The statistics prove it: Falls from heights are a veritable scourge in Quebec’s construction industry. In 2017, across all industries, 10 workers died from falls, accounting for 16.1% of all workplace fatalities—the third highest cause of death for the reference year. Furthermore, falls rank second as a cause of accidents in terms of costs: $397 million a year for 2010 to 2012.

According to the CNESST, “falls to a lower level, which are defined as a point of contact at the end of the fall lower than the surface where the person was before falling,” cause some 4,000 occupational injuries each year in Quebec.

The best strategy to protect workers against this hazard is simply to eliminate all work at heights. But when that’s just not possible, collective means of protection, such as guardrails, have to be used. Besides ensuring increased safety by preventing falls, guardrails also give workers greater mobility. Temporary guardrails, as opposed to permanent ones, are commonly erected on construction sites. They have to meet the requirements of the Safety Code for the Construction Industry (SCCI) in force in Quebec, however.

While the IRSST had already studied the design of the vertical posts and rails of temporary metal guardrails, it had never examined their method of securement, which means how they are screwed or nailed to the support structure. Wood guardrails, which are generally installed on open-web joists or prefabricated walls, had simply never been assessed at all. It was therefore hard to ensure compliance with the resistance-to-loadings requirements prescribed by the SCCI, noted Bertrand Galy, a researcher in Mechanical and Physical Risk Prevention at the IRSST and author of the research report on the strength of wood and metal guardrails. “Earlier work hadn’t looked at metal guardrails under real working conditions and had completely ignored wood guardrails, even though they’re used frequently,” the researcher said.

Hundreds of tests

To fill this knowledge gap, Bertrand Galy and his collaborators carried out three series of guardrail strength tests. The first, involving 262 tests, was conducted in the lab on wood guardrails made with boards (two-by-fours) and secured to open-web joists. A host of different combinations were tested to determine the influence of the various parameters studied, such as joist height (9.5 in. to 16 in.) and number of spans (one to three). A number of variables, including the number and type of nails (smooth, ring shank, screw shank) or screws (wood, lag), were selected to reflect the reality of work on construction sites. For others, such as guardrail height, the researchers based their choice on SCCI provisions, which stipulate that “any guardrail shall be between 1 m and 1.2 m above the surface on which the worker is working” so that it can effectively prevent a worker from falling.

The second series of strength tests (98 in all) was also conducted in the lab on wood guardrails, but ones that were secured to a prefabricated wall consisting of a structure made of two-by-fours covered with OSB panels. The influence of the same parameters mentioned above was examined. It is worth noting that prefabricated walls were incorporated into the research design following comments received from partners in the construction industry. “To begin with, we were planning to focus on the joists. But then a number of people urged us to consider prefabricated walls, as they are widely used on residential construction sites,” said Bertrand Galy.

The third and last series of tests was conducted in the field, with metal guardrails installed by roofers on two actual buildings in good condition. The purpose of these 36 tests was to assess the strength of the guardrails using the method IRSST researchers had established, but on parapets whose condition was unknown. Three guardrail models were tested and several variables were studied, including the type of screws used (black, self-tapping, lag) and the number of sides (one or two) of the vertical post securement system. “The high number of constraints in the field meant that we had to limit the number of tests,” Galy noted.

Recommendations

Some of the conclusions of the research were expected. Thus, in tests on structures erected in the lab, the strength of a 1.2 m high guardrail was systematically less than a 1 m one, simply because of the leverage effect. Similarly, a guardrail with three spans was naturally stronger than one with one span or two, with force applied directly to the vertical post. Contrary to what the researchers had assumed, however, joist height had a relatively minor effect on strength. On prefabricated walls, the base to be nailed or screwed into was not deep enough to provide maximum strength. “Pull-out occurred pretty early. Longer nails or screws need to be used in this case,” the expert recommends. In the field, the results underscored the importance of choosing the right type of screw to anchor metal guardrails to parapets, even though differences are not that significant.

In the end, the two key variables are the number of screws used to secure the foot of the vertical post of the guardrail—follow the manufacturer’s recommendations—and the system used to secure the posts, which must involve two sides of the parapet. Bertrand Galy advises opting for versatile systems that offer several methods of securing the vertical posts and can be adapted to suit small parapets. “The observations from our study can be applied in the field fairly quickly. They represent a real opportunity to reduce falls from heights,” he noted.