The immunization programme in Bangladesh: impressive gains in coverage, but gaps remain.

The paper reviews the achievements in tetanus immunization coverage and child immunization in Bangladesh. It uses data from the 1993-94 Bangladesh Demographic and Health Survey to identify and examine the programmatic and non-programmatic factors that influence the coverage of tetanus (TT) immunization during pregnancy, and full immunization among children 12-23 months old in rural Bangladesh. The purpose of this analysis is to identify the areas that need further programme attention. The logistic regression results show that the coverage of TT immunization was significantly associated with proximity to outreach clinics and the presence of a health worker in the community. Home visits by health/family planning fieldworkers and the proximity to outreach clinics had larger influences on TT coverage of poorer households compared to those better-off. The effect of distance to static clinics varied by regions. Among children, full immunization coverage (coverage of all of BCG, DPT1, DPT2, DPT3, Polio1 Polio2, Polio3) was significantly associated with distance to outreach clinics, the greater the distance to the clinics, the less the likelihood of immunization.

PIP: The authors review the achievements in tetanus immunization coverage and child immunization in Bangladesh, using data from the 1993-94 Bangladesh Demographic and Health Survey to identify and examine the programmatic and nonprogrammatic factors which influence the coverage of tetanus immunization (TT) during pregnancy, and full immunization among children aged 12-23 months old in rural Bangladesh. The research was conducted to identify which areas need additional program attention. According to logistic regression analysis, the coverage of TT immunization was significantly associated with proximity to outreach clinics and the presence of a health worker in the community. Home visits by health and family planning field workers, and the proximity to outreach clinics had more influence upon the TT coverage of poorer households than they did upon those which were more affluent. The effect of distance to static clinics varied by regions. Among children, full immunization coverage was significantly associated with distance to outreach clinics, with greater distance to the clinics reducing the likelihood of immunization.

Feeding, home-remedy practices, and consultation with health care providers during childhood illness in rural Bangladesh.

The nature of feeding, home-remedy practices, and consultation with health care providers during illnesses of children in Matlab, a rural subdistrict of Bangladesh, were studied. Data were collected between October 1986 and February 1987 from mothers of children aged 3-36 months. About half of the mothers initiated home-remedies for treating their children. The treatment included: pouring water on the head, feeding juice of herbs, using sanctified water and ORT. The type of treatment was dependent on the nature of illness. The use-rate of ORT was low. Breastfeeding was discontinued for 16% of the episodes; the reduction was the highest for fever with cough, followed by fever and diarrhoea. A reduction in intake of other foods was observed for 39% of the episodes with a total discontinuation for 10%. The reduction was the highest for dysentery, followed by diarrhoea, and fever with or without cough. Total withdrawal of foods other than breastmilk was the highest for diarrhoea. Contact with at least one health care provider of any type by the fifteenth day of illness was made for 41% of the episodes. The contact rate was significantly greater for children living in the health intervention area; for boys, for infants, for richer households, and for severe cases. A higher contact rate in cases of diarrhoeal illness was observed for children of illiterate mothers than that of literate ones.

A hazard logit model analysis of covariates of childhood mortality in Matlab, Bangladesh.

In a prospective study in Matlab, a rural area in Bangladesh, the relationship between a variety of covariates and childhood mortality was examined. Economic status of household, education of mother, sex of the children, health intervention programmes, age of mother, and live birth order of the children were identified as having a statistically significant impact on child survival when the effect of age was controlled. The effects of sex of the children, health programmes, age of mother, and birth order were found to be dependent on the age of the children, but the effect of mother's education was dependent on sex of the children.

PIP: Multivariate analysis was conducted in Matlab, a rural area of Bangladesh, to determine the absolute and relative significance of selected variables in influencing childhood mortality. Variables were grouped under the broad headings of socioeconomic, hygiene-environmental, demographic, and health program factors. Household economic status, mother's education, children's sex, health intervention programs, mother's age, and live birth order of children had statistically significant impacts on child survival when controlling for the effect of age. The sex of children, health programs, mother's age, and birth order were dependent upon the age of the children, while mother's education was dependent upon the sex of children. Having identified certain covariates of childhood mortality, efforts must now be made to identify and change the proximate determinants which negatively affect child survival.

Child morbidity patterns in Ethiopia.

This study is based on the 1983 Rural Health Survey of Ethiopia. Patterns and levels of child morbidity by age, sex, geographic region, and sanitary facilities are examined. Morbidity levels peak in the second year of life. Diarrhoeal diseases are of major importance, especially among infants and toddlers. Parasitic diseases, and respiratory diseases other than pneumonia, become increasingly important with age. There are no significant sex differentials in morbidity except for higher rates of diarrhoeal diseases among female children. Geographic differentials are quite marked with particularly high morbidity levels from all disease types in the western mountainous regions of Ethiopia. Access to high quality drinking water, a latrine, and garbage disposal, are strongly related to reduced overall morbidity levels, though not necessarily to reduced diarrhoeal disease levels.

PIP: 1983 data on 11,962 6-year old children were analyzed to examine patterns of child illness in rural Ethiopia. Overall child morbidity was high, especially among children who were at the most 1 year old. No significant difference existed between boys and girls, except diarrhea occurred more often in girls. Diarrhea accounted for most illnesses, especially among infants and 1-year old children. As children aged, parasitic and other respiratory diseases increased. Children who lived in a house with a latrine and who defecated in the latrine had the lowest morbidity rate (p.001). Only 12.7% of children who lived in a home with piped water were sick in the 2 weeks prior to the survey compared with or=20% for all other sources of water (p.001). The water supply with the highest morbidity rate was a river (30%). In addition, children who had a garbage bin in their house experienced fewer illnesses than did those who did not have a bin (17.1% vs. 29.2%; p,.001). Even though these 3 sanitation indicators were associated with reduced morbidity, they were not always associated with reduced diarrhea levels. The 3 western mountainous regions (Gonder, Ilubabor, and Wellega) had the highest morbidity rates (59.3%, 41.9%, and 45.7%, respectively). Their rates were much higher than the country morbidity rate of 28.8%. Moreover, children in these mountains regions were 2.6-4.9 times more likely to become sick than those in the other regions. Gonder region had the highest malnutrition rate and the 2nd highest diarrhea prevalence rate. Wellega had the highest diarrhea rate and the 2nd highest malnutrition rate. Ilubabor ranked 3rd for both diarrhea and malnutrition. Yet Ilubabor ranked highest in parasitic and other respiratory diseases. Lack of health services and malnutrition most likely accounted for the higher child morbidity in these 3 regions. It is concluded that the Primary Health Care Program should focus on improvements in sanitation and health education.

Socioeconomic, demographic and environmental determinants of infant mortality in Nepal.

The Nepal Fertility and Family Planning Survey of 1986 demonstrated that demographic variables, previous birth interval and survival of preceding child, still predominated as determinants of infant mortality, particularly in rural areas of Nepal. However, in urban Nepal, where the level of socioeconomic development is higher, an environmental variable, along with previous birth interval and survival of preceding child emerges as important in determining infant mortality. Separate policy measures for child survival prospects in rural and urban Nepal are suggested.

PIP: The analysis of the 1986 Nepal Fertility and Family Planning Survey, based on a representative sample of 5-29 married women 15-50 years, was consistent with the hypothesis that demographic factors are more important in early stages of development, with socioeconomic and environmental factors gaining importance as development proceeds. The 1976 Nepal Fertility Survey also reveals the pattern of high infant mortality rates as a reflection of maternal age, parity, previous birth interval, and survival of preceding child. The infant mortality rates fell between 1976-86 from 150 to 100/1000. Urban areas are beginning to show environmental variables as determinants of infant mortality. In urban Nepal, the logit linear model of the effect of socioeconomic, environmental, and demographic factors show maternal age and birth order as significant determinants of infant mortality, regardless of whether demographic or socioeconomic variables are controlled. Mothers 20 years experience a risk of infant death 51% greater than mothers 20-29 years, and 50% higher than mothers 30 years. 1st born children have the lowest risk of dying during infancy and risk increases with birth order. Drinking water and toilet facilities were also significant factors. The probability of dying is 44% higher for those drinking lake or river water than those using piped or tube well water. Those without a toilet have a 64% higher probability of dying than those with their own toilet facility. These 2 variables are significant and enhanced even after controlling for maternal age and birth order. When a logit linear model is run with only 2 nd and higher order birth with birth interval and preceding child as control variables, only toilet facility has a significant effect. Birth interval and survival of preceding child become the significant demographic variables. The joint model of birth interval, survival of preceding child, and toilet facility has significant effects on infant mortality. Children born after a short interval (19 months) with the preceding child dead have a lower mortality due to lack of competition. The policy implications are to increase the use of contraception, to reduce births of young mothers, and insure adequate birth intervals. High risk households need special attention from family planning and health services.

The changing growth of aboriginal children.

We have used data from children at an Aboriginal settlement in Queensland to demonstrate the changing patterns of growth over the last 30 years. In cohorts of children born during the early 1950s and 1960s the mean weight-for-age value was satisfactory for the first three months of life, but then fell until aged 12 months and remained at about 92% of the NCHS median value until 5 years old. In the cohort born in 1972-73 there was some improvement and in the 1982-83 cohort the mean weight-for-age value was close to international levels from 1 to 5 years. The decline in weight-for-age between 3 and 12 months was still present. Length-for-age values were not available in the 1950s and 1960s but the mean value increased from the 1972-73 cohort to the 1982-83 cohort, which was close to international levels. The mean weight-for-length was close to international levels in the 1972-73 cohort, but in the 1982-83 cohort was high from 3 to 12 months and then fell close to the international level. These data indicate that the children at this Aboriginal settlement are now growing satisfactorily. The impaired growth between 3 and 12 months needs further investigation, but most of the earlier concerns about poor growth appear to have been resolved.

Maternal education and child immunization.

This article explores the hypothesis that formal education of women results in increased child survival because of greater knowledge of the protective function of the major childhood immunizations. Education is also associated with greater awareness of proper immunization schedules. Irrespective of mother's formal education level, specific immunization knowledge is associated with an increased likelihood of using immunization. The Indonesian analysis is important as a model for preventive health campaigns among other populations with low education levels among women.

Social factors affecting use of immunization in Indonesia.

This paper examines the importance of certain social and demographic factors in the acceptance of child immunization in rural Yogyakarta, Indonesia. The major findings concern the importance of the role of the community (hamlet) leader in motivating or instructing parents to seek immunization for their children. While these leaders can be effective in conveying information about availability of immunization services, they are not in a position to provide education on immunization function or procedure. The survey results also show that knowledge of the disease-prevention function of the specific vaccines is important, and it is recommended that vaccines be given names which incorporate the name of the disease which they prevent.

A comparison of census and family planning program data on contraceptive prevalence, Indonesia.

Indonesian family planning (FP) service statistics indicate that three out of five couples of childbearing age nationwide are now using modern contraception provided by the FP program. The 1980 Indonesian Census provided a unique opportunity to cross-check the FP program's contraceptive prevalence figures for every province. The results of that comparison indicate considerable variation, particularly in Java, between the FP program estimates of numbers of current users and the census estimates. Possible explanations for these differences are sought in the FP program approach to estimation of prevalence levels from acceptor data, and by examining the possibility of census underestimation of prevalence levels.

Understanding the fertility decline in Bali.

PIP: Reasons for the sharp decline in fertility in Bali, Indonesia, during the 1970s are examined. Data are from a survey conducted in 1980 among 1,088 ever-married women aged 15-54 who were living in three villages in Klungkung regency. Analysis of the survey data shows the decline in fertility to be attributable to the use of modern contraception within marriage but does not explain why the Balinese accepted contraception to such an extent. Finally, through observation of daily life over a period of time, the author identifies "the often-noted high degree of conformity which characterizes most aspects of Balinese life" as a key factor in explaining the fertility decline.^ieng

Essential thrombocythaemia.

The results of detailed studies of 3 patients with essential thrombocythaemia (ET) are presented. Qualitative tests of platelet function were abnormal, and autologous platelet survivals were diminished. Megakaryocyte ploidy distribution analysis showed an increased proporttion of cells with high ploidy, and buoyant density distribution analysis showed increased proportions of less dense platelets. Ultrastructural studies of the platelets showed proliferation of the dense tubular system with deficiency of the surface-connected canalicular system. Megakaryocyte electron microscopy showed abnormal distribution of demarcation membrane complex and granules. The precise functional abnormality of the platelets in ET has not yet been defined, but a release defect of variable severity with corresponding morphological abnormalities of dense tubular and surface-connected canalicular systems, seems the most likely explanation.

Megakaryocytes and the heterogeneity of circulating platelets.

Megakaryocytes mature to the point of cytoplasmic disintergration in three principal ploidy classes: 8n, 16n and 32n. Cells of each of these ploidy classes have been identified, using both microdensitometry and measurement of cell volume and submitted to morphometric analysis. Mature megakaryocytes of the three ploidy classes have been shown to differ in the concentration of cytoplasmic constituents which would be expected to relate to the buoyant density of their platelet progeny. Density separated platelets have been similarly analysed. Light platelets correspond with the 32n megakarycytes and are more liberally endowed with surface connected canalicular system than the progeny of the common 16n megakaryocytes; it is proposed that they have functional characteristics related to this finding. Dense platelets, which are larger in size, correspond with 8n megakaryocytes and show a greater content of granules and mitochondria than platelets of average density. These platelets most probably show specialized function relating to release of granule constituents. Fragments of cytoplasm released from megakaryocytes represent one form of "megathrombocyte" equated by others with newly formed platelets. The differences in structure between these fragments and circulating platelets are emphasized; each such fragment must undergo further disintergration into a number of platelets. It is suggested that the heterogeneity of circulating platelets with respect to both size and density stems from the origin of platelets of varying density from the different populations of megakaryocyte and their release in the form of large cytoplasmic fragments rather than as platelets.