A 1946 study (PDF) that found that covering pipes with asbestos is not a dangerous occupation is facing a call for retraction.
The study, published in the Journal of Industrial Hygiene and Toxicology, examined four shipyards, including two operations of the US Navy, that were using high quantities of asbestos. It aimed to measure the safety of using asbestos to build warships.
The study was conducted by US Navy physician Walter E. Fleischer, Harvard University professor Philip Drinker, and other researchers. Their analysis has been heavily cited, particularly in legal documents by attorneys defending asbestos manufacturers, and came up in front of the US Supreme Court as recently as 2018.
The 1946 analysis has drawn criticism over the years. Now it’s facing a call for retraction, but editors running the modern version of the publication note that the study’s authors are deceased, leaving no one to answer questions about it.
In 2024, the US Environmental Protection Agency banned the last form of asbestos that had still been in use in the US over concerns that exposure can lead to cancer. The Asbestos Disease Awareness Organization links the chemical to around 40,000 deaths in the US each year.
Richard J. Kelly, a retired occupational health specialist, published a critique of the 1946 study last month that calls for its retraction. In his commentary in New Solutions: A Journal of Environmental and Occupational Health Policy, Kelly notes that the analysis adopted a sampling method that wasn’t the standard used in prior epidemiological studies. Those studies found a link between asbestos exposure and risk.
Fleischer and colleagues “used a method that wasn’t ideal, but then, more importantly, they just completely changed the rules for the analysis,” Kelly tells C&EN. “I was stunned that something like that had ever been published and even more stunned that something like that had never been retracted.”
For instance, Kelly notes that the gold-standard method in the field requires a researcher to place all the material under a microscope and count both fibrous and nonfibrous particulate matter. In the 1946 study, however, up to 99% of the material the authors reported was nonfibrous, Kelly says. “They had no basis for doing it,” he says of the failure to count fibrous matter. “You can’t change the rules just because you feel like it.”
The 1946 study also mostly included participants who had worked in the shipyards only for a few years — usually not long enough to develop asbestos-related diseases, Kelly notes. “Whether they ignored that because they were ignorant of that, or they had some other motivations, I don’t know,” he says.
Barry Castleman, an environmental consultant and public health expert on asbestos disease, says the 1946 paper still comes up in court cases involving people who worked with asbestos insulation in shipyards. But he says the paper hasn’t been persuasive to juries deciding cases about health effects.
“It’s kind of late to be calling for a retraction of it. Whatever harm could be done with that paper has largely been done,” Castleman says.
The Journal of Industrial Hygiene and Toxicology was run by the American Industrial Hygiene Association until 1949. After a number of mergers, the current iteration of the journal is published by Taylor & Francis.
“It is unclear who owns the archive of that journal and would therefore have the right to retract any article in it,” a spokesperson for Taylor & Francis tells C&EN by email.
“While it is not our policy to automatically rule out investigating an article where the authors have since died, such an investigation would inevitably be more difficult,” the spokesperson adds. “This particular case would be further complicated by the fact that the article was published 80 years ago.”
In such cases, the spokesperson notes, Taylor & Francis considers the fact that other papers in the scholarly record have already addressed problems with a questionable study.
