The use of wireless digital technology has grown rapidly during the last couple of decades. While in use, mobile and cordless phones emit radiofrequency (RF) radiation. The brain is the main target of exposure to RF radiation with handheld wireless phones (mobile and cordless) (1,2). An increased risk for brain tumors has been of concern for a long time. In May 2011, RF radiation in the range 30–300 GHz could be categorized in Group 2B, i.e., a ‘possible’ human carcinogen, by the International Agency for Research on Cancer (IARC) of WHO (3,4). The decision was based mainly on case-control human studies on the use of wireless phones by the Hardell group in Sweden (mobile and cordless phones; DECT) and the IARC Interphone study (mobile phones), which showed an increased risk for brain and head tumours, i.e., glioma and acoustic neuroma (3–6), which has since been confirmed (7–10), resulting in a recommendation to upgrade IARC's 2011 classification of RF radiation to Group 1, a human carcinogen. This conclusion was published in our up-dated review in 2013 (11) using the so-called Hill viewpoints on the association or causation put forward at the height of the tobacco and lung cancer controversy (12).

Due to the increasing use of the wireless technology, environmental exposure to RF radiation has been increasing, but there has been no systematic study of ambient exposure. We have measured RF radiation at Stockholm Central Station (13) and the Stockholm Old Town in Sweden (14). The results generally exceeded the levels known to have adverse biological effects. By contrast, low levels were measured at certain places in the WHO building in Geneva (15).

We have measured RF radiation in an apartment with a central location at Östermalm in Stockholm. The apartment is located on the 6th floor, with a tower including a bedroom on the first floor of the tower (7th floor) and a conference room on the second and highest floor (8th) of the tower, at the same level as the roof of the building. The measurements did not involve any human subjects, and therefore no ethical permission was needed. We also discuss laboratory studies on RF-radiation and biological effects relative to the levels of RF in question. Of particular interest are the non-thermal levels of RF radiation and biological effects. For comparison, radiation was also measured in another apartment located on the 3rd floor in a high building of 9 floors at Gärdet in central part of Stockholm.

The results were based on a large sample of measurements in the apartment at Östermalm, Stockholm representing about 83 h of recording. High levels of RF radiation were clearly measured throughout the apartment, but especially on the 3 balconies. Most of the RF radiation came from the outside base stations. The total mean level in the apartment fell from 3,810.8 to 78.8 µW/m², a reduction of ~98% occurring if the down links from base stations were excluded.

Table II shows that TV and radio communications contributed only to a minor extent. The contribution was more pronounced at the balcony outside the kitchen in the direction of the transmitting tower (mean level of 276.4 µW/m² for all TV and radio communications; data not in the Table). Most of the RF radiation in the tower bedroom was also caused by radio/TV towers when down links from base stations were excluded (see Table VIII). A minor contribution was from the DECT frequency (Table II). The source of radiation was probably from a neighboring apartment, since no DECT phone was used in the present apartment.

The highest RF radiation levels were measured on the balconies (Figs. 1 and 2 give median and mean values). Fig. 4 shows the group of base stations located only 12 m from the balcony outside the living room. It is questionable whether these balconies are suitable for any longer stay due to the high radiation; the balcony outside the living room (with mean level 24,885.9 µW/m², median 22,256.1 µW/m², and maximum level 112,317.7 µW/m²) gave the highest of all recorded measurements (Table V). Regarding all 5 balconies, the results were based on 10,461 readings (>11 h of measurements). The mean level was 16,338.7 µW/m², with a median of 13,775 µW/m² (range min 31.8, max 112,318 µW/m²). Usually balconies in Sweden are used only briefly due to the climate, but nevertheless these results far exceed levels known to be detrimental to health from RF radiation.

The use of balconies with decent views can be used for pleasure, a part of comfortable living, but most of the day and night is spent in the apartment. Thus, the level of RF radiation inside is of more concern, especially for a family with children. Notably there was high RF radiation in the bedrooms occupied by children, as also in the tower bedroom. There was little variation with time, although the level declining during night (Fig. 3) still exceeded what is known to give non-thermal biological effects from long-term RF radiation. In the boy's bedroom there seemed to be 2 steps in the reduction of radiation during night, one from approximately 7:00 p.m. and another from midnight, indicating less use of wireless communication after business hours and during late leisure time.

Interestingly measurements recorded in another centrally located apartment in Gärdet were considerably lower in RF radiation. Thus, the mean level was only 0.4% of that found in the apartment at Östermalm. The mean RF radiation on the balcony in Gärdet was much lower; only 0.3% of that on the balcony outside the living room in Östermalm. These results show extreme variation of RF radiation in 2 residential apartments in Stockholm.

The guideline for RF radiation in Sweden is set on the false assumption that adverse health effects are caused only by heating. However, human, animal and cell studies show biological effects at non-thermal exposure levels that are often exceeded during the life-time of most people, especially as shown from the measurements in the Östermalm apartment during this study. Exposure to RF radiation levels in the current study may clearly increase the risk for adverse health effects in the long run, since levels giving non-thermal biological effects are generally exceeded. Our findings will now be related to other measurements of RF radiation exposure, guidelines and studies regarding health issues.

Levels of RF radiation have increased considerably in recent years, both outdoor and indoor, due to new telecommunication technologies and protocols. In 2013, Estenberg and Augustsson (16) measured outdoor exposure with a car-based measuring system in Sweden. The median power-density for RF fields between 30 MHz and 3 GHz was measured in rural areas at 16 µW/m², urban areas 270 µW/m², and city areas 2,400 µW/m². The EME-Spy 200 exposimeter was used to measure 2 areas of high exposure to RF radiation in Stockholm. At the Central Railway Station, the mean total RF radiation level varied between 2,817 and 4,891 µW/m² (min 5.8, max 155,263 µW/m²) when walking around (13). The total RF radiation in the Stockholm Old Town varied between a mean of 404 µW/m² (min 20.4, max 4,088 µW/m²) on the streets around the Supreme Court, 756 µW/m² (min 0.3, max 50,967 µW/m²) around the Royal Castle, and 24,277 µW/m² (min 257, max 173,302 µW/m²) at Järntorget, which is a popular square with shops and outdoor restaurants (14).

Calvente et al in Spain (17) measured outside the homes of 123 boys aged 10 years. For all houses, the root mean-square of power density (S_RMS) was 286 µW/m² and the maximum power density (S_RMS) was 2,760 µW/m² for frequencies between 100 kHz and 6 GHz. The range between highest and lowest measured mean was large, ranging from 5.5 to 11,560 µW/m². The 10-year old boys with higher exposure of RF radiation in their immediate surroundings of their dwellings showed statistically significant lower scores in expression and comprehension, and higher behavioral and emotional problems, including anxiety and depressed behavior in different tests.

Usually much lower RF radiation exposure inside homes is measured since the walls can act as a shield against RF radiation. Frei et al (18) got 166 volunteers measure the frequencies 88–2,500 MHz with a body-born exposimeter (EME-Spy 120) in Switzerland. In the homes, the total mean value was 100 µW/m², and total median 44 µW/m². The maximum level in the homes was 1,212 µW/m².

Roser et al (19) got 90 adolescents to carry a body-born exposimeter (Expo-RF) for 3 consecutive days to measure frequencies from 620 to 2,450 MHz. The total mean for these measurements was 63.2 µW/m², and the total mean measured in the homes was 32.1 µW/m².

Vermeeren et al (20) measured levels using exposimeters EME-Spy 121 and 140 in schools, kindergartens, offices and homes in Belgium and Greece. In homes, the total average was 0.32 V/m (272 µW/m²) in Belgium and 0.42 V/m (468 µW/m²) in Greece. The maximum levels were 0.77 V/m (1,574 µW/m²) in Belgium and 2.08 V/m (11,482 µW/m²) in Greece. In Belgium, FM-radio, GSM 900 down link from bases stations, and DECT telephones contributed most of the radiation, and in Greece DECT and Wi-Fi 2.45 GHz contributed most.

Verlock et al (21) measured schools, homes and public places in Belgium with a Narda NBM-550 in the frequency range 100 kHz-6 GHz. Total means for the measurements was 0.45 V/m (537 µW/m²), for homes 0.08 V/m (16.9 µW/m²), and maximum value in homes was 1.08 V/m (3,096 µW/m²).

These studies, published from 2009 to 2017, show a large variation in the levels of RF radiation, with highest levels measured in Stockholm.

One obstacle to those concerned with RF radiation exposure in Sweden, as in many other countries, is that different authorities base their guideline for exposure on the International Commission on Non-Ionizing Radiation Protection (ICNIRP). This guideline is based on short-term (acute) exposure. Chronic, low-intensity cumulative exposures, possible long term health effects, and non-thermal biological effects have been ignored. The ICNIRP safety limit established in 1998 (22) was updated in 2009 (23) without change. The guideline from the ICNIRP for RF radiation is 2 to 10 W/m² (2,000,000 to 10,000,000 µW/m²) depending on frequency (22). The Swedish Radiation Safety Authority (SSM) has adopted the ICNIRP guideline. Many experts at the SSM panel are also members of ICNIRP, which suggests a conflict of interest, since they would rarely compromise the ICNIRP view by expressing critical opinions (15).

Our results on RF radiation exposure in Östermalm are orders of magnitude lower than the ICNIRP guideline, with the median level of exposure being ~10,000 times lower. Using the ICNIRP guideline gives a ‘green card’ to roll out the technology, and position mobile phone base stations on roofs on apartment houses, in close to those living in the surrounding houses, since the high exposure level by ICNIRP is rarely compromised.

In contrast to ICNIRP, the BioInitiative Reports from 2007 and updated in 2012, based the evaluation on non-thermal health effects of RF radiation (24,25). A summary of the BioInitiative Report (2007) was published in a peer-reviewed article (26). Furthermore in both reports, all chapters were based on peer-review published articles, many of them by the authors of the different chapters. Thus, it is incorrect to suggest that the reports represent views and results that are not supported by the scientific literature.

The BioInitiative Report (2012) with updated references defined the scientific benchmark for possible health risks as 30–60 µW/m². Considering also chronic exposure and the sensitivity of children, the precautionary target level was proposed at 1/10th of this, i.e., 3–6 µW/m² (25). This exposure target level has not been acknowledged by SSM in Sweden, thus making it possible to neglect results on exposure, such as in the apartment we measured, and not in providing precaution for the potential detrimental effects on health for those living therein.

We used the same exposimeter in the Geneva WHO building on March 3, 2017. The results show a low mean total exposure level of 21.5 µW/m², median 13.3 µW/m² (range min 4.8, max 433 µW/m²), i.e., a mean level below the scientific benchmark of 30–60 µW/m² that has been proposed as the ‘lowest observed effect level’ (LOEL) for RF radiation, [see Chapter 24 in the BioInitiative Report (25)]. The major sources were GSM + UMTS 900 DL (3G), GSM 1800 DL (2G) and UMTS 2100 DL (3G), i.e., down link (DL) of RF radiation from base stations. Thus these results indicate that, whether known or unknown, the WHO staff seems to protect themselves from high involuntary RF radiation levels at least within the measured areas of the Geneva building (15).

In the present study, for the whole apartment including the balconies, the measurements of RF radiation had a total mean of 3,811 µW/m² and a total median of 1,313 µW/m². For rooms inside the apartment, the tower bedroom had the highest mean (5,954 µW/m²) and median (4,504 µW/m²). Also the girl's bedroom with mean 2,531 µW/m² and median 2,271 µW/m² and the boy's bedroom with a mean of 1,471 µW/m² and median 1,122 µW/m² were high values. Time spent in a bedroom is usually many hours per night, which means long term exposure when relatively high levels of RF radiation are still present.

RF radiation exposure at or below these levels indicated above have influenced several physiological parameters in the body of mammals in laboratory studies. Effects on oxidative cell stress and DNA damage in cells, opening of the blood-brain barrier, up or down regulated proteins and microRNA in the brain, and testicular dysfunction, have been found. For people living near mobile phone base stations, effects have been seen on neurotransmitters, peripheral blood lymphocytes with DNA damage, lower antioxidant levels, decreased salivary secretion, adverse neuro-behavioral symptoms, and an increased incidence of cancer. People residing near mobile phone base stations have more often complained of sleep disturbances, headaches, dizziness, irritability, concentration difficulties and hypertension. Exposures to RF radiation were all below the reference levels in the ICNIRP guidelines. The effects were caused by non-thermal RF radiation exposure and will now be briefly discussed.

In rats exposed to RF radiation, the blood brain barrier (BBB) has opened up, leading to leakage into the brain tissues of large molecules, e.g., albumin and toxins that can damage the brain tissue. The BBB is supposed to protect and prevent pathological leakage of toxins from the blood vessels (27). Several studies have shown that the BBB can open after RF radiation from a GSM mobile phone with a peak power output of only 1,000 µW, and with an average whole body specific energy absorption rate (SAR) of down to 120 µW/kg (28). Stronger effects on health due to RF radiation at lower exposure levels than at higher exposure indicate a U-shaped response curve (28,29).

Difference between sexes exposed to different RF radiation frequencies has been reported, where only female rats showed increased BBB permeability at 900 MHz frequency, whereas male rats had increased BBB permeability at both GSM 900 and 1,800 MHz pulse-modulated RF radiation (30).

Exposure to 900 MHz for 3 h per day for 28 days caused extravasation of albumin in the hippocampus and cortex, and impaired spatial memory in rats. The hippocampus, a center for co-ordination of memory and learning in the brain, seems in particular to be a primary target for neuronal damage from RF radiation and opening of the BBB (31). Exposure for 2 h each week for 55 weeks impaired the memory in rats exposed to GSM 900 MHz, but histopathological parameters did not seem to be statistically significantly affected (32,33).

Higher sensitivity to RF radiation has been reported in growing organisms. An increase in protein synthesis in proliferating human cells has been seen after 8 h of RF radiation exposure, but not in quiescent white blood cells (34). In stems cells, the capacity to repair DNA double-strand breaks was more affected by RF radiation compared to differentiated cells (fibroblasts) (35).

Mice exposed to RF radiation from a GSM 900 MHz mobile phone for 3 h/day or to DECT base station for 8 h/day over 8 months showed an up or down regulation of 143 out of 432 proteins analyzed from the cerebellum, hippocampus and frontal lobes of the brain. Several proteins involved in the brain metabolism and different neural functions were altered (36).

Two long-term studies have exposed rats for 24 h a day for 12 months to RF radiation emitted from a Wi-Fi system at 2.4 GHz. The peak power was 100,000 µW, with the antenna 50 cm above the cage. The SAR value over 10 g of brain tissue was 1,030 µW/kg. One study examined 5 different microRNAs (miRNA) in the rat brains, 2 of which decreased at least 70%. miRNA is important in the proliferation, differentiation, function and maintenance of all cells, including neurons; and dysfunction of their pathways can contribute to pathogenesis of neurodegenerative disorders, as well as being a key indicator of epigenetic changes and cancer risk (37).

Rat testes and prostate were examined in another study, where the SAR value was 1,020 W/kg over 10 gram tissue. The Wi-Fi exposed rats showed statistically significant greater effects on testicular function and histology, and increased defects in the heads of sperms (38). DNA damage in sperms has been reported in several other Wi-Fi exposure investigations (39–41).

In a review of 100 studies, Yakymenko et al (42) showed oxidative effects of low-intensity RF radiation on living cells, with exposure down to 2,500 µW/m² (43). SAR values down to 600 µW/kg increased oxidative stress in the cells (44,45). Embryos of Japanese quails were exposed to RF radiation using GSM 900 MHz (46). The average intensity of RF radiation on the surface of hatching eggs was 2,500 µW/m² (0.25 µW/cm²). SAR was calculated to be 3 µW/kg. A statistically significant overproduction of reactive oxygen species (ROS) and oxidative damage of DNA in living cells was reported compared with the control group given no exposure. The exposure was far below the guideline of the ICNIRP for RF radiation at 2 to 10 W/m² depending on frequency and 2 W/kg to the brain. The results showed that the ICNIRP guidelines are outdated. Moreover, using a safety factor of 10 would give 250 µW/m² as a guideline, a level easily exceeded in many places, e.g., our measurements taken at the Stockholm Central Railway Station (13), in the Old Town (14) and in most rooms in the present apartment at Östermalm.

Long-term RF radiation exposure for 2 h per day 5 days per week for 30–180 days at SAR values of 595–667 µW/kg and at the frequencies 900, 1,800 and 2,450 MHz resulted in oxidative stress, increase in pro-inflammatory cytokines, DNA damage with single-strand breaks, reduced levels of neurotransmitters and downregulation of mRNA in the hippocampus in the brain of rats (45,47,48), with memory and learning also being affected (47). More deleterious effects on several of the parameters were seen with an increase in frequency; 1,800 MHz and 2,450 MHz had a statistically significant effect, not only compared with sham exposed animals, but in some cases compared with 900 MHz exposure.

Even a SAR value down to 85 µW/kg exposure from 900 MHz during 2 h/day, 5 days/week for 30 days, increased oxidative stress parameters in lipid peroxidation and protein oxidation, and resulted in a statistically significant impairment in spatial memory in rats (49).

The pancreas was examined in young rats of 6 weeks of age exposed for 3 h/day for 30 days to 2.45 MHz, pulsed 217 Hz, from a RF test generator, with similar exposure to Wi-Fi 2.45 GHz. Compared to the control group, these rats had statistically significant increased levels of glucose, lipase and amylase in the blood, degenerative changes in both endocrine and exocrine cells, increased inflammatory cells and immune-positive markers, especially in the islets of Langerhans. These findings point to a deleterious effect on both endocrine and exocrine functioning of the pancreas. Langerhans islets secret hormones like insulin, glucagon and somatostatin, which regulate blood glucose levels; insufficient secretion can lead to diabetes, which have seen a big increase during the last few decades in parallel with the fast increasing use of wireless techniques (50).

The National Toxicology Program (NTP) under the National Institutes of Health (NIH) in the USA released a report in 2016 which showed an increased incidence of glioma in the brain and malignant schwannoma in the heart in up to 2 years in RF irradiated rats. These findings support human epidemiological studies on brain tumor risk and strengthen the association between RF radiation and cancer (51). Recent results from the NTP study showed genotoxicity of RF radiation in rats and mice (52). This result supports several previous findings of DNA strand-breaks in rat brain cells exposed to RF radiation, as first published by Lai and Singh (53). RF radiation leads to oxidative stress in biological systems, including the brain, due to an increase in free radicals and changes in antioxidant defence systems.

The NTP study has greatly strengthened the evidence of cancer risk, and reaffirms that there is sufficient scientific evidence to reclassify wireless phone radiation as a carcinogenic agent in Group 1, according to the IARC classification. It confirms that the current public safety limits based only on the thermal effects are inadequate and do not protect us against the associated detrimental health effects (10,11).

In the village of Rimbach in Germany, a GSM mobile base station was built in 2004. Buchner and Eger (54) studied 60 inhabitants (aged 0–69 years), measuring the neurotransmitters adrenaline, noradrenaline, dopamine and phenyletylamine (PEA) in second morning urine samples before the base station was activated, and at 6, 12 and 18 months after activation. RF radiation was measured in peak value of the power density after the activation of the base station outside each participant's houses. The participants were divided into 3 groups based on exposure: less than 60, 60–100 or >100 µW/m². The stress hormones adrenaline and noradrenaline were significantly increased over the first 6 months after the activation of the GSM base station. The levels decreased, but were not restored to initial level, after 18 months. This was especially evident in the children and the chronically ill adults. A statistically significant decrease was seen for dopamine levels over the first 6 months, after which dopamine levels increased, but did not return to their initial level. For the participants with RF radiation exposure over 100 µW/m² at home, the 3 neurotransmitters showed a clear dose-response relationship. PEA levels decreased first for the highest exposed group, but after 18 months the 3 groups were all statistically significantly decreased. DECT, Wi-Fi and other wireless devices at home seemed to amplify the effect of GSM radiation. After 18 months, even the lowest exposed group (RF radiation <60 µW/m²) had decreased dopamine and PEA levels.

PEA is often low in patients with depression, and in adults and children with attention deficit hyperactivity disorder (ADHD). Chronic dysregulation of the catecholamine system and PEA may contribute to chronic illnesses and health problems in the long term. Several of the 60 participants had new symptoms, including headache, dizziness, concentration problems, sleep disturbances and allergies (54).

Khurana et al (55) reviewed epidemiological studies on populations living near mobile phone base stations for indications of any health risks for humans. In 80% of the studies, people living <500 m from base stations had an increased prevalence in particular of adverse neuro-behavioral symptoms and cancer. In another review of 56 studies, Levitt and Lai (56) found that exposure from base stations and other antenna arrays induced changes in immunological and reproductive systems, DNA double-strand breaks, influence on calcium movement in the heart, and increased proliferation of astrocytoma cancer cells in humans and laboratory animals. Cortisol and thyroid hormones were also affected in people living near base stations (57,58).

A study with 40 healthy persons in India living <80 m from a mobile phone base station showed that RF radiation definitely induced DNA damage, a lowering of antioxidants levels, and a higher frequency of micronuclei in peripheral blood lymphocytes compared with a control group living >300 m from the base station. Those exposed were between 20 and 40 years old, 18 being male and 22 female. All but 3 used a mobile phone every day. None of the participants had occupational exposure to RF radiation and none lived near a high tension electric power line or electric transformer station (59).

Sleep disturbances, burned-out syndromes, depressions, pain problems and sick leaves are being increasingly reported in our daily papers. Scientific studies on humans, animals and biological material show adverse effects on physiological parameters due to RF radiation. The complex picture relates to wellbeing within the stress of working life and at home, and the rapid technological development leading to more sedentary behavior, with both children and adults watching the screen of a smart phone, laptop or television for many hours each day. RF radiation may contribute to deterioration in physical and mental wellbeing and health. There seems also to be a large difference in sensitivity to RF radiation between individuals, both in humans and animals (28,60). One example is the findings of downregulation of a sleep-inducing neurohormone in the age group of 18–30 years. Its level decreased with increasing number of years in use of a wireless phone (61), whereas no effect was seen among older persons (62). Since tumors take several decades to develop, and chronic illnesses, like neurological and cardiac diseases, come in older ages, we will only know in the future if and to what degree RF radiation may influence on the incidence of these illnesses and disorders.

Children are probably more sensitive to RF radiation because of their growing bodies and more immature cells, and also because they will be exposed for probably a whole lifetime in contrast to present generation (34,35,63). High mean exposure levels in the bedrooms of growing children (one at 2,531 µW/m² and the other at 1,471 µW/m²) may have deleterious effects on their physical and mental health, based on data already obtained from humans and animals. Cognitive effects have also been found in such studies (17,32,48,49), which might affect a child's future work and memory function in older age.

This study shows high RF radiation levels in an apartment with 2 groups of base stations in the near vicinity. Of special concern is the levels in bedrooms, especially those two used by children, since they seem to be more vulnerable to adverse health effects than grown-ups. They have also a longer expected life in which illnesses may later become manifest. The results indicate that this apartment is unsuitable for long-term living based on current knowledge of the potential adverse effects on health of RF radiation. For definitive conclusions regarding the effects of RF radiation levels from the base stations, one option would be to turn off all nearby base stations and make new measurements; some exposure from more distant located base stations cannot be excluded. However, the present results show that the safest solution is to turn off permanently and dismantle the base stations.