Majia here: There are two primary risk models used to regulate risks:
1.) Cost-benefit Analysis
2.) The Precautionary Principle
US government agencies employ both of these models. The EPA is somewhat more likely to adopt the precautionary principle, whereas the FDA primarily uses cost-benefit-analysis
(although recently the EPA seems to have completely abandoned the precautionary principle).
Unfortunately, when the US Toxic Substance Control Act was passed in 1976 (enforced by the EPA) it grandfathered in approximately 62,000 chemicals with no testing or review (see http://www.pbs.org/now/shows/412/Exposed-Toxic-Chemistry.html)
Too often today, the burden of proof for regulating older chemicals resides with those who claim the chemicals are dangerous (see Ashford http://dspace.mit.edu/bitstream/handle/1721.1/1587/Vienna.pdf?sequence=1)
Thus, despite some historic applications of the precautionary principle, we today have a backwards regulatory system that too often presumes chemicals are safe until proven "guilty"
[Wikipedia: The burden of proof (Latin: onus probandi) is the obligation to shift the accepted conclusion away from an oppositional opinion to one's own position" http://en.wikipedia.org/wiki/Legal_burden_of_proof]
We saw this broken regulatory system recently with the FDA's decision not to regulate BPA.
Given the dominance of cost-benefit-analysis in risk assessment and regulation, scientists must assign economic costs and benefits to environmental harms if they aspire to have identified risks regulated.
Here is an example of an effort to assign economic costs to environmental harms.
It may seem rather cold and calculating to assign numeric values to carcinogens that cause brain damage in children... but in the absence of these calculations policy makers will not regulate....
Citation: Holtcamp W 2012. Brain Tax: Estimating the Population-Level Impact of Environmental Chemicals on IQ Scores. Environ Health Perspect 120:a165-a165. http://dx.doi.org/10.1289/ehp.120-a165b
ABSTRACT [complete abstract excerpted]
ABSTRACT [complete abstract excerpted]
The author assessed the population impact of various risk factors, including exposure to environmental chemicals, on Full-Scale IQ (FSIQ) scores. FSIQ values are a useful measure for such an assessment because past studies have shown that these scores correspond to work productivity and associated lifetime earnings. FSIQ declines therefore can be used to estimate economic consequences for various risk factors.
The data necessary for such an analysis, including information on both prevalence of exposure and effects, were available for only 3 environmental chemicals: lead, methylmercury, and organophosphate pesticides. The loss in FSIQ points was also calculated for a variety of medical conditions, neurodevelopmental disorders, traumatic brain injury, failure to thrive as a result of neglect and abuse, and iron deficiency. Loss of IQ points attributable to each risk factor was estimated based on meta-analyses or pooled analyses of existing data, then generalized to the 25.5 million children aged 0–5 years estimated to live in the United States, based on the prevalence of exposure to each risk factor.
Preterm birth resulted in the largest population health burden, with estimated losses of more than 34 million IQ points populationwide, followed by lead exposure at 22.9 million points lost. Organophosphate pesticides and attention deficit/hyperactivity disorder both resulted in estimated losses of more than 16 million points, iron deficiency in 9.4 million points, pediatric bipolar disorder in 8 million points, autism spectrum disorders in 7 million points, brain injury in 5.8 million points, and failure to thrive in 5.3 million points. The remaining risk factors resulted in estimated cumulative IQ losses of fewer than 1 million points each. Interactions between factors certainly exist, although the current study was not able to assess these.
A surprising finding was that a large fraction of the total estimated IQ loss associated with lead was contributed by children in the lower reaches of the blood lead distribution—that’s because so many children have low levels of exposure. But the author estimates that reductions in U.S. lead exposure nevertheless “saved” more than 100 million IQ points between the 1976–1980 and 2005-2006 iterations of the National Health and Nutrition Examination Survey. This suggests that similar reductions in nations where lead exposures remain high could result in dramatic improvements in neurodevelopment on a population level.