Some recent data are presented on the size and selected sociodemographic characteristics of the Afghan refugee population in Pakistan. Although the official figures show that there were 3·27 million registered Afghan refugees in Pakistan, it is estimated that the actual number may be as high as 3·6 million. There is an excess of females over males, mainly due to war-related activities and excessive casualties particularly among males. While infant and childhood mortality rates are declining and are lower than the levels prevalent in Pakistan, as well as in Afghanistan during the prewar period, the fertility levels among Afghan refugees seem very high indeed.

Data for a sample of 50 developing countries are analysed to investigate the social correlates of the teenage birth rate. Of five major factors considered as predictors of national birth rates (socioeconomic development, family planning programmes, women's status, the sex ratio, and marriage patterns), regression analyses reveal that only the average age at marriage for women has a significant effect on the teenage birth rate. In contrast, all variables except the sex ratio and the average age at marriage for women have a significant effect on the total fertility rate.

Whatever proximate variables are examined, their differential effects on rural and urban fertility are small. This indicates that no major disturbance has taken place in urban or rural reproductive norms. However, two possible reasons for the converging pattern of rural and urban fertility in Nigeria are identified. One is that urban mothers in the first half of the childbearing age range have higher fertility than their rural counterparts. The other is that breast-feeding and post-partum abstinence, which are the major determinants of marital fertility, exert a more depressing influence on rural than urban fertility.

In 1986, a survey was conducted to assess knowledge, health concerns and experience with marketing (retailers) and use (consumers) of Gulaf and Nilocon pills and Kamal vaginal tablets distributed by the Nepal Contraceptive Retail Sales Company (CRS). A sample of 763 consumers of Gulaf, Nilocon and Indian pills and Kamal vaginal tablets, and 361 retailers from a stratified sample of urban medical shops were interviewed. The CRS marketing programme is reaching people who have previously never used family planning; most of the users were practising contraception to limit, not space, births; a high proportion of pill users over 35 smoked; only about a third of CRS pill users had prescriptions or consulted a physician prior to use; CRS training of retailers was found to have increased their knowledge. Recommendations are made for improving communication, education and marketing of CRS contraceptives to ensure their safe and effective use and increase the acceptability of this mode of service delivery.

This report evaluates the decrease in maternal mortality and its relation to family planning methods in Sweden during the years 1911–80. In the 1930s fertility was low but illegal abortions were at a high level and the associated maternal death rate was 18·5 per 1000 women. With the legalization of abortion and the introduction of modern contraceptive methods, the crude reproductive mortality rate in 1965–70 was 1·7 per 100,000 women and this was reduced still further, especially for younger women, by the late 1970s. Standardized reproductive mortality was then 80% higher than the crude rate, indicating the importance of modern family planning methods. Mortality associated with oral contraceptive or IUD use in Sweden during the 1960s and 1970s was lower than in England and the US. Mortality associated with sterilization was 6·2 per 100,000 procedures.

This study of the relationship between mortality risks of siblings born to the same mother shows that, in Bangladesh, the death of the immediately preceding sibling in its infancy has a negative influence on the survival chance of the child in question in its infancy; however, death of the preceding sibling appears to have a positive influence on the index child's survival at ages 1–5 years. Similar results are found for the survival status of the two preceding siblings. Preceding birth interval length and survival status and sex of the immediately preceding sibling are also significant predictors of child mortality between ages 1 and 5 years. Possible explanations may be that the index child faces stronger competition from its immediately preceding brother than from its immediately preceding sister, or that the index child is likely to be looked after more by its preceding sister than by its preceding brother.

In 1984 a prospective study of 1645 women and 1677 births in a rural community in north-eastern Brazil showed the infant mortality rate to be 65 per 1000 live births. Neonatal, post-neonatal and infant mortality are analysed to determine the most important risk factors for each period. Post-neonatal survival depends largely on factors relating to child care, while neonatal deaths are more likely to be associated with biological factors. The principal cause of death, diarrhoeal disease, was responsible for a third of the deaths.

Data from a 1985 survey in two major population centres in Southern Sudan, Juba and Wau, were analysed in order to assess childhood mortality levels and the effect of UNICEF's health care programme. There are continuing high levels of childhood mortality. Logistic regression analysis shows significant positive associations between child survival and immunization, oral rehydration therapy and maternal education.

The influence of some household and maternal variables on three anthropometric nutritional status indices of hospitalized children in Addis Ababa, Ethiopia, are examined. On admission, only 30% of these children can be classified as being of a normal overall nutritional status. There are no significant differences in weight-for-age of hospitalized children between those residing in Addis Ababa and those residing in the rural areas. Income and father's occupation appear to be the major household factors influencing the level of two of the three indices (weight-for-age and weight-for-height). Length of last closed birth interval and, to a lesser degree, maternal age appear to have significant effects on all three nutritional status indices. Upon admission to hospital, children who will in the end survive their hospital stay are on average nutritionally normal or in a mild state of malnutrition, whereas children who will die during their stay arrive in a moderate or severe state of malnutrition. The degree of malnutrition is positively related to the risk of mortality in respiratory disease patients.

In a comparative study to examine the effect of social factors on human twinning, data on sociodemographic and other factors were collected from parents of all twins born alive in Denmark in 1984 or 1985 and from a random sample of parents of singleton infants born in the same years. A postal questionnaire was used. The twins were classified as monozygotic (MZ) or dizygotic (DZ) twins by the similarity method. A trend was found in DZ-twinning, with significantly fewer DZ-twins born in the lower social classes, but not in MZ-twinning. All results were controlled for maternal age and parity.