
225Brain Tumors and Mobile Phone Use: The Case–Control Approach
to be of poor quality; subjects for whom proxies provided the responses; subjects who
reported implausibly high amounts of mobile phone use; use of matching and conditional
analysis; and adjustment for possible confounders. For glioma, results from the sensitivity
analyses were generally similar to those from the primary analysis. Most ORs were well
within the 95% CI of the OR from the main analysis. When subjects with high reported
use were included, but with use truncated at 5 hr/day, the OR was hardly affected. When
subjects who reported >5 hr call time per day were excluded a ...