Although the general objective of this study is to examine the extent to which spouses’ socioeconomic characteristics determine whether modern contraception is used and whether family limitation (the demand for no more children) is desired, its central goal is to evaluate the degree to which the net effect of a woman’s education on those fertility decisions is altered once a control is made for the level of schooling of the husband. Individual characteristics of spouses included as controls in this analysis are on the one hand women’s attributes relating to employment, age, parity, ethnic identity, and urban residence and, on the other hand, the occupation of the husband. Data used in this research are provided by DHS surveys conducted in fourteen sub-Saharan countries: Mali, Burkina Faso, Niger, Nigeria, Cameroon, Benin, Senegal, Ghana, Central African Republic, Kenya, Zambia, Zimbabwe, Namibia and Rwanda. With two dichotomous outcome variables, logistic regression was used to estimate two nested models for each dependent variable and for each country covered by the study. DHS respondents used as units of analysis in this study are women who were married (any kind of union) and non-pregnant at the time when each national survey was conducted. The findings suggest that, while an educated wife needs the support of an educated husband to state a preference for family limitation in contemporary sub-Saharan Africa, controlling for husband’s education and other relevant covariates does little to undermine the evidence that woman’s advanced education and the adoption of modern family planning are positively related in the developing world.

The 1998 El Niño significantly reduced garden productivity in the Upper Orinoco region in Venezuela. Consequently, parents were forced to allocate food carefully to their children. Nutrition data collected from village children combined with genealogical data allowed the determination of which children suffered most, and whether the patterns of food distribution accorded with predictions from parental investment theory. For boys, three social variables accounted for over 70% of the variance in subcutaneous fat after controlling for age: number of siblings, age of the mother’s youngest child, and whether the mother was the senior or junior co-wife, or was married monogamously. These results accord well with parental investment theory. Parents experiencing food stress faced a trade-off between quantity and quality, and between investing in younger versus older offspring. In addition, boys with access to more paternal investment (i.e. no stepmother) were better nourished. These variables did not account for any of the variance in female nutrition. Girls’ nutrition was associated with the size of their patrilineage and the number of non-relatives in the village, suggesting that lineage politics may have played a role. An apparent lack of relationship between orphan status and nutrition is also interesting, given that orphans suffered high rates of skin flea infections. The large number of orphans being cared for by only two grandparents suggests that grooming time may have been the resource in short supply.

A large multicentre epidemiological study was carried out by WHO between 1991 and 1995 to analyse the duration of lactational amenorrhoea in relation to breast-feeding. The main results of this analysis, which used classical statistical modelling, have been already published. However, some specific aspects of the postpartum fertility amenorrhoea and breast-feeding covariates, and more specifically the observed progressive exhaustion of the breast-feeding inhibitory effect on the reproductive axis, may justify a closer look at the validity of the statistical tools. Indeed, as has already been emphasized, analysis of large longitudinal data sets in reproduction often faces three difficulties: (i) the precise determination of the event of interest, (ii) the way to handle the time evolution of both the studied variables and their effect on the event of interest and (iii) the often discrete nature of the data and the associated problem of tied events. The first objective of the present work was to give additional insights into the estimation and quantification of the dynamics of the effect of breast-feeding over time, considering this covariate either as fixed or time-dependent. The second objective was to show how to perform the analyses using corresponding adapted procedures in widely available statistical packages, without the need for acquiring particular programming skills.

This paper describes the efforts of Family Health International to develop a simple, standard and replicable methodology to estimate expenditures on family planning in developing countries. The study found that it is not possible, at least at this time, to develop such a methodology. Numerous problems were encountered in making expenditure estimates. First and most important, expenditure accounts are not always readily available. Even when these accounts are available, they may require some adjustments. In some instances, for example, the salaries of health workers who also provide family planning are in the health accounts, whereas in other cases the workers covered in the family planning accounts spend some of their time providing health services. Allocation variables then must be developed to separate spending on family planning from that on health. In some instances allocation variables were developed and used to separate family planning from health expenditures, but in this case allocation variables were not available and a cost analysis was performed. It is concluded that it takes considerable time and effort to estimate expenditures, and that the approach that was followed varied by country, reflecting the data available to make estimates.

This study, conducted in Kisumu District, Kenya, identified important risk factors for infection with intestinal helminths using traditional epidemiological methods. This was combined with a rapid assessment study using qualitative methods, which focused on sanitation practices and local illness perceptions. The multiple logistic regression analysis revealed that the absence of latrines was a significant predictor for hookworm infection with an odds ratio (OR) of 1·9. The analysis also revealed that households without soap had a 2·6 times higher risk of being infected with Ascaris lumbricoides compared with households where soap was available, and that the number of inhabitants living in a household was a significant predictor for infections with hookworms (OR=3.2). Furthermore, the presence of children of 5 years and under in the household was a predictor for infection with A. lumbricoides (OR=2.7), while the absence of this age group was a predictor for hookworm infection (OR=3.8). The qualitative part of the study revealed that people did not consider worms as a serious health threat, but as a nuisance. Among the population, latrines were seen as beneficial because they were believed to prevent disease, to provide privacy and to keep the environment free of faeces. Handwashing was done many times a day but usually not with soap, which was mainly used when bathing or washing the whole body. Many inhabitants in a household or the presence of children of 5 years and under were never mentioned as being a disadvantage, and these elements would therefore be difficult to include in a control strategy. Since the lack of latrines and of soap were identified as risk factors for infection, while latrines, soap and medicine were seen as assets by the population, it is suggested that helminth control interventions should be concentrated within these areas in this particular society.

This study examines the prevalence of self-reported reproductive tract infections (RTIs) and treatment seeking among married, recently pregnant women in Uttar Pradesh state, India. Associations between RTI symptom reporting and background characteristics are examined in a population-based sample of 18,506 married women with a pregnancy in the 3 years prior to a 1995 statewide survey. Logistic regression analyses are used to assess the effects of sociodemographic covariates on the probability of reporting an RTI symptom. Nearly one out of four women reports experiencing at least one RTI symptom, with the most common symptoms being abnormal vaginal discharge and pain during urination. Reporting of RTI symptoms significantly increases if the woman’s last pregnancy did not end in a live birth or if she has low economic status. Symptom reports also increase with age and decrease with parity. Only one-third of women reporting an RTI symptom sought treatment. The results indicate that survey interviews can be a cost-effective option for measuring the magnitude of RTI symptoms experienced and identifying sociodemographic influences. The findings suggest the need for improved RTI screening procedures and treatment at health facilities in this populous state of northern India.

Using cross-sectional, individual-level survey data from Maharashtra, Tamil Nadu and Uttar Pradesh collected under the Indian National Family Health Survey programme of 1992–93, statistical modelling was used to analyse the impact of a range of variables on the survival status of children during their first 2 years of life. Attention was focused on the potential impact of the mother’s autonomy. The strongest predictors of mortality were demographic and biological factors, breast-feeding behaviour, and use and knowledge of health services. Variables that can be interpreted as being related to maternal autonomy, such as the presence of a mother-in-law in the household, did not have a significant direct effect on child survival at the individual level, and their indirect effects were very limited.