Nieto-Hernandez R, Rubin GJ, Cleare AJ, Weinman JA, Wessely S. Can evidence change belief?  Reported mobile phone sensitivity following individual feedback of an inability to discriminate active from sham signals. J Psychosom Res 65(5):453-460. Aug 15, 2008 Ahead of print.

Electromagnetic hypersensitivity (EHS) is a poorly understood condition in which people report nonspecific symptoms after perceived exposure to weak electromagnetic fields (EMF). The lack of evidence in support of the association between EMF and these symptoms suggests that EMF plays little, if any, role in the pathogenesis of EHS. Little is known about the most appropriate way to treat EHS, and research regarding treatment for EHS is scarce.

The objective was to study the effect of giving individualized feedback to people who reported “sensitivity” to global system for mobile communication (GSM) mobile phones (MP). The study hypothesis was that providing disconfirming evidence to these individuals would alter their perceived sensitivity to MPs at 6 months follow-up.

Sixty-nine volunteers between 18 and 75 years old took part in a provocation study. Perceived sensitivity to MPs was assessed using a version of the Sensitive Soma Assessment Scale (SSAS) prior to the provocation study (Time 1), immediately prior to the provision of feedback (Time 2), and 6 months later (Time 3). In the provocation study, participants were exposed under double-blind conditions to two testing sessions, one involving a GSM mobile phone signal, and one - a sham signal. In order to test whether the pulsing nature of the GSM signal was important in triggering symptoms, participants also took part in a third testing session involving exposure to a non-pulsing carrier wave (CW) signal. Sessions were separated by a minimum of 24 hours, and the order of the sessions for each participant was randomized. During each session, participants recorded whether they experienced headaches or other subjective symptoms. At the end of each session, each of them recorded whether they believed a signal had been present. Following feedback at Time 2, participants were asked how they feel about the results. Their answers were coded according to whether or not the participants were questioning the underlying causes of their symptoms.

The primary dependent variable in the data analysis was the degree of self-reported sensitivity to mobile phone signals at Time 3, with the main independent variable being the feedback that each participant received as to whether they were “correct” or “incorrect” in determining the presence or the absence of the signal in the provocation study. The type of feedback was determined by their actual performance. SSAS score at Time 2 was used as a covariate to adjust for baseline differences. Other potentially important covariates (e.g. age, sex, education) were tested and only significant covariates were included in the models.


Time 2 data were obtained and feedback was provided to 61 (88%) of the 69 individuals who took part in the provocation study. Fifty-eight individuals (84%) participated in the 6 months follow-up. Thirty one respondents were categorized as being correct in the provocation study and 27 as incorrect. No significant differences in SSAS scores or in symptom severity were found between individuals told that they were correct or incorrect in detecting the presence of a mobile phone signal. Out of 61 participants interviewed at Time 2, twenty four (39%) made comments which suggested that they were reconsidering their attribution of symptoms to mobile phone signals, and 37 participants (61%) did not show any apparent reconsideration in their beliefs. No significant differences in terms of mean SSAS scores and symptom severity at Time 3 were detected between individuals whose responses suggested a change in attribution and those whose responses suggested no change.

Although 31 participants were categorized as correct for the purposes of this study, this proportion is what would be expected by chance if the participants were randomly selecting which conditions involved real signal and which were sham. The results must not be taken as implying that 31 participants were genuinely sensitive. The study found no evidence to suggest that providing individualized feedback about a person’s capacity to discriminate accurately between active and sham mobile phone signals in a provocation study altered their perception of themselves as sensitive to mobile phones or altered the severity of the symptoms that they experienced in relation to mobile phone use. Even comparison of patients who did reconsider their attributions with those who did not, failed to identify significant differences in symptom severity or perceived sensitivity. The authors noted that some methodological issues with their study might prevent from detecting a genuine effect of feedback.

Suggesting that a patient’s illness may have a psychogenic component can sometimes have a negative effect on the therapeutic relationship between doctor and patient. In this study, some of the participants were openly hostile to the feedback and others attempted to find explanations why the study results were either flawed or did not apply to them. Overall, hostile reactions were in the minority, and many participants were willing to accept that psychological factors may have been relevant to them.

The provision of accurate feedback was insufficient to change attributions or reduce symptoms in this study. However, an overly negative reaction to feedback was not observed among most participants, and some were willing to consider that factors other than EMF may be relevant in causing their symptoms. Discussing possible psychological factors with EHS patients may be beneficial for some.

Home             Links              Sitemap               Contact Us
© McLaughlin Centre for Population Health Risk Assessment