By | 2017-11-02T12:36:41+00:00 February 2nd, 2009|4 Comments

By Myra L. Karstadt, Ph.D

Whether you Twitter, IM, text or use plain old-fashioned English, here’s an important message for occupational health professionals: material safety data sheets (MSDSs) are deeply flawed, and something has to be done to change that situation. Steps are needed to improve the sheets and to ensure that the major OSHA toxic chemicals hazard communication program, the HazCom (hazard communication) standard (OSHA 1983 and amendments), which is based on those flawed sheets, is altered to provide workers with useful information that they control concerning the toxic chemicals to which they are exposed at work. Let’s hope that a new Administration sympathetic to workers will undertake the steps necessary to improve workers’ access to information on hazardous chemicals.

MSDSs as we know them really date to the late 1960s, when the U.S. Department of Labor developed what is now known as the “OSHA Form 20.” The Form 20 is a one-page (two sides) document originally intended for use in protecting the safety and health of workers in shipbuilding and shipbreaking. The Form 20 covers many aspects of occupational safety and health, with small spaces suitable for entering very brief statements on everything from appropriate respiratory protection to health hazards ranging from skin irritation to cancer. Although some companies are still using what amounts to the OSHA Form 20, there are all sorts of formats and lengths of MSDSs today.

A recent paper presents the results of what is essentially a meta-analysis of a group of peer-reviewed papers on MSDS. The authors, located at the University of British Columbia in Vancouver, identified reports of inaccurate, missing, misleading and/or difficult-to-comprehend information on MSDS from countries including the U.S., Canada, the U.K., Japan, South Africa and New Zealand.

Those results aren’t surprising. What should be surprising is how few papers have been written on MSDS, given the focus on the sheets in this country’s worker education programs and extensive use of the sheets here and abroad.

In 1980, colleagues at Mt. Sinai School of Medicine and I began a long-term study of MSDSs for products used in auto repair and body shops. The study, preliminary results from which were published in 1984, was not initially focused on MSDSs themselves, but rather on composition of the products being used in shops represented by Local 259 of the International Union, United Auto Workers (UAW). When we contacted product marketers for information on product composition, most of the time we received replies consisting of MSDSs. It became apparent soon after we started our project that the MSDSs themselves were worthy of further study. To that end, I have done sporadic checks of MSDSs for products in the original study group. A publication on the changes in MSDS over the years for those products is in preparation.

Recent checks on some MSDSs in the auto repair- auto body product group have yielded results consistent with findings in the past: sheets are still seriously deficient. A cursory review of several sheets in 2006 showed that sheets were out of date and that toxicity data, especially data on chronic toxicity, were inconsistent among marketers of the same chemical. Data were also lacking or inaccurate on toxicity endpoints such as carcinogenicity.

That MSDSs are still inadequate 26 years after promulgation of HazCom suggests two things about HazCom: the standard isn’t being enforced, and even if it were, it’s not sufficient to ensure good-quality sheets. The inadequate sheets also tell you that workers are still not getting good information on the hazards of chemicals with which they work, which was allegedly the objective of HazCom.

A good proposed rule: Hazards Identification
In the 1960s and 1970s, occupational health issues got a good deal of public attention in the United States. Asbestos, vinyl chloride, bis(chloromethyl) ether- all caused fatal cancers in workers, but if those chemicals were in products in the workplace, workers often didn’t even know the chemicals were present. At that time, it was common for workers to be handling containers marked “Formulation J” or “Mix A,” with no product composition information on the label. It was clear that workers should have access to information on the chemicals with which they work. That sentiment was the force behind the “Worker right-to-know” (RTK) movement, which resulted during the 1970s in worker RTK laws in several states.

While states took some action to get workers information on chemicals in the workplace, OSHA did nothing until the Carter Administration (1977-1981). The Carter OSHA developed the Hazards Identification (HI) proposed standard (OSHA 1981). Under HI, workers were given information on what they were being exposed to, and workers could then take that information on identities of chemicals in their workplaces and develop data on hazards and risks. HI required that product constituents appear on labels, so workers could see what was in the containers they were using and wouldn’t have to ask supervisors for MSDSs or other documents.

HI placed very severe restrictions on the ability of companies to keep product constituents confidential. That was particularly true for chemicals that caused serious health effects, such as cancer. In essence, if a chemical were a carcinogen, that chemical’s identity had to appear on the product label.

Since MSDSs were widely known to be poor documents, the HI proposed standard didn’t rely on them. Rather, HI stated that if MSDSs were available, they could be passed along, but MSDSs were neither required nor central to the proposed rule.

Once workers got information on what they were being exposed to, where could they get help figuring out what exposure to the chemicals meant for their health? In the late 1970s, union representation was much more common than it is today, most notably in the private sector. Unions had health and safety professionals on staff, and those professionals could help workers develop information on workplace hazards and risks. In addition, the Carter OSHA had introduced the New Directions program, which was designed to help workers with health and safety problems; it provides funding, usually to unions, to deal with those problems.

In addition to the stronger position of unions in the late 1970s, manufacturing was much more prevalent in the United States economy at that time. Manufacturing workers have traditionally been more sensitive to workplace health issues than workers in sectors such as service or office work.

Unfortunately, development of the HI proposal took a long time, and the proposal finally appeared as a “midnight reg,” published in proposed form at the very end of the Carter Administration.

A bad final standard: Hazard Communication (HazCom)
Chemical and manufacturing companies had been vociferously opposed to the HI standard, and the incoming Reagan Administration’s actions reflected industry’s concerns. One of the first acts of the Reagan Administration’s Secretary of Labor was to withdraw the HI proposal. Two years later, after contentious hearings, the HazCom standard was promulgated in final form.

The HazCom standard turned HI on its head, and is certainly one of the most notable exercises in regulatory cynicism in the history of the United States administrative agencies. Where HI was based on giving workers control of information concerning their exposures by putting product composition data on the label, HazCom downplayed the importance of giving workers composition data, instead basing itself on MSDSs, the discredited sheets prepared by employers or product marketers. Thus, employers controlled not only the information on which chemicals were in a product, but the information on what kind of hazard and risk workers might experience upon exposure to the chemicals. Also, instead of having product information available to workers on the label, workers would usually have had to request MSDSs from a supervisor.

The 1983 version of HazCom (final rule) gave certain health professionals the right to obtain precise product composition data- if those professionals signed liquidated damages agreements, with payment to the company whose product was involved if composition data were disclosed- but denied workers the right to those composition data, even if workers were willing to sign liquidated damages agreements. Years later, litigation succeeded in getting workers the right to sign the agreements and get the product composition data, but that is certainly no replacement for putting composition data on the label, where workers could obtain and freely discuss the chemicals to which they were being exposed.

It’s difficult to take HazCom seriously. It’s a performance standard, whose objective is to give workers employer- or product marketer-evaluated data on hazards/risks of exposure to workplace chemicals. How employers or product marketers reach that objective is, apparently, of little concern to OSHA. There are no specifications for MSDS layout or contents, and although there are guidelines as to which databases should be consulted when preparing MSDSs, there are no requirements that the steps used in preparing MSDSs be written down and kept available so that a worker or health professional could review them. There is no central repository for MSDSs. Also, there is no requirement that companies hold onto their sheets so that old sheets can be consulted for information on past exposures; that is a very important issue when doing epidemiology studies.

A major barrier to workers finding out what they’re being exposed to is HazCom’s permissive approach to disclosure of chemical composition. Any chemical, no matter how hazardous, can be claimed confidential and not disclosed. Since there is no registration of confidential ingredients with OSHA, or a requirement that OSHA approve withholding of that information, workers are simply deprived of what could be the most important information they should have: the composition of the products they are being exposed to in the workplace.

HazCom lacks enforceable requirements for preparation of MSDS and, in general, lacks enforceable criteria to ensure that workers obtain important information on possible consequences of workplace exposures. Even with the few enforceable requirements in the standard, it’s fair to say that OSHA has done little enforcement of HazCom over the years- excepting checks on whether a written HazCom plan is available in a workplace, and that plan can be as brief or as detailed as an employer chooses because of lack of specificity in the standard.

MSDS Problems Today
OSHA certainly hasn’t done much to assess the quality of MSDSs, and the lack of enforceable quality criteria in the standard is no doubt one of the reasons the sheets are still poor in quality.

There is another problem with MSDSs: their use has spread beyond the workplace. The sheets are currently being used to provide information to consumers who use products such as solvent cleaners, toners and inks for copiers and printers, and home hobby products. NLM has created the Household Products Database, an online collection of hundreds of unevaluated MSDSs for products in categories such as auto products, arts and crafts, and home maintenance. The use of sheets by consumers only adds to the need to improve the sheets’ quality.

There is also a group of workers who need information on products they’re being exposed to, but who are likely not as sophisticated about occupational health and safety as industrial workers of the past and who are ill-served by HazCom. Office workers using printer and copier products, service workers using cleaning products, child-care workers dealing with craft products and cleaners, food industry workers working with cleaning products or specialized chemicals- all need information on chemical hazards. It’s likely a good proportion of workers in some of these work categories are not good readers of English, but MSDSs are prepared in English, another barrier to improving some workers’ access to product data.

Unfortunately, unions have declined in the private sector since the late 1970s, and the occupational health staffs that were available to workers through unions are, in certain cases, less available than they were then.

It’s certainly true that these days workers have access to scientific information on chemical toxicity through the Web. However, do-it-yourself sites like Wikipedia can be less than trustworthy when it comes to toxic chemicals. I’ve found a couple of Wikis that read like the promotional material for marketers’ products. In fact, roaming the Web for information on toxic substances really does require a trustworthy guide, which may be something NIOSH could provide through support of occupational health professionals and OSHA could provide through a new version of New Directions.

We need to get started right away on fixing up MSDS. If we’re really going to reengineer American industry to bring back manufacturing, it’s going to be exceedingly important to ensure that workers get good information on the chemicals to which they are exposed at work. It’ll be difficult to use HazCom to improve MSDS, since there is so little enforceable language in the standard dealing with the content and quality of the sheets, but whatever can be done using HazCom should be done. It would also be wise to begin work on a standard that would, like the Hazards Identification proposed rule, give workers information on product composition.

There is reason to hope for positive change at OSHA under the incoming Obama Administration. A very positive change would be initiation of a program dealing with provision to workers of information they can use to protect their health when they are exposed to chemical products at work.

Myra L. Karstadt, Ph.D is a consultant in toxicology, and is located in Chevy Chase, MD.

Kaplan SA (1986). Development of MSDS.

Occupational Safety and Health Administration (1983 and amendments). Hazard communication. 29 CFR 1910.1200.

Nicol A, Hurrell A, Wahyuni D, McDowall W, Chu W (2008). Accuracy, comprehensibility, and use of material safety data sheets: a review. Am J Ind Med 51: 861-876.

Karstadt M, Bobal R (1984). Access to data on chemical composition of products used in automobile repair and body shops. Am J Ind Med 6: 359-372.

Karstadt M (2009). Material safety data sheets for products used in automobile repair and body shops: 1980-2009 (in preparation).

Occupational Safety and Health Administration (1981). Proposed rulemaking on hazard identification. 46 Fed. Reg. 4412 et seq. (January 16, 1981; withdrawn February 12, 1981).

Household Products Database. (Accessed January 19, 2009).

About the Author:

The Pump Handle


  1. Target Population February 3, 2009 at 10:46 am - Reply

    TPH reader Inkwell offers the following comment:

    “I can tell you two things (one from a worker and the other as a ommunications professional): 1) MSDSs are not available in Spanish, which is a huge issue, and 2) from a communications standpoint, they hinder direct, easy-to-access information for the reader.

    Like the medication warnings, they are written to cover the behinds of the companies and not for the actual worker on the ground. I would be interested to know just how many workers EVER READ THEM. And I wonder how many workers EVEN KNOW THEY EXIST.

    Translating the documents into Spanish would be a start, but really now! We know many Hispanic construction workers are not highly literate in their native language. Perhaps those who can read could tell the other workers what the MSDS says. But when? And who knows if they will understand it?

  2. Joan Lichterman February 11, 2009 at 12:04 am - Reply

    Thanks for this article. OSHA’s lack of attention hit me particularly strongly a week ago when I found one manufacturer’s MSDS for t-Butyl Lithium, a pyrophoric compound that recently killed a 23-year-old researcher at UCLA. Despite all the potentially serious health effects of exposure, the MSDS reads: “None of the chemicals in this product are considered highly hazardous by OSHA.” See In contrast, a technical bulletin from another company recommends using a controlled-atmosphere chamber when handling pyrophoric reagents ( Workers need understandable, reliable, and consistent information (as well as training, training, and more training).

  3. Dave Smith February 12, 2009 at 4:41 pm - Reply

    I agree – the MSDS as usually written is almost worthless in helping people effectively manage the hazards in use. They are written as product liability defense documents first and hazard info second. I like the International Chemical Safety Cards, available in lots of languages and short. However you still need the MSDS to comply with Haz Com.

  4. Jack Quinlan March 26, 2009 at 11:14 am - Reply

    Why not replace MSDS’ with New Jersey Fact Sheets, written in real English, understandable, user friendly……I could go on, but the essence is, they’re a far better source of information than a document written to comply than one intended to impart information.

Leave A Comment