An evaluation of trends for mesothelioma mortality in American women: Addressing the content of a recent Morbidity and Mortality Weekly Report (MMWR)

In their MMWR, Mazurek et al. (2022) suggested that, even in 2022, the mesotheliomas that are observed in most persons are due to exposure to asbestos. This belief is often repeated in some published papers ( Baur et al., 2023 ) and news articles ( American Industrial Hygiene Association (AIHA), 2022 ). While it may have been historically accurate for males exposed in the pre-OSHA era, the statement that “Malignant mesothelioma … is almost always caused by asbestos or other asbestiform fibers” ( Baur et al., 2023 ) is out-of-date; certainly when referring to mesotheliomas in 2023 and beyond. When considering the time periods in which significant amphibole exposures occurred and the latency of asbestos-induced mesotheliomas, it is unsurprising the number of asbestos-induced mesotheliomas has plummeted over the last decade ( Price, 2022 ). Price (2022) examined age-adjusted mesothelioma rates (for men and women) in comparison to mesothelioma use in the United States. Although the absolute number of mesotheliomas is higher for women compared to 20 years ago, and it will continue to climb due to increasing population size and life expectancy, the age-adjusted incidence for women has remained flat since the 1970s (and this would likely would be the case for the decades prior) because, by and large, they have not been exposed to asbestos. However, it should be acknowledged that the incidence of mesothelioma in men has strongly correlated to the consumption of asbestos plus latency ( Figure 1 ). When one accounts for the 30–40 years latency for asbestos to cause mesothelioma, Figure 1 supports Price’s conclusions that an estimated 75% of the current mesotheliomas in men are non-asbestos related (personal communications with Bert Price, 2022) and that by 2040, virtually all mesotheliomas will be non-asbestos related ( Price 2022 ). This is consistent with the work of ( Tomasetti et al., 2015 , 2017 ; Tomasetti and Vogelstein, 2015 ) indicating that about 67% of all cancer risk (not just mesotheliomas) is due to random mutations. By definition, cancer is a disease of the genome, meaning it arises from mutations in the genetic material within cells. The vast majority, if not all, of the malignant mesothelioma cases in the populations of women are unknown in origin or are primarily due to age and an accumulation (and lack of repair) of spontaneous genetic mutations ( Moolgavkar et al., 2017 : 58). As the MMWR authors stated, “Increases in total number, but not age-adjusted death rates, suggest that changes in underlying annual age distributions of the population over time are contributing to the observed increases in total mesothelioma deaths in women” ( Mazurek et al., 2022 : 647). Carbone et al. (2023) has noted that “the risk of developing most cancers, including mesothelioma, increases with old age because of the inevitable accumulation of genetic damage in our cells. Therefore, the older the population, the higher the incidence of cancer, including mesothelioma” ( Carbone et al., 2023 : 695). This is due to impaired DNA repair mechanisms for replication errors resulting in an accumulation of spontaneous mutations ( Garraway and Lander, 2013 ; Stratton et al., 2009 ; Tomasetti et al., 2017 ; Vogelstein et al., 2013 ). According to Tomasetti et al., 2017 , “Recent evidence from mouse models supports the notion that the number of normal cell divisions [which corresponds with age] dictates cancer risk in many organs ( Zhu et al., 2016 ).” To illustrate the importance of age on the development of mesothelioma, it has been shown that “Every doubling of age increases the risk of pleural mesothelioma approximately 30-fold and that of peritoneal mesothelioma approximately eightfold” ( Moolgavkar et al., 2017 : 46). Trends in Table 2 show that the death rate in women more than doubles from the age range of 55–64 to 65–74, and age groups over 75 years old have an incidence rate nearly five-fold greater than the 55–64-year-old age group ( Mazurek et al., 2022 : 646-648). Table 2. Number of deaths and age-adjusted death rates for malignant mesothelioma in women in the United States above the age of 25 years organized by 10-year age increments ( Mazurek et al., 2022 ). Age group (years) Number of deaths (%) *Age-adjusted death rate (per 1 million) 25–34 71 (0.6) 0.16 35–44 282 (2.3) 0.60 45–54 781 (6.4) 1.66 55–64 1,857 (15.2) 4.68 65–74 3,203 (26.2) 11.69 75–84 4,018 (32.9) 23.17 >85 2,015 (16.5) 25.10 Total 12,227 (100) 4.59 *Adjusted using 2000 U.S. standard population. Number of deaths and age-adjusted death rates for malignant mesothelioma in women in the United States above the age of 25 years organized by 10-year age increments ( Mazurek et al., 2022 ). *Adjusted using 2000 U.S. standard population. Similar to the MMWR data, the most recent World Cancer Report shows that cancers have been increasing as a percentage of all deaths ( Wild et al., 2020 : 568); primarily because we have made advances with respect to preventing cardiovascular disease through medication and the decreasing percentage of smokers. In short, progress made in eradicating, curing, or preventing many diseases in the nineteenth and early twentieth century has allowed persons to live much longer, and thus, the absolute number of all cancers is increasing. Medical treatment for cancer and other ailments using radiation therapy is a cause of secondary cancers, with mesothelioma being among them ( Attanoos et al., 2018 ). In large measure, the high radiation doses that caused secondary cancers ended by 1990–2000, but the impact of some of these treatments are just now being observed ( Cavazza et al., 1996 ; Sekine et al., 2021 ; Teta et al., 2007 ). It is noteworthy that one would rarely know if it was the treatment or old age that caused a particular mesothelioma in those populations. Unlike radiation, which directly affects DNA, chronic inflammation has been found to be a necessary precursor in the pathogenesis of some cancers and is generally considered the mode of action by which asbestos causes mesothelioma ( Cox, 2019 ; Mossman, 2018 ; International Agency for Research on Cancer (IARC), 2012 ; Thompson and Shukla, 2017 ; Cox et al., 2020 ). The inflammatory response involves the release of reactive oxygen species and immune cells, which can damage the DNA of normal cells, increasing the risk of mutations and, therefore, cancer ( Singh et al., 2019 ). With this understanding, many experts believe that chronic irritation of or around the peritoneum, such as intra-abdominal shunt catheters ( Mujahed et al., 2021 ), Crohn’s disease ( Attanoos et al., 2018 ; Butnor et al., 2017 ), and endometriosis ( Butnor et al., 2018 ; Malpica et al., 2022 ), increases the risk of peritoneal mesothelioma. All of the diseases associated with repeated irritation or cytotoxicity are believed to have a threshold dose ( Weisburger and Williams, 1989 ; Klaunig and Kamendulis, 2007 : 347; Klaunig and Wang, 2019 : 450). Although exposures are likely rare in the United States, erionite and other non-asbestos mineral fibers have gained attention in recent years as they are expected to contribute to the risk of developing mesothelioma via a similar mechanism to amphibole asbestos ( Carbone et al., 2012 ; Stevens et al., 2024 ). It has been known that genetic susceptibility can play a large role in the development of both pleural and peritoneal mesothelioma. This has been discussed at length over the past 10 years by Carbone et al. (2013 , 2019 , 2020) , Testa and Berns (2020) , and others. Having genetic mutations in DNA repair can mean that a person has a greater probability of developing mesothelioma following even low-level exposure to an amphibole asbestos fiber, or the mesothelioma (both pleural and peritoneal) can develop without any exposure to asbestos. For example, it is now known that persons develop mesothelioma simply due to inadequate DNA repair due to mutations of these specific germlines, including young children (without asbestos exposures) ( Argani et al., 2021 ; Brenner et al., 1981 ; Moran et al., 2008 ; Wolff-Bar et al., 2015 ; Mijalovsky et al., 2018 ; Silberstein et al., 1983 ; Nishioka et al., 1988 ; Lin-Chu et al., 1989 ). While research advances occur regularly, it is currently understood that genetic mutations that affect the BRCA1-associated protein 1 (BAP1) is the one most commonly associated with mesothelioma ( Carbone et al., 2013 , 2019 , 2020 ). Napolitano et al. (2016) found that germline BAP1 heterozygosity increased mesothelioma incidence by way of the peritoneal inflammatory response when exposed to low-dose asbestos fibers. Badhai et al. suggests additional mutations (Nf2 and Cdkn2ab, which are also frequently observed in human mesotheliomas), in combination with BAP1, led to the rapid onset of mesothelioma in 100% of mice and BAP1 deletion alone caused mesotheliomas in 1 out of 20 mutant mice (5%) in the absence of asbestos exposure ( Badhai et al., 2020 ). In small sample groups, Testa et al. (2011) and Bott et al. (2011) found between 22%–23% of mesotheliomas in humans had a BAP1 mutation ( Bott et al., 2011 ; Testa et al., 2011 ). For mesotheliomas the authors defined as sporadic, 8% (2 out of 26), in the Testa et al. study had BAP1 mutations ( Testa et al., 2011 ). A later study by Panou et al. determined that 26% of mesotheliomas with no known asbestos exposure had some form of germline mutations. Of these germline mutations found, 25% (6.5% of the total) were in the BAP1 gene ( Panou et al., 2018 ). A more recent study by Carbone et al. reports that 84.3% of BAP1 deletion carriers (295/350 carriers) (all ages) developed some form of cancer and 29.6% of them developed mesothelioma specifically (an even amount of pleural and peritoneal mesotheliomas) ( Carbone et al., 2020 ). Currently, the available human data leans toward the notion that germline mutations (BAP1 and others) may induce mesothelioma independent of asbestos exposure ( Attanoos et al., 2018 ). Patients diagnosed with mesothelioma are rarely tested for germline mutations (personal experience of the authors and personal communications with Dr. Victor Roggli and Dr. Steven Smith, 2024). It is apparent that this testing could lead to an improved understanding of the role of genetic susceptibility and mesothelioma.

According to the CDC, most MMWR articles are not peer-reviewed in the way typical submissions to health-related journals are evaluated for quality ( Shaw et al., 2011 ). They undergo a streamlined “multi-level clearance” within the CDC, which has allowed them to publish on relevant epidemiology and public health issues “as quickly as possible” with a “short rapid report” format ( Shaw et al., 2011 ). This journal, generally intended for health professionals, serves an important purpose for informing both the health professional and the public. It focuses on alerting them to time-sensitive issues such as food and water safety, environmental mishaps and diseases, and the spread of viruses/diseases (including Covid). As a trade-off, sometimes, these articles have been shown to have shortcomings with respect to the accuracy of the content ( Renner, 2009 ; Leoning, 2010 ; Centers for Disease Control and Prevention (CDC), 2010b ; Centers for Disease Control and Prevention (CDC), 2010a ). Asbestos is one of the most extensively studied hazards in the fields of toxicology, occupational health, and environmental science. The idea that asbestos exposures amongst American women in the 2020s is a time-sensitive issue that needed to be communicated to health profesionals and the public “as quickly as possible” seems unsubstantiated. In our view, a proper peer review of the article by Mazurek et al. (2022) may have led to more well-informed conclusions by the readers.

Like others who have raised awareness of the hazards of asbestos, we recognize that asbestos has historically had a catastrophic effect on the health of mostly male workers, dating back almost 100 years. Fortunately, the opportunity to experience cumulative exposures in the United States which might cause these diseases generally ended by the mid-1970s and, surely, not much later than 1980. The virtual elimination of all sales of all forms of asbestos in the United States was a remarkable achievement. The controls implemented after 1972 (when the OSHA asbestos regulation was promulgated) have likely prevented thousands of persons from developing asbestos-related diseases over the past 50 years. A virtual ban on imports in the United States was formally announced early in 2024 ( U.S. Environmental Protection Agency (EPA), 2024 ). The primary reason, if not the sole reason, for the increased incidence of mesothelioma deaths in the United States, especially in women, in the post-OSHA era is increased longevity. The incidence of mesothelioma in women, when corrected for age, suggests that the historical exposure for women to asbestos fibers has typically been, from a large-scale health perspective, virtually the same as background exposure since the 1930s (when considering latency) ( Price, 2022 ). The number of women, if any, that developed mesothelioma as a result of significant amphibole exposures in the factories supplying goods for WWII (1939–1945) or other exceptional circumstances was not sufficient to alter the national age-adjusted incidence rate in women in the SEER database ( Moolgavkar et al., 2009 ; Price, 2022 ; Price and Ware, 2009 ). As noted above, Price and Ware (2009) reported that “The absence of a time trend in age-adjusted female mesothelioma incidence [as seen by the flat rate in Figure 1 ] suggests that most female cases may not be a consequence of asbestos exposure and the female rate may be an estimate of the risk of background.” That was reinforced in a paper by Price (2022) , where he noted that a significant number of the current mesotheliomas in men (Price estimates 75%) are non-asbestos related and that by 2040, virtually all mesotheliomas will be non-asbestos related. This conclusion, when considering a 30–40-year lag from the time of exposure until the disease presents, coincides with government regulations promulgated by OSHA in 1972; soon thereafter, amphiboles were phased out of manufacturing and commercial products, and the restrictions on occupational exposure tightened markedly. By 2040, persons who were potentially first exposed in the early 1970s at age 20 would be around 80 to 90 years old. As scientists, it is important that we correct misunderstandings that have crept into the scientific and lay literature because we run the risk of misleading policymakers to spend resources where there is a good chance of no resulting improvement in the overall public well-being. As Kabak, a researcher at the Albert Einstein Medical school, has warned: “The distortion of information about health risks has consequences and costs … Clearly, resources devoted to a poorly-substantiated threat cannot be devoted to less sensational lines of work, which may prove to be more valuable. Another cost entails the needless anxiety about problems that were miniscule or non-existent … The amplification of hazards diverts attention from real and consequential dangers” ( Kabat, 2008 ) “The distortion of information about health risks has consequences and costs … Clearly, resources devoted to a poorly-substantiated threat cannot be devoted to less sensational lines of work, which may prove to be more valuable. Another cost entails the needless anxiety about problems that were miniscule or non-existent … The amplification of hazards diverts attention from real and consequential dangers” ( Kabat, 2008 ) There are many things to consider when assessing epidemiology data. The topics addressed in this paper; (1) shortcomings in the national databases, (2) fiber type, (3) historical exposures amongst women, (4) the likelihood of take-home and para-occupational exposures, (5) the potential for modern exposures, and (6) alternative risk factors (including genetic susceptability), if not properly considered, can lead to misleading conclusions. This illustrates why it is important to thoroughly consider the relevant variables when conducting these analyses and interpreting the data. This analysis should bring some clarity to erroneous claims that mesotheliomas in women are increasing, allegedly due to asbestos from some unknown source.

It is unfortunate that for the past 100 years, the term asbestos has been used without adequate precision of mineral type. It has brought about numerous misunderstandings in the scientific, medical and legal communities, as well as the public. Asbestos represents six distinctly different fibrous minerals. Chrysotile is the only serpentine asbestos fiber, while there are five forms of amphibole asbestos: amosite, crocidolite, asbestiform actinolite, asbestiform anthophyllite, and asbestiform tremolite ( Gunter, 2010 ; Virta, 2002 ). Each is chemically distinct and has a different potency with respect to the risk for mesothelioma ( Beckett et al., 2023 ; Darnton, 2023 ; Korchevskiy et al., 2019 ). Historically, approximately 95% of asbestos used in the world, and the related exposures, has been to chrysotile asbestos ( Budavari, 1989 : 893). It follows that if there was an opportunity for occupational asbestos exposure for women in the U.S., it would have primarily been to chrysotile and prior to the 1980s ( Hein et al., 2007 ; Loomis et al., 2009 ; McDonald et al., 1984 ). Amongst the amphiboles, amosite and crocidolite were used commercially. Specific to women, one of the known exceptions in which amphibole exposure occurred are the women who worked in certain respirator manufacturing facilities during the years surrounding World War II (WWII) ( Acheson et al., 1982 ; Schipske, 2008 ) or the use/manufacture of crocidolite containing Kent cigarette filters from 1952–1956 ( Longo et al., 1995 ). In addition to mineral type, fiber dimension has long been recognized as a determinant in the potency of asbestos ( King et al., 1946 ). Specifically, it has been known since the early 1970s that long, thin fibers present a greater risk of disease than shorter fibers ( Agency for Toxic Substances and Disease Registry (ATSDR), 2002 ; Eastern Research Group Inc., 2003a ; Eastern Research Group Inc., 2003b ; Lippmann, 1988 ; Stanton, 1973 ; Stanton et al., 1977 ). An 11 member ATSDR panel in 2002 “agreed that there is a strong weight of evidence that asbestos and short vitreous fibers shorter than 5 µm are unlikely to cause cancer in humans” based on “findings from epidemiologic studies, laboratory animal studies, and in vitro genotoxicity studies, combined with the lung’s ability to clear short fibers” ( Eastern Research Group Inc., 2003b ). Asbestos fibers are currently defined by OSHA as particles greater than 5 microns in length with a length to width (aspect) ratio of at least 3:1. In terms of aspect ratios, Korchevskiy et al. (2019) reported that chrysotile (Quebec) had a median aspect ratio of 7.09 (although chrysotile used in the textile industry could be much longer), while amosite (Transvaal) and crocidolite (South Africa, Bolivia, and Australia) were reported to have median aspect ratios of 8.59 (with a standard deviation of 2.4) and between 9.99 and 12.6 (depending on the region), respectively ( Korchevskiy et al., 2019 ). The difference in potency between fiber types for causing mesothelioma is dramatic. Numerous researchers have suggested that the relative potency for malignant mesothelioma between crocidolite, amosite, and chrysotile (the commercial forms) is approximately 500:100:1, respectively ( Berman and Crump, 2003 , 2008b ; Darnton, 2023 ; Garabrant and Pastula, 2018 ; Hodgson and Darnton, 2000 ; Korchevskiy et al., 2019 ). Asbestiform anthophyllite appears to have roughly the same potency as amosite, while Libby amphiboles, actinolite, and tremolite appear to have less than half the potency of amosite ( Garabrant and Pastula, 2018 ; Korchevskiy et al., 2019 ). Thus, when discussing exposures to asbestos fibers, it is important to identify the specific fiber type. Exposure to an equal airborne concentration of chrysotile, amosite, and crocidolite would have significantly different risks of developing mesothelioma. Consistent with these analyses on fiber type potency, Yarborough (2006) concluded that “[e]pidemiological studies investigating mesothelioma risk from exposures of cohort members to chrysotile asbestos fibers not known to be contaminated with amphiboles do not justify a conclusion of causality at this time.” Berman and Crump (2008a , 2003) supported the widely held belief that pure chrysotile may be non-potent for mesothelioma based on the epidemiology data. Darnton (2023) recently published a meta-analysis that included the most up to date occupational epidemiology data. This data is not inconsistent with the theory that pure chrysotile asbestos fibers may have near zero potency for mesothelioma. There continues to be a growing body of evidence that chrysotile does not pose a mesothelioma hazard except, perhaps, at doses greater than those that cause asbestosis (which occurs at a minimum of 25–100 f/cc-years). A “best estimate” no-observed-adverse-effect-level (NOAEL) for typical cohorts with chrysotile exposure has been estimated to be between 250 and 379 fibers/cc-year (f/cc-year) ( Beckett et al., 2023 ). This estimate is based on an epidemiological analysis of 16 cohorts that were exposed to mainly chrysotile asbestos (less than 10% amphibole) ( Beckett et al., 2023 ). Exposures of this magnitude were likely only feasible for persons who worked in an asbestos brake factory in the 1930s through 1950s ( Berry, 1994 : 540), or in an asbestos mine ( McDonald et al., 1980 : 16), raw asbestos processing factory ( Darnton, 2023 ), or as an insulator in a shipyard ( Selikoff et al., 1979 ), all in the pre-OSHA era. NOAELs for amosite and crocidolite appear in the ranges of 2–5 f/cc-years and 0.6–1 f/cc-years, respectively ( Beckett et al., 2023 ). This is, of course, relevant in understanding why asbestos has generally not been a substantial cause of mesothelioma in women in the U.S. at any time.

The Morbidity and Mortality Weekly Report (MMWR) is a weekly series published by the Centers for Disease Control and Prevention (CDC) and is available in the public domain. On the CDC’s website, the MMWR is considered “the voice of the CDC” and reflects the stance of its parent organization ( Shaw et al., 2011 ; Centers for Disease Control and Prevention (CDC), 2022a ). On May 13, 2022, there was a published report in the MMWR titled “Malignant Mesothelioma Mortality in Women – United States, 1999-2020” ( Mazurek et al., 2022 ). The authors of this MMWR relied upon the National Vital Statistic System’s Multiple Cause of Death (MCD) records to analyze trends for this disease in women. According to the CDC’s National Vital Statistics Reports, life expectancy for females in the United States has increased 0.5 years from 1999 (79.4 for females and 73.9 for males; 76.7 overall) to 2020 (79.9 for females and 74.2 for males; 77.0 overall) ( Anderson and DeTurk, 2002 ; Arias and Xu, 2022 ). The MMWR reported that the MCD records showed an increased incidence of 25% in malignant mesothelioma mortalities in women between 1999 and 2020 ( Mazurek et al., 2022 ), and this statement was placed in headlines in the popular press ( American Industrial Hygiene Association (AIHA), 2022 ). While this statement is correct, the more important observation was that the age-adjusted death rate for women remained flat throughout this time period (fluctuating between 4.12 and 4.99 per 1 million from 1999 to 2020) ( Mazurek et al., 2022 ), which suggests that the net increase in mesotheliomas amongst women was due to longer life-spans and not associated with exposures to asbestos. Mesothelioma (certainly pleural mesothelioma) has long been considered a hallmark disease of asbestos exposure in men, but this is not the case in women ( Ilgren and Wagner, 1991 ; Pavlisko and Sporn, 2014 ; Price, 2022 ; Price and Ware, 2009 ; Wagner et al., 1960 ). The exposure-response relationship for asbestos and mesothelioma is believed to be consistent across genders. However, due to historic occupational trends, few women have been exposed to significant cumulative airborne concentrations (doses) of any fiber type of asbestos. As a result, mesothelioma occurrences among women do not correlate with asbestos consumption in the United States, in contrast to the correlation observed with men ( Moolgavkar et al., 2023 ). Relevant to understanding the risk associated with various exposures, there are drastic differences in the potency and the ability of asbestos fibers to cause disease based on the chemical composition and dimension of the fibers ( Darnton, 2023 ; Garabrant and Pastula, 2018 ; Hodgson and Darnton, 2000 ; Korchevskiy et al., 2019 ; Stanton et al., 1977 ; Wagner et al., 1960 ; Yarborough, 2006 ). In order to address the underlying topics relevant to the MMWR article by Mazurek et al., a non-systematic review on the scientific community’s state-of-the-art understanding of mesothelioma incidence and asbestos exposures amongst women in the United States, as well as a critical investigation of the methods and references in the MMWR, was conducted. For purposes of our review, online tools (PubMed, Scopus, and Google Scholar) were queried for various combinations of terms including “mesothelioma,” “women,” “asbestos,” “take-home exposures”, “para-occupational exposure,” “background rates of mesothelioma,” and “background rates of asbestos exposure,” with results current through April 2024. The references cited in various published papers were examined to identify additional publications of merit. Studies of non-United States populations of women were excluded. We discuss almost 200 informative resources in this paper. We concluded that the MMWR should have provided a more thorough discussion of the many factors that need to be considered when assessing trends in the incidence of mesothelioma. In this paper, we address (1) shortcomings in the national database regarding anatomical location and occupational classification deaths, (2) the importance of asbestos fiber type, (3) the historical trends of women’s occupational exposures to asbestos in the United States, (4) the likelihood that take-home and para-occupational exposures have caused mesothelioma in women, (5) the potential for various exposures to asbestos in the post-OSHA era (after 1970), and (6) the risk factors of mesothelioma when asbestos exposure is absent.

The CDC’s Multiple Cause of Death (MCD) record is a database that contains the national mortality data from 1999 to 2020. ICD-10 codes are used to select between 113 different causes of death (which can be used to calculate crude death rates and age-adjusted death rates), and place of residence, age, race, ethnicity, and gender are also recorded for each entry ( Centers for Disease Control and Prevention (CDC), 2022b ). The ICD code was established in 1999, and screening and reporting for mesothelioma has likely increased and improved thereafter. While the MCD does not record occupation, information from the MCD can be combined with industrial and occupational data for analysis. Combined databases like these have been used in a number of papers with respect to mesothelioma ( Bang et al., 2006 ; Mazurek et al., 2017 , 2022 ). Unfortunately, at least for mesothelioma, those records are limited for characterizing incidence rates (as explained in the next two sections). An additional database that can be used to analyze mesothelioma incidence from 1975 to 2018 is the National Cancer Institute’s Surveillance, Epidemiology, and End Results Program (SEER). While the SEER database uses the ICD style coding, it also uses a list of registries that encompasses the entire SEER database and measures the incidence of diseases via patient demographics, primary tumor site, tumor morphology, stage at diagnosis, treatment, and vital status ( National Institute of Health (NIH), n.d ). Neither database considers whether diagnoses of mesothelioma were accompanied by causation assessments from pathologists or pulmonologists. Most mesotheliomas do not have causation assessments, nor should these assessments be considered reliable if they are not confirmed by pathology via lung-tissue analysis (as done in Roggli et al., 2023 ). In many cases, the absence of a national epidemiological surveillance system can lead to significant and hasty judgments regarding the inclusion or exclusion of asbestos as the cause of the mesothelioma. The recent analysis by distinguished pathologist, Dr. Victor Roggli, indicates that roughly 35% of mesotheliomas in men that were considered “asbestos-induced” in these causation assessments were inaccurate as lung-tissue analysis revealed that there were fewer asbestos fibers in the lung tissue of these individuals compared to background rates ( Roggli et al., 2023 ). One of the challenges with recordkeeping is the determination of a person’s occupational history with their stated cause of death. An inherent problem with this approach is defining a singular occupation for a person’s lifetime and which occupation, if any, could have contributed to an individual’s cause of death. The CDC’s National Center for Health Statistics and the National Institute for Occupational Safety and Health (NIOSH) collaborated to create a database in 2020 on industry and occupation (I&O) data as applicable to mortality vital statistics ( Steege et al., 2020 ). This database used the “usual occupation” from the 2012 census coding scheme to define the occupation that a person worked for most of their life ( Steege et al., 2020 ). This approach, as was done in the MMWR, does not account for an occupation where someone could have been exposed to a hazardous chemical for a significant portion of their life. This presents an opportunity for exposure misclassification. For example, a young man could have been in the Navy for 4 years and was exposed to asbestos, then worked 4 years in a textile mill where he was exposed to cotton dust, but from ages 34–65, he worked as bus driver. His primary occupation would be listed as a bus driver. At this time, this approach to coding occupations may very well be the best way to correlate between occupational history and disease incidence, but it clearly can be misleading for some persons. Neither the MCD nor SEER databases, by themselves, track occupation for either men or women. It is important to understand anatomical sites such as pleural (originating in the lining outside of the lungs), peritoneal (abdominal cavity), and pericardial (heart) mesotheliomas, which are distinct diseases with different dose-response relationships, pathology, and survival rates. The MCD records do not frequently record the distinct diseases of pleural, peritoneal, and pericardial mesotheliomas, with approximately 72% of mesothelioma amongst women being characterized as an unspecified type (shown in Table 1 ). This lack of distinction in the MCD records is also seen amongst men in which approximately 77% of men were diagnosed with an unspecified type of mesothelioma ( Mazurek et al., 2017 ). Table 1. Number of deaths and age-adjusted death rates for malignant mesothelioma in women above the age of 25 years organized by mesothelioma anatomical location ( Mazurek et al., 2022 ). Anatomical site Number of deaths (%) *Age-adjusted death rate (per 1 million) Pleura 968 (7.9) 0.35 Peritoneum 1,119 (9.2) 0.42 Pericardium 35 (0.3) 0 Other 1,385 (11.3) 0.52 Unspecified 8,842 (72.3) 3.29 *Adjusted using 2000 U. S. standard population. Number of deaths and age-adjusted death rates for malignant mesothelioma in women above the age of 25 years organized by mesothelioma anatomical location ( Mazurek et al., 2022 ). *Adjusted using 2000 U. S. standard population. There are a number of papers that use the SEER database which provide a stronger analysis regarding the anatomical location of mesotheliomas ( Moolgavkar et al., 2009 ; Price, 2022 ; Price and Ware, 2004 , 2009 ). Unlike the MCD, the SEER database does not include anatomical locations that are labeled unspecified. Historically, it is estimated that approximately 80% of total mesotheliomas in men and approximately 60% in women are pleural mesotheliomas ( McDonald and McDonald, 1980 ; Pavlisko and Sporn, 2014 ). Pleural mesothelioma in men is clearly correlated, when considering latency, with asbestos consumption in the United States from approximately the mid-1930s to the early 1970s ( Price and Ware, 2009 ). Since pleural mesothelioma make up the majority of cases, Price and Ware (2009) stated that “… the pattern over time of pleural mesothelioma was indistinguishable from the pattern for total mesothelioma defined as the sum of pleural and peritoneal cases.” Asbestos consumption peaked in the mid-1950s until 1972, when it rapidly declined in use over the years following ( Price, 2022 ). There is strong evidence that the latency period of pleural mesothelioma due to asbestos exposure is 30 to 40 years ( Pavlisko and Sporn, 2014 ). Based on the consumption trends, it makes sense that the age-adjusted incidence of pleural mesothelioma peaked in the 1990s (see Figure 1 ). Despite the massive increase in the use of asbestos between 1930 and 1970, there was no associated increase in the incidence of pleural mesotheliomas in women (see Figure 1 ). Roggli et al. (2023) reported “…that asbestos fiber content of lung tissue [in those who developed mesothelioma] has steadily decreased over the past four decades, with a concomitant increase in the relative proportion of cases that are not asbestos related.” Specifically, “[a] decreasing percentage of [pleural mesothelioma] cases with an elevated lung asbestos content [in comparison to 20 subjects with no known exposure to asbestos and normal lungs at autopsy] was once again observed: 91.8% of cases in the 1980s, 93.1% in the 1990s, 84.6% in the 2000s, and 65.1% in the 2010s.” For all mesothelioma cases [not just pleural mesotheliomas] in the 2010s, Roggli et al. (2023) found that 54% had “elevated” fibers in their lungs compared to the reference population. Cases were considered “elevated” if at least one of five possible parameters exceeded the reference population: total asbestos bodies (by EM or LM), commercial amphibole fibers (amosite and crocidolite), non-commercial amphibole fibers (tremolite, actinolite, and anthophyllite), or chrysotile fibers. Figure 1. Observed and expected SEER data regarding age-adjusted incidence of pleural and peritoneal mesothelioma from 1915 to 1995 as well as the relevant exposures from 1973 to 2012 (recreated from Mezei et al., 2017 ). When considering a 30–40-year latency period, there is a strong association between asbestos use and pleural mesothelioma amongst men. Commercial asbestos use in the U.S. did not have a significant effect on peritoneal mesotheliomas for either men or women as the age-adjusted peritoneal mesothelioma rates are generally flat. Observed and expected SEER data regarding age-adjusted incidence of pleural and peritoneal mesothelioma from 1915 to 1995 as well as the relevant exposures from 1973 to 2012 (recreated from Mezei et al., 2017 ). When considering a 30–40-year latency period, there is a strong association between asbestos use and pleural mesothelioma amongst men. Commercial asbestos use in the U.S. did not have a significant effect on peritoneal mesotheliomas for either men or women as the age-adjusted peritoneal mesothelioma rates are generally flat. “Elevated” fibers in the lungs, as it is defined, can still be well below a dose of asbestos that would have any measurable impact on a person’s likelihood of developing mesothelioma. Roggli et al. noted, “Furthermore, women and peritoneal cases accounted for an increasing percentage [of mesothelioma cases] in the last two decades, and these groups are associated with a substantially lower number of cases with elevated tissue asbestos content” ( Roggli et al., 2023 ). Historically, high cumulative airborne exposures of amphibole asbestos (i.e., 20 or more f/cc-years), only conceivable in the pre-OSHA era, have been associated with an increased incidence of peritoneal mesothelioma ( Browne and Smither, 1983 ; Darnton, 2023 ; Selikoff et al., 1979 ). Even for men, the incidence of peritoneal mesothelioma in the U.S. has been infrequently associated with occupational exposure to asbestos that occurred in the post-OSHA era ( Moolgavkar et al., 2023 ; Price, 2022 ). From approximately 1940 to 1975, insulators who had high amosite asbestos exposures (estimated time-weighted average (TWA) exposures of 4 to 12 f/cc or cumulative exposures of 160 to 480 f/cc-years) had an increased incidence of peritoneal mesothelioma ( Selikoff et al., 1965 , 1979 ). With incredibly few exceptions ( Cowan et al., 2015 ), it is unlikely workers (both men and women) in the United States have been exposed to enough amosite asbestos to have caused peritoneal mesothelioma since the mid-1970s due to regulatory and litigation pressures. When accounting for latency, mesotheliomas as a result of exposures prior to 1973 would have been apparent by 2018 at the latest (assuming up to a 45-year latency). Moolgavkar et al. (2017) stated “Using combined data from the National Program for Cancer Registries (NPCR) and SEER, Henley et al. (2013) confirmed that between 2003 and 2008, female pleural mesothelioma rates were flat, whereas male rates continued to decline [and flatten out in the mid-2010s]. Trends in some European countries also [similar to those in the U.S.] indicate that a large fraction of peritoneal mesotheliomas is unrelated to asbestos exposure, with flat age-adjusted rates in both men and women (e.g., Burdorf et al., 2007 ; Hemminki and Li, 2003 ).” If asbestos-related, peritoneal mesothelioma are associated with cumulative exposure levels much higher than those associated with pleural mesothelioma ( Beckett et al., 2023 ; Darnton, 2023 ; Goswami et al., 2013 ; Pierce et al., 2016 ; Roggli et al., 1992 ). In short, it is highly improbable that peritoneal mesotheliomas that have been observed since the mid-2010s were caused by amphibole (or chrysotile) asbestos (for men/women). Misdiagnosis is yet another possible confounder when interpreting MCD and SEER data. It is noteworthy that “[m]alignant mesothelioma must be distinguished from benign, reactive mesothelial proliferations on the one hand and on the other from various primary and secondary malignancies involving the serosal membranes” ( Pavlisko and Sporn, 2014 ). As noted by Dr. Roggli, a distinguished pathologist, “Peritoneal carcinomatosis is commonly due to another primary malignancy and may readily be confused with peritoneal mesothelioma. Such a distinction is critical in determining whether a peritoneal malignancy might be asbestos related” (Personal communications with Roggli). Almost always, a careful histological examination with immunohistochemical analysis can discriminate between peritoneal mesotheliomas and another cancer, for instance, ovarian cancer.