FILED: February 9, 2000







(SH 987089; CA A100398)

Judicial Review from Workers' Compensation Board.

Argued and submitted December 6, 1999.

Elliott C. Cummins argued the cause for petitioner. With him on the brief was Cummins, Goodman, Fish & Platt, P.C.

Judy Lucas, Assistant Attorney General, argued the cause for respondent. With her on the brief were Hardy Myers, Attorney General, and Michael D. Reynolds, Solicitor General.

Before Haselton, Presiding Judge, and Linder and Wollheim, Judges.


Citation for violation of 29 CFR 1926.502(d)(15) reversed; otherwise affirmed.


Employer Northwest Shake Tile, Inc., petitions for review of an order of the Workers' Compensation Board that affirmed two occupational safety citations and associated penalties relating to employer's roofing business. (1) Respondent Oregon Occupational Safety and Health Division (OR-OSHA) issued those citations for: (1) failing to afford adequate fall protection in violation of 29 CFR 1926.501(b)(11) (2); and (2) failing to provide "anchorages used for attachment of personal fall arrest equipment * * * capable of supporting at least 5,000 pounds (22.2 kN) per employee attached." 29 CFR 1926.502(d)(15). We reject without further comment employer's challenges to the first citation and related penalty. However, we agree with employer that the administrative law judge's (ALJ) ultimate determination that employer's anchorage devices violated 29 CFR 1926.502(d)(15) is not supported by substantial evidence. ORS 183.482(8)(c). Consequently, we reverse with respect to that citation.

On February 11, 1997, OR-OSHA safety compliance officer Timothy Nelson observed some of employer's workers on the roof of a home being built in McMinnville. Nelson saw that two anchorage points were affixed to the roof and that the self-retracting lifelines of two of the workers were attached to one of those anchorage points. (3) The anchors themselves were customized--i.e., designed and built by one of employer's principals. The parties concur in the ALJ's description of that design:

"These anchors were specifically designed to accommodate the ridge rail that runs along the top of the roof of houses where this type of roofing material was to be installed. This ridge board was a 2x3 or 2x4 attached so that it stands on the two inch edge. The anchoring device was constructed out of a heavy metal strap bent in a U shape with hinges midway down the sides so it could accommodate a variety of roof pitches. On the bottom of the U was a round eye hook. The U shape was necessary so that the device would fit down over the ridge board.

"When [Nelson] inspected this work site, he was shown one of the anchors that was similar to those that were in use by this employer. The device would be attached to the house by 16 penny nails driven through holes in the sides of the device and into the roof rafters. There were three nail holes on one side and two on the other. At the time of the hearing, the employer had drilled several more holes in each side of the anchorage device so more fasteners could be used to attach the device to the roof.

"When these devices were designed and manufactured, the employer tested them by hooking up a 300 pound weight to a self-retracting lifeline and dropping the weight in free fall simulating a fall from a roof. The employer reported that the device itself withstood this test without incident."

Nelson believed that the anchoring device was incapable of supporting 10,000 pounds (5,000 x 2 employees attached) and, pursuant to ORS 654.071, issued a citation for violation of 29 CFR 1926.502(d)(15). That provision states:

"Anchorages used for attachment of personal fall arrest equipment shall be independent of any anchorage being used to support or suspend platforms and capable of supporting at least 5,000 pounds (22.2 kN) per employee attached, or shall be designed, installed, and used as follows:

"(i) as part of a complete personal fall arrest system which maintains a safety factor of at least two; and

"(ii) under the supervision of a qualified person."

Employer contested that citation. In the ensuing administrative hearing, OR-OSHA did not offer any expert scientific or engineering assessment of the weight-supporting capacity of employer's anchoring devices. Indeed, OR-OSHA did not subject employer's anchors to any testing. Rather, OR-OSHA offered, and relied principally on, three pieces of evidence: (1) Nelson's opinion that the device could not support the prescribed weight; (2) an excerpt from the Western Woods Use Book, which described the "pull strength" of various nails, including the 16-penny nails used to affix employer's anchors (Exhibit 13); and (3) a manual for another type of anchor, which stated that that other type of anchor should never be affixed with nails (Exhibit 14). (4) OR-OSHA also presented evidence that employer had performed, and the anchoring device had passed, a 300-pound drop-weight test.

The ALJ determined that the device violated the safety standard:

"Pictures of the anchors being used by this employer on this job were submitted into the record. Those pictures show that the device would be attached to the roof by three fasteners on one side and two on the other. [Nelson] was given a 16d common nail as the type of fastener used on this project. There is an allowable load for each nail depending on the type of material into which it is attached. A 16d common nail has a maximum load of 42 pounds per inch or a total load of 105 pounds. Five nails per anchorage point would not be sufficient to meet the requirements of the rule cited as a basis for this alleged violation. This anchor itself was designed and produced by the employer. The testing done of this device would show that the device was capable of holding a load of 300 pounds when dropped to simulate a fall from a roof. This test does not, however, show that the device would support 5000 pounds per employee as the rule requires.

"The device itself was constructed out of heavy material. That material was heavier than the other devices which were submitted as exhibits at the hearing and the documentation that related to those devices indicates they would meet the requirements of this rule. Those other devices, however, were to be attached by more than five fasteners per device. Also as the literature states the wrong type of fastener was being used on this construction job. Exhibit 14, page 47, states 'Never for any reason, use nails to attach this anchor. Testing shows that most nails will pull out with less than 1,000 pounds of force. In a free fall you may exert more than 2 times that amount upon impact. If the anchor is attached with nails it will be of no value in arresting a fall.' As [Nelson] stated in his testimony, his primary concern regarding a violation of this rule was with the manner in which the anchors were attached as well as the fact that there was more than one worker attached to one anchor.

"Regardless of the design of the device, which admittedly would have to be job specific for this employer, when the device was attached to the roof by only five nails there was a violation of the rule."

On review, employer attacks that determination on several grounds. Employer's primary and essential argument is that the ALJ's determination that the anchor could not support the prescribed weight "is not supported by substantial evidence in the record"--i.e., that "the record, viewed as a whole," does not "permit a reasonable person to make that finding." ORS 183.482(8)(c). We agree. (5)

We emphasize, at the outset, that OR-OSHA bears the burden of proving a denied occupational safety violation. OAR 438-085-0820(1)(a). We examine, in turn: (1) Nelson's testimony; (2) Exhibit 13, the document describing the "pull strength" of nails; (3) Exhibit 14, the manual for another type of anchor; and (4) the results of the employer's own testing of the anchor.

Nelson had worked as an OR-OSHA safety compliance officer for two years, after receiving 12 weeks of "basic training." Before that, Nelson's experience with residential construction consisted of working as a carpenter's assistant on two homes and helping his father reroof a third. Nelson's personal experience with anchor points was limited to those three homes, which involved devices other than the employer's customized design. At the hearing, Nelson testified as to the basis for his belief that the device violated the safety standard:

"First off--I mean two employees being attached to one anchor point, that's an automatic. You just see it, and you're pretty sure it's not right, essentially on a residential-type anchor. * * *

"* * * * *

"[I]f he would have fallen * * * this hinge point would have, of course, stayed in place most likely, and it would have been a fulcrum point, for example, and it would have popped--started pulling the nails out.

"* * * * *

"Maybe with this many nails there--I doubt if it would pull, but with the amount of nails that were on the device when I was there, that really drew my attention to it."

Nelson later summarized: "The main issue was just three nails on one side, and two on the other." Thus, Nelson's opinion was ultimately based on the number of nails affixing employer's anchor. Even on direct examination, however, Nelson acknowledged that he could not address the effect of using multiple nails to affix employer's anchor: "I'm not sure if it [the nails' pull strength] goes up exp[onentially] with the amount of nails in the device * * * you'd have to be an engineer." (Emphasis added.) (6)

On cross-examination, Nelson acknowledged that he had never tested any anchoring device, much less employer's device:

"Q. [By Employer's Counsel]: Did you test [employer's anchor] at all?

"A. No, I did not. I did not have one to test.

"* * * * *

"Q. Did you ever request Northwest Shake Tile to provide one to you for testing?

"A. No, I did not.

"* * * * *

"Q. You've testified that you do--I think in answer to the Hearings Officer question, you do have experts and testing services available to you at OSHA?

"A. We hire out, yeah, for--we'd get an engineer to test equipment and stuff for us. We have done that in the past.

"* * * * *

"Q. You [personally] have never done it?

"A. No.

"* * * * *

"Q. Have you ever seen the manufacturer's test result or tested it independently yourself or through OSHA?

"A. This device?

"Q. Any device.

"A. I have not, no."

Under cross-examination, Nelson further acknowledged, specifically with reference to Exhibit 13, that he was "unsure" about the number of nails needed for the anchor to support the necessary weight:

"Q. So how many nails would it take to get 5,000 pounds?

"A. Again, I said--it goes up--I am pretty sure it goes up exp[onentially], so it's hard to tell. When I talked to technical, they said, 'You need to have an engineer test the anchor point, because of the different'--

"Q. Well, Exhibit 13 really doesn't give us a great deal of information about how many nails are appropriate or not appropriate, does it?

"A. It just shows comparing the other anchor points where it requires 12 nails, you have pretty much double the pull strength in the other anchor--

"Q. Well, how does that relate to 5,000 pounds?

"A. I am unsure.

"Q. Okay. And there's nothing in the regulations, of course, that says the number of nails, are there?

"A. No, no."

In sum, Nelson: (1) had no engineering or scientific background; (2) had limited personal experience with anchoring devices; (3) had not subjected the employer's device to testing or observed such testing, notwithstanding that OR-OSHA has obtained such test results in the past; and (4) was "unsure" about the number of nails necessary for the anchor to support the prescribed weight.

The deficiencies of Nelson's testimony concomitantly highlight the deficiencies of Exhibit 13, the Western Woods Use Book excerpt describing the "pull strength" of various types of nails under various conditions. That document shows that a single common 16-penny nail driven into "douglas fir-larch" wood has a "pull strength" of roughly 42 pounds per inch. Because the employer used three-inch nails, the "pull strength" of any single nail is approximately 126 pounds. (7) The difficulty, however, is that the exhibit does not purport to address the effect of using multiple nails--i.e., is the relationship merely additive or is it exponential? The ALJ apparently assumed, without foundation, that the relationship was additive. However, nothing in the exhibit itself or in Nelson's foundational testimony supports that assumption. Indeed, Nelson testified that he was "pretty sure it goes up exp[onentially]," and that he was "unsure" as to how the use of multiple nails "related to 5,000 pounds." Thus, from the record, the fact that a single three-inch, 16-penny common nail may have a pull strength of 126 pounds does not, rationally, support any determination as to the cumulative effect of using five nails.

Exhibit 14 is similarly deficient. That exhibit, a manual produced by the manufacturer of a different anchoring device, states:

"Never for any reason, use nails to attach this anchor. Testing shows that most nails will pull out with less than 1,000 pounds of force. In a free fall you may exert more than 2 times that amount upon impact. If the anchor is attached with nails it will be of no value in arresting a fall." (Emphasis added.)

The difficulty with relying on that material is that Nelson did not rationally relate the design and demands of the device described in Exhibit 14 to employer's device here. Moreover, Nelson at least implicitly acknowledged that if employer's anchor had been affixed with more nails, he would not have issued the citation. (8) In sum, evidence that nails may be insufficient to affix one type of design does not establish that they are inadequate as to all designs.

Finally, OR-OSHA points to--and the ALJ relied on--employer's own testing of the device. The fact that the anchor passed a test in which employer dropped a 300-pound weight from the roof to simulate a free fall reveals nothing about at what point the device might fail. We reiterate that OR-OSHA, not employer, bears the burden of proof.

We conclude that the record, read as a whole, would not permit a reasonable person to determine that employer's anchoring device could not support 10,000 pounds. Bluntly, OR-OSHA and the ALJ "could not get there from here." Accordingly, the citation pursuant to 29 CFR 1926.502(d)(15) must be reversed.

Citation for violation of 29 CFR 1926.502(d)(15) reversed; otherwise affirmed.

1. Under ORS 654.290(2)(b), the order of the administrative law judge (ALJ) whom the Workers' Compensation Board designates to conduct the contested case hearing on occupational safety citations is "deemed to be a final order of the [B]oard."

Return to previous location.

2. 29 CFR 1926.501(b)(11) provides, in part:

"Each employee on a steep roof with unprotected sides and edges 6 feet (1.8 m) or more above lower levels shall be protected from falling by guardrail systems with toeboards, safety net systems, or personal fall arrest systems."

Return to previous location.

3. Nelson observed other conditions relating to the issuance of the first citation that do not pertain to the second.

Return to previous location.

4. We address each of those items in detail in our substantial evidence analysis. See ___ Or App at ___ (slip op at 5-9).

Return to previous location.

5. Given our disposition, we assume, without deciding, the admissibility of Nelson's "opinion," Exhibit 13, and Exhibit 14, all of which were admitted over employer's objection. Consequently, we do not reach, and imply no view concerning, employer's alternative arguments on appeal challenging the admission of that evidence.

Return to previous location.

6. Earlier in Nelson's direct examination, the following colloquy had occurred:

"Q. [By OR-OSHA's counsel]: After you finished your inspection, did you do some additional investigation about the--the strength of the anchor points?

"A. Yes, I did. I--I talked to some engineers and they said that--"

At that point, employer's attorney objected on hearsay grounds, and the ALJ sustained that objection. No engineering testimony was ever admitted.

Return to previous location.

7. The ALJ inexplicably calculated that total as 105 pounds. That calculation cannot be reconciled either with the publication itself or with Nelson's testimony describing the "pull strength" per nail as approximately 120 pounds.

Return to previous location.

8. For example, Nelson testified that, "some--some anchor points do use nails, but the ones I have seen on the market require six nails to a side attached into the truss, so 12 nails total." In contrast, Exhibit 14, cautions: "Never use anything except the 'factory' supplied fasteners."

Return to previous location.

Top of page Go home

divider line

State of Oregon Seal