The Effect of Supplementary Antioxidant Therapy in Patients Who Report Hypersensitivity to Electricity: A Randomized Controlled Trial

Lena Hillert, MD, Birgitta Kolmodin-Hedman, MD, PhD, Peter Eneroth, MD, PhD, Bengt B. Arnetz, MD, PhD; Department of Environmental Health, Stockholm County Council; Division of Occupational Medicine, Department of Public Health Sciences, Karolinska Institutet, Stockholm; Unit for Applied Biochemistry, Clinical Research Centre, Novum/Karolinska Institutet, Huddinge; Section of Social Medicine, Department of Public Health and Caring Sciences, Uppsala University, Sweden

Abstract and Introduction

Context: Hypersensitivity to electricity is a proposed environmental illness of unknown etiology. Patients report a variety of symptoms that they relate to electric equipment. The afflicted individuals suffer from ill health. Many interventions have been tried but, to date, there is no one specific treatment that has been proven superior to other remedial actions. In general, there is a lack of controlled prospective studies.
Objective: To test the hypothesis that antioxidant therapy reduces symptoms and improves health in patients reporting hypersensitivity to electricity.
Design: Randomized, double-blind, crossover, placebo-controlled study.
Setting: Patients referred to the Environmental Illness Research Centre, Stockholm County Council.
Patients: Sixteen patients reporting hypersensitivity to electricity.
Intervention: Antioxidant supplementation (vitamins C and E, selenium).
Main outcome measures: Self-reported symptoms and reported degree of hypersensitivity to electricity, serum levels of uric acid and diphenylpycrylhydrazyl (DPPH).
Results: The results indicated no significant differences in reported symptoms, reported hypersensitivity to electricity, or oxidative status in serum between periods of antioxidant and placebo treatments. Serum levels of DPPH and uric acid showed no correlation with the reported degree of symptoms or hypersensitivity to electricity.
Conclusions: The study did not show any beneficial effect of antioxidant supplementation for patients reporting hypersensitivity to electricity. The results do not support the hypothesis that oxidative stress is a major contributor to ill health in patients who report hypersensitivity to electricity.

Medically unexplained symptoms, considered by some to represent "environmental illness" (EI), have been a growing concern for clinical physicians and researchers in recent decades. Nonspecific and varied multiple symptoms have been attributed to various environmental factors and situations. Typically no objective diagnostic physical findings or laboratory test abnormalities are identified, and there is a lack of agreement on the definition and whether it even is an illness.[1] Geographic differences are observed. In the United States, multiple chemical sensitivity (MCS) is one of the more common names for this cluster of symptoms.[2,3] The pathogenesis and the pathophysiology of patients with the symptoms is still to a large degree unknown.[1,4] In Sweden, patients reporting hypersensitivity to electricity (HE) constitute one of the larger groups.[5] A self-help group, The Swedish Association for the Electrosensitive, currently has about 2000 members. In a recent health survey performed by the Department of Environmental Health in Stockholm County, 1.5 % of the approximately 11,000 responders between ages 19 and 85 reported hypersensitivity to electric or magnetic fields (1 of 11 possible answers in a question phrased "Are you allergic or hypersensitive to:").[6]

The terms implying different causal factors, such as MCS and hypersensitivity to electricity, are often introduced from outside the community of standard medicine and are based on the afflicted persons' reported experience of triggering factors for their symptoms. The attribution of negative health effects to exposure to electromagnetic fields or activated electric equipment has not been confirmed in double-blind provocation studies involving persons reporting hypersensitivity to electricity.[7] Symptoms were reported as often during sham exposure as during presence of specific electric and magnetic fields or activated electric equipment such as visual display units or television sets, and subjects were not able to distinguish field exposure from sham.[5,8,9] Present knowledge suggests that many factors may be involved in symptom formation.[5] No pathophysiologic marker or diagnostic test is known. In a report to the European Commission, DG V, a European group of experts stated: "The term 'electromagnetic hypersensitivity' is used here to designate a phenomenon where individuals experience adverse health effects while using or being in the vicinity of electric, magnetic or electromagnetic field sources and devices (EMF devices). The use of the term 'electromagnetic hypersensitivity' does not -- by itself -- presuppose or indicate any causes of these adverse reactions."[5] A similar definition was agreed on by a group of Swedish experts in a report to the Swedish government.[10] The term "hypersensitivity to electricity" is in this paper used according to these definitions, ie, to describe symptoms that the afflicted individuals report as being triggered by activated electric equipment, and with the understanding that a causal relationship has not been proven.

In a situation where no certain explanation or proven cure can be offered, patients often seek help and remedies outside the standard healthcare system. In many cases, therapists practicing complementary medicine claim to have identified the cause of reported ill health and to have an effective treatment for the condition. One of the suggested causal theories in Sweden is that patients who experience symptoms they attribute to hypersensitivity to electricity are suffering from oxidative stress. Subsequently, these patients are advised to start supplemental therapies of vitamins, minerals, and antioxidants, even if no deficiency can be biologically assessed.

Oxidative stress, such as an increase in free radicals (prooxidants), compared with available antioxidants, influence cellular health and function[11] and can be induced by different factors. Some examples are low dietary intake of antioxidants, physical exercise, and exposure to environmental factors, such as cigarette smoke and air pollution.[12] So positive supplemental antioxidant treatment effects may have many different explanations.

Investigations of possible biologic effects from electromagnetic fields have reported results that might agree with positive antioxidant treatment effects, though no health-related effects have yet been proven for exposures equivalent to those from activated electric equipment in everyday life, which has been reported to trigger symptoms in persons reporting the syndrome. In vitro studies have shown an intermittent-exposure electromagnetic field effect on the production of heat shock proteins,[13] and animal studies have indicated that antioxidant therapy can influence the production of heat shock proteins.[14] Electromagnetic field effects on melatonin levels have been reported in animal and in human studies.[15,16] In a study of effects from the 50 Hz electromagnetic fields on sleep, field exposure was associated with impaired sleep and a nonsignificant decrease in circulating melatonin.[17] Melatonin is an antioxidant, and a lowered concentration might result in an increase in free radicals and an increase in risk for developing different kinds of health problems. Today, this line of reasoning is purely hypothetical regarding hypersensitivity to electricity and of interest primarily in exploring possible explanations for the patients' reported improvement due to supplementary treatment with antioxidants.

The stated suffering of the patients reporting hypersensitivity to electricity calls for remedial activities and/or help in coping until specific etiologic factors can be determined. The extent of the health problem reported as hypersensitivity to electricity clearly motivates further research. Based on patients' reports of improvement and suggested biologic, electromagnetic field effects, the present study intended to test the hypothesis that antioxidant therapy will lead to improved health and a reduction in reported symptoms in persons who report hypersensitivity to electricity. We also intended to explore the possibility of individual differences in response to the treatment, for example, if persons with indications of a higher degree of oxidative stress in the biochemical analyses were the ones to benefit from the treatment.


The subjects were recruited from patients referred to the Environmental Illness Research Centre, Stockholm County Council. The inclusion criteria for patients to be asked to participate in the study were:

  • reported hypersensitivity to electricity

  • a reported experience of change in symptoms within 24 hours after a perceived change in exposure to electromagnetic fields

  • a history of visual display units (VDUs) or fluorescent lights as the initial triggering factors

A medical examination was done before the study. Patients with known somatic or psychological disorders that could account for the present symptoms and the reported variation in symptoms were excluded. Three patients, who were already taking antioxidants, were asked to stop taking them 3 weeks before entering the study. Out of 32 persons asked to participate, 16 declined. The reasons for nonparticipation were other medical disorders (n = 6), a reluctance to discontinue their own therapies including antioxidants (n = 4), a job situation that did not allow regular meetings with the nurse (n = 3), and a complete recovery (n = 3).

The study group consisted of 16 persons (13 women and 3 men); the participants' mean age was 39.5 (range, 21 to 59). Only 1 person smoked. One person had asthma that required only occasional medication. Another person was taking low doses of an antidepressant drug but had not experienced a change in symptoms after the start of this medication. Low doses of antidepressant drugs are sometimes tried for patients reporting HE following a case report in Sweden of a remarkable improvement in 1 patient in 1992.[18] Table 1 shows the characteristics of the study group. Two previously developed and validated indices were used for characterization of the degree of skin and neurovegetative symptoms in the study group.[19]

The study was a randomized, double-blind, crossover, placebo-controlled treatment study (Figure 1). Two pills (administered in a dose cartridge with dates) were taken daily during the treatment periods. The antioxidant supplementation was a daily intake of 180 mg vitamin C and 100 mg vitamin E. In addition, 120 mcg selenium was included. Selenium is an essential trace mineral that affects the antioxidant enzyme activity of glutathione peroxidase. Antioxidants and placebo were available as pills, identical in appearance, to be swallowed without chewing (Pharmacia & Upjohn AB, Sweden). The patients, physicians, nurses, and laboratory technicians were blinded to the medication code for antioxidants and placebo. The code was broken after the initial statistical analyses were done.

Flow chart of study design. A double blind, crossover study comparing antioxidant treatment with placebo.

The patients entered into the study at 3 different times (March, September, and October 1997) because we were not able to collect blood samples from more than 6 persons during each period. The treatments with antioxidants and placebo were run in parallel during all times with an equal number of patients in each group.

Table 2 summarizes the outcome variables. The primary outcome of interest was a possible reduction in reported symptoms and reported degree of relationship to electricity. A secondary aim was to investigate if any observed improvement was paralleled by a change in the analyses of oxidative status. Daily scoring of perceived degree of suffering from stress and symptoms, and scoring of perceived change in hypersensitivity and symptoms after each treatment period, were recorded in self-administered questionnaires. The symptoms in the questionnaire were sensation of the face being swollen, itching, reddening of the skin, sensation of heat, stinging pain and/or tingling in the skin, nausea, vision problems, headache, fatigue, difficulties concentrating, aggressiveness, and depression.

Diphenylpycrylhydrazyl (DPPH)[20] and uric acid[21] in serum were analyzed for detection of change in the oxidative status in the participants' blood. The DPPH content was expressed as the red/ox equivalent of uric acid in micrometers (mcM) Samples were drawn at a standardized time (+/- 30 minutes for each patient, between 7 and 11 AM), at the participants' home or place of work. Samples were drawn on 3 separate days before each treatment period and during the last days of treatment and washout period, respectively. The mean of these values was used for the analyses to minimize the influence of day-to-day variations. Participants were instructed to have only a light lowfat breakfast without tea on the mornings of the blood sample collections. Medication known to affect the biochemical analysis was avoided and, if taken, reported on the daily questionnaire. (In only 1 case, paracetamol was taken within 2 days of drawing blood samples, and the results did not indicate any influence on the oxidative status).

It is possible that a supplementation of antioxidants might result in either an increase (due to a decrease in the use) or decrease (due to a reduction in the production) in the levels of DPPH and uric acid in blood. Our hypothesis was that a change in either direction would be recognized at the end of the treatment period of antioxidants, with normalization after the following washout period.


After careful verbal and written information about the study, written consent was obtained from all patients. The Huddinge University Hospital's ethics committee approved of the study protocol.


The Wilcoxon signed rank test for paired differences between placebo and antioxidant treatment was used for statistical analysis of the primary efficacy parameters of the crossover study with 2 treatment periods. The outcome variables were the mean, over the period studied, for the daily questionnaire and the individual change in variables measured before and after each period for the oxidative status. Spearman's rank coefficient of correlation was used to determine the relationship between different outcome variables. Student's t test and ANOVA were used for the statistical analysis of the influences of treatments, periods, and individual variations. The Wilcoxon rank sum test was used for examination of carryover and period effects.[22] The statistical analyses were performed in the SAS statistical system. A P value < .05 was regarded as statistically significant.


There were no significant differences in the pretreatment measurements (time A) in the scoring of symptoms and avoidance behavior or biochemical parameters among the 3 groups of participants entering the study at different times. The results showed no indication of any delayed effect during the treatment periods. So, we used the mean of the daily scores for the periods in the further comparisons.

Possible carryover and period effects were investigated for uric acid, DPPH, all symptoms, skin and other symptoms, respectively, as reported in the daily questionnaire and in the questionnaire after each treatment period. Except in 1 case, there were no significant results that indicated either carry-over or period effects. DPPH showed a significant period effect. There was a decrease in DPPH after the first treatment period and an increase after the second treatment period (Table 3), regardless of treatment order. The reason for this observation is unknown, and no similar effect was observed in uric acid. In an ANOVA analysis for treatment, period, and individuals there was no significant period effect. Because the outcome variables regarding symptoms and degree of relationship to electricity showed neither significant carryover nor period effect, the analyses of the outcome variables for possible treatment effects were performed.

There were no significant differences in mean scores of symptoms, perceived hypersensitivity to electricity either between periods of antioxidant and placebo treatments (Figure 2), or in evaluation of change in health status after the treatments.

Mean values of symptom scores for questions on all symptoms, skin, and other symptoms separately, estimated perceived hypersensitivity to electricity, and avoidance behaviour during different treatments (VAS); see Table 2 for description of labels. (No significant difference between treatment periods.)

The results of the analyses of DPPH and uric acid were within expected values. There were no consistent differences in the results from the different periods of sample collections (Tables 3 and 4). No significant correlation was found during treatment between DPPH and uric acid on one hand and all symptoms, skin symptoms and other than skin symptoms on the other hand. There was no indication of any subgroup or individual showing any distinct and consistent response.

Most of the participants were working, and there was a significant increase in the reported degree of symptoms during weekdays compared with weekends during all periods but one. Further analysis of differences between weekdays and weekends during the entire study period showed that the increase in symptoms during weekdays was significant for skin and symptoms other than from the skin (P < .01 in both cases). This finding is matched by a parallel significant increase in reported perceived relationship to electricity during weekdays (P < .01) and increase in avoidance behavior during weekends (P < .01).


The present study did not show any influence on self-reported health symptoms nor the oxidative status in blood due to treatment with antioxidants. Our results do not support the hypothesis that oxidative stress is a major contributing factor in environmental illness reported as HE. The oxidative balance seemed to vary more from person to person and from the daily food intake than from the change in supplementation with antioxidants during the different study periods.

The results of this study cannot rule out a possible supplementation effect as a result of higher doses of antioxidants. But it does not seem likely that higher doses should produce a different effect because there was no indication of increased oxidative stress based on the levels of DPPH and uric acid in blood in the prestudy tests of any one of the participants. Recent studies indicate that higher doses of antioxidants than were previously believed are needed for positive health effects, for example, 400 IU/day (about 400 mg/day) of vitamin E in patients with coronary artery disease.[23]

Positive health effects, eg, prevention of coronary artery disease and cancer, from antioxidant supplementation has limited support from prospective controlled trials, while epidemiologic data have shown stronger correlation between antioxidant consumption and health effects.[11,23] Considering this, a negative result in a prospective controlled trial on antioxidant supplementation is not completely unexpected. But the background of the present study was not epidemiologic data. Instead, it was specific reports of positive effects regarding symptoms and hypersensitivity from supplementation therapies.

To our knowledge, all participants took the prescribed therapy daily. We did not ask the participants to keep a daily record of their diets, and information on this might have been valuable in light of the present results. Given this information, we would have been able to evaluate the participants' total intake of antioxidants. Only 1 person in the study smoked, and it is possible that the group as a whole consisted of rather health-conscious persons with high antioxidant intake, judging from their daily diets.

The study group is comparably small but was chosen to be representative of the most typical presentation of this syndrome in Sweden. The study group is well characterized regarding belief, behavior, symptom indices, and temporal aspects of symptoms. This was necessary in order for other researchers to be able to reproduce this study or compare results from other treatment studies with ours.

In a previous study, Arnetz and coworkers evaluated other nonconventional therapy effects, namely acupuncture, in 20 patients who reported HE .[24] It showed a significant reduction in skin symptoms and in scoring on a symptom scale measuring degree of subjective sensitivity to environmental agents (various physical/chemical agents in the indoor air environment) after treatment with deep as well as superficial acupuncture. The authors of this study theorized that either superficial or deep acupuncture was efficient for subjective symptoms, even though results suggested that deep acupuncture impacted more favorably on physiologic data. Yet, the placebo effect may have been at play. We did not observe an improvement over time in our study. One difference between the former and the present study is that all the participants in the study on acupuncture received more personal attention because a therapist performed the treatment twice a week, while our therapy was self-administered at home. This might indicate that a caring environment plays an important role.

Cognitive behavioral therapy has been shown to reduce perceived degree of relationship or handicap due to what patients considered HE, but no clear effect was seen on reported symptoms as compared with environmental illness patient controls.[25,26]

The suffering from ill health and the attributions made by persons reporting the syndrome may have large implications on the afflicted persons' work ability. In the majority of cases in Sweden, the first symptoms are experienced in relation to work with VDUs. Reported symptom triggering factors, such as VDUs and fluorescent lights, are also harder to avoid at work than during free time. This was illustrated in the present study in the daily scoring of symptoms and avoidance behavior during the workweek as compared with weekends. Regardless of what contributing factors may exist for the ill health reported as HE, the attribution to work-related factors constitutes a high-risk situation for a progression to an inability to work.

The group of persons reporting symptoms related to electricity is very heterogeneous. There are several indications that persons who report mainly skin symptoms during work with VDUs are a separate group from persons reporting a general HE. Reported symptoms and prognosis differ between the 2 groups.[27,28] The prognosis in the first group is good in the majority of cases, but most cases of generalized hypersensitivity to electricity have started as VDU-related skin symptoms. We do not have any method for identification of individuals at risk for progression in illness. The present study failed to identify a positive effect of antioxidant supplementation for persons reporting hypersensitivity to electricity. Based on our present knowledge, recommendations of interventions should be based on each individual's unique situation. Medical, psychosocial, and environmental factors must be considered in the investigation of patients who report HE, MCS, or other forms of environmental illness.[5,29] Prompt actions aimed at any identified wanting condition of established importance for the well being of individuals should be taken. If symptoms persist, supportive care aimed at reducing symptoms and functional handicap should be offered.[29]


  1. American Academy of Allergy, Asthma and Immunology (AAAAI) Board of Directors. Idiopathic environmental intolerances. J Allergy Clin Immunol. 1999;103:36-40.

  2. Cullen MR. The worker with chemical sensitivity: an overview. Occup Med State Art Rev. 1987;2:655-662.

  3. Kreutzer R, Raymond NR, Lashuay N. Prevalence of people reporting sensitivities to chemicals in a population based survey. Am J Epidemiol. 1999;150:1-12.

  4. Graveling RA, Pilkington A, Geroge JPK, Butler MP, Tannahill SN. A review of multiple chemical sensitivity. Occup Environ Med. 1999;56:73-85.

  5. Bergqvist U, Vogel E, editors. Possible health implication of subjective symptoms and electromagnetic fields. A report by a European group of experts for the European Commission, DG V. Solna, Sweden: National Institute for Working Life; 1997 (Arbete och Hälsa 1997:19).

  6. Department of Environmental Health. Miljöhälsorapport 1998 (Report on environmental health 1998, in Swedish). Stockholm, Sweden: Department of Environmental Health, Stockholm County Council; 1998.

  7. Knave B. Electric and magnetic fields and health outcomes - an overview. Scand J Work Environ Health. 1994;20(special issue):78-89.

  8. Lonne-Rahm S, Andersson B, Melin L, Schultzberg M, Arnetz B, Berg M. Provocation with stress and electricity of patients with "sensitivity to electricity." J Occup Environ Med. 2000;42:512-516.

  9. Flodin U, Seneby A, Tegenfeldt C. Provocation of electric hypersensitivity under everyday conditions. Scand J Work Environ Health. 2000;26:93-98.

  10. Bergqvist U, Hillert L, Birke E. Elöverkänslighet och hälsorisker av elektriska och magnetiska fält. Forskningsöversikt och utvärdering. (Electromagnetic hypersensitivity and health risks from electric and magnetic fields. A research review and evaluation. In Swedish, English summary.) A report to the Swedish government. The Swedish Council for Work Life Research. November 2000.

  11. Halbert SC. Diet and nutrition in primary care. From antioxidants to zinc. Prim Care. 1997;24:825-843.

  12. Hollán S. Free radicals in health and disease. Haematologia. 1995;26:177-189.

  13. Smith O. Cells, stress and EMFs. Nat Med. 1996;2:23-24.

  14. Ushakova T, Melkonyan H, Nikonova L, et al. The effect of dietary supplements on gene expression in mice tissues. Free Rad Biol Med. 1996;20:279-284.

  15. Lambrozo J, Touitou Y, Dab W. Exploring the EMF-melatonin connection: a review of the possible effects of 50/60-Hz electric and magnetic fields on melatonin secretion. Int J Occup Environ Health. 1996;2:37-47.

  16. Burch JB, Reif JS, Yost MG, Keefe TJ, Pitrat CA. Reduced excretion of a melatonin metabolite in workers exposed to 60 Hz magnetic fields. Am J Epidemiol. 1999;150:27-36.

  17. Åkerstedt T, Arnetz B, Ficca G, Paulsson LE, Kallner A. A 50-Hz electromagnetic field impairs sleep. J Sleep Res. 1999;8:77-81.

  18. Jansson B. Behandlingsförsök med psykofarmaka (Treatment with psychopharmacological drugs, in Swedish). In: Ramel C, ed. Seminarium om elöverkänslighet (Symposium on hypersensitivity to electricity, in Swedish). Stockholm, Sweden: Royal Swedish Academy of Science; 1993:25-28 (KVA report 1993:2).

  19. Hillert L, Kolmodin-Hedman B, Söderman E, Arnetz BB. Hypersensitivity to electricity; working definition and additional characterization of the syndrome. J Psychosom Res. 1999;47:420-438.

  20. Dinis TCP, Madeira VMC, Almeida LM. Action of phenolic derivatives (acetaminophen, salicylate, and 5-aminosalicylate) as inhibitors of membrane lipid peroxidation and as peroxyl radical scavengers. Arch Biochem Biophys. 1994;315:161-169.

  21. Ames BN, Cathcart R, Schwiers E, Hochstein P. Uric acid provides an antioxidant defense in humans against oxidant- and radical-caused ageing and cancer. Proc Natl Acad Sci U S A. 1981;78:6858-6862.

  22. Pocock SJ. Clinical Trials. A Practical Approach. New York: John Wiley & Sons; 1997.

  23. Spencer AP, Stier Carson D, Crouch MA. Vitamin E and coronary artery disease. Arch Intern Med. 1999;159:1313-1320.

  24. Arnetz B, Berg M, Anderzén I, Lundeberg T, Haker E. A nonconventional approach to the treatment of "Environmental Illness." J Occup Environ Med. 1995;37:838-844.

  25. Andersson B, Berg M, Arnetz BB, Melin L, Langlet I, Lidén S. A cognitive-behavioral treatment of patients suffering from "electric hypersensitivity." J Occup Environ Med. 1996;38:752-758.

  26. Hillert L, Kolmodin Hedman B, Dölling BF, Arnetz BB. Cognitive behavioural therapy for patients with electric sensitivity - a multidisciplinary approach in a controlled study. Psychother Psychosom. 1998;67:302-310.

  27. Bergdahl J. Psychological aspects of patients with symptoms presumed to be caused by electricity or visual display units. Acta Odontol Scand. 1995;53:304-310.

  28. Eriksson N, Höög J, Sandström M, Stenberg B. Facial skin symptoms in office workers. A five-year follow-up study. J Occup Environ Med. 1997;39:108-118.

  29. Weaver VM. Medical management of the multiple chemical sensitivity patient. Regul Toxicol Pharmacol. 1996;24:S111-S115.