Injuries among electric power industry workers, 1995–2013

JSR-01352; No of Pages 8 Journal of Safety Research xxx (2016) xxx–xxx

Contents lists available at ScienceDirect

Journal of Safety Research journal homepage: www.elsevier.com/locate/jsr

Injuries among electric power industry workers, 1995–2013

2Q2

Vitaly Volberg, a Tiffani Fordyce, a,⁎ Megan Leonhard, b Gabor Mezei, c Ximena Vergara, d Lovely Krishen e

3 4 5 6 7Q3 8

a

9

a r t i c l e

10 11 12 13 14 19 18 17 16 15 40 Q4 41 42 43 44 45 46

Article history: Received 22 March 2016 Received in revised form 6 July 2016 Accepted 17 November 2016 Available online xxxx

F

1Q1

Exponent, 475 14th St #400, Oakland, CA 94612, United States Exponent, 15375 SE 30th Place, Suite 250, Bellevue, WA 98007, United States c Exponent, 149 Commonwealth Drive Menlo Park, CA 94025, United States d The Electric Power Research Institute (EPRI), 3420 Hillview Ave, Palo Alto, CA 94304, United States e EPRI, 942 Corridor Park Blvd, Knoxville, TN 37932, United States

R O

i n f o

O

b

a b s t r a c t

D

P

Introduction: Workers in the electric power industry face many risks of injury due to the high diversity of work tasks performed in potentially hazardous and unpredictable work environments. Method: We calculated injury rates by age, sex, occupational group, and injury type among workers in the Electric Power Research Institute’s (EPRI) Occupational Health and Safety Database (OHSD), which contains recordable injury, medical claims, and personnel data from 18 participating electric power companies from 1995 to 2013. Results: The OHSD includes a total of 63,193 injuries over 1,977,436 employee-years of follow-up, for an overall injury rate of 3.20 injuries per 100 employee-years. Annual injury rates steadily decreased from 1995 to 2000, increased sharply in 2001, and subsequently decreased to their lowest rate of 1.31 injuries per 100 employee-years in 2013. Occupations with the highest injury rates were welders (13.56 per 100 employee-years, 95% CI 12.74–14.37), meter readers (12.04 per 100 employee-years, 95% CI 11.77–12.31), and line workers (10.37 per 100 employee-years, 95% CI 10.19–10.56). Males had an overall higher injury rate compared to females (2.74 vs. 1.61 per 100 employee-years) although some occupations, such as meter reader, had higher injury rates for females. For all workers, injury rates were highest for those in the 21 to 30 age group (3.70 per 100 employeeyears) and decreased with age. Welders and machinists did not follow this trend and had higher injury rates in the 65+ age group. There were 63 fatalities over the 1995 to 2013 period, with 21 fatalities (33.3%) occurring among line workers. Conclusions: Although injury rates have decreased over time, certain high-risk groups remain (i.e., line workers, mechanics, young males, older welders and machinists, and female meter readers). Practical applications: Protective measures and targeted safety programs may be warranted to ensure their safety in the workplace. © 2016 Published by Elsevier Ltd.

R

R

E

C

T

E

Keywords: Injury surveillance Utility Electrical Occupational injury Non-fatal injury Fatal injury

49

N C O

50 48 47

1. Introduction

52 53

Workplace injuries and illnesses in the United States have declined over the past decade, but limited data on injury trends within the electric power industry are available. Although the U.S. Bureau of Labor Statistics (BLS) provides injury estimates for the utilities sectors, reporting an overall injury rate of 1.8 cases per 100 employee-years for 2013, this estimate is averaged over several diverse sub-industries including electric power generation, transmission and distribution, natural gas distribution, and water sewage systems, which are likely to have differing occupational hazards and associated risks (U.S. Bureau of Labor Statistics, 2013). Further, little is known about specific risk factors and vulnerable sub-populations that may have particularly high injury rates within the electric power industry.

56 57 58 59 60 61 62 63

U

51

54 55

20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39

⁎ Corresponding author. E-mail address: [email protected] (T. Fordyce).

The current analysis uses data gathered by the Electric Power Research Institute (EPRI) Occupational Health and Safety Database (OHSD) and is intended to update and expand upon an earlier publication characterizing injuries in the electric power industry (Kelsh et al., 2004). The OHSD program has been described previously (EPRI, 2012, 2015; Kelsh et al., 2004; Yager, Kelsh, Zhao, & Mrad, 2001). Briefly, the OHSD was created in 1999 to provide more detailed information about the occurrence of workplace injury among workers in the electric power industry (EPRI, 2001, 2004; Kelsh et al., 2004; Yager et al., 2001). Its main objectives are to: (a) monitor trends of injury and illness over time, across job characteristics, and worker demographics; (b) identify high-risk occupations and work environments; (c) quantify costs and lost time caused by work-related injuries and illnesses; (d) identify and prioritize injury/illness issues that merit focused research efforts; and (e) evaluate the effectiveness of prevention programs. Workers in the electric power industry face many potential risks of injury, including injuries from hazardous and unpredictable work environments, physically demanding maintenance and repair activities,

http://dx.doi.org/10.1016/j.jsr.2016.11.001 0022-4375/© 2016 Published by Elsevier Ltd.

Please cite this article as: Volberg, V., et al., Injuries among electric power industry workers, 1995–2013, Journal of Safety Research (2016), http:// dx.doi.org/10.1016/j.jsr.2016.11.001

64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81

110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145

C

108 109

E

106 107

R

104 105

R

102 103

O

100 101

C

95 96

N

93 94

U

91 92

3. Statistical analyses

158

F

Definitions, classification methodology, and data standardization methodology used in the OHSD have been previously described in detail (EPRI, 2012, 2015; Kelsh et al., 2004; Yager et al., 2001). In brief, the OHSD currently includes data from 18 companies, comprising a total of 1,977,436 employee-years of follow up and 63,193 reportable individual injuries. Participation in the EPRI OHSD program is voluntary. Both small and large companies are present in the database with the five largest companies comprising over 60% of all workers. Three categories of data, including personnel files, reportable injury files, and medical claim files were requested from each participating electric power company and compiled to generate the EPRI OHSD data set. Employee date of birth, sex, hire date, job code, job title, and work location or business unit were abstracted from company personnel files for each of the study years 1995 to 2013 and each employee was assigned a unique identifier. Occupation and work location were defined by the employee’s record status on January 1 of any particular year and entered into the database. Basic work history and demographic data for all company employees and not just injured employees were used to calculate injury rates. In addition to personnel data, injury event information (location, accident description, injury mechanism), data about the injury itself (body region, nature of injury), and claims information (work days lost, medical costs) were requested and incorporated into the database. Location refers to a worker’s primarily work location and may or may not represent where an injury took place. A standardized coding system for injury mechanism was developed using a combination of injury source codes (e.g., vehicle collision, fall, “struck by”) and data contained in accident descriptions. The mechanism of injury classification characterizes the event leading to the worker’s injury and usually represents the immediate or preceding cause based on temporality; however, the mechanism of injury may or may not represent the underlying or preventable cause. Data for nature of injury and body region injured were coded and classified into a standard common format based primarily on Bureau of Labor Statistics guidelines (EPRI, 2001). The OHSD contains 26 categories for nature of injury (e.g., sprains and strains, fractures and dislocations, heat and thermal burns) and 15 categories for body region injured (e.g., back and trunk, hand and finger). From over 35,000 unique reported job titles, we created 22 specific job categories using an occupational classification system previously developed for electric power industry workers (Kelsh, Kheifets, & Smith, 2000). Unclassifiable primary work location codes and missing nature of injury and injured body region information were updated based on a thorough review of the narrative accident description when the relevant information was provided. All reported lost time and “recordable” injury/illness claims have been included in the injury analyses. The Occupational Safety and Health Administration (OSHA) definition of a “lost time injury or illness”

89 90

148 149 150 151 152 153 154 155 156 157

Injury rates are expressed as the number of injuries and illnesses per 100 employees during a year of follow-up. The rate per 100 employeeyears is equivalent to that used for OSHA reporting purposes, which estimates rates per 200,000 work hours (OSHA 300 rate). Although injury rates estimate the relative occurrence and risk of injury, they do not directly reflect the severity of an injury. Time lost from work, measured by full time equivalents (FTEs), can be used as a proxy to examine injury severity. FTEs lost was defined as the total number of days lost divided by 240 workdays which assumes an average of four weeks off per year for workers (Kelsh et al., 2004). For recordable injuries where no lost time was reported, 0.002 FTEs lost, which is equivalent to one half day lost, was assigned to represent an approximate midpoint of the potential time away from work. Fatality rates are expressed per 100,000 employee-years. To date, six companies have provided data for the entire 19-year period. Six additional companies have provided data for the majority of the past 10 years. One company (Company N) provided only total employee data for the 1995 to 1999 period and did not report demographic or job description data. Thus, data for company N for this period are excluded from rate calculations, with the exception of overall OHSD injury rates. Given the deviance criteria (degrees of freedom ratio close to one) and the dispersion estimate criteria (over-dispersion parameter equal to zero), the calculation of confidence intervals assumes an underlying Poisson distribution. Upper and lower 95% confidence limits were estimated using the methods described by Fleiss (Fleiss, 1981). To investigate injury trends over time, a Poisson regression model was fit to the data, adjusting for the observation time per year. For trends in FTE loss rates over time, a negative binomial regression model fit the data best based on deviance and dispersion estimate criteria. To address the sex-specific differences in injury rates between occupations, we performed an age-adjusted Mantel–Haenszel analysis to estimate injuryrate ratios by occupation (Fleiss, 1981). For the three occupations with the highest injury rates, mechanisms of injury and body regions of injury were analyzed. Additionally, an analysis of injury by seasons was performed. We defined winter as December through February, spring as March through May, summer as June through August, and fall as September through November.

159 160

4. Results

197

The majority of electric power industry workers were male (73.4%), providing a total of 1,451,143 employee-years of observation (Table 1). Female workers accounted for 22.9% of the workforce and 452,260 employee-years. Sex was not reported for 3.7% of the study population. The majority of workers were between 41 and 60 years of age (58.9%), with 31.1% of the workforce 40 years or younger and only 5.4% 61 years or older. The most common injury type was sprains and strains, accounting for 40.9% of all injuries (Table 2). Sprains and strains were the primary contributor to reported medical costs at 43.7%. Although representing

198

O

99

88

R O

2. Methods

86 87

146 147

P

98

84 85

requires that a worker miss one full day of work (or shift) after the injury date. An OSHA recordable injury involves medical attention “beyond first aid” or loss of consciousness or results in days away from work, restricted work activity, or job transfer. Because some utilities could not provide reports on less severe, first-aid-only, or non-injury events, the EPRI OHSD database excludes such data. To ensure data confidentiality, the OHSD program policy restricts use of the data to peer-reviewed health and safety research proposals only and does not distribute personnel and individual records. In addition, all personal identifiers were removed from data records and the name of each participating company was replaced with generic identifiers.

T

97

working long shifts, working in emergency situations, and driving. The initial report using EPRI OHSD data was based on 528,133 employeeyears and 11,166 injuries over the 1995 to 2002 period and identified welders, meter readers, and line workers at highest risk of injury (Kelsh et al., 2004). Subsequent publications using OHSD data characterized risks, risk factors, and costs associated with thermal burns and neck injuries and factors distinguishing severity of sprain and strain injuries among electric utility workers (Fordyce, Kelsh, Lu, Sahl, & Yager, 2007; Fordyce, Morimoto, Coalson, Kelsh, & Mezei, 2010; Kelsh et al., 2009). The goals of the current analyses were to characterize injury and illness rates using the current OHSD data, which includes a total of 1,977,436 employee-years and 63,193 recordable injuries over the 1995 to 2013 time-period. We examined injury rates over time and by age, sex, and occupation, to determine risk factors for injury and identify vulnerable sub-populations with high injury rates.

D

82 83

V. Volberg et al. / Journal of Safety Research xxx (2016) xxx–xxx

E

2

Please cite this article as: Volberg, V., et al., Injuries among electric power industry workers, 1995–2013, Journal of Safety Research (2016), http:// dx.doi.org/10.1016/j.jsr.2016.11.001

161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196

199 200 201 202 203 204 205 206 207

V. Volberg et al. / Journal of Safety Research xxx (2016) xxx–xxx

t1:3

t2:1 t2:2

Percentage of OHSD

425,260 1,451,143 74,033

22.9 73.4 3.7

Age group (years) b20 21–30 31–40 41–50 51–60 61–65 65+ Unknown

14,944 202,767 396,159 621,654 544,667 75, 898 31,571 89,776

0.8 10.3 20.0 31.4 27.5 3.8 1.6 4.5

Table 2 Distribution of injuries and medical costs, EPRI OHSD 1995–2013. Injury type

Percentage of Injuries

Percent of Medical Costs

t2:4 t2:5 t2:6 t2:7 t2:8 t2:9 t2:10 t2:11 t2:12 t2:13 t2:14 t2:15 t2:16

Sprains, strains Cut, laceration, puncture Contusion, bruise Scratches, abrasions Fracture/dislocation CTD/RSI⁎

40.9% 16.0% 9.0% 5.7% 5.5% 4.7% 3.0% 3.0% 2.2% 1.4% 1.3% 0.7% 0.6%

43.7% 4.2% 4.2% 0.7% 12.3% 10.9% 0.9% 0.2% 0.4% 0.2% 2.9% 8.4% 2.4%

t2:17

⁎ CTD/RSI Carpal Tunnel Disorder/Repetitive Stress Injury.

215 216 217

222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239

Occupations with the highest injury rates were welders (13.56 per 100 employee-years, 95% CI 12.74–14.37), meter readers (12.04 per 100 employee-years, 95% CI 11.77–12.31), and line workers (10.37 per 100 employee-years, 95% CI 10.19–10.56) (Fig. 3). Line workers (19.5%), mechanics (12.8%), and meter readers (12.3%) accounted for the highest proportions of injuries among all of the occupational groups and were among the highest FTEs lost (61.20, 25.42, and 57.08 per 10,000 employee-years, respectively). Although welders made up a relatively small proportion of the workforce (b 1% of total employeeyears), of the injuries that had occurred (1.7%), they had the highest observed injury rate and the fifth highest FTE loss rate (25.42 per 10,000 employee-years). Occupations with the lowest injury rates were engineers (0.65 per 100 employee-years, 95% CI 0.60–0.69) and managers (0.42 per 100 employee-years, 95% CI 0.38–0.45).

240 241

7. Injury rates and FTEs lost by sex

254

D

6. Injury rates and FTEs lost by job classification

E T C

E

R

213 214

R

211 212

only 2.0% of injuries, burns had the highest cost per incident, accounting for 11.3% of total medical costs. Commonly affected body regions were back and trunk (17.8%); hand and finger (14.3%); head (excluding eyes, 9.9%); upper extremities, including arm, forearm and elbow (8.3%); neck and shoulder (8.2%); and knees (8.0%) (Fig. 1). There was a statistically significant increase in the proportion of injuries to the head, excluding eyes, with increasing age (p b0.01). The unadjusted distributions of injuries were similar between males and females, with a few notable exceptions. Injuries to the wrist and upper extremities were more frequent among female

N C O

209 210

The overall injury rate over the 1995 to 2013 period was 3.20 per 100 employee-years (95% CI 3.17–3.22) (Fig. 2). Annual injury rates steadily decreased from 1995 to 2000, increased sharply in 2001, and subsequently steadily decreased to their lowest rate of 1.31 injuries per 100 employee-years in 2013. For 2013, the current data-reporting year, injury rates (1.31 per 100 employee-years, 95% CI 1.22–1.40) were significantly lower than the peak injury rates from 1995 (4.70 per 100 employee-years, 95% CI 4.57–4.82). They were also significantly lower compared to the 2012 injury rate of 2.06 per 100 employee-years (95% CI 1.97–2.16). For the entire 19-year study period, the annual injury rate declined by an average 5% per year (p b 0.01). Annual total FTEs lost have shown substantial variability and no consistent trend over the 19-year reporting period (p = 0.11). However, there was a steady decline from the peak in 2003 of 28.19 FTEs per 10,000 employee-years to the current, 2013, rate of 6.83 FTEs lost per 10,000 employee-years (p b 0.001); the lowest in the history of the OHSD (data not shown).

242 243 244 245 246 247 248 249 250 251 252 253

Over the 1995 to 2013 period, males had higher injury rates (2.74 255 per 100 employee-years, 95% CI 2.71–2.76 vs. 1.61 per 100 employee- 256

U

208

221

P

t2:3

Hearing loss or impairment Bite Respiratory Dermatitis/skin Burn heat/thermal Burn, flashburn Electric shock, electrocution

5. Overall injury rates and FTEs lost

F

Employee-years Sex Female Male Unknown

O

t1:4 t1:5 t1:6 t1:7 t1:8 t1:9 t1:10 t1:11 t1:12 t1:13 t1:14 t1:15 t1:16 t1:17

workers (14.1% and 12.2% vs. 3.3% and 8.1%, respectively), while injuries 218 to the back and trunk (17.5% vs. 12.6%), head (11.1% vs. 4.6%), and eyes 219 (5.0% vs. 1.8%) were more frequent among male workers. 220

Table 1 Distribution of sex and age, EPRI OHSD 1995–2013.

R O

t1:1 t1:2

3

Fig. 1. Distribution of injuries by injured body region, EPRI OHSD 1995–2013.

Please cite this article as: Volberg, V., et al., Injuries among electric power industry workers, 1995–2013, Journal of Safety Research (2016), http:// dx.doi.org/10.1016/j.jsr.2016.11.001

Fig. 2. Injury rate per 100 employee-years by year, EPRI OHSD 1995–2013.

For all workers, injury rates were highest among those aged 21 to 30 years, at 3.70 per 100 employee-years (95% CI 3.62–3.79) (Fig. 5). Workers in the 41 to 50 and 51 to 60 age groups made up the majority of the worker population (61.8%) and had injury rates of 3.19 per

C

E

R

R

278 279 280 281 282 283 284 285 286 287 288 289 290 291 292

Welders, meter readers, and line workers had the highest injury 294 rates of all occupations in the OHSD. For meter readers and line workers, 295 over half of all injuries were classified as sprains and strains or cuts, 296

O

276

293

C

274 275

9. Additional analysis of welders, meter readers, and line workers

N

268 269

U

266 267

R O

273

264 265

P

8. Injury rates and FTEs lost by age

262 263

277

D

272

260 261

100 years (95% CI 3.14–3.23) and 2.54 per 100 employee-years (95% CI 2.50–2.58), respectively. Injuries in these age groups accounted for the most total FTEs lost, 872.3 and 896.9, respectively. Injury rates for trade occupations tended to decrease with age and were lowest in those aged 65 or older (0.94 per 100 employee-years, 95% CI 0.84– 1.05). Welders did not follow this trend and had higher injury rates among the youngest population and oldest population (71.43 per 100 employee years and 50.00 per 100 employee years, respectively). However, welders less than 20 and welders older than 65 combined represented less than 1% of the total employee-years for that occupation. The majority of injuries to welders over 65 were due to falls on the same level (66.7%) with hands/fingers being most commonly injured. The majority of injuries to welders under 20 were indicated as struck by (60.0%), with injuries to the head and eyes. Injury rates across most other occupational age groups, including office-based staff, were relatively constant across workers aged 31–60 years.

T

270 271

years, 95% CI 1.57–1.65) and more FTEs lost (13.86 per 10,000 employee-years, 95% CI 13.25–14.46 vs. 10.63 per 10,000 employeeyears, 95% CI 9.67–11.57) compared to females. Several occupations had higher rates of injury among females compared to males; these occupational groups included line workers (11.36 per 100 employeeyears, 95% CI 9.14–13.48 vs. 8.66 per 100 employee-years, 95% CI 8.49–8.83), meter readers (14.10 per 100 employee-years, 95% CI 13.25–14.93 vs. 9.12 per 100 employee-years, 95% CI 8.87–9.38), and plant and equipment operators (3.31 per 100 employee-years, 95% CI 2.90–3.72 vs. 2.48 per 100 employee-years. 95% CI 2.40–2.57). An age-adjusted Mantel–Haenzel analysis by occupation indicated that females have higher injury rates than males for three non-office related occupations: meter readers, security, and plant and equipment operators (Fig. 4). Custodians and cooks had slightly higher rates, which were not statistically significant.

258 259

E

257

F

V. Volberg et al. / Journal of Safety Research xxx (2016) xxx–xxx

O

4

Fig. 3. Injury rate per 100 employee-years by job classification, EPRI OHSD 1995–2013.

Please cite this article as: Volberg, V., et al., Injuries among electric power industry workers, 1995–2013, Journal of Safety Research (2016), http:// dx.doi.org/10.1016/j.jsr.2016.11.001

5

R O

O

F

V. Volberg et al. / Journal of Safety Research xxx (2016) xxx–xxx

309 310

D

E

T

C

307 308

E

305 306

R

303 304

R

301 302

most prevalent body regions injured for overexertion, body motion injuries were back/trunk (36.3%), neck/shoulder (14.6%), and knees (10.2%). For struck by injuries, head excluding eyes (25.7%), hand/finger (15.2%), eyes (10.9%), and feet/toes (10.0%) were the most common body regions affected. For a fall on the same level the knees (21.9%), ankle (17.9%), or back/truck (17.0%) were most common. Amongst meter readers, the top three mechanisms of injury, animal or insect bite (30.1%), overexertion, body motion (23.4%), and fall on the same level (19.0%), accounted for over 70% of all causes of injury. Animal or insect bite injuries were most common to other lower extremities (34.2%) and hand/finger (20.4%). Overexertion, body motion injuries were most frequently to the back/trunk (25.2%), feet/toe (15.7%), and knees (14.7%). For falls on same level ankles (24.4%), knees (17.6%), back/trunk (15.9%) were the most common body regions injured.

N C O

299 300

lacerations, or punctures. For welders, over half of all injuries were classified as sprains and strains; scratches, abrasions; or cuts, lacerations or punctures. Amongst welders, the three most common mechanisms of injury, struck by (30.9%); overexertion, body motion (18.4%); and contact with temperature extremes (7.1%), accounted for over half of all injuries. The majority of struck by injuries had body region listed as eyes (60.3%), followed by hand/finger (10.1%). For overexertion, body motion, back/trunk (46.3%), hand/finger (10.7%), and neck/shoulder (10.7%) were the most common body regions injured. Upper extremities including the arm, forearm, and elbow comprised a quarter (25.0%) of all contact with temperature extreme injuries and hands/ fingers represented 19.1%. The top three mechanisms of injury for line workers, overexertion, body motion (39.4%), struck by (12.2%), and fall on the same level (12.1%), account for over 60% of all injury. The

U

297 298

P

Fig. 4. Female to male injury rate ratios controlled for age and 95% confidence intervals by job classification, EPRI OHSD 1995–2013.

Fig. 5. Distribution and injury rate per 100 employee-years by age group, EPRI OHSD 1995–2013.

Please cite this article as: Volberg, V., et al., Injuries among electric power industry workers, 1995–2013, Journal of Safety Research (2016), http:// dx.doi.org/10.1016/j.jsr.2016.11.001

311 312 313 314 315 316 317 318 319 320 321 322 323 324

V. Volberg et al. / Journal of Safety Research xxx (2016) xxx–xxx

343

352 353

The fatality rate was 3.18 per 100,000 employee-years for the entire 19-year study period (95% CI 2.37–3.95). Of the 63 total fatalities, line workers (21 deaths, 33.3%), maintenance workers (8 deaths, 12.7%), and mechanics (4 deaths, 6.3%) accounted for the highest proportion of deaths. The highest fatality rate was for line workers (18.03 per 100,000 employee-years, 95% CI 11.16–27.56), which was significantly higher compared to most other occupational groups. Most of the fatalities involved male workers (n = 52, 82.5%), two involved female workers (3.2%), and a total of nine had an unknown sex. The most common sources of fatality were motor-vehicle collisions (n = 16, 25.4%) and contact with an electric current (n = 16, 25.4%).

354

12. Discussion

355 356

Data from OHSD indicate that injury rates among electric power industry workers have tended to decrease over the 1995 to 2013 period. Although there was an increase in total injury rate in 2001, this was likely due at least in part to a change in OSHA reporting requirements that was announced in January 2001 (Occupational Safety and Health Administration, 2001). Injury rates for the current reporting period (2013) are similarly low when compared to the 2013 BLS injury rate for the entire utilities sector (U.S. Bureau of Labor Statistics, 2013). Although injury rates have decreased over time, certain high-risk groups remain. Injury rates varied more than 30-fold across occupational groups examined, with the highest risk of injury among workers in the craft/trade occupations and the lowest injury risk among office-based staff. Line workers and meter readers tended to have the highest FTE loss rates, potentially reflecting the severity of the injuries occurring among these workers, or the mobile nature of the work. Other highrisk groups included male workers overall, younger workers (aged 21– 30 years), and older welders. In comparing these results to an earlier analysis of OHSD data spanning the 1995 to 2002 period, many of these trends and high risk populations have persisted (Kelsh et al., 2004). Male workers, meter readers, welders, and line workers have continued to have higher risk of injury, although the rank order between meter readers and welders has switched. Sprains and strains continue to be a common injury type amongst these occupations, indicating the continued need for development and implementation of prevention programs accounting for the majority of medical costs. For these occupations, the back and truck was the primary body region associated with overexertion, body motion injuries. An analysis of occupation risk factors and back injury determined that weight lifted per hour, trunk twists per hour, weight lifted per day, frequency of lift, trunk motions per hour, and trunk flexions per hour were significantly associated with occurrence of back

350 351

357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385

C

348 349

E

346 347

R

344 345

R

339 340

O

337 338

C

335 336

N

333 334

U

331 332

F

11. Fatalities

330

O

342

328 329

R O

341

Of all injuries in the OHSD, 88.0% provided the data necessary to determine season of injury. There was little change in injury by season with only a slightly higher proportion of injury in summer (27.2%) and a slight reduction in injury seen in the winter months (22.8%). This pattern did not differ between sexes. Workers under 60 years of age had the lowest proportion of injury in the winter months (22.5%), while workers 61 and older had the lowest proportion of injury in the fall (19.0%). Summer had the highest frequency of injury for those 50 years and younger (27.7%). For workers ages 51 to 64, spring had the highest proportion of injury (27.1%). Workers 65 and older experienced the highest proportion of injury in winter (39.6%). Of these winter injuries, 37.0% were indicated as due to falls on the same level. Of the three highest risk occupations, meter readers and line workers followed the same pattern as the overall cohort with the highest proportion of injury in summer and lowest in winter. Welders, on the other hand, experienced the highest proportion of injury in the spring (28.7%).

P

326 327

injury among laborers performing manual material handling tasks (Craig et al., 2013). Although narrative descriptions of injury are inconsistently present and/or complete in the EPRI OHSD, future analyses of this field may provide further information on tasks being performed when injured or equipment being used. Welders were most likely to have a “stuck by” mechanism of injury with the eyes as the body region. Lombardi reported that almost 72% of struck by injuries to the eyes among welders were due to airborne objects (Lombardi et al., 2005). OSHA reports that helmet protection alone is not sufficient to prevent eye injuries to welders and that proper eye glasses are a warranted prevention measure suggesting development and implementation of an eye protection program (Braun, 2007). Over a third of all stuck by injuries among line workers were to the head and eyes, indicating that similar head and facial protective wear policies could reduce the occurrence of injuries. In the present analysis, meter readers, line workers and mechanics had high proportions of FTEs lost. A recent study of this population assessing injury severity reported that meter readers had the highest severe injury rate (2.26 per 100 employee-years), followed by line workers (1.99 per 100 employee-years) and mechanics (1.17 per 100 employee-years) (Fordyce et al., 2016). Welders in the youngest and oldest age groups had elevated injury rates and may represent vulnerable sub-populations. Welders 20 and under were most likely to be struck by an object to the head or face, indicating that increased emphasis on helmet and safety glasses use may benefit this population. On the other hand, welders over 65 were primarily injured by falls on the same level, an injury mechanism which represents a small proportion of total injury among all welders. Thus, prevention strategies may need to be specifically tailored for this population. Decreased mobility among older welders may contribute to the increased risk of falls and further exploration of factors leading to fall injury in this population is warranted. However, prevention strategies targeted primarily to welders in these age groups are not likely to lead to considerable reductions in injuries amongst all welders since these age groups represent less than 1% of all welder employee-years. Data on fatalities in the utility sector remain sparse. The fatality rate of 3.18 per 100,000 employee-years estimated in the present analysis is substantially smaller than that of results from a study using death certificate data, which reported a fatality rate of 13.2 per 100,000 employeeyears (Loomis, Dufort, Kleckner, & Savitz, 1999). Follow-up in that cohort covered 1950 to 1986, however, and likely reflects higher mortality risk compared to the 1995 to 2013 period of EPRI OHSD monitoring. As in the previous analysis using 1995 to 2002 OHSD data, line workers had the highest risk of death compared to all other occupations, accounting for 15 of the 53 deaths after 2002, indicating that increased safety measures may be justified. When stratified by sex, injury rates in certain trade occupations (line workers, meter readers, and plant and equipment operators) were higher for females compared to males. Among non-office type occupations, after controlling for age, only the injury risk ratio comparing female to male meter readers remained significant. Reasons for this sex-dependent difference in injuries among meter readers are unclear; ergonomic issues and/or training (formal or informal) may be contributing factors and further research is required. Injuries to the wrist and upper extremities were more prevalent among females than males. This is partially explained by the association between CTD/RSI and office-related job types which are more common amongst women. In comparison to the analysis of OHSD data from 1995 to 2002, some of the present female to male injury risk ratios have undergone drastic shifts (Kelsh et al., 2004). For example, representatives shifted from an odds ratio less than one, to an odds ratio of 1.4. These changes may be due to the shifting gender composition of certain job titles over time, indicating that periodic updating Analysis of injuries by season indicated that, overall, there is little variation in the proportion of injury due to seasonal factors. Reductions in the proportion of injury during winter may be due to fewer days

D

10. Seasonal analysis

T

325

E

6

Please cite this article as: Volberg, V., et al., Injuries among electric power industry workers, 1995–2013, Journal of Safety Research (2016), http:// dx.doi.org/10.1016/j.jsr.2016.11.001

386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451

V. Volberg et al. / Journal of Safety Research xxx (2016) xxx–xxx

473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517

O

F

534

R O

471 472

Acknowledgement

520 521 522 523 524 525 526 527 528 529 530 531 532 533

We thank all the participating utility companies and their health and 535 safety staff who assisted in assembling the data for the EPRI OHSD pro- 536 gram. 537 538

Funding source

539 540

This research was funded by the Electric Power Research Institute (EPRI), an independent private and nonprofit center for energy and environmental research, which is supported principally by electric utility companies. Although electric utility companies contribute to funding for EPRI, EPRI independently determines avenues for research funding. Data voluntarily provided by participating electric power companies to EPRI was used for this analysis. Primary development of study design, analysis completion, interpretation of data, and manuscript writing were undertaken by Exponent employees.

541

References

550

Braun, T. (2007). Occupational health & safety. Preventing eye injuries when welding. [cited 2016 July 1]; Available from: https://ohsonline.com/Articles/2007/02/ Preventing-Eye-Injuries-When-Welding.aspx Brooks, P. (2013). The seasonal timing of work-related injuries. Joint statistical meetings (Montreal, Quebec, Canada). Capelli-Schellpfeffer, M., Floyd, H. L., Eastwood, K., & Liggett, D. P. (2000). How we can better learn from electrical accidents. IEEE Industry Applications Magazine, 5, 16–23. Chau, N., Wild, P., Dehaene, D., Benamghar, L., Mur, J. M., & Touron, C. (2010). Roles of age, length of service and job in work-related injury: a prospective study of 446 120 person-years in railway workers. Occupational and Environmental Medicine, 67(3), 147–153. Craig, B. N., Congleton, J. J., Beier, E., Kerk, C. J., Amendola, A. A., & Gaines, W. G. (2013). Occupational risk factors and back injury. International Journal of Occupational Safety and Ergonomics, 19(3), 335–345. EPRI (2001). Injury and illness among the electric energy workforce, 1995–2000. Palo Alto, CA: Electric Power Research Institute. EPRI (2004). Electric energy industry workforce—Occupational health and safety trends 2004. Palo Alto, CA: Electric Power Research Institute. EPRI (2012). Occupational health and safety annual report, 2012: Occupational health and safety trends among electric energy workers, 1995–2011. Palo Alto, CA: Electric Power Research Institute. EPRI (2015). EPRI Occupational Health and Safety Annual Report, 2014 Injury and Illness Among the Electric Energy Workforce, 1995–2013. Palo Alto, CA: Electric Power Research Institute. Farrow, A., & Reynolds, F. (2012). Health and safety of the older worker. Occupational Medicine (London), 62(1), 4–11. Fleiss, J. L. (1981). Statistical methods for rates and proportions. New York: John Wiley & Sons. Floyd, H. L., II, Andrews, J. J., Capelli-Schellpfeffer, M., Neal, T. E., Liggett, D. P., & Saunders, L. F. (2004). Safeguarding the electric workplace. IEEE Industry Applications Magazine, 10, 18–24. Fordyce, T. A., Kelsh, M., Lu, E. T., Sahl, J. D., & Yager, J. W. (2007). Thermal burn and electrical injuries among electric utility workers, 1995–2004. Burns, 33(2), 209–220. Fordyce, T. A., Leonhard, M. J., Watson, H. N., Mezei, G., Vergara, X. P., & Krishen, L. (2016). An analysis of fatal and non-fatal injuries and injury severity factors among electric power industry workers. American Journal of Industrial Medicine. Fordyce, T. A., Morimoto, L., Coalson, J., Kelsh, M. A., & Mezei, G. (2010). Neck injuries among electric utility workers, 1995–2007. Journal of Occupational and Environmental Medicine, 52(4), 441–449.

551 552 553 554 Q5 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588

P

469 470

518 519

D

467 468

The EPRI OHSD has several important strengths and advantages over larger nationwide and statewide injury reporting systems (Sorock, Ranney, & Lehto, 1996). The OHSD contains basic work history and demographic data for all employees and not just for those injured, allowing for more precise calculation of injury risks and rates. The detailed information and circumstances for each injury occurrence can be used to better understand relationships between occupation, injury type, injury source, and other factors. Overall, the OHSD data can be used to identify risk factors associated with injury and help characterize high injury-risk sub-populations, providing employers with information to prioritize health and safety efforts and identify areas where further research or interventions are required. Our results reinforce some of the associations seen in other industries between age, sex, and risk of injury (U.S. Bureau of Labor Statistics, 2014c) and contribute to understanding the etiology of occupational injuries among workers in the electric power industry.

E

465 466

T

463 464

C

461 462

E

459 460

R

458

R

456 457

N C O

454 455

worked due to holidays. However, in an analysis of occupational injury rates by seasons, the Bureau of Labor and Statistics reports that lower end of year injury rates may be due to difficulty in identifying the injury, or difficulty identify the injury as work-related, or inability to identify in time for reporting (Brooks, 2013). In our analysis of seasonality by age group, workers over 65 experienced the highest proportion of injury during winter months, the majority of which were due to falls. Raising awareness of increased fall risk, primarily amongst older workers, during winter months and determining contributing factors leading to these falls may reduce this increased risk of injury during winter among older workers. Several studies of aging construction worker populations showed that older workers typically experience fewer injuries and accidents, although when they do occur, injuries are more severe (Farrow & Reynolds, 2012; Jones, Latreille, Sloane, & Staneva, 2013; Schwatka, Butler, & Rosecrance, 2012). Similar to the present findings, Lipscomb et al. (2003) reported that carpenters 45 and older were more likely to experience falls on the same level compared to workers 30 years and younger (Lipscomb, Li, & Dement, 2003). A study of male railway workers indicated that older employees ages 50 to 55 had a higher risk of fall, injury resulting from handling equipment, and injuries from a collision with moving objects (Chau et al., 2010). Given that the relative proportion of older workers in the labor force has been increasing in the United States, it is important to further characterize risk factors for injury among older workers, especially in this under studied electric power worker population, to guide creation of targeted safety measures (Rogers & Wiatrowski, 2005). Injury rates over all EPRI OHSD companies have declined over the 19-year study period. This corresponds with trends observed by the BLS in nationally collected data on utilities as well as other industries such as manufacturing and retail trades (U.S. Bureau of Labor Statistics, 2014a). The observed decline over time may be due to various aspects of company health and safety programs, potentially including improved safety awareness, increased attention to health and safety by management, and improved safety program implementation. The decline may be, in part, a result of the increasing use of contractors (historically employed for high injury-risk work) whose injury and illness records are not maintained by electric power companies and are thus not included in the EPRI OHSD database. A limitation of this study is the use of voluntarily reported industry data. The quality, representativeness, and accuracy of these data are dependent on several factors. The OHSD data depends on the quality of information captured and provided by the participating companies to the EPRI research program. Given variations in coding of injury reports and medical claims across different U.S. states and electric power companies, database development and standardization of variables has been dynamic and classification error may have occurred. In addition, there are recognized reporting biases that cause underreporting of workplace injuries including worker perceptions that an injury is not serious enough to report and/or that their job security will be affected, and an employer perception that the injury was not work-related (Capelli-Schellpfeffer, Floyd, Eastwood, & Liggett, 2000; Floyd et al., 2004). Due to heavy regulation and the unionized nature of electric power companies, however, these types of reporting bias may be less likely to occur (Kelsh et al., 2004; U.S. Bureau of Labor Statistics, 2014b). Missing data in fields such as age or occupation could introduce bias, particularly in our analysis of age adjusted female to male injury rate ratios by occupation. Unknown self-selection factors may affect our study since only 18 of over 200 investor-owned electric utility companies in the United States provide data to the OHSD. Furthermore, events that do not result in an injury to a worker, such as damage to equipment or near-misses, are not reported in the OHSD due to incomplete and less systematic reporting of these events. Despite these limitations, the EPRI OHSD provides a unique resource to examine occupational injury/illness and characterization of severity of injury/ illness unique to the electric utility industry.

U

452 453

7

Please cite this article as: Volberg, V., et al., Injuries among electric power industry workers, 1995–2013, Journal of Safety Research (2016), http:// dx.doi.org/10.1016/j.jsr.2016.11.001

542 543 544 545 546 547 548 549

Dr. Vitaly Volberg is a Senior Scientist in Exponent's Health Sciences Center for Epidemiology, Biostatistics, and Computational Biology. He has 8 years of experience with epidemiological study design, data collection and analysis, and grant and manuscript preparation. His prior work has focused on risk factors for childhood obesity, with emphasis on environmental exposures. Dr. Tiffani Fordyce is a Managing Scientist in Exponent's Health Sciences Center for Epidemiology, Biostatistics, and Computational Biology. She has 15 years of experience with research protocols and procedures, epidemiological study design and execution, statistical analyses, database development, data collection and management. Dr. Fordyce has conducted many occupational health and safety studies, including multiple large cohort mortality studies and case-control studies.

F

Ms. Megan Leonhard is a Scientist in Exponent's Health Sciences Center for Epidemiology and Computational Biology. Ms. Leonhard has expertise in epidemiological study design, data analysis and management, data analysis program use, and surveillance system use. She also has extensive experience in literature review particularly related to injury, trauma, and rare disease.

R O

O

Dr. Gabor Mezei has over 25 years of experience in health research including epidemiological studies of both clinical outcomes and environmental and occupational health issues. Previously, at the Electric Power Research Institute, he was responsible for leading a multidisciplinary scientific research program aimed at addressing potential human health effects associated with residential and occupational exposure to power frequency and radiofrequency EMF.

D

P

Ximena Vergara leads the EPRI Electric and Magnetic Fields and Radiofrequency Fields Health Assessment and Safety Program. Dr. Vergara received a Bachelor of Arts in chemistry from the University of Chicago and a Master of Public Health in environmental health science (industrial hygiene) from the University of California, Los Angeles (UCLA), School of Public Health. In 2012, she completed her Ph.D. in epidemiology at UCLA. Dr. Vergara is a member of the Society for Epidemiologic Research, American Industrial Hygiene Association and American Public Health Association. Her research focuses on electromagnetic and radiofrequency field exposure assessment and epidemiology, occupational hygiene and injury surveillance. Dr. Lovely Krishen is a senior technical leader and program manager of health and safety at the Electric Power Research Institute. She is an experienced R&D manager with a background in technology management, research and development, and training for industry, government, and academia. Her previous research experience has primarily focused on: Systems-level Safety, Reliability, and Quality Assurance, Human Health, Environments and Safety Risk Management for Human Space flight programs at NASA.

N

C

O

R

R

E

C

T

Jones, M. K., Latreille, P. L., Sloane, P. J., & Staneva, A. V. (2013). Work-related health risks in Europe: Are older workers more vulnerable? Social Science & Medicine, 88, 18–29. Kelsh, M. A., Fordyce, T. A., Lau, E. C., Mink, P. J., Morimoto, L. M., Lu, E. T., & Yager, J. W. (2009). Factors that distinguish serious versus less severe strain and sprain injuries: an analysis of electric utility workers. American Journal of Industrial Medicine, 52(3), 210–220. Kelsh, M. A., Kheifets, L., & Smith, R. (2000). The impact of work environment, utility, and sampling design on occupational magnetic field exposure summaries. AIHAJ, 61(2), 174–182. Kelsh, M. A., Lu, E. T., Ramachandran, K., Jesser, C., Fordyce, T. A., & Yager, J. W. (2004). Occupational injury surveillance among electric utility employees. Journal of Occupational and Environmental Medicine, 46(9), 974–984. Lipscomb, H. J., Li, L., & Dement, J. M. (2003). Falls among union carpenters. American Journal of Industrial Medicine, 44(2), 148–156. Lombardi, D. A., Pannala, R., Sorock, G. S., Wellman, H., Courtney, T. K., Verma, S., & Smith, G. S. (2005). Welding related occupational eye injuries: A narrative analysis. Injury Prevention, 11, 174–179. Loomis, D., Dufort, V., Kleckner, R. C., & Savitz, D. A. (1999). Fatal occupational injuries among electric power company workers. American Journal of Industrial Medicine, 35(3), 302–309. Occupational Safety and Health Administration (2001). OSHA Revises Recordkeeping Regulations. [cited 2016 July 1]; Available from: https://www.osha.gov/pls/ oshaweb/owadisp.show_document?p_table=NEWS_RELEASES&p_id=200 Rogers, E., & Wiatrowski, W. (2005, October). Injuries, illnesses, and fatalities among older workers. Bureau of Labor and Statistics, Occupational Safety and Health, Monthly Labor Review, 24–30. Schwatka, N. V., Butler, L. M., & Rosecrance, J. R. (2012). An aging workforce and injury in the construction industry. Epidemiologic Reviews, 34, 156–167. Sorock, G. S., Ranney, T. A., & Lehto, M. R. (1996). Motor vehicle crashes in roadway construction workzones: An analysis using narrative text from insurance claims. Accident Analysis and Prevention, 28(1), 131–138. U.S. Bureau of Labor Statistics (2013). Incidence rate and number of nonfatal occupational injuries by industry and ownership. [cited 2016 July 1]; Available from: http://www. bls.gov/iif/oshwc/osh/os/ostb3966.pdf U.S. Bureau of Labor Statistics (2014a). Employer-reported workplace injury and illness summary. [cited 2016 July 1]; Available from: http://www.bls.gov/news.release/ osh.nr0.htm U.S. Bureau of Labor Statistics (2014b). U.S. Bureau of Labor Statistics. Union Members – 2014. [cited 2016 July 1]; Available from: http://www.bls.gov/news.release/pdf/ union2.pdf U.S. Bureau of Labor Statistics (2014c). Occupational employment statistics — Power generation, transmission and distribution. [cited 2016 July 1]; Available from: http://www.bls.gov/oes/current/naics4_221100.htm Yager, J. W., Kelsh, M. A., Zhao, K., & Mrad, R. (2001). Development of an occupational illness and injury surveillance database for the electric energy sector. Applied Occupational and Environmental Hygiene, 16(2), 291–294.

U

589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634

V. Volberg et al. / Journal of Safety Research xxx (2016) xxx–xxx

E

8

Please cite this article as: Volberg, V., et al., Injuries among electric power industry workers, 1995–2013, Journal of Safety Research (2016), http:// dx.doi.org/10.1016/j.jsr.2016.11.001

635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677