The authors respond:
We thank Pira and colleagues1 for their interest in our article2 on extremely low-frequency magnetic fields (ELF-MF) and neurodegenerative diseases. Their main questions were related to: (1) the correctness of combining contributing cause with primary cause of death; (2) the potential for confounding, particularly from metals, as a result of the characteristics of the study population; and (3) the definition of exposure groups.
Pira and coworkers1 argue that >0.530 μT (90th percentile) is not a relevant limit for “very high exposure.” It should be further clarified that the μT value corresponds to the geometric mean of the mean values observed within an occupation, and that a majority of the study subjects in this group had estimated mean values above 1 μT. The geometric mean of mean values for welders was estimated at 1.12 μT based on 13 independent measurements.2,3 When interpreting the μT level, it is also important to realize that the estimates refer to workshift averages of intermittent exposure, including considerably higher peak values. According to the reference distribution of ELF-MF exposure from a population-based sample of male workers, fewer than 5% had mean values exceeding 0.53 μT. Considering the origin of the reported μT levels, we maintain that “very high exposure” is an adequate expression.
The justification for combining primary and contributing cause of death was addressed in the discussion of our paper.2(p.424) Because there was no clear association between ELF-MF exposure and overall mortality, the probability of detecting Alzheimer's disease as a contributing cause was not related to exposure. When writing the death certificate for individuals who suffered from Alzheimer's disease, it is usually difficult to decide whether the disease had a direct link to the final cause of death (often cardiovascular or pulmonary complications), and therefore Alzheimer's disease will be recorded more often as a contributing cause than as primary cause. Pira et al.1 refer to our relative risk (RR) of 0.94 (95% confidence interval [CI] = 0.10–8.57), based on one single case with the disease as primary cause in the very high exposure group. We find it more accurate to emphasize the results showing increasing relative risks for combined causes across the exposure groups with an RR of 4.04 (1.40–11.6) in the very high exposure group.
Pira and coworkers1 suggest that workers inhaling “metal fumes containing definite neurotoxicants such as lead, aluminum and manganese, that have well established neurotoxic potential” explain our results. Furthermore, they suggest that this type of exposure explains the comparatively large number of cases of amyotrophic lateral sclerosis (ALS). The large number of ALS cases in the cohort relative to the number of Alzheimer's disease cases is mainly the result of the age and sex distribution. Mean age at inclusion was 35 years, and just a few individuals were above 65 years at entry.
The cohort based on engineering industries was created to reduce the potential for confounding from risk factors prevalent in the traditional welder environment. Furthermore, the suggestion of an exposure–response relationship for ALS and Alzheimer's disease should reduce the possibility that confounding explains the association. Nevertheless, we have recognized the possibility of residual confounding—also from metals. Based on the restricted study population, the results of the tests for trend and the consistency with previous studies involving ELF-MF-exposed occupations from other work environments, we said that “It could be that exposure to metals may affect the study results by confounding, a question that, along with the possibility of interaction between metal exposure and ELF-MF, should be further explored in future studies.”2(p.424) We have not neglected or dismissed a possible role of other risk factors, but our focus was on ELF-MF.
1. Pira E, Discalzi G, Herroro Hernandez E. Occupational exposures and neurodegenerative diseases [Letter]. Epidemiology
2. Håkansson N, Gustavsson P, Johansen C, Floderus B. Neurodegenerative diseases in welders and other workers exposed to high levels of magnetic fields. Epidemiology
3. Floderus B, Persson T, Stenlund C. Magnetic field exposures in the workplace: Reference distribution and exposure in occupational groups. Int J Occup Environ Health