Evaluating Fall Safety Compliance among Skilled Trades in Construction

| |
Amanda McQueen , Melissa J. Perry , SeungSup Kim
CPWR - The Center for Construction Research and Training

Summary Statement

This CPWR Small Study evaluates fall safety among members of five constructions trades: Carpenters, Electricians, Ironworkers, Painters, and Roofers, and their use of safety compliance equipment or practices. There is a confirmed high level of safety compliance across the skilled trades, with Ironworkers potentially at higher risk and personal fall systems deserving of further attention.
January 2015

ABSTRACT

Falls are one of the leading causes of workplace death, lost work time, and costs to the construction industry. The goal of this study was to develop an assessment tool to evaluate fall safety in general construction among five construction trades and among five types of equipment throughout different stages of construction. The GW Audit of Fall Risk (GAFR) was developed by modifying three existing assessment tools (according to applicable OSHA standards), validated by a panel of experts, tested during a two-week pilot study, and used during a 12-month observation period. The overall mean safety compliance was 98.9%. Lowest mean safety compliance was found with ironworkers (97.0%), the use of personal fall arrest systems (96.3%), and during the concrete pouring/placement phase (97.3%). The findings indicate there was a high level of safety compliance throughout the project and across the skilled trades. Ironworkers may be at higher risk and deserve increased attention and support on the job. Use of personal fall arrest systems, especially during the earlier phases of construction was also a context that emerged deserving continued and heightened attention.

KEY FINDINGS

  • The study context afforded a unique opportunity to perform multiple repeat assessments on one general contractor site during the bulk of construction activity.
  • The GAFR was constructed to respond to a gap in systematic fall hazard assessment tools for general construction, and yielded valuable data on which various site characteristics and work practices could be scored.
  • The high mean safety compliance overall was primarily due to the strong safety culture fostered by the site superintendent. Site superintendent training should emphasize the importance of cultivating a safety culture on each project to make personal safety an inherent part of the job for each worker.
  • The use of personal fall arrest systems resulted in the most recurring issues throughout the observation period. These issues include the misuse or lack of use of a proper personal fall arrest system (e.g. not tying off to an appropriate structure, extending a retractable lifeline too far and negating swing fall clearance). Because these systems are essential for safely working with different types of suspended platforms, focused and regular training should be given to all workers.
  • Ironworkers were found to have the highest number of recurring issues with the use of personal fall arrest systems and scaffolding, and may therefore be at higher risk of falling. The site superintendent should work with the ironwork subcontractors to ensure adequate training and consistent reminders of proper fall safety procedures for the use of personal fall arrest equipment, aerial lifts, scaffolding, and any other pieces of equipment. Furthermore, subcontractors should ensure adequate supply and accessibility of equipment for these workers.
  • Significant differences were found when mean safety compliance for carpenters was compared to that of electricians and painters. Carpenters may be at a higher risk of falling and should be monitored by the site superintendent and carpentry subcontractor to ensure these workers are able to perform their work safely.
  • Lowest overall safety compliance was found during the concrete pouring/placement phase during the first and second months of observation. Because observation of the site began just as this phase ended, this finding was not robust enough to assume that lower safety compliance was consistent during the entire phase. However, because ladder and lift usage is inherently intense during this phase workers’ are likely to be at heighten risk of falling during this early stage of a project and fall safety protection should receive added attention.

INTRODUCTION

Prevalence of Fall Hazards in the Construction Industry

Falls are one of the leading causes of workplace death, lost work time, and costs to industry, particularly in construction (Leamon & Murphy 1995, Courtney, Sorock et al. 2001, Courtney, Matz et al. 2002). In fact, falls are considered one of construction’s “Fatal Four” (along with struck-by-object, electrocutions, and caught-in/between) and contribute roughly 35% of these injuries that are responsible for more than half of deaths within general construction, as of fiscal year 2012 (OSHA, 2013). One study of 1,025 carpenters found that 16% had personally fallen in the past year and 51% knew someone who had fallen from a height at work (Kaskutas, Dale et al. 2010). Falls in construction incur the highest workers’ compensation and hospitalization costs (Derr, Forst et al. 2001). Furthermore, the duty to provide fall protection (29 CFR 1926.501) is one of the ten most frequently cited OSHA standards violated, along with the OSHA standards to ensure safety protection and safe construction of scaffolding (29 CFR 1926.451) and ladders (29 CFR 1926.1053), as of fiscal year 2013 (OSHA, 2013).

Previous Literature on Fall Safety within the Construction Industry

Only a few surveillance studies have been conducted to evaluate fall safety practices among construction workers. The unit of observation in these studies has typically been either at the worksite or at the individual worker-level to determine compliance with fall prevention practices. For example, a Washington University construction safety team based in St. Louis, MO studied worksite fall safety by developing a tool to assess fall hazards and control practices in residential construction sites based on the Occupational Safety and Health Administration’s (OSHA) fall prevention standards for residential construction (Kaskutas, Dale et al. 2008). The tool was successfully utilized in measuring fall prevention practices in 197 residential sites. Data collected determined that truss settings met the safety criteria on average 28% of the time and use of personal fall arrest and monitoring of unguarded floor openings were rare at the worksites studied (Kaskutas, Dale et al. 2009). Likewise, Becker and his colleagues at West Virginia University developed an audit tool to assess fall safety practices in general construction and administered the tool to evaluate the impact of their organizational intervention on improvement of fall prevention practices in general construction setting (Becker, Fullen et al. 2001).

Alternatively, some studies have focused on fall safety practices at the individual-worker level. Lipscomb and her colleagues analyzed the fall injuries among union carpenters over a three-year period using an active injury surveillance system of individual injured workers as well as worksites where fall injuries occurred (Lipscomb, Dement et al. 2003). Because ladders are a major source of falls in construction, our previous work has developed a tool to assess individuallevel ladder safety practice; it includes 24 within 5 ladder use domains and was tested with 771 stepladder observations (Perry and Ronk 2010, Ronk, Dennerlein et al. 2011).

However, these previous studies have limitations (Becker, Fullen et al. 2001, Kaskutas, Dale et al. 2008, Sparer & Dennerlein 2013). First, although both significantly contribute to fall risks, to our knowledge, no previous study has sought to assess worksite- and individual-level fall prevention practices simultaneously. One previous study worked to identify contributing factors of fall injury at the individual- and worksite-level, but this study interviewed individual workers and visited worksites to assess their safety practice after fall injuries occurred (Lipscomb, Dement et al. 2003). In addition, none of the previous surveillance studies have considered specific construction trades, such as electricians, painters, and carpenters, as differing in their risks of falling (Wang 1999, Derr, Forst etal. 2001, Dong, Fujimoto et al. 2009) as compared to other trades. Finally, fall risks have not been systematically quantified among the skilled trades and this is necessary for developing more tailored and effective intervention strategies to reduce fall injury among general construction workers.

OBJECTIVES

Specific Aims

Specific aims of this study were to:

  1. Develop and evaluate an effective fall practices assessment instrument relevant to the commercial construction setting.
  2. Determine heightened risk of falling for each of the seven types of equipment targeted throughout different stages of construction.
  3. Identify vulnerable trades at specific periods with higher risk of falling, which may be targets for intervention.

METHODS

Setting and Study Population

The study was conducted at the new George Washington University Milken Institute School of Public Health building site, 950 New Hampshire Avenue, Washington, DC 20052. The study population consisted of all construction workers (Whiting-Turner and other trade-specific contractors) on the site during each observation. Only those workers of the carpentry, electrician, ironwork, paint, and roofing trades working from a height of six feet or more are to be observed.

Unit of Observation

This study had two units of observation. First, safety compliance related to the use of ladders, aerial lifts, personal fall arrest systems, and scaffolding (mobile) was assessed by observing individual employees within the five occupational construction trades targeted for this study (i.e. carpenters, electricians, ironworkers, painters, roofers) working from elevations of six feet or higher at the time of observation. During the course of observation, if a worker was observed to be working at heights in more than one context, each occurrence will be treated as a separate observation. Second, safety compliance was assessed in relation to scaffolding (fixed), guardrails, safety net systems, and roof sheathing through the observation of the work environment.

Assessment Tool

The instrument was developed by modifying three existing construction safety audit tools: a fall safety assessment tool developed for general construction (Becker, Fullen et al. 2001), the St. Louis Assessment of Fall Risks (SAFR) for residential construction (Kaskutas, Dale et al. 2008), and the Ladder Assessment Tool (Dennerlein, Ronk et al. 209, Perry and Ronk 2010). Based on the review of these tools, an extensive list was developed of items to assess fall safety practices in general construction in a comprehensive way. Modifications were made to this list and choice items based on OSHA construction standards to reflect fall hazards for general construction settings (e.g. the incorporation of work platforms on mobile scaffolding due to their frequent use in general construction). These modifications were determined by first reviewing the OSHA standards and other relevant literature to identify fall hazards specifically for general construction settings. The draft instrument was then shared with a convening expert panel, including the on-site superintendents, to review and provide feedback on the modified instrument. Finally, a two-week pilot test was conducted using the draft instrument to determine ease-of-use, as well as areas for improvement and further refinement. Further modifications occurred after the observations ceased and the researcher assessed the relevancy of the audit tool items and its overall usability.

Assessment Protocol

Each item of observation was scored dichotomously as to whether or not it was observed, and if observed, whether or not it meets the established definition of best safety practice. In each domain for audit, the average number of items not meeting the safety criteria was calculated to estimate potential heightened risk (as noted as lower safety compliance) for that domain.

The average duration of each site visit was one hour spent going through each of the building’s nine floors (including the two basement levels) and the roof, when safe to inspect, as directed by the Whiting-Turner site superintendent, Mike Whitmore. The date, floor number, phases of construction, entry for each applicable item on the assessment tool, the profession (if applicable), and any notes were recorded during each site visit. Whiting-Turner utilizes a subcontractor logging system that identifies the workers and their company according to the number of their hard hat. Thus, the researcher did not need to interact with the workers in order to identify their trade and no individual persona data will be collected. Mr. Whitmore confirmed the accuracy and appropriateness of phases of construction upon completion of this project. In addition, he authorized the use of pictures to be used in future presentations and reports.

Data Analysis

Data were collected using the assessment tool and scored dichotomously, from which, the following comparisons will be made: safety compliance for each equipment domain (ladder, aerial lift, personal fall arrest equipment, guardrails, scaffolding, roof sheathing, safety net system) across each phase of construction (concrete pouring/placement, skin, roofing, interior rough-end, interior finishes), as well as among each trade (carpenter, electrician, ironworker, painter, roofer) across each phase of construction, each trade per domain, and overall safety compliance among each domain, trade, and phase of construction. These comparisons were documented in individual Microsoft Excel spreadsheets (accompanied with a data dictionary) and were analyzed using SAS 9.3 software. One-way ANOVA and unpaired (independent) t-test analyses were derived to evaluate differences in fall safety (as evidence of lower safety compliance) among types of equipment, occupational trades, and phases of construction.

RESULTS

Trades Targeted for this Study

The purpose of this first analysis was to determine whether there was a relationship between the profession of a worker (trade) and their risk of falling from a significant height (six feet or higher), as determined by lower mean safety compliance. Specifically, this analysis was carried out to determine whether there are significant differences between carpenters, electricians, ironworkers, painters, and roofers. Thus, the null hypothesis states that in the subject population, there is no difference between subjects of one trade versus those of another trade, with respect to their mean safety compliance scores. The alternative hypothesis states that there is a significant difference between at least one of the targeted trades and another targeted trade. Total worker compliance for this project was 98.5%. Figure 1 displays the mean safety compliance for each trade during this study.

Percent Safety Compliance vs. Targeted Trades
Figure 1: Mean Safety Compliance among Targeted Trades

Compared to the other four trades that were targeted for this study, ironworkers were clearly at the lower end of compliance, and therefore, at higher risk of falling, supporting the research hypothesis. Carpenters and roofers were also found to have lower safety compliance compared to electricians and painters, indicating their heightened risk of falling. To determine the presence of significant differences between trades, a one-way ANOVA (Analysis of Variance) was conducted. Results were analyzed with one between-subjects factor and did not reveal any significant treatment effect for safety compliance by trade, (F(4, 322) = 1.21, MSE = 0.007, p = 0.3081) nor did Tukey’s HSD test with alpha set at 0.05 (see Table C-1 in Appendix C for results). Therefore, each relationship was compared using an unpaired (independent) t-test, results of which displayed in Table 1.

Table 1: Mean Safety Compliance and Independent (Unpaired) T-Test Results
Trade Comparison Mean Safety Compliance (%) Difference Between Means (95% Confidence Limits) p-value
Carpenters –
Electricians
97.9
99.4
-0.015 (-0/025, -0.006 0.002
Carpenters –
Ironworkers
97.9
97.0
-0.008 (-0.021, 0.038) 0.572
Carpenters –
Painters
97.9
99.5
-00016 (-0.027, -0.006) 0.003
Carpenters –
Roofers
97.9
98.5
-0.006 (-0.039, 0.028) 0.719
Electricians –
Ironworkers
99.4
97.0
0.024 (-0.005, 0.052) 0.105
Electricians –
Painters
99.4
99.5
-0.001 (-0.009, 0.006) 0.773
Electricians –
Roofers
99.4
99.5
0.009 (-0.007, 0.025) 0.276
Ironworkers –
Painters
97.0
99.5
-0.025 (-0.054, 0.004) 0.094

Ironworkers –
Roofers

97.0
98.5
-0.014 (-0.049, 0.020) 0.400
Ironworkers –
Roofers
99.5
98.5
0.010 (-0.006, 0.026) 0.215

This analysis involved one predictor variable and one criterion variable. The predictor variable was the targeted occupational trade (carpenters, electricians, ironworkers, painters, and roofers), as noted as a dichotomous variable for each pairing (e.g. carpenters versus electricians, carpenters versus painters). The criterion variable was a continuous variable measuring mean safety compliance for these trades. This analysis revealed a significant difference in safety compliance between carpenters and electricians, t (184) = -3.14, p = 0.0021, and between carpenters and painters, t (118) = -3.06, p = 0.0028. The sample means for safety compliance are displayed in the second column of Table 1 and show that carpenters scored lower in terms of safety compliance compared to electricians and painters. The observed difference between the means for carpenters and electricians was -0.0151 and the 95% confidence interval for the difference between means extended from -0.0247 and -0.00560. The observed difference between the means for carpenters and painters was -0.0162 and the 95% confidence interval for the difference between means extended from -0.0267 and -0.00571. The remainder of the results from this independent-samples t-test analysis revealed non-significant differences between the other trades. This information is also displayed in Table 1.

Equipment Targeted for this Study

The purpose of this second analysis was to determine whether there was a relationship between the types of equipment or a worksite safety element and a worker’s risk of falling from a significant height (six feet or higher), as determined by mean safety compliance. Specifically, this analysis was carried out to determine whether there is a difference in safety compliance between the use of a ladder, an aerial lift, a personal fall arrest system, scaffolding, and the presence of guardrails.

Total compliance based on equipment usage/presence was 98.9%. Figure 2 displays the total safety compliance values for each type of equipment observed. The frequency tables that informed Figure 2 for each assessment item and mean safety compliance can be found in Appendix A. Mean safety compliance values for each domain are also displayed in Figure 2.

Percent Mean Safety Compliance per Targeted Equipment

Compared to the other four types of equipment/worksite safety elements, use of a personal fall arrest system was the lowest in compliance compared to the other targeted equipment observed. Additionally, lower safety compliance was found with the use of scaffolding was lower than other types of equipment. To determine the presence of significant differences between these equipment/worksite elements, a one-way ANOVA (Analysis of Variance) was conducted. Results were analyzed with one between-subjects factor and did reveal a significant treatment effect for safety compliance and equipment used/present on the worksite, (F(4, 639) = 3.32, MSE = 0.005, p = 0.0105). Tukey’s HSD test revealed significant differences between personal fall arrest systems and guardrails, as well as personal fall arrest systems and scaffolding, with alpha set at 0.05 (see Table C-2 in Appendix C for results). Therefore, these relationships and those between the other types of equipment were compared using an unpaired (independent) t-test, results from which displayed in Table 2.

Table 2: Mean Compliance and Independent (Unpaired) T-Test Results For Equipment
Equipment Comparison Mean Safety Compliance (%) Difference Between Means (95% Confidence Limits) p-value
Ladders – Aerial Lifts

99.2
99.5

0.0021 (-0.0060, 0.0103) 0.6016
Ladders – Personal Fall Arrest 99.2
96.3
0.0295 (-0.0018, 0.0608) 0.0640
Ladders – Guardrails 99.2
99.3
0.0002 (-0.0084, 0.0089) 0.9510
Ladders – Scaffolding 99.2
99.0
0.0021 (-0.0039, 0.0080) 0.4935
Aerial Lifts – Personal Fall Arrest 99.5
96.3
0.0317 (-0.0003, 0.0636) 0.0524
Aerial Lifts – Guardrails 99.5
99.3
0.0019 (-0.0092, 0.0130) 0.7397
Aerial Lifts – Scaffolding 99.5
99.0
0.0042 (-0.0049, 0.0133) 0.3631
Personal Fall Arrest – Guardrails 96.3
99.3
0.0298 (-0.0024, 0.0619) 0.0691
Personal Fall Arrest – Scaffolding 96.3
99.0
-0.0274 (-0.0590, 0.0041) 0.0874
Guardrails – Scaffolding 99.3
99.0
0.0023 (-0.0073, 0.0120) 0.6338

This analysis involved one predictor variable and one criterion variable. The predictor variable was the type of targeted equipment (ladders, aerial lifts, personal fall arrest systems, guardrails, scaffolding), as noted as a dichotomous variable for each pairing (e.g. ladders versus aerial lifts, ladders versus guardrails). The criterion variable was a continuous variable measuring mean safety compliance for these domains. This analysis did not reveal a significant difference in safety compliance between any of the types of equipment/worksite elements, although the relationship between aerial lifts and personal fall arrest systems was very close to being considered significant (p = 0.0524). The sample means for safety compliance are displayed in the 10 second column of Table 2 and show that personal fall arrest system use was scored lower in terms of safety compliance compared to the other types of equipment. The remainder of the results from this independent-samples t-test analysis revealed non-significant differences between the other domains. This information is also displayed in Table 2.

Compliance to safety protocols was lowest with the use of personal fall arrest systems, with recurring issues regarding the lack of use of appropriate personal fall arrest equipment for specific jobs. Other issues of non-compliance regarded the use of ladders (e.g. climbing up and down the correct way without tools in-hand) and regarding the use of scaffolding (e.g. wheels locked during use). Furthermore, the presence of guardrails, especially toeboards, was a recurring issue of non-compliance.

Phases Targeted for this Study

The purpose of this third analysis was to determine whether there was a relationship between the phases of construction and a worker’s risk of falling from a significant height, as indicated by lower safety compliance. Specifically, this analysis was carried out to determine whether there is a difference between the risk during the concrete pouring/placement phase, the skin phase, the roofing phase, and interior rough-end phase, and the interior finishes phase. Figure 3 displays the trend in overall compliance across the months of observation (April 2013 to March 2014) and the trend with differentiations of the span of each phase. Table 3 displays the time period of observation and mean safety compliance for each phase of construction.

Osha compliance over monthscompliance over the months
Figure 3: Mean Safety Compliance Across Months of Observation Highlighting Span of Phases of Construction

Table 3: Compliance Across Targeted Phases of Construction
Phase Time Period Mean Safety Compliance (%)
Concrete Pouring/Placement Before April 2013 - Mid-April 2013 97.3
Skin Before April 2013 - January 2014 98.4
Interior Rough-End Before April 2013- Mid-December 2014 98.9
Roofing Before April 2013- Mid-December 2013, January 2014 - February 2014 98.9
Interior Finishes July 2013 - March 2014 99.1

To determine the presence of significant differences between these equipment/worksite elements, a one-way ANOVA (Analysis of Variance) was conducted. Results were analyzed with one between-subjects factor and did not reveal any significant treatment effect for safety compliance by phase of construction, (F(4, 2189) = 0.98, MSE = 0.004, p = 0.4159) nor did Tukey’s HSD test with alpha set at 0.05 (see Table C-3 in Appendix C for results). Therefore, each relationship was compared using an unpaired (independent) t-test, results from which displayed in Table 4.

Table 4: Mean Compliance and Independent (Unpaired) T-Test Results among Phases of Construction
Phase Comparison Mean Safety Compliance (%) Difference Between Means (95% Confidence Limits) p-value
Concrete Pouring/Placement
– Skin

97.3
98.4

-0.0161 (-0.0365, 0.0043) 0.3964
Concrete Pouring/Placement
– Interior Rough-End
97.3
98.9
-0.0158 (-0.0367, 0.0052) 0.4070
Concrete Pouring/Placement
– Roofing
97.3
98.9
-0.0164 (-0.0365, 0.0038) 0.3881

Concrete Pouring/Placement
– Interior Finishes

97.3
99.1
-0.0185 (-0.0380, 0.0009) 0.3293
Skin
– Interior Rough-End
98.4
98.9
0.0003 (-0.0074, 0.0081) 0.9309
Skin
– Roofing
98.4
98.9
-0.0003 (-0.0077, 0.0071) 0.9410
Skin
– Interior Finishes
98.4
99.1
-0.0024 (-0.0101, 0.0052) 0.5205
Interior Rough-End
– Roofing
98.9
98.9
-0.0006 (-0.0082, 0.0070) 0.8731
Interior Rough-End
– Interior Finishes
98.9
99.1
-0.0028 (-0.0106, 0.0051) 0.4772
Roofing
– Interior Finishes
98.9
99.1
-0.0022 (-0.0097, 0.0054) 0.5646

This analysis involved one predictor variable and one criterion variable. The predictor variable was the targeted phases of construction (concrete pouring/placement, skin, interior rough-end, roofing, and interior finishes), as noted as a dichotomous variable for each pairing (e.g. skin versus interior rough-end, skin versus interior finishes). The criterion variable was a continuous variable measuring mean safety compliance for these domains. This analysis did not reveal a significant different between any two phases of construction. The sample means for safety compliance are displayed in the second column of Table 4 and show that safety compliance during the concrete pouring/placement phase was lower compared to the other phases. The results from this independent-samples t-test analysis are displayed in Table 4.

Additional Results

Consistent with the first goal of this study, mean safety compliance was computed for each of the five trades using each of the four types of targeted equipment (i.e. ladders, aerial lifts, personal fall arrest systems, fixed scaffolding). Results are displayed in Table 5.

Table 5: Mean Compliance for each Trade using each Type of Targeted Equipment
  Ladders % mean safety compliance (number of observations) Aerial Lifts % mean safety compliance (number of observations) Personal Fall Arrest System % mean safety compliance (number of observations) Scaffolding (mobile) % mean safety compliance (number of observations)
Carpenters 98.26% (23) 97.22% (9) 87.04% (36) 89.47% (19)
Electricians 99.43% (82) 97.22% (9) 100% (4) 91.67% (4)
Ironworkers 99.67% (20) 100% (24) 94.23% (52) 94.44% (6)
Painters 99.74% (26) 100% (4) 100% (2) 66.67 % (1)
Roofers 100% (5) Not Observed Not Observed 100% (1)

Results from this table show the lowest overall compliance among the use of these types of equipment and the targeted trades was the use of mobile scaffolding by painters (66.67% compliance). However, due to this being a single observation, the most notable lowest mean compliance was attributed to carpenters not using personal fall arrest systems correctly (87.04%). Overall, ironworkers were the most frequent users of personal fall arrest systems and aerial lifts, with safety compliance over 90%. Electricians were the most frequent users of ladders by far, but with higher compliance than the other targeted trades, including carpenters, with the lowest compliance of 98.26%. Carpenters contributed to the lowest compliance and most consistent use of mobile scaffolding (excluding the single observation of a painter using a mobile scaffolding system with 66.67% compliance to safety protocols).

Consistent with the second goal of this study, mean safety compliance was computed for each of the five types of targeted equipment included in this analysis (i.e. ladders, aerial lifts, personal fall arrest system, guardrails, scaffolding) across each phase of construction (i.e. concrete pouring/placement, skin, interior rough-end, roofing, interior finishes). Results are displayed in Table 6.

Table 6: Mean Safety Compliance for each Type of Targeted Equipment over each Phase of Construction
 

Concrete Pouring/Placement Phase

Mean Safety Compliance (Number of Observations)

Skin Phase

Mean Safety Compliance (Number of Observations)

Interior Rough-End Phase

Mean Safety Compliance (Number of Observations)

Roofing Phase

Mean Safety Compliance (Number of Observations)

Interior Finishes Phase

Mean Safety Compliance (Number of Observations)

Ladders 98.2% (5) 99.2% (136) 99.1% (125) 99.2% (141) 99.3% (114)
Aerial Lifts 100% (2) 99.3% (38) 99.6% (35) 99.3 (38) 99.2% (33)
Personal Fall Arrest System 88.5% (12) 96.1% (90) 96.0% (87) 96.1% (90) 98.6% (52)
Guardrails 100% (20) 99.6% (229) 98.8% (217) 99.7% (238) 99.4% (153)
Scaffolding 99.5% (18) 99.1% (88) 99.2% (82) 99.1% (89) 98.8% (62)

Of the 644 observations of the five types of equipment/worksite elements during the five phases of construction, the lowest compliance was found with the use of personal fall arrest systems during the concrete pouring/placement phase (88.5%). Likewise, safety compliance regarding the use of personal fall arrest systems was consistently lowest across all phases of construction. Overall mean safety compliance was lowest during the concrete pouring/placement phase, although the mean safety compliance for both aerial lifts and guardrails was 100% during this phase.

CHANGES/PROBLEMS THAT RESULTED IN DEVIATION FROM THE METHODS

Originally, the intent was to have two on-site observations per week, but given the necessity of the site superintendent to escort the researcher around the site, it was unreasonable to ask for two site visits per week given the site superintendent’s schedule.

APPLICABLE RESULTS AND RELEVANCE

The fall safety record of this construction project was excellent, as no accidents resulted from a fall of six feet or higher. However, this study did illuminate some opportunities to improve fall safety. First, the most commonly observed issues included the improper use of mobile scaffolds (specifically, not locking wheels while the scaffold is in use) and the improper use of safety harnesses, which includes issues ranging from not tying off to a structurally-sound point or the omission of use altogether while performing a task that requires the use of one. Additional recurring issues observed on the worksite included workers demonstrating improper use of ladders, including climbing techniques (i.e. climbing with tools in hand, not facing the ladder while climbing), working from the top rung, or choosing to use a ladder that is inappropriate for the task.

Worksite elements that provide fall safety were found to also have issues throughout the course of the project, including damage to or absence of guardrails along the perimeter of the atrium and along the stairwells. Roof sheathing was only observed during the first five observations with no issues. However, when assessing the usability and relevancy of the GAFR assessment tool, it is of the researcher’s opinion that this worksite element does not provide any semblance of fall safety and should therefore not be included in the analysis nor the next iteration of the GAFR. Likewise, safety net systems were not used on this project and were not observed; therefore, this element was not included in the analysis.

The phases of construction overlapped throughout the majority of the project, providing a difficult way to measure risk throughout each individual phase. According to frequency of issues resulting in heightened risk of falling due to some instance of non-compliance to safety protocols, the issues with safety non-compliance occurred during the concrete pouring/placement phase, although it’s brief period of observation may account for this. Likewise, safety compliance was lowest regarding equipment usage/presence during this period with the use of personal fall arrest systems. However, safety compliance did not appear to differ greatly across phases of construction.

The workers found to be most at risk of falling were ironworkers, carpenters, and roofers, as they accounted for the majority of instance of non-compliance to safety protocols. Along with ironworkers, carpenters were observed using personal fall arrest systems, accounting for lowest compliance among trades using the targeted construction equipment. Therefore, ironworkers and carpenters using personal fall arrest systems can be considered worker populations more vulnerable to falls and should therefore be targeted for fall prevention education. The most frequently recurring observation was the use of ladders by electricians, and although the mean safety compliance remained above 95%, they are considered more vulnerable to the risk of falling.

The unique architectural design of this building presented some opportunities for innovation in terms of construction fixed scaffolding or ladder usage and worker’s methods for following safety protocols, such as using a harness and tying off to an appropriate structure. Ironworkers were most frequently involved in accessing difficult-to-reach locations, and were more likely to make compromises in order to effectively complete their tasks.

FUTURE FUNDING PLANS

Funding opportunities to expand this work by using the GAFR in repeat assessments in multiple sites will be explored next.

PRESENTATIONS AND PUBLICATIONS

This project was presented twice at The George Washington University’s Research Days (2013 and 2014), first as an introduction to this project and then with preliminary findings.

A manual describing the research study has been uploaded to the CPWR website and continues to be distributed throughout various construction safety networks.

DISSEMINATION PLAN

The research team will work with CPWR’s dissemination group to market this work to major stakeholders, particularly in the general construction trades. Likewise, the research results and the tool will be presented at a professional association or construction safety conference (e.g. Washington Metro Area Construction Safety Association (WMACSA) event or the American Society of Safety Engineers (ASSE) conference).

REFERENCES

Becker, P., M. Fullen, M. Akladios, M. Carr and W. Lundstrom (2001). "Use of a handheld computer to audit construction fall prevention effectivenes." Int J Comput Integrated Des Construct 3: 16-24.

Becker, P., M. Fullen, M. Akladios and G. Hobbs (2001). "Prevention of construction falls by organizational intervention." Inj Prev 7 Suppl 1: 64-67.

Courtney, T. K., S. Matz and B. S. Webster (2002). "Disabling occupational injury in the US construction industry, 1996." J Occup Environ Med 44(12): 1161-1168.

Courtney, T. K., G. S. Sorock, D. P. Manning, J. W. Collins and M. A. Holbein-Jenny (2001). "Occupational slip, trip, and fall-related injuries -can the contribution of slipperiness be isolated?" Ergonomics 44(13): 1118-1137.

Dennerlein, J. T., C. J. Ronk and M. J. Perry (2009). "Portable ladder assessment tool development and validation – Quantifying best practices in the field." Safety Science 47(5): 636-639.

Derr, J., L. Forst, H. Y. Chen and L. Conroy (2001). "Fatal falls in the US construction industry, 1990 to 1999." J Occup Environ Med 43(10): 853-860.

Dong, X. S., A. Fujimoto, K. Ringen and Y. Men (2009). "Fatal falls among Hispanic construction workers." Accident Analysis & Prevention 41(5): 1047-1052.

Fullen, M. and W. Lundstrom (2008). Fall Hazard Identification and Control Audit of the Mandarin Building City Center Project CPWR - The Center for Construction Research and Training.

Kaskutas, V., A. M. Dale, H. Lipscomb, J. Gaal, M. Fuchs and B. Evanoff (2010). "Fall prevention among apprentice carpenters." Scand J Work Environ Health 36(3): 258-265.

Kaskutas, V., A. M. Dale, J. Nolan, D. Patterson, H. J. Lipscomb and B. Evanoff (2009). "Fall hazard control observed on residential construction sites." Am J Ind Med 52(6): 491-499.

Kaskutas, V. K., A. M. Dale, H. J. Lipscomb and B. A. Evanoff (2008). "Development of the St. Louis audit of fall risks at residential construction sites." Int J Occup Environ Health 14(4): 243-249.

Leamon, T. B. and P. L. Murphy (1995). "Occupational slips and falls: more than a trivial problem." Ergonomics 38(3): 487-498.

Lipscomb, H. J., J. M. Dement, J. Nolan, D. Patterson, L. Li and W. Cameron (2003). "Falls in Residential Carpentry and Drywall Installation: Findings From Active Injury Surveillance With Union Carpenters." Journal of Occupational and Environmental Medicine 45(8): 881-890.

Lombardi, D. A., G. S. Smith, T. K. Courtney, M. J. Brennan, J. Y. Kim and M. J. Perry (2011). "Work-related falls from ladders - a follow-back study of US emergency department cases." Scand J Work Environ Health.

Perry , M. J. and C. J. Ronk (2010). Preventing Falls from Ladders in Construction Boston, Harvard School of Public Health.

Ronk, C. J., J. T. Dennerlein, E. Hoffman and M. J. Perry (2011). "Is renovation riskier than new construction? An observational comparison of risk factors for stepladder-related falls." American Journal of Industrial Medicine 54(8): 579-585.

Smith, G. S., R. A. Timmons, D. A. Lombardi, D. K. Mamidi, S. Matz, T. K. Courtney and M. J. Perry (2006). "Work-related ladder fall fractures: Identification and diagnosis validation using narrative text." Accident Analysis & Prevention 38(5): 973-980.

Sparer, E. H. and J. T. Dennerlein (2013). "Determining safety inspection thresholds for employee incentives programs on construction sites." Safety Science 51(1): 77-84.

Wang, E. (1999). "Mortality Among North Carolina Construction Workers, 1988-1994." Applied Occupational and Environmental Hygiene 14(1): 45-58.

Wickstrom, G. and T. Bendix (2000). "The "Hawthorne effect" - what did the original Hawthorne studies actually show?" Scandinavian Journal of Work Environment & Health 26(4): 363-367.

 


GW Audit of fall risk

Ladder portion of survey

Aerial Lift and Personal Fall arrest portion of survey

Guardrail and scaffold safety device questions on survey

Safety Net System and Roof Sheating categories on survey

 


Evaluating Fall Safety Compliance among Skilled Trades in Construction
CPWR Small Study
Final Report
Melissa J. Perry, Amanda McQueen, SeungSup Kim
George Washington University
© 2015, CPWR – The Center for Construction Research and Training.

CPWR logo

CPWR, the research and training arm of the Building and Construction Trades Dept., AFL-CIO, is uniquely situated to serve construction workers, contractors, practitioners, and the scientific community. This report was prepared by the authors noted. Funding for this research study was made possible by a cooperative agreement with the National Institute for Occupational Safety and Health, NIOSH (OH009762). The contents are solely the responsibility of the authors and do not necessarily represent the official views of NIOSH or CPWR.