This study investigates the effect of both total and unsupplemented breast-feeding in conjunction with birth interval on early childhood mortality, using longitudinal data from Matlab, Bangladesh. A discrete hazard model approach shows that it is not the duration of total breast-feeding but the duration of unsupplemented breast-feeding which increases child survival. Unsupplemented breast-feeding appears as such a crucial determinant of early childhood mortality that its effect could not be substantially attenuated even when important demographic and socioeconomic factors were controlled. Each of the covariates—supplementation, previous birth interval and onset of a subsequent conception—has an independent influence on early childhood mortality.

A new analysis of a randomised double-blind trial confirms the fact that vitamin–mineral supplementation has a statistically significant effect on non-verbal IQ as measured by the WISC-R test. Further studies may show if this effect is large enough to be of practical importance.

Data from the 1991 Belize Family Health Survey show differentials in the use of maternal and child health services between ethnic groups (Creole, Mestizo, Maya/Ketchi and Garifuna). Multivariate analysis is used to explore whether such differentials can truly be attributed to ethnicity or to other characteristics that distinguish the ethnic groups. Health services considered are: family planning, place of delivery (hospital/other), postpartum and newborn check-ups after a birth, and immunisations for children. The language usually spoken in the household is found to be important for interpreting ethnic differentials. Mayan-speaking Maya/Ketchis are significantly less likely to use family planning services or to give birth in a hospital. Spanish-speakers (Mestizos and Maya/Ketchis) are less likely to use newborn and postpartum check-ups, after controlling for other characteristics. There are no ethnic differentials for immunisations. Programmatic implications of these results are discussed.

The relationship between chronic parasitic loads and psychopathology is examined. Research suggests that: (1) psychiatric disorders are more prevalent in less developed countries where parasitic loads are more endemic; (2) individual patients with parasitic loads are more likely to exhibit mental status changes; (3) there is an improvement in the mental status of a subset of psychiatric patients following treatment for parasites. Inasmuch as the prevalence of parasitic content is underestimated in both more developed and less developed countries, an awareness of this relationship should increase understanding of the needs of infected individuals, and be taken into consideration when evaluating psychiatric patients.

A survey of 673 consecutive Papua New Guinea parturients carried out at the Port Moresby General Hospital between May and June 1990 showed that socioeconomic and educational factors played a part in predicting perinatal death. Mothers who have previously experienced a perinatal death are more likely to experience a second one.

This paper investigates the effects of continued breast-feeding after resumption of menses on fertility, using data from two retrospective surveys in India and single decrement life table and multivariate time-dependent hazards analyses. Breast-feeding even after the return of menses is found to be associated with longer birth intervals. The interaction of breast-feeding duration after resumption of menses and postpartum amenorrhoea has a significant effect on the risk of conception after return of menses.

The relationships between 29 cephalic and post-cephalic body dimensions and the length and regularity of the menstrual cycle, as well as the duration of menstrual bleeding, have been investigated in 327 Austrian women. Five of the eight head measurements and seventeen of the 21 body measurements correlated statistically significantly with the individual menstrual variables. Measures of facial and post-cephalic robustness and the amount of subcutaneous fat tissue correlated negatively with the length of cycle, as well as with the difference between minimum and maximum cycle length, a measure of the cycle regularity. The post-cephalic height and length dimensions correlated significantly positively with bleeding duration. The results indicated a marked influence of body type, especially the amount of subcutaneous fat, on the length and regularity of the menstrual cycle.

A biosocial model of the effects of early adolescent testosterone levels and religiosity on adolescent males' sexual attitudes and activity over a 3-year period was examined. Using panel data for approximately 100 boys who were 12·5/13·0 years old at study entry, significant additive effects of free testosterone and frequency of attendance at religious services were demonstrated on the transition to first intercourse and other aspects of sexual behaviour and attitudes. No interactive effects of the two predictors were found. Boys with higher free testosterone levels at study entry who never or infrequently attended religious services were the most sexually active and had the most permissive attitudes. Boys with lower free testosterone who attended services once a week or more were the least active and reported the least permissive attitudes. For some behaviours, differences between free testosterone/attendance groups increased over time, resulting in substantial behavioural differences by the final round of measurement 3 years later.

A comparison of contraceptive and fertility data for 1985–91 with data for 1983 shows that fertility has continued to decline in Bangladesh, in all segments of society. The magnitude of decline varied according to educational level, region and urban–rural locality. The percentage decline in total marital fertility rate was somewhat higher among urban than rural residents; educated women showed greater declines than uneducated, increas-ing the overall educational differences in total fertility by 1991. Factors contributing to the recent decline in fertility are discussed.

This study examines the effects of birth order and interpregnancy interval on birthweight, gestational age, weight-for-gestational age, infant length, and weight-for-length in a sample of 2063 births from a longitudinal study in the Philippines. First births are the most disadvantaged of any birth order/spacing group. The risks associated with short intervals (<6 months) and high birth order (fifth or higher) are confined to infants who have both attributes; there is no excess risk associated with short previous intervals among lower-order infants, nor for high birth order infants conceived after longer intervals. This pattern is observed for all five birth outcomes and neonatal mortality, and persists in models that control for mother's age, education, smoking, family health history and nutritional status. Since fewer than 2% of births are both short interval and high birth order, the potential reduction in the incidence of low birthweight or neonatal mortality from avoiding this category of high-risk births is quite small (1–2%).

A retrospective study of 21,442 Singaporean women who gave birth at the National University Hospital, Singapore, between January 1986 and November 1991 is used to assess the effects of increasing age on obstetric performance. The results show that reproduction after the age of 35 years in Singapore is associated with a higher incidence of antenatal complications such as hypertension and diabetes and a higher rate of obstetric intervention. However, given the current level of obstetric and neonatal care in Singapore, these adverse features do not prejudice the obstetric and neonatal outcomes.

It is usually supposed that androgens play a major role in the aetiology of prostatic cancer. This note suggests that the association of the disease with low (as well as high) coital rates and its apparent association with vasectomy can both be reconciled with the hypothesised androgenic involvement.

This paper examines the current fertility of men and women in the Western area of Sierra Leone and the prospects for future fertility behaviour. Probably due to the effect of rapid economic decline in Sierra Leone since 1980, the desired family size has fallen. But indications are that the preferred completed family size is lower than the desired family size