Construction

Electrical deaths in the US construction: an analysis of fatality investigations

Dong Zhao a *, Walid Thabet

a , Andrew McCoy

a and Brian Kleiner

b

a Department of Building Construction, Virginia Polytechnic Institute and State University, Blacksburg, VA 24601, USA;

b Department of

Industrial and Systems Engineering, Virginia Polytechnic Institute and State University, Blacksburg, VA 24061, USA

(Received 13 June 2013; final version received 8 July 2013)

Electrocution is among the ‘fatal four’ in US construction according to the Occupational Safety and Health Administration. Learning from failures is believed to be an effective path to success, with deaths being the most serious system failures. This paper examined the failures in electrical safety by analysing all electrical fatality investigations (N ¼ 132) occurring between 1989 and 2010 from the Fatality Assessment and Control Evaluation programme that is completed by the National Institute of Occupational Safety and Health. Results reveal the features of the electrical fatalities in construction and dis- close the most common electrical safety challenges on construction sites. This research also suggests the sociotechnical system breakdowns and the less effectiveness of current safety training programmes may significantly contribute to work- er’s unsafe behaviours and electrical fatality occurrences.

Keywords: electrical safety; construction; fatality investigations; accident prevention

1. Introduction

The construction industry remains a dangerous product sec-

tor in fatality and injury world widely (Gholipour, 2004). In

the United States, electrocution is among the ‘fatal four’ in

construction industry according to the Occupational Safety

and Health Administration (OSHA). Contact with electric-

ity ranks 4th in the top 10 accident events that lead to con-

struction fatalities, following falls to a lower level,

roadway vehicle incident, and being struck by object or

equipment (see Table 1). In 2011, 69 deaths were due to

electrocution, accounting for 9.3% out of 738 construction

fatalities (US Bureau of Labor Statistics, 2013).

Also as shown in Table 2, the electrical fatalities in

the construction industry have been approximately as

many as the sum of all other non-construction industries

for years. During the recent nine years, the smallest pro-

portion of electrical deaths that the US construction

engaged was 39.7%, indicating that among every five

electrocutions at least two occurred in construction. On

the contrary, the construction sector was involved in

smaller proportions of fatality due to other top events. For

instance, the US construction industry incurred 38.4% of

the overall falling fatalities, 8.6% of roadway fatal inci-

dents and 15.3% of being-struck fatal injuries in 2011. In

other words, the construction industry has the highest per-

centage of electrocution and its workers encounter the

highest risk from electrical injuries in workplaces (Zhao,

Thabet, McCoy, & Kleiner, 2012). Like many construc-

tion incidents, many of these are preventable.

The frequency of being killed by electricity in con-

struction continues to be higher than the industry average.

The electrocution rate is a measure of the number of

deaths due to electrocution within a population in a given

time period. For 1992 to 2002, Cawley and Homce (2008)

claimed that the electrocution rate for construction was

over five times that for all industry level. For 2003

to 2011, the electrocution rate for construction was

1.5 deaths per 100,000 full-time construction workers,

while the all industry average rate was 0.2 per 100,000

workers (US Bureau of Labor Statistics, 2012, 2013).

The construction electrocution rate has risen to be more

than seven times that the average of all industry as a

whole.

Losses due to electrical accidents in construction are

significant in both physical and societal traumas. Electri-

cal accidents involve electric shocks, electrical burns, arc

blasts, of which most result in severe tissue damage or

even mortality as high as 15% (Lee & Dougherty, 2003).

In terms of financial expense, average losses due to elec-

trocution were $948,844 per fatal case and $86,829 per

non-fatal case, which respectively, ranked the first highest

per-fatal-case cost and the second highest per-

nonfatal-case cost in construction (National Institute of

Occupational Safety and Health, 2006; Waehrer, Dong,

Miller, Haile, & Men, 2007).

Efforts to explore either electrical injuries or construc-

tion accidents have been undertaken by some researchers.

Niskanen and Saarsalmi (1983) conducted a frequency

*Corresponding author. Email: [email protected]

� 2013 Taylor & Francis

International Journal of Injury Control and Safety Promotion, 2014

Vol. 21, No. 3, 278–288, http://dx.doi.org/10.1080/17457300.2013.824002

analysis using different factors to investigate construction

accidents; Haslam et al. (2005) examined construction

accidents and distilled contributing factors based on both

of the site-based and off-site-based investigations; Hale,

Walker, Walters, and Bolt (2012) explored the construc-

tion fatal accidents using a standard of four-level classifi-

cation based on the science of human factors analysis;

Sawacha, Naoum, and Fong (1999) analysed the impacts

of the historical, economical, psychological, technical,

procedural, organisational and the environmental issues to

the construction safety; Behm (2005) linked the construc-

tion injuries to the design for construction safety or pre-

vention through design concept. As well, several other

studies emphasising the electrical injury in terms of medi-

cine and health (Taylor, McGwin, Valent, & Rue, 2002;

Cawley & Homce, 2008) did not concentrate on the US

construction industry. However, a gap of accident analysis

on the combined area of the electrical fatality in construc-

tion still to some extent exists.

The present study follows prior research but with an

attempt to fill the gaps. The objective of the study is,

through an examination of electrical fatality investiga-

tions, to explore the features of electrical fatalities in con-

struction and disclose the most common safety challenges

on construction sites. Further, accident investigations

include more hidden information that could be retrieved

rather than mere statistical numbers, which this work

collects and analyses as well. Although not funded

directly, the objective of this study supports the National

Occupational Research Agenda (NORA Construction

Sector Council, 2008), in which electrical hazards serve

as the number two focus area within the construction

sector.

2. Data preparation

The electrical fatality data source in this research is the

Fatality Assessment and Control Evaluation (FACE) pro-

gramme which is compiled by the National Institute of

Occupational Safety and Health (NIOSH, 2010). The

FACE programme provides full text of hundreds of fatal-

ity investigation reports since 1982, allowing of the identi-

fication of contributing factors to fatal injuries as well as

the comprehensive recommendations for preventing simi-

lar deaths. The reasons for choosing FACE as data source

of this study are: (1) it provides more explicit narratives,

detailed contexts and professional investigations

compared to only statistical numbers; (2) it is a key occu-

pational safety and health surveillance resource for con-

struction that is recommended by the NORA; (3) learning

from failures is believed to be an effective path to success

as fatalities are the most serious system failures (Beavers,

Table 1. Top 10 accident events leading to construction fatalities in 2011.

Accident events N Pct. (%)

Falls to a lower level 255 34.6 Roadway vehicle incident 95 12.9 Stuck by object or equipment 73 9.9 Contact with electricity 69 9.3 Pedestrian vehicular incident 65 8.8 Crushed in collapsing structure, equipment or material 30 4.1 Non-roadway vehicle incident 25 3.4 Exposure to other harmful substance 26 3.5 Intentional injury by person 22 3.0 Caught in or compressed by equipment or objects 18 2.4 Misc. 60 8.1 Total 738 100

Note: Data source is from the Census of Fatal Occupational Injuries (US Bureau of Labor Statistics, 2013).

Table 2. Electrical fatalities in construction and all other industries, 2003–2011.

2003 2004 2005 2006 2007 2008 2009 2010 2011

Construction industry N 132 122 107 126 108 89 89 76 69 Pct. (%) 53.7 48.0 42.6 50.4 59.9 46.4 52.4 46.6 39.7

All other industries N 114 132 144 124 104 103 81 87 104 Pct. (%) 46.3 52.0 57.4 49.6 49.1 53.7 47.7 53.4 60.3

Total counts 246 254 251 250 212 192 170 164 171

Note: Data source is from CFOI (US Bureau of Labor Statistics, 2013).

International Journal of Injury Control and Safety Promotion 279

Moore, & Schriver, 2009; Chi, Yang, & Chen, 2009;

Health and Safety Executive, 1988) and (4) its representa-

tiveness was validated through a pre-conducted T-test on

victims’ age, gender and occupation between the datasets

from FACE and the Census of Fatal Occupational Injuries

(CFOI).

FACE investigations are composed of two divisions:

the NIOSH FACE that is conducted by NIOSH and the

state FACE that is conducted by NIOSH’s cooperative

state partners (for example, the Massachusetts FACE is

known to be an exemplary programme). Both divisions

use the same FACE model. The NIOSH FACE began in

1982 and targeted traumatic occupational fatalities result-

ing from the death causes of confined spaces, electrocu-

tion, machine-related, falls from elevation and motor

vehicles. The state FACE began in 1989 and investigated

fatal accidents of both NIOSH-level targets and state-

level targets, which included falls, electrocutions, suicides

and homicides, transportation fatalities, worker deaths

involving toxicological issues and chemical-related fatali-

ties. With the exception of authors, these two divisions

are neither different in format nor overlapped in content.

The scope of data collection was confined to the

FACE investigations with the cause of electrocution in

construction from 1989 through 2010. FACE reports can

be indexed by industry, fatality cause or populations. Con-

struction is one category that could be indexed by indus-

try, and Electrocution is another category indexed by

fatality cause. Reports under the category of Construction

are cases of which the victims belong to the construction

industry. As well, the category of Electrocution is the col-

lection of different types of electric shocks and electrocu-

tion accidents that occurred. Therefore, the overlapped

cases under category Construction and category Electro-

cution from both NOISH FACE reports and State FACE

reports were consequently targeted as research objects,

since they were eligible to present the research scope of

construction electrocution. In addition, electrical deaths

occurring in 2010 were the most updated cases used in

this study as of December 2012.

The data preparation was manually conducted on the

897 construction fatality investigations that were public

assessable from the FACE programme website (NIOSH,

2010). Since FACE reports did not have a combined

category for electrocutions in construction, three basic fil-

tering criteria were applied: (1) the victim died at work;

(2) the cause of death was electrocution and (3) the

employer of victim belonged to the construction industry.

The definition of being caused by electrocution was

according to the decedent’s death certificate. The defini-

tion of construction complied with the 2010 North Ameri-

can Industry Classification System (NAICS) in which the

construction industry was defined between code 230000

and 238990 (US Census Bureau, 2012). In this way, a total

of 132 qualified FACE reports with 140 fatalities were

selected for following analysis (some accidents resulted in

multiple victims).

3. Methods

3.1. Factor framework

A factor framework of 15 factors was established for con-

tent analysis and information organising. The framework

was created and refined through literature reviews and

expert consultations. At the beginning, an initial version

of the factor framework was developed using a fishbone

diagram to cover the fatality time, entities, circumstances,

media and causes, all of which can fully imply the fatality

features in terms of ‘when’, ‘who’, ‘what’, ‘how’ and

‘why’. According to the data accessibility and research

objectives, after several iterations, the finalised factor

framework was shown in Table 3. The factor categorisa-

tion was determined with reference to existing classifica-

tions and industry regulations, such as NAICS, the

Standard Occupation Classification (SOC) and the Inter-

national Electrotechnical Commission (IEC) standards.

Previous research supports the use of iterative pro-

cesses to establish a final set of relative factors. Ling, Liu,

and Woo (2009) chose factors F1, F2, F6, F7 and F8 as

factors in their research on construction fatalities in Singa-

pore. Beavers et al. (2009) identified factors F3, F4, F5

and F13 while investigating steel erection fatalities in the

construction industry. Hinze, Pedersen, and Fredley

(1998) established factors F1, F2, F3, F7, F8 and F10 for

the analysis on root causation of construction injuries.

Huang and Hinze (2003) examined factors F1, F2, F6, F7,

F8, F9 and F10 in their study on fall accident analysis of

construction workers. Janicak (2008) analysed factors F6,

F7 and F13 in his research on occupational fatalities due

to electrocutions. Chi et al. (2009) selected F7, F8, F11

and F13 as factors to determine the cause of electric shock

in the construction industry. Mullins (2005) utilised fac-

tors F1, F4, F5, F7 and F8 to evaluate safety climate defi-

ciencies in construction fatalities.

3.2. Content analysis and information extraction

As part of the review process for cases, the research team

analysed information, retrieved from the FACE content,

which might not be obvious, but could be important as a

factor. Two specific methods of text analysis were used in

this process: (1) using text patterns or key words that

match such regular expressions to identify small or large-

scale structure, e.g. ‘safety training programme’ and

(2) using text analytics to attempt to understand the text

and link it to other information. Taking FACE investigat-

ing report #10MA019 (Massachusetts FACE, 2011) for

example, the accident time from text that ‘3 August 2010’

in this case was captured and categorised into factors

280 D. Zhao et al.

Table 3. Factor framework.

No. Factor name Elements

F1 Month From January to December F2 Weekday From Monday to Sunday F3 Employer’s industry

a Residential BC (2361) Non-residential BC (2362) Utility system construction (2371) Land subdivision (2372) Highway, street and bridge construction (2373) Other heavy and civil Engineering construction (2379) Foundation, structure and building exterior contractors (2381) Building equipment contractors (2382) Building finishing contractors (2383) Other specialty trade contractors (2389)

F4 If the employer has a written safety policy? Yes No

F5 If the employer provides a safety training programme? Yes No

F6 Victim’s occupation b

Electricians (47-2110) Line installers and repairers (49-9050) Supervisors (47-1010) Carpenters (47-2030) Cement masons, concrete finishers, and terrazzo workers (47-2050) Construction Labours (47-2060) Construction equipment operators (47-2070) Pipe layers, plumbers, pipefitters and steamfitters (47-2150) Roofers (47-2180) Structural iron and steel workers (47-2220) Painters and paperhangers (47-2140) Insulation workers (47-2130)

F7 Victim’s age c

16–19 20–24 25–34 35–44 45–54 55–64 65þ

F8 Victim’s gender Male Female

F9 Agent(s) that victim touched Direct contact with electrical wire Dump truck, boomed crane or other mechanical equipment Pipes, poles, or other conductive objects Energised ladders, scaffolds, or other support equipment Antenna, arrestors, gates, or other building parts

F10 Physical work environment Exposed Unexposed

F11 Project type Residential building construction Non-residential building construction Heavy and civil construction

F12 Voltage level (volts) d

Below 1000 1000–15,999 16,000–34,999 35,000 and above

F13 Electricity origin e

Power lines (both of overhead and underground) Transformers, conductors, panels or other electrical components Powered machinery, tools, appliances, equipment or light fixture.

F14 Human error origin Victim self Third person

F15 NIOSH prevention recommendations See Table 5.

a F3 elements were classified based on NAICS 2012 and numbers in the parentheses refer to NAICS codes. b F6 elements were classified based on SOC 2010 and numbers in the parentheses refer to SOC codes.

c F7 elements were classified based on the NIOSH age classification. d F12 elements were classified based on the IEC standard 60038.

e F13 elements were classified based on the CFOI electrical source classification.

International Journal of Injury Control and Safety Promotion 281

F1 (August) and F2 (workdays); the victim’s demographic

information from text that ‘23-year-old male roofer’ was

captured and stored into factors F6 (roofers), F7 (age 16–

24) and F8 (male).

3.3. Exploratory analysis

An exploratory data analysis (EDA) technique was chosen

for data analysis in this study. Different from conventional

statistics, inductive reasoning begins with specific observa-

tions and measures, then detects patterns and regularities,

and finally ends up developing some general conclusions

or theories. EDA is advantageous as it does not rely on pre-

conceived notions on fatal accidents (Ling et al., 2009).

Specifically, frequency analysis and chi-squared test were

conducted to elaborately describe and interpret extracted

and categorised information from electrical fatality investi-

gations. It is important to note that frequencies presented in

this research do not necessarily reflect the risk level to

workers, but rather describe the problem’s proportional

magnitude. The team used IBM SPSS V20.0 as the techni-

cal software tool to generate analysis results.

4. Results: features of construction electrocutions

4.1. Time of electrocution occurrence

Based on data analysis, the frequencies of electrical fatal-

ity incidents occurring by month and weekday are pre-

sented in Figure 1. The number of electrical deaths in

construction peaked in August at 17.1% and bottomed in

January at 1.4% of the entire electrocutions. Summer was

the season with the most electrical accidents (39.3%)

while winter contained the fewest (10.7%). Also, 122

electrical fatalities (87.1%) occurred during workdays

and most of them were at the beginning of the week, on

Monday (20.7%) and Tuesday (22.9%).

Summer has the most favourable weather conditions

for outside activities and August has the highest average

temperature in North America. Based on findings, hot

weather conditions could contribute to lower awareness of

potential electric hazards for workers, especially when

conducting tasks in an open area. On the other hand, the

high volume of construction fatalities might simply be a

result of the high concentration of projects in warm areas

and during warm seasons. Combining features of work

and the rate of work together may also contribute to

effects that increase the number of accidents in summer

for certain types of work.

4.2. Employers of electrocution victims

Information about victims’ employers is presented in

Table 4, in which 28.6% of electrical deaths were associ-

ated with building equipment contractors categorised as

‘Specialty Trade Contractors’. These contractors install

and operate specialised building equipment, such as

cranes, boomed vehicles, elevators, escalators, service sta-

tion equipment and central vacuum cleaning systems.

Work scopes may include new work, additions, altera-

tions, maintenance and repairs. Also, special trade con-

tractors contain the highest percentage of electrical

fatality at 46.4%, while the category of ‘building con-

structors’ (BC) represents the lowest percentage at 18.6%.

Figure 2 shows that more than 50% of employers nei-

ther had a written safety policy nor provided a safety

training programme, according to the statements in FACE

investigations. This finding is surprising, as standards of

the OSHA explicitly require the employer to train

employees in the safety and health aspects of their jobs.

Hence, a lack of safety policy and training programmes

could affect the likelihood of electrical injuries to con-

struction workers.

Figure 1. Electrocution distribution by occurrence timing.

282 D. Zhao et al.

4.3. Victims’ demographic characteristics

Figure 3(a) shows the seven occupations with the highest

number of electrical fatalities. Line installers and repairers

(24.3%) and electricians (20.0%) account for the first and

third largest portions of electrical deaths, although they

typically received extensive training in the electrical

safety and the hazards associated with electrical energy.

Moreover, construction labourers (21.4%), who generally

receive little or no electrical training, rank the second

highest occupation for electrocutions. Ten percent of

victims are construction equipment operators such as

construction crane operators, dump truck drivers and

boom-mounted vehicles operators. All the occupations

contributing less than 3% merged into the category of

‘Others’, include plumbers (2.9%), carpenters (2.1%),

Table 4. Electrocution distributions by victims’ employer.

F3: Employer/industry N a

Pct. (%)

Building constructions (BC) 26 18.6 Residential BC 14 10.0 Non-residential BC 12 8.6

Heavy and civil engineering construction 49 35.0 Utility system construction 35 25.0 Land subdivision 1 0.7 Highway, street and bridge construction 7 5.0 Other heavy and civil engineering construction 6 4.3

Specialty trade contractors 65 46.4 Foundation, structure and building exterior contractors 10 7.1 Building equipment contractors 40 28.6 Building finishing contractors 6 4.3 Other specialty trade contractors 9 6.4

Total 140 100.0

a Rounding off error may occur in calculating percentage.

Figure 2. Electrocution distribution by employer’s safety status.

Figure 3. Electrocution distribution by victim’s occupation and age group. Note: Missing data may cause the accumulated percentage shown above to total less than 100%.

International Journal of Injury Control and Safety Promotion 283

structure steel workers (1.4%) and cement masons (1.4%).

Of the 12 various occupations involved in construction

electrical accidents, the top 5 occupations cumulatively

represent 81.4% of the total electrical fatalities. Rather

than line installers/repairers and electricians, non-

electrical occupations, who generally encounter few elec-

tric sources in their regular work, account for 55.7% of

total electrical fatalities.

Victims’ ages range from 16 to 61 years old with the

mean of 35.2 years. Among all victims, there is only one

under 18, a 16-year-old electrical-contractor labourer who

was electrocuted in 1996 due to energised power lines.

As shown in Figure 3(b), victims in age group of 25–34

dominate the electrical fatalities at 30.71%, followed by

the group of 35–44 at 28.57%. Young victims dying prior

to age 55 are substantial, accounting for over 90%. For

the premature fatalities due to electrocution in this

research, the years of potential life lost (YPLL) before

age 65 equalled 4021 years (Gardner & Sanborn, 1990).

Pertaining to the victim’s gender, data from factor F8

(gender) indicated 100% of them were male, the dominant

gender of the industry.

4.4. Circumstances of electrical accidents

As shown in Figure 4(a), live electrical wires are the agent

of 40% of electrocution victims, who directly touched

them in some way. Electrical wires include overhead

power lines and underground power lines. Construction

machines such as dump trucks, boomed cranes and aerial

buckets are the agents that the second most victims were

touching, accounting for 28.6%. Interestingly in these

data is that the substitution of aerial buckets for ladders as

a safety control does not seem to eliminate the risk of

Figure 4. Electrocution distributions by electrocution circumstance. Note: Missing data may cause the accumulated percentage less than 100% shown above.

284 D. Zhao et al.

electrocution. The data also indicate that supporting

equipment such as aluminium ladders or scaffolds and

construction materials such as pipes rank as the third and

fourth riskiest agents in being electrocuted, respectively,

accounting for 16.4% and 12.1% of electrical deaths.

Figure 4(b) indicates that more than three-quarters

(76.4%) of electrical fatal accidents occurred in exposed

construction sites. Such outdoor working environments

might be impacted by natural conditions such as weather,

sunshine, temperature and humidity (similar to findings of

‘timing’ above). The minimum clearances of live electri-

cal source also differ between outdoor and indoor working

environments since the outdoor clearance is usually lon-

ger than the indoor (National Fire and Protection Associa-

tion, 2008).

Regarding the occurrence of construction incidents

by type of work (see Figure 4(c)), heavy civil projects

possess the largest portion of fatalities (45.0%), which

was followed by residential building projects (30.7%)

and non-residential building projects (24.3%). The most

frequent heavy civil projects that involve incidents are

power transmission, distribution substation, road, bridge

and gas station constructions. Residential building proj-

ects include the construction and repair of houses,

apartments and condominiums. No-residential building

projects involve manufacturing facility, warehouse, shop-

ping mall, store, school, commercial office and recrea-

tion facilities.

As shown in Figure 4(d), 61.4% of electrical fatalities

in construction involve the alternating current of 1000–

15,999 V. The voltages within this range, such as standard

voltage of 4160 V, 7200 V, 12,470 V, 13,200 V or

14,470 V, are usually used for the local power distribu-

tion. Specifically, voltage of 7200 V dominates for 23.6%

of all electrocutions and 23.6% of electrical accidents that

involved low voltages are less than 1000 V (National Fire

and Protection Association, 2008; Reese & Eidson, 2006;

International Electrotechnical Commission, 2002). The

combined voltage level for long-distance distribution

(16,000–34,999 V) and transmission (35,000 V and

above) approximately account for 11% of all electrical

fatalities.

4.5. Origins of electrical accidents

The research team also examined the origins of the electri-

cal fatality, with human error as a possible origin. As

shown in Figure 5, 52.9% of the electrical fatalities in

construction derive from energised power lines and

89.3% of those fatalities result from the victim’s improper

operation or insufficient hazard awareness. These data

suggest that power lines, especially overhead ones, are the

primary electrical hazard in US construction. It is impor-

tant to note that 10.7% of electrical deaths were related to

the conduct of a third person, where the victim was not

necessarily at fault.

4.6. Prevention controls

FACE investigators provided several recommendations at

the end of each report. These recommendations targeted

incident causes and system defects, and thus could be

used for injury control and safety promotion. Each listed

recommendation might not be stated exactly in a similar

way, but the meaning of each one can be generalised

(Kunadharaju, Smith, & DeJoy, 2011). In this regard, the

team read every recommendation from every FACE

report and examined and categorised these statements as

well.

The results (see Table 5) show that within the exam-

ined 132 electrocution reports, 62.9% incidents (n ¼ 83)

Figure 5. Electrocution distributions by origin.

International Journal of Injury Control and Safety Promotion 285

suggest a need to provide adequate and effective safety

training on electrical hazard identification and prevention.

Providing appropriate equipment and conducting jobsite

hazard surveys tie for the second most recommendations

at 47.7%. This content suggests that the enhancement of

electrical hazard awareness can be critical for construction

workers to avoid being electrocuted and that the training

is an axiomatic part of injury prevention strategy.

To further explore the impact of safety training to

other electrocution features, the team conducted chi-

squared tests between F5 (safety training) and each of the

other factors. Results statistically support the differences

between F5 and F1 month (p ¼ 23.124, significance ¼ 0.017); F5 and F3 employer (p ¼ 0.251, significance ¼ 0.012); F5 and F4 safety policy (p ¼ 0.601, significance < 0.001); F5 and F7 age group (p ¼ 0.267, significance ¼ 0.045); and F5 and F11 project type (p ¼ 18.321, signifi- cance < 0.001). Differences between F5 and each of the rest factors (F2, F8, F9, F10, F12, F13 and F14) could not

be supported. As suggested in the FACE investigator

statements, safety training (F5) is a critical part of the

industry that must be considered by employers, age

groups, project types and be part of the written training

policy.

5. Conclusions and discussion

Electrocution is among the ‘fatal four’ in US construction,

according to the OSHA. Learning from failures is

believed to be an effective path to success, with fatalities

being the most serious system failures. As a result, this

paper explored failures in electrical safety by analysing

all electrical fatality investigations from the FACE pro-

gramme completed by the National Institute of

Occupational Safety and Health (NIOSH). A total of 132

FACE investigations with 140 victims from 1989 to 2010

were selected and examined. The data are partially repre-

sentative, which is supported by the pre-conducted T-test

on victim’s age, gender and occupation between the data-

sets from FACE and the CFOI. Nevertheless, possible sta-

tistical limitation ascribed from the limited number of

FACE cases remains.

Results reveal the typical features of the electrical

fatalities in construction and disclose the most common

electrical safety challenges on construction sites. Extra

care with electrical hazards should be taken when working

in hot weather timing since electrical fatalities were sig-

nificantly dense in summer, especially in August. Both

exposed working environments in construction and rela-

tively high frequency of construction projects during this

season pose another explanation. Firms such as construc-

tion equipment contractors, utility construction contrac-

tors and residential builders are commonly involved in

electrical accidents and should pay particular attention to

electrocution prevention efforts for their employees, at a

minimum the OSHA required. Occupations particularly

susceptible to electrocution include line installer and

repairer, construction labourer, electrician and construc-

tion machine operator. Data suggest that young male

workers within the age 25–44 bear higher risk of getting

electrically shocked. Such age data might also include

young construction workers (within the lower part of this

age range) that are less matured in hazards awareness and

lack safe practical experiences, which could be a topic of

future research. Outdoor tasks involving power lines,

boomed vehicles and supporting equipment such as lad-

ders and scaffolds, are exposed to a relatively higher elec-

trical risk and thus require additional safety training and

Table 5. Top-ranked NIOSH recommendations.

NIOSH recommendation N Pct. (%)

Adequate safety training and periodic specialised electrical safety training programmes should be implemented to enhance the electrical hazard cognition and the avoidance of unsafe conditions in workplace.

83 62.9

Well-designed non-conductive personal protective equipment (PPE), communication equipment and supporting equipment should be provided and enforced to workers in workplace.

63 47.7

An electrical hazard survey should be conducted at jobsite to identify potential electrical hazards and intervention measures before work.

63 47.7

Compliance with safety procedures that required by existing federal and state standards and regulations should be ensured, such as the proper grounding, minimum clearance and lock-out/ tag-out procedures.

56 42.4

Power lines should be de-energised or insulated before all works start. 45 34.1 On-site safety procedures, safety meeting and safety inspection should be enforced at construction

site on a routine base. 44 33.3

Electrical safety procedures and preventions should be thoroughly considered and improved at the construction planning stage.

23 17.4

Guarding co-workers, warning signs and the supervisory guidance should be ensured on site. 18 13.6

286 D. Zhao et al.

possible countermeasures. More than half of construction

electrocutions originated from power lines for local distri-

bution systems with voltage ranging from 1 kV (1 kV ¼ 1000 V) to 16 kV, which are worthy of special attention

in terms of hazard surveys and safety inspections.

Of interest, when comparing these electrical fatality

findings with census statistics (e.g., CFOI), a similarity of

occurrence time, victim’s demographic characters exists

and is methodologically supported. Current findings on

fatality entities, accident circumstances and electrocution

origins also supplement missing data from other previous

studies. As a result of consistency across data types and a

lack of coverage, the reported electrocution features and

safety challenges can also be used as a basis to initiate fur-

ther thinking on the fatality mechanisms and preventions

for the larger construction industry.

Outside of the statistical findings, sociotechnical sys-

tem breakdowns seem to provide an essential contribution

to electrocution occurrence. Haslam et al. (2005) sup-

ported this concept with findings in which worker actions

and behaviours, as an involving factor, determines 49% of

construction accidents. Coupled with NIOSH prevention

recommendations, findings suggest that these sociotechni-

cal system breakdowns are commonly associated with a

failure to identify electrical hazards involved in complet-

ing a task or the incorrect use of equipment (Strauch,

2002). Errors can result from the lack of knowledge, task

inexperience and deficiencies in training (Read, Lenn�e, & Moss 2012; Hasan & Jha, 2012), which also confirms

NIOSH’s top recommendation of implementing effective

safety training for electrical hazard cognition and unsafe

condition avoidance in the workplace. It is especially

important for hazard awareness training to be a major

goal in reducing electrical accidents (Zhao, Lucas, &

Thabet, 2009). The authors suggest such breakdowns to

often be system-based rather than only worker error

(Kleiner, Smith-Jackson, Mills, O’Brien, & Haro, 2008).

From a hierarchy of controls design perspective, risk

should be designed out of features of work, with adminis-

trative controls such as training and personal protective

equipment (PPE) as mandatory measures, whose reduc-

tion of risk is a final layer of protection and not a protec-

tion strategy necessarily. Trained victims account for

approximately half of the victims in this study, which

implies that basic accident prevention might not be

enough if other factors such as project hazard level and

safety culture level present a high risk (Feng, 2013). Also,

the chi-squared tests here cannot statistically validate dif-

ferences between trained cases and untrained cases for

many factors, suggesting that the effectiveness of current

safety training programmes for electrical safety in con-

struction might be inadequate and could have the ability

to decrease unsafe behaviour and mitigate differing types

of accidents. Further, as concluded by Huang and Hinze

(2003), traditional safety training may not be sufficient to

enable construction workers to detect and eliminate the

broad array of potential hazards, and therefore, innovative

training approaches should be considered.

This study proposes challenges of electrical safety in

construction, but several limitations exist and some areas

need to be addressed in future research. One limitation is

due to the relatively small data size of FACE investiga-

tions. To minimise any possible bias, which may be

caused by this limitation, a T-test was conducted whose

results supported an avoidance of this bias. Moreover,

data from fatality reports are strong in a comprehensive

context, which includes more information than census sta-

tistics. Another limitation is the possibility of subjective

opinions from investigators. The team, therefore, used

precautions for mitigating this limitation when designing

the factor framework. To a great extent, more objective

factors were chosen for analysis. For future studies, partic-

ular concerns on safety training and hazard design-

for-safety may need to be further investigated to address

significant fatality controls.

References

Beavers, J.E., Moore, J.R., & Schriver, W.R. (2009). Steel erection fatalities in the construction industry. Journal of Construction Engineering and Management, 135(3), 227– 234.

Behm, M. (2005). Linking construction fatalities to the design for construction safety concept. Safety Science, 43(8), 589– 611.

Cawley, J.C., & Homce, G.T. (2008). Trends in electrical injury in the US, 1992–2002. Industry Applications, IEEE Transac- tions on, 44(4), 962–972.

Chi, C., Yang, C., & Chen, Z. (2009). In-depth accident analysis of electrical fatalities in the construction industry. International Journal of Industrial Ergonomics, 39(4), 635–644.

Feng, Y. (2013). Effect of safety investments on safety perfor- mance of building projects. Safety Science, 59(0), 28–45.

Gardner, J.W., & Sanborn, J.S. (1990). Years of potential life lost (YPLL) – what does it measure? Epidemiology, 1(4), 322–329.

Gholipour, Y. (2004). Injury minimization in construction proj- ects. International Journal of Injury Control and Safety Promotion, 11(1), 63–65.

Hale, A., Walker, D., Walters, N., & Bolt, H. (2012). Developing the understanding of underlying causes of construction fatal accidents. Safety Science, 50(10), 2020–2027.

Hasan, A., & Jha, K.N. (2012). Safety incentive and penalty pro- visions in Indian construction projects and their impact on safety performance. International Journal of Injury Control and Safety Promotion, 20(1), 3–12.

Haslam, R.A., Hide, S.A., Gibb, A.G., Gyi, D.E., Pavitt, T., Atkinson, S., & Duff, A.R. (2005). Contributing factors in construction accidents. Applied Ergonomics, 36(4), 401– 415.

Health and Safety Executive (HSE). (1988). Blackspot construc- tion: Study of five years fatal accidents in the building and civil engineering industries. London: HSE Books.

Hinze, J., Pedersen, C., & Fredley, J. (1998). Identifying root causes of construction injuries. Journal of Construction Engineering and Management, 124(1), 67–71.

International Journal of Injury Control and Safety Promotion 287

Huang, X., & Hinze, J. (2003). Analysis of construction worker fall accidents. Journal of Construction Engineering and Management, 129(3), 262–271.

International Electrotechnical Commission. (2002). Interna- tional standard – IEC 60038. Geneva: IEC.

Janicak, C.A. (2008). Occupational fatalities due to electrocu- tions in the construction industry. Journal of Safety Research, 39(6), 617–621.

Kleiner, B., Smith-Jackson, T., Mills, T. III, O’Brien, M., & Haro, E. (2008). Design, development, and deployment of a rapid universal safety and health system for construction. Journal of Construction Engineering and Management, 134(4), 273–279.

Kunadharaju, K., Smith, T.D., & DeJoy, D.M. (2011). Line- of-duty deaths among US firefighters: An analysis of fatality investigations. Accident Analysis & Prevention, 43(3), 1171–1180.

Lee, R.C., & Dougherty, W. (2003). Electrical injury: mecha- nisms, manifestations, and therapy. Dielectrics and Electri- cal Insulation, IEEE Transactions on, 10(5), 810–819.

Ling, F.Y.Y., Liu, M., & Woo, Y.C. (2009). Construction fatali- ties in Singapore. International Journal of Project Manage- ment, 27(7), 717–726.

Massachusetts, FACE. (2011). Immigrant roofer electrocuted when an aluminum ladder platform hoist contacted overhead power line – Massachusetts [online]. Retrieved from http:// www.cdc.gov/niosh/face/pdfs/10MA019.pdf

Mullins, J.N. (2005). Evaluating the effectiveness of a company’s online health and safety training program by assessing its employees perceptions and knowledge. 1428872 (M.S.E.H.). East Tennessee State University.

National Fire and Protection Association. (2008). National elec- trical code. Quincy, MA: NFPA.

National Institute for Occupational Safety and Health. (2010). Fatality assessment and control evaluation (FACE) program [online]. Retrieved from http://www.cdc.gov/niosh/face/

National Institute of Occupational Safety and Health. (2006). NIOSH fatal occupational injury cost fact sheet: Construc- tion [online]. NIOSH. Retrieved from http://www.cdc.gov/ niosh/docs/2006-153/pdfs/2006-153.pdf

Niskanen, T., & Saarsalmi, O. (1983). Accident analysis in the construction of buildings. Journal of Occupational Accidents, 5(2), 89–98.

NORA Construction Sector Council. (2008). National occupa- tional research agenda: National construction agenda. Atlanta, GA: The National Institute for Occupational Safety and Health.

Read, G.J.M., Lenn�e, M.G., & Moss, S.A. (2012). Associations between task, training and social environmental factors and error types involved in rail incidents and accidents. Accident Analysis & Prevention, 48(0), 416–422.

Reese, C.D., & Eidson, J.V. (2006). Handbook of OSHA construc- tion safety and health (2nd ed.). Boca Raton, FL: CRC Press.

Sawacha, E., Naoum, S., & Fong, D. (1999). Factors affecting safety performance on construction sites. International Journal of Project Management, 17(5), 309–315.

Strauch, B. (2002). Investigating human error: Incidents, accidents and complex systems. Burlington, VT: Ashgate.

Taylor, A., McGwin, G., Valent, F., & Rue, L. (2002). Fatal occupational electrocutions in the United States. Injury Prevention, 8, 306–312.

US Bureau of Labor Statistics. (2012). Employment, hours, and earnings from the current employment statistics survey [online]. Retrieved from http://data.bls.gov/cgi-bin/survey- most?ce

US Bureau of Labor Statistics. (2013). 2011 Census of fatal occupational injuries [online]. Washington DC: US Bureau of Labor Statistics. Retrieved from http://www.bls.gov/iif/ oshcfoi1.htm#2011.

US Census Bureau. (2012). North American Industry Classifica- tion System (NAICS) [online]. Retrieved from http://www. census.gov/eos/www/naics/index.html

Waehrer, G.M., Dong, X.S., Miller, T., Haile, E., & Men, Y. (2007). Costs of occupational injuries in construction in the United States. Accident Analysis & Prevention, 39(6), 1258– 1266.

Zhao, D., Lucas, J., & Thabet, W. (2009). Using virtual environ- ments to support electrical safety awareness in construction. In M. Rossetti, R. Hill, & B. Johansson (Eds.), Proceedings of the 2009 winter simulation conference (pp. 2679–2690). Austin, TX: IEEE.

Zhao, D., Thabet, W., McCoy, A., & Kleiner, B. (2012). Manag- ing electrocution hazards in the US construction industry using VR simulation and cloud technology. Proceedings of the European conference on product and process modelling (pp. 759–764). Reykjavik, Iceland: CRC Press.

288 D. Zhao et al.

Copyright of International Journal of Injury Control & Safety Promotion is the property of Taylor & Francis Ltd and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.