Funch DP, Rothman KJ, Laughlin JE, Dreyer NA, Chou C-K, Morgan R, et al.
These papers are amongst the few which have attempted to address cellular phone users specifically, from an epidemiological standpoint.
The first paper described the accuracy and feasibility of using telephone company records as a measure of exposure by comparing cellular telephone records with responses from a questionnaire survey to account holders. They found that users' recall of cellular telephone use was closely correlated with company records.
The second paper examined the overall mortality rate in 256,000 phone users. Comparisons were made between portable (i.e. handheld) and mobile (i.e. in automobiles) phone users. The death rates were no greater than the expected mortality rate in the general population. However, there are several weaknesses in the study. The major one is that only one third of the potential population could be studied, since there was inadequate information on the remainder. In addition, the specific causes of death were not available, and the time between exposure and death was short. Further analysis of the data was reported by the authors (Dreyer NA, et al, JAMA 1999;282:1814-1816). They looked at individual causes of death, and found that "the only category for which there was an indication of increasing risk with increasing minutes of use was motor vehicle collisions".
The third paper reviews in detail the difficulties of exposure assessment in research on cellular telephones.
The last paper reviews the methodology employed in the studies, and points out that the 1996 Telecom Act led to the cessation of the process that had been established for data collection from phone customers' records. In fact, "the follow-up was halted before it was complete for several reasons, including a lawsuit" (Rothman KJ, Lancet 2000;356:1837-1840).