Meta-analysis of in vitro and in vivo studies of the biological effects of low-level millimetre waves

Abstract

Background: With the roll-out of new technologies such as 5G, there has been renewed community concern regarding the adequacy of research on possible health effects from associated radiofrequency radiation, mainly in the millimetre wave (MMW) band.

Objective: We conducted a meta-analysis of in vitro and in vivo studies investigating bioeffects of MMWs at low exposure levels.

Methods: We identified 107 in vitro and in vivo studies investigating MMWs and biological effects in which the power density employed has been below 100 W/m², which is below the current standards for occupational local exposures. Where possible, we estimated the magnitude of the principal effect reported or set this magnitude to zero in studies reporting no significant effects. We also estimated the quality of the studies, based on a methodology used in previous analyses.

Results: We show a negative correlation between effect size and both power density and specific absorption rate. There was also a significant negative correlation between effect size and quality score. A multivariate analysis revealed that there is an increase in the effect size for certain biological systems being investigated and laboratories in which the work was carried out whilst the quality score for some of these tends to be low. We note that many of the studies were motivated by a desire to elucidate the possible mechanisms in therapeutic devices rather than assessing the safety of telecommunications systems. Finally, it appears that the presence or absence of modulation does not influence the reported effect size.

Significance: Many of the findings of this meta-analysis have not been reported before and have important implications for overall interpretation of in vitro and in vivo data. Overall, the results of this study do not confirm an association between low-level MMWs and biological effects.

Keywords: Disease; Empirical/Statistical models; Radiation.

Fig. 1. Effect size as a function of power density from 84 studies where both were reported.

Effect size (ES) is defined as reported maximum difference between exposed and sham, divided by the standard deviation of sham; power density (PD) is in W/m² (log scale).

Fig. 2. Effect size as a function of specific absorption rate in 31 studies were both were reported.

Specific absorption rate (SAR) is in W/kg (log scale).

Fig. 3. Effect size as a function of reported frequency (n = 88).

Where multiple frequencies were investigated, the one at which the greatest ES was reported is plotted.

Fig. 4. Effect size as a function of quality score.

Quality score (QS) was judged by two independent scorers (scale 0–5).