‘Statistical Malpractice’: The ALA’s Attack on Pittsburgh

Member Group : Allegheny Institute

The American Lung Association (ALA) has revived its smear campaign against the Pittsburgh region’s air quality with the release of its latest hyped up “State of the Air” report.

“Fear-mongering” and “disingenuous” is an apropos way to characterize the ALA’s annual exercise in misrepresenting and/or ignoring data to issue dire health predictions, says the president-emeritus of the Allegheny Institute for Public Policy.

“The problem for the ALA is that the health experience in states with widely disparate pollution levels does not match up with its rhetoric,” says Jake Haulk, also a senior advisor at the Pittsburgh think tank (in Policy Brief Vol. 19, No. 19). “Nor do the health statistics and pollution levels match up in Pennsylvania.

“What is in evidence is statistical malpractice,” he says.

It was on April 24 that the lung association issued its 20th-annual report, offering this portentous warning:

“Residents of Pittsburgh and the metro area should be aware that we’re breathing unhealthy air, driven by local emissions, upwind sources and extreme heat as a result of climate change, placing our health and lives at risk,” said Kevin Stewart, the group’s director of environmental health for advocacy and public policy.

But there are four critical lapses in the ALA pronouncements.

“First, it fails to mention that readings in all but one of the region’s air quality monitors measuring particulate matter (pm2.5) concentrations are below the Environmental Protection Agency’s (EPA) maximum annual average guidelines with many well below,” notes Haulk, a Ph.D. economist.

“Second, (the ALA) fails to mention that the EPA limits on pm2.5 have declined massively over the years, dropping from an annual average of 60 micrograms per cubic meter (mcg/m3) of air in 1970 to 15 in 1997 to 12 in 2012,” Haulk says, reminding that Pittsburgh’s air has seen equally large declines in pm2.5 concentrations.

Third, “and most telling,” the think tank scholar says, is that the American Lung Association violates EPA rules in calling the region’s air some of the worst in the nation based on monitor readings near known generators of particulate matter.

Simply put, per the EPA, “The EPA monitoring data represent the quality of air in the vicinity of the monitoring site and, for some pollutants, may not necessarily represent urban-wide air quality.”

Yet, the American Lung Association, wielding a large, overloaded and sloppy brush, regularly violates that admonition.

“In short, neighborhoods in Allegheny County and other surrounding counties are being tarred with the brush of measurements at a handful of monitors with readings above an annual average of 11 mcg/m3 and only one out of compliance (above 12) in Liberty Borough,” Haulk says.

“This unfair maligning of areas away from the offending monitors, by lumping them in with the areas in proximity to those monitors, is detrimental because of the taint it puts on them and almost certainly hurts their ability to recruit new businesses.”

Then there are those lung association claims of potential local adverse health effects – lung cancer among them — that simply are not supported by the data.

To wit, the Centers for Disease Control (CDC) pegs the U.S. death rate from lung cancer at 41 per 100,000 people. For Pennsylvania, it’s 43 per 100,000. Calculated by county, age-adjusted deaths per 100,000 ranged from 32 per 100,000 in Snyder County to 59.5 in Forest County.

But none of those counties has air quality monitors, typically placed in areas with known or suspected pollution problems.

“Allegheny County, with its all-monitor average of 9.1 mcg/m3, had a lung cancer death rate of 48 per 100,000,” Haulk says. “Across the state in Philadelphia, the eight-monitor average was 8.6 mcg/m3 and the age-adjusted lung cancer death rate was 55.7 per 100,000.”

Similar and even more striking incongruities can be found in heavily polluted California, with only 30 aged-adjusted lung cancer deaths per 100,000 people.

Then there’s relatively lower-polluted Maine, which has higher death rates from lung cancer (50.3 per 100,000) than Pennsylvania (43), California (30) and the nation (41).

“Obviously, prevalence of lung diseases in specific geographic areas are the result of far more determinative factors than simply air quality readings at EPA monitors,” Haulk says.

“The extreme examples of California – with its overall worst metro area air quality yet better than national lung disease statistics – and Maine’s very good air quality — but bad lung disease statistics — suggest strongly that the ALA take a closer look at is long-running assault on Pittsburgh air and warnings about health effects,” the think tank scholar says.

“The failure to do more investigation into the differences in lung disease prevalence that do not track as it claims with pollution levels is unprofessional.”

Colin McNickle is communications and marketing director at the Allegheny Institute for Public Policy ([email protected]).