Models of Exposure and Risk
Radiation, specifically ionizing radiation is, by its nature, harmful to human health (and that of most living things). There's a no-threshold understanding of radiation safety that is well established in related policies for medicine, nuclear power, national and international guidelines. Most noticeably, it's what defines evacuation zones around nuclear power plants in the event of a disaster--like the wide area around Fukushima Daiichi that was evacuated in 2011. In an effort to avoid unnecessary exposure to radiation, this wider evacuation was responsible for death and hardship of its own. As a result, some believe the evacuation zone caused unnecessary hardship(1) while others think it was the right response; some former residents are still concerned about returning(2). There was even public fear of contaminated waters in other nations. So what was the risk from radiation and what were people still exposed to? These decisions are guided by how we see radiation; two perspectives assessing risk being the Linear No-Threshold or radiation Hormesis models. This reference doesn't provide exact risk information from exposure levels, as the scientific understanding is constantly being refined; but will provide the underlying concept and reasoning for different models along with external links to scientific references with more information.
As different forms of ionizing radiation pass through an object molecules are ionized, disrupting cells and causing damage to an exposed person's DNA. The damage is measured in the amount of energy left behind per weight of living tissue as doses (such as Gray [1 Joule / 1 Kilogram] or Rad [100 ergs / 1 gram]) or dose equivalents (Sievert, equal to 1 Gray of high penetration x-ray, or REM). The reason for a dose equivalent (and why high penetration x-ray is used as a benchmark for the equivalent) is because, while all ionizing radiation causes damage, not all forms of such radiation produce the same effect.
There are several ways we can be exposed to radiation in our daily lives; some of it is natural, background radiation, and some of it is human caused. In the United States, these natural and human caused sources are actually evenly balanced, according to an NRC analysis(3). The average person is exposed to 620 millirem (mrem) (NRC) or 2.4 milliSievert (UNSCEAR(4)). But radiation exposure varies widely based on activity and residence. For example, Denver has one of the highest levels of natural background radiation--to the point that living there for just two days would expose someone to a dose equivalent of 1 mrem(5). You'd be exposed to the same dose equivalent from one coast-to-coast flight or living near a normally operating nuclear power plant for a year. To see how you compare to the average, the NRC(6) does have a
To convert from REM to Sievert or back, or to better understand any of the measures mentioned throughout this page, a unit conversion website such as the
As I mention above, ionizing radiation causes damage to any living cells exposed to it. For each cell that damage can have different outcomes; according to the NRC "radiation may have one of three biological effects, with distinct outcomes for living cells: (1) injured or damaged cells repair themselves, resulting in no residual damage; (2) cells die, much like millions of body cells do every day, being replaced through normal biological processes; or (3) cells incorrectly repair themselves, resulting in a biophysical change." Of these three outcomes, it's the second and third which are most concerning and publically addressed (and for good reason). Changes to cells and the potential for cell death beyond natural levels, are what underlie the horrifying consequences most feared, from large radiation doses. The most clear and devastating cases, like that at Tokai-Mura in 1999 or the several other, equally disastrous exposures, make clear what's at risk with radiation exposure. But, these kinds of events don't expose large populations, and are rare when existing standards are followed. So while the effects of high dose exposure are clear, what is less understood are the effects of low dose exposures.
For an understanding of how people are more generally affected, from even the most minute exposures, the Linear No-Threshold model assumes that any exposure will always increase risk. The base of this assumption comes from the largest groups of people exposed to a wide range of radiation: survivors of the atomic bombs used in Japan. Mortality rates and cancer incidence for survivors were used to verify the possibility that any level of increasing radiation exposure meant an increasing risk(9)(10). Such studies use information on the doses received by each survivor and their health to create a reliable reference across exposure levels. The study by Gilbert found a 1% increased risk of cancer for those who received a low (0.1 Gy) exposure, with a rising increased risk supporting the LNT model. The risk is not only higher for larger doses, but also for younger people. The general dose-risk relationship for the LNT model, including that for three other radiation exposure models, can be seen in this graph from a 2012 UNSCEAR report(11).
When intentional exposure is considered (for an x-ray or similar medical procedure), the FDA lays out a clear guideline using the reasoning of the LNT model for its recommendation. "In the field of radiation protection, it is commonly assumed that the risk for adverse health effects from cancer is proportional to the amount of radiation dose absorbed... A CT examination with an effective dose of 10 millisieverts (1 mSv = 1 mGy in the case of x rays) may be associated with an increase in the possibility of fatal cancer of approximately 1 chance in 2000..." The risk is weighed against all causes for screening and the impact of less accurate tests(12). Because both undergoing or not undergoing any medical procedure has its risks, your doctor can help you make the best case for your health when considering such procedures.
The model offers a definite exposure-response course of action: any additional exposure, no matter how small, is going to cause more cancer and cancer related deaths. It's straight-forward and allows for clearly defined evacuation procedures like those employed around Fukushima Daiichi. The result, according to an UNSCEAR study following the tragedy, was that "the evacuations greatly reduced (by up to a factor of 10) the levels of exposure that would otherwise have been received by those living in those areas(13)." According to the LNT model, that will have saved evacuees from a much higher risk of cancer, death, and even genetic defects than they now face. There was, as the same UNSCEAR study states "evacuation-related deaths and the subsequent impact on mental and social well-being (for example, because evacuees were separated from their homes and familiar surroundings, and many lost their livelihoods)." But, if radiation exposure didn't result in risk in such a linear fashion, or if there was a threshold at which low levels of radiation, evacuation might have been unnecessary for some of the least exposed victims.
Though we know ionizing radiation causes damage and the potential outcomes of that damage, it's also a natural force. Just as Denver has a larger background radiation than similar places, a lot of common foods have naturally occurring radiation (a portion of the potassium in bananas, for example(14)). At the same time, natural cell damage is massive compared to that of a low radiation dose. According to UNCEAR data, 1 mSv per year radiation induced DNA damage is 6 million times lower than natural, spontaneous DNA damage (Cuttler 2012). Because it's natural and unavoidable to some level, there are also natural responses to radiation damage. Two models, radiation hormesis and linear threshold, take into account the possibility of natural cell healing and resilience. As one study describes: "the LNT model considers only the initial interaction of radiation that causes the oxidative damage and mutation. The LNT model completely ignores the body's defensive adaptive responses that may be triggered by the low dose radiation(15)." The idea that an adaptive response could be triggered by low dose radiation is key to hormesis, which states that this adaptive response means exposed individuals will have a noticeably lower incidence of cancer from low doses.
Natural DNA damage is similar to the worst implications of ionizing radiation, so the natural processes for repairing it could also be responsible for repairing damage from ionizing radiation. But as a 2006 Biological Effects of Ionizing Radiation committee (BEIR) report cautions that "oxidative damage is much more complex..." and goes on to explain that understanding of a hormetic effect is not clear(16). With that in mind, 1 cGy (10 times as much as 1 mSv, if we're talking about x ray) of radiation causes approximately 100 measurable DNA changes per cell--which sounds like a lot for just one cell. But there are as many as 10,000 measurable DNA changes in each cell every hour(17). Cuttler (2015) notes that there are more than 150 genes responsible for the activation of the natural protection from or, repair, replacement and removal of cells when damaged by natural endogenous changes or other causes. He asserts that "a low radiation dose causes stresses, which turn on genes at specific radiation dose thresholds. These genes increase the activities of protection systems." He goes on to note that these genes turn off or fail at too high of a radiation dose, so the potential benefit could only be observed at low doses. Two likely low doses that he says would cause more hormetic benefit are 50 rem (0.5 Gy) for a short-term total dose, and a lifetime dose rate of about 0.7 Gy per year.
This makes evacuation around a damaged nuclear reactor, such as around Fukushima, a little more uncertain. This isn't to say that those immediately around the plant shouldn't have evacuated. But evacuation may have been a needless hardship for those in the farthest parts of the evacuation zone, who weren't as significantly affected by the tsunami and earthquake and were only likely to receive low doses of radiation. One source states that two-thirds of evacuees received an external radiation dose of 1 mSv, 98% below 5 mSv, and 10 people exposed to more than 10 mSv (each of these dose equivalents were in reference to normal international exposure limits across an entire year(18)). Conversely, another analysis states that as much as 95% of the population around Fukushima only received around 1 mSv(19). Compared to the acute hormetic dose Cuttler mentions (.5 Gy) the 10 people who received the most radiation were still below harmful dose levels by an order of magnitude. But these doses are after evacuation, which UNSCEAR noted likely reduced doses received by a factor of 10 (note 13). In addition, the uncertainty of what exposure levels may have been should be taken into account. The World Nuclear Association report (note 18) mentions that "people living around the plant were unlikely to exceed 30 mSv/yr in the first year. This was based on airborne measurements between 30 March and 4 April, and appears to be confirmed." As a yearly dose, this would still be less than half of the value Cuttler mentions.
It's easier, however, for someone who doesn't live in the area of such a disaster to say that evacuation was unnecessary. The situation could have been far worse, and hormesis may not be as readily achieved in humans as proponents believe. But a better understanding of nuclear power and radiation in medicine, by knowing how their most dangerous sides are understand, can help us make more informed decisions about our health and environment.