Guilt by association isn't a very reliable standard of proof for convicting people of crimes, so it's not used. Anti-chemical activists often get away with an even more dubious standard of proof to convict chemicals of harming people: guilt by implication. 

Anti-chemical activists at the National Environmental Trust and Physicians for Social Responsibility just released their latest alarmist report, Polluting Our Future: Chemical Pollution in the U.S. that Affects Child Development and Learning

The groups estimate about 360,000 children — 1 in every 200 — suffer from developmental or neurological problems caused by an estimated 24 billion pounds of chemicals released into the environment every year. 

Polluting Our Future calls for a number of policy changes to "lower the risks from these substances, [including] pre-market screening of new chemicals, mandatory testing of existing chemicals, product labeling, better pollution reporting and toxic chemical controls for electric power plants." 

These recommendations might be justified if the case can be made that chemicals in the environment are causing the alleged child health problems. But does Polluting Our Future satisfy its burden of proof? 

There is no question that a multitude of chemicals compounds are released into the environment every day through industrial and consumer activities. Exposure to these chemicals — usually at very low levels — is unavoidable. 

There is also no question that a small percentage of children suffer a variety of developmental and neurological problems. But the key question is: Are these exposures related to the health problems? 

In trying to link chemicals with child health problems, Polluting Our Future relies on an estimate in the National Research Council report "Scientific Frontiers in Developmental Toxicology and Risk Assessment" that 3 percent of all developmental defects — mostly birth defects, but also including low birth weight, mental retardation, miscarriage — are caused by exposure to "toxic chemicals and physical agents." The 3 percent statistic is then applied to a Census Bureau estimate that nearly 12 million children under 18 (17 percent of children) suffer from one or more developmental, learning or behavioral disabilities. 

Problems abound with the report use of these estimates, though. 

First, the Census Bureau estimate of 12 million children with developmental and neurological disabilities cannot be taken at face value. The Census Bureau data was collected by polling parents on whether their children have developmental or neurological disabilities. The responses were not validated or verified and such "self-reported" data are notoriously inaccurate. 

For example, a recent University of Minnesota study reported that in a survey of 1,059 subjects, 40 percent erroneously reported they had suffered heart attacks. If people can't accurately report whether they've had a heart attack, how can they reliably diagnose developmental and neurological effects that are often more subtle? 

Next, the NRC report's "3 percent" statistic, even if accurate, only applies to developmental — not neurological — effects. So based on the NRC report's estimate of 120,000 annual live-born infants with major developmental defects, about 3,600 cases per year may be associated with exposure to "toxic chemicals and physical agents" — a far smaller problem than claimed in Polluting Our Future

The "toxic chemicals and physical agents" related to the "3 percent" estimate include such exposures as cigarette smoke, alcohol, and drugs (prescribed and illicit). But Polluting Our Future misleadingly implies that these defects are due to industrial chemicals in the environment. 

The only environmental pollutants identified in the NRC report as associated with developmental problems in humans are lead, mercury and PCBs. And even with those, the NRC admits, uncertainty abounds. There is no hard evidence that typical, low-level exposures to these chemicals in the environment are associated with any health problems whatsoever. 

Polluting Our Future, in contrast, goes on to list 323 chemicals as known or suspected developmental and neurological toxicants. These designations, though, are derived primarily from experiments in which laboratory animals are essentially poisoned with chemicals at high doses that humans would never experience. 

A fair read of the NRC report indicates that, even if some chemicals in the environment are responsible for some developmental problems, the occurrence is so rare that it cannot be reliably identified or measured — despite more than a century of widespread use of and exposure to a multitude of chemicals. 

Considering that Polluting Our Future is sounding the alarm about 24 billion pounds of chemicals released into the environment annually, the report virtually debunks itself. If chemicals in the environment are so dangerous, why are the alleged health effects unidentifiable and unmeasurable? Given the tremendous amount of scientific research on the topic, the absence of evidence can't be blamed on a lack of effort. 

Polluting Our Future is premised on the familiar, but faulty syllogism of the anti-chemical alarmists: Chemicals are in the environment. Children have health problems. Therefore, chemicals cause the health problems. 

This syllogism has intuitive appeal to many in the public protective of its children and already made suspicious of industrial chemicals by two decades worth of media hype such as this week's CNN headline, "We're poisoning our kids, toxins report says." 

The report rushes to convict chemicals. But the facts don't. Perhaps that's why the National Environmental Trust and Physicians for Social Responsibility put their conclusion in the report's subtitle, "Chemical Pollution in the U.S. that Affects Child Development and Learning." Once you get past the cover, though, the accusation just doesn't stand up. 

— Steven Milloy is a biostatistician, lawyer, adjunct scholar at the Cato Institute and publisher of Junkscience.com.