Tuesday, June 30, 2009

I'm Back...

In the last week the Pump Handle has posted two interesting articles, one on how slow the administration has been to put forth nominations for DOL, and the other on a FOIA request at MSHA. The first post I'll let you read and draw your own conclusions about.

The second post is the one that really made me sit up and laugh, it reminds me of our own Ergonomic Case Development Procedures. Does anyone remember reading these, you know, before your RA decided that they were ridiculous and that your region wouldn't be doing any ergonomic cases? My copy says "SOL DRAFT 6/27/02 - Privileged and Confidential." As far as I know, these procedures haven't changed by so much as one letter in 7 YEARS! How can something still be a draft if we're expected to use it, yet nothing has changed in 7 YEARS! I'll tell you how, someone doesn't want those procedures released to the public.

I don't get it, we publish almost everything so businesses know what we expect from them, why not this? I would love for some one to do a FOIA request under the DOJ's new "when in doubt, release it" policy. When SOL denies it, I would hope the requester would point out the procedures were in draft just to keep them from the public and bring the case to court.

I also want to clarify a couple of things said over on the OSHA Underground with their last two posts.

First, the number of inspections a CSHO conducts were removed from our performance elements after Congress passed the legislation in 1998, not in the late 1980's.

Second, this was, in part, a result of the NCFLL (National Council of Field Labor Locals, our union) pushing for it. Kennedy may have been the sponsor, but the NCFLL pushed it. There is some history here, in the old days an IH could do 40 inspections and get an Outstanding, but if a safety did one PSM inspection or a big fatality, they could end up with 40 inspections and get a Fails to Meet on that element. It was a pretty unfair system. I'm not saying today's system is any better, but removing the number of inspections from our evaluation was at least an attempt to fix a problem.

For those of you who are new to the agency or don't work for OSHA, contrary to what Kane says, don't expect big changes in OSHA anytime soon. This is my 5th president and will be my 6th Assistant Secretary, and, while the changes will eventually be noticeable, they won't happen fast and they won't be dramatic. Not withstanding the fact that Kane continues to give the Secretary credit for the work that CSHOs are doing, the simple fact is you can't turn a bureaucracy 90 degrees, much less 180 degrees. And you certainly can't force the change in direction overnight.

Yes, we will have more CSHOs, and yes, we should finally have the money for training, but I also expect more emphasis programs. Now we'll have less time to do more inspections, and the additional inspections are going to be more complex.

You have to keep in mind that most of those 130 new CSHOs don't know anything. They may have masters degrees, but they're still relatively stupid and won't be capable of conducting an inspection on their own for almost a year. Once they're out on their own, they're likely to get just the simplest of inspections for the next 1-2 years. This means that each of us are going to have to do more inspections, and each of those inspections will be, on average, more complex than what we're doing now. Right now I get to mix in a few B.S. complaints that we all know before I even go out are going to be in compliance, but now the new CSHOs are going to get those and I'm going to be stuck with extra SST or NEP inspections.

Not one of those 130 CSHOs have been hired yet, so we don't actually have a change, what we have so far is the promise of change. The only actual change I've seen is in our own attitude. We seem to believe that this administration is behind us to support us, not in front of us slowing the ship down. Only time will tell.

Finally, big ups to the Eau Clair AO, not the Secretary, for Milk Specialties Company.

Monday, June 22, 2009

Quick follow-up

Quick follow-up on a comment from the June 15, 2007 post. The commenter said:
feasible abatement is NOT a requirement to support a citation where a standard is being used.
It may however be an affirmitive defense from the employer. All the CSHO has to establish is hazard, exposure and knowledge.

Feasible abatement is part of the 5a1 supportive documentation, but not when there is a standard.
This statement is not quite true. In order to establish a prima facie case, the agency must demonstrate all four conditions have been met, a hazard exists, employees are exposed to the hazard, the employer knew of the conditions, and a feasible means of abatement exists. We have to do that for all cases, period.

OSHA standards, however, allow the CSHO to skip documenting that a hazard exists and that there is a feasible means of abatement, because the rulemaking process does that for us. A standard is written because a hazard exists, and the standard itself is the means of abatement (if you read the preamble of any standard, and I don't know why you would, it documents all of this). This means that we do not document that a hazard exists, we document that a violative condition exists. That the guard is not on the saw does not prove a hazard, in order to do that we would have to provide research into the impact of saw blades on human hands and this happens when the machine isn't guarded, etc. But, not having a guard on is a violation of the standard and also the means of abatement.

Thursday, June 18, 2009

Following up the Follow up

Here are a couple more comments from RT I want to share and respond to, but in reverse order.
Additional thought - what keeps an OSHA inspector from going down the Scenario 2 route?

E.g., citing an employer for 5 different 5(a)(1) citations/items if they didn't follow 5 different provisions of a consensus standard like the NFPA, and rack up $35k in penalties?
We can't cite a company for violating different provisions of a consensus standard, because we can't cite a consensus standard. That is was I was trying to get at in my first post on 5(a)(1). We're citing for exposing employees to a hazard, not for failure to follow a particular abatement method. Remember, a standard (whether OSHA promulgated or consensus) is nothing more than a detailed abatement method.
Thanks for the response. Can you elaborate on why you think if a consensus standard was the only evidence you had for employer recognition, the ALJ would toss the case? Do the memos from the National Office ever explain why you can't cite for failure to follow a consensus standard?
You're welcome.

A consensus standard is not, by itself, recognition of a hazard. In order to use a consensus standard as a piece of the evidence, we have to link it to the company or industry. You previously mentioned API standards, so let's say I inspect a refinery and find that a relief valve on a reactor isn't sized properly, meaning if there is an over pressurization of the vessel, the relief valve can't handle the release and everything goes boom. Let's suspend the PSM standard for a minute and say I decide to cite 5(a)(1). The charging language will say the government is accusing the refinery of exposing their employees to the hazard of an explosion, not to an undersized relief valve. The AVD will show that the relief valve is improperly sized which could lead to an explosion. The last paragraph will will mention that using the API standard to size the valve would prevent the hazardous condition.

In order for the citation to be upheld, I can either link that API standard to the company, or find some other way to demonstrate recognition. Since we're talking about a refinery, there's a pretty good chance that they are a member of API, so that's one of the things I'm going to check. I'm also going to check to see if that API standard is widely used within the industry. Now I've confirmed that the company is a member of API and every other refinery that's a member of API uses that standard, I have recognition.

Let's flip it, let's say the API standard exists, but nobody in the industry uses it, let's say the industry engineers uses a rule of thumb they all learned in college, and the valve at this refinery meets that rule of thumb. Now, even though the API standard may be the best engineering practice, it's worthless to the case because no one uses it. That doesn't mean you can't establish recognition in other ways, for example, if they keeping blowing the reactor up because it can't release the built up pressure, that's a good way to demonstrate recognition.

As for the memos, I can't find mine right now, but there is a long discussion on 5(a)(1) in the new Field Operations Manual (FOM) starting on page 4-14 you should check out, just be warned, it's a very large .pdf file (about 3.5 mb).

Wednesday, June 17, 2009

5(a)(1) Follow-up

I received the following comment from RT that I want to bring out and respond to:
"Good description of 5(a)(1), I agree. But I still see it all the time where OSHA uses consensus standards to prove up a 5(a)(1).

For example, in the example you mention, at the hearing, when OSHA is asked to show that the hazard was recognized by the employer or the employer's industry, the SOL attorney will typically attempt to introduce the NFPA standard as evidence of industry recognition of a hazard. "Everyone in the industry follows the NPFA, therefore this employer (in this industry) recognizes the hazard cited." Isn't that using the NPFA standard to prove a 5(a)(1)?

This happens frequently to the oil and gas industry, where OSHA trots out American Petroleum Institute (API) recommended practices in support of their 5(a)(1) citations."
The question RT asked, "Isn't that using the NPFA standard to prove a 5(a)(1)?" is a good one, and the answer is yes, but if a consensus standard was the only evidence we had for employer recognition, I think we would get laughed out of court. Also, using a consensus standard as evidence of employer recognition or as one feasible means of abatement is not the same as citing a company for failure to follow a specific consensus standard. This may seem like a distinction without a difference, but it's not. Let me give you a practical example of citing 5(a)(1) versus citing a standard.

Let's start with a fictional workplace called Abel's Metal Fab Shop. Abel's is a metal fabrication shop that specializes in metals with high cadmium content. They do typical metal fab work there, cutting, welding and grinding. Abel doesn't believe that science has adequately proven that cadmium is dangerous so he does nothing to protect his workers.

Now suppose a CSHO walks into Abel's and sees black dust everywhere, people eating lunch right out in the shop, people smoking in the shop area, etc., every IH's worst nightmare. The kind of place you're going to spend weeks in. For the purposes of this example let's assume we can't cite willful or instance-by-instance, and that there will be no reductions for size, history or good faith.

Scenario 1: No OSHA standard exists that addresses cadmium, no expanded health standard and no PEL, but there is a consensus standard. Assuming I can prove all of the necessary elements of a violation, I will get to issue a 5(a)(1) citation for one violation with a $7,000 penalty to Abel for failure to protect his employees from exposures to dangerous levels of cadmium. That's the end of the cadmium related citations.

Scenario 2: OSHA has an expanded health standard for cadmium and my AD and RSOL trust me enough to classify the citations myself. Assuming I can prove all of the necessary elements of a violation, I will get to issue a citation with multiple violations for over exposures, not conducting initial monitoring, not establishing regulated areas, not having a written program, not implementing engineering controls, no emergency response plan, not providing hygiene facilities, housekeeping, not conducting medical surveillance, training, etc., etc., etc. The cadmium standard also has respirator requirements, which call in part of the respirator standard, so I get to cite some of that standard too. Citing each of those as separate violations at $7,000 each means I've gotten into the significant case range just on cadmium violations.

The difference between the number of violations and penalties for a 5(a)(1) versus an expanded health standard is pretty obvious. If we were allowed to cite a consensus standard, the citations would look like the second scenario not the first.

Monday, June 15, 2009

General Duty Clause Citations

I want to clarify for the non-OSHA people just how we use the General Duty Clause to issue a citation (we generally refer to these as 5(a)(1) citations in favor of the term General Duty Clause), and specifically the role that consensus standards play in those citations (this is for you John).

Section 5 of the OSH Act

Below is Section 5(a)(1) of the OSH Act:
(a) Each employer --
(1) shall furnish to each of his employees employment and a place of employment which are free from recognized hazards that are causing or are likely to cause death or serious physical harm to his employees;
This is the section of the OSH act that we use when we have no written standard but the employer is still subjecting their employees to a hazard.

Documenting Citations

Let's first backup a little for a lesson in CSHO 101. In instances where OSHA has a standard a CSHO must document four elements for each violation: 1) That a hazard exists; 2) That employees are exposed to the hazard; 3) That the employer knew, or should have know with the exercise of reasonable diligence, of conditions that exposed the employees to the hazard (referred to as employer knowledge); 4) That a feasible means of abatement exists to protect employees.

Let's take an example. Suppose I'm driving along and I see people working on a scaffold at a height over 15 feet, and the scaffold has no rails. I am, of course, going to stop and conduct an inspection. This is a simplified version of what I'm going to document for my case:
  1. A hazard exists: The hazard in this case is a fall from height. Establishing this as a hazard has been taken care of by the fact that we have a standard that states that employees must be protected from falls when working at heights over 6 feet in construction.
  2. Employees are exposed to the hazard: I saw, video taped and interviewed employees working on the scaffold.
  3. The employer knew, or should have know, of the conditions: I'll be able to show that the employer (usually in the form of the site super or foreman) saw or walked by the scaffold while it was up with no rails.
  4. A feasible means of abatement exists: Following the standard is usually the means of abatement, all the company has to do is put up the top and mid rails.
Documenting 5(a)(1) Citations

When we're dealing with 5(a)(1), those same four elements have to be met, plus we add a fifth element, employer recognition. Employer recognition follows from employer knowledge, first we show that the employer knew of the conditions of work, then we have to show that the employer knew (or should have known) the conditions were hazardous.

With a 5(a)(1) showing the first element, that a hazard exists, becomes a bigger task and can get quite involved, depending on what the hazard is. For example, showing that aerosolized wood flour can explode is relatively easy (there is a lot of work to do, but the CSHO just needs to work through the process), showing that nano-particles cause respiratory illness, not so easy (yet).

Showing the second and third elements are the same for a standard as for 5(a)(1).

The fourth element, feasible means of abatement, can be very difficult, or very easy, depending on the hazard, and this is where consensus standards may come in.

What is a Hazard?

Before I get to consensus standards, however, first it's important to understand what a hazard really is and what the purpose of a standard is.

Let's take a hazard close to John Astad's heart, combustible dusts. Dusts in and of themselves are not usually hazardous, yes there are a few temperamental ones like magnesium, but by and large they are not inherently hazardous when left alone. Dusts can become an inhalation hazard when they are aerosolized in high enough concentrations, or they can become an explosion hazard when they are aerosolized in high enough concentrations AND there is an ignition source (assuming it is combustible).

What's the Purpose of a Standard?

The purpose of a standard, whether it has been promulgated by OSHA, MSHA, EPA, or any other federal/state agency or whether it is a consensus standard, to to prevent employees from being exposed to a hazard.

How Does 5(a)(1) Differ?

OSHA standards tend to be very proscriptive when it comes to abatement, telling an employer in detail what must be done to protect their employees from a hazard. The OSH Act lists no specific abatement requirements, which means that an employer is free to use any method they want to abate the hazard.

What does it Mean?

What does this mean for 5(a)(1)? That failure to follow a consensus standard is not a hazard, the aerosolized dust blowing up is the hazard. One feasible means of abatement may be to follow a consensus standard, but any abatement that protects employees is acceptable.

OSHA 5(a)(1) Citations

Let's look quickly at how OSHA 5(a)(1) citations are written. Below is a real citation, although the name has been left off, we all know who it is.
TEXT FOR CITATION: 01 ITEM/GROUP: 005 HAZARD: EXPLOSION P.L. 91-596 Section 5(a)(1) of the Occupational Safety and Health Act of 1970: The employer does not furnish employment and a place of employment which are free from recognized hazards that are causing or likely to cause death or serious physical harm to employees in that employees are exposed to the hazard(s) of explosion and fire:
a. Top of Silos 1, 2, & 3 (9th floor) and Bottom of Silo (tunnel) - On or about 02/07/2008, bulk material steel conveyor belts on the top and bottom of the silos that convey granulated sugar were not equipped with bearing temperature, belt alignment, and vibration detection monitors at the head and tail pulleys to shut down equipment and/or notify the operator before the initiation of a fire and/or explosion, exposing employees to explosion and fire hazards.

b. South Packing House & Bosch Packing House - On or about 02/07/2008, inside legs (bucket elevators) used to convey granulated sugar were not equipped with bearing temperature, belt alignment, and vibration detection monitors at the head and tail pulleys to shut down equipment and/or notify the operator before the initiation of a fire and/or explosion, exposing employees to explosion and fire hazards.
One feasible means of abatement is to install bearing temperature, belt alignment, and vibration detection monitors at the head and tail pulleys of all steel belt conveyors and inside legs (bucket elevators), as referenced in National Fire Protection Association (NFPA) 61, 2008 & 2002, Chapter 7.3. Pursuant to 29 C.F.R. 1903.19, within ten (10) calendar days, the employer must submit an abatement plan describing the actions it is taking to prevent the hazards described in instances a and b, above.
The first paragraph is the charging language, it's the part of the citation that says "we, the government, accuse you, the company, of exposing your employees to a hazard." Notice it doesn't say we accuse you of failure to follow a consensus standard.

The next two paragraphs (a. and b.) are the Alleged Violation Description, which says "these are the circumstances under which you exposed your employees to the hazard." Again notice it doesn't say we accuse you of failure to follow a consensus standard.

The final paragraph is the part that shows we have done our jobs and identified a feasible means of abatement (remember element number four from above?). In that final paragraph we do put forth an NFPA standard as a feasible means of abatement, but keep in mind that the employer is free to ignore that and use any means they decide will protect their employees. They can choose to stop processing sugar, they can follow a different consensus standard, they can try to automate the process so no people are ever around, whatever, it's up to them, they just have to stop exposing their employees to the hazard.

As long as that was, it was a very simplified discussion of 5(a)(1), as anyone who makes their living doing this can tell you. I hope, John, that answers your question. If not ask again, if it leads to other questions, feel free to ask those.

Wednesday, June 10, 2009


If anyone wonders why I get so frustrated at the OSHA Underground, just read Mondays post and then follow along below.
"The Press release is posted on the OSHA site.
Who ever is running the site is keeping up with the times."
The press release came out the same day the citations were issued! We can't issue a press release before the citations, can we?
"Why is it so hard to put these General Duty Clause violations up?"
This case is still open, the company hasn't even had their 15 working days to respond yet, so of course the citation language isn't available yet. Any one who works for OSHA should know that we do not release that kind case file information until the case is closed. It has to do with that pesky Constitution and an employers rights.
"No industry consensus standard used."
These are the kind of statements that make me question whether or not Kane has ever actually worked for OSHA. Every CSHO should know, as the National Office has memoed us numerous times, WE CAN NOT CITE FOR FAILURE TO FOLLOW A CONSENSUS STANDARD! We can't do it, it's not allowed, it's forbidden, it's anathema, and it's going send your AD into a tizzy.

Beyond that, I can't even find a consensus standard for crowd management.
"No prior OSHA or NIOSH guidance on the issue."
Why would we have ever written guidance on this? How many times has it happened? The other instances Kane mentioned were The Who concert in 1979 and the AC/DC concert in 1991, yet in both instances the people who died were concert goers, not employees. Certainly the fact that they were concert goers doesn't help the families with their grief, but it does preempt us from having jurisdiction. So I ask again, with our limited resources, why would we have ever even have thought to work on guidelines?
"This will be a tough one to affirm.
WE salute NY for trying this.
Too many people would have washed their hands clean of this.
No one would blame them."
This I agree with, go Long Island.

I just wish that the OSHA Underground would, at least once in awhile, stop just throwing negative comment after negative comment out there and take a realistic look at our budget and staffing levels and help to find better ways of doing things. If we went by what they've posted, we would working on so many standards and so many guidelines, receiving so much training, and conducting so many inspections targeted towards so many hazards, that we would need 300,000 CSHOs doing no more that three inspections per year, but spending 14 weeks doing each of them.

Let's get real, let's role up our sleeves and try to figure out what works, what could work better, and what doesn't work. That is how we're going to make things better.

Monday, June 8, 2009

Mea culpa, mea culpa, mea maxima culpa

When I started this blog, I promised myself that I would never delete any posted comments (just SPAM). Part of what I had hoped to accomplish here was to have a frank discussion about safety and health and OSHA's role in protecting workers, but the last few comments that people have left have been mean spirited and way off topic, so I broke my promise and deleted two comments.

From this point forward, I'm going to refocus the blog towards OSHA's role in safety and health (including what OSHA could do better), which means no politics beyond those that relate directly to OSHA, and I will be the sole arbiter of what is or is not directly related. If you want to have a discussion of government's role in or lives, you need to go somewhere else, because it won't happen here. I am not singling any one commenter, the two posts I deleted were from opposite sides of the political spectrum.

This is where the mea culpa comes in, this blog got off track because I allowed it to, but the bigger sin was that I made it worse, and for both of those reasons I'm sorry.

To the commenter I dubbed The Anarchist, calling you that was every bit as mean spirited as what I accuse you and the other commenter of being, and I'm sorry. If you want to continue a discussion about governments role in our lives, please e-mail me or start you own blog and I'll subscribe. Maybe in the future I can act more like an adult.

I said this once before, and I hope I mean it this time: And now, back to our regularly scheduled programming.

Tuesday, June 2, 2009

Here are the two comments I received on how OSHA could measure success:
"The use of the data collected for the Site Specific Targeting Program might be a starting point since the same sic codes are used most of the time A comparision of older establishment lists with newer lists could help establish repeats and provide an indicator of improvement.
Just a thought...."
This is a good thought. We could do it in two ways (maybe concurrently?), first by looking at the industries and seeing how their injury rates change over time, and second by looking at individual sites that show up year after year. I'm just not sure how we would pull the second part off, tracking something like this is important, yet our IMIS system is so antiquated that we couldn't use it for something like this, and who knows when or if the new system will ever be rolled out.

Someone pointed out to me that the Silica NEP calls for mandatory follow-up of sites where overexposures were found, which is a similar kind of thing.

The downside to these two measures is that they are micro-measures instead of macro-measures. In other words, they measure only the direct impact of inspections and not the secondary impact or the impact of outreach.

The second comment:
"OSHA success may be so small that it is hard to measure! No one high in OSHA wants to measure this because it might cause the OSHA Administrators to have a very red face. Granted, workplace safety has improved over the decades but it might be a tough sell if the results of OSHA efforts are singled out from other workplace safety forces such as the high cost of workplace accidents from worker's comp rates, lawsuits, etc. So, I don't think OSHA management really wants the answer to this question cause they might not like it and then they would be put in the position of having to defend the OSHA program. For now, OSHA management wants to let sleeping dogs lie and to not disturb them."
Sorry, I don't buy it, or the OSHA Underground's anti-national office portion of today's post (although I do appreciate the mention). I know too many people who work in the NO, most of whom believe in what we're trying to do just as strongly as I do.

Do you know the biggest difference between the NO and an AO? It's who we fight with. As CSHOs our fights are usually with employers and our own attorneys, but the NO has to fight politicians, lobbyists and the press. Even though we're probably involved in more directly confrontational fights, I think it's still easier than fighting politicians, lobbyists and the press.

Don't misunderstand, I'm not trying to suggest that the NO should be worshiped from afar, after all, I've spent more than my share of time mocking them, sometimes because they deserve it and sometimes just because it's fun, but I do appreciate the fact that they buffer us from direct assault by politicians and the national press (I should also acknowledge that the RO also does some of this, even though I mock them almost as much as the NO).

No, I don't think the problem is that we're afraid to look at ourselves, I think the problem is we may not know how to look at ourselves.