Flawed Analysis Blocks Official Release of CDC Study

Outdated data fails to reflect subsequent remediation


Hold that call to Erin Brockovich. Allegations that the U.S. Centers for Disease Control and Prevention covered up a damning environmental report says less about science — in this case, pollution in the Great Lakes basin — than the need to politicize and point fingers.

According to the abandoned study, "Public Health Implications of Hazardous Substances in the Twenty-Six U.S. Great Lakes Areas of Concern,"[1] the air near the Saginaw River is thick with paint strippers[2], pesticides[3] and carbon tetrachloride, a compound once used in fire extinguishers. The soil and water are worse. Decades of industrial pollution have fouled them with toluene, which is used to make nail polish, and naphthalene, a key ingredient in mothballs.

And it isn’t just the Saginaw River. The study identifies 71 waste sites that could increase the risk of breast, colon and lung cancers in two dozen communities, including Detroit, Chicago and Milwaukee.

Yet a close read of the 400-page study reveals that much of the data is old, drawn from reports filed in 1982 and 1996 by the CDC’s Agency for Toxic Substances and Disease Registry. Several of the spill locations, including the E.I. DuPont de Nemours plant in Muskegon County, the Auto Ion Chemicals plant in Kalamazoo County, and the Berlin and Farrow Superfund site in Genesee County[4], have since been remediated and deleted from the Environmental Protection Agency’s National Priorities List.[5]

"It’s not an epidemiological study," said Hugh McDiarmid, spokesman for the Michigan Environmental Council. "It paints in some pretty broad brushstrokes … If people are going to use it to say the contamination in the Great Lakes is giving people cancer, then that’s a stretch."

The CDC never finished the report. In February 2008, a draft copy was leaked by the Center for Public Integrity, a nonprofit watchdog group based in Washington, D.C.[6]

The center alleged that the CDC blocked publication of the report and demoted Christopher De Rosa, the scientist who oversaw the work, when he pushed for its release.

The CDC responded that the science was flawed.[7]

The media smelled a scandal. The story was picked up by The Detroit News[8], The Washington Post[9] and the Milwaukee Journal Sentinel.[10] The headlines took a side: "Great Lakes Health Report Withheld by Agency" and "Delay of Report Is Blamed on Politics."


The author of CPI’s piece was Sheila Kaplan, a journalist whose work has appeared in Mother Jones, U.S. News & World Report and on the PBS series "Frontline."

Her study of the CDC report was funded in part by the Nation Institute, a decidedly partisan nonprofit based in New York.

Kaplan did not return calls for this article.[11]

However, some members of Michigan’s environmental-science community questioned the link Kaplan made between the CDC report and cancer rates across the Great Lakes.

"Often when you come out with reports, it’s a lot of information to break down in a one-page press release," said Robert McCann, a spokesman for the Michigan Department of Environmental Quality.

"That’s usually the biggest issue people have when it comes to pollution: ‘How does this affect my family? What does it mean for our health?’" McCann said. "The trick is to put it in context for them, to expand on it and really explain what it means and how it affects people."

He said the department would review the leaked version of the CDC report.

George Wolff, the principal scientist at the General Motors Public Policy Center,[12] found other weaknesses in the research.

"What bothers me about it is they just show two sets of statistics," he said. "One on the chemicals and one on the health effects. But they don’t relate them. There’s no cause and effect involved.

"They’re just out there," he said. "It’s just a case of co-location of chemicals."


There is some substance to the CDC report. Several of the waste sites it identifies have polluted communities and put residents at risk. Among the worst:

The Shiawassee River in Livingston County
The Cast Forge Company discharged hydraulic fluids into the South Branch of the Shiawassee River from 1969 to 1973. Approximately 2,600 pounds of PCBs[13] were removed from one mile of the river when it was dredged in 1982. Remediation of the floodplain "will mitigate, but not eliminate, the contamination," the CDC study warns.

Dow Chemical Company, Midland County
Waste ponds on the 1,900-acre Dow property emptied into the Tittabawassee River at regular intervals beginning in 1915. A flood in 1986 overwhelmed the company’s wastewater treatment plant and discharged more waste[14] into the river. Tissue cancers among white women living in Midland County were between 3.8 and 4 times higher than the national average between 1960 and 1978, the CDC says. Fish-consumption warnings were issued in 2006.

Allied Paper Inc., Kalamazoo County
This site, which includes a three-mile stretch of Portage Creek, was contaminated with PCBs discharged by the paper-making industry. In 2006 the EPA estimated that 110,000 pounds of PCBs were in the sediment of the Kalamazoo River, which links to Portage Creek. Consumption of fish from the river brought "unacceptable risks to public health," the EPA said in 2006.

Roto-Finish Company, Kalamazoo County [15]
Some 83,000 gallons of industrial waste were pumped into two storage lagoons, which often overflowed beginning in 1960. The soils have been remediated and are being monitored, and the CDC says there is little risk to the community. But a study of the homes of Roto-Finish employees found traces of a suspected bladder carcinogen[16] in vacuum cleaner dust, dryer lint and urine samples.

That’s probably not news to people who live near the sites.

"It’s not really a surprise that we’ve got areas of toxic concern," said Jeff Skelding, the national campaign director for the Healing Our Waters-Great Lakes Coalition. "They’ve been there for a long time. But they need to be cleaned up. Congress needs to act on this soon."

It already has. U.S. Reps. John Dingell and Bart Stupak, both Michigan Democrats, have begun an investigation of the CDC report. They requested the personnel records of De Rosa, the scientist who was demoted, and any record of the requests he made to publicize the study.

click to enlarge
A series of detailed maps created by the CDC labeled “Great
Lakes Areas of Concern” were part of the unpublished report.

"ATSDR’s[17] apparent withholding of this report raises grave questions about the integrity of scientific research at CDC and ATSDR, as well as the treatment of its scientists," the Congressmen wrote in a joint letter.[18] "The validity of the findings of this report deserves a fair and open debate within the scientific community"

That letter, however well-intentioned, reflects a weakness not only in the CDC report, but in the "gotcha" journalism that exposed it: All too often during the environmental review process, as a report passes through state, federal and public-sector checks, science is hijacked for political effect.

There is no doubt that this particular CDC report was flawed. But the response to it — politically charged and spread without benefit of all the facts — was arguably far more harmful.

[1] Available online at www.publicintegrity.org/GreatLakes/index.htm?source=home.

[2] Trichloroethylene, a paint stripper and dry-cleaning solvent that also was used to decaffeinate coffee.

[3] Chlorobenzene.

[4] The 40-acre site near Swartz Creek was polluted by a waste incinerator that operated from 1971 to 1978. It was removed from the National Priorities List in 1998.

[5] Eighty-four Michigan sites are now on the NPL. The list is available online at www.epa.gov/superfund/sites/npl/mi.htm#statelist.

[6] Other recent studies by the group examined prescription-drug lobbying and the influence technology companies have at the Federal Communications Commission. Information is at www.publicintegrity.org

[7] “Significant scientific shortcomings were identified,” the agency said in a March 7 “Statement of Scientific Concerns,” “Health and environmental data were presented in ways likely to be misinterpreted.”

[8] Trowbridge, Gordon. "Great Lakes Health Report Withheld by Agency," Feb. 8, 2008. Available online at www.detnews.com/apps/pbcs.dll/article?AID=/20080208/ METRO/802080374.

[9] Lydersen, Kari. "Delay of Report is Blamed on Politics," Feb. 18, 2008. Available online at www.washingtonpost.com/wp-dyn/content/ article/2008/02/17/AR2008021702186.html.

[10] Rust, Susanne. “Area’s Toxins May be Sickening People,” Feb. 11, 2008. Available online at www.jsonline.com/story/index.aspx?id=716884.

[11] Messages were left at office and mobile phone numbers provided by Steve Carpinelli, the media manager for the Center for Public Integrity. He referred all questions to Kaplan.

[12] Full disclosure: Wolff is on the Mackinac Center for Public Policy Science Advisory Board.

[13] Polychlorinated biphenyls, oily compounds used as industrial coolants or lubricants until 1977.

[14] Polychlorinated dibenzo-dioxins (PCDDs) and polychlorinated dibenzo-furans (PCDFs).

[15] The company manufactures polishing, buffing and grinding machines. Information is available online at www.jobwerx.com/tooling/directory/Roto_Finish_Company_Inc.htm.

[16] 4,4-Methylene bis(2-chloroaniline), which is used as a curing agent for polyurethane.

[17] The Agency for Toxic Substances and Disease Registry.

[18] February 28 letter to CDC Director Julie L. Gerberding.



Frank Bevacqua
International Joint Commission

Steve Carpinelli
Center for Public Integrity

Hugh McDiarmid
Michigan Environmental Council

Robert McCann
Michigan Department of Environmental Quality

Paul Mohai
University of Michigan

Jeff Skelding
Healing Our Waters—Great Lakes Coalition

George Wolff
General Motors Public Policy Center