The results of a study on public exposure to radiofrequency (RF) radiation at the Stockholm Central Railway Station in Sweden were recently published (1). The exposimeter EME Spy 200 that covers 20 different radiofrequency bands from 87 to 5,850 MHz was used. The results were based on 1,669 data points recorded in November, 2015. The median value for total exposure was 921 µW/m² (0.092 µW/cm²), with certain outliers >95,544 µW/m² (6 V/m, which is the upper detection limit). One example of such very high measured power density was from a global system for mobile communications (GSM) + universal mobile telecommunications system (UMTS) 900 downlink band from a base station located at the Stockholm Central Station lower level (1). People standing at that area or passing by are involuntarily exposed to high RF radiation without their knowledge. It was concluded that this represented an improper location of a base station with an unnecessary high downlink level.

In European countries, the Old Town is a point of a national heritage, a place and source of cultural and historical development throughout centuries. Stockholm Old Town has already been retrofitted with several existing antennas to accommodate voice and data transmission. The aim of the present study was to characterize RF radiation already in place and its effect on the public. The antenna grid is expected to be further expanded to accommodate the rollout of 5G mobile networking as the next wave of mobile technology is implemented. As mobile base station antennas are placed on rooftops, outer walls of buildings and other places, visual perturbations appear in the form of antenna casings, cables and other peripheral devices. Not only does this damage the aesthetic appearance of the historic districts of the Old Town, but it is associated with exposure of the public to invisible RF radiation. Even where hidden antennas have been architecturally camouflaged (stealth installations), the RF radiation exposure is still present and remains a growing concern for environmental health.

The most drastic change brought about by the mobile telephony network is the change in the electromagnetic climate. The streets of old towns are filled with visually imperceptible RF radiation. As modern mobile telephony and data networking provide services at several different bands, there are different frequencies that also require band-specific antennas, therefore multiplying the number of antennas.

One major concern regarding RF radiation exposure in Sweden, as well as in a number of other countries, is that various authorities base their guidelines for exposure on the International Commission on Non-Ionizing Radiation Protection's (ICNIRP) guidelines that are based on short-term (acute) exposure. Chronic, low-intensity cumulative exposure, possible long-term health risks and non-thermal biological effects are not taken into consideration. The ICNIRP safety limit was established in 1998 (2) and updated in 2009 (3), without any further changes, and it still recommends 10 W/m² (10,000,000 µW/m²) as the reference level for limiting the public's exposure to RF radiation (2–300 GHz) (2). The Swedish Radiation Safety Authority (SSM) has adopted the ICNIRP guidelines. Several members of the SSM expert panel are also members of ICNIRP, which may entail a conflict of interest, since these members would rarely compromise the ICNIRP view so that critical opinions are not heard.

Our results on public RF radiation exposure at the Stockholm Central Station are orders of magnitude lower than the ICNIRP guidelines, where the median level of exposure was ~10,000 times lower. Wireless internet access in schools is also a concern (4), although measurements of levels are scanty.

In contrast to ICNIRP, the BioInitiative Reports from 2007 (5), updated in 2012 (6), also based their evaluation on the non-thermal effects of RF radiation. The BioInitiative 2012 Report assessed ~1,800 new peer-reviewed studies published between 2006 and 2012 that documented the new and lower scientific benchmark for health harm to be 30–60 µW/m² (0.003–0.006 µW/cm²), which is 0.0003–0.0006% below ICNIRP's guidelines. Also considering chronic exposure and sensitivity among children, the precautionary target level was proposed to be one-tenth of this value, namely 3–6 µW/m² (6). This exposure target level is not acknowledged by SSM, thus making it possible to overlook results on exposure, such as those at the Stockholm Central Station, and not providing precautions against potential detrimental effects on public health.

Interestingly, scientific evidence on the carcinogenic potential of RF radiation in laboratory studies has long been accumulating, but has mostly been ignored or entirely overlooked by decision-makers.

In 1982, Szmigielski et al published a study on the co-carcinogenic effects of RF radiation exposure and benzopyrene in mice (7) and demonstrated that 2,450 MHz of RF radiation at either 50 or 150 W/m² promoted carcinogenesis. The authors concluded that the resulting acceleration of development of spontaneous and chemically induced cancers indicated the carcinogenic potential of RF radiation.

Two studies published in 1990 demonstrated that non-thermal 2,450 MHz continuous-wave RF radiation exerted a biphasic effect on glioma cells (8) and lymphocytes (9). Cell proliferation was found at a specific absorption rate (SAR) of ≤50 W/kg, whereas a higher SAR suppressed DNA and RNA synthesis. These effects were reported to be non-thermal.

In 1992, Chou et al published a study on 200 rats exposed to 2,450 MHz pulsed RF radiation for 21.5 h/day for 25 months and 200 controls (10). SAR ranged between 0.144 and 0.4 W/kg, depending on the rat's weight. A statistically significantly increased incidence of primary malignant diseases was found in exposed animals compared with sham exposure. It should be noted that malignant lymphoma and thyroid cancer were among the malignancies found in exposed rats. These findings are interesting in view of the low SAR values in that study compared with the ICNIRP safety limit of SAR 2 W/kg in the brain for use of mobile phones (2).

In 1997, a study on mice carrying a lymphomagenic oncogene exposed to RF radiation was published (11). A total of 100 mice were sham-exposed and 101 were exposed for two 30-min periods per day for up to 18 months to 900 MHz pulsed RF radiation with power densities of 2.6-13 W/m² (SAR 0.008–4.2 W/kg; mean, 0.13–1.4 W/kg). The lymphoma risk was found to be statistically significantly higher in the exposed mice compared with that in the controls.

Those results were not confirmed in the study by Utteridge et al (12), which has been criticized for not being a replication study. However, the findings on lymphoma risk by Repacholi et al (11) and Chou et al (10) are of relevance in terms of increased non-Hodgkin lymphoma (NHL) risk in epidemiological studies on humans associated with the use of wireless phones. Thus, a statistically significantly increased risk of T-cell NHL was found in a previous study [odds ratio (OR)=8.75] (13) and in a second study for all NHL cases among subjects with >6 years of mobile phone use (OR=4.4 in men) (14), although based on limited numbers. Furthermore, the finding of thyroid cancer risk in the Chou et al study (10) is interesting in view of the sharp increase in the incidence of thyroid cancer during recent years (15). The thyroid gland is among the organs with highest exposure to RF radiation during use of handheld wireless phones, particularly smartphones (16,17).

The effects on tumour susceptibility in mice exposed to a UMTS test signal from the fetal period for up to 24 months was studied by Tillman et al (18). Animals were exposed to UMTS fields with intensities of 0, 4.8 and 48 W/m². The low-dose group (4.8 W/m²) was subjected to additional prenatal ethylnitrosourea (ENU) treatment. The group that was ENU-treated and UMTS-exposed at 4.8 W/m² exhibited an increased rate of lung tumours and an increased incidence of lung carcinomas as compared with the controls treated with ENU alone. The authors concluded that the study demonstrated a co-carcinogenic effect of lifelong UMTS exposure in female mice subjected to pretreatment with ENU (18).

A follow-up study of the Tillman et al investigation was published in 2015 (19). The exposure levels were 0 (sham), 0.04, 0.4 and 2 W/kg SAR. The numbers of lung and liver tumours in exposed animals were statistically significantly higher compared with those in sham-exposed controls, as were the numbers of malignant lymphoma. A tumour-promoting effect of RF radiation was found at low to moderate levels (0.04 and 0.4 W/kg SAR), which were well below the exposure limits for users of mobile phones (19).

A report was released from The National Toxicology Program (NTP) under the National Institutes of Health (NIH) in USA on the largest ever animal study on cell phone RF radiation and cancer (20). An increased incidence of glioma in the brain and malignant schwannoma in the heart was found. Acoustic neuroma or vestibular schwannoma is a similar type of tumour as the one found in the heart, although benign. The use of mobile as well as cordless phones has been associated with increased risk for glioma and acoustic neuroma in humans, as described below.

The same RF radiation that led male rats to develop brain tumours in the NTP study also caused DNA breaks in their brains. Female rats, which did not have statistically significantly elevated tumour counts, had fewer DNA breaks (http://microwavenews.com/news-center/ntp-comet-assay). According to Ron Melnick, who led the team that designed the NTP study, the results provide ‘strong evidence for the genotoxicity of cell phone radiation’ and this ‘should put to rest the old argument that RF radiation cannot cause DNA damage’. The study is currently under peer review.

These new NTP study results appear to corroborate the findings of the 1995 study by Lai and Singh regarding increased levels of DNA single-strand breaks in brain cells from rats acutely exposed to low-intensity 2,450 MHz microwaves (21). A dose-rate-dependent increase in DNA single-strand breaks was found in brain cells of rats at 4 h post-exposure to 0.6 and 1.2 W/kg whole-body SAR. In rats exposed for 2 h to continuous-wave 2,450 MHz microwaves (SAR 1.2 W/kg), increases in brain cell DNA single-strand breaks were observed immediately as well as at 4 h post-exposure (21).

Our research group published the first results on the increased risk for brain tumours associated with the use of mobile and cordless phones ~15 years ago (22–25). More results have been obtained from our subsequent studies, as well as from studies conducted in other countries. The International Agency for Research on Cancer (IARC) at WHO concluded in 2011 that RF radiation is a possible human carcinogen, Group 2B (26,27). The IARC decision was based on human Swedish studies by our group (28–30) and the IARC Interphone Study (31–33) giving evidence of increased risk for brain tumours, mainly glioma, and acoustic neuroma in subjects using mobile phones. These results have been replicated in our subsequent studies (34–37) and the French CERENAT study on glioma and meningioma (38). IARC concluded that there is limited evidence from experimental animal studies supporting the carcinogenicity of RF radiation. However, later published human and animal studies reported that there is sufficient evidence of the carcinogenicity of RF radiation. That conclusion is supported by our review (39) using the Bradford Hill viewpoints (40) on the association or causation on environment and disease. Mechanistic studies that support a causal inference are also currently available and discussed below.

As areas with high RF radiation were previously recorded at the Stockholm Central Station in Sweden (1), the aim of the present study was to continue with outdoor measurements in locations visited by numerous individuals in Stockholm. Stockholm Old Town is a popular place for tourists that is located in Stockholm City within walking distance from the Central Railway Station. Several people reside in the Old Town and a number of shops and restaurants are located within this area. In addition, certain government buildings, such as the Supreme Court and the Royal Castle and the nearby Swedish Parliament, are located in that region as well.

As this was a measurement study with no involvement of test subjects, no ethical permission was required.

The results of the Old Town measurements were based on 10,437 measurements in total during five separate measurement rounds, with a total of 696 min of measurements (~12 h). The results are displayed in Table II. The mean total level of RF radiation was 4,292.7 µW/m². The highest mean values were obtained for GSM + UMTS 900 downlink and LTE 2600 downlink (1,558.2 and 1,264.9 µW/m², respectively). It should be noted that the highest maximum levels were from base stations (5 bands) varying from 24,384.7 (LTE 800 downlink) to 95,522.5 (LTE 2600 downlink) µW/m². The highest maximum uplink (from a mobile phone) was found for GSM 1800, yielding 19,136.9 µW/m².

In Fig. 2, the box plot shows the results for five old town measurements and all exposure (total). The total median level was 534.0 µW/m² with several outliers of >100,000 µW/m².

In Fig. 3. the RF exposure variation over time is displayed for one of the measurement rounds. There was a large variation over time during these >2 h measurements covering major parts of the Old Town (streets and squares). The mean level was 5,371.4 µW/m² (median, 701.8 µW/m²; range, 4.2-83,348.8 µW/m²). Only for few moments the exposure was <100 µW/m², whereas several measurements were >10,000 µW/m².

The Stockholm Old Town RF radiation levels were measured on five occasions, yielding a total of 10,437 readings for the Old Town. This study showed a large variation in outdoor RF radiation level in the Stockholm Old Town. Furthermore, it should be noted that the measurements mainly characterize downlink exposure, without considering personal use of wireless devices, such as mobile phones. Indoor RF levels are likely to be somewhat lower with respect to inbound radiation, since the walls and windows of buildings attenuate the signals. On the other hand, on the premises on the opposite side of the transmitter antennas, the exposure levels may be even an order of magnitude higher if the beam happens to be directed towards these premises. For accurate information it would be necessary to make measurements inside buildings such as the Royal Castle, the Supreme Court, restaurants, apartments and the Swedish Parliament. Several buildings contain Wi-Fi routers for wireless communication, which would increase exposure at Wi-Fi bands. Furthermore, personal use of wireless phones (mobile as well as cordless), use of tablets and wireless internet via computers, add to RF radiation exposure.

This study provides information on the passive RF radiation exposure in a popular area of Stockholm, which is a destination for a number of visitors, such as tourists, citizens contacting different authorities, employees and residents. The study determined low levels at the surroundings of the Royal Castle, with a mean level of 755.6 µW/m², which is 17.6% of the total mean level in the Old Town (4,292.7 µW/m²). An even lower mean level was measured for the Supreme Court (403.6 µW/m), representing 9.4% of the total mean level for the Old Town. It should also be noted that certain hotspots had somewhat higher RF radiation levels, both for the Royal Castle and the Supreme Court (Figs. 5 and 7). This is also illustrated by the box plots (Figs. 4 and 6) with third quartiles of <1,000 µW/m², with the exception of one of the Royal Castle measurements.

The abovementioned results are in large contrast to the squares of the Old Town. The three major squares were measured, namely Stortorget in the area of the Royal Castle, Kornhamnstorg and Järntorget at the other end of the Old Town. Particularly striking is the RF radiation level at Järntorget, although high mean levels were measured in all squares.

The mean RF radiation level at Järntorget was 24,277.1 µW/m², which is 60 times higher compared with that for the Supreme Court. It should also be noted that the maximum level for Järntorget was 173,301.8 µW/m², compared with 4,088.1 µW/m² for the Supreme Court, which is 42 times higher.

The high RF radiation exposure at the squares is of major concern due to the numerous shops, outdoor restaurants and cafés. The measured RF radiation is almost exclusively from mobile telephony base stations (e.g., see Table VI for Järntorget). Some of the base stations appear to be located in buildings of rather low height, with the radiation directed towards the sitting area of restaurants and coffee shops. The location of a base station at one of the measured squares is shown in Fig. 12; it was placed rather low, with the main emission directed towards an outdoor restaurant at a short distance.

Our results are significantly lower compared with the ICNIRP exposure guideline of 10,000,000 µW/m². However, that guideline is outdated, since it does not include current scientific findings in humans and laboratory studies on RF radiation exposure. On the other hand, most measurements were higher than the scientific benchmark of 30–60 µW/m² that has been proposed to be the lowest observed effect level for RF radiation in Chapter 24 of the BioInitiative Report (6). Furthermore, our result on a total mean RF radiation level of 4,292.7 µW/m² in the Old Town (Table II) is of major concern, considering the oxidative effects on living cells from exposure to a level of RF radiation as low as 2,500 µW/m², as described by Yakymenko et al (41). This is also of concern, since reactive oxidative species are crucial in carcinogenesis.

In Sweden, the dominating sources are currently exposure from mobile communication with UMTS, 3G and LTE, 4G with addition of Wi-Fi indoors, but also in certain outdoors locations. Our study from Stockholm Central Railway Station in 2015 yielded a mean total radiation level of 2,817–4,891 µW/m² for each measurement round and a median value of 921 µW/m² for all rounds. The mean for all rounds was 3,860 µW/m². In addition, hotspots were registered yielding >95,544 µW/m², which exceeded the exposimeter EME-Spy 200′s upper detection limit. GSM+UMTS 900 downlink and UMTS 2100 downlink contributed the most, but LTE 800 downlink, GSM 1800 downlink and LTE 2600 downlink were also in the higher range of measurements (1). Thus, these results are similar to those of the present study.

The results of the present study show a large variation in passive outdoor RF radiation exposure in Stockholm Old Town. One explanation may be that there are no outdoor places with sitting areas to accommodate people during lunch or coffee breaks around the Royal Castle or the Supreme Court. This also applies to the Swedish Parliament, which is located on a separate island close to the Old Town. Thus, base stations are located around the squares to enable good network access. However, a lower level of radiation would still enable good network access (6). Comparatively low levels of RF radiation were measured for the Swedish Parliament compared with Järntorget. However, no indoor measurements were performed at the Swedish Parliament, where the RF radiation level would be quite different due to wireless internet access within the Parliament.

Unlike in other city districts, where mobile phone base station antennas are placed at a certain height, in the Old Town some antennas are located significantly closer to humans, sometimes only a few meters from the ground level. This creates areas of significantly high RF radiation, particularly in the immediate vicinity of such antennas. The exposure situation is worsened by the narrow streets and tall walls, which tend to trap the radio waves, hence creating reflections and increasing the exposure. The highest radiation zones may be proposed at buildings with windows facing the mobile base station antenna just across the street at the same height. The most affected are the individuals that have to stay longer in such places, also referred to as hotspots. Individuals whose living quarters, workstation or work area happen to be located in hotspots, are exposed to the heaviest RF load, whereas those who walk by hotspots only undergo momentary exposure and are less affected.

As the mobile telephony network is currently under development, the base station antennas are positioned following the current legal safety limits. However, the current limits only provide protection from short-term health effects (heating), but do not exclude possible long-term effects, which are currently under investigation. To grant better service coverage, the base station antennas are continuously placed in new locations, and the mobile telephony network becomes denser annually. Thus, the RF exposure of the population is increasing, particularly in densely populated areas.

As the research on long-term health hazards is ongoing, it is mostly focused on clearly identifiable biological effects. Knowledge on the effect of RF fields on mental health is of utmost importance, as in chronic exposure to RF radiation levels, such as those measured in hotspots in the Old Town. It should also be noted that long-term health effects, such as cancer, have not been investigated for this type of environmental contamination.

EME Spy 200 is a band-specific exposimeter and it enables identification and measurement of the majority of RF radiation bands currently used in Sweden. This study describes measurements mostly from far-field RF radiation and the exposure that the citizens may undergo without using personal wireless devices. Near-field exposure from personal mobile phones held near the ear or in the hand when on the internet will likely contribute considerably to an individual's total exposure.

The shielding effect from the body of a person carrying an exposimeter may be considerable, as shown by Bolte et al (42) when comparing a body-worn exposimeter with an exposimeter mounted on the roof of a car. This was partly compensated in our study by holding the exposimeter at a certain distance from the body.

Other studies have measured the exposure of RF radiation in public places. In 2012, Estenberg and Augustsson (43) used a car-mounted device to measure the frequency range 30–3,000 MHz in public places (rural, urban and large cities). The arithmetic mean measured exposure was 6,700 µW/m² in Stockholm, 1,500 µW/m² in 4 urban towns and 230 µW/m² in 2 rural areas. Similar to our study, the major sources were GSM and UMTS downlinks.

Earlier studies, prior to the use of smartphones, when 3G and 4G protocols became more common, reported lower exposure. Joseph et al (44) performed measurements with a personal exposimeter (EME-Spy 120–121) in different urban areas across Europe during 2007–2009. They found mean values for outdoor areas of 372–569 µW/m², mostly from mobile communications. In 2010, Bolte and Eikelboom (45) also performed measurements in The Netherlands with a personal exposimeter (EME-Spy 121) and found a mean value in outdoor activities of 208 µW/m². Visits to pubs, cafés, discos and snack bars yielded the highest mean value in their study (526 µW/m²), mostly from mobile uplink at 1,800 MHz. Urbinello et al (46) used the same exposimeter in three different European cities (Basel, Ghent and Brussels), every month for 1 year (April, 2011-March, 2012). The total from all sources of RF radiation in downtown outdoor areas increased by 56% in Basel, 27% in Ghent and 25% in Brussels. The mean levels were 0.32–0.58 V/m (271–892 µW/m²). In 2012, Aerts et al (47) conducted measurements in an urban/suburban area of Ghent, Belgium covering an area of 1 km². They selected 0.7 V/m (1,300 µW/m²) as a threshold for a hotspot and found five hotspots in the area. GSM 900 base station signals contributed the most to all five hotspots with 45–100%, with GSM 1800 and 3G present in two of the hotspots.

Verlock et al (48) used a Narda NBM-550 in public places in Belgium during 2012–2013. The total average from all RF radiation sources between 100 kHz and 3 GHz was 0.62 V/m (1,020 µW/m²), with a maximum of 2.4 V/m (15,287 µW/m²).

Calvente et al (49) conducted measurements outside the homes of 123 families in Granada, Spain during 2012–2013. The arithmetic mean root square (RMS) for power density was 799 µW/m² and the median value was 285 µW/m². For the different homes, the median RMS ranged from 5 to 11,559 µW/m². In 2015, Gonzalez-Rubio et al (50) used an EME-Spy 140 located inside the plastic basket of a bicycle. With this bicycle they covered all 110 regions in the city of Albecete, Spain, performing measurements. The average for the 3 bands of mobile telephone antennas (GSM, Digital Combat Simulator and UMTS) in the different regions varied from 0.04 V/m (4.2 µW/m²) to 0.89 V/m (2,102 µW/m²).

The mean values for exposure to RF radiation from the earlier studies indicate a clear increase in the more recent exposimeter studies. It is clear from our present and previous (1) studies that the level of ambient RF radiation exposure is increasing. With the development of mobile communications technologies and the widespread use of wireless services, the exposure will continue to increase with higher exposure levels and also other frequency bands, despite the already proven adverse health effects.

Long-term studies with laboratory animals at or below the levels in the present study have shown the effects of RF radiation on several physiological parameters in the body of mammals. These are non-thermal effects and are discussed briefly below, including a few human studies.

Several studies on rats have demonstrated that the blood-brain barrier (BBB) may open when exposed to RF radiation, resulting in pathological leakage of large molecules that may be toxic to the brain, with condensed dark neurons in the brain representing a sign of damage (51–54). The hippocampus, a center for memory and learning in the brain, appears to be particularly sensitive to neuronal damage from RF radiation and an opened BBB. Long-term exposure to 900 MHz RF radiation has been associated with extravasation of albumin in the hippocampus and cortex and impaired spatial memory in exposed rats (55–57).

A long-term study with exposure from GSM 900 MHz 3 h/day or DECT base station 8 h/day during 8 months reported a statistically significant downregulation or an overexpression for one-third of the 432 analyzed proteins from the brains of the RF-irradiated mice. Several neural function-related proteins, such as apolipoprotein E, heat shock proteins and cytoskeletal proteins, as well as proteins involved in brain metabolism, were altered (58).

Up to 12 months of exposure to Wi-Fi exerted an effect on microRNA in the brain (59) and in the testes, with head defects and DNA damage in sperm cells (60–62).

Yakymenko et al (41) published a review of 100 studies investigating the oxidative effects of low-intensity RF radiation in living cells, and demonstrated that exposure down to 2,500 µW/m² (63) with SAR values as low as 600 µW/kg (64,65) may increase oxidative stress in the cells. Long-term, low-intensity RF radiation exposure has also been associated with reduced levels of neurotransmitters, downregulation of microRNA, increase in pro-inflammatory cytokines and DNA damage with single-strand breaks in the hippocampus and cognitive impairments in learning and memory in exposed rats (65–67). An increase in frequency appears to exert more deleterious effect on several of the parameters (66).

Exposure levels in rats down to a SAR of 85 µW/kg, for 900 MHz during 2 h/day, 5 days/week for 30 days, increased the oxidative stress parameters in lipid peroxidation and protein oxidation. In addition, there was a statistically significant impairment in cognitive function in terms of spatial memory in the rats (68).

RF radiation has been shown to increase protein synthesis in proliferating human cells after 8 h of exposure, but not in quiescent white blood cells. This indicates a higher sensitivity of growing organisms (69). Furthermore, the capacity to repair DNA double-strand breaks was more affected by RF radiation in stem cells compared with differentiated cells, such as fibroblasts (70).

Epidemiological studies of mobile phone base stations have indicated health risks for humans. In particular, increased prevalence of adverse neurobehavioral symptoms or cancer in populations living at a distance of <500 m from base stations was found in 80% of the available studies in a review by Khurana et al (71). In another review, Levitt and Lai (72) included 56 studies. Exposure from base stations and other antenna arrays brought about changes in the immunological and reproductive systems of animals and humans, changes in biological material, DNA double-strand breaks, changes in calcium movement in the heart and increased proliferation rates in human astrocytoma cancer cells.

Long-term studies on low exposure to RF radiation of humans have demonstrated an effect on the neurotransmitters adrenaline, noradrenaline, dopamine and phenyletylamine when a GSM 900 MHz base station was installed in the village of Rimbach in Germany (73), as well as on cortisol and thyroid gland hormones in individuals living near base stations (74,75). Chronic dysregulation of psychobiological stress markers may contribute to health problems and chronic illnesses.

The results from these studies report levels of environmental exposure to RF radiation below the ICNIRP target level (2,3), but far above the BioInitiative Report level for biological effects (5,6). It has been argued that the BioInitiative Reports are not peer-reviewed. However, peer-reviewed and published studies also document health effects at low RF radiation levels (76,77). A recent article highlights the need for accurate information on the safety of exposure to RF radiation (78).

In conclusion, the aim of our study was to assess outdoor exposure to RF radiation at the Old Town and the nearby Parliament in Stockholm, Sweden. The Old Town is a part of Stockholm that is very popular among tourists, but also a destination for locals visiting the official buildings located in the area. In this study, real-time band-specific exposure measurements showed the highest contributors to the exposure to be download frequencies from the base stations of GSM + UMTS 900, UMTS 2100, LTE 800, LTE 2600 and GSM 1800 bands. The highest mean levels were found for GSM + UMTS 900 (3G) downlink and LTE 2600 (4G) downlink. The differences in exposure levels between the Supreme Court and Järntorget were striking, with the mean level 60 times higher at Järntorget. Unfortunately, studies on human risk from long-term environmental RF radiation based on personal exposure monitoring do not exist to the best of our knowledge. Future studies on cancer risk and other health effects from such exposure are imperative. The results of this study revealed unnecessarily high RF radiation areas in several parts of Stockholm Old Town. Using unnecessarily high power levels and placing mobile phone base station antennas too close to the ground and in heavily frequented areas pose an excess health risk to a significant part of the population.