CEYLON J . M E D . SCI. (D) VOL. VII, P T . I l l and IV. (December, 1 9 5 0 ) . An Analysis of the Vital Statistics of Ceylon B Y H. CULLUMBINE (Department of Physiology and Pharmacology, University of Ceylon, Colombo, Ceylon). Introduction The health of a population can be assessed in two main ways, namely, (i) in a negative fashion by studying the causes of illness and of death within the population and (ii) positively by defining the nutritional status, physique and functional capabilities of the people. The attempt at a positive assessment of the health of the people of Ceylon has already been described in previous papers from this department. In the following discussion the negative assessment is presented, using the Ceylon Registrar-General's statistics for the years 1937 to 1948 inclusive. This period avoids the major malaria epidemic in 1934-35 and covers about 10 years before D.D.T. was used in Ceylon and 2 years following its use. While the statistics were being collected, the 1949 report of the Registrar-General appeared and some relevant data from this source have also been included. Mortality rates for various diseases and occurring in both sexes and in the different ethnic groups and regions of Ceylon have been calculated for each year and then averaged for the period 1937-48. In some cases only the data for male subjects are given; where this is so, it can be assumed that the statistics for females are substantially similar. (Lack of space has prevented the reproduction of the standard deviations, etc. of many of the cal­ culated figures). It is not possible to judge the accuracy of these statistics given by the Registrar- General. Only about 15 per cent, of deaths are registered with medical registrars and a minority of all certifications are made by practitioners of' western medicine '. For this reason groups of major diseases have been considered rather than specific causes. In any case, while it is usually agreed that Ceylon's vital statistics are more accurate than those of the rest of South-East Asia, they are all that are available to us at present. The information about Ceylon has been compared with the latest statistics available from certain countries, and particularly Eire, in the Western hemisphere. Eire was chosen because it has a superficial resemblance to Ceylon. Not only has it a similar area but it has an agricultural economy and possesses only one city of appreciable size. There are more differences than resemblances between Eire and Ceylon, as we shall see, but a comparison between the two countries is helpful in assessing the mortality trends in Ceylon. 2 0 0 6 — A 92 H. OULJjTJMBINE Other sources of information include the Census Reports for Ceylon, and especially that of the 1946 Census by Ranasinha, and the Administration Reports of the Director of Medical and Sanitary Services. I am deeply grateful to my clerical assistants, Mr. V. Abeywardene and Mr. V. Somaskandan, who have helped in the laborious collection and compilation of the statistical tables. CEYLON J. M E D . Sen, (D) VOL. V I I , Fx. m and TV. (December, 1960JVV ' j . j PART I Some Recent Population The Ceylonese nation is a composite of many ethnic groups and the distribution (see map I) of the several communities or racial groups in Ceylon reflects their past history. The Sinhalese, who are the largest community, are to be found chiefly in the central, western and southern parts of the island- The Ceylon Tamils comprise the great majority of the population in the northern parts of Ceylon, although some have migrated south and to the Eastern Province. The Indian Tamils are more recent migrants, who have been introduced into Ceylon since the latter part of the nineteenth century to work on the tea and rubber estates. They are to be found, therefore, mainly in the estate area of the Central and Uva Provinces. The Ceylon Moors have a mixture of Tamil and Arabian blood and are said to be descendants from the ancient Arab traders. The Indian Moors are traders, temporarily domiciled in Ceylon for their business and commercial activities. The Malays are descendants of people from Malaya and the islands of Oceania, who were recruited as soldiers by the British, the Dutch and the last Eandyan Kings. The Dutch Burghers have a mixed Dutch and Sinhalese ancestry and the Europeans are, in general, British subjects temporarily resident in Ceylon as government officials, planters and commercial agents. The Veddhas are a primitive race having Australoid traits. They are a reminder of the pro-Sinhalese era of Ceylon's history and a few are said still to exist in the remote jungle areas. TABLE 1 The Population of Ceylon by Race Groups, 1921-1946. Race POPULATION Per cent Increase ( + ) or Race 1921 1946 Decrease (-), 1921-46 Sinhalese 3,016,154 4,620,507 +53 Ceylon Tamil 517,324 733,731 +42 Indian Tamil 602,735 780,589 +30 Ceylon Moor 251,938 373,559 +48 Indian Moor 33,026 35,624 + 8 Burghers & Eurasians 29,439 41,926 +42 Malays 13,402 22,508 +68 Veddhas 4,510 2,361 -48 Europeans 8,118 5,418 -33 Others 21,959 41,116 . + 87 SOME RECENT POPXTLATION TRENDS 95 At the 1946 Census, the total population of Ceylon was enumerated as being 6, 693,945 persons. This population is rapidly increasing, as the figures for the Census years 1921 and 1946 show (Table 1). All the 'native' races (except the Veddhas who are now practically indistinguish­ able from their neighbours) have increased their numbers by about 50 per cent, during thisintercensal period. This fast growing population is confined within an area of 25, 332 square miles. (Ceylon lies between 5° 55' and 9° 50' north latitudes, and 79° 42' and 81° 53' east longitudes. The greatest north to south length is 270 miles and the greatest breadth, east to west, is 140 miles). Not all this land is readily available for occupation or cultivation. The central area of the island is hilly and some peaks reach a height of over 7,000 feet. The rest of the island is for the most part flat and extensive portions of the north, central and eastern plains are still covered with jungle. While history has largely determined the racial-distribution of Ceylon's people, it is the rainfall, influencing the fertility of the soil and the endemicity of malaria, which has largely determined the density-distribution of the population. The well- watered and fertile regions of the west and the south have the greatest concentration of people, while the dry lands of the north-central area are but sparsely occupied (Map II). For administrative purposes, Ceylon is divided into 9 provinces and these are further sub-divided into a total of 21 Revenue or Administrative Districts (MapIII). Province Western Central Southern Northern Eastern North-Western Sabaragamuwa Uva North-Central Districts (1) Colombo, (2) Kalutara, (3) Negombo (4) Kandy, (5) Matale, (6) Nuwara Eliya (7) Galle, (8) Matara, (9) Hambantota (10) Jaffna, (11) Mannar, (12) Vavuniya (13) Batticaloa, (14) Trincomalee (15) Kurunegala, (16) Puttalam, (17) Chilaw (18) Ratnapura, (19) Kegalle (20) Badulla (21) Anuradhapura Local administration functions through Village Committees (about 400 in 1946), Urban District Councils (38 in 1946) and Municipal Councils (at Colombo, Kandy and Galle in 1946). The provinces and the districts differ in their population density and in the composition of their population (Tables 2 & 3), these differences merely illustrating the above remarks about density-distribution and the racial-distribution of the population. > SOME RECENT POPULATION TRENDS 10 12 14 21 13 Map 3.—Administrative Districts of Ceylon. 98 H. OuXLTJMBINE TABLE 2 The Population, Population Density and the Urban/Rural Population Ratio of the Districts of Ceylon at the 1946 Census. Province District Total Population (thousands) Population Density (thousands/ sq. mile) Urban/Rural Population Ratio Western — 1876-9 1-311 0-577 Colombo & Negombo 1420-3 2-400 0-650 Kalutara 456-6 0-748 0 125 . Central — 1135-3 0-496 0-109 Kandy 711-4 0-775 0 1 1 7 Matale 155-7 0 1 7 3 0-099 N'Eliya 268-1 • 0-504 0 038 Southern —- 961-4 0-44S 0 122 • Gallo 459-8 0-705 0 147 • Matara 351-9 0-732 0-112 ' • Hambantota 149-7 0-147 0 070 Northern — 479 G 0-140 0 159 Jaffna 424-8 0-426 0 146 Mannar 31-5 0 033 0-218 Vavuniya 23-2 0 0 1 0 0-221 Eastern — 279-1 0 073 0-256 Batticaloa 203-2 0 072 0 130 Trincomalee 75-9 0-066 0-623 North-Western -- 667-9 0-221 0 055 Kurunegala 485-0 0-264 0 033 Puttalam 43-1 0 047 0 027 Chilaw 139-8 0-533 0 069 Sabara- gamuwa — 745-4 0-394 0 030 Ratnapura 343-6 0-275 0-037 Kegalle 401-8 0-623 0-019 Uva Badulla 372-2 0-114 0 046 North-Central A nuradhapura 139-5 0 035 0-097 • SOME RECENT POPULATION TRENDS 99 TABLE 3 The Racial Composition of the Population of each District in Ceylon. District Per Cent of Total Population who are District Sinhalese Ceylon Tamil Indian Tamil Ceylon Moor Indian Moor Burghers and Eurasians Colombo and Negombo 81-1 4-10 4-30 4-20 1-25 2-02 Kalutara 8 0 S 0-83 6-44 5-42 0 0 9 0-15 Kandy 57-8 4-20 29-2 6-79 0-65 0-53 Matalo 6 8 1 3 1 4 21-9 * 5-24 0-60 0-30 N'EIiya 37-8 1-97 57-3 1-44 0-46 0-35 Galle 94-5 0-68 1-49 2-98 0-03 0-16 Matara 94-4 0-67 2-12 2-55 0 0 1 0 1 1 Hambantota 96-6 0-53 0-17 1-59 0 0 1 0-07 Jaffna 1 0 7 96-3 0-99 1-21 0-11 0 0 9 Mannar 3-82 51 0 11-2 30-2 2-87 • 0-13 Vavuniya 16-6 69-3 4 1 6 8-72 0-54 0-33 Batticaloa 5-83 49-7 0-59 42 0 0-21 0-57 Trincomalee 20-7 40-1 4-43 29-2 1-43 1-55 Kurunegala 92-2 1-52 1-60 3-85 0-36 0-14 Puttalam 52-0 12-8 2-13 30-8 0-81 0-23 Chilaw 87-0 5-17 3-31 2-92 0-32 0 1 6 Ratnapura 75-8 1-17 20-6 1-36 0-34 0 1 6 Kegalle 82-1 0-80 12-9 3-51 0-24 0 1 1 Badulla 57-4 4-18 34-2 2-72 0-47 0-30 Anuradhapura 79-7 6-67 2-22 10-1 0-65 0-16 It is not intended to discuss in detail the demographic position in Ceylon, but is necessary to indicate the present population distribution and rate of growth as a background to the mortality and morbidity figures. (A full discussion can be found in the General Report, Census of Ceylon, 1946). The ancient records suggest that the population at the height of Ceylon's prosperity was many times greater than at present. If this be true then the Ceylonese must have approached near-extinction during the seventeenth and eighteenth centuries and have later apparently enjoyed a renascence during the-prolonged period of British rule. Thus the population has been given at over 70 million persons at one period of the island's history (Denham ; Ceylon at the Census of 1911, p. 9). Arunchalam calculated a population of at least 10 millions in the early 14th century A.D. (Census of Ceylon, 1901, p. 23) but the Census of 1827 enumerated only 889,584 100 H . ' C U L L U M B I N E persons. The accuracy of these figures cannot now be assessed and it is only since the first decennial census of 1871 that reliable data on the island's population are available. These later figures show that, in common with the rest of the world, the population of Ceylon is increasing rapidly. In the countries of North-Western Europe (except Holland in which rapid growth is still continuing) and in those countries inhabited mainly by people of European descent (U.S.A., Australia, New Zealand, Canada) the rate of growth of the population is now declining. This is not true for Ceylon (Table 4), where the rate of growth seems to be still increasing. The Western countries had their maximum rates of population growth in the 19th century ; Ceylon has probably still to reach her maximum rate. TABLE 4 The Population of Certain Countries at Various Dates. DATES Total Population in Thousands Increase ( + ) Decrease (-) or a Percentage of Population as 20 years before Great Britain Ceylon Eire Great Britain Ceylon Eire 1801 10,601 — — — — — 1821 14,092 — — 34 H — — 1841 18,534 — — 32 — — 1861 23,128 4,402 25 — — - 1881 29,710 2,760 3,870 28 — -12 1901 37,000 3,566 3,222 25 29 -17 1921 42,769 4,499 2,972 16 26 - 8 1941 46,605* 6,657f 2,992 9 48 0 * Estimated figures t 1946 Census The increase in population during the latter part of the 19th century was due largely to the immigration of Indian labourers first to coffee and then to the tea estates of Ceylon. More recently and for several reasons this immigration has been substantially reduced. The Indian immigrant labourers became resident on the estates and their descendants have formed a natural source of labour for replacement and expansion. In addition, further extension of the estates is not possible to any considerable degree so that the need for new immigrants is small. Moreover, in an effort to ' Ceylonise ' the local industries, in 1939 the Ceylon Government banned the immigration of labour from India. Despite this great reduction in the number of immigrants, the rate of growth of the population is still increasing (Table 5). SOME RECENT POPULATION TRENDS > 101 TABLE 5 The Recent Increase in the Population of Ceylon. Natural and Actual Increase of Population 1871-46 PERIOD Natural Increase (Excess of births over deaths) Net Gain ( + ) or Loss (-) by Migration Actual Increase 1871-1881 120 +240 360 1881-1891 144 + 104 248 1891-1901 225 +333 558 1901-1911 362 + 178 540 1911-1921 313 + 80 393 1921-1931 663 + 145 808 1931-1946 1,286 + 64 1,350 A comparison of the rates of growth during the last two intercensal periods suggests, however, that the rate of growth may have passed its peak, e.g. Years Actual Increase per Year (Thousands) Per cent. In­ crease per Year Natural Increase per Year (Thousands) Migration Increase per Year (Thousands) 1921-31 8 0 - 0 1 - 8 0 6 5 - 7 1 5 1 1931-46 9 0 0 1 - 6 9 8 5 - 4 4 - 6 The difference between the actual increases for these intercensal periods is due entirely to the greater number of births in the later period, e.g. Years Average Births per Year Average Crude Birth Bate Average Deaths per Year Average Crude Death Bate 1921-30 194 ,612 3 9 8 128 ,917 2 6 5 1 9 3 1 ^ 5 213 ,968 3 6 8 128 ,574 2 2 - 2 The average crude birth rate and death rate have decreased but, because of the different population sizes and the greater rate of decrease of the death rate, the average number of deaths per year is about the same while the average births per year have increased. It is probable that migration will not affect, to any substantial extent, the population of Ceylon in^the near future and that the rate of population growth will depend almost entirely upon the natural balance between births and deaths. It is only since the 1946 Census that the deaths in Ceylon have shown any marked 102 H. CULLUMBINE decrease so that it is only recently that the improved health of the people has influenced the rate of growth of the population. In Table 6 it will be seen that whereas the number of births per year has gradually increased during the last 2 0 years, the number of deaths has only declined during the last 3 years. TABLE 6 Annual Births and Deaths in Ceylon for the Period 1932-1949. Year Births Deaths Natural Increase ( + ) or Decrease (-) Per cent. Natural Increase ( + ) or Decrease (-) 1032 199,370 110,649 + 88,621 + 1-6G 1933 209,032 114,690 + 94,342 + 1-75 1934 206,512 127,069 + 79,443 + 1-46 1935 192,755 204,823 - 12,068 - 0 - 2 2 1936 192,060 123,039 + 69,021 + 1-23 1937 216,072 124,210 + 91,862 +.1-63 1938 208,389 122,299 + 86,090 + 1-51 1939 212.111 128,611 + 83,500 + 1-44 1940 212,980 122,738 + 90,242 + 1-53 1941 219,864 113,003 + 106,861 + 1-79 1942 221,064 112,044 +109,020 + 1-81 1943 248,820 131,001 + 117,759 + 1-95 1944 232,827 133,985 + 98,842 + 1-61 1945 238,494 142,931 + 96,561 + 1-54 1946 256,886 135,937 + 120,949 + 1-86 1947 271,191 98,544 + 172,647 +2-58 1948 287,695 93,711 + 193,984 + 2-82 1949 291,191 91,889 + 199,302 ' + 2-82 In other words, we can distinguish 3 phases in the growth of the population of Ceylon, viz. (1 ) between 1871 and 1901 when emigration from India was the major contributory factor ; (2) between 1901 and 1946 when the increase depended mainly on the increase in the number of births ; and (3) a phase which is commencing and in which the death rate is declining rapidly. It is this recent new factor which complicates any estimates of Ceylon's future population. Until this trend can be studied over a period of years any population predictions must be looked upon with caution. Ranasinha (Census of Ceylon, 1946) has estimated that the population of Ceylon will be nearly 11 million in 1981 , but his estimates are derived from population trends existing before the death rate began to decline. His estimates are, therefore, probably too low and, indeed, whereas he estimated a population of 7 ,147 ,000 for 1951 , the Registrar-General of Ceylon gives an estimated population 7,297,000 for SOME RECENT POPULATION TRENDS 103 31 32 33 34 3S 36 37 38 39 40 41 42 43 44 45 46 47 48 49 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 i i i i • i i i I I I i i i i PROJECTION INTO THE FUTURE OF THE RECENT TRENDS OF THE B I R T H S A N D D E A T H S I N C E Y L O N 140 325 290 250 L l L_ - I ' • 104 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 ' 1 Figure 1. 104 H. OUVLU1XBIXT& AGE PYRAMIDS FOR CEYLON. 1891-1946 MALES FEMALES i i 400000 300000 200000 100000 MALES i i i 400000 300000 200000 100000 0- 4 Age 1891 6S-over 60-64 55-59 50-54 45-49 40-44 35-39 30-34 25-29 20-24 15-19 10-14 5- 9 0 - 4 Age 1946 i i 100000 200000 300000 400000 FEMALES 100000 200000 300000 400000 SOME RECENT POPULATION TRENDS 105 1949. - Assuming that the present trends in births and deaths continue (Figure 1), we can estimate the population as being 10,180,000 in 1961 (allowing 24,000 gain by migration each year, which is the average for 1946-49); Ranasinha estimates the 1961 population at 8,197,000. It is possible that our estimate is also too low since we have assumed only a very gradual rise in total births each year and an even slower fall in the number of deaths. For one thing the recent great reductions in infant and maternal mortality rates is going to increase the fertility of the Ceylon female' population but it is not possible to estimate the increase accurately until the present mortality trends become stabilized. The Change in Age Distribution:— The Age Pyramids for the census years 1891 and 1946 are shown in Chart 1. In the population of 1946 every age group is larger than the next older group and an age pyramid is obtained which is typical of those found in Western countries fifty years ago. The proportionate age distribution of the Ceylon population in 1946 is also very similar to that in Great Britain in 1891 (Table 7). However, the proportion of young people has fallen steadily during the past 65 years and, although this ageing process is not occurring as rapidly as in, for example, Great Britain, yet it will presumably be accelerated in the future owing to the greater expectation of life resulting from the recent decrease in the general death rate. TABLE 7 The Age Distribution of the Population of Ceylon in Various Census Years. CEYLON (Thousands) Year MALES FEMALES 0-19 20-39 40-59 60 & over 0-19 20-39 40-59 60 & over 1881 774-2 451-9 190-1 50-2 725-0 386-4 143-8 32-1 1801 834-2 498-8 204-3 55-4 785-3 433 1 160-9 34-8 1901 972 0 624-5 239-5 58-0 . 917-8 528-3 184-9 38-5 1911 1050-4 717-5 306-8 100-3 976-2 627-1 •252-1 76-0 1921 1131-6 788-9 350-1 111-2 1053-7 695-2 281 1 84-8 1946 1628-3 1132-7 565-3 I 9 6 0 1530-6 994-6 435-8 164-0 Per cent, of Total Population. 1881 53 31 13 3 56 30 11 3 1891 52 31 13 4 56 31 11 2 1901 61 . 33 13 3 55 32 11 2 1911 48 33 14 5 51 32 13 4 1921 47 33 15 5 50 33 13 4 1946 46 32 16 6 49 32 14 5 10.6. H . - O ULLTJM BINE G reat Britain— Per cent. D istribution of Population (M ales and Fem ales). Y ear 0-19 20-39 40-59 60 & over 1891 45 30 17 7 1947 21 30 26 15 The changes in the m ortality rates for the past years can be seen by a com parison, of the death rates at various ages for the years 1937 to 1948 (Table 8). oo n e1 to >3 0Q CO o CO C O to "3 » •A r T A B L E 8 ( B ) Death Bates for each Age Period per 1,000 of the Population at the same Age. HALES Age Periods Y E A R S Age Periods 1937 1938 1939 1940 1941 1942 • 1943 1944 1945 1946 1947 1948 Under 5 years 65-7 63-2 Q4-2 59-0 52-5 4 8 - 7 . 61-0 62-4 61-6 60-4 48-2 46-9 5 - 7-12 6-36 6-76 6-40 5-90 5-85 7-75 8-18 7-71 6-20 4-20 3-80 10- 3-63 3-28 3-51 3-46 3-37 3-29 3-51 3-74 3-60 3-18 2-08 1-89 16 - 5-58 5 0 6 5-75 5-22 4-88 5-02 5-65 5-93 6-49 5-10 2-93 2-60 2 0 - 7-10 6-87 6-98 6-49 6 0 7 6-39 7-05 6-98 7-50 6-51 4-27 3-89 2 5 - 9 0 3 8-33 8-27 8-04 7-43 7-73 8-32 7-84 8 0 7 7-34 5-44 4-45 3 5 - 13-6 12-9 13-3 1 3 0 11-7 12-8 13-5 12-7 1 3 1 11-7 7-69 6-45 4 5 - 21-8 20-9 • 21-3 21-3 19-3 2 1 0 22-8 21-5 23-4 20-8 12-6 11-3 55 + 65-1 65-8 70-7 7 0 0 66-0 71-9 76-6 74-2 78-2 68-4 43-0 38-9 to O 3 H E a § H •d O 1-3 h-t o as a CO O - 4 o 00 T A B L E 8 ( C ) Death Bates'for each Age Period per 1,000 of ihe Population at the same Age. FEMALES Age Periods Y E A R S 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 .1948 Under 5 years 64-7 63-5 66-9 59-5 52-8 . 49-3 63-0 64-3 63-1 67-6 47-6 46-1 5 - 8-00 7-80 7-83 7-71 6-71 6-97 9-17 9-40 8-98 7-11 4-70 4-30 10- 4-09 3-84 4-14 4 1 2 3-78 3-74 4-19 4-20 4-50 3-52 2-28 1-88 16- 7-05 7-08 6-95 6-53 6-26 5-99 6-22 6-65 7-31 6-10 4-18 3-36 20 - 11-7 11-3 10-9 9-96 9 1 7 8-93 9-36 9-31 10-8 10-3 7-37 6-19 2 5 - 1 4 1 13-1 12-8 11-8 n o 10-3 10-3 10-4 11-3 11-2 8-78 7-35 35 - 15-9 15-1 14-7 14-6 1 3 0 13-4 13-5 12-8 14-4 13-4 8-78 7-87 4 5 - 1 8 0 16-5 1 7 0 1 7 1 15-2 15-6 15-8 15-8 17-2 15-5 10-5 9-45 55 + 76-6 75-7 82-3 82-3 76-4 77-9 82-8 81-8 90-1 78-9 44-6 42-4 w Q a SOME REGENT POPuXATTON TRENDS 109 These figures show that the improvement in the death rates has occurred with more or less equal emphasis at all ages and for both sexes, so that for all age-periods the expectation of life must have increased. •There are, at present, no official life-tables available for the Ceylon population and the data presented in the Registrar-General's Reports are not sufficiently detailed to allow accurate estimates to be made. The compilation of life-tables is an actuarial feat involving elaborate calculation, presumptions and deductions which the accuracy of the basic data available in Ceylon would hardly seem to justify. Some of the omissions from the Registrar-General's Reports which make life- expectancy calculations so difficult include the absence of information on births by sex for each quarter of the calendar year and the fact that deaths are not classified by each year of life. Despite these handicaps, it did seem necessary to make some assessment of the change in life expectancy which is occurring in Ceylon today as a result of the rapid alterations in the death rate. The graduated rates of mortality at infantile a g e 3 and the average length of life experienced by those infants have been calculated by the methods detailed in the Registrar-General's Decennial Supplement, England and Wales 1931, the assumption being made that the sex-ratio at birth was the sams for each quarter of the calendar year as that given for the whole' year. It did not seem justifiable to continue this assumption into the calculations of the death rates for each year for the ages 1 to 5 years, so these rates were calculated direct from the recorded deatbs for each year of life and the population for that age. For persons over 5 years of age, deaths are only enumerated for age-periods of 5 or 10 years so that here the assumption has been made that the death-rate calculated for each age-period was the same as that existing at each year of age of that period. In view of the possible errors in recording ages in Ceylon this assumption probably gives a fair estimate and certainly more elaborate procedures are not justified. The Life-Tables cannot, therefore, be absolutely accurate but, since gross changes are being looked for, they probably give a reasonable estimate of the mortality experience in Ceylon. ' For the whole of Ceylon, separate Life-Tables have been estimated for males and females and three such tables have been compiled for each sex. In the first one, the practice followed in compiling the English Life-Tables has been used and the death rates for the years, 1945,1946 and 1947 have been applied to the 1946 Census data. However, it is since 1946 that the great reduction in mortality has occurred so that such a procedure, while having a sound statistical basis, will probably give erroneous estimates. To assess the alterations in life expectations that may have occurred recently, therefore, the death rates for the years 1946 and 1949 have also been applied to the 1946 Census. These Life-Tables are given in detail in the Appendix, but some of the more important estimated life-expectations are presented in Table 9. Up to the age of 45 years, males have a longer expectation of future life than have females ; in late-middle life the females have a slightly greater expectation although this changes in old age to favour males again. The enormous difference which the new mortality rates have made to the average future life time is readily evident from the comparative figures for 1946 and 1949. At all ages, both males and females now have a chance to live longer, the increase in life-expectation being of 110 H. CLTLLr/MBINE TABLE 9 Calculated Life-Expectations in Ceylon. Age in Years Expectation of Life in Years for the 1946 Population using the Mortality Bates for the Years. Age in Years 1945-47 1946 1949 Age in Years Males Females Males Females Males Females 0 47-2 42-5 43-8 41-5 54-2 52-8 5 55-5 51-4 52-4 50-2 59-9 58-5 10 52-1 48-2 49-1 47-0 5 6 0 54-7 20 43-9 40-2 40-8 39 0 47-0 45-9 30 36-6 34-2 33-5 33-2 38-7 38-5 45 25-3 24-3 22-7 23-7 26-6 27-1 60 1 6 0 14-3 13-8 1 4 0 1 6 0 ' 15-9 the order of 10 to 11 years. Even with the 1949 mortality rates males can still expect to live longer but the increase in future length of life has been greater for the females. One result of this changing mortality risk in Ceylon is that the proportion of aged people in the population will increase. This ' ageing' has been progressing for some time in most countries and was even evident in Ceylon before the 1946 Census (Table 10) . Ceylon's population is still younger than those in Eire and Great Britain but the proportions in the various age groups are altering and this alteration will be accelerated by the recent change in life-expectation. TABLE 10 Changes in Age Distribution of the Population of Ceylon and the British Isles. Country Year Proportion Per cent, of Total Population Aged— Country Year 0-19 20-39 40-59 60 & over Ceylon 1891 54 31 12 3 Ceylon 1946 48 32 15 5 Eire 1891 45 27 19 9 Eire 1941 30 29, 21 14 Great Britain 1891 45 30 17 7 Great Britain 1947 28 30 26 15 Marriage:— ' Since nearly all births occur to married women it is obvious that the proportion of people who marry before they are too old to have children is a factor of basic importance to the trend cf the population '. The proportion of people aged 4 5 - 5 4 who have been married may be taken as a reasonably good index of the marriage trend of the population, while the proportion of people aged 2 0 - 2 4 who are married may be taken to indicate changes in the ages at which the people are marrying (Table 11) . SOME RECENT POPULATION TRENDS 111 TABLE 11 Proportion of People of each Sex Aged 4 5 - 5 4 and 2 0 - 2 4 Years who were or had been married (Ceylon). Year Aged 45-54 Aged 20-24 Males Females Males Females 1901 87-6 90-4 37-1 79-1 1911. 88-2 01 -6 23-2 72-8 1921 89-0 92-6 18-6 69-0 1946 92-6 ' 96-4 19-6 70-6 There has been a steady rise during the century in the proportion of both males and females, aged 45-54 , who have been married. Part of this rise for females may have been due to the increased expectation of life of married women due to the falling maternal mortality rates, but a greater possibility is the improved method of enumeration of conjugal condition with each successive census. In the census reports of 1911 and 1921 it is stated that the parties to unregistered marriages were often recorded as ' unmarried '; it is claimed that this error was eliminated in the 1946 Census. (There are three recognised systems of marriage in Ceylon. That for the general population recognises marriages solemnized by a Christian Minister or by the Registrar of Marriages. The Kandyan (Sinhalese Marriages may be in diga (patrilocal marriage) or in binna (matrilocal marriage). Fraternal polyandry also occurred until recently among the Kandyans, but it is now believed to have died out. The registration of Muslim Marriages is regulated by a law passed in 1886. Limited polygamy is allowed but is said to be infrequent among Ceylon Muslims). Too much stress must not, therefore, be placed upon the figures, purporting to marriages, in the previous census years. Probably the proportion of people marrying has not altered to any great degree. This is certainly true for example in Great Britain (Table 12) . TABLE 1 2 Proportion of People Aged 2 0 - 2 4 and 45 -54 who were or had been married (Ceylon and Great Britain). Aged 20-24 years Aged 45-54 years Year Coylon Per cent. Great Britain Per cent. Ceylon Per cent. Great Britain Per cent. 1901 57-5 22 89 0 87-1 1911 48-1 19 89-9 85-5 1921 43-8 22 90-6 85-2 194.6 44-3 32* 94-2 86-9* * 1947 figures. 112 H. OULLtTMBINE However, the figures for the population aged 20-24 years do indicate that females and especially males are marrying later than they did in 1901, although even this trend appears to be altering in 1946. Females marry much sooner than males and there is, in the general population, a larger proportion of married females than married males. (In Great Britain, at 45 to 54, there have always been a greater proportion of married males than married females but the influence of emigration, war, etc., complicates the picture there). TABLE 13 The Marital Condition of Population of Ceylon (1901-46). 1901 1911 1921 1946 Total Per cent. Total Per cent. Total Per cent. Total Per cent. Single 2132-4 59-8 2459-8 59-9 2091-6 59-7 3783-5 56-8 Married 1200-1 33-7 1378-2 33-6 1495-3 33-3 2498-5 37-6 Widowed 233-2 6-5 268-3 6-5 310-8 7-0 375-3 5-C (Totals are in thousands) TABLE 14 Annual Births, Marriages and Birth Rates in Ceylon 1933-1949. Years Total Births (Thousands) Total Marriages (Thousands) Total Persons Married (Thousands) Marriage Rates 1933 209-0 27-2 — 5 0 2 1034 206-5 29-8 * — 5-36 1935 192-8 2 5 0 — 4-46 1936 192-1 28-3 56-8 5-02 1937 216-1 34-3 69-4 6 0 7 1938 208-4 35-5 70-8 6 0 9 1939 2 1 2 1 32-6 64-8 5-51 1940 2 1 3 0 33-5 66-6 5-62 1941 2 1 9 0 36-5 72-8 6 04 1942 221 1 48:5 96-5 8 0 4 1943 248-8 50-2 • 100-2 8-18 1944 232 • 8 46-9 93-6 7-46 1945 238-5 44-3 87-9 6-82 1946 256-9 43-0 85-6 6-42 1947 271-2 42-4 84-4 6-15 1948 287-7 44-7 89-3 6-30 1949 291-2 45-7 91-3 6-26 SOME REGENT POPULATION TRENDS 1 1 3 We have already noted that the number of births occurring each year is steadily increasing (Table 6 ) . This increase may be due to two factors, a change in the number of married couples or a change in the fertility of married couples. The number of mairied people in the population has increased steadily since 1 9 0 1 (Table 13 ) though it is only since 1921 that the proportion of married people in the population has increased. In other words the increasing numbers of people married reflects tbe increasing size of the population but, in the last intercensal period, there has also been a rise in the proportion of people married. This can be studied in more detail by com­ paring the annual total marriages and marriage rates since 1933 (Table 14) . The annual marriages have increased in a fluctuating manner. There was a high peak during the war ( 1943 ) and a fairly high rate has been mere or less maintained up to the present time. The variation in the annual number of marriages is closely paralleled by the variation in the annual marriage rate, and we can conclude that a greater proportion of people are now marrying and this greater rate, together with the increasing size of the population, has produced a considerable (68 per cent.) increase in the number of annual marriages since 1933. This increase in the marriage rate would itself readily account for the increased number of annual births over the same period. A comparison between tbe proportion of people in the various age groups who were or had been married in 1921 and 1946 is also of interest (Table 15) . TABLE 1 5 Proportion of People who were or had been married, by age, in Ceylon ( 1 9 2 1 and 1946) . Age Group 1921 Per cent. 1046 Per cent. 15-10 13-9 1 2 1 20-24 43-7 44-5 25-34 73 0 77-3 35-44 86-8 91-9 45-54 90-7 94-3 It will be seen that people have been marrying slightly later and there would also seem to be more people eventually marrying. If the trend towards later marriages continues then this may become an important factor which may decelerate the rate of population increase. Fertility :— This is the second important influence affecting the number of births in a given population and it can be estimated in several ways. The crude Fertility Ratio (proportion of children under 5 years of age to the number of women aged 1 5 to 4 4 years inclusive) is the simplest measure and is sufficiently accurate for most purposes. When estimated in this way, the fertility of Ceylon women has apparently decreased 114 H. OuXLUMBINE steadily over the past sixty years (Table 16) . Thematic- is, however, still greater than that found in Western countries, e.g. the British Isles. TABLE 16 Fertility Ratios (Unstandardized) in Ceylon. RACES Years All Ceylon Sinha­ lese ' Ceylon Tamil Indian Tamil Moors Burghers Eire United Kingdom 1881 882 923 r~ 6 14 1035 — 508 1871 580 . 1891 835 — — — 448 1901 817 913 . i- 0 79 936 • - 421 1901 460 1911 686 — — — — — 450 1921 648 703 541 493 677 — 460* . 427 | 1921 313 1946 594 525 511 644 624 473 460* . 427 | 1946 339 * 1926 t 1941 The reasons for the decline in fortuity are difficult to find. In the West it is known that urbanisation, improved education, increased incomes, etc., are factors which tend to reduce fertility by affecting the age of marriage, promoting birth control measures, and creating an awareness of social and personal responsibilities. These influences cannot, as yet, be accurately assessed, although it would appear that, as in Western countries, so in Ceylon the urban population has a lower fertility. This is true despite the smaller proportion of rural women who are in the reproductive age group (15-44 years), e.g. TABLE 17 The Differential fertilities of Urban and Rural population in Ceylon, 1946. Rural— Urban— Average Number of children born per mother 4 - 4 3 - 8 Average Number of children living per mother 3 2 2 - 9 Proportion of Proportion of women Female Population i n Reproductive Ago in Reproductive Group who had Age Group 5 0 - 2 per cent. 54" 3 per cent. borne children 5 9 "2 per cent. 4 9 - 0 per cent. It has not been possible to make more refined estimates of fertilities since the births of children by age of mother are not available. The 1946 Census, however, does give estimated reproduction rates for Ceylon, these being a Gross Reproduction Rate of 2- 3 0 2 per woman and a Net Reproduction Rate of 1 • 5 8 8 per woman. Again these rates are higher than those to be found among the population of England and Wales. SOME RECENT POPULATION TRENDS 115 Sea; Ratio :— TABLE 18 Ratio of Males and, Females at Various Age-Periods in the Population of Ceylon. Age All Ceylon 1891 1946 Sinhalese 1946 Ceylon Tamil 1946 Indian Tamil 1946 Ceylon Moors 1946 Birth 109 104 104 104 102 102 0-4 106 103 104 102 100 103 5-9 115 103 104 102 100 100 10-14 128 106 107 104 99 110 15-19 82 115 116 109 110 111 20-24 101 104 96 108 128 110 25-29 116 114 106 113 124 115 30-34 152 121 115 114 135 124 35-39 120 126 117 119 153 133 40-44 144 130 123 119 156 127 45-49 93 134 129 121 169 130 50-54 150 113 105 108 163 115 55-59 157 136 135 114 186 138 60-64 212 103 122 103 149 105 65-69 141 118 120 108 144 120 It is normal for the number of male births to be greater than the number of female births and, in Ceylon during the period 1937-45, there were 1 ,035 male live births to every 1,000 female live births. In most countries too, the male infant death rate exceeds that for females and, in Ceylon during the period 1937-45,1,144maleinfants died to every 1,000 deaths of female infants. In most countries in the West this higher male death rate persists throughout life but in Ceylon, after the first year of life, females have a consistently greater mortality rate. There has been a marked reduction in all the death rates in Ceylon since 1946 and this reduction has been proportionately similar for males and females in all age groups (Table 8 ) . The absolute reductions for females have been greater than those for males and the maternal mortality rate has been markedly reduced. If these trends continue then the proportion of females to males in tbe population may be in creased in the future and this, by providing more females in the reproductive ages, may produce a rise in the fertility of the population. Ceylon in common with most other Eastern countries has a preponderance of males in all age-groups of the population. This is true for all the main ethnic groups in Ceylon too (Table 18). In the countries of Western Europe, due to the devastations of war, the hazards of industrial occupation, better maternal care and to migration, females exceed males in the populations. 116 H. OULLTJMBIKB It is possible to conclude from this short analysis of the recent trends in the growth of the population of Ceylon : (1) Because estate expansion is now negligible and because of recent legislation, immigration will not play an important part in determining Ceylon's future population. (2) Since the beginning of this century, the increase in the size of the population has been due almost entirely to a rise in the number of births each year. This increase in births is caused by a rise in the number of marriages and also in the marriage rate. The process is a cumulative one. An increase marriage rate means increased births which cause an increased population. This in turn means even more people marry which produces still more births and so on. (3) There are evidences of limiting factors to this process. People are beginning to marry later and the fertility rate is declining. • (4) During the last three years there has been a sudden decline in the general death rate and this is another process which is going to accelerate the growth of the population. Not only has the general expectation of life increased but the infant mortality rate has been reduced so that more people will now live to the ages of marriage. In addition, the maternal mortality rate has also been decreased so that more mothers can live to have further babies. (5) As the trend is so recent in origin any population projections and predictions into the future would be mere guesses. It is important, however, to study in more detail the mortality experience in Ceylon. / CEYLON J. MED. SOT. (D) VOL. VII, P T . I l l and IV. (December, 1950). PART II The Mortality Pattern in Ceylon One method of comparing the distribution of the pathology of disease in different countries is to study the proportion of deaths from the principal causes. The proportion of deaths from principal causes in Ceylon, Eire, the United States of America and England and Wales for certain years have, therefore, been detailed in Table 19. TABLE 19 Proportions of Deaths from Principal Causes in 1948 and for the period 1937-1948 for Ceylon. PROPORTION PER 1,000 DEATHS Causes of Death Ceylon Eire U.S.A. England & Wales 1947 1937-48 1948 1947 1947 England & Wales 1947 All causes 1,000 1,000 1,000 1,000 1,000 Enteric 9-1 9-2 — 0-2 0 0 6 Influenza 14-4 12-1 1 5 0 5-3 — Dysentery 1 6 1 7-7 — 0-6 0-15 • Phthisis 25-8 40-2 04-0 3 1 0 38-3 Other forms of tuberculosis 3 1 3-2 2-0 2-5 6-45 Malaria & Malarial Cachexia 56-2 35-7 — 0 1 0-02 Ankylostomiasis 12-5 9-9 — — — Convulsions under 5 years 93-5 86-6 — — 0-67 Cancer 6-5 9-8 90-0 132-4 154-5 Diabetes Mellitus 5-3 5-3 5 0 26-2 6-84 Intracranial lesions of vascular origin 17-7 23-3 6 9 0 91-4 132-1 Diseases of the heart 18-3 29-6 252-0 321-4 280-0 Other diseases of the circulatory system 5-5 5-3 20 0 1 9 0 37-7 Bronchitis 12-0 12-6 30 0 — 6 1 1 Pneumonia and Broncho, pneumonia 73-1 86-4 4 2 0 37-8 .63-9 Diarrhoea and Enteritis 55-2 53-7 19-0 5-6 11-4 118 H. CULLUMBINE PROPORTION PER 1,000 DEATHS Causes of Death Ceylon Eire 1947 U.S.A. 1947 England & Wales 1947 Causes of Death 1937-48 1948 Eire 1947 U.S.A. 1947 England & Wales 1947 Nephritis 12-7 13-7 25-0 56 0 24-5 Deaths during puerperium and child-birth 28-7 25-4 3 0 3-5 1-78 Diseases peculiar to first year of life 108 154 57-0 45-5 30-2 Senility 38-6 73-8 147 0 19-2 30-0 Pyrexia 91 0 4 6 0 — — — It is evident that tropical, agricultural Ceylon differs from Western countries with a more temperate climate in the following main aspects :— (a) The main causes of death in Ceylon are the parasitic and infective diseases, which still account for about 1 in every 8 deaths. (Eire 1 in 1 2 ; TJ.S.A. 1 in 25) . The main causes of death in Western countries are from diseases of the Cardiovascular system (with the accompanying or attendant dangers of vascular intracranial lesions and possibly nephritis), which there account, for one in three to one in two of all deaths (Ceylon 1 in 2 0 ) . These latter diseases occur mainly in the middle and later years of life and may possibly be more prevalent in Western countries because of the greater length of life there. (b) Similarly, cancer, another hazard of the later years of life and which accounts for 9 to 1 5 per cent, of all deaths in Western countries, produces less than 1 per cent, of the deaths in Ceylon. (c) By contrast the dangers to life during the first post-natal year are 3 to 4 times greater in Ceylon. (d) Maternal risk of death during the puerperium and child-birth is 8 to 1 5 times greater in Ceylon. (e) Other points to note are the similarity of the proportion of deaths due to phthisis and diabetes (except in U.S.A.) in all the countries, while bronchitis is more prevalent as a cause of death in Western countries and convulsions in children and pneumonia are commoner causes in Ceylon. (/) Eire (temperate climate and agricultural economy) exhibits a distribution of causes of mortality which tends to be intermediate between that of Ceylon (tropical and agricultural) and those shown by America and Britain (mainly temperate and more industrialized). (