Manganese and Boundary Work
Recently I was invited to present at the 2016 International Neurotoxicology Conference on Manganese in New York. While I am, at best, a Sunday driver on manganese toxicity, my claimed expertise in the evolution of scientific knowledge and public health response—occupational exposure limits—justified my place on the program. There’s a simmering controversy, and litigation, over the association of manganese exposure in welding and Parkinson’s-like disease, which made it worth my while to take the subway across town to Mt. Sinai. My main interest was in manganese as a paradigm for evaluating OELs, and for how research evolves—and who evolves it.

Parkinson’s disease (PD) afflicts about 1 million Americans and about 1 in 100 people over 60, and may be underdiagnosed. PD, a movement disorder, also has a cognitive component, associated with degeneration of a brain structure that releases the neurotransmitter dopamine. Manganism, another disease associated with manganese exposures, presents with similar debilitative signs and symptoms but can be distinguished by specialists. Manganism has been reported since 1837 in mining, refining, and steelmaking, then in battery making, and now there is evidence of neurological effects from environmental exposures.
The toxic potential of manganese has long been recognized, but the recognized toxic potency, as reflected in occupational exposure limits, has evolved over time. The OSHA PEL of 5 mg/m3, a ceiling value for manganese and compounds, was promulgated in 1971 based on the ACGIH Threshold Limit Values of 1968. The documentation for the TLVs from 1971 recommended a ceiling limit of 5 mg/m3 and included only13 references published from 1940 to 1969. My reading is that the numerical result was basically a lowest observed adverse effect level (LOAEL) from a 1943 study in a steel plant. The TLV was reduced to 1 mg/m3 in 1977. The NIOSH recommended exposure limit (REL) of 1 mg/m3 appears to have been derived from the TLV and then from testimony on the OSHA PEL update in 1989, although documentation for the REL and its actual promulgation date are not apparent. (The PEL update was flushed in 1992 when a Court of Appeals sided with industry.) In 2001, citing publications from 1937 to 1992, ACGIH adopted a TWA-TLV of 0.2 mg/m3 for manganese, elemental and inorganic compounds—a 25-fold reduction. The TLV was further reduced to 0.02 mg/m3 in 2013—250-fold less than the PEL—based on documentation with about 100 references through 2011.
Internationally, exposure limits vary from 0.02 mg/m3 respirable and 0.2 mg/m3 inhalable in Germany to 0.5 mg/m3 in the U.K. to 1 mg/m3 in Japan. Dates when these standards were promulgated are not readily available. The EPA IRIS reference concentration (RfC) of 5 x10-5 mg/m3 dates to 1995. The heavily documented and recent recommendation is the 2012 ATSDR minimal risk level (MRL) of 0.3 µg/m3 for chronic exposure. (Equating the RfC and MRL to 8-hour exposure limits requires taking into account duration of exposure.)
There’s a simmering controversy, and litigation, over the association of manganese exposure in welding and Parkinson’s-like disease.

The breakpoint in OELs for manganese is based on two Belgian studies published in 1987 and 1992. These found dose response for adverse effects on neurobehavioral tests, which yielded a LOAEL of 0.05 mg/m3 for workers with an average duration of exposure of 5 years. In its IRIS review published in 1995, EPA applied an uncertainty factor of 1,000 based on the absence of an identified NOAEL (no observed adverse effect level), short duration of exposure, and other considerations. Several additional studies by other investigators supported the LOAEL. It would appear that the 2001 ACGIH TLV was set above the LOAEL observed in workers. The 2013 TLV is slightly below the LOAEL, also based on neurobehavioral studies.
Welding electrodes consumed during welding contain manganese, as does fume from steel being welded, which is released into workroom air. Two major studies published recently, one based on original data sets (
Journal of Occupational and Environmental Hygiene
, December 2010) and one based on literature reports mostly from studies of health effects (
Annals of Occupational Hygiene
, January 2011) reported average manganese samples grouped around 0.25 mg/m3—significantly higher exposures than the 0.02 mg/m3 TLV.
There are about 1 million American workers in occupations that include welding. Literature reports of associations between welding, manganese, and Parkinson’s disease go back at least to 1981; there are a lot of these reports, including several published after or not included in the 2011 closing of the ACGIH documentation. Plaintiffs’ lawyers brought lawsuits by afflicted welders aimed at the manufacturers of the welding electrodes. Product liability law permits workers to sue product suppliers for these effects, while workers’ compensation law is a barrier to gaining full compensation from employers. Although a few lawsuits achieved large judgments for plaintiffs, these were later reversed. According to a, a defense lawyers’ website, a multidistrict litigation panel established in 2003 for manganese-related lawsuits was dissolved in 2013.
Null reports appeared in the literature, published by authors disclosing prior consulting relationships with industry or funding from entities such as General Electric, a “group of manufacturers of welding equipment,” a series of commercial consulting firms, and the “welding industry defense group.” The author lists for most of these papers include consultants from firms for hire or with no affiliations, although a few include faculty with academic affiliations or even NIOSH. In contrast, reports finding associations between welding with neurological conditions or Parkinson’s were mainly authored by NIOSH scientists or university-based workers with federal funding.
The assessment of authority in scientific questions invokes the concept of “boundary work,” well known in communications theory and philosophy of science. Factions invoke a boundary to establish dominance in disputes, such as the 19th century debate over evolution, or the current debate over global warming. “Scientists” have authority; non-scientists may only contribute opinions and observations to the debate.
Clearly inside the boundary are investigations funded following competitive peer-reviewed applications, published in peer-reviewed journals, and complying with norms of scientific ethics. Maybe the best way to define “outside the boundary” is economic conflict of interest and the manner in which the research question is framed. The nature of funding and the organizational affiliation of authors should be a factor in weighing a study. A paper in the March 2016
American Journal of Industrial Medicine
shows that the source of funding is associated with investigators reporting null results on occupational studies.
I admit that I am stumped about defining the boundary. I’m not stumped about the weight of the evidence regarding the potential of welding-related manganese exposures to cause neurological effects indicative of Parkinson’s risk.
Particle size remains an issue. The work defining the point of departure, the 1987 and 1992 studies, was in facilities that produced chemicals or batteries. I would expect “total” particulate measurements in this environment to underestimate the “inhalable” particulate being ingested. Also, in a welding environment, I would expect particulate to be mostly 1-micron particles and below, likely more potent than respirable and potentially penetrating the brain.
The major take-home lesson from this review is that the OSHA PEL for manganese is distinctly not protective, and the newest ACGIH TLV is well supported by science, showing that welders are at distinct risk of overexposure, and probably at risk for neurological effects of manganese. Claims that publications of projects supported by industry funding dispel concerns for work-related Parkinson’s-like disorders are suspect.
is a professor in the CUNY School of Public Health in New York.
He can be reached at (212) 396-7782 or
American Journal of Industrial Medicine
: “Influence of author's affiliation and funding sources on the results of cohort studies on occupational cancer” (March 2016).
American Sociological Review
: “Boundary-work and the Demarcation of Science from Non-science: Strains and Interests in Professional Ideologies of Scientists” (December 1983).
Annals of Occupational Hygiene
: “Estimation of Particulate Mass and Manganese Exposure Levels among Welders” (January 2011).
Journal of Occupational and Environmental Hygiene
: “Manganese, Iron, and Total Particulate Exposures to Welders” (February 2010).