The birth, still-birth, death and fertility rates in the coloured population of Antigua, West Indies, from 1857 to 1956

The birth, still-birth, death and fertility rates in the coloured population of Antigua, West Indies, from 1857 to 1956


582KB Sizes 0 Downloads 0 Views





1. January,











BY K. H.


M.A., M.D., D.T.M. ~; H. #

Senior Medical Officer, Antigua Antigua is a small island of 108 square miles situated in the Caribbean Sea, and is one of a chain of six known as the L e e w a r d Islands between Puerto Rico to the west, and G u a d a loupe, the W i n d w a r d Islands and Barbados to the south-east. The inhabitants are mainly of pure African descent, though a minority are of mixed blood. The white population has never during the period under review exceeded 5 per cent. of the total, and frequently was much less. The people are essentially rural, being engaged in growing and processing sugar, the production of cotton and in fishing. The only town contains about 20 per cent. of the total population. At present the inhabitants number 55,000. In 1861 the number was 36,400. It fell slowly until 1921 when it was 28,800, but has risen rapidly since the early 1930's, owing to the reduction in the infant mortality rate. The population is very largely a self-contained one, but emigration of young adults, mostly males, has taken place at times, e.g. to the Panama Canal Zone during the building of the canal in the first decade of this century, to the United States on various occasions, and at the present moment to the British Isles. As regards immigration the censuses show that between 5 per cent. and 10 per cent. of the population claim to have been born outside the Colony. An accurate calculation of the birth and death rates a m o n g the people of African descent in this island has not hitherto been made. As is the case with m a n y of the smaller West Indian islands, the official annual returns in Antigua have at different times been vitiated b y inaccuracies due to one or both of the following, the degree of error varying f r o m year to year, or f r o m decade to decade • (i) Failure to differentiate records of white persons from those of African descent. (ii) Failure to separate still-births from live births ; sometimes still-births were included among live births, sometimes among deaths, and frequently no indication was given as to where they were recorded. I t occurred to m e that it might be worth while to investigate all births and deaths that have taken place in Antigua over the last hundred years, in an effort to obtain some accurate figures. T h e island is very favourably placed for obtaining such information, because : A. I n Antigua a Births and Deaths Registration Ordinance was introduced in 1856, and, because of the island's small size and good internal communications, nearly all births and deaths came to be registered within a couple of decades. As regards deaths, no body could be buried without a burial order, which in turn could not be issued without the presentation of a doctor's or coroner's certificate of the cause of death. B. T h r o u g h o u t the period u n d e r discussion there has been a sufficiency of physicians trained in the British Isles or in Canada, and since 1880 there has been a midwifery service, * This paper is one of a series assisted by a grant from the Standing Advisory Committee for Medical Research in the British Caribbean, for which I wish to express my thanks.



which in so far as the n u m b e r of midwives is concerned, in m y opinion has been m o r e than sufficient for the island's needs right f r o m its inception (e.g. one midwife per 1,200 persons, or to each 3½ square miles of a fairly undulating open country). C. T h e Ordinance included still-births within its control ; a still-birth being defined as any child " who shall have arrived at a period of viability." D. Censuses were taken in the years 1861, 1871, 1881, 1891, 1911, 1921, and 1946. T h e y have probably been reasonably accurate in so far as n u m b e r s and sex are concerned, but there are b o u n d to be inaccuracies in such surveys of uneducated and rural populations who frequently have little or no idea of their ages. E. I have personally examined all the parish registers f r o m •856 onwards and extracted all the relevant data about births, still-births and deaths that have occurred a m o n g persons of African descent, excluding similar material about white or other races. PART I BIRTH RATE AND FERTILITY RATE TABLE 1. Antigua, 1857-1956.

Birth rates and ratio of male to female live births, in decades. England and Wales live birth birth rate*

Per 1,000 living Live birth rate

Total birth rate (live and still-births)

Males per 1,000 female live births



1857-66 1867-76

30.8 39.5

35.1 45.2

1067 1064

1851-60 1861-70

34.2 35.2

1877-86 1887-96

37.8 35.0

42.3 38.3

1053 1008

1871-80 1881-90

35.4 32.5

1897-1906 1907-16

34.4 34.7

37.4 37.7

1046 1039

1891-1900 1901-10

29.9 27.3

1917-26 1927-36

34.3 30.9

36.9 32.8

1053 1033

1911-20 1921-30

21.3 18.2

1937-46 1947-56

34.5 32.8

36.2 33.9

990 1006

1931-40 1941-50

14.9 17.0




Live the M F T

births for century 62,175 60,139 122,314

Total stillblahs 10,838

Average for the century 1034

* From Enc. Brit. (1953) and from M.O.H. (1956). T h e figures for the first decade after registration was introduced are incomplete. N o census was taken between 1921 and 1946, so that the birth rates given above for the decade 1927-1936 m a y be too low, possibly due to emigration during that decade, though I have no direct evidence for this. T h e rate for the last decade is correct, but is unbalanced



as compared with the previous ones, because of the large proportion of infants and children in the population, occasioned by the impact of modern health and clinical measures on the infant mortality since the late 1930's. T h e current Antigua birth rate of around 33 corresponds with that for England and Wales in the 1880's, when it was 32.5, but there has been no drop in Antigua to correspond with that which occurred in England and Wales during the last 75 years, when it has fallen to less than half of its former figure.

Fertility rate T h e fertility rate is a more accurate criterion of a race's reproducive power than is tile crude birth rate, since it takes into consideration only the females of child-bearing age. The following is the fertility rate for Antigua : TABLE II. Antigua, 1857-1956. Fertility rate, or births, per 1,000 females aged 15-44 years.


Births per 1,000 females aged 15-44 years

England and Wales, births per 1,000 females aged 15-44 years*

1857-86 1867-76

140 172

149 135

1877-86 1887-96

168 146

147 128

1897-1906 1907-16

146 143

115 98

1917-26 1927-36

140 128t

79 64

1937-46 1947-56

143 144

67 76

* From data kindly supplied by the Chief Medical Officer to the Ministry of Health, London, 1958. t This is an estimate, because there was no census between 1921 and 1946. The figures suggest that the fertility rate, which was around 170 per 1,000 adult females during the third quarter of the 19th century has remained remarkably stationary at about 140-145 since the late 1880's, whereas that for England and Wales began by being about 150 and fell steadily to 64 in the 1930's since when it has risen. Whether this rise is a temporary one or not, it is too early to say. There is no suggestion that in Antigua a drop in the fertility rate has started. Social conditions are usually considered to have a very important influence on the fertility rate of a c o m m u n i t y - - t h e fall in the British rate coincides with the great improvement in the people's conditions of living that has taken place over the last hundred years. There is no doubt that conditions have improved in Antigua (as in the West Indies generally) over the last century, but certainly not to the extent that has occurred in Europe. The people are essentially a rural population with no outside interests after the day's work is done. It is therefore not surprising that the rate has not yet begun to fall.



Sex-ratio of live births The average ratio over the century of male to female live births has been 1034 in a series of 122,314 consecutive live births (Table 1). There is a swing of about 3 per cent. or 4 per cent. on either side of the mean, with a fall during the last two decades. This ratio, which is lower than that for England and Wales where it is around 1060, associated as it is with a higher mortality in male than in female children, has resulted in a larger number of females in the population than males, especially in adult life, where the proportion is 754 males to 1,000 females. No doubt the emigration of young adult males plays a certain part in this unbalance. A female preponderance is met with in many of the West Indian Islands. SUMMARYAND CONCLUSIONS I) A survey has been carried out of all births registered in Antigua since 1857, the first completeyear of the operation of the Births and Deaths Ordinance, in which the records of all births in the coloured population have been scrutinized. 2) There were 62,175 male live births and 60,139 female live births, with 10,838 stillbirths. 3) The live birth rate has varied between 30.8 and 39.5 per 1,000 living ; during the last 80 years it has been falHy steady at about 34. The total birth rate (for live and stillbirths) has been between 32.8 and 45.2. 4) There has been no fall in the birth rate in Antigua, such as has occurred in England and Wales during the last 75 years. 5) The ratio of male to 1,000 femalelive births has averaged 1034 with a fall during the last two decades to around parity. 6) The fertility rate (births per 1,000 females aged 15-44 years), was around 170 during the first 30 years of the century under review, but it has been steady at 140-146 for the last 70 years. There is no suggestion that this rate will fall in the near future, as it has fallen in England and Wales since the 1870's. REFERENCES Encyclopaedia BHtannica, 1953 Ed. (Article on Birth rate). Report of the Ministry of Healthfor 1956. Part 2; On the state of the Public Health. p. 227. London: H.M.S.O. PART II THE STILL-BIRTH RATE Following the introduction of the Births and Deaths Ordinance in Antigua in 1856, the registration of all still-births was made compulsory. The definition of a still-birth was a child " who shall have arrived at a period of viability." The following Table shows the still-birth rate over the century, together with the ratio of males to females among the still-born infants. The figure of 122 for the decade 1857-66 is almost certainly lower than the true value, because it is unlikely that registration became complete for many years. There was a steady fall in the rate for each decade until the current figure of 32 was reached. This is higher than the figure for England and Wales of 23.7 for the decade 1947-56, but compares favourably with 42.7, the England and Wales figure for 1928-36.


Antigua, 1857-1956.


The still-birth rate in decades and the ratio of males to females among the still-born. England and Wales



Average annual [ Still-born males per 1,000 still-born still-birth rate, per females 1,000 live and stillbirths

Still-birth rate

Males per 1,000 females*

1857-66 1867-76

122 125

1179 1174

1877-86 1887-96

108 98

1253 1232

1897-1906 1907-16

79 80

1113 1257

1917-26 1927-36

69 58

1185 1052

1928-36t 42.7



1937-46 1947-56

47 32

1103 1156

1937-46 1947-56





Average for the century Total number of still-births

34.0 23.7


* Sutherland (1949) p. 23. t 1928 was the first year in which still-births were registered m England and Wales. The figures are compiled from data kindly supplied by the Chief Medical Officer, Ministry of Health, London. T h e sex ratio of still births is about 1180, with a variation of up to 10 per cent. on either side of that figure ; it does not show any tendency to fall as has that for England and Wales, which fell from 1243 in 1928 to 1156 in 1940-45.

Illegitimacy and the still-birth rate An analysis of the ratio of the legitimate to the illegitimate still-births shows that 27.5 per cent. occur in married women. T h e census figures for the last census in 1946 show that 26.9 per cent. of females over the age of 15 were married ; the proportion of married to unmarried women aged 15 and over, therefore, was 1 to 2 . 7 and the ratio of legitimate to illegitimate still-births is 1 to 2.6, so that it is reasonable to conclude that illegitimacy as such plays little part in the still-birth rate. T h e census figures for 1946 also show that as regards the Leeward Islands generally, the average n u m b e r of children born per married woman of post-fertile age, i.e. of 55 years and over, was 5.98, but that the n u m b e r born to unmarried women of the same age-group was 4 . 5 4 (LAMPE). ( T h e figures are not as straightforward as they might appear, because many a woman does not marry until she has had several children). I am not in a position to argue from these figures as regards what influence these two variables--(i) marriage and (ii) n o n - m a r r i a g e - - m a y have on the stillbirth rate, but if, as the figures in the earlier part of this paragraph suggest, illegitimacy as such does in fact play little part in the still-birth rate, this is contrary to what happens in Great Britain where the unmarried mother is at a great social disadvantage, but it is what



one would expect in the West Indies. The explanation for this difference requires a short digression into the subject of the West Indian family and its mores. The unmarried mother outnumbers the married mother by two to one, and there is no social or other stigma associated with this state of affairs. Although the older woman who has had several pregnancies would often prefer the married status, the young teenage girl embarking on her first pregnancies usually does not desire the legal implications of marriage because she considers that if she is not legally tied to her " boy friend " and can leave him if he does not treat her well, then he is more likely to be a good partner ; if she is married to him, her freedom of action is considerably restricted. It is not until the mother is much older and has had several children by perhaps several boy friends, that she comes to change her view-point, but by then her value on the marriage market is very much reduced. As far as the young man is concerned, he normally does not feel any obligation to marry his girl friend and to accept what he considers to be unnecessary responsibilities. In a minority, marriage between the two partners is the ultimate object, but weddings are expensive affairs and it may take many years before enough money is saved. In the meanwhile the children arrive. Other factors affecting the still-birth rate

There is no local information about the causes of still-births in Antigua, either as regards the average interval between successive pregnancies, or as to the amount of maternal employment outside the home during pregnancy. It is well known that age and parity of the mother influence the rate, that it is higher in the young primipara than in the multipara of the same age, and that after the middle 20's, the rate increases in both groups of women. In Antigua where fertility is high, early teenage pregnancy is general, and birth control unknown, it would be reasonable to assume the fact that most primiparae are in their teens would be an important contributory factor to the high still-birth rate, though no statistical evidence has been collected to support this view. In any case, however, such a view would not account for the fall in the still-birth rate over the century, because there has been no decrease in the number of very young primiparae. It is only in recent years that a middle class with its smaller size of family and its later average age of marriage has begun to appear "in Antigua, and in any case it is still too small a group to have any noticeable effect on the data. CONCLUSIONS

1) In Antigua the still-birth rate has fallen steadily to one-quarter of what it was a century ago, and is now around 32 per 1,000 total births. The figure for England and Wales is about three-quarters of this. 2) The ratio of male per 1,000 female still-births over the century in Antigua has varied within 10 per cent. on either side of 1181. The figure fell to 1052 in the early 1930's, but has risen somewhat over the last two decades and is now 1155. 3) In Antigua where two-thirds of those born are illegitimate, and where there is no stigma associated with illegitimacy, there is no significant difference between the still-birth rate of the legitimate and the illegitimate infant. 4) The fall in the still-birth rate began long before modern techniques of public health and treatment began to be applied to the mother. REFERENCES SUTHERLAND,I. (1949). Still-births, their Epidemiology and Social Significance. London: LAMPE, P. H. I- (Undated). A study of Human Fertility in the British Caribbean Colonies. Caribbean Commission. Port-of-Spain, Trinidad.


K. H. U T T L E Y


Crude death rate T h e following is a statement of the crude death rate, given in 10-year averages : TABLE I.

Antigua, 1857-1956.

Crude death rate, averages for 10-year periods.


Crude death rate

1857-66 1867-76 1877-86

36.8 37.2 35.5

1887-96 1897-1906 1907-16

31.8 30.7 27.1

1917-26 1927-36 1937-46

27.1 18.3 17.7

1947-56 1957-58

12.4 9.4

Standardized death rates have been calculated for each census year from 1861, using the England and Wales 1901 population as the standard population. Table I I shows the results : TABLE II.

Antigua, 1857-1956. Standardized death rates for census years*, with the Crude Death Rate, and the England and Wales S.D.R. for comparison.

Antigua standardized death rate

Crude D.R.





1861 1871

38.5 42.4

31.7 39.5

35.0 40.9


1881 1891

37.0 32.8

33.8 37.9

35.3 35.4

37.0 33.9

1911 1921

27.2 30.2

30.3 30.0

28.8 30.1

28.4 29.9







England and Wales standardized death rate* M






21.1 (for 1851 onlyt) 20.3

1881-90 1891-1900 1901-10 1911-20 1921-30

18.6 18.1 15.2 13.5 10.6 7.6

* Standardized against the 1901 population of England and Wales. den, p. 228.


6.7 (for 1951 onlyt)

~"From Taylor and Knowel-

I n 1946 the S.D.R. had fallen in Antigua to 34 per cent. of the 1871 value, whereas in 1951 the figure for England and Wales had fallen to 32 per cent. of a m u c h lower value in 1851. It has not been possible to calculate the S.D.R. for any year later than 1946, but because the C.D.R. has fallen in the interim to around 9 per 1,000, it is reasonable to infer that the



S.D.R. has fallen correspondingly, and that at present the S.D.R. for Antigua, like the C.D.R., must be around one-quarter of what it was in 1871.

Age and sex specific death rates TABLE III.



Antigua, 1857-1956.

1861 M F

1871 M F

84 151 9 13

Death rate per 1,000 at ages and by sex at census years.

1881 M F

154 158 8 8

132 104 12 9

England and Wales, 1951'* M F

1911 M F

1921 M F

1946 M F

124 7

96 3

86 6

33 3

31 1

7.4 0.6

5.7 0.4

1891 M F

108 7

82 8


113 10


13 24

14 26

10 23

15 18

11 19







13 17









1.1 1.6

0.8 1.3


26 45

25 35

26 43

27 31

26 33

16 25

30 36

27 24

16 23

19 18

19 34

18 19

11 23

9 12

2.9 8.6

2.3 5.3


86 98

45 64 59 109

58 88

43 56

41 57

55 95

47 91

34 99

49 59 100 127

48 106

43 68

17 50

24.1 59.1

13.1 37.0



159 186



181 150

175 149

192 191







141 186





136 155.0



* From data in M.O.H. (1951) p. 200. T h e infant mortality will be discussed in a later paper ; apart from this the Table shows that Antiguan experience is similar to that of the world in general, namely, that the male mortality at all ages, and with few exceptions, is greater than that of females. Although Antigua is considered a healthy island, a glance at the figures shows that at no age, except in extreme old age, can the rates bear comparison with corresponding figures in England and Wales. I f we now take the age and sex specific death rates in 1871 as being 100 for each group and express the corresponding figures for 1946 as percentages of the 1871 figures, we have the following : 1946 Age group

TABLE IV. Antigua. Age and sex specific death rates for 1946 expressed as a percentage of those for 1871.



22 23

20 13

15 25

50 43

33 28

35 45

42 53

33 39

55 m 65

67 62

29 57

75 -






T h e Table shows the outstanding improvement that has come about in the infant and childhood rate over the period. Allowing for the smallness of the figures, we can state : (i) After infancy and until late middle life the improvement in the rate for females is one and a half times to twice that for males. (ii) No change has occurred in the death rate for the elderly. T h e view is held by many epidemiologists that, where comparisons over a long term of years are being made, a very useful standardization is that of the equivalent average death rate, which is the mean of the death rates for 13 5-year groups from 0 to 64 years of age. It is convenient because it eliminates the distortion caused by the falling off in numbers of the population aged 65 years and over. T h e E.A.D.R. for the census years in Antigua over the last century is given b e l o w : "I~ABLEV.


Equivalent average death rate, for the census years 1861o1946.





1861 1871

40 3q

30 36

33 38

1881 1891

34 31

29 30

31 31

1911 1921

24 29

26 24

25 27








England and Wales, 1951" * Taylor and Knowelden, p. 287.

Except in 1911, the male rate has exceeded that for females, sometimes, as in 1946, by a considerable margin. Over the century the rate for females has fallen irregularly, but in the last 50 years it has fallen faster than has that for males. Although a considerable fall has occurred, a glance at the corresponding figures for England and Wales shows what a long way Antigua has to go before the rate becomes as good as in the latter country. SUMMARY AND CONCLUSIONS

1) T h e crude death rate has been worked out for the coloured population of Antigua for the years 1867-1958. Its highest value was in the decade from 1867 to 1876 when it was 37.2 ; it fell very slowly to 27.1 in the decade 1917-1926, but since then it has fallen rapidly and in 1958 it was 9.4. 2) T h e standardized death rate for each sex and for persons, using the England and Wales 1901 population structure as a basis, has also been calculated for each of the seven census years between 1861 and 1946. It follows fairly closely the crude rate, the highest figure was 40.9 in 1871, and the lowest 14.8 in 1946. 3) Death rates per 1,000 at ages have also been calculated for each census year, and the drop in the mortality in 1946 has been expressed as a percentage of the 1871 figures. 4) Male mortality in general was somewhat higher than that for females in each age group.



5) The equivalent average death rate for males, females and persons, calculated for census years, shows that the fall in the mortality for females under 65 years of age has been greater than that for males of the same age, and that that for persons is still nearly three times higher than that for England and Wales. REFERENCES TAYLOR, I. & KNOWELDEN,J. (1949). Principles of Epidemiology. London : Report of the Ministry of Health, 1951. Part II. On the State of the Public Health, p. 200. London : H.M.S.O.