University Hospital, Örebro, Sweden, email@example.com (Hardell)
National Institute for Working Life, Umeå, Sweden (Mild)
To the Editor:
Lönn et al1 reported an association between use of cellular telephones and acoustic neuroma confirming our previous results.2–4
Their study area encompassed 3 medical regions in Sweden for cases but controls were from 4 regions (with the addition of the Umeå region, as part of the larger INTERPHONE study). Thus, the population base was different for cases and controls. According to the Swedish Cancer Registry, 122 histologically verified cases were reported in the 3 medical regions, but 160 cases were included in the study (M Talback, Swedish Cancer Registry, personal communication). Furthermore, analog phones have been more common in the Umeå region. Assessment of exposure was not uniform for cases and controls and was not done blinded as to case or control status.
Table 2 is confusing since no explanation is given for the differing numbers of cases and controls among the variables. In particular, it is not clear how there could be 85 controls with regular use of analog phones but a total of 86 in the duration-of-use categories. Table 2 gives the total number of cases as 148 and the number of controls as 604. In Table 3, the numbers are 138 cases and 601 controls, although the footnote states that only 1 case (and 3 controls) was excluded; thus 9 cases are not accounted for. Furthermore, as in Table 2, some of the subgroup tallies are less than the total number.
From Tables 2 and 3 it can be calculated that of the cases, 3 (3%) used both ears compared with 33 (9%) of the controls. In these calculations the authors have used a method similar to that first described by us.2–4 However, data should have been presented for ipsilateral, contralateral, and equally both sides. The unexposed group should include only cases and controls with no use of cellular or cordless phones.
Only 58 (39%) of the cases were histologically verified, as compared with 100% in our studies. Clinical observation is 1 option for patients with acoustic neuroma, and so our studies are not comparable as to tumor induction period.
University Hospital, Örebro, Sweden, firstname.lastname@example.org
Kjell Hansson Mild
National Institute for Working Life, Umeå, Sweden
1. Lönn S, Ahlbom A, Hall P, et al. Mobile phone use and the risk of acoustic neuroma. Epidemiology
2. Hardell L, Hallquist A, Hansson Mild K, et al. Cellular and cordless telephones and the risk for brain tumours. Eur J Cancer Prev
3. Hardell L, Hansson Mild K, Carlberg M. Further aspects on cellular and cordless telephones and brain tumours. Int J Oncol
4. Hardell L, Hansson Mild K, Sandström M, et al. Vestibular schwannoma, tinnitus and cellular telephones. Neuroepidemiology
© 2005 Lippincott Williams & Wilkins, Inc.