Editor, Cellular phone use in most hospitals has been restricted since the 1990s following a number of early studies highlighting the possible risks of their inadvertent interference with medical equipment (Medical Devices Agency 1997; Robinson et al 1997; Fetter et al. 1998). None of the devices tested are routinely used in the ophthalmology setting, meaning that the extension of these policies to eye departments may be unjustified. This study aims to address this lack of data by evaluating the electromagnetic interference (EMI) patterns of cellular phones in the ophthalmic setting. The field strength and electromagnetic radiation (EMR) generated by cellular phones was examined using an appropriately calibrated spectrum analyser. Cellular phones receive and transmit signals with base stations through microwaves at carrier frequencies between approximately 900 MHz and 1800 MHz. Electromagnetic interference occurs when the phone (an intentional radio transmitter) causes a piece of electronic equipment to act as an unintentional radio receiver. The phone has adaptive energy and thus can increase or decrease its energy output to compensate for signal strength and proximity to the network antenna. This investigation used a standard electromagnetic testing protocol (Medical Devices Agency 1997). Ophthalmic equipment was tested in the usual working environment with clinically sensible settings in both active clinical use and idle modes. Each surface of the equipment was examined for EMI at 10 different distances (within 0−2 m), while the cellular telephones were in standby, dialling and then connected modes (representing different energy levels). Two fully charged phones, operating at 900-MHz and 1800-MHz carrier frequencies, were used. The sweeps were repeated three times to ensure a genuine effect. The field strength emitted by the phones (in milli-Watts of peak power above background random noise) ranged from 10.18 mW to 243.82 mW. The higher end of the energy spectrum was used to fully assess susceptibility of ophthalmic devices. Interference was demonstrated with only two pieces of equipment (Table 1). The B-scanner showed horizontal banding on the display screen at 0-m test distance (phones touching scanner), with both the 900-MHz and 1800-MHz frequencies (226.09 mW and 243.82 mW, respectively). The interference was present only in the dialling and connected (not standby) modes. The YAG laser emitted a buzzing sound from within the unit when the 900-MHz phone was within 10 cm in connected and dialling modes (≥ 201.93 mW). The 1800-MHz phone did not cause any problems with this device. There was no evidence that the interference induced any clinically significant malfunctioning while the devices were operational. However, the buzzing sound from the YAG laser could be a cause for concern as it demonstrates a potential for EMI with potentially random effects on the circuitry. Ophthalmology is a very technology-dependent specialty and many of its tools are electronic, which makes information on EMI highly relevant. Our study found that a form of EMI was demonstrated with only two of the 20 pieces of ophthalmic equipment tested but they suffered no clinically detectable malfunction as a result. This indicates a relatively safe profile. However, further investigation by electrical engineers and manufacturers should be encouraged. Previous studies have described effects on operating theatre, anaesthetic, emergency medicine, clinic and intensive care equipment (Fetter et al. 1998), thus including the majority of non-ophthalmic devices that the ophthalmologist might encounter. These studies found a 98% safety profile at 1 m and concluded that ‘it would rarely be clinically important’ at this distance (Robinson et al 1997; Irnich & Tobisch 1999). As our data were collected according to the standardized protocol of the Medical Devices Agency (1997), the results can be used in conjunction with the existing body of evidence to help appraise safety in the ophthalmic setting. Caution against even the occasional potential interference (and very unlikely malfunction) of life support equipment seems intuitive as the consequences of failure could be fatal. No such dire effects could occur with B-scanner or YAG laser interference. The growing importance of cellular phones in medical communication is leading to a reappraisal of restrictions. Most ophthalmic units are located out of the main path of core hospital equipment and this study found that non-clinically significant EMI occurs only at or within 10 cm of ophthalmic equipment. Data from this study can now aid a more evidence-based platform for ophthalmic equipment and purely ophthalmic areas (Saleh 2004). Rather than applying blanket bans, perhaps clearly defined cellular-free zones could be implemented in a manner analogous to that of no-smoking areas.