Accident Mortality Among Children

Bull. Org. mond. Sante 1956, 123-163 Bull. Wld Hlth Org. 15,

ACCIDENT MORTALITY AMONG CHILDREN

S. SWAROOP Chief, Statistical Studies Section, World Health Organization R. M. ALBRECHT Epidemiological Studies Section, World Health Organization B. GRAB Statistical Studies Section, World Health Organization

SYNOPSIS The authors present statistics on mortality from accidents, with special reference to those relating to the age-group 1-19 years. For a number of countries figures are given for the proportional mortality from accidents (the number of accident deaths expressed as a percentage of the number of deaths from all causes) and for the specific death-rates, per 100000 population, from all causes of death, from selected causes, from all causes of accidents, and from various types of accident. From these figures it appears that, in most countries, accidents are becoming relatively increasingly prominent as a cause of death in childhood, primarily because of the conquest of other causes of death-such as infectious and parasitic diseases, which formerly took a heavy toll of children and adolescents-but also to some extent because the death-rate from motor-vehicle accidents is rising and cancelling out the reduction in the rate for other causes of accidental death. In the authors' opinion, further epidemiological investigations into accident causation are required for the purpose of devising quicker and more effective methods of accident prevention.

Introduction The First International Conference of National Committees on Vital and Health Statistics, held in London in October 1953,23 pointed out that accidents constituted one of the major elements of morbidity and mortality in highly developed countries and accordingly recommended that the World Health Organization continue or initiate statistical studies on this problem. In January 1955 the Government of Sweden, in a com- munication submitted to the WHO Executive Board,a made the following

a Unpublished working document EB15/53 504 - 123- 9 124 S. SWAROOP, R. M. ALBRECHT & B. GRAB

statement: " In many countries, death through accidents is now the main cause of deaths among children 1-15 years. In some countries and age- groups the mortality rate for accidents alone is as high as the mortality rate for all diseases together." A proposal was accordingly made that the importance of the problem be studied through a statistical investigation in different countries. It was felt that in some countries it would probably be necessary to stimulate interest in this field for obtaining at least some information concerning the importance of accidents as a factor of mortality. The Board adopted a resolution requesting the Director-General of WHO to examine this problem.b During the discussion of the problem it was remarked that deaths from accidents among children in the age-group 1-15 years should be split up into various sub-groups by age as well as by type of accident, in order to focus attention on the type of action which was needed in each case. The variation from one country to another was also suggested for examination. The subject again came up for discussion in September 1955 during the fifth session of the WHO Regional Committee for Europe, which included in its 1956 programme a meeting to study the prevention of accidents in childhood and also chose for its Technical Discussions in 1956 the subject, " The Prevention of Accidents in the Home ". During the seventeenth session of the WHO Executive Board, the Netherlands Government referred to the question of accidents in childhood as constituting a serious health problem, pointing out that in the Netherlands violent death ranked first among causes of death in the age-group 1-14 years. In this age-group it formed 25 % of the causes of death, while in the age-group 5-14 years the percentage even rose to 33. Further, the number of children dying through accidents was six times as high as the number of children dying from tuberculosis and more than three times as high as the number of children dying from diphtheria. The significance of this public health problem is further increased when it is remembered that for each fatal accident there are many non-fatal ones which cause prolonged or even permanent disability. In so far as the prevention of road accidents is concerned, interest in the subject at the international level dates back to the year 1934, when the League of Nations sent a letter to governments asking whether they kept statistics on road accidents and their causes, and if so, on what basis they were compiled. The Committee for the Unification of Statistics Relating to Road Traffic Accidents was established in 1935 under the auspices of the League of Nations Organization for Communications and Transit. The Committee met in Geneva in September 1936 and May 1937,14 and its recommendations covered the collection and compila- tion of statistics in a uniform manner, including data on the number and kinds of vehicles involved in accidents, the number of persons killed or

b Resolution EB15.R64 (see Off. Rec. Wid Hlth Org. 60, 26) ACCIDENT MORTALITY AMONG CHILDREN 125

injured, and the nature, circumstances and causes of accidents. Further, it recommended that countries should adopt the International List of Causes of Death, drawn up in Paris in 1929 (Fourth Decennial Revision), and that the road-accident statistics should be accompanied if possible by statistics of deaths due to means of transport, compiled according to this scheme. Interest in the prevention of road accidents has been revived recently by the United Nations Economic Commission for Europe through its Sub-Committee on Road Transport, which in 1950 set up a Working Party on the Prevention of Road Accidents. The Ad Hoc Committee on Inland Transport of the Economic Commission for Asia and the Far East also set up a working party in October 1950 to consider, inter alia, the problem of accident statistics. The International Union for Child Welfare has also been interested in the problem of- prevention of children's accidents and has published a review of the problem.3 13 It has been observed by many previous writers on this subject-Godfrey,9 Gordon,'0' 11 Armstrong,' Cruickshank,6 Logan & Gissane,15 Charles,4 and McFarland,16' 17 to name only a few-that the study of accidents is amenable to the same type of epidemiological approach as is applied to disease processes, because an explanation of accident causation can also be sought in the interactions of the host, the agent and the environment. Indeed, the frequency of accidental injury shows geographical variations, seasonal changes, and differences by place of occurrence, sex and age, as well as variations by other population characteristics. Accidents can thus be interpreted as resulting from a totality of factors involved in the competition between the individual and the environment. As is revealed in the subsequent discussion relating to the classification of causes of accidental deaths by age, sex, etc., infants and pre-school children or schoolchildren are found to be affected by different kinds of injury. A clear appreciation is therefore necessary of which particular type of injury is more frequent in which particular population segments; where, when and how injuries occur and what are the special circumstances or factors contributing to their increase. The host, i.e., the individual affected, is of primary concern from the medical point of view, because of the physical, physiological and psychological characteristics at his various stages of development and under various conditions of exposure. Addi- tional information is needed not only in regard to the host but also in respect of the agent and the environmental factors determining or influencing the accident. Inevitably, many details are suppressed when statistical tables for a country as a whole are compiled and submitted to WHO. Indeed, mortality data by themselves give little indication of the complex factors whose interaction results in accidental injury. In undertaking the present study we are therefore handicapped by the absence of all such detailed knowledge as would enable us to present the problem in its proper perspective. Nevertheless, the statistical information made available to 126 S. SWAROOP, R. M. ALBRECHT & B. GRAB

WHO by a large number of countries in different stages of development does throw valuable light on the relative importance of accidents of various kinds and in different types of population groups as a cause of death, and also indicates certain features of the problem which would appear to be common to all countries. The information supplied to WHO by Member States has therefore been analysed statistically, and the results are presented here in the belief that they will be of interest to workers in the field of accident prevention. The World Health Organization has already published a series of tables on mortality from transport accidents,21 followed by another series on mortality from accidents other than transport accidents.22 The present study supplements these tables and gives the results of statistical analysis of the mortality data which had been made available to WHO up to the time of writing.

Available Statistical Material As has been briefly stated already, the only statistical material we have had access to at the international level relates to deaths from accidents. Different countries have, to some extent, followed different procedures in classifying deaths by cause. In order to attain uniformity in the classification of causes of death, WHO has recommended the use of the Manual of the International Statistical Classification of Diseases, Injuries, and Causes of Death (Sixth Revision of the International Lists of Diseases and Causes of Death), which was adopted by the First World Health Assembly in 1948.20 In the Manual, three lists of causes of death are given: a detailed list; an intermediate list (A) of 150 categories; and an abbreviated list (B) of only 50 categories. Further, especially in respect of deaths resulting from accidents, poisoning and violence, provision has been made for the classification of deaths simultaneously in two ways-first, according to the external cause of the injury (" E" classification) and, secondly, according to the nature of the injury (" N " classification). The E and N classifications are independent, and either or both can be used. For instance, a fracture of the base of a motorcyclist's skull which was caused by collision with another motor vehicle can be coded under category E815 " Motor vehicle traffic accident to rider or passenger of motorcycle in collision with other motor vehicle " as well as under category N801 " Fracture of base of skull ". It would have been of considerable interest if we could have undertaken this study on the basis of both E and N lists, but unfortunately the figures appearing under the N list include not only deaths from accidents but also those from violence such as self-inflicted injury, homicide or injuries resulting from operations of war. For this reason, and because in many countries only the E code is used for classification, the material used in this paper relates only to deaths classified under the E classification. ACCIDENT MORTALITY AMONG CHILDREN 127

The abbreviated list of 50 categories provides for only two items under which deaths from accidents may be classified in the E code-namely, BE47 " Motor vehicle accidents " and BE48 " All other accidents ". The Intermediate A List provides for the following ten categories under E classification for deaths from accidents: AE138 Motor vehicle accidents AE139 Other transport accidents AE140 Accidental poisoning AE141 Accidental falls AE142 Accident caused by machinery AE143 Accident caused by fire and explosion of combustible material AE144 Accident caused by hot substance, corrosive liquid, steam, and radiation AE145 Accident caused by firearm AE146 Accidental drowning and submersion AE147 All other accidental causes Many countries tabulate their deaths from accidents in accordance with this intermediate list. The detailed list of course provides a much larger number (124) of categories for accidents, but unfortunately information has not been available under this list for many countries. Therefore it is the Intermediate List that is used in this paper. The definitions by means of which deaths may be classified in different categories are given in the Manual of the International Statistical Classifi- cation of Diseases, Injuries, and Causes of Death. 20 In order to ensure, as far as possible, uniformity and comparability of statistics, the Manual makes recommendations for the certification of the cause of death by the attending physician through the use of a uniform medical certificate of death and the selection of the underlying cause of death as the cause to be used for tabulation in mortality statistics. The underlying cause of death is defined in the Manual as (a) the disease or injury which initiated the train of morbid events leading directly to death, or (b) the circumstances of the accident or violence which produced the fatal injury. In studying deaths from accidents, the time that elapses between the occurrence of the accident and the death is important. The Manual provides that deaths occurring one year or more after the date of the accident be classified separately in the detailed classification under " Late effects of injury and poisoning" in categories E960 to E962. In the Intermediate List, nine types of accident are separately classified (AE138 to AE146). In this list, deaths from late effects of accidents are included in the miscellaneous category AE147. Thus categories AE138 to AE146 of the Inter- mediate List refer to deaths within 12 months of the accident, while the 128 S. SWAROOP, R. M. ALBRECHT & B. GRAB miscellaneous category AE147 is not restricted with respect to the period elapsing between the occurrence of the accident and the death. Reference may also be made here to comments, which have appeared in the WHO reports referred to earlier, on the inter-country comparability of data. For instance, it is stated in the report on mortality from accidents other than transport accidents 22 that, in France, "the title AE138, ' Motor vehicle accidents', covers only cases in which death takes place imme- diately or within 24 hours; deaths occurring after that time are entered under the residual title 'All other accidental causes ' (AE147). . .. Conse- quently, a certain number of deaths officially entered under titles for other accidents (AE140-AE147), and particularly under the residual rubric AE147 ... should be removed and transferred to ' Transport accidents' (AE138, AE139)". Differences in medical practice, in the availability of physicians, etc. may also affect comparability. Further, as has been men- tioned in the review on mortality from transport accidents,19 in France a complete picture of the extent of deaths due to transport accidents is not obtained from statistics based on medical certificates of death, because the certificates " do not always mention the external cause of the injury ", with the result that " these statistics of causes of death only throw light on a certain proportion of motor vehicle accidents, the others being necessarily classified with the accidents of an unspecified nature under BE48 ". Another factor affecting comparability is that some of the deaths ascribed to senility or ill-defined and unknown causes may really be due to accidents. It may also be pointed out that mortality statistics are based on the defacto population in some countries and on the normally resident population in others. Thus it is possible that the death in an accident of a traveller journeying in a foreign country may either be counted twice or not be counted at all. In view of these considerations it may be stated that, in spite of certain procedures recommended and adopted for ensuring comparability of accident statistics, the figures are not strictly comparable from one country to another or during any long period within a single country. Yet, despite these limitations of the statistical material, it is felt that some of the results emerging from the study stand out so boldly that the possible incompara- bility of the statistics may not have seriously affected them. As stated already, we have not made use of statistical information in respect of non-fatal injuries because such information is not generally available on a national scale. The nature of the accident problem is, of course, not identical in each country or within different parts of the same country, and varies with host, agent and environmental factors. The problem should therefore be studied separately by each area. In order to study the causation of accidents and the extent of the problem, and to determine the effectiveness of control programmes, information is required on non-fatal as well as on fatal accidents, since the death-rate from accidents according to external cause will be rather unstable because of the small ACCIDENT MORTALITY AMONG CHILDREN 129

numbers involved. Moreover, the death-rate is affected not only by the prevention of accidents, but also by the quality of the treatment given for the injuries that occur. The leading causes of accidents may differ, depending on whether trivial injuries are counted, as in some morbidity surveys, or only fatal or serious accidents. For example, one might expect that falls, a common occurrence in childhood, would be one of the leading causes of accidents to children, if trivial injuries are included, whereas they account for only a small proportion of deaths in this age-group. On the other hand, an important cause of death may not occupy a prominent position in the causes ofnon- fatal accidents. For example, Collins, Phillips & Oliver5 pointed out that drowning caused 8.4 % of all the accidental deaths in Baltimore, Md., during the years of the morbidity study in the Eastern Health District of that city, whereas drownings and near-drownings amounted to only 0.1 % of all the accidents and 0.2 % of the disabling accidents occurring during the period of the study. The record of a nearly-drowned person who is rescued may, in fact, never get into the accident statistics.

Importance of the Problem From the statistical point of view the problem ofaccidental deaths could be studied in at least two different ways: (1) estimation of the extent of loss of life; (2) assessment of the relative risks of different population groups exposed to various forms of hazard. A study of the second type would require knowledge of the population groups actually exposed to various accident risks-information that is ordinarily not available and is not easy to obtain without detailed epidemiological studies. The extent of loss of life from accidents and its relative importance in various age and sex groups have been studied by previous workers by means oftwo statistical indices-namely, the percentage ofaccidental deaths to total deaths (proportional mortality from accidents) and the specific accident death-rates per 100 000 population. Both these indices are shown for a number of countries in Table I. The figures relate to the average for the period 1951 to 1953-the latest period for which information is available. In view of the general importance of fatal motor-vehicle accidents, columns 4, 5 and 6 in Table I provide additional information on this problem. Before commenting on these figures, it should be pointed out that while proportional mortality does enable us to see the importance of accidents in relation to other causes of deaths, it does not provide a measure of the absolute importance of accidental deaths. The latter is measured by specific 130 S. SWAROOP, R. M. ALBRECH[r & B. GRAB

TABLE I. ANNUAL AVERAGE PERCENTAGE OF ACCIDENTAL DEATHS TO TOTAL DEATHS, AND ANNUAL AVERAGE ACCIDENT DEATH-RATE PER 100000 POPULATION IN CERTAIN COUNTRIES AND TERRITORIES, 1951-53

Percentage Death-rate Death-rate of fatal Percentage Death-rate from from all motor- Country or territory of accident from all motor- other vehicle deaths to accidents vehicle accidents accidents total deaths per 100000 accidents acients000 to all population per 1000 per 100000 accidental population population deaths

Australia (excluding full- blooded aborigines) 6.1 57.3 23.7 33.6 41.3 Austria 4.4 54.0 5.3 48.7 9.8 Belgium (1952-53) 3.4 40.6 10.6 29.9 26.3 Canada 6.7 58.4 20.2 38.2 34.6 Ceylon 2.7 32.0 3.0 29.0 9.3 Chile (1950-51) 4.2 66.6 1.4 65.2 2.1 Colombia 3.2 43.8 7.0 36.7 16.1 Denmark 4.7 41.9 10.7 31.3 25.4 Dominican Republic 1.7 16.0 5.7 10.3 35.7 Egypt (localities with health bureau) 2.6 55.6 10.2 45.4 18.4 Finland (1952-53) 4.9 47.2 8.3 38.9 17.6 France 4.3 54.7 10.3 44.4 18.8 Germany, Federal Republic (1952-53) 4.7 50.3 16.6 33.7 33.0 Hong Kong 2.4 21.1 4.4 16.7 21.0 Ireland 2.3 28.5 6.0 22.5 21.0 Israel (1951-52) 5.7 33.1 5.9 27.2 17.8 Italy 3.2 32.5 11.0 21.5 34.0 Japan 4.1 38.0 4.8 33.2 12.5 Mauritius 2.2 32.7 10.2 22.5 31.3 Mexico (1951-52) 3.1 49.6 2.6 47.0 5.2 Netherlands 4.8 36.1 9.9 26.3 27.3 New Zealand (excluding Maoris) 4.7 43.7 14.2 29.4 32.6 Norway 5.4 45.6 5.5 40.1 12.0 Peru (1951-52) 2.9 32.9 5.7 27.2 17.2 Singapore 3.9 42.9 23.2 19.7 54.0 Spain (1949-50) 2.7 30.4 1.2 29.2 3.9 Sweden 4.1 39.8 11.4 28.4 28.6 Switzerland 5.5 56.0 15.6 40.5 27.8 Union of South Africa (Europeans) 5.7 48.8 18.4 30.4 37.7 United Kingdom: England and Wales 2.8 33.2 9.8 23.4 29.6 Scotland 3.7 45.3 10.0 35.3 22.0 Northern Ireland 3.2 36.1 10.0 26.1 27.8 United States of America 6.4 61.4 24.1 37.3 39.3 Venezuela (1951-52) 3.8 41.6 9.9 31.7 23.9 Il J ACCIDENT MORTALITY AMONG CHILDREN 131 death-rates. Indeed, while the importance of accidents in childhood has been brought to light by a study of the proportional mortality, it should not be forgotten that the total number of deaths from accidents per 100 000 population is generally lowest in the age-group 10-14 years and is known gradually to increase with age (see Fig. 2), reaching its maximum during the last years of life. Other limitations of using proportional mortality figures for this study should also be recognized. Even though the proportional mortality may be relatively low for any particular population group as compared with others, it is possible that accidents may still be the leading cause of death in that group. For instance, as is commented on later in respect of Fig. 1, in the USA the proportional mortality figure for the age-group 30 to 35 is much lower than that for the younger ages, and yet accidents still constitute the leading cause of death. Again, in the next age-group, 35 to 40 years, the proportional mortality shows a further decrease yet accidents remain the leading cause of death in this group also. A mere study of proportional mortality figures, therefore, is not sufficient by itself to define properly the importance of the problem of accidents. Further, it is doubtful whether a study of proportional mortality figures is valid for inter-country comparisons, especially among countries at different stages of development, because even though the death-rate from accidents alone may be high, a still higher death-rate from other causes may considerably minimize the value of the proportional mortality figure. An interesting illustration of this is provided by Egypt which, as column 3 of Table I shows, has about as high an accident death-rate (55.6 per 100 000 population) as Canada (58.4 per 100 000 population). But because of the relatively high mortality from other causes, the proportional mortality from accidents in Egypt is only 2.6% as compared with 6.7 % for Canada. Chile is another interesting example, having the highest death- rate from accidents (66.6 per 100 000 population) of the 34 countries or territories shown in Table I, but a proportional mortality from accidents of only 4.2%. On the other hand, Table I provides several illustrations of high accident death-rates in developed countries-for example, Australia, Canada, Switzer- land, and the USA (the United Kingdom being a remarkable exception), in which the proportional mortality from accidents is also relatively high. To what extent the efforts of the Royal Society for the Prevention of Acci- dents, established as early as 1916, may have contributed to the difference in the figures for the United Kingdom is difficult to assess here. Therefore, while a study of proportional mortality figures may in special cases be helpful in focusing attention on the relative importance of this problem within each country, its value for inter-country comparison is clearly subject to severe limitations. In any case, proportional mortality figures should be studied in conjunction with specific death-rates for each population group. 132 S. SWAROOP, R. M. ALBRECHT & B. GRAB

c: " 04M0M )040 EN C-Cr- 11- M0 ')0D C- "0) M(0 M CM)c C6Y51-:C6c 040404 - c04-, --0404c~C6)04C6) _N0t-: C0 0 U')

I- 15) tD 1- OD N CO ( CO C)O o o LO (*c66o1-iC6O o O)(r0 1- 1- LO o0 rr-a-6OC5C600 0-t - 6O0r)LO)L0LO O5 z 0

C z C) ov '-o co no6r- 1iON-mnoc6 ao 1CJC5 NC)C or-cO, Or Of)N1- )1 0 - 1-t 0100000 O a;

u- oo CY q cq _- _i .q CY l - q Ci q I i ( Ulw or o u\'00Ov- ) '-(D 0--0 LO ) 00 a) LO CIO z ._

Ul C\a u, C)- 04 0 1 -)Ll04 C04 ) C4- - ,)a) 0>N C 1 CO1- - :) UDtX) _CO 0, LO00 le - a) NLO V CV) tt C-4 C-4 0 _C) CO CT Q C-4 C- CM Le) Li4 C)cr;r--6 j Lcoc c a;4 0)0)01c 0) U)

1o _- ur r- o4 C O u- _) _- Lo cn l a m _ NtF_¢o Ce lr N CO CM V) N Ql tn M X tN Cs sls w a) U) A c o r O - ac q0 Loca"tto_9L co coC? Ile"It CI ce-r Coo: c0?bobo.COcecOolo.t1ceo" t ot 00rlo CM ' ...... * >s v C) 1W -4 CO N - le C4 N, NV 04 ce v C CO NNN -c

0 O!a 4CYr: OR-'I-tDcot 1D0O"IcC-PX ORcm t CQr-. CYLO -D000CR cD ",tLO0C! !-t0X (0ON7 CO PI00 O U.IL -C

C:2 0 , 0 _04404 O _)04 _ Os-0U C 0C_4 04 N0 n () CY CY CY CY N ) CO cq _- cl CY _1- r- (D I'l LO _CNr etco O I.j0

0 C)\ Ac- NXN c- LO oN_ a) 0, U) . (6r- oic Nc'5 O 1 041 r'i6(_ 6c'iF o NO c I- CY -)or- CYNC- CO- CY)00 _ CY)t r\ _ O\J 0 1CY )0) COO) CCY) - q C-( c 0r- 0 cLf c\j c) ct -t -^'O < N 1- -ON C q CY)r- ) CY) CtO 0 C J CY1- ._ LU S-C c " C U) CO C =z CYO CY') CCY) c\0 C) 0404 uiJ U o U) wm a) . 0) zai aM r-Cr-r-C--O 0)CY) ) r- C N - ITlU) C- or- c It OLL) LI CO CD N~ 4 N~ 4 4 4J N~ W) 'i i 4 [) 4 6S U-3 Nl CY5 C5 6 mlA- a) M a) "0 0 *0 >1 co 0 U a) -0 0 c a) U) C C c -0 - C./)a)< :tfc a) -c 0 (/)0 0 co -IC c (t 0 00. a) a2) c 0 Co " - -c vo N U) co a) %- C5 o E a) a) CD 0 -t a), C > C Q 3 ¢ .H C 0 a) a) 0 W U1) z D aoa) B B C < <0 U 0 3Z ZZZ ) U) D) .D ACCIDENT MORTALITY AMONG CHILDREN 133

a 0 JR CO 04 0 ig -" I I- V$

z 0 0

z au

'U zJ u Aux w

z I-0 E l2 PI- (I)z I-- 'U I -a 0 zI- 'U (a U

I-

] I 0D 0t 134 S. SWAROOP, R. M. ALBRECHT & B. GRAB

FIG. 2. SPECIFIC DEATH-RATES PER 100000 POPULATION FOR ACCIDENTS AND FOR ALL CAUSES OF DEATH, BY AGE AND SEX, IN TWELVE COUNTRIES, 1951-53

Ret* R2000 20000 20000

10000 lO000 8000 6000 6000 4 000 4 000

2000 2000

1 000 1000 D00 0OO 600 600 400 400

200 200

100 100 80 80 60 60 40 40

20 20

10 10 6

20 000

8 000 6 000 4 000

2 000 2 000

1000 1000 800 8000 600 600 400 400

200 200

100 100 s0 s0 60 60 40 40

20 10 10 8 6 6

w4*

- 5 -10 - 15 - 20 -25 -30-35 -40 -45 -50 -55 -60 -65 -70 -75 (I - 5 -10 -I15-20 -25 -30 -35 -40 - 45 -50 - 55 -60 Agv-W..

Aisle All ca s of death Accidw do-ths _-__ F0l0 F-cle ACCIDENT MORTALITY AMONG CHILDREN 135

FIG. 2 SPECIFIC DEATH-RATES PER 100 000 POPULATION FOR ACCIDENTS AND FOR ALL CAUSES OF DEATH, BY AGE AND SEX, IN TWELVE COUNTRIES, 1951-53 (continued)

000 20 000 am FRANCE GERMANY 000 f FEDERAL REPUSLIC 000 . t(952-53) fj 800000

200 64WD| t

D00 _ 000_ . ^r *s 20w _ _ E

Soo ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~600

100 --\/ 200 ?00-0000 zuu~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~0 100~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~0 60 \___th S 60 40 40

D00 ITALY } 20 JAPAN J lOOD0~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~O~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~20 ROOQ_/_ 809D 10 ~ ~ ~~ ~ ~ ~ ~ ~ ~~~~~l *~-10 6 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~6 4 4

2 2

,I 5 10 15 -20 -25 -30 35 -40 455-50 -55 60 -65 -70 -75 5 -10 15 0 25 530-35540 -45 50 -55 -60 -65 -70 -75 100 --20002

00 A/ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~10000 00 6000 100 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~45000

00 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~20200

woo00

co 4~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~00

050

80 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~80 60

20 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~20

10 0 -

6 -..6 4 4

M.I. All o... d 4t.1 - - F. ___ F.-Itt.l.I Accid.,t dJehs 136 S. SWAROOP, R. M. ALBRECHT & B. GRAB

FIG. 2 SPECIFIC DEATH-RATES PER 100 000 POPULATION FOR ACCIDENTS AND FOR ALL CAUSES OF DEATH, BY AGE AND SEX, IN TWELVE COUNTRIES, 1951-53 (concluded)

Rat Rt.e 20000 20000 NETHERLANDS SWEDEN 10000 10000 8000 6000t 4000 4000 / / 4000

2000/ 2000 10Wlt a~~~~/ 1000 8000 600 / 600 400 / 400

200 500 011~~~~~~~~~~~~0 100 Xx '00 z 100 20 / ,~~ 60 40 8106 6620 200 40 4~~~~~~~~~ a10 z_'>,_s ~~~~~~~~~~~~~6 ___>sx 4~~~~~~~~~~~ 10

2 1 1 10 1 (1-5 -10 -15 -2012-~-.2 -10-3-404045-50Sl34 -55 -,I-65I65--0 -75 1-5 -10 -15-20 -25 -30 -35 -40 -45 -S0-S5-0 6-65 -70 -75 A w 20000 SWITZERLAND 20000 10000 *000 U SA X 8000 2000 6000 / 8000 Soo /7

400

Soo 200 400 100 100 m 60 20060 40

200

40 - ~~~~~~40

01 - -10'\k~~/-1S -20 -25 -30D-35 -40 - 45- SO -SS - 60 -65 -70 - n *1-5 -10 -15 -20 -25 -30 -355-40 -45 -- -55 -600-65 -70 -75 A

Awe MI Acusf t Accident deaths ___ Femle ___ Female ACCIDENT MORTALITY AMONG CHILDREN 137

With these reservations in mind, a few special features of the problem of accident mortality, as revealed by a study of proportional mortality figures, may be given. Table II shows the average figures for the percentage of accident deaths to total deaths in special age-groups for a number of countries over the period 1951-53. Those instances in which accidents constitute the leading cause of mortality have been indicated by printing the percentage figure in bold type. The special features revealed by an examination of these percentages are as follows: 1. In almost all cases the proportional mortality is the highest among the young, the peak generally being reached between 5 and 25 years of age, with considerable variations from one country to another. This point is further brought out by the frequency with which accidents appear as the leading cause of death in these age-groups. 2. Among infants under one year of age, accidents are not relatively as important a cause of death as other causes. This is partly because neonatal deaths predominate in this age-group. Thus, the child population in the age-group 1-19 years is relatively the most hard hit by accidents. The word " relatively " needs emphasis, because the chance of dying from accidents is certainly much greater at higher ages than in childhood. For a study of sex differences in the relative importance of accidents in different age-groups, Fig. 1, which is based on the proportional mortality percentages given in Annex 1 (see page 159), provides a series of diagrams for different countries and shows the proportional mortality in each age- group separately for males and females. As a rule, the proportional mortality among males is consistently much higher than that among females. Exceptions occur at the two extremes of life; thus the difference is negligible for infants and is again small in old age. It is also worthy of note that the peak is reached at different ages by the two sexes, that for females occurring earlier than that for males. An indication of the loss of life from accidents by age and sex is provided by Fig. 2, which shows the age and sex specific death-rates from accidents for twelve countries. For purposes of comparison, corresponding death- rates for all causes are also shown on this figure. The corresponding rates are set out in Annexes 2 and 3 (see pages 160 and 161). The specific death-rates have been charted on a logarithmic scale. A comparison of the age specific death-rates for all causes with those for accidents alone shows that, at the advanced ages of life, the male accident death-rate, even though showing some increase, does not increase in the same manner as does the total death-rate for all causes. The female death- rate for accidents in most cases increases in the same manner as the age specific death-rate for all causes. The maximum accident death-rate occurs in old age in both sexes. The most remarkable difference in the death-rate 138 S. SWAROOP, R. M. ALBRECHT & B. GRAB

for accidents is to be noticed in respect of sex. While the specific death- rates for all causes for the two sexes run almost parallel courses, the curves for accident specific death-rates for the two sexes are wide apart, except at the two extremes of life. The male accident death-rate is consistently much higher than the female accident death-rate. As a matter of fact, the curve for the male accident death-rate rises so high on the logarithmic scale that in many cases it crosses the female death-rate from all causes around 15-25 years, thus emphasizing the heavy loss of lives of male children and young people in this age-group. This phenomenon is seen in the case of Australia, Canada, Sweden, Switzerland and the USA. The diagram for Australia shows that the age specific death-rates from all causes run fairly parallel for the two sexes from the age of 30 onwards, but that there is a bulge in the male curve between 10 and 30 years which is generally explicable in terms of the high traffic-accident death-rates for males at these ages. In a discussion of the importance of the problem of accidents, the fact that it is children or adolescents who are relatively the most hard hit has considerable economic and social significance, as has been pointed out by Dickinson & Welker.7 These authors note that between such leading causes of death as heart disease, cancer, intracranial lesions of vascular origin and nephritis on the one hand and accidents on the other, there is a marked difference between the average ages of death-a difference of the order of about two decades. Accordingly they estimate the " working years lost " through death from different so-called " leading " causes of death. These working years lost are calculated on the basis of the life expectancies of survivors of the same ages as the decedents, but excluding the years before the twentieth and after the sixty-fifth birthdays. In the year 1945 in the USA, " when the seven leading causes of death were ranked by working years lost, accidents replaced heart diseases in the number one position. In 1945, a year of nationwide gas rationing, the working years lost from accidental deaths numbered 1,750,000 while those from heart diseases numbered only 1,680,000. Pneumonia ranked third with 1,100,000 years and cancer fourth with 1,040,000 years." In 1940 also, accidents were the chief destroyer of the working years of life; they ranked second in 1935 and third in 1930. In regard to British Columbia, Strong 18 says that in 1952 accidents were responsible for only 7.9% of all deaths, but accounted for 17.5% of "life years " lost from all causes of death. Heart disease, the leading cause of death, was responsible for 36.20% of all deaths but for only 12 % of life years lost. Cancer, the second leading cause of death, was responsible for 15.7 % of all deaths, but for only 10% of life years lost from all causes. The Registrar General for England and Wales 8 presents data concerning deaths from certain causes in 1954 in terms of mortality rates, years of " working life ' lost (taken as ages 15-64) and years of " total life " lost. His statistics show that the importance of accidents increases when based ACCIDENT MORTALITY AMONG CHILDREN 139 on years of total life lost, and still more when based on years of working life lost, because of the comparatively young mean age at death. For example, in males, accidents ranked after heart disease (including coronary disease), cancer, vascular lesions of the central nervous system, and bron- chitis and pneumonia, in terms of deaths per 10 000 population. Heart disease caused 32.0 % of deaths from all causes, and accidents 3.3 %. How- ever, in terms of years of total life lost per 10 000 population, the proportion caused by heart disease (the leading cause) was 23.6 % and that caused by accidents was 6.7%, although accidents were still in the fifth place. In terms of years of working life lost, accidents moved up next to cancer and heart disease, in that order. Cancer caused 14.6 % of the years of working life lost from all causes of death, and accidents 11.0%. The same tendency is shown in the figures for females, although the difference is small because the average age at death from accidents was higher in females. These observations suggest that the causes or factors affecting the two sexes at various ages should be studied separately. From this point of view each population group would require a special study. Further, as Table II has brought out the relatively heavy fatality among children, the present discussion is mostly confined to the consideration of accidents among children under 19 years of age. Data for higher age-groups are nevertheless provided for comparative purposes.

Relative Importance of Accidental Deaths in Relation to Deaths from Other Causes among Children 1-19 Years of Age The progressive reduction in deaths due to infectious and parasitic diseases, which at one time took a heavy toll of children and adolescents, has in recent years placed deaths from accidents in a particularly important, if not a leading role. In Table III the relative importance of different groups of causes of death, classified according to the 1948 Abbreviated List, is shown for twelve countries in terms of percentage of deaths from individual causes or groups of causes to total deaths, for the age-group 1-19 years. In order to facilitate comparisons and to focus attention especially on the importance of deaths from accidents, diagrams have been prepared in which the first four leading causes of death in each country are shown in order of magnitude, different causes of death being distinguished by different shading (see Fig. 3). In each case the leading cause of death is measured from the upper vertical radius towards the right-hand side and is followed by the cause of death next in importance. For instance, from the first three diagrams, relating to Australia, Canada and Ceylon, it can be seen that in only the first two countries do accidents occupy the leading place. In Ceylon the first position is taken by deaths from pneumonia (11.7%), accidents appearing third in order of importance (3.9%). Fig. 3 shows

10 140 S. SWAROOP, R. M. ALBRECHT & B. GRAB

< r-O c LO O ) c C Y)O 0)0CC (D aO D O r4CoO CoC00Y) , 8

CO >-a 000)CO0N4C0 CD CO 15) ) CDN 0 L(CC 0) COOCODNCO CO - 8 C/) V) cl cs oD r- I5 c ce Lo r-4 0 a 0 w

o CO r- C~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~C)0CIcs rQ - (OCo COYO c O O C) "I N 8

6c~~~~6oo~~~~- c~~~~ o56~:tco? cl ,,o r

(-z AJ~~~ *D * 0 Na,NF No 04 r- Co 1 -- z E _u0oNFa) 0 N - CD -Co )' 8 lU a 00 a 0 (N a CO aCo)N D C') CO C aO n I-EL NMO) NCY)C C N OO0 ) CY) O E-aD~Y Co- 0'- O C O r-- Co C O- O'- O CY) 8

r-o5c5i6'266 6 o5a~'2 ~ r- CU~ ~ ~ ~~~ ~~~~~~~~~~~~~~~~~C CO 0 d- a 05 CO 0) cO 60 CO 0 -

a, ' 6 6 ( ~CC'C '2c C r- N N8 E CO C CO C-C\ C) NCo D)- C O 0 U. z ': 0U)1-

a a, C') M m 0) (DO a .t ,mCD 0 cic'-0 It COO 0- ' C Y?- CO? C') Col r- 8 0: 0

U) CD- =~~~~- 'lt CO 0 C LC.) CY CDm 0 wAZ

A. - -''-----~~~~~~U c o a a)_ Q C ) _ Cn 0 0 Za ch a co 4- E n M 0 W6 a~~~C L C~~~~~~a' EE CDCC-C a- E5 2 0l Uc If, Co~~~~~~ aC.L) 0) e4 HQQc < ao Z D c0 U< 0) L- 0 0~~~~~~~~~~~~~~ Ul rn- CUn C-cw >3L D + CDa, CO-'):- m CD a,- CD a, '-'- '- C) CO ~~~~~~~~~COC CO -- W5 4-U a, m to to to to to to to tooE Ei. 3 0 ACCIDENT MORTALITY AMONG CHILDREN 141

FIG. 3. PERCENTAGE OF DEATHS FROM THE FOUR LEADING CAUSES AND FROM THE REMAINING GROUP OF CAUSES TO TOTAL DEATHS IN THE AGE-GROUP 1-19 YEARS IN TWELVE COUNTRIES, 1951-53

Ameooo T.bwad..i, *11 f... @3 bolwis E C .do -I..uNm ami "N1K deedgitHs. eutmiMs I

l Mor md umk. CmesOe AUSTRAUA CANADA CEYLON

Ni; 1'

GERMANY PFDERAL REPUBLIC ENGLAND AND WALES FRANCE (11912524)

jKM.ii. V17

ITALY JAPAN NETHERLANDS

'I'l ~~~~13.S~ ~~~~~~~~1 1

11.9 .0 M 7

SWEDEN SWMTZERLAND 14.4~~ s. i

&4 Ii

......

Wm Gin 142 S. SWAROOP, R. M. ALBRECHT & B. GRAB that in nine of the twelve countries, deaths from accidents occupy the first place, the percentage being highest for the USA, followed by Canada (36.7 %), Sweden (36.4%), Switzerland (34.4%), Australia (34.0%), Ger- many (33.7 %), the Netherlands (33.1 %), England and Wales (25.2 %) and France (19.1 %). In Italy and Japan accidents occupy the second place. While these diagrams serve to illustrate that deaths from accidents are now a relatively leading problem in the broad age-group 1-19 years, they are not adequate for properly assessing the absolute magnitude of the problem without reference to the specific death-rates for each cause. The information given in Fig. 3 should therefore be studied in conjunction with that given in Table IV, in which specific death-rates by cause are shown. From Table IV we see that even though in Ceylon accidents occupy the third place in importance, the specific death-rates are high, so much so that the Ceylon rate for females (26.1) is the highest rate for females in all the twelve countries. The rate for males (30.3) is also higher than that for England and Wales, Italy and France. This comparison once again illustrates the fact that for a proper understajiding of the importance of accidental deaths, proportional mortality and specific death-rates should be considered together.

Relative Importance of Various Types of Accidental Death in Age-group 1-19 Years In the foregoing sections we have examined the loss of life from accidents of all forms, considered as a single group. In order to show which different types of accident are responsible for the majority of deaths in the age-group 1-19 years, specific death-rates for various types of accident are presented by sex for certain countries in Table V. The following observations are of interest. (a) With the exception of Ceylon, in all the countries the male accident death-rate is at least twice that of the corresponding female rate. In such countries as Sweden, Germany and Australia, it is about three times the female accident death-rate. The highest male accident mortality is recorded for Canada (57.1) and the lowest for England and Wales (25.6). The highest female accident mortality is recorded for Ceylon (26.1) and the lowest for England and Wales (10.1). (b) In the majority of countries motor-vehicle accidents are responsible for the largest share of accidental deaths. This is particularly so in the case of Australia, Canada and the USA, each country having a specific mortality rate of over twenty for males. In England and Wales and in Germany, motor-vehicle deaths are also more important than other causes of accidental death. But in Japan and France deaths from accidental drowning and ACCIDENT MORTALITY AMONG CHILDREN 143

N- CO 0 CO UC D 05605 (- co U\D in CO N- CC) CS) CtCo N CO O C (A 0 CM 0 0)

0 05 CY? 0 Co CO) CN 0) ({

0 -o Co 0- N- cO N CO? N- o o 0 o - 0 - IL N-j N- N-

0 co co CO LO CO) N- CO CO I- C5 CO C O NOt Lo5D6r-ICO(Y) Co jCo)(6 c) .... co oo 0 0 cq 0~~N N- Co

co Co 'U N- COj CO CO COCo O u. u-0 r-~ u it CO r'- 00 (05 N- 0 COJ 05 COi

CO 0 CS5 05 CO (0 ao a) CO CO N- CO Co N- C) z o ar Co Co -u CD CO z ui 0 > q cq U: cS 0 CO

Coi 4. -o CO N- C( co C6 co cM CO? CO Co U) U 04z LU ~LL CD Co 0 co C) CI r- CO 0 CO CO 0 co OD (0-o co LO SW _O or- CO-Co a-0 0 CD Coi U) 0 cna) LI? UC CO CO? 05CO co CY MU) ro ca 0) CY) 'IT U? O0 CO co o5

"I . _ a a; 0Co 0 (0 u ~ %u (Eu (0 iI. U + E U0E 0 _U -o o E c 7 C 0 (0 U)0 (0U~(a0 0 (0 u. ,_. ~ C- 0E-0 0 L) < C) Co 0 (n (0 0 Co )r cn U) 0m00a-m?mm m E (0 I a) :. o (0 >s0 C)U co00-OC C Q (0 a _ : 6i -0 L] ~~ coo-Ec-lE ( .C- w -Z O < C 3 E tC E E us (0Co a a E AI D (0 =

az Q)0 :3 0) co~N- 0) o -E.- E CO- COj c) ) CO C\j ILU* 0 co C *L C~o C co C O) CO CI 144 S. SWAROOP, R. M. ALBRECHT & B. GRAB

_o )OC-0)0, t ',0D00 0) ~- 10 r- NI 0 1010)LO ) >I- 0L ~- 6o6c~i ui U) < a) C'5 a. C0 CI

LO wI- -C4 )0 0) NCtO C-L

0 C?t0N N 0#C_r" 0 UIL 0- 0 -0 0 oI6c'c'6I 0 LL 0c0000 0 '-ON0CO N 0

0 C 11. 0)'CO000-- C\ - CD 0' cNc C 0) c: 0)a I?V~ri ~ U 666 c5 cic0N6roi 6 NN0U) "OO(r-0 LO -U)0 C-~~~~~t10100~~~~~c\

(D C000) coCC) ,tt 0) cn 11 00 '-0 I U1 0 c1 0)') 0 'It -0co(0 0a) 0)O, CY) ~: 6666C UcNl 0 66 c-ic C' 0- o O~ Oq( 10 t 00) C) C') 10 0 N a) 0) N - 0010t Nq 2 0)0 U) cn L M 00 I- 0 0 4,, J '-0To0 0) 001010 -7 0 C)1.T0 C- C ) CD0) 00000 0) 00 0, D Li- .-C O O O 0 0 0z N 0 u a) 6 U 0 N10 C' 0)0)010''C')o C') 00.-CCD 6 a) O NCO o 4> C') z 2 _ Z 0)CO- t C-.- a)? 00 0" n N LL =-0.-0G C') OON.- C') 11 N - 000 10 0o c -U) U) O,0 cN _ z 00co1C-O1t 10 tNN(D O X I N )- LO N C')

N st0)N10N C- 0)OC')0 0 1010D.- 10) ON0)10 LL 1-0-00C OON'- 0) NOO-0 0 00- N 0A_. T(0 >1 0 -I~ N_ -F Ou C-0C-1N 10 OC'-C'0) L X 4f aD 0 co O~ ,,: J 0 0 J0 I. CD .C2 Ci U) .22 C. 4- 0 0C- L Z O (Z.c o0M. a x D 0 0 40 0.2 :3 E I a)x-o U)Do 0 M a) C a) 0) 0)_ , 0 U) @°°~~~~~~~CU)Un '-0 *-*-.0OE>na , a)u)Ba a) wo °) -o =0-00) U)0.0z 0cL Z 0o 0 c C C0 C co a) c a) 0o 0 --00-0-00c <4 0 to 0O 0 00Pa) a a) .2~j0 00-0 0 2W )a E)ao7a) a a E 'W c cn E 0E 0 a) 0 0 07: F- a) ' H a o oa a U) (n a) a) a) 0 Ch, CO M M a) H . = .)'0n _ CO -0 > HC 0 0 c7o*5 0- 0 ILU) CIO M :3 :3 !, CO. O- a) U cmJ 04- W W 0 0 0 a) "1.1.o--0-00uo1-00+. 4oU 0 0)U) 0 o u0.5 5= O Q0 0 O Q 0D 0 = 0 0)0 20o«« < <<< < 0 a,0 0 0CS)0a)Q) 0 ) a C a} jI M) C-._ 09 0 -c 20<44<0< < <<<< 4 <4<<

IaI *0-I wfl ZEu IDo-iL.. -zL -I >i±p it duSLII f I MI.~.L(ow .. Ut~~6 F L . A

A I.

EJILUL1L

0 z C ______I Cc LU~~~~~~~~~~~~~ LL.a~~~~~~~~~~~~~~ U0) t r Ii- (05 LAI a thu. x,f 0 U) LUki IC~~~~~~~~~~~~~~~~~~~~~~~~~

.0 EihLEtili

.9"I-~~~~~~~~F3t ACCIDENT MORTALITY AMONG CHILDREN 147

I~~~~~~I

Zc 3

az I

A M*

flU

(IW6

ml 1--~ ~ ~~~~~~~~~~~~~~~~11 0 MlEE

WL)~~~~~~~~~~~~~~~~~~~~~~~~M 148 S. SWAROOP, R. M. ALBRECHT & B. GRAB

------

IL~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 1 8W la~~~~~~~~~~~~~~~~~~~~~~~~

Z z ~ ~ ~~ ~ ~ ~ ~ ~~~~~~~~2 UuU 2s~~~~~~~~~~~~*LjIJLJLJ2I~~~~~~~~~~~~~~~~~~~~~~~~~~~~ua IE~~~~~~~~~~~~~

(IU) vZmam M__ [M

ILI z EEl

0ui LiELJO1

00~~~10

00.1- o ACCIDENT MORTALITY AMONG CHILDREN 149

TABLE VI. DEATHS FROM TRANSPORT ACCIDENTS AND ACCIDENTAL DROWNING AND SUBMERSION (AE138, AE139, AE146) AS PERCENTAGE OF TOTAL ACCIDENTAL DEATHS, BY SEX AND AGE, IN TWELVE COUNTRIES, 1951-53

Male Female Country age-group (years) age-group (years) 1-4 5-9 10-14 15-19 1-4 5-9 10-14 15-19

Australia 59.0 79.2 69.7 80.3 48.3 70.7 67.1 86.8 Canada 61.0 77.5 71.2 74.7 51.9 71.5 74.7 79.6 Ceylon 53.4 47.6 40.1 40.0 45.1 37.7 43.0 31.1 England and Wales 64.5 80.3 70.4 76.8 49.9 65.8 66.4 79.2 France 45.9 58.8 56.9 53.1 41.6 47.6 58.6 56.9 Germany, Fed. Rep.* 50.2 73.0 67.9 67.8 48.4 73.6 73.6 79.7 Italy 41.1 62.7 62.7 63.5 33.8 47.6 51.1 54.7 Japan 77.1 80.9 68.4 49.5 74.1 68.5 66.3 53.2 Netherlands 67.1 74.9 65.9 59.7 60.7 61.7 48.8 44.2 Sweden 73.0 90.1 75.9 82.6 64.6 86.6 84.1 91.7 Switzerland 57.6 69.1 56.9 63.5 48.6 70.4 58.5 69.8 USA 50.4 69.2 65.1 77.5 41.0 54.5 68.0 81.6

* For the years 1952-53 only Table VI shows that in the age-group 5-19 years the majority of accidental deaths are due to transport accidents and drowning. In fact, these two groups together sometimes constitute more than 80% of the total accidental deaths; in Sweden this proportion is 90.1 % for boys in the age- group 5-9 years and 91.7% for girls in the age-group 15-19 years. In any case, such deaths may on the average be regarded as being responsible for two-thirds of the total accidental deaths in the age-group 5-19 years. There are of course exceptions, an example being Ceylon, where, as we have already seen, deaths from fire and explosion and from drowning constitute the major causes. But if the problem of accidental deaths among children is viewed on a global basis, it would seem that transport accidents and drowning take priority. Within the transport-accident group, deaths arising from motor-vehicle accidents constitute the major part.

Trends While it is of interest to examine how the importance of deaths from various causes of accidents has varied with time, the study is complicated by the lack of comparability of the tabulated statistical material. Inasmuch 150 S. SWAROOP, R. M. ALBRECHT & B. GRAB as the International Lists for classifying causes of death have undergone changes, the figures recorded under " accidental deaths" are also affected by changes in the allocation of death to different groups. For instance, until the 1938 Revision of the International Statistical Classification of Causes of Death, adequate importance had not been given to the classification of deaths under transport accidents, even though it had been stipulated at the 1929 Revision that some of the accidental deaths should also be separately classified as follows: I. Accidents in mines and quarries II. Accidents caused by machinery III. Transport accidents: (1) railways and tramways (2) automobiles, motorcycles (3) other land transport (4) water transport (5) aerial transport However, only a few countries used this special tabulation. At the 1938 Revision of the International Lists, it was stipulated that among others, accidental deaths due to railways, automobiles, other road transport, water transport and air transport should primarily belong to categories specified for transport-accident deaths, whatever the nature of injury. Comparable categories were preserved in the 1948 Revision of the International Statistical Classification. In a study of trends, therefore, even though figures for the periods beginning from 1900 could possibly be quoted, it would be unsafe to relate them to more recent data. For this reason, in Table VII figures are given at decennial intervals only for three periods, beginning about 1931. For the countries for which figures have been quoted it is believed that the figures around 1931-based as they are on the 1929 Revision of the Inter- national Classification-are fairly comparable. It has been possible to substantiate this because, at the time of the 1938 Revision, many countries gave figures based on the 1938 classification for previous years also. Table VII gives, for eight countries, a series of indices for studying the various trends in operation since about 1931. A note of caution is necessary when considering the rates for the period around the year 1940, when the accident death-rate was indirectly affected by war conditions in some countries. The relatively high rates of 37.6 for the period 1941-42 in England and Wales, and of 33.4 for the period 1942-43 in Italy are examples. A similar remark applies also to the death-rate from motor-vehicle accidents in most of the European countries. In the Nether- lands, Sweden and Switzerland, the corresponding rate was lower around ACCIDENT MORTALITY AMONG CHILDREN 151

E cm 0'-

soa)~~~~~~~~rLO 00)L)L 0s 0C)C'occ U-L)o0 1- "1-Wt0 -(Or 'lOC C CO 000) c u 0 '-006oc'-5 co r-~ oCo\4io u r-:-c-6u6o~o o%5b0o6 -i Ou sC)O cIcucCuC\J - c \ N - - - )tCIJ-- M M M CQ --- N U0 E~ 00u) -m w 0CU) U)

Om (c0) c o'--- (r000 cOO r-) 0')m(0o0 10(0) (00>- _

UJ w I > ) - z)U-0 z o t) tO C) 1t ul0 N-4 LO Iq r- stCul ul- 0 - (D C C) It r-st r

1u) 00 ) . E z C) U D -r--(0o \CO - C\J ) oJ- CC\ '- C$0( 000$ COU) c

0 00 a ZZ, c- 0o8 Eui (10C O NCC C)-( D "T ) 0 - >--(N CC')0) U) - o a U) n 3j _ O WU):s0 c8 cooo- co-r-Caco 00D) (l)O1 0 00>0 c0D(0 UU) ~00 0 Ul 0 LU E c'ic~ q VO53 5-l~r-~-l c cool-. orcuI'l: -rio5 OC?7 6c'i(q C O O lD (O tr I0 N4 l£J\"co(D 0\C\I 0\ c\)cl CI Ot) > EU)8(D Q C >,Z, c

0-C co CO(0I( 1C) 0r- I-1m1 0 0(Co)N 1- 0 )N\ - 00)

-0U)0 -0) co C\ 0 C J 0) .0) C\0F t 0) 0) LOJ Ct N rt 0) O q 1) coD Cf)t 0)C00 OD1-- 0 (D LO O) r-Nco 1 coa)S ) (O co) wi- 000 () C\J N C\l C\j U) 'N C CQ N C\j CI) 0 0 I) 0

L U)8 0

\NN C') C\J(NC) C') NC')C' (coN\ NN ---CO 0C\-'- 0'- .0 co) 1 Lo (n 11t LO co) 'l IS) co) T U-) CY) q U-) c st Ur) L co "t U')I6co "T U-) *Z T; tU ) CO0Z UC:;O J-U) co-T U) co ~~:;S co 1L .1- U- !O) O) O) o) o-5)C O-) 0) (7 o) O> 0) CO 0) 0) m C) cDn00 000cn s

a) ZI-

c) -0

0 v) u a) U) U7) -0 _c_- U) a) 0) a) < u LU v) 152 S. SWAROOP, R. M. ALBRECHT & B. GRAB 1941 than around 1931. Nevertheless, if due allowance is made for these exceptional circumstances, conclusions based on the general trend between the first and the third period remain valid. In order to facilitate an understanding of Table VII, some of the indices have been usedin the preparation ofFig. 5. These show that so far as all forms of accident are concerned, the specific death-rate in the age-group 1-19 years has remained practically unchanged during the last 20 years in Aus-

FIG. 5. TREND IN ACCIDENT DEATH-RATE AND IN PROPORTIONAL MORTALITY FROM ACCIDENTS IN CERTAIN COUNTRIES SINCE ABOUT 1931 IN THE AGE-GROUP 1-19 YEARS ACCIDENT MORTALITY AMONG CHILDREN 153 tralia, Canada and Sweden. However, a decreasing trend is observed in other countries. The decrease is rather small in Switzerland (about 15%), but an appreciable reduction (between 25 % and 30 %) appears to have been recorded in Italy, the Netherlands and the USA, and a decrease of over 35 % has been observed in England and Wales. In any case it can be concluded from Fig. 5 that the total accidental deaths have not increased in absolute importance in the last 20 years in this age-group. If, however, the relative importance of accidental deaths as a percentage ofall deaths is studied, a corresponding indication of increase is noticeable-an increase which can be ascribed largely to the fact that the death-rates from other causes in this age-group have been brought down at a faster rate. It has already been pointed out that, in Australia, Canada and Sweden, the fall in the death-rates from all causes has not been accompanied by a fall in the death-rate from accidents. Column 3 of Table VII gives the specific death- rate (all causes) for the same periods. In about 20 years this rate has registered a decrease of 47 % in Australia and of 76 % in England and Wales. These two extreme percentages determine the range ofthe reduction observed in the other countries. Within the group of deaths due to accidents, if we examine only deaths from motor-vehicle accidents, a clear trend towards an increase is noticeable, especially in Australia, Canada and Sweden. The corresponding rates in the USA have remained at about the same level or have decreased slightly. It is significant to note that the rate for motor- vehicle accidents in England and Wales declined by almost half during the period 1941-42 to 1950-53. If allowance is made for the variations noticed in the case of motor-vehicle accident death-rates, it can be seen that the death-rate from the remaining causes of accidents shows a decrease for all countries in Table VII. This is reflected by the increasing trend in the proportion of motor-vehicle accident deaths to total accident deaths, as shown in column 8 ofTable VII. The proportional mortality is approximately unchanged in England and Wales, because the decline in motor-vehicle and in other accidental deaths has been of about the same magnitude. Fig. 5 shows that in most countries the decline in the death-rate from all accidents is due to the fact that the increasing death-rate from motor- vehicle accidents is more than compensated for by the decline in the death- rate from other causes of accidents. The exceptions are Australia, Canada and Sweden, in which the death-rate from all accidents remains fairly stable because the rise in the motor-vehicle accident death-rate counter- balances the decline in the death-rate from other accidents ; and England and Wales, and perhaps, the USA, where the decline in the death-rate from all accidents is due to a fall in both the motor-vehicle accident death-rate and the death-rate from other accidents. The increase in motor-vehicle accident deaths should be studied not only per 100 000 population but also in relation to the number of vehicles on the road, the number of miles driven, etc. There is evidence in the 154 S. SWAROOP, R. M. ALBRECHT & B. GRAB

literature to show that if the motor-vehicle accident death-rate is adjusted for the increase in the number of motor vehicles or in mileage, it is likely that, even in those countries where an increase in the absolute rate is recorded, a decrease is actually being observed. In this connexion, reference may be made to a report published in 1955 19 which shows that, in the USA, the death-rate from motor-vehicle accidents is, in fact, on the decrease. Similarly, Berfenstam & Vahlquist 3 have shown that in Sweden the increase in motor-vehicle accidents is occurring at a slower rate than the increase in the number of vehicles themselves. The same conclusion has been arrived at by Harris 12 in respect of Canada. The following figures, quoted from Harris, are of interest in illustrating that while the deaths per 100 000 population have increased, the rate calculated per 100 000 miles driven has actually decreased.

Year Deaths per 100 000 Deaths per 100 000 Deaths per 000 population vehicles miles driven100 1934 10.5 9.8 14.9 1939 14.1 11.0 15.4 1944 11.5 9.2 15.4 1949 16.8 9.8 11.8 1953 21.6 9.4 10.1 As has already been stated, drowning is generally the type of accident second in importance for the age-group 1-19 years. The trend of the corresponding specific rate per 100 000 population is shown in column 6 of Table VII for three periods at decennial intervals. It may be observed that for all countries the rate has decreased during the last 20 years. In column 9 of the same table, deaths from accidental drowning are expressed as a percentage of all accidental deaths, showing the relative importance of this type of accidental death twenty years ago and today. It is of interest to note that around 1931 the death-rate for accidental drowning was in many countries higher than the death-rate for motor-vehicle accidents in the age-group 1-19 years, being about 10 per 100 000 population in Canada, the Netherlands and Sweden. On the other hand, in countries like Australia, England and Wales, and the USA, motor-vehicle accidents were even then the major cause of accidental deaths. Among the eight countries for which data are given in Table VII, Canada, the Netherlands and Sweden still have the highest death-rate for accidental drowning. The lowest rates are recorded for England and Wales (3.0) and Italy (4.0).

Summary and Conclusions The main features of the mortality from accidents in the age-group 1-19 years have been elucidated by an examination of the statistical information available to WHO for certain countries. Attention is drawn to the ACCIDENT MORTALITY AMONG CHILDREN 155 fact that such a study, based as it is merely on data relating to mortality, is in many ways inadequate for assessing properly the magnitude of the public health problem involved. An elucidation of the epidemiological and economic aspects would require inter alia morbidity data, which unfortunately are not at present recorded on a national scale and consequently are not available to WHO. The mortality statistics for certain countries have been examined, with particular attention to the sex and age of the population, in two ways: (a) by the proportional mortality from accidents (percentage of accidental deaths to total deaths) and (b) by the specific death-rates per 100 000 population. It has been emphasized that a study of proportional mortality figures alone does not give a correct view of the magnitude of the problem. The proportional mortality figures indicate that accidents have on the whole assumed the leading role as a cause of death in children between 1 and 19 years, but that this is not so among children under 1 year or among old persons. The specific death-rates, on the other hand, show that accidents take their highest toll of life in old age, and a not inconsiderable toll in infancy. However, it is the predominance of other causes of death at the extremes of life which overshadows the role of accidents in these two groups. It has also been pointed out that such causes as heart disease and cancer, which account for more deaths at all ages than accidents do, may nevertheless be less important when judged in terms of the years of working life lost or the years oftotal life lost, because on the average an individual who is the victim of a fatal accident is younger that the patient who dies of heart disease or cancer. An examination of the main causes of accidents reveals that they differ somewhat from country to country and more by age and sex, thus emphasizing the need for an epidemiological consideration of the accident problem. In general, however, the two most important causes, in the age-group 1-19 years, in the countries studied are motor-vehicle accidents and drowning. It should, however, be pointed out that the case-fatality rate varies considerably from one type of accident to another and that morbidity surveys have shown that other types of accidental injuries may also be frequent in this age-group. Apart from the question of frequency, there is also the question of the severity of the injury and the duration of disability (permanent or temporary), which can be answered only by morbidity statistics. An examination of the trends in the death-rate from all causes, from accidents in general, and from motor-vehicle accidents and drowning in particular, in the age-group 1-19 years, shows that the death-rate from all causes has fallen greatly since 1931 in the countries studied, and that in most of them the death-rate from accidents has also decreased, although ur some there is no appreciable change. In any case, the fall in the death- fate from causes other than accidents (most particularly in the death-rate irom infectious diseases) has been so great that accidents are now respon-

I1 156 S. SWAROOP, R. M. ALBRECHT & B. GRAB sible for a relatively higher percentage of deaths in this age-group. For example, in England and Wales, the average specific death-rate from infectious and parasitic diseases for the age-group 1-19 years was 127.3 per 100 000 population in the period 1931-32. The corresponding rate for the period 1951-53 is only 12.9, a decrease of 89.9 %. During the same interval of about 20 years the accident death-rate in this age-group decreased by only 36.3 %, so that in 1951-53 accidental deaths accounted for 25.2% of all deaths at these ages in contrast to 9.4% in the period 1931-32. The trend in the accident death-rate in the age-group 1-19 years has been affected mainly by a decline in the death-rate from accidents other than those due to motor vehicles and by a rise, in most of the countries studied, in the death-rate from motor-vehicle accidents. In some countries the decline in the death-rate from accidents not caused by motor vehicles has been greater than the rise in those due to motor vehicles, so that the death-rate from all accidents has fallen. In Canada, Australia and Sweden, the decline and rise have been of approximately equal magnitude, so that the death-rate from all accidents has changed little. In England and Wales, and to a lesser extent in the USA, the death-rates from motor-vehicle accidents have fallen, as well as the death-rates from other types of accident. In general, therefore, accidents are becoming relatively an important cause of death among children primarily because of the conquest of other causes of death, but partly also because the rising tide of motor-vehicle accidents in most of the countries studied is nullifying to a greater or lesser extent the gains obtained by the control of other forms of accidental death. Although the number of motor vehicles and road mileage have been increasing, the motor-vehicle accident death-rate for the age-group 1-19 years has fortunately not increased in the same proportion. Still it is obvious that the trend of the specific death-rate from accidents in this age-group will continue to be affected by the trend of motor-vehicle accident mortality, which is related, more or less, to the increase in motor transport. The over-all conclusion to be drawn from the available statistical evidence is that, although as a result of the speedier reduction in the death- rate from other causes, the relative importance of accidental deaths in the age-group 1-19 years has increased, there is no justification for taking an alarmist view. In absolute magnitude, accidental deaths are on the whole decreasing, but this decrease is not occurring as fast as it is in the case of many other causes of death. A complacent attitude is therefore also not justifiable. The motor-vehicle problem is clearly not under control. There is also need to determine effective and practicable means of accelerating the decrease in other forms of accidental death. It is hoped that the above analysis, although limited to the mortality aspect of the problem, will to some extent help to clarify the situation and be ofuse in the efficient planning of appropriate preventive measures in childhood. ACCIDENT MORTALITY AMONG CHILDREN 157

Rl-SUMt La presente etude porte sur l'analyse de quelques statistiques nationales officielles relatives aux d6ces accidentels selon le sexe et l'age, classes d'apres les types d'accidents proposes par le Manuel de classement statistique international des maladies, trauma- tismes et causes de deces adopte par 1'OMS en 1948. Le calcul du pourcentage des deces dus aux accidents par rapport au total des deces montre qu'une faible proportion (generalement moins de 4 %) des deces survenant avant l'age d'un an sont la consequence d'accidents. Cette proportion augmente ensuite tres rapidement avec l'age. Ainsi, pour la periode 1951-53, elle se montait chez les garqons du groupe d'age 15-19 ans 'a 63,3% en Australie, a 60,7% aux Etats-Unis, a 46,3% en Suisse et 'a 34,2% en Italie. L'augmentation est moins prononcee chez les filles, et le maximum est gen6ralement observe dans le groupe d'age 5-9 ans. Pour la periode 1951-53, 33,7 % des deces survenant au Canada chez les filles de ce groupe d'age etaient dus aux accidents; cette proportion atteignait 31,0% aux Etats-Unis, 28,3% en Suisse, 17,7% en France et 12,0% en Italie. Un classement des deces par ordre d'importance de leurs causes revele que dans de nombreux pays les accidents viennent au premier rang pour l'ensemble des garrons et filles de 1 a 20 ans. Viennent ensuite, le plus souvent, les tumeurs malignes, la pneumonie, la tuberculose et les malformations congenitales. Les accidents 'a eux seuls sont fr6quem- ment responsables d'un plus grand nombre de deces que ces quatre autres causes reunies. L'examen de la mortalite par accidents a l'aide des taux sp&ifiques montre que, dans le groupe d'age 1-19 ans, le risque de mort accidentefle est de 2 a 3 fois plus eleve chez les gargons que chez les filles. Toujours pour la periode 1951-53, les taux par 100 000 etaient en Australie de 49,4 pour les garrons et de 18,0 pour les filles; ils atteignaient respectivement 44,6 et 22,7 au Japon, 28,8 et 12,7 en France. Des taux remarquablement faibles etaient enregistres en Angleterre et Pays de Galles (25,6 chez les gargons et 10,1 chez les filles). Les accidents de transport et les noyades sont les causes les plus importantes de d6c&s accidentels entre 1 an et 20 ans. Dans de nombreux pays, les deux tiers au moins des d6c6s accidentels sont dus La ces deux types d'accidents. En Suede par exemple, de 1951 a 1953, cette proportion a meme atteint 90,1 % chez les garrons ages de 5 a 10 ans et 91,7% chez les filles de 15 'a 20 ans. I1 est a remarquer que la majorite des accidents de transport sont dus aux vehicules La moteur circulant sur route. Parmi les autres causes de d&ks accidentels les plus frequentes, on peut citer les chutes et les accidents causes par des armes a feu chez les gargons, et les accidents causes par une substance brulante ou par le feu chez les filles. De 1930 a nos jours, et pour le groupe d'age 1-19 ans, on observe dans la plupart des pays une nette hausse des taux specifiques de mortalite par accidents dus aux vehicules a moteur. Par contre, les autres types d'accidents sont en baisse tres sensible durant la meme periode. L'6volution du taux de mortalite relatif La l'ensemble des accidents est determinee par la resultante de ces tendances opposees. En fait, la mortalit6 par accidents n'a que peu varie chez les enfants et les adolescents au cours des 20 dernieres ann6es. On a enregistre une tres legere hausse dans quelques pays et une reduction plus ou moins prononc&e dans d'autres. Par contre, la mortalite due a l'ensemble des autres causes non accidentelles (en particulier aux maladies infectieuses) a considerablement baisse chez les enfants. Ceci explique l'augmentation proportionnelle des deces par accidents et leur importance relative actuelle. Entre la periode 1930-31 et la periode 1951-53, le taux de mortalite g6nerale par 100 000 personnes du groupe d'age 1-19 ans a passe en Suede de 241,8 a 71,9, alors que, dans le meme laps de temps, le taux de mortalite sp&i- fique par accidents passait de 27,4 a26,2. Ainsi, 11,3% seulement des deces de ce groupe d'age etaient causes par des accidents en 1930-31, tandis qu'en 1951-53 cette proportion atteignait 36,4%. 158 S. SWAROOP, R. M. ALBRECHT & B. GRAB

En r6sume, les accidents constituent chez les enfants une cause de deces dont l'importance relative va en augmentant parce que la mortalite due aux autres causes est en r6gres- sion continuelle et parce que l'augmentation du nombre des accidents provoques par les v6hicules A moteur reduit presque A neant les gains realises sur la mortalite relative aux autres accidents.

REFERENCES 1. Armstrong, D. B. (1949) Publ. Hlth Rep. (Wash.), 64, 355 2. Berfenstam, R. (1954) Courrier, 4, 419 3. Berfenstam, R. & Vahlquist, B. (1955) Z. Kinderheilk. 76, 489 4. Charles, J. (1954) Practitioner, 172, 613 5. Coffins, S. D., Phillips, F. R. & Oliver, D. S. (1953) Publ. Hlth Monogram, 14, 1 6. Cruickshank, R. (1952) Proc. roy. Soc. Med. 45, 621 7. Dickinson, F. G. & Welker, E. L. (1948) What is the leading cause ofdeath ? Chicago (Bureau of Medical and Economic Research, American Medical Association, Bulletin No. 64) 8. England and Wales, General Register Office (1955) The Registrar General's quarterly return for England and Wales: births, deaths and marriages, infectious diseases, weather, population estimates, No. 426, 2nd quarter 1955, London 9. Godfrey, E. S. (1937) Amer. J. publ. Hlth, 27, 152 10. Gordon, J. E. (1949) Amer. J. med. Sci. 217, 325 11. Gordon, J. E. (1949) Amer. J. publ. Hlth, 39, 504 12. Harris, F. F. (1955) Canad. Serv. med. J. 11, 439 13. International Union for Child Welfare (1955) Int. Child Welf. Rev. 9, 2, 78 14. League of Nations Organization for Communications and Transit (1937) Report on the first and second sessions of the Committee for the Unification of Statistics Relating to Road Traffic Accidents, Geneva (Document LON C.276.M.179.1937. VIII) 15. Logan, W. P. & Gissane, W. (1953) Proc. roy. Soc. Med. 46, 499 16. McFarland, R. A. (1955) Amer. J. publ. Hlth, 45, 1302 17. McFarland, R. A. (1955) Milit. Med. 116, 426 18. Strong, G. F. (1955) J. Amer. med. Ass. 158, 905 19. United States, National Safety Council, Inc. (1955) Accident facts, Chicago 20. World Health Organization (1948) Manual of the International Statistical Classifi- cation of Diseases, Injuries, and Causes of Death, Geneva, vol. 1 21. World Health Organization (1953) Epid. vital Statist. Rep. 6, 257 22. World Health Organization (1956) Epidem. vital Statist. Rep. 9, 1 23. World Health Organization, First International Conference of National Committees on Vital and Health Statistics (1954) WId Hlth Org. techn. Rep. Ser. 85 ACCIDENT MORTALITY AMONG CHILDREN 159

-C)- Ca) (0(D 0 C\l (C LOC0 OD- C (C C\ 0 01- 04'- O(o 00O 04 r>C.-00 0-C0 u- COLC) OC)C(00C C O) (i ((C) C-(C0C) 0)C

0) 60 C\)'--O c-F04C\ 04'C\J - '-''- 04'-C\0 04-nC\- '-00 - N-0-0O C\J04L)'- 0 -- 0)') - CCC\ C\J

(0 Cy? 6 o>ci6 r6oi c6 - c'i i-qcqj,) CCi

CY) *I (0( oooCO0 or~ 00,-'-N (00Co((04coCO 0)c)r( 04(0(C0 (cnO"t r4~C) OCO'-'- 0-(00 .-cOc0C't

-) CO (00 - 0Itc 0 (C ) N0 (0( N0(0( C'4 ) L'- OON\N U)-0 C (0(N(D CY) 0) CO 0 (NO

Ez 1CYV0 U() C (00 00- 0 O) Ot 0 (0(C) 'I 0) ) N C( Co- rD N rA 0 CtS)- C - "4 -4 - () N- 0't 0 I N0 0t C -o 04C- 0)c c' (0 '4(0 0-t - (004 '-(0 'V OVL Ev en) 0

6S) N- 6O2)-- O C (D O. CO)(0 - CO) 0N "It C-OV 0 - t 00 Lo C\) CO 4 04 CO 04 1- C\ CO-

0 (0) N 1-\ '-) < tC\l ) CO l 0- 't 4( 0\O V'- 0D C 6-0 Or-- 0-04 004 N-04 0-0 0 N0 -0 004Ci-) 04V c-4 CY) CY) CY) ° ° \ 1 \ C) 1-\t 1 - OtOC\i N \ co ) 1-0NI x N\i LC) - (0 CY) C0 ) l;T 0 ) ) C\) C-) - N ) (0 - E) t- C\ - - 4 10- N0 N NO 00 - V (0 (004 (0(O0'-04) - CVt0-(0- CO 0 C\C (' V0- NC C r o (64cic6 6 o. 6.i c-56 4o5~ 'er- 6 r--i A-:P- 6 66 (66 CO 0- 040) 'tO- 00 C CCY) ( 0 CO (00) (00C\ N 00 ~o 6 c' o)rf( LOTi O- 0)0 10(0 C O 04O- 6 S- 04)0N\J l6- CTO- CO-(66C\JL) C0) N LD 0 \ UL-L)O "0IN (0040

Lq t) 1": CN()~ Cy 0q "4 cr? 1D N tC14 '14 cr?) qCy (D U) ()0 0 00 r-) ) 0) C) 'IT N\j LC)0-0-tN\j - CO0040 CY)C()"0t0 N (0'' - 0)\-t(00C" N04It CO04CO COO :z LOCO CVL -V 0)0- 0)04 11-01(0) 0)O0 0)C (C0 0 COO) CV) 1-'-0 00'Co) '-- O0C - 0LO 0 0 040 O40CO L)o 1LO 00()O CY') C\N "t0CY) t N Nt0 It L C\j ()NtC\JY

0~ Ctr -: 0 ca l V)O) C'D N~ a (D O? " Cy? O) 't C) __,040\N 0 CY)C\l 04- N N CO04CY) 04 N '\ CY0)04N\ CY)c\C) CY)04CY) N\ N04\ 0 \

(0 0404 00 0) N 0N)04 CO tOC 0'- 04ON- 0N)- It N015> --C 040 OJCtOO 60 COCOO '-' N-' 00 '-0 004 - 04 COCO

0 7"2 Cu cu v- o (0 C LO - -U) 2 c ELL ELL 2E 2111-L -21 c c- LL C) -~~~~~~~ .2(0 ~~~~~~~~~~ E~~~~> ~~~~ coN-~~~1 0 C>.: - U) C) (L co( 0) C>CD3. co a) < C C) j u- (D --, Z ) C D o 160 S. SWAROOP R. M. ALBRECHT & B. GRAB

E N "Itr- - 0 O'0 09 r-O '- 01 C- 00 10 COO 0 (00 co > r-:o6 o5,i 6g6 6 ~ 6r-. ic i t6,- 4ir-. oic6 (DC-. gc6 > (0 C9 L (D0 a ) M -' C-C')rC"C M0MCC 't f)0 Mc)0 0 (0C ICY) C\ C\CO0CN)CO

o 10CO)r- CC- .O (OCO '0-10 COO (D- 00)3CD0 -CO

C) CoO 'tr- ; (D O C- CO- LO)0) 0)l-0 CO ~ o () 00 r c () c)o "t It c C 'It CY)r0)1010'CO '- C- 0 -CO CY)) CO 1O '-q oco LCCo 0cqa 001 co cc0 C- (D0t LO

0) (D1-: COCO(c N "r-:S 0)09ON r0(DC-"LNO C,-m COm ~aCD010)tO 00)OO? M (0)CO3t t-(OcqCO ? 100 CO0CO '(0)C> 0 '- - 0) COO 0 ¢ 0~C'It"(OD C-TOCO(C C\)OCO LC )01 0C-C- C-o C C-) COO) 00 OCO- C- 0C)1-0 C.- (0.-

? C C C r-- 6 ~oi r. f 6.i .6o5 .6 . . . . 6. .'i. cn 0 CO- C.- .'- - It - 'I LO CO')- C---CO- 0)t ) .c 1£)0Dct (c? q Ut 0) CtLI? r- (J c oq rl a) OD- LI CY) C) 07 (c 0)_ COO) 0) CO)O C-Co Cr-CO CON0 'CO .--CO LO-N )CO C-Co Co ° t O C>l ~~~~~1~1' CO (I )r- C- LaO oCt) F ) t)LO (D - CD COCOC- LCCor 10CO O- 0).CO) 0)01 C--01 COCOO 'CIOCO.- CO 0 C- CO-(DOIO C O

0 oc-i 6cj 6 OCO 'C;O 6 LOC- C-0) C-- CY) ,C C-LO -) IC O) CO--CY) C--)1O 0 1 COD C)lC- CO N C-qOCO

u)4 CO CO CD CO OCO 1 -O C CO COCtOO ) o en CO COC - C\oC C- urC 10 CO '- OO CO C- -4 R as CO 0-u10 C-1 r-C- COY)c CY)Ot CY lO 00 CO't C- C- C CO)10J 6 t g5v1)oc N8\ (DC- rD- -C- Co oi CO g. 0 r- - CO N s tO 10O cO 1tr 8SO t~~~~~~~~~~~~r- It N L)O u-t) 1£ M r- Nq tS() r- 1SO O? 0ttt) cO c6 1- c5 i u 1- u (6 o6 t6 o Cq a) o stu- cor-( F 0 10 1 CO OCON O CO CI C- O O )'- 10CCrOCO) QQ X c 1 COOt 0 M - 0 -It (010D0) 10CO0-c)1NC 10C-) 10C-CO C 0 0)01-OCO 0)- x 3 ¢ _ C) 1C- 01 -CC)0 1 C0 O C-0 LC- O O COCO CO '- NCCO m) CO 0'10 1)0 COO e489 CO ~'- C').-- - - (D Co c- C') C') DC) CO'- v 0 OCO 00 6 OCO CO1 C-CO C- CO-- (6-- 6 - 6'-10 c-z6c6 1g6c ) 10' 0 10a( CO) ) -0 C-C') CO 0(0 CO10 10CO 10'- v c' 0 C C C ' C' ' 0 CO)1'- 610 CO6 )-COO)o COO) -CO r616 6 166 61:6 6 ci6 c616 r-6

4 aicoo c6c6 6coi 66COO.) COO) C-C- COc"- "It COCOr )10 Co. - U' W IT- C- -COC-C COO LO~ OCOM L(01 C(OC- 0 (0'- l-CLO 11L- NL 00- NtO NL LL0 1-0 LL- NO CO0 00- Cd - COO N( O 0COCOC"616ll) (COCOC IOOCOCO) C'- CC-10 10C-) O COO0 CLO0)1) ((0(CO

. ,'-: COy COC ic6r-: oi i 166 6 cy 166 .5--. 1616 7E D N st a 00NC) a U)tr )O D x C) U N~ N 0 O) tO E CO 0 r- t )Os [)C >r Le) D )C)tO )Q(i) C))C >.S01)DC -o~~~~~~~~~~~~~~a~~~~~~~~~~~~ ) () D * sr O0 co ( O 0N Q D 0 o-0 0a) 0 0 Nc .0 0 c NCC CO c N-0L- >1 N a)0 0 0 a)I£)ON1£)>1 0Cy,> -- ao)O ACCIDENT MORTALITY AMONG CHILDREN 161

040 (0'- 0)0) 0)0) '- 04li(0 >i OCO) 0- Cy-0- 00-0) C\- LO O~C6-C~I1 161614Oi6r LO"6: r- c- 0',-CY)-)CO ~> 0)0Cf) 0- 0 0)0C)ciC)C41')0 04- Nt C r c -LO -

't Lq q 0? ( CY?0 f04 11 160) ML)10 c')oMO 00o5c'iJ 6c6SW l6o c 'D6 10(0 6 1 C'10tClU)LOU) LO- -c\1 101') UC) (0'-) t'0 0-4 0 CC)00 (0IT 10- 00-CY) 00(D 0 C

IC') CJ CY C Oc6 O O -1tC )r-r O 0C Yc6c6 r- c CY - C) D - ~t0) -0)C 0-0LO ) LO li L U)q '-(0R LC)O C01- t U)1co 0)0 ) 0- CY) 04):;0)~ 10103c\ (

0)0 (00C)( (00 CY(00 0-c' 0-'C\ 0-0 C')0 0f)04r CY)0 '-04 C')-

OL- 10) 101') C L04 C)Y) U Cy)0L)0)-0 0)r-C)O " lU. ~ C6- \ 10\')I -04 10(004( 001 )0 (1) 0- 4 -04 *) (0-QC C')(qq O(0-r 4rl040-l 0) '-0 0)0 010 00 011) 0 0( 0(0 c0) 0(03 CY)( 00L)(0 ( 040-) 10 -0 U')10 L OC)- U? CY)OD n'- (0~ OC 00i- LO)4- (0-(D 0 C') 00)r (00-LC0r 0 U) r-r-4') 0) 0) 0)Ur0)O)0)O0 0)0 0)- '-C') O LO(00) F z 0 0 CD)m0)C0 010 C')1 0)0D (01 C-( 0-0- (0' (0'ct- "0c '-04\ 0) 040 0)It10 (00) (01') C')0)- O-) 0- & 0) C'0- (001' (004O 0 10(0 100) (0 '-r- 0)1' (0 C'0 LO)101 0C) ( C04'- P en C' - c6164 66r, o6 6 o6 60)10- ,0- C- '-04C, 10j (0 U' (0) -* 00"C) 04(-0040 0)0U) C 0)0 10T:0 C)'- 0'1- 0)0) p 0% g 0) 0)4 0410t 0-0- U,) CY) 0(0-L 100 c'-( (00 10(0 I'* c L (0Y)00 10 r-0) 0-- 0-M 0) 0)) )' (00U) (004 C\04) 0)i0) C-0O)' C)~ C) co 0)0 00 ) Y)010) ~0 C(0 0' r0-0)D100 101 U-10'It )1

0) 0- 0) ( 0)0 ()(Y0)0- 10Y 0)0 10'- 0'-0- 010 0)0 (00-O 0(0 % 04 o04 6- --~ r--y L6-'-'--' 0~a6()04'_- '-6 D '-_ 04 z LO L) 0)OC U' r- M (0 MM M I 3 LOL16 ItA oi1 (C U 0 U )) 100L)C 0-04 040) (0C 0'-L 101 U)0 '10U(0O 0-C) 0)r U CY 0)4 1~c, 66 6 01 6 0:06 cj6 o r6 o56 c~i L64 6c' U (0m 0-) 0)10 100)1M" 0) (00 0- LO0 100 10)MM - w;b 0) P- 10) 0-0 )1(001-(1 0-10 0(0 0-) (00(0)0

0-0- r-00- L(0(0 0404 00 (00 100) 0'- 0)0) C)0 ) ~ 0-4o6 074 '-c, 4 10 0106L6c4---1 6c c-i6 0.:6c 0-'- (010.- 0 04'- Dr-010 CY) 040404' 0)0 ) 1010C\ -CY C t )0

E '-("'-0l 10(0C 010CLO )0) 0)0rO) C0 10- 10 0 '-0- '- L 0)L- C) c'i1 co CO)(01 CO~ 66V)C 0)1') LO1 0)0- 8'-C 101 ')-' 0 co v -) 0-04 0(0q 0-0U) 0-) 110 0)0-) C0 (00)(0( 040 v4 U) 610. . 6 .16 r-c' 04. c g6 o 6 6 U')~om( U -CO U- (04 0- D - -L '- 0)1 0-10O"I r

C4 ~~~~~~-,

0 U)C " CL D -I'CCU) U) &MC cx, S U' - U)m( '.OwU)L COr O a or U') Wm "-U, co 0> ~ '~ - C Ca <( U" UorU 'It 0( -, Z m cm U'

0 La- 162 S. SWAROOP R. M. ALBRECHT & B. GRAB

Annex PROPORTIONAL MORTALITY FROM VARIOUS TYPES OF DEATHS, BY AGE AND SEX,

AUSTRALIA CANADA Type of accident (Intermediate List, 1948) Sex age-group (years) age-group (years) 1-4 5-9 10-14 15-19 1-4 5-9 10-14 15-19

Motor vehicle accidents M 26.5 39.6 30.3 60.9 33.9 44.5 30.9 42.3 F 27.2 50.3 39.0 71.1 36.7 59.6 47.1 67.8 Other transport accidents M 3.9 9.4 14.8 7.5 2.3 5.7 10.2 11.4 F 2.0 5.4 11.0 8.3 2.0 3.1 5.7 5.2 Accidental poisoning M 9.9 0.6 1.3 0.3 3.8 0.6 0.8 1.5 F 11.5 2.4 - 1.7 6.3 0.9 1.0 7.0 Accidental falls M 2.5 0.6 4.0 3.2 4.2 2.5 4.1 3.0 F 2.8 3.0 6.1 2.5 4.0 2.2 2.9 1.7 Accident caused by machinery M 0.5 1.1 1.3 1.3 2.3 2.0 2.5 4.7 F 2.2 - - - 2.1 1.1 1.0 0.7 Accident caused by fire and explosion M 5.0 3.7 - 0.6 15.2 8.9 5.7 1.5 of combustible material F 12.6 8.4 9.8 2.5 21.3 18.2 10.5 6.3 Accident caused by hot substance, M 9.4 0.9 0.3 0.1 4.1 0.4 0.1 0.1 corrosive liquid, steam, and radiation F 9.3 1.8 - - 4.8 0.2 - - Accident caused by firearm M 0.5 2.3 10.1 7.0 0.4 1.6 7.9 5.8 F 1.1 2.4 3.7 2.5 0.4 1.8 1.9 1.4 Accidental drowning and submersion M 28.6 30.2 24.6 11.9 24.8 27.3 30.1 21.0 F 19.1 15.0 17.1 7.4 13.2 8.8 21.9 6.6 All other accidental causes M 13.1 9.7 13.1 7.1 8.9 6.5 7.6 8.7 F 12.1 11.4 13.4 4.1 9.1 4.2 8.1 3.1

ITALY JAPAN 1-4 5-9 10-14 15-19 1-4 5-9 10-14 15-19

Motor vehicle accidents M 13.3 35.8 32.8 37.7 10.8 13.5 17.2 16.5 F 10.0 34.1 32.0 36.1 10.3 17.0 12.4 17.6 Other transport accidents M 4.7 8.2 8.6 6.9 6.5 6.9 12.2 20.3 F 3.5 4.3 7.6 6.7 5.6 7.2 9.0 20.5 Accidental poisoning M 4.4 1.8 1.0 1.4 1.4 0.7 1.0 3.7 F 3.3 2.2 1.2 6.7 1.3 1.3 1.3 8.9 Accidental falls M 8.6 13.0 14.7 12.7 2.9 5.3 8.7 11.6 F 6.7 9.9 13.9 13.3 2.5 4.6 4.7 4.2 Accident caused by machinery M 0.4 0.3 0.9 1.1 0.3 0.5 0.8 3.3 F 0.2 0.2 0.9 0.8 0.2 0.4 0.7 0.7 Accident caused by fire and explosion M 2.7 0.7 1.8 1.0 4.2 2.2 2.3 2.8 of combustible material F 4.7 7.0 3.0 3.6 5.3 5.5 4.8 4.9 Accident caused by hot substance, M 32.4 4.0 1.0 0.6 5.8 0.9 0.6 0.7 corrosive liquid, steam, and radiation F 40.9 13.5 4.5 3.1 7.5 2.7 1.3 1.1 Accident caused by firearm M 0.8 3.0 3.7 2.6 0.1 0.3 1.3 1.1 F 0.6 2.0 3.0 2.2 0.1 0.2 0.1 0.5 Accidental drowning and submersion M 23.1 18.7 21.3 21.2 59.8 6.5 39.0 12.7 F 20.3 9.2 11.5 13.1 58.3 4.3 45.0 15.2 All other accidental causes M 9.7 14.5 14.2 14.9 8.3 0.1 16.9 27.2 F 9.6 17.5 22.4 14.4 9.0 1.8 20.8 26.6

* For the years 1952-53 only ACCIDENT MORTALITY AMONG CHILDREN 163

4 ACCIDENT AS PERCENTAGE OF TOTAL ACCIDENTAL IN TWELVE COUNTRIES, 1951-53

CEYLON ENGLAND and WALES FRANCE GERMANY, Fed. Republic * age-group (years) age-group (years) age-group (years) age-group (years) 1-4 5-9 10-14 15-19 1-4 5-9 10-14 15-19 1-4 5-9 10-14 15-19 1-4 5-9 10-14 15-19

2.6 6.7 8.1 8.2 42.0 48.0 44.9 48.6 14.8 33.2 27.0 26.0 20.7 38.1 30.7 40.8 3.4 4.4 1.8 2.8 35.8 56.9 45.9 60.1 15.9 37.9 33.9 29.9 23.0 48.2 41.3 55.7 1.6 0.7 3.2 6.9 2.1 3.5 7.5 16.0 1.6 1.7 3.1 4.5 6.8 10.4 10.2 10.0 0.6 0.9 - - 2.0 2.1 8.3 13.1 1.1 1.3 3.8 4.5 6.3 10.0 13.8 13.9 6.0 4.0 0.8 2.1 6.5 1.1 0.6 1.1 3.5 2.0 1.9 2.1 3.7 1.5 1.2 1.5 3.8 2.2 1.3 5.5 7.2 1.7 2.9 3.7 4.8 4.5 6.7 10.1 5.3 3.4 3.4 3.4 2.6 5.7 21.3 21.6 6.1 5.6 8.7 5.8 3.6 3.9 3.5 4.0 4.2 5.7 8.4 6.4 1.3 1.1 1.3 4.1 6.8 4.2 4.4 3.4 3.1 3.2 1.7 2.9 5.1 5.3 5.4 3.6 - - - 0.5 0.1 0.3 1.0 1.8 0.3 0.1 0.3 0.2 0.4 0.7 1.0 1.5 - 0.2 0.4 - 0.2 0.4 - 0.4 0.3 - - - 0.3 0.6 - 0.4 14.9 12.6 6.7 3.3 4.5 3.1 1.8 0.9 1.1 0.8 - 0.3 3.5 2.0 2.1 1.2 29.0 43.8 26.8 30.3 15.9 21.4 15.1 6.3 1.0 0.5 0.4 0.2 3.0 2.1 3.0 1.6 6.6 1.0 0.3 0.5 7.1 0.1 0.3 0.1 19.7 3.1 0.4 1.2 25.8 1.9 0.7 0.1 7.0 0.7 0.4 1.4 7.9 0.4 0.5 - 23.4 8.9 3.8 2.3 25.3 5.5 1.7 1.1 0.6 0.5 2.7 2.1 - 1.0 3.0 2.8 0.3 2.1 3.7 3.0 0.1 0.8 2.1 0.9 0.2 - - 0.7 - 0.2 1.0 0.4 1.1 1.3 0.8 2.1 0.1 0.4 - 0.4 49.2 40.2 28.8 24.9 20.4 28.8 18.0 12.2 29.5 23.8 26.8 22.6 22.7 24.5 27.0 17.0 41.1 32.4 41.2 28.3 12.1 6.8 12.2 6.0 24.6 8.4 20.9 22.5 19.1 15.4 18.5 10.1 15.9 28.6 28.0 29.8 11.2 8.5 14.2 10.5 25.5 29.2 33.3 35.9 12.2 14.4 16.6 20.8 13.5 14.4 26.8 26.9 12.0 5.7 9.8 6.7 24.6 33.9 28.0 25.6 12.4 9.0 13.1 9.7

NETHERLANDS SWEDEN SWITZERLAND USA 1-4 5-9 10-14 15-19 1-4 5-9 10-14 15-19 1-4 5-9 10-14 15-19 1-4 5-9 10-14 15-19

21.6 34.2 35.0 27.5 22.8 37.9 29.5 57.5 21.8 39.3 31.7 37.3 33.4 41.5 31.7 57.5 21.9 36.4 33.6 24.8 31.7 58.8 59.1 75.0 23.4 52.0 46.3 44.2 30.7 43.3 44.4 72.5 6.2 8.6 9.9 12.3 5.2 11.7 16.3 12.8 4.8 7.3 7.2 9.2 1.7 5.5 7.3 5.9 7.7 13.0 8.0 10.1 2.5 10.3 6.8 10.4 7.8 9.2 2.4 11.6 1.2 2.6 6.4 3.5 2.0 1.2 0.9 2.0 1.3 - 0.6 1.1 2.6 1.0 0.7 1.5 7.8 0.8 0.6 1.5 2.0 0.4 0.8 6.2 4.3 1.0 4.5 4.2 2.8 3.1 2.3 8.7 1.4 1.7 3.6 3.8 3.2 6.1 7.0 3.6 2.0 4.8 3.7 8.8 10.5 17.3 17.7 4.6 4.3 3.1 2.7 4.4 2.7 4.0 1.6 5.0 3.1 2.3 - 7.3 4.1 17.1 14.0 4.9 3.4 2.8 1.6 0.6 1.2 1.5 2.0 1.6 0.5 1.2 2.0 0.9 0.7 1.5 2.1 1.8 4.1 2.8 0.2 1.1 0.8 1.4 1.0 1.1 1.4 1.2 0.5

1.5 0.6 0.3 2.2 6.2 1.8 - 0.9 2.0 1.0 3.6 1.5 16.7 9.3 4.1 1.7 1.1 1.5 0.8 13.0 1.0 - - 2.3 1.0 25.8 26.7 13.3 6.3

6.3 0.2 0.3 4.6 - 0.6 0.3 12.2 2.6 3.8 0.4 0.2 0.1 7.0 1.1 1.6 4.3 1.0 - - 20.6 4.1 2.4 2.3 5.0 1.7 0.8 0.4

0.2 0.6 - 0.3 3.6 1.7 0.5 3.6 1.8 1.8 4.8 12.4 7.1 2.0 4.7 1.4 2.8 4.2 2.5 39.3 32.1 21.0 19.9 45.0 40.5 30.1 12.3 31.0 22.5 18.0 17.0 15.3 22.2 26.1 14.1 31.1 12.3 7.2 9.3 30.4 17.5 18.2 6.3 17.4 9.2 9.8 14.0 9.1 8.6 17.2 5.6 18.7 18.5 25.1 26.6 9.8 5.3 13.3 7.7 15.9 15.2 17.3 12.5 13.0 9.3 10.4 6.6 24.6 31.4 45.6 45.7 8.7 7.2 4.5 4.2 17.0 14.3 22.0 7.0 12.2 8.0 8.1 3.6