- Email: [email protected]
Causes of infant and early childhood deaths in Sierra Leone
027-I-9536186 53.00 + 0.00 Pcrgamon Journals Ltd
Sot. Sci. Med. Vol. 23, No. 3, pp. 297-303. 1986 Printed in Great Britain
CAUSES OF INFANT AND EARLY CHILDHOOD DEATHS IN SIERRA LEONE ERROR
SAMA KANDEH
Department of Geography and Environmental Studies, Njala University College, University of Sierra Leone, P.M.B., Freetown, Sierra Leone Abstract-Information about the causes of infant and early childhood (1-4 years) mortality was compiled from the vital registration system for the Western Area, records from the children’s Hospital in Freetown and two demographic sample surveys conducted in various chiefdom headquarters towns around the country. The leading causes of infant mortality are tetanus, fevers, measles and diarrhoea. A breakdown of certified deaths in infancy showed that tetanus is quite important in the neonatal period accounting for as much as 68% of neonatal deaths. Measles and diarrhoea were the leading causes of death in the last 6 months of infancy. The leading causes of early childhood deaths were measles, diarrhoea and fevers. Nutritionally related diseases such as measles and diarrhoea were seen to account for up to 40% of all early childhood deaths. The major factors affecting these causes of death were childbirth and childcare practices in the case of tetanus and the nutritional status of the children in the case of measles and diarrhoea. The majority of deliveries were still being performed by Traditional Birth Attendants in very unhygienic surroundings which it was felt contributed significantly to the high incidence of neonatal tetanus. At the other childhood ages the poor nutritional status of the majority of children in Sierra Leone as shown by the results of the 1978 National Nutrition Survey was seen as the significant factor. The effects of the identified major causes of infant and early childhood mortality (tetanus, fevers, measles, and diarrhoea) can be largely diminished by effective intervention programmes such as oral rehydration therapy and the training of Traditional Birth Attendants.
INTRODUCITON
fatality rate of British colonial officers from yellow
Sierra Leone is a small country of about 28,000 square miles which lies on the South West coast of West Africa between latitudes 6.55’, and 10.0” N and between longitudes 10.16” and 13.18”W [l]. Sierra Leone has four major administrative units, the Western Area, the Southern Province, the Eastern Province and the Northern Province. Each of the three provinces is divided into districts which total twelve and which are in turn subdivided into chiefdoms which number 148. The Western Area is also divided into four very small rural districts and the city of Freetown but its local government patterns are very different from that seen in the three provinces. Sierra Leone’s demographic picture is gradually falling into place with the information provided by the last two censuses (1963 and 1974) which for the first time covered the entire country. The 1974 census results showed a total population of 2,735,154 persons. This was later adjusted to take into consideration an estimated 8.9% underenumeration yielding a new total of roughly 3,002,426 persons. When the unadjusted population totals for 1963 (2,180,355) and 1974 (2,735,159) are compared, the resulting average annual intercensal growth rate is 1.9%. The age structure of the population is youthful with over 35% of the population < 15 years of age, Only 5% of the population is 65 years and over in age. The estimated crude birth rate for the country is 48.7 based on the 1974 census results. Mortality levels are quite high with an estimated crude death rate of 28 and an infant mortality rate of 240 based on the 1974 census results. High mortality conditions have been recorded for Sierra Leone since the start of the colonial era (1807) when Sierra Leone was referred to as- the ‘White Man’s Grave’ because of the high
fever [2]* METHODOLOGY
Information about the major causes of death in Sierra Leone is available from different sources. These sources include Government hospitals in the Western Area where the main hospitals for mothers and children are located (Princess Christian Maternity Hospital and Children’s Hospital), provincial hospitals and private hospitals such as the Catholic Hospital at Serabu, the Nixon Memorial Hospital at Segbwema and the Netland Nursing Home in Freetown. This hospital based information is available because collection procedures are more rigorous in the above mentioned hospitals than in other hospitals. Most of this information eventually ends up in the Medical Statistics Unit of the Ministry of Health where it is compiled into annual reports. Even then the reports are largely for the Western Area. One of the limitations of these hospital-based reports of causes of death is that they do not accurately reflect the distribution of causes of death on the national scene because of the low user rates of hospitals as well as the probability that only very serious cases are usually taken to hospitals. The value of the hospital based data is that over half of the causes of death are certified by a Medical OfBcer. The distribution of causes of death for these certified deaths can then be compared with that for non-certified deaths. More representative information about the major causes of death is provided by reports of causes of death by parents in several sample surveys. This is because such surveys have contacted all households in an area or a representative sample of households.
297
298
BARBER SW
Table I. Principal certified causes of infant deaths in the Western area, I%+1979 (pooled figure) Rank
1 : 4 5 6 7 : 10
% Of all infant deaths
Cause of death Tetanus Pneumonia Diarrhoeal diMeasles Malaria Anaemia Malnutrition Accidents PulmOtUry tuberculosis Other infectious and parasitic diseases
Total Number
17.01 14.64 10.22 5.76 5.42 2.13 1.67 0.51 0.36 42.28 100.00 4186
KANDEH
Jimmi, Bumpe and Njala Komboya).‘It was aasingle round retrospective survey which interviewed 2233 currently married women between the ages of 15-49 years. It was carried out between February and July 1977. (d) A National Mortality Survey which was carried out in ten chiefdom headquarter towns. The five chiefdoms with the highest infant mortality rates and the five with the lowest infant mortality rates, based on the 1974 census results were selected. A total of 1464 households, 1820 women (12-49 years) and 9770 household members were contacted in a single round retrospective survey. The field survey was carried out in early 1980.
Source: Adapted from Table 3. Horton [3, p. 811.
INFAM
that such surveys include households which make use of medical facilities as well as those who do not. The problem with data from this source is that the information is often collected by persons who do not have any training in the certification of deaths and they therefore rely exclusively on retrospective reports and descriptions provided by parents. One advantage of this data source in that is provides information for other parts of the country outside the Western Area. These two data sources should therefore complement each other and provide one with the opportunity of understanding what are the major causes of death in Sierra Leone. In the analysis of the causes of death that follows, the causes of deaths will be presented according to the data source although in the end the discussion of these causes will be integrated. The data for this paper has been drawn largely from: (a) The Vital Registration system in the Western Area. (b) Reports from the Ministry of Health. (c) An Infant and Child Mortality Survey in Bo District. This survey was restricted to the District Headquarter Bo and four other towns (Yamandu, So
MORTALITY
Information on the principal certified causes of infant deaths in the Western Area for the period 1969-1979 [3] is presented in Table 1. According to the information provided in Table 1, the five leading causes of infant deaths in 1969-1979 were tetanus (17.01%), pneumonia (14.64%), diarrhoeal diseases (10.22%), measles (5.76%) and malaria (5.42%). One must, however, take into consideration the fact that other infectious and parasitic diseases account for 42.28% of all infant deaths. So that as a group, infectious and parasitic diseases account for over 50% of all infant deaths when one takes into consideration the fact that a disease like tetanus is an infectious disease. An opportunity to get some insight into what these other infectious and parasitic diseases are, is provided in Table 2 where the certified causes of infant deaths are distributed by cause and age at death for the same Western Area but only for the period 1969-1971. A total of 3783 infant deaths were recorded in the Western Area during this period but only 1772 (46.8%) were medically certified and it is these medically certified infant deaths that are presented in Table 2. From the information provided in Table 2, it can be. seen that tetanus is the leading cause of infant deaths accounting for 25.7% of all certified infant deaths. The results further show that deaths due to
Table 2. Percentage distribution of certified infant deaths by age and cause for the Western area, 1969-1971 (pooled figure) Age at death A
B
C
D
E
F
Pneumonias Tetanus
25.7 15.6
18.0 4.5
68.9 5.6
26.2 17.6
36.6 0.8
29.2 0.3
Anoxic and hypoxic conds Lliarrhoeal diseases Measles Birth Infection Acute respiratory infections Malaria Congenital anomalies Anaemias Septicaemia Avitarninosis Other p&natal causes All other causes
11.0
31.4 0.3
2.5 3.1
10.1
29.2 3.5
19.7 15.8
7.3 0.5
1.3 0.8 0.2 1.6
14.3 2.4 1.3
2.3 0.2 11.8 1.7 100.0 483 27.3
4.8 3.6 6.0 13.1 100.0 84 4.7
5.4 4.3 1.2 3.5 0.8 2.7
3.0 a.5 0.6 6.0 1.6 3.0
22. I 100.0 257 14.5
12.3 100.0 366 20.7
Cause
Rank : 3 4 5 6 7 8 9 10 II I2 13 14 Total Number %
9.8 3.8 2.7 2.5 2.5 2.4 1.9 I.5 1.2 13.4 6.0 100.0 1772 100.0
4.8 0.3 0.7 30. I 2.1 loo.0 582 32.8
Key: A-total; B-under 1 we& C-l-3 weeks; W weeks to under 2 months; E-2-S Source: Adapted from Table 4 Sierra Leone Government [6, p. 61.
months; F-&l
I months.
Causes
of infant and early childhood deaths
Table 3. The percentage distribution of reported causes of infant deaths (Bo town and other towns*) (pooled fig&e 1950-1977) Cause
Fever related Tetanus Convulsions Diarrhoea Pneumonia Measles Other causes Total Number
Bo town
21.6 20.7 18.3 8.0 3.0 7.4 IS.0 100.0 517
(1) (2) (3) (4) (5) (6)
Other towns* 18.6 45.8 3.5 4.8 1.9 6.8 18.5 99.9 550
(2) (1) (5) (4) (6) (3)
*Other towns: Jimmi, Bumpe. Yamandu and Njala Komboya. ( ) Relative ranking. Source: Adapted from Kandeh [4, Tables 3.9 and 3.101.
tetanus were concentrated in the l-3 week period where they accounted for 68.9% of all infant deaths and that is in fact the only period in infancy when tetanus is the leading cause of infant deaths. After the second month of life the importance of tetanus is drastically reduced for the remaining period of infancy. The second leading cause of infant deaths is accounting for 15.6% of all infant pneumonia, deaths. It becomes the leading cause of death in the 2-l 1 month period when the importance of neonatal tetanus has been reduced. The third leading cause of infant deaths are the anoxic and hypoxic conditions of the new born, accounting for 11.0% of all infant deaths and restricted to only the first 3 weeks of life. Diarrhoeal diseases are the fourth leading cause of infant deaths (9.8%) while measles is fifth (3.8%). When the five leading causes of infant deaths for 1969-1971 are compared with those for 1969-1979, tetanus and pneumonia maintain their positions as the leading and the second leading cause of infant deaths, but the anoxic and hypoxic conditions which are the third leading cause of infant deaths in 1969-1971 are not listed for the 1969-1979 data. They have probably been subsumed under the other infectious and parasitic diseases category. Diarrrhoeal diseases still take precedence over measles in both tables although their rank order is higher for the 1969-1971 data (Table 2) than for the 1969-1979 data (Table 1). When the distribution of infant deaths in the various segments of infancy is taken into consideration (Table 2), one observes that 64.84% of all certified infant deaths occurred in the first three weeks of life again underscoring the importance of neonatal tetanus, anoxic and hypoxic conditions of the new born, other perinatal causes and pneumonia. Of these four causes tetanus and pneumonia are more likely to be reduced by intervention programmes aimed at reducing infant mortality levels. For the remaining 35.6% of infant deaths in the 2-l 1 month period, again pneumonia, diarrhoeal diseases and measles are the important causes that intervention programmes should be aimed at for the reduction of infant mortality levels. Based on this one could therefore conclude that the five leading causes of infant deaths based on certified reports are tetanus, anoxic and hypoxic conditions, measles, pyrexia and diarrhoeal diseases.
299
Now turning to noncertified reports of infant deaths, the survey of infant and child mortality in Bo district carried out in 1977 obtained retrospective reports of causes of infant deaths from mothers interviewed [4]. The results have been separated for the town of Bo (a provincial headquarter town and very urban in outlook) and other towns; Jimmi, Bumpe, Yamandu and Njala Komboya (these are chiefdom headquarters and are more rural in outlook). The results are shown in Table 3. The causes of infant deaths shown in Table 3 appear to be similar for both Bo town and the other towns except that whereas tetanus is the leading cause of infant deaths in the other towns (45.8%), it is only the second leading cause of infant deaths in the town of Bo (20.7%). This can not be attributed entirely to the presence of major medical facilities within the town of Bo. It is probably due to a wrong diagnosis of causes of death as the Mende words for tetanus and convulsions are similar. Notice that convulsions have a very high percentage (18.3%) and that the symptoms are similar to those for tetanus. Also up to now convulsions have not emerged as a major cause of infant deaths. So it is likely that in both areas, tetanus is the leading cause of infant deaths even though one would expect a higher percentage of neonatal tetanus deaths in the other towns. The second leading cause of infant deaths shown in Table 3 for both areas are the fever related diseases which could include all diseases that exhibit fevers as a symptom such as malaria, yellow fever, etc. A comparative mortality survey of two contrasting mortality areas was carried out in 1980 in ten chiefdom headquarters [5]. Five of the headquarters were from the chiefdoms with the highest infant mortality rates according to the results of the 1974 census and the other five were from the chiefdoms with the lowest infant mortality rates. The chiefdom headquarters in the high mortality area were Fairo, Zimmi, Futa, Banta Mokelle and Njala Komboya all in the Southern Province of Sierra Leone. Those in the low mortality area were Mahera, Batkanu, Mateboi, Musaia and Falaba, all in the Northern Province. Birth histories from mothers 15-49 years of age were used to compile the causes of death presented in Table 4. The results show that tetanus is the leading cause of infant deaths in both areas although the incidence is somewhat higher in the high mortality area. Fever related diseases are again the second leading cause of infant deaths. Measles and
Table 4. Leading causes of infant deaths (pooled figure 1940-1980) High mortality Rank : 3 4 5
Cause Tetanus Fevers Headpain Measles Diarrh. Stomac. Others
Total Key: diarrh.diarrhoea;
Low mortality
N
%
197 165 97 57 7 7 73 603
32.7 27.4 16.1 9.4
1.2 1.2 12.0 100.0
Cause
N
0%
Tetanus Fevers Measles Diarrh. Stomac.
42 36 29 14 8
25.8 22. I 17.8 8.6 4.9
34 163
20.8 100.0
Others
stomac.-stomach headache; N = number of deaths. Source: Kandeh and Dow [A.
ache; head+pain/
BORBORS.4hf.4KANDEH
300
Table 7. The percentage distribution of the most common causes of adminission and death at children’s hospital, Freetown
Table 5. Principal certified causes of early childhood (I-4 years) deaths in the Western area, 1969-1979 (pooled figure) Rank
QUX
%
Measles Diarrhoeal diseases Pneumonia Malnutrition Malaria Anaemia Pulmonary tuberculosis Accidents Tetanus Other infectious and parasitic diseases
I 2
3 4 S 6 : 9 IO Total Number
1980 Cause
14.1. 14.1. 13.2 7.0. 5.6 4.6’ I.4 I.2 0.6
Bronco-pneumonia Measles Diarrhoeal diseases Neonatal tetanus Rot. cal. malnutr. Meningitis Malaria Anacmia Others Total
38.2 100.0 3825
B
A
B
16.0 (2) 21.7 (I) 14.1 (3) 5.1 (5) 6.5 (4) 3.5 (8) 4. I (7) 4.1 (6) 24.9 100.0
10.2 (3) 28.0 (I) 4.9 (2) 12.0 (4) 9.2 (5) 4.5 (6) 3.8 (7) 3.6 (8) 23.8 100.0
15.2 (2) 10.3 (3) 17.6 (I, 5.4 (6) 6.1 (5) 6.3 (4) 3.3 (8) 4.7 (7) 31.1 100.0
6.4 (6) 13.7 (3) 20.2 (1) 14.9 (2) 8.5 (4) 7.3 (5) I.8 (8) 3.8 (7) 23.4 100.0
Numbers in parentheses indicate the relative ranking of diseases. Key: A-% of all admissions; B-% of all dmths. Source: M. L. Wright (undated).
*Nutritionally related diseases cause a total of 39.8%. Source: Adapted from Horton (3, p. 261, Table 31.
Another interesting aspect of Table 5 is that the nutritionally related diseases (measles, diarrhoea, malnutrition and anaemia) together account for 39.8% of all early childhood deaths. This underscores the importance of nutritional starus as a major contributor to the health of children in these early years. Poor nutritional status it has been suggested inhibits the formation of antibodies which provide a natural protection against infectious and parasitic diseases. The Results in Table 6 [6] show the same pattern to that already described in Table 5 but specifically for the years 1974-75 and 1975-1976. Here also measles, diarrhoea and pneumonia are the leading causes of early childhood deaths. Another report of certified causes of early childhood deaths is provided for the Western Area by Wright [I. This report lists the common causes of admissions and deaths at Children’s Hospital in Freetown for the period 1980-1981 (Table 7). In terms of the percentage of admissions, measles is the leading cause in 1980 and is replaced by diarrhoeal diseases in 1981. The leading cause of death for these admissions in 1980 is measles and in 1981 diarrhoea. So here again the nutritionally related diseases (measles and diarrhoea) are the leading causes of death for the admissions into Children’s Hospital in Freetown. Included among the common causes in Table 7 are protein calorie malnutrition and anaemia which are also nutritionally related.
diarrhoea are also important but their importance is diminished in the high mortality area where pain and severe headaches are the third leading cause of infant deaths. It is probable that this pain and severe headache could really be symptoms of either tetanus or fever related diseases especially since one is looking at infants. A combination of the results of the certified and non-certified causes of infant deaths suggests that the important causes of infant deaths are tetanus, diarrhoeal diseases and measles in order of decreasing importance. Furthermore it also appears that these causes of death are important in certain stages of infancy. Tetanus is important in the neonatal period while the diarrhoeal diseases and measles are important in the last 6 months of infancy. EARLY CHILDHOOD
1981
A
MORTALITY
The principal certified causes of early childhood deaths in the Western Area for the period 1969-1979 are presented in Table 5. The leading causes of early childhood deaths are measles and diarrhoeal diseases which each account for 14.1%. Pneumonia follows very closely as the third leading cause of death accounting for 13.2% while malnutrition is fourth (7.0%). These four leading causes of early childhood deaths together account for 48.4% of the deaths.
Table 6. Ten leading causes of certified early childhood deaths for the Western area for 1974-1976 causes I :
Measles* Influenza and pneumonia Diarrhoeal diseases* Avitaminosis, etc.’ Malaria Anaemias* Dysentery Accidents Tubercolosis Meningitis Other causes
4 5 6 7 8 9 10 II Total Number % of total contributed by nutritionallv related diseases
lNutritlonalIy
19741975 23.4 (I) 20.4 (2) 12.7 (3) 11.0(4) 4.6 (6) 10.2 (5) I.1 (9) I.1 (9) 2.2 (8) 2.0 (7) IO.5 100.0 498 (57.3%)
1975-1976 Table 8. The percentage distribution of reported causes of early childhood deaths (Bo town and other towns’) (waled figure 1950-1977)
21.1 (2) 24.4 (1) 13.6 (3) 9.4 (4) 4.2 (6) 9.2 (5) 0.5 (IO) 1.6 (5) 4.2 (6) 1.6 (8) IO.1 99.9 426
CCIUSC
Fever-related Tetanus Convulsions Diarrhoea Pneumonia Measles Other causes Total Number
(53.3%)
related disease.
( ) Ranking. Source: Sierra Leone Govermnent [6. Table AS. p. IS. 19761.
I
Bo town 26.3 (I) 6.4 (4) 6.1 (5) 14.7 (3) 2.3 (6) 18.5 (2) 19.0 93.3 373
Gther towns* 18.2 (2) 129 (3) 2.6 (5) 125 (4) 1.6 (6) 29.2 (1) 23.0 loo.0 287
*Other towns: Jimmi. Bumpe, Yamandu and Njala Komboya. Source: Adapted from Kandeh 14, Tabks 3.9 and ,.lO].
Causes
of infant and early childhood deaths
Table 9. Leading causes of early childhood deaths (pooled figure 1940-1980) High mortality Rank
I : 4 5 Total
Low mortality
Cause
N
%
Fevers Measles Tetanus Diarrh. Headpain Others
40 30 IS 6 3 30 124
32.2 24.2 12.1 4.8 2.4 24.2 99.9
Key: diarrh.--dianhoca; stomac-stomach headache; N = number of deaths. Source: Kandeh and Dow [5].
Cause
N
%
Measles Fevers Diarrh. Stomac. Tetamis Others
38 20 IS 8 4 21 106
35.8 18.9 14.1 ::: 19.8 99.9
ache; headpain-pain/
From the information provided in Tables 5, 6 and 7 one can conclude that based on certified causes of death in hospitals, the five leading causes of early childhood deaths are measles, pneumonia, diarrhoea, malnutrition and anaemia. Furthermore four of these five leading causes are nutritionally related. The leading causes of early childhood deaths for the survey carried out in Bo district in 1977 are shown in Table 8. Looking at all childhood deaths in Bo town, the leading cause of death are the fever related diseases such as malaria, accounting for 26.3%, followed by measles (18.5%) and diarrhoea (14.7%). In the other towns the leading cause of all early childhood deaths is measles (29.2%), followed by the fever related diseases (18.2%), tetanus (12.9%) and diarrhoea (12.5%). It is worth noting that not only is measles the leading cause of early childhood deaths in the other towns but that the percentage for the other towns is almost double that for the town of Bo. This is probably due to the rural nature of the other towns, so that the nutritional status of the children in these other towns would be relatively lower compared to Bo town. Furthermore the presence of superior medical facilities in Bo town would reduce the proportion of children dying from an attack of measles, while also reducing the incidence of measles through immunization. The results from the comparative mortality survey (Table 9) also show measles and fever as the leading causes of early childhood deaths with measles dominating in the low mortality zone and fevers dominating in the high mortality zone. From the information in Tables 8 and 9 one can conclude that the leading causes of early childhood deaths from non-certified reports are fever related diseases, measles and diarrhoea, Here again the importance of the nutritionally related diseases as significant contributors to early childhood deaths is shown. A combination of the results from the certified and non-certified causes of early childhood deaths suggests that the four leading causes of death are measles, diarrhoea, pneumonia, and fever related diseases. However of these four, measles and diarrhoea are consistently among the leading causes of early childhood deaths whether one is looking at certified or non-certified causes of early childhood deaths. Another consistent pattern from both sources is the high percentage of early childhood deaths caused by nutritionally related diseases, which in the case of certified causes of death is at least 40% and
301
for the non-certified causes of death, measles and diarrhoea account for at least 30% of ah ;arly childhood deaths. The dominance of these nutritionally related diseases as leading causes of early childhood deaths shows the importance of the direct and indirect effects of malnutrition in the early childhood years. DISCUSSION
In trying to understand why this sort of pattern has emerged one needs to examine the practices surrounding childbirth and childcare in the neonatal period, child feeding practices, and the endemicity of malaria in Sierra Leone. In the case of childbirth, one is faced with the situation wherein at least 70% of all deliveries in Sierra Leone are still being carried out by traditional birth attendants (TBAs) [8]. These TBAs carry out deliveries in circumstances that are far from hygienic and also indulge in practices that expose both the mother and the new born child to infection. For example, according to West, the instrument used to sever the umbilical cord is generally cleaned with a piece of cloth or rinsed in warm water or hot water but sufficient sterilization is rarely accomplished, thus providing an avenue for tetanus to be introduced to the new-born child. Furthermore, the cord stump is left about three inches long and may be dressed with any number of the following: “juices squeezed from a banana leaf, spittle from chewing kolanut, tobacco leaf or snuff, petroleum jelly, deep ashes from the home cooking fire, talcum powder, chicken manure or scrapings from the bottom of a clay drinking water pot. After these medicines are applied to the stump, it is then wrapped with a strip of rag”. With this sort of unhygienic environment one can perhaps begin to see why there is such a high incidence of deaths from neonatal tetanus, and why current efforts aimed at reducing infant mortality are putting a lot of emphasis on the training of TBAs. They would then be able to continue making deliveries but within a more hygienic environment. In addition, emphasis is also being put on immunization of pregnant mothers so that protection against tetanus can be provided for infants for up to 6 months after birth, since it will be sometime before all TBAs are trained. Furthermore, it has been shown that even after training and the provision of a delivery kit containing equipment for more hygienic delivery conditions, the TBAs still apply to the stump, dressings such as those already described which serve as a source of tetanus infection (91. Another description of the unhygienic environment within which TBAs perform deliveries is provided by Williams [lo]. According to her when the baby is born, most TBAs wait for the expulsion of the placenta before cutting the cord and separating the baby. A razor blade or knife or a pair of scissors or a long blade of grass is then used for cutting the cord. Then native cotton thread or machine thread boiled or unboiled is used to tie the cord. The dressings used on the cord include: “juice from never die leaves, black tumbla leaves, goat fat, the dirt from under the country pot where water is stored, chewed kolanuts, orie or salt or ash”. These dressings are applied until the cord drops off. The attitude of
302
~ILSOR
bt.4
TBAs ,‘o neonatal tetanus is also interesting. Some believe that the new born child develops tetanus because the mother ate some of the prohibited foods during pregnancy. Another belief is that the child has been attacked by a devil, while others believe that the use of a dirty blade or knife to cut the cord is responsible. In the case of measles and diarrhoeal diseases, these are linked to the nutritional status of the child. The results of the National Nutrition Survey indicate that almost one-quarter of all children in Sierra Leone up to 5 years of age suffer from chronic undernutrition. The incidence of malnutrition if lowest for infants aged 3-5 months after which it increases steadily to a peak at 21-23 months [1 11. For example the prevalence of chronic undernutrition was 10.6% for infants aged 3-5 months and it increased to 14.1% for those aged 6-8 months and to 14.3% for infants aged 9-l 1 months. Furthermore, almost onethird of the children are underweight. The percentage of infants were were underweight for their age was 13.8% for 3-5 month old infants, 27.3% for 6-8 month old infants and 37.5% for 9-11 month old infants. With regard to the causes of this poor nutritional status, the survey concluded that poverty, ignorance and disease interacted with diet and other factors in the child’s environment to determine his nutritional status. Another researcher argued that it was not clear that the existence of food taboos affected the nutritional status of children. She concluded that the taboos might be related to the availability of food, which when it was in short supply meant that children would receive smaller amounts since they were fed last [12]. For example, with regards to vegetable and animal proteins and dark green leafy vegetables, only 33% of the children aged 6-l 1 months and 72% aged 12-17 months received these foods when they were eaten by the family. It is possible that inflation during recent years, which reduces the amount of available food within households is a very important contributor to the poor nutritional status of the children. The WHO carried out a malariometric survey in Sierra Leone between 1976 and 1979, the results of which classified Sierra Leone as a hyper to holoendemic area [13]. According to this survey crude parasite rates ranged from 31.9 to 81.5% for all ages and from 35.3 to 81.5% for 2-9 year old children. The survey concluded by identifying malaria as the commonest cause of morbidity and mortality in Sierra Leone. Although malaria by itself is often not listed as an important cause of childhood mortality, yet it is felt that the fever-related diseases which have been indentified throughout this paper as one of the leading causes of infant and early childhood mortality are dominated by malaria. Finally, it must be pointed out that the effects of the diseases identified (tetanus, fevers, measles and diarrhoea) as the leading causes of infant and early childhood mortality, can be reduced by intervention programmes as has now been clearly demonstrated with for example oral rehydration therapy. So a policy which focuses on these major causes initially will yield more immediate dividends after which efforts could then be concentrated on the other causes of infant and early childhood mortality.
KANDEH
CONCLUSION
Information about the Ieading causes of infant and early childhood mortality was compiled from hospital and sample survey records. The leading causes of infant mortality are tetanus, fevers, measles and diarrhoea. A breakdown of certified deaths in infancy showed that tetanus is quite important in the neonatal period accounting for as much as 68 percent of neonatal deaths. Measles and diarrhoea were the leading causes of death in the last 6 months of infancy. The leading causes of early childhood deaths were measles, diarrhoea and fevers. Nutritionally related diseases such as measles and diarrhoea were seen to account for up to 40% of all early childhood deaths. The major factors affecting these causes of death were childbirth and childcare practices in the case of tetanus and the nutritional status of the children in the case of measles and diarrhoea. The majority of deliveries were still being performed by Traditional Birth Attendants in very unhygienic surroundings which it was felt contributed significantly to the high incidence of neonatal tetanus. At the other childhood ages the poor nutritional status of the majority of children in Sierra Leone as shown by the results of the 1978 National Nutrition Survey was seen as the significant factor. The effects of the identified major causes of infant and early childhood mortality (tetanus, fevers, measles and diarrhoea) can be largely diminished by effective intervention programmes such as oral rehydration therapy and the training of Traditional Birth Attendants.
REFERENCES
1.
2.
Clarke J. I. (Ed.) Sierra Leone in Maps. University of London Press, London, 1969. UNFPA. Sierra Leone: Report of a Mission on Needs Assessment for Population Assistance, No. 66, 1984. U.S. Census Bureau, World Population 1983. Washington DC., 1983. Rankin F. H. The White Man’s Grave: A Visit to Sierra Leone in 1834, Vols I and II. Richard Bently, London,
1936. 3. Horton S. A. Health problems in Sierra Leone and its implications for development planning. Contributed paper to the Seminar on Population, Employment and Development Planning in Sierra Leone, 249-286. MODEP, F&town. 1980.
Vol. II, pp.
4. Kandeh H. B. S. Infant and child mortality in Sierra Leone. Ph.D. dissertation, University of Pennsylvania, 1979. 5. Kandeh H. B. S. and Dow T. E. &rer!r and Determinants of Infant and Child Mortality in Sierra Leone: Patterns and Policy Implications The Rockefeller Foun-
dation, New York, 1985. 6. Sierra Leone Government. Infant Mortality Rate and Causes of Infant Deaths in the Western Area 196%1971.
Medical Statistics Unit, Freetown, 1973. Sierra Leone Government. Vital Statistics and Causes of Death for the Western Area 1971-1972 to 1975-1976. Ministry of Health, Freetown, 1976. Sierra Leone Government. General Medical and Statistical Information. Ministry of Health Freetown, 1978. 7. Wright M. L. Some aspects of the pediatric situation in Sierra Leone. Mimeo, undated. 8. West K. M. Traditional Birth Attendant Case Study for Sierra Leone. WHO, Geneva, 1979.
Causes of infant and early childhood deaths 9. ColeS R. W. iteonatal tetanus and training traditional midwives at Segbwema hospital. Unpublished mimeo, Segbwema, 1975. 10. Williams B. E. 0. The Traditional Birth Attendant: Training and Utikarion. Ministry of Health, Freetown. 1979.
303
11. USAID. Sierra Leone National .Vutrition Survey. Washington, D. C.. 1978. 12. Kolasa K. M. The Nutritional Situation in Sierra Leone. Michigan State University, East Lansing. 1978. 13. WHO. Malaria Control Profile in Sierra Leone. WHO, Freetown. 1982.
Sot. Sci. Med. Vol. 23, No. 3, pp. 297-303. 1986 Printed in Great Britain
CAUSES OF INFANT AND EARLY CHILDHOOD DEATHS IN SIERRA LEONE ERROR
SAMA KANDEH
Department of Geography and Environmental Studies, Njala University College, University of Sierra Leone, P.M.B., Freetown, Sierra Leone Abstract-Information about the causes of infant and early childhood (1-4 years) mortality was compiled from the vital registration system for the Western Area, records from the children’s Hospital in Freetown and two demographic sample surveys conducted in various chiefdom headquarters towns around the country. The leading causes of infant mortality are tetanus, fevers, measles and diarrhoea. A breakdown of certified deaths in infancy showed that tetanus is quite important in the neonatal period accounting for as much as 68% of neonatal deaths. Measles and diarrhoea were the leading causes of death in the last 6 months of infancy. The leading causes of early childhood deaths were measles, diarrhoea and fevers. Nutritionally related diseases such as measles and diarrhoea were seen to account for up to 40% of all early childhood deaths. The major factors affecting these causes of death were childbirth and childcare practices in the case of tetanus and the nutritional status of the children in the case of measles and diarrhoea. The majority of deliveries were still being performed by Traditional Birth Attendants in very unhygienic surroundings which it was felt contributed significantly to the high incidence of neonatal tetanus. At the other childhood ages the poor nutritional status of the majority of children in Sierra Leone as shown by the results of the 1978 National Nutrition Survey was seen as the significant factor. The effects of the identified major causes of infant and early childhood mortality (tetanus, fevers, measles, and diarrhoea) can be largely diminished by effective intervention programmes such as oral rehydration therapy and the training of Traditional Birth Attendants.
INTRODUCITON
fatality rate of British colonial officers from yellow
Sierra Leone is a small country of about 28,000 square miles which lies on the South West coast of West Africa between latitudes 6.55’, and 10.0” N and between longitudes 10.16” and 13.18”W [l]. Sierra Leone has four major administrative units, the Western Area, the Southern Province, the Eastern Province and the Northern Province. Each of the three provinces is divided into districts which total twelve and which are in turn subdivided into chiefdoms which number 148. The Western Area is also divided into four very small rural districts and the city of Freetown but its local government patterns are very different from that seen in the three provinces. Sierra Leone’s demographic picture is gradually falling into place with the information provided by the last two censuses (1963 and 1974) which for the first time covered the entire country. The 1974 census results showed a total population of 2,735,154 persons. This was later adjusted to take into consideration an estimated 8.9% underenumeration yielding a new total of roughly 3,002,426 persons. When the unadjusted population totals for 1963 (2,180,355) and 1974 (2,735,159) are compared, the resulting average annual intercensal growth rate is 1.9%. The age structure of the population is youthful with over 35% of the population < 15 years of age, Only 5% of the population is 65 years and over in age. The estimated crude birth rate for the country is 48.7 based on the 1974 census results. Mortality levels are quite high with an estimated crude death rate of 28 and an infant mortality rate of 240 based on the 1974 census results. High mortality conditions have been recorded for Sierra Leone since the start of the colonial era (1807) when Sierra Leone was referred to as- the ‘White Man’s Grave’ because of the high
fever [2]* METHODOLOGY
Information about the major causes of death in Sierra Leone is available from different sources. These sources include Government hospitals in the Western Area where the main hospitals for mothers and children are located (Princess Christian Maternity Hospital and Children’s Hospital), provincial hospitals and private hospitals such as the Catholic Hospital at Serabu, the Nixon Memorial Hospital at Segbwema and the Netland Nursing Home in Freetown. This hospital based information is available because collection procedures are more rigorous in the above mentioned hospitals than in other hospitals. Most of this information eventually ends up in the Medical Statistics Unit of the Ministry of Health where it is compiled into annual reports. Even then the reports are largely for the Western Area. One of the limitations of these hospital-based reports of causes of death is that they do not accurately reflect the distribution of causes of death on the national scene because of the low user rates of hospitals as well as the probability that only very serious cases are usually taken to hospitals. The value of the hospital based data is that over half of the causes of death are certified by a Medical OfBcer. The distribution of causes of death for these certified deaths can then be compared with that for non-certified deaths. More representative information about the major causes of death is provided by reports of causes of death by parents in several sample surveys. This is because such surveys have contacted all households in an area or a representative sample of households.
297
298
BARBER SW
Table I. Principal certified causes of infant deaths in the Western area, I%+1979 (pooled figure) Rank
1 : 4 5 6 7 : 10
% Of all infant deaths
Cause of death Tetanus Pneumonia Diarrhoeal diMeasles Malaria Anaemia Malnutrition Accidents PulmOtUry tuberculosis Other infectious and parasitic diseases
Total Number
17.01 14.64 10.22 5.76 5.42 2.13 1.67 0.51 0.36 42.28 100.00 4186
KANDEH
Jimmi, Bumpe and Njala Komboya).‘It was aasingle round retrospective survey which interviewed 2233 currently married women between the ages of 15-49 years. It was carried out between February and July 1977. (d) A National Mortality Survey which was carried out in ten chiefdom headquarter towns. The five chiefdoms with the highest infant mortality rates and the five with the lowest infant mortality rates, based on the 1974 census results were selected. A total of 1464 households, 1820 women (12-49 years) and 9770 household members were contacted in a single round retrospective survey. The field survey was carried out in early 1980.
Source: Adapted from Table 3. Horton [3, p. 811.
INFAM
that such surveys include households which make use of medical facilities as well as those who do not. The problem with data from this source is that the information is often collected by persons who do not have any training in the certification of deaths and they therefore rely exclusively on retrospective reports and descriptions provided by parents. One advantage of this data source in that is provides information for other parts of the country outside the Western Area. These two data sources should therefore complement each other and provide one with the opportunity of understanding what are the major causes of death in Sierra Leone. In the analysis of the causes of death that follows, the causes of deaths will be presented according to the data source although in the end the discussion of these causes will be integrated. The data for this paper has been drawn largely from: (a) The Vital Registration system in the Western Area. (b) Reports from the Ministry of Health. (c) An Infant and Child Mortality Survey in Bo District. This survey was restricted to the District Headquarter Bo and four other towns (Yamandu, So
MORTALITY
Information on the principal certified causes of infant deaths in the Western Area for the period 1969-1979 [3] is presented in Table 1. According to the information provided in Table 1, the five leading causes of infant deaths in 1969-1979 were tetanus (17.01%), pneumonia (14.64%), diarrhoeal diseases (10.22%), measles (5.76%) and malaria (5.42%). One must, however, take into consideration the fact that other infectious and parasitic diseases account for 42.28% of all infant deaths. So that as a group, infectious and parasitic diseases account for over 50% of all infant deaths when one takes into consideration the fact that a disease like tetanus is an infectious disease. An opportunity to get some insight into what these other infectious and parasitic diseases are, is provided in Table 2 where the certified causes of infant deaths are distributed by cause and age at death for the same Western Area but only for the period 1969-1971. A total of 3783 infant deaths were recorded in the Western Area during this period but only 1772 (46.8%) were medically certified and it is these medically certified infant deaths that are presented in Table 2. From the information provided in Table 2, it can be. seen that tetanus is the leading cause of infant deaths accounting for 25.7% of all certified infant deaths. The results further show that deaths due to
Table 2. Percentage distribution of certified infant deaths by age and cause for the Western area, 1969-1971 (pooled figure) Age at death A
B
C
D
E
F
Pneumonias Tetanus
25.7 15.6
18.0 4.5
68.9 5.6
26.2 17.6
36.6 0.8
29.2 0.3
Anoxic and hypoxic conds Lliarrhoeal diseases Measles Birth Infection Acute respiratory infections Malaria Congenital anomalies Anaemias Septicaemia Avitarninosis Other p&natal causes All other causes
11.0
31.4 0.3
2.5 3.1
10.1
29.2 3.5
19.7 15.8
7.3 0.5
1.3 0.8 0.2 1.6
14.3 2.4 1.3
2.3 0.2 11.8 1.7 100.0 483 27.3
4.8 3.6 6.0 13.1 100.0 84 4.7
5.4 4.3 1.2 3.5 0.8 2.7
3.0 a.5 0.6 6.0 1.6 3.0
22. I 100.0 257 14.5
12.3 100.0 366 20.7
Cause
Rank : 3 4 5 6 7 8 9 10 II I2 13 14 Total Number %
9.8 3.8 2.7 2.5 2.5 2.4 1.9 I.5 1.2 13.4 6.0 100.0 1772 100.0
4.8 0.3 0.7 30. I 2.1 loo.0 582 32.8
Key: A-total; B-under 1 we& C-l-3 weeks; W weeks to under 2 months; E-2-S Source: Adapted from Table 4 Sierra Leone Government [6, p. 61.
months; F-&l
I months.
Causes
of infant and early childhood deaths
Table 3. The percentage distribution of reported causes of infant deaths (Bo town and other towns*) (pooled fig&e 1950-1977) Cause
Fever related Tetanus Convulsions Diarrhoea Pneumonia Measles Other causes Total Number
Bo town
21.6 20.7 18.3 8.0 3.0 7.4 IS.0 100.0 517
(1) (2) (3) (4) (5) (6)
Other towns* 18.6 45.8 3.5 4.8 1.9 6.8 18.5 99.9 550
(2) (1) (5) (4) (6) (3)
*Other towns: Jimmi, Bumpe. Yamandu and Njala Komboya. ( ) Relative ranking. Source: Adapted from Kandeh [4, Tables 3.9 and 3.101.
tetanus were concentrated in the l-3 week period where they accounted for 68.9% of all infant deaths and that is in fact the only period in infancy when tetanus is the leading cause of infant deaths. After the second month of life the importance of tetanus is drastically reduced for the remaining period of infancy. The second leading cause of infant deaths is accounting for 15.6% of all infant pneumonia, deaths. It becomes the leading cause of death in the 2-l 1 month period when the importance of neonatal tetanus has been reduced. The third leading cause of infant deaths are the anoxic and hypoxic conditions of the new born, accounting for 11.0% of all infant deaths and restricted to only the first 3 weeks of life. Diarrhoeal diseases are the fourth leading cause of infant deaths (9.8%) while measles is fifth (3.8%). When the five leading causes of infant deaths for 1969-1971 are compared with those for 1969-1979, tetanus and pneumonia maintain their positions as the leading and the second leading cause of infant deaths, but the anoxic and hypoxic conditions which are the third leading cause of infant deaths in 1969-1971 are not listed for the 1969-1979 data. They have probably been subsumed under the other infectious and parasitic diseases category. Diarrrhoeal diseases still take precedence over measles in both tables although their rank order is higher for the 1969-1971 data (Table 2) than for the 1969-1979 data (Table 1). When the distribution of infant deaths in the various segments of infancy is taken into consideration (Table 2), one observes that 64.84% of all certified infant deaths occurred in the first three weeks of life again underscoring the importance of neonatal tetanus, anoxic and hypoxic conditions of the new born, other perinatal causes and pneumonia. Of these four causes tetanus and pneumonia are more likely to be reduced by intervention programmes aimed at reducing infant mortality levels. For the remaining 35.6% of infant deaths in the 2-l 1 month period, again pneumonia, diarrhoeal diseases and measles are the important causes that intervention programmes should be aimed at for the reduction of infant mortality levels. Based on this one could therefore conclude that the five leading causes of infant deaths based on certified reports are tetanus, anoxic and hypoxic conditions, measles, pyrexia and diarrhoeal diseases.
299
Now turning to noncertified reports of infant deaths, the survey of infant and child mortality in Bo district carried out in 1977 obtained retrospective reports of causes of infant deaths from mothers interviewed [4]. The results have been separated for the town of Bo (a provincial headquarter town and very urban in outlook) and other towns; Jimmi, Bumpe, Yamandu and Njala Komboya (these are chiefdom headquarters and are more rural in outlook). The results are shown in Table 3. The causes of infant deaths shown in Table 3 appear to be similar for both Bo town and the other towns except that whereas tetanus is the leading cause of infant deaths in the other towns (45.8%), it is only the second leading cause of infant deaths in the town of Bo (20.7%). This can not be attributed entirely to the presence of major medical facilities within the town of Bo. It is probably due to a wrong diagnosis of causes of death as the Mende words for tetanus and convulsions are similar. Notice that convulsions have a very high percentage (18.3%) and that the symptoms are similar to those for tetanus. Also up to now convulsions have not emerged as a major cause of infant deaths. So it is likely that in both areas, tetanus is the leading cause of infant deaths even though one would expect a higher percentage of neonatal tetanus deaths in the other towns. The second leading cause of infant deaths shown in Table 3 for both areas are the fever related diseases which could include all diseases that exhibit fevers as a symptom such as malaria, yellow fever, etc. A comparative mortality survey of two contrasting mortality areas was carried out in 1980 in ten chiefdom headquarters [5]. Five of the headquarters were from the chiefdoms with the highest infant mortality rates according to the results of the 1974 census and the other five were from the chiefdoms with the lowest infant mortality rates. The chiefdom headquarters in the high mortality area were Fairo, Zimmi, Futa, Banta Mokelle and Njala Komboya all in the Southern Province of Sierra Leone. Those in the low mortality area were Mahera, Batkanu, Mateboi, Musaia and Falaba, all in the Northern Province. Birth histories from mothers 15-49 years of age were used to compile the causes of death presented in Table 4. The results show that tetanus is the leading cause of infant deaths in both areas although the incidence is somewhat higher in the high mortality area. Fever related diseases are again the second leading cause of infant deaths. Measles and
Table 4. Leading causes of infant deaths (pooled figure 1940-1980) High mortality Rank : 3 4 5
Cause Tetanus Fevers Headpain Measles Diarrh. Stomac. Others
Total Key: diarrh.diarrhoea;
Low mortality
N
%
197 165 97 57 7 7 73 603
32.7 27.4 16.1 9.4
1.2 1.2 12.0 100.0
Cause
N
0%
Tetanus Fevers Measles Diarrh. Stomac.
42 36 29 14 8
25.8 22. I 17.8 8.6 4.9
34 163
20.8 100.0
Others
stomac.-stomach headache; N = number of deaths. Source: Kandeh and Dow [A.
ache; head+pain/
BORBORS.4hf.4KANDEH
300
Table 7. The percentage distribution of the most common causes of adminission and death at children’s hospital, Freetown
Table 5. Principal certified causes of early childhood (I-4 years) deaths in the Western area, 1969-1979 (pooled figure) Rank
QUX
%
Measles Diarrhoeal diseases Pneumonia Malnutrition Malaria Anaemia Pulmonary tuberculosis Accidents Tetanus Other infectious and parasitic diseases
I 2
3 4 S 6 : 9 IO Total Number
1980 Cause
14.1. 14.1. 13.2 7.0. 5.6 4.6’ I.4 I.2 0.6
Bronco-pneumonia Measles Diarrhoeal diseases Neonatal tetanus Rot. cal. malnutr. Meningitis Malaria Anacmia Others Total
38.2 100.0 3825
B
A
B
16.0 (2) 21.7 (I) 14.1 (3) 5.1 (5) 6.5 (4) 3.5 (8) 4. I (7) 4.1 (6) 24.9 100.0
10.2 (3) 28.0 (I) 4.9 (2) 12.0 (4) 9.2 (5) 4.5 (6) 3.8 (7) 3.6 (8) 23.8 100.0
15.2 (2) 10.3 (3) 17.6 (I, 5.4 (6) 6.1 (5) 6.3 (4) 3.3 (8) 4.7 (7) 31.1 100.0
6.4 (6) 13.7 (3) 20.2 (1) 14.9 (2) 8.5 (4) 7.3 (5) I.8 (8) 3.8 (7) 23.4 100.0
Numbers in parentheses indicate the relative ranking of diseases. Key: A-% of all admissions; B-% of all dmths. Source: M. L. Wright (undated).
*Nutritionally related diseases cause a total of 39.8%. Source: Adapted from Horton (3, p. 261, Table 31.
Another interesting aspect of Table 5 is that the nutritionally related diseases (measles, diarrhoea, malnutrition and anaemia) together account for 39.8% of all early childhood deaths. This underscores the importance of nutritional starus as a major contributor to the health of children in these early years. Poor nutritional status it has been suggested inhibits the formation of antibodies which provide a natural protection against infectious and parasitic diseases. The Results in Table 6 [6] show the same pattern to that already described in Table 5 but specifically for the years 1974-75 and 1975-1976. Here also measles, diarrhoea and pneumonia are the leading causes of early childhood deaths. Another report of certified causes of early childhood deaths is provided for the Western Area by Wright [I. This report lists the common causes of admissions and deaths at Children’s Hospital in Freetown for the period 1980-1981 (Table 7). In terms of the percentage of admissions, measles is the leading cause in 1980 and is replaced by diarrhoeal diseases in 1981. The leading cause of death for these admissions in 1980 is measles and in 1981 diarrhoea. So here again the nutritionally related diseases (measles and diarrhoea) are the leading causes of death for the admissions into Children’s Hospital in Freetown. Included among the common causes in Table 7 are protein calorie malnutrition and anaemia which are also nutritionally related.
diarrhoea are also important but their importance is diminished in the high mortality area where pain and severe headaches are the third leading cause of infant deaths. It is probable that this pain and severe headache could really be symptoms of either tetanus or fever related diseases especially since one is looking at infants. A combination of the results of the certified and non-certified causes of infant deaths suggests that the important causes of infant deaths are tetanus, diarrhoeal diseases and measles in order of decreasing importance. Furthermore it also appears that these causes of death are important in certain stages of infancy. Tetanus is important in the neonatal period while the diarrhoeal diseases and measles are important in the last 6 months of infancy. EARLY CHILDHOOD
1981
A
MORTALITY
The principal certified causes of early childhood deaths in the Western Area for the period 1969-1979 are presented in Table 5. The leading causes of early childhood deaths are measles and diarrhoeal diseases which each account for 14.1%. Pneumonia follows very closely as the third leading cause of death accounting for 13.2% while malnutrition is fourth (7.0%). These four leading causes of early childhood deaths together account for 48.4% of the deaths.
Table 6. Ten leading causes of certified early childhood deaths for the Western area for 1974-1976 causes I :
Measles* Influenza and pneumonia Diarrhoeal diseases* Avitaminosis, etc.’ Malaria Anaemias* Dysentery Accidents Tubercolosis Meningitis Other causes
4 5 6 7 8 9 10 II Total Number % of total contributed by nutritionallv related diseases
lNutritlonalIy
19741975 23.4 (I) 20.4 (2) 12.7 (3) 11.0(4) 4.6 (6) 10.2 (5) I.1 (9) I.1 (9) 2.2 (8) 2.0 (7) IO.5 100.0 498 (57.3%)
1975-1976 Table 8. The percentage distribution of reported causes of early childhood deaths (Bo town and other towns’) (waled figure 1950-1977)
21.1 (2) 24.4 (1) 13.6 (3) 9.4 (4) 4.2 (6) 9.2 (5) 0.5 (IO) 1.6 (5) 4.2 (6) 1.6 (8) IO.1 99.9 426
CCIUSC
Fever-related Tetanus Convulsions Diarrhoea Pneumonia Measles Other causes Total Number
(53.3%)
related disease.
( ) Ranking. Source: Sierra Leone Govermnent [6. Table AS. p. IS. 19761.
I
Bo town 26.3 (I) 6.4 (4) 6.1 (5) 14.7 (3) 2.3 (6) 18.5 (2) 19.0 93.3 373
Gther towns* 18.2 (2) 129 (3) 2.6 (5) 125 (4) 1.6 (6) 29.2 (1) 23.0 loo.0 287
*Other towns: Jimmi. Bumpe, Yamandu and Njala Komboya. Source: Adapted from Kandeh 14, Tabks 3.9 and ,.lO].
Causes
of infant and early childhood deaths
Table 9. Leading causes of early childhood deaths (pooled figure 1940-1980) High mortality Rank
I : 4 5 Total
Low mortality
Cause
N
%
Fevers Measles Tetanus Diarrh. Headpain Others
40 30 IS 6 3 30 124
32.2 24.2 12.1 4.8 2.4 24.2 99.9
Key: diarrh.--dianhoca; stomac-stomach headache; N = number of deaths. Source: Kandeh and Dow [5].
Cause
N
%
Measles Fevers Diarrh. Stomac. Tetamis Others
38 20 IS 8 4 21 106
35.8 18.9 14.1 ::: 19.8 99.9
ache; headpain-pain/
From the information provided in Tables 5, 6 and 7 one can conclude that based on certified causes of death in hospitals, the five leading causes of early childhood deaths are measles, pneumonia, diarrhoea, malnutrition and anaemia. Furthermore four of these five leading causes are nutritionally related. The leading causes of early childhood deaths for the survey carried out in Bo district in 1977 are shown in Table 8. Looking at all childhood deaths in Bo town, the leading cause of death are the fever related diseases such as malaria, accounting for 26.3%, followed by measles (18.5%) and diarrhoea (14.7%). In the other towns the leading cause of all early childhood deaths is measles (29.2%), followed by the fever related diseases (18.2%), tetanus (12.9%) and diarrhoea (12.5%). It is worth noting that not only is measles the leading cause of early childhood deaths in the other towns but that the percentage for the other towns is almost double that for the town of Bo. This is probably due to the rural nature of the other towns, so that the nutritional status of the children in these other towns would be relatively lower compared to Bo town. Furthermore the presence of superior medical facilities in Bo town would reduce the proportion of children dying from an attack of measles, while also reducing the incidence of measles through immunization. The results from the comparative mortality survey (Table 9) also show measles and fever as the leading causes of early childhood deaths with measles dominating in the low mortality zone and fevers dominating in the high mortality zone. From the information in Tables 8 and 9 one can conclude that the leading causes of early childhood deaths from non-certified reports are fever related diseases, measles and diarrhoea, Here again the importance of the nutritionally related diseases as significant contributors to early childhood deaths is shown. A combination of the results from the certified and non-certified causes of early childhood deaths suggests that the four leading causes of death are measles, diarrhoea, pneumonia, and fever related diseases. However of these four, measles and diarrhoea are consistently among the leading causes of early childhood deaths whether one is looking at certified or non-certified causes of early childhood deaths. Another consistent pattern from both sources is the high percentage of early childhood deaths caused by nutritionally related diseases, which in the case of certified causes of death is at least 40% and
301
for the non-certified causes of death, measles and diarrhoea account for at least 30% of ah ;arly childhood deaths. The dominance of these nutritionally related diseases as leading causes of early childhood deaths shows the importance of the direct and indirect effects of malnutrition in the early childhood years. DISCUSSION
In trying to understand why this sort of pattern has emerged one needs to examine the practices surrounding childbirth and childcare in the neonatal period, child feeding practices, and the endemicity of malaria in Sierra Leone. In the case of childbirth, one is faced with the situation wherein at least 70% of all deliveries in Sierra Leone are still being carried out by traditional birth attendants (TBAs) [8]. These TBAs carry out deliveries in circumstances that are far from hygienic and also indulge in practices that expose both the mother and the new born child to infection. For example, according to West, the instrument used to sever the umbilical cord is generally cleaned with a piece of cloth or rinsed in warm water or hot water but sufficient sterilization is rarely accomplished, thus providing an avenue for tetanus to be introduced to the new-born child. Furthermore, the cord stump is left about three inches long and may be dressed with any number of the following: “juices squeezed from a banana leaf, spittle from chewing kolanut, tobacco leaf or snuff, petroleum jelly, deep ashes from the home cooking fire, talcum powder, chicken manure or scrapings from the bottom of a clay drinking water pot. After these medicines are applied to the stump, it is then wrapped with a strip of rag”. With this sort of unhygienic environment one can perhaps begin to see why there is such a high incidence of deaths from neonatal tetanus, and why current efforts aimed at reducing infant mortality are putting a lot of emphasis on the training of TBAs. They would then be able to continue making deliveries but within a more hygienic environment. In addition, emphasis is also being put on immunization of pregnant mothers so that protection against tetanus can be provided for infants for up to 6 months after birth, since it will be sometime before all TBAs are trained. Furthermore, it has been shown that even after training and the provision of a delivery kit containing equipment for more hygienic delivery conditions, the TBAs still apply to the stump, dressings such as those already described which serve as a source of tetanus infection (91. Another description of the unhygienic environment within which TBAs perform deliveries is provided by Williams [lo]. According to her when the baby is born, most TBAs wait for the expulsion of the placenta before cutting the cord and separating the baby. A razor blade or knife or a pair of scissors or a long blade of grass is then used for cutting the cord. Then native cotton thread or machine thread boiled or unboiled is used to tie the cord. The dressings used on the cord include: “juice from never die leaves, black tumbla leaves, goat fat, the dirt from under the country pot where water is stored, chewed kolanuts, orie or salt or ash”. These dressings are applied until the cord drops off. The attitude of
302
~ILSOR
bt.4
TBAs ,‘o neonatal tetanus is also interesting. Some believe that the new born child develops tetanus because the mother ate some of the prohibited foods during pregnancy. Another belief is that the child has been attacked by a devil, while others believe that the use of a dirty blade or knife to cut the cord is responsible. In the case of measles and diarrhoeal diseases, these are linked to the nutritional status of the child. The results of the National Nutrition Survey indicate that almost one-quarter of all children in Sierra Leone up to 5 years of age suffer from chronic undernutrition. The incidence of malnutrition if lowest for infants aged 3-5 months after which it increases steadily to a peak at 21-23 months [1 11. For example the prevalence of chronic undernutrition was 10.6% for infants aged 3-5 months and it increased to 14.1% for those aged 6-8 months and to 14.3% for infants aged 9-l 1 months. Furthermore, almost onethird of the children are underweight. The percentage of infants were were underweight for their age was 13.8% for 3-5 month old infants, 27.3% for 6-8 month old infants and 37.5% for 9-11 month old infants. With regard to the causes of this poor nutritional status, the survey concluded that poverty, ignorance and disease interacted with diet and other factors in the child’s environment to determine his nutritional status. Another researcher argued that it was not clear that the existence of food taboos affected the nutritional status of children. She concluded that the taboos might be related to the availability of food, which when it was in short supply meant that children would receive smaller amounts since they were fed last [12]. For example, with regards to vegetable and animal proteins and dark green leafy vegetables, only 33% of the children aged 6-l 1 months and 72% aged 12-17 months received these foods when they were eaten by the family. It is possible that inflation during recent years, which reduces the amount of available food within households is a very important contributor to the poor nutritional status of the children. The WHO carried out a malariometric survey in Sierra Leone between 1976 and 1979, the results of which classified Sierra Leone as a hyper to holoendemic area [13]. According to this survey crude parasite rates ranged from 31.9 to 81.5% for all ages and from 35.3 to 81.5% for 2-9 year old children. The survey concluded by identifying malaria as the commonest cause of morbidity and mortality in Sierra Leone. Although malaria by itself is often not listed as an important cause of childhood mortality, yet it is felt that the fever-related diseases which have been indentified throughout this paper as one of the leading causes of infant and early childhood mortality are dominated by malaria. Finally, it must be pointed out that the effects of the diseases identified (tetanus, fevers, measles and diarrhoea) as the leading causes of infant and early childhood mortality, can be reduced by intervention programmes as has now been clearly demonstrated with for example oral rehydration therapy. So a policy which focuses on these major causes initially will yield more immediate dividends after which efforts could then be concentrated on the other causes of infant and early childhood mortality.
KANDEH
CONCLUSION
Information about the Ieading causes of infant and early childhood mortality was compiled from hospital and sample survey records. The leading causes of infant mortality are tetanus, fevers, measles and diarrhoea. A breakdown of certified deaths in infancy showed that tetanus is quite important in the neonatal period accounting for as much as 68 percent of neonatal deaths. Measles and diarrhoea were the leading causes of death in the last 6 months of infancy. The leading causes of early childhood deaths were measles, diarrhoea and fevers. Nutritionally related diseases such as measles and diarrhoea were seen to account for up to 40% of all early childhood deaths. The major factors affecting these causes of death were childbirth and childcare practices in the case of tetanus and the nutritional status of the children in the case of measles and diarrhoea. The majority of deliveries were still being performed by Traditional Birth Attendants in very unhygienic surroundings which it was felt contributed significantly to the high incidence of neonatal tetanus. At the other childhood ages the poor nutritional status of the majority of children in Sierra Leone as shown by the results of the 1978 National Nutrition Survey was seen as the significant factor. The effects of the identified major causes of infant and early childhood mortality (tetanus, fevers, measles and diarrhoea) can be largely diminished by effective intervention programmes such as oral rehydration therapy and the training of Traditional Birth Attendants.
REFERENCES
1.
2.
Clarke J. I. (Ed.) Sierra Leone in Maps. University of London Press, London, 1969. UNFPA. Sierra Leone: Report of a Mission on Needs Assessment for Population Assistance, No. 66, 1984. U.S. Census Bureau, World Population 1983. Washington DC., 1983. Rankin F. H. The White Man’s Grave: A Visit to Sierra Leone in 1834, Vols I and II. Richard Bently, London,
1936. 3. Horton S. A. Health problems in Sierra Leone and its implications for development planning. Contributed paper to the Seminar on Population, Employment and Development Planning in Sierra Leone, 249-286. MODEP, F&town. 1980.
Vol. II, pp.
4. Kandeh H. B. S. Infant and child mortality in Sierra Leone. Ph.D. dissertation, University of Pennsylvania, 1979. 5. Kandeh H. B. S. and Dow T. E. &rer!r and Determinants of Infant and Child Mortality in Sierra Leone: Patterns and Policy Implications The Rockefeller Foun-
dation, New York, 1985. 6. Sierra Leone Government. Infant Mortality Rate and Causes of Infant Deaths in the Western Area 196%1971.
Medical Statistics Unit, Freetown, 1973. Sierra Leone Government. Vital Statistics and Causes of Death for the Western Area 1971-1972 to 1975-1976. Ministry of Health, Freetown, 1976. Sierra Leone Government. General Medical and Statistical Information. Ministry of Health Freetown, 1978. 7. Wright M. L. Some aspects of the pediatric situation in Sierra Leone. Mimeo, undated. 8. West K. M. Traditional Birth Attendant Case Study for Sierra Leone. WHO, Geneva, 1979.
Causes of infant and early childhood deaths 9. ColeS R. W. iteonatal tetanus and training traditional midwives at Segbwema hospital. Unpublished mimeo, Segbwema, 1975. 10. Williams B. E. 0. The Traditional Birth Attendant: Training and Utikarion. Ministry of Health, Freetown. 1979.
303
11. USAID. Sierra Leone National .Vutrition Survey. Washington, D. C.. 1978. 12. Kolasa K. M. The Nutritional Situation in Sierra Leone. Michigan State University, East Lansing. 1978. 13. WHO. Malaria Control Profile in Sierra Leone. WHO, Freetown. 1982.