“Life is a journey and not a destination.”  — Lynn H. Hough

When Richard Nixon signed OSHA into law in 1970, the United States was looking at 14,000 work-related fatalities per year. With a workforce of about 70 million full-time equivalents, the work-related fatality rate was about 20 fatalities per 200,000,000 hours worked.

In 2018, the most recent year reported by the Bureau of Labor Statistics, the size of the workforce has doubled, and the fatality rate has dropped to 3.5 fatalities per 200,000,000 hours worked.

OSHA is celebrating its 50th anniversary this year, and with good cause. The overall work-related fatality rate has fallen almost 6-fold in that time.

Wouldn’t It Have Fallen Anyway?

OSHA is an agency that many love to hate. “They have all these ridiculous rules.” And when told that the workplace in the United States has become much, much safer as a result of the efforts of OSHA, the response is sometimes, “Yeah, but wouldn’t fatality rates in the U.S. workplace have fallen anyway? Isn’t that what the market demands?”

Work-related fatality rates would have fallen, I’m sure. But 6-fold? Not likely. Consider the self-employed, which includes gig-workers. In 2018, the work-related fatality rate for wage and salary workers—employees who OSHA regulations are meant to protect—was 2.9 fatalities per 200,000,000 hours worked. That almost 7 times lower than the work-related fatality rate in 1970.

On the other hand, the work-related fatality rate for the self-employed—workers to whom OSHA regulations do not apply—was 12.7 fatalities per 200,000,000 hours worked. That is clearly less than the rate in 1970, but only by about a third. It is hard to argue that OSHA regulations and OSHA’s enforcement of those regulations are not making a difference.

Process Safety Management

The Process Safety Management (PSM) standard, 29 CFR 1910.119, was not among the first of OSHA’s regulations. It hit the books in 1992, 22 years after Congress first established OSHA. It was largely a response to the disaster in Bhopal, India, where the December 3, 1984 release of methyl isocyanate from a Union Carbide pesticide plant resulted in the deaths of 3,787 according to official records, although there are estimates that put the toll at as high as 20,000 deaths.

I remember as a young chemical engineer who had started his career working for Union Carbide being shocked and dismayed at the death toll and at the realization that I could be responsible for decisions that could result in the deaths of thousands.

I also remember at the time the rationalizing in our industry. At the time, many argued that the chemical industry in the U.S. should not be judged by a disaster in a third world country, insisting that nothing like that could ever happen here.

That, despite an explosion at the Union Oil refinery on July 23, 1984 in Romeoville, Illinois that killed 19.

Fortunately, leaders from 17 companies in the process industries came together to form the Center for Chemical Process Safety (CCPS), chartered less than four months after Bhopal on Mary 23, 1985, under the auspices of the American Institute of Chemical Engineers (AIChE). That organization assisted OSHA in drafting the PSM standard.

If we needed reminders that something like the PSM standard was necessary, there were plenty.  In the time between the Bhopal tragedy and the implementation of the PSM Standard, the U.S. experienced:

  • August 12, 1985 – Institute, West Virginia
    Release containing MIC from Union Carbide’s Bhopal sister-plant in West Virginia
    No fatalities
  • May 4, 1988 – Henderson, Nevada
    Pepcon fire and explosion
    2 fatalities, 372 serious injuries
  • May 5, 1988 – Norco, Louisiana
    Shell Oil refinery explosion
    7 fatalities, 42 serious injuries
  • October 23, 1989 – Pasadena, Texas
    Phillips refinery explosions and fire
    23 fatalities, 315 serious injuries
  • July 5, 1990 – Channelview, Texas
    ARCO Chemical explosion and fire
    17 fatalities, no survivors, so no injuries
  • May 1, 1991 – Sterlington, Louisiana
    Angus Chemicals/IMC Fertilizer explosions
    8 fatalities, 120 serious injuries

Have Things Gotten Better?

The pace of fatal process safety disasters has slowed. A list of notable industrial disasters includes less than a half dozen in the United States during the last 15 years:

  • March 23, 2005 – Texas City, Texas
    BP Refinery explosion
    15 fatalities, 180 injured
  • February 7, 2008 – Port Wentworth, Georgia
    Imperial Sugar dust explosions
    13 fatalities, 42 injuries
  • April 17, 2013 – West, Texas
    West Fertilizer ammonium nitrate explosion
    15 fatalities, 160 injuries
  • November 15, 2014 – LaPort, Texas
    Du Pont mercaptan release
    4 fatalities, 1 serious injury
  • June 21, 2019 – Philadelphia, Pennsylvania
    Philadelphia Energy Solutions refinery explosion
    No fatalities, minor injuries
  • January 24, 2020 – Houston Texas
    Watson Grinding propylene explosion
    3 fatalities, 18 “self-reported” minor injuries

It is interesting to note that incidents in the U.S. are increasingly rare, so that the Philadelphia Energy Solution made the list despite there being no fatalities or serious injuries. The threat alone was sufficient.  On the other hand, the Du Pont incident did not make the list—it’s included here because it obviously should be—and the Watson Grinding explosion is too recent to have made the list.

It is also interesting to note that most of the disasters on the list above involved processes not covered by the PSM standard. And while the frequency of fatal process safety disasters in the U.S. is lower now than at the time the PSM standard was implemented, the frequency of process safety disasters globally has not diminished. The disasters are simply happening in parts of the world that are developing a chemical industry but have not yet implemented equivalent regulations.

Are We There Yet?

The PSM Standard was implemented in 1992. Except for some tweaks to correct clerical errors and to adjust language to make it consistent with the Globally Harmonized System of Classification and Labelling of Chemicals, the standard is unchanged since it was first implemented. Surely, with almost 30 years to comply, the need for this regulation would be behind us.


OSHA still finds many opportunities to cite for violations of the PSM standard. In part, that is because while the standard has not changed, enforcement has become more rigorous. Things OSHA would let slide in 1995 are no longer given a pass. But also, it’s because some organizations still don’t get it, and some organizations are new to industry and need to understand their responsibilities for process safety management.

We Have Come A Long Way…

In the U.S., which has a fairly mature and sophisticated chemical process industry, process safety incidents are increasingly less frequent, and the consequences are increasingly less severe. The same is true of the chemical process industries in other developed industrial nations. In the rest of the world, where the chemical process industry is just emerging from its infancy, however, they are experiencing the growing pains we have already experienced.

For some, it may be tempting to believe that we have arrived. We haven’t. In terms of overall rates for work-related fatalities, the U.S. is at 3.5 fatalities per 200,000,000 hours worked. We’ve been stuck there for a decade. Is that the end? Is that simply as low as we can go?


In the European Union, the work-related fatality rate is much lower. The reported values exclude transportation, but assuming that work-related transportation fatalities in the EU is proportional to the work-related transportation fatalities in the U.S. (~40%), the overall rate in the EU is 2.1. In the United Kingdom, it is 0.8. Some member countries – Croatia, Bulgaria, Latvia, Romania, Lithuania, Estonia, and France—have worse rates than the U.S. But there are many countries with lower rates, which tells us that the U.S. can do better.

The same is true for process safety.

…But There is No “There” To Get To

Process safety, like life, is not a destination but a journey. It’s a road we must continue to travel, no matter how far we’ve come. It’s nice to look back and acknowledge that we have come a long way, because that confirms for us that our efforts truly make a difference. But our efforts only make a difference as long as we continue to make the effort. For everyone who is dedicated to improving process safety, thank you. Keep up the good work. Even though you will never get there, it’s still worth the journey.


  • Mike Schmidt

    With a career in the CPI that began in 1977 with Union Carbide, Mike was profoundly impacted by the 1984 tragedy in Bhopal and has been working on process safety ever since.