions. Mice treated in this manner exhibited twice the lymphoma incidence as controls. In contrast, however, in California Adey (1990) reported that digital cell phone signals reduced lymphoma incidence among rats previously exposed to a chemical carcinogen. Like the laboratory evidence, epidemiological evidence associating cancer

with EMFs broadcast by television, radio, and cell tower transmitters has been equivocal (Alpert 2000; Goldsmith 1997; Repacholi 1997). Such research, on the positive side, has motivated Goldsmith (1997) to question the traditional wisdom that only ionizing radiation can cause cancer:

TELECOMMUNICATIONS, ELECTROMAGNETIC FIELDS, AND HUMAN HEALTH 119

Available data suggest that RF [radio frequency] radiation be considered a carcinogenic risk, a position already taken by an internal U. S. EPA document in 1990 when there was much less evidence of the potential harmfulness of RF radiation (Goldsmith 1997, page 1585).

This review casts some doubt on efforts to distinguish ionizing from non-ionizing radiation with respect to their health effects (Goldsmith 1997, page 1586). Diverse evidence supports Goldsmith’s conclusions, but several studies seem to have been most influential. Mutational activity of RF EMFs in incubated white blood cells was evinced from studies of Moscow embassy workers, who had been irradiated by Soviet transmitters, but quantification of the exposure regime, including intensity and duration, was unreliable. A second study involved overexposure of several air traffic controllers to microwaves. Goldsmith’s inference of an EMF/cancer association based

upon mutagenesis producing chromosome damage must be interpreted in the context of repair of the damage, with a reported half-time of about 15 weeks. Even so, chromosome damage repair is an imperfect process, with efficiency typically under 100 percent. Several studies focused directly on cancer. In one study, the incidence of leukemia and other cancers was compared in communities in Honolulu, Hawaii with vs. without radio broadcasting towers. According to Goldsmith (1997), communities with broadcasting towers exhibited higher standard incidence ratios for leukemia (SIR ¼ 2.08 vs. 0.56) and all cancers (SIR ¼ 1.88 vs. 1.07) when data were adjusted by race rather than by age. The apparent absence of an association in age-adjusted data casts doubt upon the association with race, given the possibility that demographic and cultural fac-

tors reasonably might be expected to produce potentially different age structures among races. Goldsmith (1997) also associated leukemia with low-frequency radio towers based upon a second study in Hawaii. That study, however, apparently was based upon only a small number of EMF measurements, and only along roads rather than in the vicinity of homes. The authors, acknowledging that “closeness to the low-frequency radio towers at Lualualei Naval Station may have a weak association with leukemia,” nonetheless suggested that exposure to benzene and/or other chemical leukemogens constituted the more likely cause(s) of the cluster. Finally, leukemia incidence returned to normal eventually, but no change in EMF sources or residence locations relative to them was reported. Thus, the cancer association with proximity

to radio towers in Hawaii must be deemed inconclusive at best, and more likely casual rather than causal. Similarly, another study of relative cancer and leukemia incidence in communities in North Sydney, Australia revealed higher leukemia incidence in communities near television broadcast towers (Hocking et al.

120 R. A. MICHAELS

1996). Incidence among children exhibited a relative risk (RR) of 1.58 (95% confidence interval RR ¼ 1.07 to 2.34), and among adults RR was 1.24 (95% confidence interval RR ¼ 1.09 to 1.40). According to Goldsmith, “the authors were unsuccessful in identifying confounders to explain these results” (Goldsmith 1997; page 1583). Neither the absence of confounders

nor the causality of the association, however, appears to have been established convincingly. Other studies also were cited by Goldsmith (1997l Dolk et al. 1997a, b). The cited studies are suggestive but still inconclusive with respect to the issue of whether electromagnetic radiation from broadcast facilities can cause cancer in general, or leukemia in particular. As a final issue, Goldsmith acknowledged that “most of the exposures studied are relevant to occupation” (rather than residence). In addition to the higher exposure levels typical of occupational settings, radio and television broadcast facili-

ties may operate in the power range of 50,000–100,000 watts. In contrast, cellular telecommunication facilities typically operate at less than 50–100 watts, a power range that is three orders of magnitude (1,000 times) lower. Transmitters in the hypothetical wireless network considered here typically would operate at lower power than the 50- to 100-watt range reported above. One might adopt the worst-case assumption that 50,000–100,000 watts indeed posed a real (though statistically difficult-to-discern) cancer risk. In that case, EMF exposure levels reasonably expected in connection with the hypothetical network would be three orders of magnitude (1,000

times) lower than the difficult-to-discern effect levels reported above.

Cancer vs. Cell Phone Use Concern about possible cancer causation by cell phones arose decades ago, in 2000 culminating in a lawsuit involving a plaintiff claiming that his cell phone caused his brain cancer (Parascandola 2001, 2002). In support of the

lawsuit, in 2001 Hardell et al. conducted a questionnaire study (unpublished) involving 1429 brain cancer patients and 1470 controls (Hardell et al. 2002; Parascandola 2001). The study revealed elevated brain cancer risk among subjects using analog cell phones but, enigmatically, not among users of digital cell phones. Today’s cell phones are digital. Further studies augmented the 2001 findings (e.g., Hardell et al. 2002, 2003) The association at most could be regarded as casual rather than causal, failing to satisfy all nine “Hill criteria” for demonstrating causation in epidemiology (Hill 1965). Most notably, the finding of association was unaccompanied by

a showing of biological plausibility, assuming that only ionizing radiation can break bonds in DNA molecules, and thereby cause mutations and cancer, at least via a genotoxic mechanism.

TELECOMMUNICATIONS, ELECTROMAGNETIC FIELDS, AND HUMAN HEALTH 121

The unpublished 2001 Hardell et al. evidence ultimately did not prevail, as it also had been contradicted, principally by two epidemiology studies (Inskip et al. 2001; Muscat et al. 2000). Further, the increased brain cancer incidence was enigmatic in that the duration of cell phone use by study subjects was only 1–3 years, whereas the reported brain cancers typically

would be expected to take 10–15 years to develop (Parascandola 2001). Moreover, the increased brain tumor incidence was statistically significant only if benign brain tumors (acoustic neuromas) also were included among positive carcinogenicity responses. The judge in the case finally rejected the Hardell et al. data because of doubts about their validity in the face of poor peer review ratings, and accusations of data distortion. Considering the full body of evidence, the United Nations International Agency for Research on Cancer in 2011 classified RF radiation in Group 2b, possible human carcinogen (UN IARC 2011). Similarly, under the 1986 US EPA carcinogenesis classification system, electromagnetic radiation would be placed in Category C, “possible human carcinogen” (US EPA

1986). U. S. EPA guidelines were revised in 2005 (US EPA 2005). Under the later guidelines, evidence of human carcinogenicity of EMF would be classified either as “suggestive” or “inadequate.” The weight of evidence as of 2011, described above and resolving to zero cancer risk in formal health risk assessments, eventually must be updated. This might be done by UN IARC, perhaps in 2021 after passage of a decade from the 2011 assessment. If it were done today, the update most notably would incorporate three recent bioassays involving rodents exposed to cell phone-type EMFs (Falcioni et al. 2018, US NTP 2018a, 2018b). The bioassay organisms were mice (US NTB 2018b) and rats (Falcioni et al. 2018; US NTP 2018a). In both rat studies males, but

(oddly) not females, exhibited elevated incidence of heart schwannomas (cancers of Schwann cells, which insulate nerve cells), revealing a doseresponse relationship and suggesting RF EMF causation. The mouse studies produced only tentative associations with cancer. Specifically, NTP concluded that the RF exposures may have caused tumors in the skin and lungs of male mice and malignant lymphomas in female mice … [and] tumors in the liver of male mice and malignant lymphomas in female mice (US NTP 2018b, emphasis added). In the (more interesting) NTP rat study, leaked prematurely in 2016 (Maron 2016; Schmidt 2018), radiation power levels were controlled to avoid tissue heating, while being increased gradually to maintain constant

whole-body dosage as the rats increased in body size and body weight from birth to 2 years. Treatment groups of 90 rats were exposed for life to power levels of 1.5, 3, or 6 watts per kilogram (W/kg). U. S. cell phone companies allow exposure to power levels of up to 1.6 W/kg. The study, involving

122 R. A. MICHAELS

thousands of rats, reported association of near-field RF EMF exposure, similar to human exposure via cell phone use, with elevated and doserelated incidence of heart schwannomas. In contrast, Falcioni et al. (2018) studied far-field RF EMF exposure, similar to human exposure to ambient RF fields such as those produced by tablet and laptop computers. The study

likewise revealed association with heart schwannomas in male, but not female, rats. Both studies also revealed “some evidence” of brain cancers known as gliomas (Schmidt 2018), but in the Falcioni et al. study (oddly) only in females.

Discussion

This report assesses, and draws conclusions about, potential EMF impacts, such as those arising from wireless communication networks of mobile phone and stationary emergency communication system transmitters and repeaters. Impacts assessed include four primary components:

1. quantifying the EMF power density at a ’critical receptor’, 2. evaluating compliance with Federal Communication Commission (US FCC) standards

3. assessing potential cancer risks, and 4. assessing potential non-cancer risks.

EMF power density at a “critical receptor” EMF power densities were found to depend upon the following factors:

(cid:6) mounting height: pole-mounting vs. tower-mounting, (cid:6) radiation type: radio vs. microwave transmission, (cid:6) power to transmitter: realistic scenario vs. maximum-power scen-

ario, and (cid:6) co-location: presence vs. absence of other transmitters on the same mounting structure.

In all scenarios, multiple transmitters were considered. In all scenarios, each transmitter generated EMF power densities in the range of a small fraction of a milliwatt per square centimeter at the location of a hypothetical “critical receptor” situated below the transmitter. In all scenarios, co-

location of multiple transmitters operating simultaneously and at maximum power resulted in power densities of less than 0.1 milliwatt per square centimeter at the location of a hypothetical ’critical receptor’.

TELECOMMUNICATIONS, ELECTROMAGNETIC FIELDS, AND HUMAN HEALTH 123

Compliance with Federal Standards

The issue of whether the hypothetical proposed transmitters, all transmitting simultaneously, would contravene applicable standards was addressed. To accomplish this, calculated power densities at the critical receptor were compared with US FCC standards promulgated under the governing Telecommunications Act of 1996 (US Congress 1996). This analysis revealed that, under all scenarios, total transmitter power at the critical receptor would be less than that allowable under US FCC standards, with a comfortable margin of safety between calculated and allowable power densities.

Potential Cancer Risks Peer-reviewed carcinogenicity literature was evaluated. Studies suggesting a possible association of leukemia with television or radio broadcast towers

in communities were inconclusive. Further, the transmitters in those studies were three orders of magnitude (1,000 times) more powerful than transmitters evaluated here. Early epidemiological research on EMFs has revealed casual, possibly causal association of ambient EMFs with cancer. These epidemiology studies have revealed higher exposure to analog (though not digital) cell phone EMFs among brain cancer patients than among controls. Direct statistical detection of human carcinogenicity of cell phone EMFs, however, has not been possible, suggesting that human exposure to cell phone EMFs is at most only weakly carcinogenetic. Cell phone EMFs have been termed “non-ionizing,” meaning that they are insufficiently energetic to damage (“ionize”) DNA and cause mutations. In

that case, the mechanism(s) by which RF and MW EMFs generated by cell phones might be carcinogenetic must be non-genotoxic. This means that the mechanism(s) would have a power threshold below which they would be unable to cause cancer. In contrast, genotoxic mechanisms may lack a threshold, and therefore may be associated with cancer risk at any non-zero exposure level, albeit with potency diminishing with power level (dose). Since the early epidemiology studies described above, evidence supporting the hypothesis of cancer causation by low-power RF EMFs has evolved. Recent bioassays associating cell phone-type EMFs with cancer in male rats together justify updating the 2011 UN IARC classification of RF radiation, which place RF EMFs in Group 2b, possible human carcinogen (UN IARC

2011). Revisiting the IARC classification could result in qualitative reclassification, perhaps to probable human carcinogen, and possibly also quantitative re-estimation of potency to a non-zero value. The tasks of characterizing and quantifying cell phone EMF carcinogenicity will constitute a major analytical undertaking by members of an

124 R. A. MICHAELS

international committee seeking consensus. The task of forecasting the eventual action of such a future committee therefore will remain beyond the scope of the present report. Analytic issues that will face such a committee will include determining whether the recent bioassays reported above demonstrated RF EMF causation of cancer(s) in rats, and what sig-

nificance causation in rats might have for humans. The latter issue will be influenced by the degree to which rats used in the bioassays might have been especially susceptible to contracting the cancers that were observed to be elevated (schwannomas). If people prove to be less susceptible than the rats, the potency estimate for humans would be expected to be reduced. This is an important consideration in view of the “precautionary principle,” which otherwise (in the absence of evidence to the contrary) would require assuming that people will be more susceptible than animals. This issue is especially important, given that cell phone use has become ubiquitous. The

clock cannot feasibly be turned back to pre-cell phone days. A conclusion that people are at elevated cancer risk from using (currently digital) cell phones would raise urgent issues. If it is drawn ultimately, such a conclusion would raise three general issues:

1. whether cancer risks were estimated to be sufficiently high to require mitigation, 2. Whether, in that case, risk management strategies feasibly might be adopted to reduce risks to within traditionally acceptable levels, and 3. whether cell phone technology advances might reduce power densities,

and thereby also reduce cell-phone-related cancer risks.

Potential Non-cancer Risks

Available peer-reviewed literature on non-cancer effects was evaluated. Emerging research on EMFs, including component electric fields and magnetic fields, revealed no report(s) definitively linking ambient EMFs from cell phone or at power densities expected to be generated by wireless networks of microwave or radio transmitters to causation of any adverse noncancer effect(s). This conclusion applies to realistic worst-case scenarios and also to more extreme, maximum-power scenarios including co-location scenarios in which many transmitters are assumed to operate simultan-

eously from the same mounting structure (pole or tower).

Risk Management and Risk Communication

Telecommunications equipment is ubiquitous, and exposure prolonged, raising the issue of possible health risks. Such risks, if any, cannot feasibly

TELECOMMUNICATIONS, ELECTROMAGNETIC FIELDS, AND HUMAN HEALTH 125

be eliminated, but they can be managed. For example, epidemiology studies reported higher exposure to analog cell phone EMFs among brain cancer patients than among controls, but analog phone risks have been “managed” via replacement with today’s digital phones, which have not been associated with cancer. Indeed, risk management would appear to be the best if not

the only available approach. Examples of risk management strategies include using speakers, storing cell phones off the body when possible, and using wired headsets to reduce RF exposure. Health concerns likewise have motivated public opposition to siting transmitters in communities. Credible, objective explanation of technical information to primarily non-technical audiences constitutes a necessary risk communication strategy for supporting informed public participation and dispassionate weighing of telecommunications risks and benefits in decision-making.

Conclusion Telecommunication generates EMFs at radio (RF) and microwave (MW)

frequencies. Transmitters have proliferated in communities with siting of wireless communication networks, often co-located among other transmitters. “Cell” phones also have proliferated, representing small transmitters used in contact with human heads, and stored on human bodies. Telecommunications equipment is ubiquitous, and EMF exposure prolonged, raising the issue of possible health risks. Such risks, if any, can be managed. For example, epidemiology studies reported higher exposure to analog cell phone EMFs among brain cancer patients than among controls, but those risks were “managed” via replacement of analog phones with today’s digital phones, which have not been associated with human cancer.

Challenges remain, recently from rodent bioassays that show dose-related association of lifetime exposure to cell-phone-type EMFs with heart schwannomas (cancers of Schwann cells, which insulate nerve cells) in male rats, though not in female rats. Human cancer risk, if any, remains to be characterized and quantified, which partly will depend upon whether EMFs indeed are non-ionizing as has been assumed, and whether a threshold or non-threshold (genotoxic) mechanism caused the cancers in the male rats. Health concerns have motivated further exposure reduction suggestions, and sometimes opposition to siting transmitters. Credible, objective explica-

tion of technical information to primarily non-technical audiences is necessary to support informed public participation and dispassionate weighing of telecommunications risks and benefits in community decision-making. Ultimately, experts and non-experts should adhere to the “precautionary

126 R. A. MICHAELS

principle,” requiring adoption of reasonably (but not excessively) pessimistic exposure and risk assumptions, whether or not they are likely to materialize.

ORCID Robert A. Michaels http://orcid.org/0000-0002-3557-1562

References Able, K. P., and M. A. Able. 1993. Daytime calibration of magnetic orientation in a migratory bird requires a view of skylight polarization. Nature 364(6437):523–6. doi: 10.1038/ 364523a0. Adair, R. K. 1991. Constraints on biological effects of weak extremely-low-frequency electromagnetic fields. Physical Review A 43:1,039–48. Adams, J. G., J. Zhang, M. G. Morgan, and I. Nair. 1995. A method for evaluating transmission line magnetic field mitigation strategies that incorporates biological uncertainty. Risk Analysis 15(3):313–8. doi: 10.1111/j.1539-6924.1995.tb00324.x. Adey, W. R. 1990. Joint actions of environmental non-ionizing electromagnetic fields and chemical pollution in cancer promotion. Environmental Health Perspectives 86:297–305. doi: 10.2307/3430968. Adey, W. R., S. M. Bawin, and A. F. Lawrence. 1982. Effects of weak amplitude-modulated microwave fields on calcium efflux from awake cat cerebral cortex. Bioelectromagnetics 3(3):295–307. doi: 10.1002/bem.2250030302. Algers, B., and K. Hennichs. 1986. The effects of exposure to 400-kV transmission lines on the fertility of cows: a retrospective cohort study. Preventative Veterinary Medicine 3:352–61. Algers, B., and J. Hultgren. 1986. Cows under 400-kV power lines: effects on ovulation and fertility. Svensk Veterinartigning 38:229–35. Algers, B., and J. Hultgren. 1987. Effects of long-term exposure to 400-kv, 50-Hz line on estrous and fertility in cows. Preventative Veterinary Medicine 5(1):21–7. doi: 10.1016/ 0167-5877(87)90003-1. Alpert, M. 2000. Worrying about wireless. Scientific American 283(3):20–1. American Heritage Medical Dictionary. 2007 New York City, Houghton-Mifflin; 2004. Amstutz, H. E., and D. B. Miller. 1980. A study of cattle near 765-kV transmission line. 11th International Congress on Disease in Cattle. Reports and Summaries vol. 1, pp. 609–21. Anderson, L. E. 2000. Effects of 50- or 60-hertz, 100 uT magnetic field exposure in the DMBA mammary cancer model in Sprague-Dawley rats: possible explanations for different results from two laboratories. Environmental Health Perspectives 108(9):797–802, September. doi: 10.2307/3434986. Angell, R. F., M. R. Schott, R. J. Raleigh, and T. D. Bracken. 1990. Effects of a high-voltage direct-current transmission line on beef cattle production. Bioelectromagnetics 11(4): 273–82. doi: 10.1002/bem.2250110404. Baker, R. R. 1980. Goal orientation by blindfolded humans after long-distance displacement: possible involvement of a magnetic sense. Science 210(4469):555–7. doi: 10.1126/ science.7423208. Balcer-Kubiczek, E. K., and G. H. Harrison. 1991. Neoplastic transformation of C3H/10T1/2 cells following exposure to 120-Hz modulated 2.45-GHz microwaves and phorbol ester tumor promoter. Radiation Research 126(1):65–72. doi: 10.2307/3578172.

TELECOMMUNICATIONS, ELECTROMAGNETIC FIELDS, AND HUMAN HEALTH 127

Balode, S. 1996. Assessment of radio-frequency electromagnetic radiation by the micronucleus test in bovine peripheral erythrocytes. Science of the Total Environment 180(1): 81–5. doi: 10.1016/0048-9697(95)04923-1. Bawin, S. M., L. K. Kaczmarek, and W. R. Adey. 1975. Effects of modulated VHF fields on the central nervous system. Annals of the New York Academy of Sciences 247:74–81. Beniashvili, D. S., V. G. Bilanishvili, and M. Z. Menabde. 1991. Low-frequency electromagnetic radiation enhanced the induction of rat mammary tumors by nitrosomethyl urea. Cancer Letters 61(1):75–9. doi: 10.1016/0304-3835(91)90079-W. Bennett, W. R. 1994. Cancer and power lines. Physics Today 47(4):23–9. doi: 10.1063/ 1.881417. Berman, E., L. Chacon, D. House, B. A. Koch, W. E. Koch, J. Leal, S. Løvtrup, E. Mantiply, A. H. Martin, and G. I. Martucci. 1990. The effects of a pulsed magnetic field on chick embryos. Bioelectromagnetics 11(2):169–87. Blackman, C. F., S. G. Benane, J. A. Elder, D. E. House, J. A. Lampe, and J. M. Faulk. 1980. Induction of calcium ion efflux from brain tissue by radiofrequency radiation: effect of sample number and modulation frequency on the power-density window. Bioelectromagnetics 1(1):35–43. doi: 10.1002/bem.2250010104. Blackman, C. F., L. S. Kinney, D. E. House, and W. T. Joines. 1989. Multiple power-density windows and their possible origin. Bioelectromagnetics 10(2):115–28. doi: 10.1002/ bem.2250100202. Blank, M. 1995. Biological effects of environmental magnetic fields: molecular mechanisms. Biosystems 35(2-3):175–8. doi: 10.1016/0303-2647(94)01509-6. Burchard, J. F., D. H. Nguyen, L. Richard, and E. Block. 1996. Biological effects of electric and magnetic fields on productivity of dairy cows. Journal of Dairy Science 79:1549–54. Chen, G., B. L. Upham, W. Sun, C. C. Chang, E. J. Rothwell, K. M. Chen, H. Yamasaki, and J. E. Trosko. 2000. Effect of electromagnetic field exposure on chemically induced differentiation of friend erythroleukemia cells. Environmental Health Perspectives 108(10):967–72. doi: 10.2307/3435056. Cleary, S. F., L. M. Liu, and R. E. Merchant. 1990. Glioma proliferation modulated in vitro by isothermal radiofrequency radiation exposure. Radiation Research 121(1):38–45. doi: 10.2307/3577561. Coleman, M. P., C. M. J. Bell, H.-L. Taylor, and M. Primic-Zakelj. 1989. Leukemia and residence near electricity transmission equipment: a case-control study. British Journal of Cancer 60(5):793–9. doi: 10.1038/bjc.1989.362. Dolk, H., G. Shaddick, P. Walls, C. Grundy, B. Thakrar, I. Kleinschmidt, and P. Elliott. 1997a. Cancer incidence near radio and television transmitters in Great Britain. I. Sutton Coldfield transmitter. American Journal of Epidemiology 145(1):1–9. doi: 10.1093/ oxfordjournals.aje.a009025. Dolk, H., P. Elliott, G. Shaddick, P. Walls, and B. Thakrar. 1997b. Cancer incidence near radio and television transmitters in Great Britain. American Journal of Epidemiology 145(1):10–7. doi: 10.1093/oxfordjournals.aje.a009026. Dutta, S. K., A. Subramoniam, B. Ghosh, and R. Parshad. 1984. Microwave radiationinduced calcium ion efflux from human neuroblastoma cells in culture. Bioelectromagnetics 5(1):71–8. doi: 10.1002/bem.2250050108. Falcioni, L., L. Bua, E. Tibaldi, M. Lauriola, L. De Angelis, F. Gnudi, D. Mandrioli, M. Manservigi, F. Manservisi, I. Manzoli., et al. 2018. Report of final results regarding brain and heart tumors in Sprague-Dawley rats exposed from prenatal life until natural death to mobile phone radiofrequency field representative of a 1.8 GHz GSM base station

128 R. A. MICHAELS

environmental emission. Environmental Research 165:496–503., doi: 10.1016/j.envres. 2018.01.037. Feychting, M., and M. Alhbom. 1993. Magnetic fields and cancer in people residing near Swedish high voltage power lines. American Journal of Epidemiology 138(7):467–81. doi: 10.1093/oxfordjournals.aje.a116881. Feychting, M., and A. Ahlbom. 1995. Childhood leukemia and residual exposure to weak extremely low frequency magnetic fields. Environmental Health Perspectives 103(Supplement 2):59–62. doi: 10.1289/ehp.95103s259. Fulton, J. P., S. Cobb, L. Preble, L. Leone, and E. Forman. 1980. Electric wiring configurations and childhood leukemia in Rhode Island. American Journal of Epidemiology 111(3): 292–6. doi: 10.1093/oxfordjournals.aje.a112899. Goldsmith, J. R. 1997. Evidence relevant to radar (microwave) effects. Environmental Health Perspectives 105(Supplement 6):579–88. Graham, C., M. R. Cook, A. Sastre, M. M. Gerkovich, and R. Kavet. 2000. Cardiac autonomic control mechanisms in power-frequency magnetic fields: a multistudy analysis. Environmental Health Perspectives 108(8):737–42. doi: 10.2307/3434727. Guyton, A. C. 1971. Textbook of medical physiology. 4th edition. Philadelphia, Pennsylvania; W. B. Saunders Company. Hardell, L., A. Hallquist, K. Mild, M. Carlberg, A. Påhlson, and A. Lilja. 2002. Cellular and cordless telephones and the risk for brain tumours. European Journal of Cancer Prevention 11(4):377–86. doi: 10.1097/00008469-200208000-00010. Hardell, L., K. H. Mild, and M. Carlberg. 2003. Further aspects on cellular and cordless telephones and brain tumours. International Journal of Oncology 22(2):399–407. Hill, A. B. 1965. The environment and disease: association or causation? Proceedings of the Royal Society of Medicine 58:295–300, Hocking, B., I. R. Gordon, H. L. Grain, and G. E. Hatfield. 1996. Cancer incidence and mortality and proximity to TV towers. Medical Journal of the Australia 165(11-12): 601–5. Holmberg, B. 1995. Magnetic fields and cancer: animal and cellular evidence–an overview. Environmental Health Perspectives 103(Supplement 2):63–7. doi: 10.1289/ehp.95103s263. Hsu, C.-Y., and C.-W. Li. 1994. Magnetoreception in honeybees. Science (New York, N.Y.) 265(5168):95–7. doi: 10.1126/science.265.5168.95. Inskip, P. D., et al. 2001. Brain cancer study: 782 patients, 799 controls; negative results. Not obtained yet. Kaune, W. T. 1993. Introduction to power-frequency electric and magnetic fields. Environmental Health Perspectives 101(suppl 4):73–81. doi: 10.1289/ehp.93101s473. Kaune, W. T., R. G. Stevens, N. J. Callahan, R. K. Severson, and D. B. Thomas. 1987. Residential magnetic and electric fields. Bioelectromagnetics 8(4):315–35. Lai, H., A. Horita, and A. W. Guy. 1994. Neural mechanisms involved in microwaveinduced deficit in radial-arm maze performance. Department of pharmacology and center for bioengineering. University of Washington, Seattle, Washington. Paper presented at the Bioelectromagnetics Society meeting, February 1993 (cited in GAO). Lee, K. Y. C., J. F. Klingler, and H. M. McConnell. 1994. Electric field-induced concentration gradients in lipid monolayers. Science 263(5147):655–8. doi: 10.1126/science. 8303272. Lee, J. M., F. Stormshak, and J. Thompson. 1997. Studies on melatonin, cortisol, progesterone, and interleukin-1 in sheep exposed to EMF from a 500-kV transmission line. In The melatonin Hypothesis - Breast cancer and use of electric power, ed. R. G. Stevens, B. W. Wilson, and L. E. Anderson, 391–427 Columbus, OH: Battelle Press.

TELECOMMUNICATIONS, ELECTROMAGNETIC FIELDS, AND HUMAN HEALTH 129

Lejbkowicz, F., and S. Salzberg. 1997. Distinct sensitivity of normal and malignant cells to ultrasound in vitro. Environmental Health Perspectives 105(Supplement 6):575–8. Liboff, A. R. 1985. Geomagnetic cyclotron resonance in living cells. Journal of Biological Physics 13(4):99–102. doi: 10.1007/BF01878387. Liburdy, R. P., and W. Lo€scher. 1997. Laboratory studies on extremely low frequency (50/60-Hz) magnetic fields and carcinogenesis. In: The Melatonin Hypothesis - Breast cancer and use of electric power, ed. R. G. Stevens, B. W. Wilson, and L. E. Anderson, 585–667. Columbus, Ohio; Battelle Press. Lohmann, K. J., and A. O. D. Willows. 1987. Lunar-modulated geomagnetic orientation by a marine mollusk. Science 235:331–4. doi: 10.1126/science.3798115. London, S. J., D. C. Thomas, J. D. Bowman, E. Sobel, T.-C. Cheng, and J. M. Peters. 1991. Exposure to residential electric and magnetic fields and risk of childhood leukemia. American Journal of Epidemiology 134(9):923–37. doi: 10.1093/oxfordjournals. aje.a116176. Lo€scher, W., M. Mevissen, W. Lehmacher, and A. Stamm. 1993. Tumor promotion in a breast cancer model by exposure to a weak alternating magnetic field. Cancer Letters 71(1-3):75–81. doi: 10.1016/0304-3835(93)90100-N. Lo€scher, W., and G. K€as. 1998. Conspicuous behavioural abnormalities in a dairy cow herd near a TV and radio transmitting antenna. Practical Veterinary Surgeon 79(5):437–44. http://www.teslabel.be/001/documents/Conspicuous%20behavioural%20abnormalities%20 in%20a%20dairy%20cow%20herd.pdf. Lo€scher, W., and M. Mevissen. 1997. Magnetic fields and breast cancer: experimental studies on the melatonin hypothesis. In: The melatonin Hypothesis - Breast cancer and use of electric power, ed. R. G. Stevens, B. W. Wilson, and L. E. Anderson, 555–84. Columbus, OH: Battelle Press. Lyle, D. B., P. Schechter, W. R. Adey, and R. L. Lundak. 1983. Suppression of T-lymphocyte cytotoxicity following exposure to sinusoidally amplitude-modulated fields. Bioelectromagnetics 4(3):281–92. doi: 10.1002/bem.2250040308. Marks, T. A., C. C. Ratke, and W. O. English. 1995. Stray voltage and developmental, reproductive and other toxicology problems in dogs, cats and cows: a discussion. Veterinary and Human Toxicology 37:163–72. Maron, D. F. 2016. Major cell phone radiation study reignites cancer questions. Exposure to radio-frequency radiation linked to tumor formation in rats. Scientific American, http://www.scientificamerican.com/article/major-cell-phone-radiation-study-reignites-cancerquestions/?WT.mc_id¼SA_WR_20160601, 27 May Martin, F. B., A. Bender, G. Steuernagel, R. A. Robinson, R. Revsbech, D. K. Sorensen, N. Williamson, and A. Williams. 1986. Epidemiologic study of Holstein dairy cow performance and reproduction near a high-voltage direct-current powerline. Journal of Toxicology and Environmental Health 19(3):303–12. doi: 10.1080/15287398609530930. McLean, J. R. N. 1993. The effect of magnetic fields on tumor co-promotion in SENCAR mouse skin. Paper presented at the 1993 Bioelectromagnetics Society meeting. Meinert, R., and J. Michaelis. 1996. Meta-analyses of studies on the association between electromagnetic fields and childhood cancer. Radiation and Environmental Biophysics 35(1):11–8. doi: 10.1007/BF01211237. Mevissen, M.,. A. Stamm, S. Buntenko€tter, R. Zwingelberg, U. Wahnschaffe, and W. L€oscher. 1993. Effects of magnetic fields on mammary tumor development induced by 7, 12-dimethylbenz(a)-anthracene in rats. Bioelectromagnetics 14(2):131–43.

130 R. A. MICHAELS

Meyer, R. E., T. E. Aldrich, and C. E. Easterly. 1989. Effects of noise and electromagnetic fields on reproductive outcomes. Environmental Health Perspectives 81:193–200. doi: 10.2307/3430829. Miller, M. A., J. R. Murphy, T. I. Miller, and A. J. Ruttenber. 1995. Variation in cancer risk estimates for exposure to power line frequency electromagnetic fields: a meta-analysis comparing EMF measurement methods. Risk Analysis 15(2):281–7. Muscat, J., et al. 2000. Brain cancer study: 469 patients, 422 controls; negative results. Not obtained yet. December. Myers, A., A. D. Clayden, R. A. Cartwright, and S. C. Cartwright. 1990. Childhood cancer and overhead power lines: a case-control study. British Journal of Cancer 62:1008–14. Nelson, B. K., D. L. Conover, P. B. Shaw, D. M. Werren, R. M. Edwards, and A. M. Hoberman. 1994. Interactive developmental toxicity of radiofrequency radiation and 2methoxyethanol in rats. Teratology 50(4):275–93. doi: 10.1002/tera.1420500403. Nelson, B. K., D. L. Snyder, and P. B. Shaw. 2001. Developmental toxicity interactions of methanol and radiofrequency radiation or 2-methoxyethanol in rats. International Journal of Toxicology 20(2):89–100. NIEHS 1999. Health effects from exposure to Power-Line frequency electric and magnetic fields. Research Triangle Park, North Carolina; U. S. Department of Health and Human Services (DHHS), National Institutes of Health, National Institute of Environmental Health Sciences. NIH Publication No. 99-4493. Parascandola, M. 2001. Cell phone lawsuits face a scientific test. Science 294:1440–2. Parascandola, M. 2002. Scientific evidence. Judge rejects cancer data in Maryland cell phone suit. Science (New York, N.Y.) 298(5592):338. doi: 10.1126/science.298.5592.338. Repacholi, M. H. 1997. Radiofrequency field exposure and cancer: what do laboratory studies suggest?. Environmental Health Perspectives 105(Supplement 6):565–8. Repacholi, M. H., A. Basten, V. Gebski, D. Noonan, J. Finnie, and A. W. Harris. 1997. Lymphomas in El-Pim1 transgenic mice exposed to pulsed 900 Megahertz electromagnetic fields. Radiation Research 147(5):631–40. doi: 10.2307/3579630. Sagan, L. A. 1992. Epidemiological and laboratory studies of power frequency electric and magnetic fields. JAMA: The Journal of the American Medical Association 268(5):625–9, 5 August. doi: 10.1001/jama.1992.03490050073029. Salford, L. G., A. Brun, B. R. R. Persson, and J. Eberhardt. 1993. Experimental studies of brain tumor development during exposure to continuous and pulsed 915-MHz radiofrequency radiation. Biochemistry and Bioenergetics 30:313–8. doi: 10.1016/0302-4598(93)80090-H. Savitz, D. A. 1993. Epidemiologic studies of electric and magnetic fields and cancer: strategies for extending knowledge. Environmental Health Perspectives 101(Supplement 4): 83–91. doi: 10.1289/ehp.93101s483. Savitz, D. A. 1995. Overview of occupational exposure to electric and magnetic fields and cancer: advancements in exposure assessment. Environmental Health Perspectives 103(Supplement 2):69–74. doi: 10.1289/ehp.95103s269. Savitz, D. A., H. Wachtel, F. A. Barnes, E. M. John, and J. G. Tvrdik. 1988. Case-control study of childhood cancer and exposure to 60-Hz magnetic fields. American Journal of Epidemiology 128(1):21–38. doi: 10.1093/oxfordjournals.aje.a114943. Schmidt, C. 2018. New studies link cell phone radiation with cancer. Scientific American, https://www.scientificamerican.com/article/new-studies-link-cell-phone-radiation-with-cancer/ Severson, R. K., R. G. Stevens, W. T. Kaune, D. B. Thomas, L. Heuser, S. Davis, and L. E. Sever. 1988. Acute nonlymphocytic leukemia and residential exposure to electromagnetic fields. American Journal of Epidemiology 128(1):10–20. doi: 10.1093/oxfordjournals.aje. a114932.

TELECOMMUNICATIONS, ELECTROMAGNETIC FIELDS, AND HUMAN HEALTH 131

Shaw, G. M., and L. A. Croen. 1993. Human adverse reproductive outcomes and electromagnetic field exposures: review of epidemiologic studies. Environmental Health Perspectives 101(Supplement 4):107–19. doi: 10.1289/ehp.93101s4107. Siemiatycki, J. 1993. Problems and priorities in epidemiologic research on human health effects related to wiring code and electric and magnetic fields. Environmental Health Perspectives 101(Supplement 4):135–41. doi: 10.1289/ehp.93101s4135. Smith, O. 1996. Cells, stress and EMFs. Nature Medicine 2(1):23–4. doi: 10.1038/nm0196-23. Stedman, T. L. 1982. Stedman’s medical dictionary. 24th ed. Baltimore, MD; Williams & Wilkins, Stevens, R. G. 1993. Biologically based epidemiological studies of electric power and cancer. Environmental Health Perspectives 101(Supplement 4):93–100. doi: 10.1289/ehp. 93101s493. Stevens, R. G., B. W. Wilson, and L. E. Anderson. (Eds). 1997. The melatonin hypothesis: Breast cancer and use of electric power. Columbus, OH; Battelle Press. Stuchly, M. A., J. R. N. McLean, R. Burnett, M. Goddard, D. W. Lecuyer, and R. E. J. Mitchel. 1992. Modification of tumor promotion in the mouse skin by exposure to an alternating magnetic field. Cancer Letters 65(1):1–7. doi: 10.1016/0304-3835(92)90205-A. Tomenius, L. 1986. 50-Hz electromagnetic environment and the incidence of childhood tumors in Stockholm County. Bioelectromagnetics 7(2):191–207. doi: 10.1002/ bem.2250070209. Trichopoulos, D. V. 1994. Response to Wertheimer and Leeper (103.5) re: are electric and magnetic fields affecting mortality from breast cancer in women?. Journal of the National Cancer Institute 86(23):798. UN IARC. 2011. Non-ionizing Radiation, Part 2: Radiofrequency Electromagnetic Fields. Lyons, France, United Nations International Agency for Research on Cancer, Monograph 102, https://monographs.iarc.fr/wp-content/uploads/2018/06/mono102.pdf. US Congress. 1996. Telecommunications Act of 1996. Public Law 104-104. u. p. US EPA. 1986. Guidelines for carcinogen risk assessment. Washington, DC; U.S. Environmental Protection Agency, Risk Assessment Forum, EPA/630/R-00/004, 38 pages, US EPA. 1992. Electric and magnetic fields: an EPA perspective on research needs and priorities for improving health risk assessment. Washington, DC; Office of Research and Development, EPA/600/9-91/016F, i.p., December US EPA. 2005. Guidelines for Carcinogen Risk Assessment. Washington, DC; U.S. Environmental Protection Agency, Risk Assessment Forum, https://www.epa.gov/risk/ guidelines-carcinogen-risk-assessment, 166 pages, EPA/630/P-03/001F; US FCC. 1997a. Evaluating compliance with FCC guidelines for human exposure to radiofrequency electromagnetic fields. Washington, DC. U. S. Federal Communications Commission, Office of Engineering and Technology, OET Bulletin 65 (Edition 97-01), 63 pp. plus appendices, November US FCC. 1997b. Evaluating compliance with FCC guidelines for human exposure to radiofrequency electromagnetic fields. Washington, DC. U. S. Federal Communications Commission, Office of Engineering and Technology, OET Bulletin 65 (Edition 97-01), Supplement B (Edition 97-01), 45 pp. plus appendices. US NTP. 2018a. Technical report on the toxicology and carcinogenesis studies in Hsd: Sprague Dawley Sd rats exposed to Whole-Body radio frequency radiation at a frequency (900 MHz) and modulations (GSM and CDMA) used by cell phones. Research Triangle Park, North Carolina; National Toxicology Program, National Institutes of Health, Public Health Service, U.S. Department of Health and Human Services, NTP TR 595, https://ntp.niehs.nih.gov/ntp/htdocs/lt_rpts/tr595_508.pdf, 384 pages.

132 R. A. MICHAELS

US NTP. 2018b. Technical report on the toxicology and carcinogenesis studies in B6c3f1/N mice exposed to Whole-Body radio frequency radiation at a frequency (1,900 MHz) and modulations (GSM and CDMA) used by cell phones. Research Triangle Park, North Carolina; National Toxicology Program, National Institutes of Health, Public Health Service, U.S. Department of Health and Human Services, NTP TR 596, https://ntp.niehs. nih.gov/ntp/htdocs/lt_rpts/tr596_508.pdf. Vignati, M., and L. Giuliani. 1997. Radiofrequency exposure near high-voltage lines. Environmental Health Perspectives 105(Supplement 6):569–74. Weast, R. C., and D. R. Lide. 1989. (editors). Handbook of chemistry and physics. 70th ed (1989-1990). Boca Raton, FL; CRC Press. Weaver, J. C., and R. D. Astumian. 1990. The response of living cells to very weak electric fields: the thermal noise limit. Science 247(4941):459–62. doi: 10.1126/science.2300806. Wertheimer, N., and E. Leeper. 1979. Electrical wiring configurations and childhood cancer. American Journal of Epidemiology 131:763–73. Wertheimer, N., and E. Leeper. 1982. Adult cancer related to electrical wires near the home. International Journal of Epidemiology 11(4):345–55. doi: 10.1093/ije/11.4.345. Wertheimer, N., and E. Leeper. 1987. Magnetic field exposure related to cancer subtypes. Annals of the New York Academy of Sciences 502(1 Environmental):43–54. doi: 10.1111/ j.1749-6632.1987.tb37645.x. Wertheimer, N., and E. Leeper. 1994. Re: are electric and magnetic fields affecting mortality from breast cancer in women?. Journal of the National Cancer Institute 86(23):797. Wiltschko, W., U. Munro, H. Ford, and R. Wiltschko. 1993. Red light disrupts magnetic orientation of migratory birds. Nature 364(6437):525–7. doi: 10.1038/364525a0. Youngston, J. H., A. D. Clayden, A. Myers, R. A. Cartwright. 1991. A case-control study of adult hematological malignancies in relation to overhead power lines. British Journal of Cancer 63:977–98. About the author Robert Michaels, PhD, CEP, is President of Schenectady-based RAM TRAC Corporation, and a toxicologist specializing in assessment and management of risks to public health potentially posed by environmental contaminants and other stressors. Dr. Michaels has conducted numerous risk assessments relating to wireless communication network technologies such as pole-mounted and tower-mounted transmitters for FM radio and television stations, cellular telephone communications, and communications among emergency responders. He has served many corporate clients, the U.S. Congressional Office of Technology Assessment, and public interest organizations such as the Natural Resources Defense Council (NRDC). Dr. Michaels chaired the State of Maine Scientific Advisory Panel, and for 20 years chaired the Certification Review Board of the Academy of Board Certified Environmental Professionals (ABCEP). He has been Secretary of the NFPA Committee on Classification and Properties of Hazardous Chemicals, Board Member of the National Association of Environmental Professionals, and Member of the Editorial Advisory Boards of Springer-Verlag and Cambridge University Press journals. He earned his doctorate at SUNY at Stony Brook in 1979, and in 2004 was awarded ABCEP’s Kramer Medal recognizing his professional contributions.

Publication citation

Michaels, RA. Telecommunications, electromagnetic fields, and human health. Routledge, Taylor & Francis Group, Environmental Claims Journal, 31(2):93-132, doi: 10.1080/10406026.2019.1603442, online 17 May 2019;

Read the publication

View DOI record

Robert A. Michaels on ResearchGate