Nevermind cancer. Cellphones turn you into a zombie. Which is worse?
It's interesting, but I'm going to pick holes in the study because the results are quite surprising. I'm not going to say wrong
- results are what they are - but it's possible they're not measuring what they think they're measuring.
Sprague-Dawley rats were housed in custom-designed reverberation chambers
Why? Nobody lives in a "reverberation chamber" or goes into one to operate a cellphone. The test model should approximate normal usage conditions as closely as possible.
Actually, it becomes clear why they've done this: they monitor RF power as absorbed power rather than radiated power, so they have to make sure that a known amount of energy is absorbed. While I can see the logic in this, it's so different to what happens in real life that it may induce effects that would not normally occur. The 6 W/kg level is at least 5x higher than everyday exposure. No harm in measuring what happens (unless you're a lab rat) but the results, as they say, should be treated with caution.
All RF exposures were conducted over a period of approximately 18 hours using a continuous cycle of 10 minutes on (exposed) and 10 minutes off (not exposed), for a total daily exposure time of approximately 9 hours a day, 7 days/week.
This never happens, to anybody. Not even close. Again, we can see why they've done it - they wanted to get some sort of outcome within a reasonable timeframe, and rats don't live long anyway. But with both ionizing and non-ionizing radiation, timescales matter. If I zap my hand in a microwave oven for 50 milliseconds at a time once a month for ten years, I'll probably be just fine; the hand will repair whatever (minor) damage occurs. If I zap my hand for 6 seconds, once, I will probably cause irreparable damage to nerves and muscles.
You can actually get a very accurate estimate of exposure from modern smartphones: they all have power-usage monitors that tell you how much of the battery capacity has been used for screen, CPU, RF subsystem, etc. If you know the battery capacity, radiation efficiency, RF poweramp usage time, and typical recharge interval, you know the average RF energy being radiated and can estimate absorption using models. With the call log, you can estimate peak radiation. Emulating typical usage patterns would be more useful, IMO, rather than using very high exposure levels to force a positive result.
Throughout the remainder of the chronic study, no RFR exposure-related effects on body weights were observed in male and female rats exposed to RFR, regardless of modulation (Figures 1 and 2). At the end of the 2-year study, survival was lower in the control group of males than in all groups of male rats exposed to GSM-modulated RFR (Figure 3). Survival was also slightly lower in control females than in females exposed to 1.5 or 6 W/kg GSM-modulated RFR. In rats exposed to CDMA-modulated RFR, survival was higher in all groups of exposed males and in the 6 W/kg females compared to controls (Figure 4)
So, um, data does not support conclusions?
If you look at the "cancer" results, the incidence is so low (1-3 rats out of P=90) that it could not be described as statistically significant - even if, mathematically speaking, some of the observations reach the p<0.05 threshold. Remember, that 0.05 magic number is literally plucked out of the air; it's convention, nothing more. This is real borderline stuff: the authors point out that some of the results are significant and some are not. The fact that these rats all survived longer than controls suggests a statistical anomaly. Doesn't mean the effect doesn't exist - simply that there weren't enough test subjects to be sure either way.
Finally, there is no theoretical reason why low-level RF should cause cancer. Again, that's not to say the results are illusory, but without such a model there's no obvious direction for future research.
I'm not picking fault with the authors here, incidentally. They are scrupulously honest about their method and results in this paper, and they provide an excellent analysis of the stats. It's a very good paper. One just gets the impression they've been told to slant the conclusion in a preferred direction.