In recent years, the education in many schools has changed considerably, from classes where the teacher holds lectures in front of a class, where every student has his/her own school books to read and write in, to classes where all students use their own laptop and no printed text books. The teacher may act as a mentor in the individually based education where the students work at their own pace and capacity directly through different programs on their personal laptops. Access to the Internet is, in most schools, available through wireless connection, wireless fidelity (Wi-Fi). There is usually a network of access points for Wi-Fi in classrooms and corridors to make it possible to teach and keep in contact everywhere in the school building. Most students and teachers also bring their own mobile phones, usually a smartphone, with them to class and use them for social communication and entertainment throughout the day. Besides Wi-Fi, other mobile networking protocols are also actively used—subscription to 3G and 4G wireless connections find widespread use among school children.

Information and communication technology (ICT) is an important field in teaching in our schools, both for every student to learn how to get information in different subjects and as a tool to learn and practice knowledge. Most students and teachers appreciate the development of more technology in school education and are very positive to all advantages it brings.

However, the benefits of ICT in education seem to be largely unproven and there are also concerns that there may be side effects (1). Results from the OECD’s PISA performance surveys from 2012 in reading and mathematics show decreasing results in countries that have invested the most in introducing computers in schools. Sweden is the country that deteriorated the most in the PISA performance surveys in mathematics and reading between 2003 and 2012. At the same time, Sweden was one of the countries that had invested the most in introducing computers in school (2). Results from the PISA performance surveys 2015 from Sweden in reading and mathematics show better results and are now just above the OECD average. However, Sweden is still well below its scores in 2003 (3).

Multitasking, to do several things at the same time, means that your attention has to shift focus between different activities. The brain can only concentrate on one matter at the time. Swift shifting of focus between different tasks may lead to impaired ability to store memories and knowledge (4–7). Even to have nearby classmates surfing on the Internet can impair learning (8). The time used to check their smartphone for new messages can give a positive feeling of being updated but takes time and concentration from school work (9).

Apart from during school time, most students use their smartphones, laptops, and tablets during evenings for several hours. In the yearly inquiry 2015/2016 in the schools in the county of Norrbotten in Sweden, 80% of the 16-year-old students spent at least 3 h in front of a screen in their free time. Five hours or more were spent by 35% (10). If the students also use a laptop during most classes in school, it can add up to many hours of screen time every day. A study in the United States concluded that college students spent nearly 9 h a day on their cell phones (11).

Screen time can affect us in many different ways. Long hours of sitting in front of a screen can lead to aches in the neck, back, and shoulders. In the yearly inquiry in Norrbotten, almost one-fourth of the 16-year-old girls often had aches in their neck, back, or shoulder (10).

Screen time has been shown to have a higher correlation to overweight and obesity than lack of physical activity (12), since prolonged sedentary activities can be a metabolic risk factor (13) and less physical activity may reduce cardiorespiratory fitness (14). There are also concerns about the emotional development for especially young children with a high amount of screen time. Time is taken from play, physical activities, and being with friends, parents, and siblings, which are important parts when growing up. The development of mirror neurons in the brain, which give us the possibility to imitate the behavior of other people, can be affected (15). Empathetic concern among American college students has dropped sharply since 1979 and especially since the year of 2000 according to a meta-analysis on studies performed between 1979 and 2009 (16).

High ICT usage among young adults in Gothenburg led to an increased risk for depression, sleep problems, and stress (17). Among adolescents in Japan using a mobile phone after lights out at night gave sleep disturbances, increased tiredness during day time, and several had suicide thoughts and worse mental health (18, 19). In a review, bedtime access and use of portable screen-based media devices showed a statistically significant association with inadequate sleep quality, poor sleep quality, and excessive daytime sleepiness (20). Increased duration of mobile phone use has been linked to higher risks for depressed mood and mobile phone addiction (11, 21, 22).

Adolescents at risk for mental health problems showed an association between both the amount of time spent using digital technologies and the numbers of text messages sent and increased same-day symptoms from attention deficit hyperactivity (ADHD) and conduct disorders (23). One study in Switzerland raised the question “whether problematic mobile phone use is the consequence of unfavorable conditions or whether and to what extent problematic mobile phone use reinforces behavioral problems as well as decreased mood and psychological well-being” (24).

In the WHO study, Health Behavior in School-aged Children (HBSC) conducted during 2013/2014 results from Sweden showed that girls’ mental health has deteriorated compared with the same study made during 2009/2010. The study has since 1985/1986 taken place around the world every fourth year among 11-, 13-, and 15-year-old girls and boys in over 40 different countries. In Sweden, sleep problems in particular have increased among 13- and 15-year-old girls. Of the girls, 39% have sleep problems more than once a week. Also, boys at the same age have sleeping problems, though to a lesser extent (around 27%). To feel depressed, nervous, or irritated has increased among Swedish, 13- and 15-year-old girls and were about twice as common compared with boys at the same age. The conclusion of the HBSC report brings up that “the fast technical development has changed children’s and adolescents’ everyday life in many ways, which may have influenced their mental well-being” (25). In the latest yearly inquiry in Norrbotten, Sweden, there is a statistically significant correlation between school districts with high amount of screen time and how many students in the community that feel depressed (10).

Considering the long hours of screen time and deteriorating health with increasing problems with sleep, depression, and aches among students, there are concerns regarding whether radiofrequency (RF) radiation from wireless networks, laptops, and mobile phones is harmless below reference values of RF radiation and if there can be potential adverse health effects from this exposure in the long term.

The reference values for RF radiation 10 MHz to 300 GHz were recommended in 1998 by the International Commission on Non-Ionizing Radiation Protection (ICNIRP) to 2–10 W/m² depending on frequency (26). This reference value protects against injuries caused by a heating effect over 1°C after an exposure of 30 min, and with a safety factor of 50. Injuries caused by other biological mechanisms than heating or from chronic effects of RF radiation are not taken into account in the ICNIRP’s reference values. Sweden and many other countries apply the reference value of 10 W/m² (10,000,000 µW/m²) for frequencies 2–300 GHz. Some countries, like Russia, Poland, Italy, India, and China have chosen lower reference values down to 0.1 W/m² (100,000 µW/m²) (27). In 2012, the Bioinitiative Working Group proposed a precautionary target level of 3–6 µW/m², with a safety factor of 10, since research studies on biological effects have shown biological effects of RF radiation down to 30–60 µW/m² (28).

The Council of Europe recommended in a resolution from 2011 that schools and other buildings where children spend their time should give preference to wired Internet connections and implement information and awareness-raising campaigns on the risks of potentially harmful long-term biological effects on human health from wireless technology, especially targeting children, teenagers, and people of reproductive age (29). In France, mobile phones are not allowed up to sixth grade in school. Israel and Cyprus have recommendations to not have wireless networks in preschool and to turn off Wi-Fi when not used in primary schools. Many schools in Sweden have started to ban mobile phones during school time since they disturb the students’ concentration on school work.

Exposure to RF radiation was classified as a possible human carcinogen, Group 2B, by the International Agency for Research on Cancer (IARC) at WHO in 2011. The decision was mainly based on case–control human studies on use of wireless phones by the Hardell group from Sweden and the IARC Interphone study, which showed increased risk for brain tumors, i.e., glioma and acoustic neuroma (30–33). Further research has confirmed the increased risk for brain tumors and mobile phone use (34–37). A report released in 2016 from the National Toxicology Program under the National Institutes of Health in the United States strengthened the association between RF radiation and cancer. An increased incidence of glioma in the brain and malignant schwannoma in the heart was found in life-long RF radiated rats (38), thereby supporting human epidemiological studies on brain tumor risk. Several laboratory studies have shown mechanistic effects in carcinogenesis such as oxidative stress, down regulation of mRNA and DNA damage with single strand breaks (39–43).

The risk for cancer may be accentuated for children partly because of their likely longer life-time use of wireless devices, but also since their smaller size and thinner skull bone give higher RF radiation to the brain (44). Children are also growing and have more immature cells which can be more sensible to RF radiation (41, 45).

Beside the cancer risk, studies with laboratory animals have shown that RF radiation may open the blood–brain barrier and may thereby increase the rate toxic molecules enter the brain tissue, hurt neurons in the hippocampus (the brain center for memory), impair spatial memory in exposed rats (46–49), and down or up regulate essential proteins in the brain engaged in the brain’s metabolism, stress response, and neuro-protection (50). Long-term, low intensity of RF radiation exposure has also shown reduced levels of neurotransmitters and key-regulating enzymes in the rat brain (51). Increase in frequency seems to have more deleterious effect on several of the parameters (42).

Long-time exposure from a new GSM base station in Rimbach, Germany revealed adverse effects on neurotransmittors like Epinephrine, Norepinephrine, Dopamine, and Phenyletylamine which after 18 months had not normalized, especially not in children and chronically ill adults (52). Chronic dysregulation of psychobiological stress markers may contribute to health problems and chronic illnesses. More behavioral problems have been seen in children with higher exposure to RF radiation compared with children with lower exposure (53–56).

Several studies on humans indicate an influence on the electrical activity in the brain seen in EEGs after exposure to RF radiation during both sleep and in active memory tests (57, 58). Other studies on people exposed to RF radiation from mobile phones have shown disturbed glucose metabolism in the brain (59), effects on endocrine system (60), and in young adults with high-cumulative amounts of hours of mobile phone use, decreased β-trace protein, which is a key enzyme in the synthesis of a sleep-promoting neurohormone (61). DNA damage has also been found in hair root cells after 30 min of mobile phone talk (62) and buccal mucosal cells in high-mobile phone users with more than 10 h of mobile phone use per week for over 5 years (63).

Wireless fidelity exposure on testes and sperms has led to decreased sperm mobility, more head defects, and DNA damage (39, 64–66). Wi-Fi signals in studies on laboratory animals have had effects on heart rhythm and blood pressure after 1 h of exposure (67). Low-intensity long-term exposure to Wi-Fi up to 12 months in animals induced oxidative stress in the lens (68), increase in pro-inflammatory cytokines, oxidative stress, and DNA damage with single strand breaks in the hippocampus in the brain (42), down regulation of microRNA expression in brain tissue (40), and cognitive impairment and DNA damage (69). In humans, Wi-Fi exposure during a language test showed gender-related effects on EEG in a large area of the brain (70), but no effect on EEG or on a psychomotor vigilance test during a 60-min Wi-Fi exposure (71).

It is important to collect information about the RF radiation exposure in schools to consider the best technological devices used in education regarding convenience and economy, but also to consider the potential long-term health effects as mentioned above.

The aim of this study was to assess RF emissions in the classroom by measuring the teachers’ RF exposure and thereby to approximate the children’s exposure but also to identify the main sources of exposure. The investigated schools were equipped with a wireless infrastructure. We wanted to evaluate the measured RF radiation levels in relation to ICNIRPs’ and Sweden’s reference values, but also in relation to countries with stricter safety limits and to the Bioinitiative Working Group’s suggested level. Furthermore, we wanted to compare our measured results with exposimetric studies in schools in other countries. Finally, based on measurements results, we discuss strategies for RF radiation risk management in schools.

This is a descriptive study with measurements of exposure to RF radiation in schools with Wi-Fi and one laptop for each student. School principals were contacted by phone and if interested, information was sent by mail. The principal, or in some cases interested teachers, recruited the teachers willing to participate. All teachers were informed both by written information and personally at the start of measurements by one of the researchers. All teachers signed a statement of agreement to participate before starting the measurements in accordance with the declaration of Helsinki. The Ethical Committee approved the study (Uppsala University DNR 2015/485).

In total, 18 teachers were recruited from seven schools, both rural and urban. The schools were located in the city of Örebro in the middle of Sweden and in the county of Norrbotten in the northern part. The included schools had from 28 to 1,300 students (Table 1). The teachers carried the exposimeter (weight 440 g) in a small and comfortable shoulder bag. The exposimeter was carried during the whole working day. If the teacher also carried a mobile phone on the body, the teacher was advised to place it on the opposite side from the exposimeter. Signals from the mobile phones did not affect the results as the exposimeter is capable of discriminating between upload and download bands. The teachers were carrying the exposimeter from 1 to 4 days.

The teachers handed over their work schedule and also filled in a questionnaire about (1) the school building, (2) classroom for each lesson, teachers’ office or lunch room, (3) if there was any access point for Wi-Fi in the room, (4) the number of students in each class, (5) if they used their laptops, (6) if they were connected to the Internet and if so for how long time, and (7) if the students were allowed to use their mobile phones in school and if so if they could connect to the school’s Wi-Fi network.

If the teacher in the questionnaire reported that he/she was near a microwave oven in the lunch room and there also was a peak signal at the frequency band 2.400–2.4835 GHz, we excluded the measurement data during that time. The signal from the oven (2.45 GHz) could be interpreted as a Wi-Fi signal by the exposimeter.

The information from the teachers, their work schedule, and questionnaire made it possible to relate different activities during the day with the measured values. The measurements took place during school time during the time period March–November 2016.

For the measurements of RF radiation, the exposimeter EME-Spy 200 from Satimo (MVG Industries, Brest, France) with a valid calibration was used. The exposimeter measures 20 predefined frequency bands, as presented in Table 2. These cover the frequencies of most public RF radiation emitting devices currently used in Sweden. The exposimeter covers frequencies of 87–5,850 MHz. For frequency modulation (FM), TV3, TETRA, TV4&5, Wi-Fi 2.4 GHz and Wi-Fi 5 GHz, the lower detection limit is 0.01 V/m (0.27 µW/m²); for all other bands, the lower detection limit is 0.005 V/m (0.066 µW/m²). For all bands, the upper detection limit is 6 V/m (95,544 µW/m²). The sampling time was every fourth second.

The exposimeter measures different telecommunications protocols: FM-radio broadcasting; TV broadcasting; TETRA emergency services (police, rescue, etc.); GSM second generation mobile communications; UMTS third generation mobile communications, 3G; long-term evolution (LTE) fourth generation mobile communications standard, 4G; digital European cordless telecommunications (DECT) cordless telephone systems standard; Wi-Fi 2.4 and 5 GHz wireless local area network (WLAN) protocol; worldwide interoperability for microwave access (WIMAX) wireless communication standard for high-speed voice, data, and Internet.

EME-Spy 200 utilizes 3-axis antenna to capture RF radiation from all possible directions. The unit reports the exposure after statistical processing, since each reported value is the sampling outcome, where many samples are collected and statistically processed including minimum, mean, median, and maximum values.

In total, 18 teachers carried the exposimeter during one to four working days in school. Every teacher measured from 6 to 31 h resulting in 5,321–28,238 readings for each teacher. In total, 230,100 readings were assessed corresponding to 255 h of measurements of RF radiation. The total results for each teacher are displayed in Table 3 with mean, median, minimum, and maximum values for a total of all RF radiation exposures.

The mean for each teacher ranged from 1.1 to 66.1 µW/m² and medians from 0.6 to 51.8 µW/m². The maximum values varied between 241.8 and 82,857.3 µW/m². Figure 1 shows box plot for each of the 18 teachers with median values, boxes with the first and third quartiles and outliers. Table 4 shows all measurements divided into the 20 different frequency bands from 87 to 5,850 MHz. The highest peaks ranging from 43,938.7 to 82,856.6 µW/m², came from mobile phones uplink from 4G 800 MHz, GSM 1,800 MHz, and 3G 2,100 MHz. The highest max downlink value, 3,285.9 µW/m², came from 4G 800 MHz, while max for Wi-Fi 2.4 GHz was 4,482.8 µW/m² and Wi-Fi 5 GHz 3,321.4 µW/m². All separate means added up to a total of 22.5 µW/m². Total median was 4.6 µW/m².

The teacher MH3 with the lowest mean, 1.1 µW/m², taught in a classroom far from any Wi-Fi access point and had a wired connection to her own laptop for the education. The students used their laptops only in flight mode and mobile phones for students were banned during school time. The teachers in the school were allowed to carry and use their own mobile phones which can be seen by some higher peaks of GSM, 3G and 4G uplinks up to 151.1 µW/m², Table 5 and Figure 2.

The teacher SG1 with the highest mean 66.1 µW/m² and median 51.8 µW/m² taught in a school with a weaker Wi-Fi network which entailed that both students and teachers connected their mobile phones to base stations outside school. Here, the different mobile phone networks had a high impact on the total RF radiation exposure, see Table 6 and Figure 3. Exposure from Wi-Fi 2.4 and 5 GHz was low, partly because there was a wired connection to the teacher in each classroom.

The highest mean during a whole lesson, 107.3 µW/m², is shown in Table 7 and Figure 4. The teacher OG1 stood right below an access point while using her Wi-Fi connected laptop during the lesson. None of the students used their laptops during this lesson. A short high mean came from a part of a lesson when another teacher, MH2, let the approximately 20 students in the class stream and watch YouTube videos. For a few minutes, the mean exposure went up to 396.6 µW/m², mostly from Wi-Fi 2.4 GHz, Table 8 and Figure 5.

This study was carried out in accordance with the recommendations of The Ethical Committee, Uppsala University, with written informed consent from all subjects: all subjects gave written informed consent in accordance with the Declaration of Helsinki. The protocol was approved by The Ethical Committee, Uppsala University, Sweden (DNR 2015/485).

LKH and LH made contributions to the conception and design of the study. LKH, LH, and TK handled the measurements and information to schools and teachers. TK contributed with technical advice for the exposimeters and the measurements. MC made the statistical analyses, the tables, and the figures. All the authors agreed on the risk management principles. LKH drafted the manuscript. All authors revised the manuscript and gave approval to the final version to be published.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. The reviewer HM and handling editor declared their shared affiliation.