Publication bias is rampant. Bad science is everywhere. You have to read the papers. The last reference is indeed worth noting. But even there, look at the main

Publication bias is rampant. Bad science is everywhere. You have to read the papers. The last reference is indeed worth noting. But even there, look at the main finding. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3907669/ https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3907669/ This is a correlational study with multiple comparisons and no statistically significant results. In the only outcome reported between cases and controls, cell phone use was (insignificantly) protective for tumor incidence. As for the case-case comparisons, in the best case those are highly susceptible to confounding, and, contrary to the implication in the abstract, they didn't find anything. https://www.ncbi.nlm.nih.gov/pubmed/29530389 https://www.ncbi.nlm.nih.gov/pubmed/29530389 This is a rat model study wherein rats were exposed to cell-range RF from before birth to death, for 19 hours per day. The positive finding was at 50 v/m , 40% higher than the highest measured value (during an outgoing call)[1] that I could find offhand. So yeah, if you strap your phone to your head and have an outgoing call going all day every day, and your phone's field strength is crazy high, and you're a rat, you might have a problem. https://www.ncbi.nlm.nih.gov/pubmed/28472042 https://www.ncbi.nlm.nih.gov/pubmed/28472042 "No association between mobile phone use and glioma risk was observed across the total population (OR 0.98; 95% CI = 0.88–1.10)..., there was an association between long-term mobile phone use and glioma. However, the analysis was limited by significant heterogeneity between trials and the paucity of data on long term use (10 years)." The funnel plot, an indication of publication bias, does not look good. This is in a reputable journal and appears to be a legit systematic review. Their positive finding can't be dismissed offhand. [1] https://thescipub.com/pdf/10.3844/ajeassp.2009.771.774 https://thescipub.com/pdf/10.3844/ajeassp.2009.771.774