Introduction

Establishment of electromagnetic compatibility (EMC) is important for safe use of electronic medical equipment in hospitals because of possible electromagnetic interference (EMI) with this equipment caused by radio waves coming from outside the hospital [1–5]. EMI may also occur between electronic medical equipment and radio waves emitted from cellphones, wireless local area networks (W-LANs), and radio frequency identification (RFID) tags [6–9].

Use of cellphones is restricted in most Japanese hospitals based on the “Guidelines for Use of Mobile Handsets to Prevent EMI with Electronic Medical Equipment”, published in 1997 by the Council for Countermeasures Against Unnecessary Electromagnetic Waves Japan [10]. However, usage regulation has liberalized gradually due to progression of electronic medical equipment immunity and improvement of cellphone performance. By July 2012, all second generation (2G) cellphone services had terminated, and only third (3G) and subsequent generation cellphones are now used in Japan. Electricity in the cellphone terminal varies depending on the distance from the base station. The maximum electrical radiation emitted from a Japanese 2G cellphone (Personal Digital Cellular) terminal is 800 mW, and this could potentially invoke EMI with electronic medical equipment [10]. However, the maximum emission from a Japanese 3G cellphone terminal is only 200 or 250 mW. This value also decreases by a few mW if the terminals receive base station signals with good reception. Emission is much lower with a Personal Handyphone System (PHS, maximum of 80 mW and average of 10 mW). Thus, EMI with electronic medical equipment is now relatively less likely.

In August, 2014, the old guidelines were replaced by the “Guidelines for Use of Mobile Phones and Other Devices in Hospitals for Secure, Safe Use of Wireless Communication Devices in Hospitals”, which were announced at the Japan Electromagnetic Compatibility Conference (EMCC) [11]. These new guidelines address key issues with regard to safe and secure use of cellphones, including separation, public use, protecting personal and medical information, an improved structure for EMC, and setting area-specific usage rules. Based on the new guidelines, use of cellphones in Japanese hospitals is expected to increase markedly.

Despite these improvements, the potential for EMI between electronic medical equipment and a cellphone in a poor signal area is still a concern. For example, it was reported that a ventilator can breakdown due to maximum electricity radiation from a Japanese 3G cellphone terminal at a distance of 50 cm [12]. This suggests that rapid deregulation may increase the potential for EMI and indicates that knowledge of signal strength from cellphone base stations in a hospital is important to ensure greater safety in use of cellphones and electronic medical equipment. Additionally, use of W-LAN and RFID-based clinical support systems, such as those used for electronic health records, biological monitoring, location detection, and physical distribution management, has also expanded in recent years [13–18]. Thus, proof of the compatibility of electronic medical equipment with communication devices is needed as hospitals introduce such wireless technology and products.

Each hospital needs to understand their current circumstances to take adequate measures with regard to the background electromagnetic environment. However, accurate measurement of this environment is difficult because radio waves are also emitted in operation of electronic medical equipment and communication devices, which are constantly running, and patients and visitors bring electronic communication devices that emit radio waves, such as cellphones, mobile routers, personal computers, and handheld games. Therefore, to establish the basis for promoting greater safety in use of cellphones and to evaluate the electromagnetic environment, measurements should be made after completion of construction of a hospital, but before the opening of the hospital to patients, and again after opening of the hospital. In this study, we had an opportunity to measure electromagnetic waves under these conditions, which allowed investigation of the electromagnetic environment at a newly-built hospital. Herein, we report the results of these measurements.

Methods

Location and Study Protocol

The hospital in which we measured the electromagnetic environment was the newly-built Kitasato University Hospital (KUH), located in Sagamihara, Kanagawa, in eastern Japan. Sagamihara is a mid-sized city with a population of 700,000, but KUH was built outside the metropolitan area. KUH has 1000 beds in 21 wards, 20 operation rooms, a rooftop heliport, and about 2200 staff members. The new building has 14 above-ground floors and one basement floor, with floors 7 to 14 housing wards. All floors are 100 m in length and oriented east to west. The total floor area is 92,776 m2 and the building is 70 m tall. The hospital opened in May 2014. The medical school of Kitasato University (a nine-story building) is located next to KUH. There are no buildings exceeding 70 m in height around KUH.

The study included wideband measurements and cellphone base station measurements. Both types of measurements were conducted before and after opening of the hospital, during the day (from 9 a.m. to 5 p.m.). Wideband measurements were made in February to March 2014 and August 2014. Cellphone base station measurements were made in February 2014 and February 2015. After the hospital opened, we had to consider the care and privacy of patients. Therefore, we did not measure locations in which medical care was provided, such as patient rooms, intensive care units (ICU), consultation rooms, physiological laboratories, and X-ray examination rooms.

Wideband Measurements

To determine the pure electromagnetic environment before the hospital opened (propagating only from incoming radio waves, since electronic equipment was not in operation), the frequency distribution of the electric field intensity induced by radio waves coming from outside was measured at 73 locations in the building, including outpatient clinics, wards, corridors, operation rooms, treatment rooms, and ICUs. The same measurements were performed 6 months after the hospital opened to determine the affect of operation of electronic equipment and traffic. In both measurements, the frequency ranged from 30 MHz to 3 GHz. The electric field intensity was measured in vertical and horizontal polarizations with a spectrum analyzer (MS2721B, Anritsu) and specific antennas. The antenna was erected 150 cm above the floor. Antenna types, resolution bandwidth (RBW) and video bandwidth (VBW) in each frequency range are shown in Table 1. The strongest electrical intensity for all measured frequencies was recorded using a maximum hold function for half a minute. After measurement, we determined the usage of recorded signals from a frequency table and analyzed the frequency distribution and electric intensity related to cellphone terminals and base stations at each location [19]. Data were recorded in dBm units and the results were converted to V/m to match the immunity evaluation unit for electronic medical equipment.

Table 1 Measurements detail in the wideband measurements
Full size table

Cellphone Base Station Measurements

Single frequency bands from cellphone base stations were recorded to determine the change of signal strength due to cellphone terminals in each location. The measured frequency ranged from 2.13 to 2.15 GHz. This frequency band is assigned to the 3G cellphone system (Wideband Code Division Multiple Access). Electric intensity was measured with a spectrum analyzer (MS2713E, Anritsu) and dipole antenna (ED-B033S-C3, ANTTEC). Data were recorded in dBμV units. Measurements were performed at 230 locations before the hospital opened and at 112 locations after the hospital opened. Measured locations were selected for as many floors as possible.

Results

Wideband Measurements

Radio waves detected at each location before and after the hospital opened are shown in Tables 2 and 3, respectively. Frequency modulated (FM) radio signals (70 to 85 MHz), aeronautical radios (278 to 285 MHz), community wireless systems (270 to 275 MHz), multimedia broadcasting services for cellphones (207 to 220 MHz), police scanners (360 MHz), ultra-high frequency (UHF) televisions (488 to 566 MHz), cellphone base stations (850 to 900 MHz, 1475.9 to 1510.9 MHz, 1844.9 to 1879.9 MHz, and 2.11 to 2.17 GHz), and Worldwide Interoperability for Microwave Access base stations (2.55 to 2.65 GHz) were recorded at high intensity at the windows of upper floors. Cellphone base stations were particularly detected at all points on upper floors. On the lower floors, cellphone base stations were detected at many locations, but broadcast waves were rarely recorded. Automatic terminal information service (126.3 MHz) was detected at windows on the south side. This signal was coming from Camp Zama (a United States Army base located in the cities of Zama and Sagamihara, about 5 km south of KUH). There were no incoming radio waves in spaces enclosed by metallic materials, such as operation rooms and radiological areas. There was no significant difference in these recorded levels before and after the hospital opened. However, cellphone terminal signals and W-LAN using the 2.4 GHz band were frequently detected at various locations after opening of the hospital.

Table 2 Measured radio waves before the hospital opened
Full size table
Table 3 Measured radio waves after the hospital opened
Full size table

The frequency, value, polarization, and radio wave usage of detected electric field intensities of >0.1 V/m (100 dBμV/m) are shown in Table 4. The maximum electrical intensity measured in this study was 108.796 dBμV/m (0.28 V/m) associated with cellphone base stations (2127 MHz), which was observed on the south end of the 2nd floor before the hospital opened. After the hospital opened, cellphone base station signals were recorded at many locations, but there were very few signals in the operation rooms, dialysis treatment center, center of the building, and basement. Cellphone terminal signals were recorded at 42 locations and 210 signals were detected, of which 13 did not correspond to base station signals (Table 5).

Table 4 Measured radio waves with intensity ≥0.1 V/m
Full size table
Table 5 Detected cellphone unit frequencies with no corresponding base station signals
Full size table

Cellphone Base Station Measurements

Measured electric field intensities on each floor before and after the hospital opened are shown in Figs. 1, 2, 3, 4, and 5. The color of circles indicates the received strength: blue indicates good reception with intensity of 60 to 70 dBμV; red indicates poor reception, but possible communication, with intensity of 20 to 30 dBμV; and black indicates no reception or out of service, with intensity <20 dBμV. Before the hospital opened, electric field intensity indicating good reception was detected at various locations on the upper floors. On the lower floors, the intensity was high near the windows, but decreased in the center of building. Out of service measurements were made at most locations in the basement. After the hospital opened, the overall electric field intensity was lower, but notable intensity was detected in the center of the building on the lower floors. From before to after the hospital opened, signals decreased by a mean of 3 dBμV at all locations, but those on floor 14 decreased by a mean of 6 dBμV.

Fig. 1
figure 1

Cellphone base station signal strength (B1F) before (upper panel) and after (lower panel) the hospital opened

Full size image
Fig. 2
figure 2

Cellphone base station signal strength (1F) before (upper panel) and after (lower panel) the hospital opened

Full size image
Fig. 3
figure 3

Cellphone base station signal strength (2F) before (upper panel) and after (lower panel) the hospital opened

Full size image
Fig. 4
figure 4

Cellphone base station signal strength (4F) before (upper panel) and after (lower panel) the hospital opened

Full size image
Fig. 5
figure 5

Cellphone base station signal strength (12F) before (upper panel) and after (lower panel) the hospital opened

Full size image

Discussion

Previous measurements of the electromagnetic environment in hospitals have been conducted at only one time point, either before or after opening of the hospital. In this study, we measured electromagnetic waves at a newly-built hospital prior to opening and then repeated the measurements after opening of the hospital. Changes in the electromagnetic environment were found after opening of the hospital due to operation of electrical and communication devices and turnover in the population. Japanese radiofrequency clearance has changed significantly over the past years, in part through expansion of frequency bands allocated for cellphones resulting from termination of the analog TV broadcasting service in 2012. Thus, our results provide a current model for evaluation of the electromagnetic environment at a modern Japanese hospital.

Before the hospital opened, incoming radio waves were frequently recorded, but there were few signals from communication devices brought in by patients, visitors or medical staff. None of the measuring crew used a cellphone, PHS, or other communication devices. Thus, the results before the hospital opened reflect the electromagnetic environment based only on incoming radio waves in the hospital, with no effects of communication device signals and device-driven radio waves or noise. At some locations, PHS and 2.4 GHz band radio waves were occasionally detected before the hospital opened, but these were relatively weak and came from an old building next to the new building.

After the hospital opened, PHS, cellphone terminals, and radio waves in the 2.4 GHz band were detected frequently and relative strongly. Medical staff also used a PHS. Many W-LAN access points had been installed at KUH and these radiated 2.4 GHz band radio waves for use of the hospital information system, computerized patient and nursing records, and a picture archiving and communication system in most locations. Medical staff, patients, and visitors also brought their own communication devices, including cellphones, which are almost always powered, even if not in use.

A survey at a modern urban university hospital in Japan found a maximum electrical intensity of 200 V/m at 2.79 GHz from airport surveillance radar waves [4]. This value is extremely high and could invoke EMI with electronic medical equipment. However, other radio waves such as UHF television signals, land mobiles, and cellphone base stations were almost always <1 V/m in this hospital. At an urban university hospital located about 600 m from the Tokyo Tower (a communication tower that radiated broadcasting waves until 2013), the maximum electrical intensity was <1 V/m in the 170 to 222 MHz band from analog TV broadcasting [5]. At a few US hospitals, the average electrical intensity due to incoming radio waves was found to be about 0.03 V/m [3]. Our results are similar to these findings. The maximum electrical intensity detected in this study was 109 dBμV/m (0.28 V/m), associated with cellphone base stations (2127 MHz). This value is lower than the EMC marginal value of general electronic medical equipment, as specified in IEC 60601-1-2 (3 V/m) [20]. Therefore, a serious problem of EMI with electronic medical equipment is extremely unlikely in this situation. Moreover, after the hospital opened, the maximum electric intensities were 0.14 V/m, associated with PHS terminals and W-LAN. This value was higher than that for incoming radio waves, which indicates that radio waves from inside the hospital were more important than incoming radio waves.

Electrical intensity associated with cellphone terminals is subject to the electrical intensity of the base station radio wave. While out of service, Japanese 3G cellphone terminals radiate a maximum electrical intensity (200 or 250 mW) searching for base stations with which to communicate. This electrical intensity could potentially invoke EMI with electronic medical equipment. In KUH, use of a cellphone (for e-mail, internet, and games) is allowed in all areas, except for consultation rooms, operation rooms, ICUs, and emergency medical care centers. This is because electronic medical equipment susceptible to EMI, such as remote patient monitors, infusion and syringe pumps, and electrocardiographs, are used in these rooms. Telephone calls are allowed in cafeterias, waiting rooms, corridors, elevator halls, and patient rooms. However, using a smartphone as a Wi-Fi router, commonly known as “tethering”, is restricted in KUH to avoid interference with the hospital W-LAN.

In this study, weak signals radiating from the 2.13 to 2.15 GHz band from cellphone base stations were recorded after the hospital opened, but not before opening. Depending on traffic and installation of certain equipment, incoming radio waves reaching the center of the building were proportionally damped. There were no cellphone base station signals in the basement, in part of the corridor (in the center of the hospital), in operation rooms, and in the hemodialysis unit. However, strong cellphone terminal signals were detected in several locations, including these areas, in which electronic medical equipment is routinely used. Therefore, our results indicate that EMI with electronic medical equipment by a cellphone terminal is a potential concern.

The EMCC suggested a separation distance of about one meter from a cellphone to electronic medical equipment [11]. van Lieshout et al. suggested that the policy of keeping cellphones one meter from critical care bedside equipment in combination with easily accessed areas of unrestricted is sufficient to eliminate most EMI of 3G cellphone terminals with electronic medical devices [21]. The EMCC suggested that if hospitals can confirm safety based on their own results and using instruction manuals for specific electronic medical equipment, a separation distance of lower than one meter may be acceptable [11].

Establishing femtocell base stations is likely to expand coverage inside hospitals with poor signal quality [22]. Around a femtocell base station, the radiation level of cellular access to the base station becomes lower under specific conditions, compared with a distant base station. There is less EMI with radio waves from a cellphone for surrounding electronic medical equipment. KUH plans to install these base stations in the near future. However, this installation is costly and improvement of the electromagnetic environment may be limited. If the hospital permits the use of all types of cellphones, base stations will have to correspond to all frequencies of signals and all communication systems, excluding those before 2G. Presently, there are 6 frequency bands provided by 4 service carriers in Japan. It would be useful for each hospital to have femtocell base stations, but the cost, limited improvements, and the need to facilitate all services prohibit introduction of these base stations in many hospitals.

Our results suggest that use of cellphones was safe on the upper floors, but that this use requires consideration on the lower floors, and especially in the basement and in the center of the building. To promote greater EMC safety, it would be desirable for each hospital to measure the electrical intensity of cellphone base station signals in areas in which electronic medical equipment is used and to establish appropriate rules for use of cellphones based on measurement of the electromagnetic environment.

Conclusion

The electromagnetic environment in a new university hospital building was compared before and after the hospital opened. No radio waves with extremely strong intensity signals were detected. There were also no cellphone base station signals, but very strong cellphone terminal radio waves were detected on several floors. This is a concern for EMI and it would be desirable to improve these conditions.