VIEWS: 3 PAGES: 6 POSTED ON: 5/7/2010
A job to die for A job to die for Katherine Marshall I n 1992, Canadians were stunned when 26 miners were buried alive in the Westray Mine in Nova Data sources and limitations These missing counts are not likely to affect the results of this paper since the Fatality data are supplied to Human Scotia. Accidental death related to Resources Development Canada by analysis centres on fatality rates, and work happens to relatively few Ca- provincial Workers’ Compensation adjustments were made to compensate nadians, but commands attention Boards and Commissions (WC); data on for the excluded data; that is, when because it happens unexpectedly to the number of paid workers (used in the 1992 and 1993 national and regional calculation of rates) originate from Sta- fatality rates were calculated, paid otherwise healthy people. In addi- tistics Canada’s Labour Force Survey worker counts for Newfoundland and tion to accidents, death can result New Brunswick were excluded from the from illness caused by exposure to (LFS). (Statistics Canada received detailed fatal work injury data from WC denominator to match the missing environmental hazards in the counts in the numerator. (Similar exclu- for 1993 and 1994.) Fatalities are workplace. reported for workers aged 14 and over, sions were made in 1993 to adjust for while paid worker data from the LFS missing Quebec data.) Work-related death also stands out because of its large financial cover only those persons aged 15 and The recording of fatalities is not uni- over. This discrepancy is negligible as form across Canada: Quebec reports by and personal cost. Apart from its work-related fatalities among 14 year- year of compensation; Ontario by year effect on family and friends, a olds are extremely rare. death at work can reduce produc- of death; and all other provinces by year The fatal work injury counts are a of accident. Fatalities caused by expo- tivity because of temporary shut- census of all reported* WC cases and are sure are the most likely to be affected downs, loss of morale and the by these variations, because some expo- training of replacement workers. therefore not affected by sampling vari- ability errors. However, the data con- sure deaths are gradual and will there- This article traces job-related cern only workers covered by WC, fore have different accident and death mainly paid workers. Since 1995, the dates. For example, if a person died of deaths over three 6-year periods Workers’ Compensation Acts in Canada asbestos poisoning in 1992 but compen- (1976 to 1981, 1982 to 1987 and sation was awarded in 1993, Ontario 1988 to 1993), and examines dif- provide compulsory coverage for most workers in most industries, ranging would report the fatality in 1992, Que- ferences in fatality counts and rates bec in 1993 and the remaining provinces from 70% to 100% depending on the by industry, region, occupation, jurisdiction (AWCBC, 1995). Persons in the year (estimated) the poisoning age, sex, and cause of death. It con- most likely not covered by WC are the began. However, the estimated number cludes with data on the financial self-employed, unpaid family workers of exposure cases with dates that fall cost associated with these deaths and workers in professional offices, outside the study periods are suffi- (see Data sources and limitations although all may, and some do, apply ciently small as not to affect national and Definitions). for coverage. Therefore, WC fatality results. counts are in fact an under-representa- Finally, the data were aggregated Many deaths occur in tion of the total number of work fatali- over three 6-year periods (1976-1981, ties in Canada. In particular, work 1982-1987, 1988-1993) because of the manufacturing deaths prior to 1995 may be seriously limited number of observations available From 1976 to 1993, almost 17,000 under-counted in the agriculture indus- in any one year. try, where WC coverage has been tradi- Canadians were fatally injured in * Nationally, over 90% of reported tionally excluded. However, the trend the course of, or as a result of, their to expand compulsory coverage now cases are accepted for compensation by employment (Table 1) 1 – an aver- includes workers in agriculture, who, Workers’ Compensation Boards and age of more than two deaths a day. since 1995, have had compulsory cov- Commissions. The remaining cases may Four industries – manufacturing; erage in most jurisdictions. not be compensated for a number of rea- construction; transportation and sons; for example, the job is not covered Newfoundland and New Brunswick by the legislation, or the illness is not storage; and mining, quarrying and have not provided fatality reports since covered or deemed work-related. oil wells – accounted for 63% of all 1991, nor has Quebec since 1992. fatalities (over 10,000 deaths). In each of the three periods examined, the highest number of work-related From 1988 to 1993 (the most deaths were traced to manufactur- recent period examined), 863 Katherine Marshall is with the Labour ing (one in 6); but, as in many other people died while working in and Household Surveys Analysis Division. She can be reached at (613) industries, fatality counts have manufacturing or as a result of hav- 951-6890. dropped over time. ing once been employed there. 26 / Summer 1996 PERSPECTIVES Statistics Canada - Catalogue no. 75-001-XPE A job to die for Definitions Work-related fatality: Any fatal in- group (for example, an industry, occu- The ratio is multiplied by 100,000, jury, disease or illness resulting from pation or age range) who died from a which allows the rate to be expressed a work-related incident that has been fatal injury, disease or illness resulting in whole numbers, and thus represents accepted for compensation by a from their employment, in a specific the number of deaths per 100,000 Workers’ Compensation Board or period, divided by the sum of the paid workers. Commission. An accepted case annual average paid worker counts for means the death was connected with, that group during that period. (In some An employment-based fatality or directly related to, the performance provinces, the year of accident or claim rate, such as the one used in this pa- of the worker’s current or past job. acceptance determines the time frame per, measures the fatality risk for assigned. See Data sources and limita- those employed at the time the LFS A death may occur at a worker’s tions.) As an example, the national was carried out. However, the annual usual place of work (for example, a fatality rate in the construction industry average number of paid workers does mine) or anywhere else duties are from 1988 to 1993 is calculated as fol- not reflect the movement of workers performed (an outdoor hydro line for lows: in and out of the workforce, nor does instance). The death may be instanta- it account for the volume of work per- neous (caused by a fall, for example) (N / W) x 100,000 formed (for example, part-time or or may happen much later (years where part-year employment), which may after exposure to carcinogens at N = the total number of work-related differ for various groups. A number work). fatalities in construction from of factors may affect fatality rates: 1988 to 1993 (from WC Boards worker experience, number of hours Fatality rate: Although a fatality count worked, physical and other demands is important, a more telling indicator and Commissions) of the job, occupational environment, of risk is the fatality rate, which ex- W = the sum of the annual average condition of equipment, human error, presses the number of work-related counts of paid workers in con- legislation, and presence and effec- deaths per 100,000 paid workers. struction from 1988 to 1993 tiveness of health and safety regula- More specifically, the rate represents (from the LFS). tions. the number of people in a particular Table 1 Fatalities and fatality rates by industry, 1976 to 1993 Fatalities Fatality rates * 1976-93 1976-81 1982-87 1988-93 ** 1976-93 1976-81 1982-87 1988-93 ** All industries 16,668 6,260 5,437 4,971 9 11 9 7 † Agriculture 295 84 113 98 10 9 11 10 Fishing and trapping 363 120 128 115 145 182 155 113 Logging and forestry 1,022 405 338 279 90 97 89 82 Mining, quarrying and oil wells 2,228 836 748 644 70 79 69 63 Manufacturing 2,942 1,094 985 863 8 9 8 8 Construction 2,750 1,047 874 829 27 32 28 23 Transportation and storage 2,520 1,092 744 684 29 36 26 25 Communication and other utilities 343 128 112 103 5 6 5 4 Wholesale trade 891 317 339 235 10 11 11 7 Retail trade 395 158 127 110 2 2 2 1 Finance, insurance and real estate 132 47 41 44 1 1 1 1 Government services 986 356 360 270 7 8 7 5 Other services †† 1,404 448 408 548 2 3 2 2 Industry not stated 397 128 120 149 .. .. .. .. Sources: Workers’ Compensation Boards and Commissions, and Labour Force Survey * Work-related deaths per 100,000 paid workers (see Definitions). ** Fatality counts are missing for Newfoundland and New Brunswick in 1992 and 1993, and for Quebec in 1993; paid worker counts for these provinces and years were excluded from the denominators in fatality-rate calculations. † Agricultural workers have traditionally been excluded from Workers’ Compensation coverage, so the fatality counts and rates shown are an under-representation (see Data sources and limitations). †† Includes business services; educational services; health and social services; accommodation, food and beverage services; and other services, such as amusement and recreational services, personal and household services, and membership organizations. Statistics Canada - Catalogue no. 75-001-XPE Summer 1996 PERSPECTIVES / 27 A job to die for However, because of this indus- added the dangers of handling haz- lower-than-average absence rates try’s high number of paid workers,2 ardous materials or heavy machin- (Akyeampong, 1995). its fatality rate was relatively low ery. Such work inevitably places (8 per 100,000). In contrast, many people at greater risk of injury or Rates vary by region other industries were responsible death. for fewer deaths but had much The most recent six-year period higher fatality rates, because of while rates in most service indus- (1988 to 1993) shows that British their relatively small workforces. tries have always been low Columbia had the highest death The safest industries are in the rate at 15 per 100,000 paid work- Although fatality rates have service sector.4 With one exception, ers, followed by the Atlantic prov- been falling ... fatality rates were below 12 per inces with 8, the Prairies and 100,000 paid workers in all three Quebec both with 7, and Ontario Over the 18 years covered by this with 6. Although all regions saw a periods studied. For example, in re- study, there was a substantial re- tail trade and in finance, insurance steady decline over the study’s duction in the national fatality rate: time frame, the Prairie provinces and real estate, the death rates were from 11 per 100,000 paid workers and British Columbia experienced only one per 100,000 between in the 1976 to 1981 period, to 7 per 1988 and 1993. An exception was the most dramatic improvements: 100,000 between 1988 and 1993 from 14 to 7 deaths per 100,000, the transportation and storage in- (Table 1). In industries with ini- and from 20 to 15, respectively dustry, which registered a rate of 25 tially low fatality rates, figures re- during this period; nevertheless, (Chart). mained relatively unchanged over this figure was an improvement As previously noted, fatality the three periods; in industries with over death rates seen in earlier pe- rates vary by industry, which high rates between 1976 and 1981, riods. Another study, which looked death rates steadily decreased. For accounts in part for regional differ- at absences from work due to ill- ences. In the 1988 to 1993 period, example, the fishing and trapping ness or disability, also found that one in 7 paid workers in British industry fatality rate improved the service sector, particularly from 182 in the first period to 113 Columbia, as well as in the Prairies trade and the finance, insurance and the Atlantic provinces, were in the last. and real estate industries, had employed in fishing and trapping; jobs in primary industries are still risky ... But in spite of these gains, fatality Chart rates remained high in most Work-related fatality rates have been falling in all regions. primary industries: fishing and trapping (as mentioned above); Fatality rate * logging and forestry (82 per 20 100,000 paid workers between 1988 and 1993); and mining, quar- 16 rying and oil wells (63). The rela- tively low fatality counts and rates for agriculture are an under-repre- 12 sentation because until recently this industry has been excluded 8 from compulsory Workers’ Com- pensation coverage (see Data 4 sources and limitations). American and regional Canadian studies in- 0 dicate that agricultural work is in British Columbia Prairies Ontario Quebec Atlantic fact riskier than average. 3 1976-1981 1982-1987 1988-1993 ** Work in the primary industries usually requires physical exertion Sources: Workers’ Compensation Boards and Commissions, and Labour Force Survey and dexterity; in addition, many * Work-related deaths per 100,000 paid workers (see Definitions). employees must work outside in all ** Fatality counts are missing for Newfoundland and New Brunswick in 1992 and kinds of weather, as well as in rug- 1993, and for Quebec in 1993; paid worker counts for these provinces and ged terrain, on water or under- years were excluded from the denominators in fatality-rate calculations. ground. To these difficulties are 28 / Summer 1996 PERSPECTIVES Statistics Canada - Catalogue no. 75-001-XPE A job to die for logging and forestry; mining, quar- rying and oil wells; construction; Table 2 and transportation and storage – The 10 most dangerous jobs, 1988-1993 the five most dangerous industries. Fatalities Fatality rates * In comparison, one in 10 employ- ees in both Quebec and Ontario worked in these industries. All occupations 4,971 7 Top 10 occupations combined ** 1,354 58 Workers in British Columbia were the most likely to be em- Mining and quarrying: cutting, handling, loading 212 281 ployed in high-risk industries (16% Construction: insulating 65 246 of all paid workers in this prov- Mining and quarrying: labouring 56 139 Air pilots, navigators and flight engineers 77 137 ince). They also experienced Timber cutting 108 123 higher-than-average fatality rates Log hoisting, sorting and moving 64 116 in most of them. For example, from Net, trap and line fishing 67 110 1988 to 1993 the province posted Truck drivers 496 38 Construction: labouring 122 35 the highest fatality rates of all the Construction: pipefitting and plumbing 87 31 regions involved in logging and forestry (131 per 100,000 paid Sources: Workers’ Compensation Boards and Commissions, and Labour Force Survey workers, compared with a low of 28 * Work-related deaths per 100,000 paid workers (see Definitions). ** Based on the 1980 Standard Occupational Classification, coded to the 4-digit level; in Ontario), construction (37 com- each occupation listed had at least 50 deaths between 1988 and 1993. pared with 8 in Quebec), and trans- portation and storage (46 compared with 15 in the Atlantic provinces). British Columbia’s rugged terrain Fatality rate is highest for 100,000 for men aged 65 and over. probably makes logging, road con- older men The aging process, which reduces struction and highway driving more agility, stamina and overall health, Most workers whose deaths were is one factor that puts older men at dangerous than in other regions. registered in the 1988 to 1993 pe- greater risk of injury or death, par- riod were men (96%); this propor- ticularly if workers are in the more Occupations in mining are tion varied little by age group. physically demanding blue-collar the most dangerous Presumably, more men than women jobs. In addition, many older peo- In order to determine which occu- die from job-related causes be- ple, who have long since retired, cause of their over-representation die from exposure to environmen- pations are the most dangerous it is in the higher-risk industries and tal hazards encountered earlier in important to focus on fatality rates, rather than counts. For example, occupations. Women are more their working lives. likely to work in the “safer” serv- between 1988 and 1993 a relatively ice industries. Causes of death low number of deaths (64) were as- sociated with log hoisting, sorting The largest number of work One in five work-related fatalities and moving, but because relatively deaths (929) was among people accepted by Workers’ Compensa- few people worked in this job, it aged 25 to 34, accounting for 19% had a high fatality rate of 116 per tion Boards and Commissions for of all deaths between 1988 and the 1988 to 1993 period were 100,000. Similarly, truck driving 1993. However, the fatality rate of caused by exposure to harmful sub- accounted for 10% (496) of all fa- this age group was low, hovering talities (4,971) in the 1988 to 1993 stances, including poisons, chemi- just above zero for women and cals, allergens, carcinogens and period, but had a death rate of only reaching 8 per 100,000 paid work- radiation. Almost as many involved 38 per 100,000 paid workers, plac- ers for men (Table 3). ing it eighth among the 10 occupa- transportation vehicles, mostly Among women, fatality rates trucks and cars (65%), but this tions with the highest fatality rates remained below 5 per 100,000 re- category also covers aircraft and (Table 2). gardless of age. The picture was watercraft. Workers also died from Two occupations had strikingly quite different for men, however. being struck or caught by objects, high fatality rates in this period: Their rates initially rose slowly with and from falls, overexertion, fire mining and quarrying (cutting, age – from 6 per 100,000 for 15 to and explosion, electrocution and handling and loading) and con- 24 year-olds, to 12 for 45 to 54 violence5 (Table 4). struction (insulating) had rates of year-olds, and to 28 for 55 to 64 The two leading causes of death 281 and 246 per 100,000 paid year-olds – then jumped to 203 per workers, respectively. affect certain age groups differ- Statistics Canada - Catalogue no. 75-001-XPE Summer 1996 PERSPECTIVES / 29 A job to die for which to be employed. In terms of Table 3 specific high-risk occupations, Fatalities and fatality rates by age and sex, 1988-1993 miners who cut, handled and loaded material had the highest fa- Fatalities * Fatality rates ** tality rate (281 per 100,000). High- Men Women Men Women risk jobs tend to be held by men, who accounted for 96% of all work- Total 4,788 181 13 1 related deaths. 15-24 † 371 24 6 - Occupational deaths are tragic 25-34 888 40 8 - and expensive to compensate. The 35-44 800 42 9 1 monitoring and surveillance of 45-54 733 27 12 1 55-64 898 21 28 1 such fatalities can shed light on the 65+ 887 11 203 †† 4 effectiveness of safety policies and Age not stated 211 16 .. .. programs over time, and help pin- Sources: Workers’ Compensation Boards and Commissions, and Labour Force Survey point work environments that may * Excludes two cases where sex was not reported. require further preventive inter- ** Work-related deaths per 100,000 paid workers (see Definitions). vention. Certainly, it is encourag- † Fatalities are reported for workers 14 years and over. ing that Canada’s fatality rates †† This high rate reflects the large number of retirees who died from earlier exposure to have progressively improved since harmful substances. the mid-1970s. ently. Exposure-related death, Summary which can happen gradually, ac- Acknowledgement counted for 55% of all work fatali- From 1988 to 1993 the manufactur- ties among persons aged 65 and ing industry experienced the high- The author wishes to thank John est number of work-related Hemingway, Occupational Safety over and 13% among persons un- fatalities among paid workers and Health Branch, Human Re- der 65. In contrast, transportation sources Development Canada, for accidents were the leading cause of (863); nevertheless, this industry had one of the lowest death rates (8 his time and effort in providing data death among those under 65 years and documentation used in this (26%), but accounted for only 3% per 100,000). At 113 deaths per article. of deaths for persons 65 and over. 100,000, the fishing and trapping industry was the most dangerous in o Fatality costs are deadly The cost of a fatality, in terms of survivors’ benefits, is huge. In Table 4 1993, $5.6 billion was awarded for Causes of work fatalities, 1988-1993 881,512 injury and fatality claims Number % for all provinces combined. Although fatalities accounted for All fatalities 4,971 100 only 735 (or 0.1%) of these claims, they cost $361 million in benefits Exposure to harmful substance * 982 20 (6.4% of the total). As a result, Involving transportation vehicle 933 19 Struck by object 891 18 the cost per fatality averaged Fall from elevation 425 9 $492,000; in comparison, the Bodily exertion and heart attack 311 6 cost per injury averaged $6,000. Caught in, on, between object(s) 255 5 Beyond the costs assumed by Struck against 194 4 Workers’ Compensation, there may Fire and explosion 139 3 Contact with electric current 138 3 have been increased insurance Violence 93 2 premiums for the employer as Fall from same level 78 2 well as safety violation fines. Of Other ** 532 11 course, the emotional cost to the Sources: Workers’ Compensation Boards and Commissions victims’ families and friends is * Half of all exposure-related deaths (497) happened to people aged 65 and over. incalculable. ** Includes past injuries and cases not elsewhere classified. 30 / Summer 1996 PERSPECTIVES Statistics Canada - Catalogue no. 75-001-XPE A job to die for n References Newfoundland and Labrador. Workers’ Com- pensation Commission. Nineteen Hundred n Notes Akyeampong, E.B. “Missing work.” Perspec- and Ninety Three Annual Report. St. John’s, tives on Labour and Income (Statistics Canada, 1994. 1 The figures in this study are for the Catalogue no. 75-001-XPE) 7, no.1 (Spring provinces only. The Yukon and Northwest 1995): 12-16. Territories are excluded because the Labour Nova Scotia. Workers’ Compensation Board. Force Survey does not provide their paid 1993 Annual Report. Halifax, 1994. Alberta. Workers’ Compensation Board. worker counts, which are required in the Annual Report 1993. Edmonton, 1994. calculation of fatality rates. Ontario. Workers’ Compensation Board. 1993 Annual Report. Toronto, 1994. Association of Workers’ Compensation 2 Between 1988 and 1993, 17% of all paid Boards of Canada (AWCBC). Workers’ Com- workers were employed in manufacturing. Prince Edward Island. Workers’ Compensa- pensation Industry Classifications, Assess- tion Board. Forty-fifth Annual Report for ment Rates, and Experience Rating Programs the Year Ended December 31, 1993. 3 The 1993 Census of Fatal Occupational in Canada. Edmonton: AWCBC, 1995. Injuries in the United States showed a fatality Charlottetown, 1994. rate of 34 per 100,000 agricultural workers Brison, R.J. and W. Pickett. “Fatal farm inju- (Toscano and Windau, 1994), and a Canadian Quebec. Commission de la santé et de la ries in Ontario, 1984 through 1992.” Cana- sécurité du travail. Report of Activities 1993. study of fatal farm injuries in Ontario yielded dian Journal of Public Health 86, no.4 (July- a rate of 56 per 100,000 farms from 1984 to Québec, 1994. August 1995): 246-248. 1992 (Brison and Pickett, 1995). Saskatchewan Workers' Compensation Board. British Columbia. Workers’ Compensation 1993 Annual Report: Statistical Supplement. 4 The service sector includes transporta- Board. Annual Report 1993: Part 2 - Statis- tion and storage; communication and other Regina, 1994. tics ’93. Victoria, 1994. utilities; wholesale trade; retail trade; fi- nance, insurance and real estate; government Statistics Canada. Standard Occupational Grayson, J.P. “Perceptions of workplace haz- Classification, 1980. Catalogue no. services; and other services. ards.” Perspectives on Labour and Income 12-565-XPE. Ottawa, 1981. (Statistics Canada, Catalogue no. 5 Canadian and American work-related fa- 75-001-XPE) 6, no.1 (Spring 1994): 41-47. tality data follow similar patterns, with one Toscano, G. and J. Windau. “The changing exception. In Canada, violence in the character of fatal work injuries.” Monthly Manitoba. Workers’ Compensation Board. Labor Review 117, no.10 (October 1994): workplace has remained stable and rare (2%); Annual Report 1993. Winnipeg, 1994. in the United States, violence (predominantly 17-28. shootings) was the second leading cause of New Brunswick. Workers’ Compensation death at work in 1993 (following highway Board. Seventy-fifth Annual Report of the fatalities), accounting for 21% of work-re- Workers’ Compensation Board. Saint John, lated fatalities (Toscano and Windau, 1994). 1994. Statistics Canada - Catalogue no. 75-001-XPE Summer 1996 PERSPECTIVES / 31
"A job to die for"