Prevalence and Psychiatric Co-Morbidity of Self-Reported Electromagnetic Field Sensitivity in Taiwan: A Population-Based Study

Background: Environmental illness has been demonstrated to be crosslinked with psychiatric disorders. No published population-based studies have investigated the relationship between electromagnetic hypersensitivity and psychiatric morbidity. Objective: In this study, we aimed to investigate the prevalence and characteristics of people with self- reported electromagnetic field sensitivity (SREMFS), especially the psychiatric co- morbidity, in adult population of Taiwan. Methods: Questions regarding hypersensitivity to EMF were included in a telephone survey of environmental health risk perception. Participants were adults selected from nation- wide households registered in the database of the computer- assisted telephone interviewing (CATI) system via two-stage geographically stratified systematic sampling. A total of 1251 persons successfully completed telephone interviews adminstered by trained interviewers ( response rate = 17.76% and cooperation rate = 32.5%). Data analyses were conducted among subjects with complete information (n = 1197). SREMFS was defined as self- referred to be Allergic or very sensitive to getting near any electromagnetic field”.SREMFS questions included presence of electromagnetic hypersensitivity, their most sensitive EMF sources, and severity of hypersensitivity ( with a four- point Likert scale). People with psychiatric morbidity were identified by a screening questionnaire, the Brief Symptom Rating Scale-5 (BSRS-5), with a cutoff score at 6. Sample weight was adjusted according to age, gender and geographical distribution of nation-wide population. Statistical analyses were performed using STATA 8.0 ( StataCorp, 2003). The prevalence of SREMFS in Taiwan (13.4%, 95% C.I. = 11.4–15.3) was higher than that of Western countries.The psychiatric co-morbidity rate among people with SREMFS was 28.8% (95% C.I. = 21.6–36.0). Dose-response relationship existed between percentage of psychiatric morbidity and severity of SREMF. No statistical significance in gender, geographical area, education, and marital status between individuals with SREMFS and those without SREMFS. Associated risk factors for SREMFS included age above 65 (OR = 0.20, 95% C.I. = 0.08–0.56 ), poor selfreported health condition ( OR = 6.03, 95% C.I. = 1.99–18.22) , unable to work (OR = 2. 09, 95% C.I. = 1.21–3.60), as well as psychiatric morbidity (OR = 2.18, 95% C.I. = 1.43-3.33) after controlling for all other factors using multiple logistic regression. We further restricted the analyses on people with SREMFS in the severe end. Age and unable to work became borderline significant, while psychiatric morbidity (OR = 4.68, 95% C.I . = 2.50–8.78) remained statistically significant in their relationship with SREMFS. Discussion and Conclusion: This study demonstrated that people with psychiatric morbidity were more likely to report hypersensitivity to EMF after controlling for the risk factors including age and working status. The effect of psychiatric morbidity on SREMFS needs to be studied further