Infant mortality in Nigeria: effects of place of birth, mother's education and region of residence

This paper examines the effects of a child's place of birth, mother's education, region of residence and rural and urban residence on infant mortality in Nigeria between 1965 and 1979, using data from the 1981/82 Nigeria Fertility Survey. Infant mortality rates declined in all regions between 1965 and 1979. Children born in modern health facilities, irrespective of their mothers' place of residence, experienced significantly lower rates of infant mortality than those born elsewhere. Logistic regression analysis showed that all other variables tested were also significant, although some to a lesser degree. Efforts to reduce infant mortality in Nigeria should include policies that rectify rural and urban differentials in the distribution of health facilities and encourage their use.

Some correlates of child mortality in the refugee populated regions in Tanzania

Data from the 1988 Tanzania census were used to examine child mortality in three regions populated with Burundi refugees. Logistic and least squares analyses show that for both Tanzanian nationals and refugees low levels of maternal education are associated with high child mortality levels. Children born to mothers who are housewives are associated with low levels of mortality compared to those born to employed mothers, though the results were not statistically significant for the refugees. Maternal demographic status, computed from age and parity, has a strong effect on child survival. Unexpectedly, child mortality was lower where the water source was a well outside the village. Tanzanian mothers who are at highest risk of childbearing are roughly 6.4 times more likely to have a child death than those at lowest risk; the corresponding figure for the refugees is 36.8. This emphasises the need to intensify family planning programmes in these regions.

User characteristics and oral contraceptive compliance in Egypt

Results from the 1988 Egypt Demographic and Health Survey show that many women are not taking oral contraceptives in a manner that ensures full protection by the method. Reports from 1,258 current pill users show a range of incorrect use; 63 percent of women surveyed reported an interruption in their use of the pill in the past month, and of those women, only 40 percent took the correct action when they missed a pill. The majority (89 percent) did not wait the correct number of days between packs. Multivariate analysis revealed that rural women were more likely to take pills out of sequence, compared with their urban counterparts. Failure to take a pill within the previous month was strongly associated with the experience of side effects. The younger women surveyed were more likely to know the correct interval between pill packs than were older women; and women who wanted more children were more likely to know the correct interval than those who did not. The use patterns exhibited by the pill users may be the result of their receiving confusing, incorrect, or incomplete information, and highlight the need to provide women with accurate, updated, and comprehensible information about oral contraceptives.

Determinants of couple agreement in U.S. fertility decisions

An analysis of data from 8,450 women interviewed in 1988 for the National Survey of Family Growth finds that teenagers, never-married women, black women and those with less than a high school education are less likely than other women to have a birth that is jointly desired by both partners; 29%, 35%, 45% and 51% of births, respectively, are wanted by both partners, compared with an overall average of 69%. Third and higher order births are also less likely than earlier births to be jointly planned--58%, compared with 69% of first births and 76% of second births. In situations in which the birth is not jointly planned, black women, unmarried women, teenagers and women having third or higher order births are all significantly more likely than other women in their race, marital status, age and birth-order categories to have a birth when the man's preference is unknown. Never-married women are significantly more likely than married women to have a birth when the woman desires one but the man does not, while black women are significantly more likely than white women to have a birth that the man wanted but that the woman did not.

Factors affecting desired family size in Bangladesh

Factors affecting desired family size in rural Bangladesh are examined using data from contraceptive prevalence surveys conducted between 1983 and 1991. The analysis suggests that mothers having two sons and one daughter are more inclined to perceive their family as complete than those having three sons and no daughter. Logistic regression analysis indicates that important determinants of desire for more children are age of woman, current contraceptive use status, work status, and family planning worker's visit. The policy implications of these findings are discussed.

Education and assortative marriage in northern and urban Sudan: 1945-79

This study examines the extent of assortative mating for education in Northern Sudan and urban Khartoum. More males than females were found at higher levels of education. Increasingly, people tend to marry persons of equal educational level, but the unequal educational opportunities for males and females have led to the emergence of educational exogamy in which members of different educational levels are more likely to marry from the adjacent educational category than from distant categories. The increasing level of education for both sexes, and especially for females, may in part explain the rising trend in age at marriage.

Use of obstetric services in rural Nigeria

The pattern and determinants of maternal service utilization were studied in a rural Nigerian community. The study sample consisted of 488 randomly selected women who had a childbirth or an abortion between May 1987 and September 1989. Although 93% registered for prenatal care in a health care institution, only 51% delivered in a health institution while 49% delivered at home mainly under the care of traditional birth attendants. Factors found to be most consistently associated with the use of health institutions for delivery were maternal education and occupation, religion, and occupation of the husband. Maternal age, parity and marital status and place of the residence were not significantly associated with the choice between home and institutional delivery. Logistic regression analysis was used to determine the odds ratio and to quantify the weight of these independent variables found to be significantly associated with the place of delivery as the outcome variable.

The socio-cultural context of high fertility among Igbo women

The Igboland of Nigeria has been under the influence of socio-economic change since the turn of the century, as typified by a high literarcy rate, a highly migratory population, the predominance of Christianity, and built-up towns and villages. Yet Igbo fertility has remained high even by Nigerian standards. Part of the explanation for high Igbo fertility is the prevalence of peculiar socio-cultural institutions which tend to encourage or support high fertility. In this study fertility differentials, reproductive behaviour and fertility preferences and intentions are examined as a function of three well-recognised cultural institutions or contextual factors: the bestowal of high fertility honour or title to women of a given family size, patriarchal relations, and patrilinearity and son preference, together with individual status indicators. Our findings suggest that socio-cultural institutions establish or condition relationships and behaviours among the Igbo; in other words, individual fertility behaviour takes place within the context of complex social organisation and under the influence of multiple social, cultural and ideological realities.

Age-specific patterns of association between breast cancer and risk factors in black women, ages 20 to 39 and 40 to 54

Data from the 1980 to 1982 population-based Cancer and Steroid Hormone case-control study of women 20 to 54 years old afforded the opportunity to investigate risk factors for breast cancer among black women younger than 40 years (177 patients and 137 control subjects) and to compare the results to black women 40 to 54 years old (313 patients and 348 control subjects). Information on exposure variables was obtained by in-person interviews. The logistic regression results indicated that the risk of breast cancer among black women younger than 40 years was nearly three times greater for those who used oral contraceptives for more than 10 years relative to never-users (odds ratio, 2.8; 95% confidence interval, 1.2 to 6.8) and more than four times greater for severely obese women (body mass index > or = 32.30 kg/m2) relative to women whose relative weights were less than 24.90 kg/m2. Patterns of association for the two age groups were similar for surgical menopausal, age at first full-term pregnancy, and multiple births, but differed for age at menarche.

Associations between social and environmental factors and perinatal mortality in Jamaica

Social and environmental factors in Jamaica were compared between 9919 mothers delivering in a 2-month period a singleton who survived the early neonatal period and 1847 mothers who were delivered of a singleton perinatal death in a contiguous 12-month period. Logistic regression showed independent positive statistically significant increased odds of having a perinatal death among mothers who lived in rural parishes, older mothers (aged 30 +), single parents, no other children in the household, large number of adults in the household, mother unemployed, the major wage earner of the household not being in a managerial, professional or skilled non-manual occupation, the household not having sole use of toilet facilities, smaller mothers and those classified as obese or undernourished. Variations were found for different categories of death. Intrapartum asphyxia deaths were not related to union (marital) status, occupation of major wage earner, number of adults nor to the use of the toilet. Antepartum fetal deaths did not vary significantly with occupation of major wage earner or maternal height, but did show a relationship with maternal education, mothers with lowest levels having reduced risk. Deaths from immaturity were significantly related only to occupation of major wage earner, number of children in the household, number of social amenities available (negative relationships) and maternal age (< 17 at highest risk). In conclusion there was little to indicate that social deprivation per se was related to perinatal death, although specific features of the environment showed strong relationships.

Nutritional status and early warning of mortality in southern Ethiopia, 1988-1991

OBJECTIVE

Ethiopian Government guidelines on nutritional survey data and relief provision call for intervention once mean weight-for-length (W/L) in an area falls below 90% of reference, on the basis that mortality is unlikely to rise until this level is reached. In this paper the appropriateness of the 90% cut-off is examined using data from Wolayita, southern Ethiopia. SURVEY DESIGN AND SUBJECTS: Fifteen to 25 villages are selected at random for survey each year, with all children 70-110 cm in length being followed up every 2 months. During the 3 years covered by these analyses 21,701 W/L measurements were made on 5455 children from 65 villages. 126 of the children died.

RESULTS

In the first and third survey years, rapid declines in mean W/L were recorded, with mortality increasing very roughly three-fold (compared to year 2, P < 0.01), even though mean W/L remained at or above 90% of reference at all times. A logistic regression analysis relating mortality to W/L indicates that between 20% and 35% of the greater mortality in years 1 and 3 can be explained by the observed changes in W/L. The remainder occurred because of an increase in underlying or background risk (which might perhaps be expected in the circumstances of generally deteriorating nutritional status).

CONCLUSIONS

The results suggest that child mortality is likely to increase before area mean W/L falls to 90% of reference, indicating that emergency interventions should be triggered earlier than at present.

Does women's literacy affect desired fertility and contraceptive use in rural-urban Pakistan?

The 1984-85 Pakistan Contraceptive Prevalence Survey showed that urban wives had more than twice the literacy rate of rural wives. The present study explored the relationship of the rural-urban gap in female literacy to differences in contraceptive use. In rural areas, literacy did not increase women's perceptions of having reached a 'sufficient' number of living children, although the opposite was true for urban areas. Yet rural women with an 'insufficient' number of living children were more likely to use contraception if they were literate, as did their urban counterparts. Thus, raising the literacy rate in rural Pakistan would not narrow the rural-urban gap in contraception to cease childbearing but would narrow the rural-urban gap in contraception used to space wanted births further apart. Recommendations for government policy are made.

State income tax policy and family size: fertility and the dependency exemption

Tax exemptions for dependents are a subsidy to having children and therefore create incentives for fertility. This article explores the relationship between the changing tax value of the state exemption for dependents—combined with the federal exemption—and the observed fertility choices of married couples. Using data from the Panel Study on Income Dynamics (PSID), the article finds that the federal exemption does have a positive impact on differential period fertility for the sample of 229 married couples in this study. This effect is dampened by an offsetting labor force participation and marginal tax rate effect, but it does provide evidence that fiscal policy can influence fertility. State income tax exemptions do not appear to significantly influence the fertility decisions of the sample.

Adolescent life events and their association with the onset of sexual intercourse

This study examined the timing of several events marking the transition from adolescence to young adulthood and their correlation with the age at first sexual intercourse. The model included parental disruption, dropping out of school, working, drug use, and dating. Socioeconomic status, gender, and race were included as control variables. Data are taken from the National Survey of Children, a longitudinal study of children ages 7-11 in 1976 who were reinterviewed in 1981 and 1986. Factors associated with first intercourse include age, tobacco use, marijuana use, dating, and parental divorce during the child's adolescent years. Age interacts with dating, working, and the use of illegal substances, and race interacts with dating and dropping out of school in their influence on first intercourse. Other transitions, such as alcohol use, parental divorce before adolescence, and parental marriage, had little influence on first sexual intercourse. Events such as dating appear to increase the risk of first intercourse, while the use of drugs may reflect a set of behaviors that occur simultaneously with sexual initiation. The correlated transitions identify a group of adolescents at risk for early sexual activity.

Hierarchical logistic regression models for imputation of unresolved enumeration status in undercount estimation

"In this article we describe a logistic regression modeling approach for nonresponse in the [U.S.] Post-Enumeration Survey (PES) that has desirable theoretical properties and that has performed well in practice.... In the 1990 PES, interviews were not obtained from approximately 1.2% of households in the sample, and approximately 2.1% of the individuals in interviewed households were considered unresolved after follow-up....The missing binary enumeration statuses for these unresolved cases were replaced with probabilities estimated under a statistical model that incorporated covariate information observed for these cases. This article describes an approach to modeling missing binary outcomes when there are a large number of covariates."

Factors influencing infant mortality in Vietnam

Selected determinants of overall infant mortality in Vietnam were examined using data from the 1988 Vietnam Demographic and Health Survey, and factors underlying neonatal and post-neonatal mortality were also compared. Effects of community development characteristics, including health care, were studied by logistic regression analysis in a subsample of rural children from the 1990 Vietnam Accessibility of Contraceptives Survey. Infant neonatal and post-neonatal mortality rates showed comparable distributions by birth order, maternal age, pregnancy intervals, mother's education and urban-rural residence. Rates were highest among first order births, births after an interval of less than 12 months, births to illiterate mothers and to those aged under 21 or over 35 years of age. Logistic regression analysis showed that the most significant predictor of infant mortality was residence in a province where overall infant mortality was over 40 per 1000 live births. In the rural subsample, availability of public transport was the most persistent community development predictor of infant mortality. Reasons for the low infant mortality rates in Vietnam compared to countries with similar levels of economic development are discussed.

Educational outcomes in teenage pregnancy and parenting programs: results from a demonstration

A comparison of five in-school educational and service approaches offered at seven sites in Arizona to 789 pregnant and parenting teenagers shows that except for those who enroll in a program in their third trimester, pregnant and parenting teenagers who attend a comprehensive, school-based, community-linked program are significantly more likely to continue in school than are those who have no access to a special program. The comprehensive program's impact is greatest among Hispanic students, younger students, those in grades 9-10, those who are living with their partner and those who enter the program in the first trimester. Two of the program components--strong outreach efforts and case management-are believed to have an especially favorable impact on continuation in school.

The socioeconomic consequences of teenage childbearing: findings from a natural experiment

A study based on census data from 1970 and 1980 examines the socioeconomic effects of unplanned teenage childbearing by comparing teenage mothers whose first birth was to twins with those whose first birth was to a single infant. Among black women, an unplanned teenage birth--represented by the secondborn twin--results in significantly lower rates of high school graduation and labor-force participation and significantly higher rates of poverty and welfare recipiency. Ten years after giving birth, black women who have an unplanned child are also significantly less likely than women who have not to be currently married, but are not less likely to have ever been married. Like black women, white women who have an unplanned teenage birth have significantly higher rates of poverty and welfare recipiency; they also have significantly lower family earnings and household income.

Women's labour force participation and socioeconomic development: influences of local context and individual characteristics in Brazil

We address several key hypotheses about the effects of socioeconomic development on women's labour force participation during the transition from agriculture to industrialism. To this end, we explore differences in women's labour force participation in Brazil by education, marital status, age, and urban or rural residence. We also show how socioeconomic development affects the overall level of women's participation and the differentials by education, etc. Our data are drawn from a large 1973 PNAD (Pequisa Nacional por Amostra de Domicilos) survey conducted by the Brazilian census bureau. Socioeconomic development in different parts of Brazil ranges from pre-industrial agriculture to heavy industry. Using logistic regression, we show that the general level of women's labour force participation does not change with the level of development. Highly educated women are much more likely than the less educated to be in the labour force (net of other influences); this difference is substantially greater than in post-industrial societies. Somewhat surprisingly, the influence of education is the same across the range of development levels in Brazil. Single women are more likely to be in the labour force than married women, and the difference grows during development. Age has a curvilinear relationship to labour force participation, and the old are much less likely to participate in more developed places. Rural women are slightly more likely to be in the labour force at all levels of development.