DBDavid W. Brew

David W. Brew

*This is an algorithmically generated author record based on publicly available information.

David W. Brew has a total of 17 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Risk assessment of trace hydrazine in a 2020 bottled water-associated nonviral hepatitis outbreak

    by Blake M. Deckard, Kylie M. McCauley, Michael E. Stevens, David W. Brew, Dennis J. Paustenbach

    Between August 1 and December 3, 2020, a 124-day outbreak of nonviral hepatitis occurred in geographically limited regions of Nevada and California, resulting in 21 hospitalized cases (18 in Nevada, 3 in California), including 18 intensive care unit admissions and one fatality in an adult female with preexisting conditions. In March 2021, a multi-agency investigation led by the Southern Nevada Health District (SNHD), Centers for Disease Control and Prevention (CDC), U.S. Food and Drug Administration (FDA), and others identified consumption of Real Water, an alkalinized bottled water product, as the likely cause. On March 24, 2021, a recall was issued and the company ceased production of Real Water. Despite extensive investigation, no definitive causative agent was identified. However, trace concentrations of hydrazine, a known hepatotoxin, were detected in select product batches, prompting concerns regarding its potential role in the outbreak. This study presents a quantitative risk assessment of hydrazine exposure from daily consumption of retail bottled Real Water products during the outbreak period. Estimated intake was calculated based on age-, sex-, and body weight-specific water consumption patterns and measured hydrazine concentrations (mean: 0.0505 mg/L; maximum: 0.3 mg/L). Sub-chronic acceptable daily intakes (ADIs) of 0.09 mg/kg-day for adults and 0.04 mg/kg-day for children were developed based on toxicological data from isoniazid (INH) studies. Hydrazine and INH share similar profiles regarding hepatotoxicity, with hydrazine acting as a major mediator of INH-induced liver injury. Calculated hazard quotients for all demographic groups were via retail Real Water consumption could not have caused the observed hepatic injuries, supporting the conclusion of the Centers for Disease Control (CDC) that the etiologic agent remains unidentified.

  • Open Access

    United States Environmental Protection Agency's Perfluorooctanoic Acid, Perfluorooctane Sulfonic Acid, and Related Per- and Polyfluoroalkyl Substances 2024 Drinking Water Maximum Contaminant Level: Part 2 – Fifteen Misconceptions About the Health Hazards

    by Dennis Paustenbach, Kylie McCauley, Jacob Siracusa, Sarah Smallets, David Brew, Michael Stevens, Blake Deckard, My Hua

    This paper examines widely held beliefs about the six per- and polyfluoroalkyl substances (PFAS) addressed in the final U.S. Environmental Protection Agency's (EPA) rule on PFAS in drinking water (e.g., the Maximum Contaminant Levels - MCLs). Based on our understanding of the scientific literature and the comments submitted by stakeholders regarding the EPA's regulation that was promulgated in April 2024, we identified 15 misconceptions that had a weak scientific foundation. These are now memoralized in the MCLs for the six PFAS but remain debated due to ongoing ambiguous research findings. Many critics of the MCLs found the EPA's systematic review of the published relevant information, particularly the toxicology of perfluorooctanoic acid (PFOA) and perfluorooctane sulfonate (PFOS), to be inadequate. The following seven views are among the most important. First, the EPA asserted that the toxicology of these six chemicals was poorly understood and lacked sufficient data to determine a safe daily intake level for chronic health effects; nonetheless, they promulgated what may be the costliest environmental regulation to date. Notably, adverse effects remain difficult to demonstrate in occupationally exposed individuals even at blood concentrations 50-100 times higher than current background PFAS levels. Second, the Agency indicated that the epidemiology data showed that exposure to PFOA and PFOS caused kidney and potentially other cancers, yet the data were equivocal and do not support that assertion. Third, it was stated that specific non-cancer effects, such as heart disease, would be prevented under the promulgated rule; however, the studies that they relied upon do not show an increased incidence of heart disease even in highly exposed populations. Fourth, the Agency relied on animal data to support its views on the likely toxic effects in humans, despite ample toxicology data that animals, particularly rodents, are poor predictors of the human response to PFAS exposures. Fifth, the EPA predicted a reduction in healthcare expenditures that would offset much of the cost of complying with the MCL, but, they did not have adequate data to support this prediction. Sixth, the EPA suggested that these six PFAS act through a shared mechanism of action (i.e., PPARα pathway induction); however, data indicate that PPARα induction in humans may be 80% less than what is observed in rodents. Also, induction of the PPARα pathway is not a cause of systemic disease. Seventh, the Agency failed to disclose that achieving the new MCL would yield negligible reductions in blood PFAS levels even among highly exposed populations, given drinking water accounts for only 20% or less of total PFAS exposure. The survey that could answer that question, the EPA's fifth Unregulated Contaminant Monitoring Rule, was only 25% complete at the time the MCL was promulgated. Overall, our analysis concluded that while the EPA's intent to regulate these chemicals due to their environmental presence was necessary, the derivation of the MCLs and the alleged health effects was based on the application of the precautionary principle rather than robust scientific evidence.


  • Top Cited Papers
    Open Access

    United States Environmental Protection Agency's Perfluorooctanoic Acid, Perfluorooctane Sulfonic Acid, and Related Per- and Polyfluoroalkyl Substances 2024 Drinking Water Maximum Contaminant Level: Part 1 – Analysis of Public Comments

    by My Hua, Kylie McCauley, David Brew, Jonathan Heywood, Jacob Siracusa, Michael Stevens, Dennis Paustenbach

    In March 2023, the EPA proposed a 4.0 ppt maximum contaminant level (MCL) for perfluorooctanoic acid (PFOA) and perfluorooctane sulfonate (PFOS) (each) and a hazard index approach for four other PFAS. The EPA sought public feedback on the proposed MCL in early 2023 and received 1626 comment submissions via the PFAS docket website (Docket ID: EPA-HQ-OW-2022-0114). Final MCLs were promulgated on April 10, 2024. Our analysis of the PFAS docket identified 128 comments that had a reasonable degree of scientific merit, with 57 comments endorsing the regulations and 71 questioning the MCLs public health utility. Critics noted the lack of evidence for adverse health effects at low PFAS exposures, the rule's significant impact on the economy, and the EPA's selection of published papers which the Agency chose to support their views. Many well-substantiated comments highlighted that few, if any, adverse health effects were reported at doses as much as 100-1000 times above those associated with the proposed drinking water guidelines. We found that the comments which discussed the evidence linking PFAS exposures below 200 ppt in drinking water to adverse health effects were equivocal. Most of the well-documented science based comments indicated that the data did not justify setting a 4.0 ppt MCL. It was noted that the EPA MCL was quite different from drinking water standards in other countries (up to 8-140 fold lower). During the review, it became apparent that a 4.0 ppt MCL may have little effect on PFAS blood concentrations in most Americans since drinking water accounts for less than 20% of their total PFAS intake. Additionally, a significant portion of the American population consumes minimal amounts of tap water. Commenters noted that the financial burden for treatment and cleanup was much higher than what was reported in the justification for the final MCL which was submitted to the Office of Management and Budget (OMB) and eventually promulgated. It is possible that EPA underestimated the financial impact on the nation by up to 100 to 200-fold. Our analysis indicates that many, if not most, of the scientifically rigorous comments on the EPA's proposed MCL were not acknowledged or considered by the Agency. We conclude the article by offering sixteen recommendations for the EPA to consider if Congress or the courts choose to reopen the evaluation of these MCLs. These included convening an international expert panel, reevaluating the appropriateness of the LNT model for PFAS, ensuring adequate time for study quality assessment and cost-benefit analysis, considering an approach to implementing a series of MCLs, critically reevaluating scientific studies, adhering to EPA risk assessment guidelines, addressing SDWA compliance concerns, revisiting the Hazard Index approach, and ensuring thorough and transparent review of public comments.


  • Open Access

    A risk assessment of mechanics who changed chrysotile asbestos containing brakes and other vehicle components in the 1950s–early 2000s era: an update on the 2004 evaluation

    by David W. Brew, Michael E. Stevens, Arthur M. Langer, Dennis J. Paustenbach

    For the past 50 years, there has been an ongoing interest in understanding the potential health hazards, if any, to vehicle mechanics who worked with asbestos-containing brakes in the 1950s-early 2000s era. Two reviews have been published on this topic, one by Langer (2003) ("Reduction of the biological potential of chrysotile asbestos arising from conditions of service on brake pads") and another by Paustenbach, et al. (2004) ("Environmental and occupational health hazards associated with the presence of asbestos in brake linings and pads (1900 to present): a 'state-of-the-art' review"). This analysis is an update on those papers since a considerable amount of research has been published over the past 20 years on this topic. The following important aspects are addressed in this review: new information on the toxicology of chrysotile, toxicology studies of brake dust associated with grinding, additional epidemiology studies and meta-analyses published on auto mechanics of the era, previously unfound data on how brakes (during the era when chrysotile was used) were manufactured, and new work describing the transformation of chrysotile to various degradation products during vehicle braking. This update also addresses questions about the health hazards associated with asbestos in vehicle clutches, transmissions, and gaskets. The exposure data indicate that the airborne concentrations of chrysotile fibers associated with vehicle mechanic work when asbestos was in auto brakes were, on average, less than 0.04 f/cm3 (8-h TWA) and the average lifetime cumulative dose was in the vicinity of 0.5-3 f/cm3-year for mechanics of that era. Although many of these fibers may have no toxicity due to thermal degradation and the conversion to degradation products, 31 epidemiology studies have evaluated the risks of mesothelioma for vehicle mechanics of this era and all but one indicate that there was no increased incidence of this disease in these workers. The weight of evidence continues to indicate that the asbestos-related health risks to vehicle mechanics from asbestos-containing components were de minimis. The risks associated with take-home and bystander exposure of a mechanic were also addressed and they were found to pose a de minimis or zero health risk to those potentially exposed. Based on our evaluation, there is no indication that asbestos from asbestiform tremolite was present at detectable concentrations in bulk samples of brakes or in the air during brake work. The recent U.S. Environmental Protection Agency (EPA) risk assessment of 2024 on chrysotile and their views of the hazards of asbestos-containing brakes were discussed. Their analyses did not alter our views that exposures to mechanics posed no increased risk of asbestos related disease. The latest knowledge about the role of genetic susceptibility on the development of mesothelioma is also addressed.


  • Open Access

    Occupational exposure to asbestos in the steel industry (1972–2006)

    by Dennis J. Paustenbach, Michael E. Stevens, Brett P. Tuttle, Ross A. Shore, Sabina Ligas, David W. Brew

    Background: Historically, the use of asbestos in steelmaking has been limited to a few applications. Due to its physical and chemical properties, asbestos was not necessary or suitable for most purposes in a steel mill. The few applications where asbestos were used (i.e., certain gaskets, brakes, protective cloth, refractory materials, insulation materials, and hot top products) were replaced by alternative materials as they became available. Objective: We discuss historical uses of asbestos in steel manufacturing and the associated airborne asbestos concentrations collected at sixteen U. S. Steel facilities between 1972 and 2006. Methods: A total of 495 personal airborne asbestos samples from the U. S. Steel industrial hygiene records were analyzed across four time periods corresponding to changes in the OSHA permissible exposure limit (PEL) for asbestos. 68% of the samples (n = 337) were considered representative of an employee’s workday. The remaining samples (n = 158) represented task samples. Samples were grouped by facility, department, and job category within the four time periods. Results: The average fiber concentrations measured for each facility and department over time were below the contemporaneous OSHA PEL. The mean representative workday asbestos air concentration from 1972 and 1975 was 1.09 f/cc. The mean representative workday concentration decreased to 0.13 f/cc between 1976 and 1985, then decreased again to 0.02 f/cc between 1986 and 1993 and 0.03 f/cc between 1994 and 2006. For task samples, the mean air concentration from 1972 to 1975 was 3.29 f/cc. The mean task sample concentration decreased to 0.48 f/cc between 1976 and 1985, then decreased again to 0.01 f/cc between 1986 and 1993 and 0.03 f/cc between 1994 and 2006. Only eleven out of the 495 samples (2.2%), for both task and representative workday samples, were in exceedance of the contemporaneous PEL(as an 8-hour TWA), ten of which occurred prior to 1978. Eight of these eleven PEL exceeding samples were task samples. Of the remaining three representative workday samples, two had unknown sampling times. Impact: This paper presents an analysis of all the available personal sampling data for airborne asbestos across 16 facilities of the U. S. Steel Corporation between 1972 and 2006. This dataset has previously never been publicly shared or analyzed. It represents one of the more complete industrial hygiene datasets from a corporation to be presented in a scientific journal and, due to the similarities in the processes at each mill, it should reflect analogous exposures throughout the steelmaking industry in the United States. One of the benefits of presenting these data is that it also provides insight into where asbestos-containing materials (ACMs) were used in the steel making process. This is just one example of a large firm that released information that had previously remained in file cabinets for decades. We believe that another benefit of publishing this paper is that it may encourage the largest firms in industry to assemble and analyze their industrial hygiene data to benefit the occupational hygiene, medical, and epidemiology communities. This can support future epidemiology studies and improve the design of future industrial hygiene programs.


  • Metabolomic Investigations of the Temporal Effects of Exposure to Pharmaceuticals and Personal Care Products and Their Mixture in the Eastern Oyster (Crassostrea virginica)

    by David W. Brew, Marsha C. Black, Marina Santos, Jackson Rodgers, W. Matthew Henderson

    The eastern oyster (Crassostrea virginica) supports a large aquaculture industry and is a keystone species along the Atlantic seaboard. Native oysters are routinely exposed to a complex mixture of contaminants that increasingly includes pharmaceuticals and personal care products (PPCPs). Unfortunately, the biological effects of chemical mixtures on oysters are poorly understood. Untargeted GC‐MS metabolomics was utilized to quantify the response of oysters exposed to fluoxetine, N, N‐diethyl‐meta‐toluamide (DEET), 17α‐ethynylestradiol (EE2), diphenhydramine, and their mixture. Oysters were exposed to 1 µg/L of each chemical or mixture for ten days, followed by an eight‐day depuration period. Adductor muscle (n = 14/treatment) was sampled at days 0, 1, 5, 10 and 18. Trajectory analysis illustrated that metabolic effects and class separation of the treatments varied at each time point and that overall, the oysters were only able to partially recover from these exposures post‐depuration. Altered metabolites were associated with cellular energetics (i.e., Krebs cycle intermediates), as well as amino acid metabolism and the urea cycle. Exposure to these PPCPs also affected metabolic pathways associated with anaerobic metabolism, osmotic stress, and oxidative stress, in addition to the physiological effects of each chemical's postulated mechanism of action. Following depuration, there were fewer metabolites altered, but none of the treatments returned to their initial control values, indicating that metabolic disruptions were long‐lasting. Interestingly, the mixture did not directly cluster with individual treatments in the scores plot from partial least squares discriminant analysis, and many of its affected metabolic pathways were not well‐predicted from the individual treatments. This research highlights the utility of untargeted metabolomics in developing exposure biomarkers for compounds with different modes of action in bivalves.