INDUSTRIAL ACCIDENTS. In the term industrial acci dent, the adjective implies that the accident results in injury to a person and has some causal connection with work, employ ment or occupation. The relevant wording of the Workmen's Compensation Act, 1923, of Great Britain runs ". . . accident arising out of and in the course of employment"; this definition has been incorporated in laws enacted in the British Dominions and the United States, while in other countries the definition though different in wording is similar in intent. There is a dis tinct tendency towards making the term cover several "health risks" also, for these, though "accidental" only in a limited sense, have an eminently "industrial" character. Experience shows that industrial accidents (including industrial diseases) occur with great regularity in all industries and cause a loss to persons engaged in it. They therefore constitute a risk to labour. This risk may be regarded under different aspects. First, it may be qualified as an "economic risk"; secondly, accidents imply a human cost, a "personal risk," because attended with physical and mental suffering, both on account of the injury itself and because wage earners exposed to the risk generally have a very small margin between income and necessary expenditure.
For Great Britain, no comprehensive statistics of all industrial accidents exist. The statistics of workmen's compensation cover seven principal groups of industries, and all accidents disabling for at least three days and for which compensation was paid for the first time during the year. It is estimated that not more than half of the employees under the Workmen's Compensation Act are thus covered by the statistics, the principal exclusions being the building industry, agriculture and domestic service. The figures for 1926 are as follows: The German statistics cover practically all industries except small undertakings and domestic service, agriculture being given apart. In 1926, the number of insured persons in industries was 10,794,031, the number of fatal accidents, 5,429, and that of non fatal accidents, 826,899, the rates being 0.5o fatal and 76.61 non fatal cases per i,000 persons ; in agriculture, there were 14,068,000 insured persons, 2,682 fatal and 176,119 non-fatal cases; or 0.19 and 12.52 cases respectively per 1,00o persons. In France, the total number of accidents in all industries, including agriculture, commerce, transport and domestic service, but excluding mines and railways, in 1926 was 2,392 fatal and 981,244 non-fatal; the number of coal-mining accidents (in 1925) was 354 fatal and 93,516 non-fatal, while the railway employees sustained (in 1923) 387 fatal and 615 severe non-fatal accidents. The situation in a small highly industrialized country is illustrated by the Belgian statistics; they covered (in 1921) 601,841 workers and recorded 209 fatal and 95,146 non-fatal accidents, i.e., o.35 fatal and 158.09 non-fatal cases per i,000 workers.
There is one important industry in respect of which an inter national comparison may be attempted, viz., coal-mining. Owing to the fact that work underground involves quite particular occupational risks, and coal-mining actually is the most hazardous industry (with the possible exception of shipping), special statis tics relating thereto are more developed than others. The accident exposure may be expressed• for the several countries in uniform terms (though in Germany allowance is not made for overtime). Nevertheless, the reporting of accidents is different, and while the fatal accident rates are fairly comparable, those of non-fatal cases are less so. The figures for 1925 are as follows: In the United States, too, no statistics for all jurisdictions and all industries are compiled. Summing up the information avail able for 38 States in 1925, it is •found that the total number of fatal cases was 10, 53 7, and that of non-fatal cases, 1,68 7,95 7. It should be noted that these figures exclude agriculture, domestic service, maritime and railway employments in all States, and all accidents in 14 States and territories. More accurate statistics are compiled for a few important industries; the number and the frequency rate per i,000 full-time workers in 1925 were as : The total number of employees killed in steam railway acci dents in 1925 was 1,594, and of those injured, 118,874; the frequency rates among trainmen were 2.14 fatal and 87.66 non fatal cases per i,000 men in service.
Finally, as regards industrial diseases assimilated to industrial accidents, the scope of the statistics is determined by the diseases compensated. The most comprehensive statistics are those for Great Britain which cover 32 diseases. The total number of cases for which first payment was made in the seven groups of indus tries was in 1925, 41 fatal, 8,792 non-fatal, in 1926, 27 fatal, 5,816 non-fatal. The bulk of these cases (87 and 79 per cent. respectively) occurred in the mining industry, and the consider able decrease noted in 1926 is largely due to the long coal stop page in that year.
In Germany, the statistics refer to 11 diseases; the total number of cases reported in the year 1925-26 was 3,310, of which 2,781 (72 per cent.) were due to lead poisoning. In France, only two industrial diseases are compensated; in 1925 there were cases of lead poisoning and 5 of mercury poisoning of occupational origin.
The British statistics do not contain precise information as to the extent of disability caused by accidents, beyond the distinc tion between fatal and non-fatal cases. According to an estimate, based on the statistics of several other European countries, of the general distribution of industrial accidents usually reported, less than 1 per cent. result in death, about 5 per cent. in perma nent disability, and about 94 per cent. in temporary disability. It should, however, be remembered that both from the human and the economic point of view, the few cases of death or per manent disability of high degree are of more importance than the bulk of minor accidents. This point is the basis of the calcula tion of "severity rates," which show the number of working days lost on account of accidents, related to the accident exposure.
In order to know the working time lost, an arbitrary value is fixed for cases of death (6,000 days lost in the United States, 7,50o days in Sweden and Switzerland); the time loss entailed by permanent disability is assessed at a given fraction of this basic value (e.g., loss of a thumb equal to 600-1,000 days lost) ; the time loss resulting from temporary cases is recorded directly. The accident exposure is the number of workers or "full-time workers" or man-hours, as in the case of frequency rates.
The severity rate gives a truer idea of the actual risk of indus trial accident than the frequency rate; but it is rather trouble some to calculate, and has therefore as yet been employed neither in the British nor in most other statistics. The following instances drawn from the American records will illustrate the results which have been obtained by the calculation of severity rates as corn pared with those given by frequency rates. In the United States iron and steel industry, during the five-year period 1920-24, the total number of full-time workers was 1,839,818, and the number of cases, as well as the frequency and severity rates per million hours' exposure, were as follows: The United States National Safety Council has calculated on similar lines the accident experience of 13 industries employing 828,028 workers and 74,367 cases of accident for 1925. The frequency and severity rates per million man-hours were as follows : Causes of Industrial Accidents.—The principal causes of accidents in factories and workshops are shown in the report for 1924 of the Chief Factory Inspector of Great Britain (covering accidents) and in the reports for the same year of corn pensation authorities of 21 American States (covering 838,042 accidents) ; the figures in the table in next column represent per centages of all accidents.
As regards particularly coal-mining accidents, the statistics show that the largest part of them are due to falls of roofs (in 1926, 33 per cent. in Great Britain; 43 per cent. in the United States), haulage and explosions underground coming next. The accidents to railway employees are classed in the British statistics in (i.) "train accidents," mostly caused by collisions and derail ments; (ii.) "movement accidents," the most frequent causes of which are shunting operations, including coupling and uncoupling of vehicles, working on permanent way, and attending to engines in motion; and (iii.) "non-movement accidents" attributable par ticularly to loading and unloading of wagons, attending to engines at rest, falling of wagon doors and lamps, working on the line, etc. In the American statistics, accidents to (i.) trainmen, occur most frequently to the class of brakemen (55 cases per million hours' exposure, the general average being 30), while accidents to (ii.) non-train employees are principally attributable to handling (7.4 out of 20.7 cases per million hours) . It may be noted that shunting operations present a conspicuously high risk, a fact which has given rise to the question of introducing automatic couplings for trains. Finally, the British statistics of shipping accidents for the 5-year period 1921-25 show that the yearly average of accidental deaths amongst the crews of (mercantile and fishing) vessels was 724, of which 327 were due to casualties to vessels and 395 to other accidents at sea or in river or harbour.
These classifications, however, do not refer to "causes" prop erly so-called. The determination of the real and most important causes is a difficult task, but it may be briefly stated that each accident is the outcome of a long train of events which is ulti mately traceable to some failure of human foresight or insight. Accident statistics have attempted to find out these ultimate causes on two different lines.
Some earlier statistics classified accidents by what may be called their "moral causes" or responsibility for the occurrence of acci dents. For instance, according to the German statistics for 1907, 12 per cent. of accidents were attributed to the fault of the employer, 41 per cent. to that of the worker, 7 per cent. to both of them or to a third person, while in not less than 4o per cent. of cases no fault was established.
The failure of care, attention and skill which underlies the occurrence and incidence of the great majority of accidents, results from the combined action of numerous factors, one of the most important of which is inexperience. Evidence obtained from the American metal trades shows that the accident frequency of new workers engaged diminished rapidly from week to week and from month to month, so that the accident rate during the first week was 90 times higher than six months later, while, among men having worked for six months, it was 2.5 times higher than among those employed for from three to five years. Further, accident frequency is larger during the night than the day, night work being always more or less abnormal. Experiments made in the British munition factories during the World War and in the American iron and steel industry showed that the hour-to-hour accident incidence was almost completely reversed during the day and during the night; the day accident rate reached its maximum toward the end of the shift, while the night rate fell for the most of the shift. Hence it appears that the accident incidence is influenced by the mental state of the workers rather than fatigue. Fatigue, however, also plays a part in accident causation when working hours are particularly long, e.g., in docks and railway service, and when heavy muscular effort is required, especially in the case of women. On the other hand, alcoholism has been found to be a potent cause of accidents, especially chronic alco holism, i.e., excessive regular indulgence; excessive drinkers have been found to be about three times more liable to accident than other persons.
By the economic consequences are meant the loss of working days and loss in earning capacity of the injured persons. With a view to showing the consequences of accidents taken as a group, in this respect, these are classified according to the "ex tent" and "degree of disability." The main classification by the extent of disability was discussed above in connection with acci dent severity. The "degree of disability" is fixed according to different criteria in different groups of disabilities. In fatal cases it is indicated by the age of the killed person; moreover, these cases are classified by the number of his dependents. Non-fatal cases resulting in "permanent disability" are often classified according to the percentage of incapacity for work, total dis ability (e.g., loss of both eyes) being taken as equal to Ioo. Cases of "temporary disability" which form about 94% of all accidents, are subdivided according to the number of working days lost. According to an estimate based on the statistics of a number of countries, this loss is in one-third of the cases less than a week; in two-thirds, less than a month; and in about 95% of cases, less than three months.
Finally, industrial accidents involve important financial con sequences. The loss of earnings consequent upon industrial acci dents affects in the first place the workers, but a part of this loss is, in practically all countries, a charge on the employers in virtue of workmen's compensation and accident insurance legislation ; moreover, the State participates in many countries by bearing the cost of the administration of insurance insti tutions. International comparisons are still less feasible in this field than in other parts of accident statistics, and for several countries, e.g., the United States and France, no information is available. In Great Britain, the regular statistics only show a part of the cost ; a recent official estimate, however, puts the total charge on employers in 1925 (including compensation, adminis trative expenses, commissions, profits, etc.) at £12,117,000. In Germany, the total of payments of the industry and of the State to the accident insurance institutions was in the same year RM. 266,916,000 (L12,934,000). And it should be remembered that these amounts represent hardly more than one-third to one half of the aggregate burden which industrial accidents consti tute on labour and industry as a whole.
Official Statistics of the United States: U.S. Department of Labor, Bureau of Labor Statistics. Bulletin No. 157 (1915) . Industrial Accident Statistics. No. 234 (1918). Safety Movement in the Iron and Steel Industry, 1907 to 1917. No. 251 (1919). Preventable Deaths in the Cotton Manufacturing Industries. No. 256 (1919). Accidents and Accident Prevention in Machine Building. No. 276 (192o). Standard ization of Industrial Accident Statistics. No. 298 (1922). Causes and Prevention of Accidents in the Iron and Steel Industry, 1910-1919. No. 339 (1923) . Statistics of Industrial Accidents in the United States. No. 425 (1927). Record of Industrial Accidents in the United States to 1925. U.S. Department of Commerce, Bureau of Mines. Coal-mine fatalities in the United States, 187o-1914 and subsequent years. Metal mine Accidents in the United States. Quarry Accidents in the United States. Accidents at Metallurgical Works in the United States. Coke oven accidents in the United States. Interstate Commerce Commission, Bureau of Statistics. Quarterly Accident Bulletins, Railroad Accidents in the United States.
International Labour Office (League of Nations) . Studies and Reports. Series F. (Second section) No. i (5924). Automatic Couplings and the Safety of Railway Workers (1928). Series N, No. 3 Methods of Statistics of Industrial Accidents. Series N, No. 14 Q[928). Statistics of Occupational Diseases. Prevention of Industrial Accidents. Joseph L. Cohen. Workmen's Compensation in Great Britain (1923) Boyd Fisher. Mental Causes of Accidents (1922) . P. Sargant Florence. Economics of Fatigue and Unrest (1924). Edouard Fuster. Statistique internationale des accidents du travail (Bulletin des assurances sociales 1913). T. M. Rubinow. Standard Accident Table as a Basis for Compensation Rates (1915) . H. M. Vernon. Industrial Fatigue and Efficiency (1921) ; The Human Factor and Industrial Accidents (Inter national Labour Review 1926) . T. Voionmaa, Alcoholism and Indust rial Accidents (International Labour Review 1925) . (T. V.)