Conduct of a personal radiofrequency electromagnetic field measurement study: proposed study protocol

Table 5 Exposure predictions for different strata.

Variable	Category	n	Coefficient	95%-CI	p-value
Age	young adults (20-34 y)	56	reference	-	-
	adults (35-64)	69	0.77	0.59;1.01	0.06
	retired people (>64)	6	0.75	0.39;1.42	0.37
Gender	Female	74	reference	-	-
	Male	57	0.93	0.72;1.20	0.58
Place of residence	Urban	76	reference	-	-
	Suburban	55	1.27	0.97;1.66	0.08
Ownership of mobile phone	Yes	119	reference	-	-
	No	12	0.70	0.44;1.11	0.13
Ownership of cordless phone	Yes	79	reference	-	-
	No	52	0.91	0.68;1.21	0.51
Ownership of W-LAN	Yes	50	reference	-	-
	No	81	0.95	0.72;1.25	0.72
Socio economic status	Low	21	reference	-	-
	Middle	17	0.87	0.54;1.39	0.55
	High	93	1.10	0.77;1.58	0.59

Coefficients of a multiple loglinear regression model using data from a Swiss RF-EMF population survey [15]. This model allows predicting average RF-EMF exposure in different population strata
Intercept of the model: 0.11 mW/m² (95%-CI: 0.08-0.17) (exposure during the day of a female person aged 20-34 living in an urban environment, owning a mobile phone, a cordless phone and wireless LAN at home, with the lowest socioeconomic status).
To calculate total exposure of a woman with the same characteristics but who does not own a mobile phone, the value has to be multiplied by 0.70 resulting in an exposure of 0.08 mW/m². Note that this is only an example to demonstrate the principle of an exposure prediction model. Lack of significance of coefficients for potentially relevant parameters may indicate that a larger sample size is needed for this type of exposure prediction model.

ISSN: 1476-069X