This paper presents the results of a longitudinal intervention study carried out in the Davao del Norte province of the Philippines. The intervention, tested through a quasi-experimental design, consisted of training of family planning service providers in information exchange and training of their supervisors in facilitative supervision. The training intervention significantly improved providers' knowledge and quality of care received by clients. Moreover, good quality care received by clients at the time of initiating contraception use increased the likelihood of contraceptive continuation and decreased the likelihood of both having an unintended pregnancy and an unwanted birth. However, comparison of women in the experimental group with those in the control group did not show any significant effect of provider-level training intervention on these client-level outcomes. The reasons for this conundrum and the implications for quality of care are discussed.

This paper argues for the inclusion of ethnography as a research methodology for understanding the effects of public health policy. To do this, the implementation of DOTS (Directly Observed Therapy, Short-course) – the World Health Organization (WHO) prescribed policy for the control of the infectious disease tuberculosis – is explored in the context of Nepal. A brief history of DOTS and its implementation in Nepal is outlined, and the way it has been represented by those within the Nepal Tuberculosis Programme (NTP) is described. This is followed by an outline of the research done in relation to this, and the ethnographic methods used. These ethnographic data are then interpreted and analysed in relation to two specific areas of concern. Firstly, the effects around the epidemiological uses of ‘cases’ is explored; it is argued that a tightening of the definitional categories so necessary for the programme to be stabilized for comparative purposes has profound material effects in marginalizing some from treatment. Secondly, the paper examines some of the implications and effects relating to the way that the ‘directly observed’ component was implemented. The discussion explores how current debate on DOTS has been played out in some medical journals. It argues for the importance of ethnography as a method for understanding certain questions that cannot be answered by particular, and increasingly dominant, research ideologies informed by randomized controlled trials. This raises important issues about the nature of ‘evidence’ in debates on the relationship of research to policy.

Puberty is defined in the Oxford English Dictionary as the state or condition of having become functionally capable of procreating offspring, suggesting that it is a discrete event that occurs at a particular point in time. Whilst it is easy to determine the age at puberty in animals from the time of completion of spermatogenesis in the male, or the first ovulation in the female, these are particularly difficult criteria to evaluate in our own species. Thus human puberty is generally described in terms of a sequence of changes in secondary sexual characteristics occurring over an extended period of time, and these have been reviewed in detail by Marshall & Tanner (1974).

Nuñoa is a high altitude rural Peruvian community characterized by socioeconomic stratification and differential access to the market economy. Nursing practices and the effects of nursing are also stratified; this translates into a risk of menses at 12 months post-partum nearly seven times greater in wealthier than in poorer women.

Most nursing occurs in the morning, among those who practise on-demand breast-feeding. Nursing episodes are clustered into sessions; the amount of breast-feeding is regulated by varying the number of episodes per session rather than by changing the duration of suckling episodes or the frequency of sessions per hour. Thus, resumption of ovulation is not dependent on the variable spacing of nursing episodes or sessions. The components of nursing activity most likely to be responsible for variation in the duration of post-partum amenorrhoea in these nursing women are mean session duration and mean number of episodes per session.

The mean duration of morning nursing sessions is negatively associated with infant's age, reflecting the greater reliance of younger children on breast-milk. The mean duration of afternoon nursing sessions is positively associated with mother's age, independent of infant's age, possibly reflecting maternal age-related variation in milk production capabilities. Baby minding by older daughters may also help to explain variation in afternoon nursing.

Consanguineous marriage has been the culturally preferred form of marriage in Iran. This paper examines the extent to which education, urbanization and changes in modes of economic production have affected the incidence of consanguineous marriage and attitudes towards consanguineous marriages. The 2002 Iran Fertility Transition Survey conducted in the four provinces of Gilan, Sistan and Baluchistan, Yazd and West Azarbaijan provides information on the degree of relationship of marriage partners from around 6550 ever-married women aged 15–49. Attitudinal data were also obtained. Overall, the level of marriage to biological relatives ranged from 23% in Gilan to 78% in Sistan and Baluchistan. The paper finds that the practice of marriage to biological relatives has remained surprisingly resilient in the face of modernizing influences and that ethnicity, province and area of residence remain important determinants. On the other hand, attitudes have shifted towards marriage with a non-relative. Anthropological research would illuminate the processes of consanguineous marriage in Iran.

This study examines the relationship between blood pressure and the religious practices of Kuwaitis as members of a Muslim society. Religious variables were measured via a sociocultural questionnaire. Blood pressure measurements were taken with a sphygmomanometer. Non-opportunistic samples were taken from 223 Kuwaitis. The difference in religious commitment between Muslim Sunnis and Muslim Shiites was examined using a t-test. Matrix correlation was used to examine the relationship between religious commitment and some other variables. Multiple regression was conducted to determine the effect of religiosity on blood pressure, as well as statistically controlling for other variables such as body mass index, socioeconomic status, smoking, gender and age. It was found that both systolic and diastolic blood pressure were affected by religious commitment and religious activities.

A comparison was made between 1308 mothers delivered of twins in Aberdeen between 1950 and 1969 and mothers of singleton infants. There was a greater number of tall women amongst mothers of twins. This was more marked in primigravidae than in multiparae. The rate of twinning was less in thin women than in those who were obese or of normal build. The social class of the husband did not appear to influence the incidence of twins. The rate of weight gain in twin pregnancies, even in early pregnancy, was high compared with the rate in singleton pregnancies. The more frequent occurrence of twinning in heavier, taller women, suggests that nutrition might be responsible but this is not borne out by the findings about social class.

The 1988 Malawi Traditional and Modern Methods of Child Spacing Survey data are used to identify determinants of infant mortality in Malawi. The logistic binomial analysis shows that socioeconomic factors are significant even during the neonatal period while the length of the preceding birth interval is significant in the post-neonatal period only. There is a strong familial correlation of mortality risks during both the neonatal and post-neonatal periods but the effect of geographical area of residence is stronger in the post-neonatal period.

This paper examines the effect of son preference on the hazards of having a second and a third birth. With data from the Two-per-thousand National Sample Survey on Fertility and Contraception conducted in 1988 by the State Family Planning Commission of China, the hazard of having a second birth among 62+ thousand married women who have had a first birth, and the hazard of having a third birth among 43+ thousand married women who have had two births was examined. These two hazards (i.e. the hazard of moving from the first to the second birth, and the hazard of moving from the second to the third birth) were analysed by estimating Cox proportional hazard models. The major covariate in the first analysis is whether or not the first-born was a daughter. In the second analysis the main covariate is whether both of the first two children were girls. In both models seven covariates known to have independent effects on the transition to a second (or third) birth are controlled for, namely, whether the woman is a Han, whether she is a farmer, her age at the birth of the first (or second) child, whether she had her first (or second) birth prior to the initiation in 1979 of the one-child policy, and three dummy variables reflecting her level of education. The results show the important influence of son preference on the hazard of having another birth.

This paper explores the relationship between body mass and risk of death among US adults. The National Health Interview Survey–Multiple Cause of Death linked data set is used for the years 1987–1997, and Cox proportional hazard models are employed to estimate the association between obesity, as measured by the body mass index (BMI), and overall, circulatory disease-specific and diabetes-specific mortality. A U-shaped relationship is found between BMI and overall mortality. Compared with normal weight individuals, mortality during the follow-up period is 34% higher among obese class II individuals and 77% higher among obese class III individuals, controlling for age and sex. A J-shaped relationship exists between circulatory disease mortality and obesity, with a slightly higher risk of death for all categories of BMI. The relationship between BMI and diabetes mortality is striking. Compared with normal weight individuals, obese class I individuals are 2·8 times as likely to die, obese class II individuals are 4·7 times as likely to die, and obese class III individuals are 9·0 times as likely to die of diabetes during the follow-up period, controlling for age and sex. These results demonstrate that obesity heightens the risk of overall and circulatory disease mortality, and even more substantially increases the risk of diabetes mortality. These mortality findings, together with the substantial recent increases in obesity, lend urgency to public health programmes aimed at reducing the prevalence and consequences of obesity.

Uganda is invoked as a metaphor for a host of arguments and insights about HIV/AIDS. However, much of what has been asserted about the country is not based on the available evidence. This paper reviews findings by epidemiologists and anthropologists, and draws on the author’s experiences of researching in the country since the early 1980s. It comments on various myths about HIV/AIDS in Uganda, including myths about the origin and dissemination of the disease, about the links between HIV/AIDS and war, and about declining rates of infection. It shows that much less is known about Uganda than is commonly supposed, and it offers some alternative hypotheses for interpreting HIV prevalence and incidence data. In particular it draws attention to the importance of mechanisms for social compliance. It concludes by raising concerns about the current enthusiasm for provision of anti-retroviral drugs.

Seven hundred and ninety six Thai women who stopped using the long-acting injectable contraceptive depot medroxyprogesterone acetate (DMPA, Depo Provera), 437 women who stopped using oral contraceptives and 125 women who had an IUD removed to have a planned pregnancy, were followed up to ascertain the delay to conception after the end of contraception and to determine the proportion of women who did not conceive in the 4 years after discontinuation. The median delay to conception was 5·5 months plus the estimated duration of the effect of the last injection of DMPA, 3 months for oral contraceptives and 4·5 months after discontinuing the IUD. The proportion of women who did not conceive within 9 months after discontinuation of DMPA is similar to that of ex-IUD users, and by 3 years to that of the ex-pill sample. There is no evidence to suggest that prolonged use of DMPA increases the delay to conception, and the return of fertility among never pregnant ex-users resembled that of ever pregnant ex-users.

There were comparable proportions of live births among ex-DMPA users and ex-pill users and both of these showed higher proportions of live-births than ex-IUD users. There was no evidence to suggest that previous use of DMPA had any significant adverse effect on the outcome of pregnancy of the subsequent births. This study did not show any association between infertility and the previous use of DMPA or other contraceptives.

A postal questionnaire was sent out in September 1977 to all women aged between 40 and 60 years in Ede, the Netherlands. The design of the study was similar to that used by Jaszmann et al. in 1967 in the same locality. The response rate was 66·6%. Data were analysed using theprobit analysis, a segmentation analysis and by applying chi-squared tests to cross tabulations. The median age at menopause was 51·7 years, 3 months later than in 1967. When, however, women who were taking oral contraceptives at the time of the survey were excluded the median age at menopause was found to be virtually the same as 10 years previously. The effect of factors associated with early or late menopause was studied. The results are discussed. The concept of a secular trend is briefly commented upon.

The aim of this paper is to account for the effect of mother's death on child survival in a historical population. Using comprehensive data on the early French Canadian population of Quebec, evidence is provided for a higher risk of dying for motherless children that remains significant over all childhood and long after the death of the mother. The specific effect of the loss of maternal care was estimated by comparing mortality before and after mother's death, furnishing a means to control for family heterogeneity. No differential in investment between genders was detected before age 3, but older girls suffered a three-fold higher susceptibility to mother's death than their male counterparts. This suggests that grown-up girls assuming the responsibilities of the missing mother had a lower chance of survival.

The remarkable decline in fertility in Iran, which saw the total fertility rate fall from 7 children per woman in 1986 to 2 in 2000, has received only limited analysis in the demographic literature. Using the 2000 Iran Demographic and Health Survey and Bongaarts’ age-specific fertility model, this paper examines the role of the major proximate determinants of fertility in bringing about the rapid decrease in fertility in Iran. The analysis indicates that contraception had the largest effect on fertility, accounting for 61% of the reduction in fertility from its theoretical maximum. The fertility-inhibiting effect of marriage patterns accounted for an additional 31% reduction, and was most important among the young. Further analysis of contraceptive behaviour suggests that the current period fertility rate of 2·0 children per woman is an outcome of a synchronization of delaying and spacing of births among younger women with stopping of childbearing among women in the middle and late reproductive ages. The policy implications of the results are discussed.

A study has been made of twinning rates throughout the world over the last two decades. There has been a decline in age-specific dizygotic twinning rates in almost all of the developed countries during this period. Exceptions to this trend are provided by the United States, where the only decline in twinning rates in the last two decades was in births to older women in the 1960s, and by Czechoslovakia, Hungary and Bulgaria, where an initial decline in twinning during the 1960s was followed by stable twinning rates.

It is suggested that the cause may be some form of environmental pollutant, perhaps a pesticide, which has been the subject of restrictive legislation first in the United States and later in the three Communist countries. Possibly the cause of this decline may be identified by a study of such legislation.

This seems to be the first study of twinning rates in Central and South America. The low twinning rates in some of the countries there may indicate the genetic affiliations of their inhabitants with those (of Mongoloid origin) of countries in the Far East.

While the ‘immigrant health paradox’ posits better health behaviours and outcomes for immigrants upon arrival to the US, research suggests that this advantage may deteriorate over time. This study analysed the relationship of acculturation and breast-feeding initiation and duration among a sample of predominantly Latina, low-income women in the US. The four measures of acculturation included: mother’s nativity (foreign born vs US born), mother’s parents’ nativity (foreign born vs US born), years of US residence (<8 years vs ≥8 years) and a dichotomous measure of language acculturation adapted from three items on Marin’s acculturation scale (preferred language spoken at home, reading language and writing language) as exclusive use of native language versus non-exclusive use (mixed or English only) (Marin et al., 1987; Marin & Gamba, 1996). Final multivariable models showed that mothers who exclusively used their native language were more likely to initiate breast-feeding as well as breast-feed for longer duration compared with mothers with non-exclusive use, whereas years of US residence and mother’s nativity were not significantly associated with breast-feeding initiation or duration. Mother’s parents’ nativity also emerged as a significant predictor of breast-feeding duration, both within final models for immigrants and across study participants. Programmes providing nutrition education to low-income women may wish to consider the role of language as an important determinant of breast-feeding. The role of mother’s parents’ nativity on breast-feeding practices deserves exploration in future studies, as the cultural practices taught by family members born outside the US may exert strong pressure within immigrant families now living in the US.

This paper focuses on the clinical and social diagnostics of stroke-like symptoms in Limpopo Province, South Africa. The research questions addressed here are: what are the lay understandings of strokelike symptoms and what are the health-seeking behaviours of Tsongan Mozambican refugees and South Africans in this area? The study site is ten villages in the Agincourt sub-district of Limpopo Province which are within the health surveillance area of the Agincourt Health and Population Unit (AHPU) of the University of Witwatersrand. The population are Tsongan who speak Shangaan and comprise self-settled Mozambican refugees who fled to this area during the 1980s across the nearby border and displaced South African citizens. The latter were forcibly displaced from their villages to make way for game reserves or agricultural development and moved to this area when it was the former ‘homeland’ of Gazankulu. The team collected data using rapid ethnographic assessment and household interviews as part of the Southern Africa Stroke Prevention Initiative (SASPI). The main findings are that stroke-like symptoms are considered to be both a physical and social condition, and in consequence plural healing using clinical and social diagnostics is sought to address both these dimensions. People with stroke-like symptoms maintain their physical, mental and social well-being and deal with this affliction and misfortune by visiting doctors, healers, prophets and churches.

During the first half of this century, the seasonal pattern of births in European countries showed a major peak in the spring and a minor peak in the autumn. In contrast, the pattern in the US was of a minor peak in spring and a major peak in autumn. Over the last 20 years, the pattern in England and Wales has changed to resemble the US pattern, and the same seems to be true of several other European countries. A hypothesis is offered to account for the difference between the European and the US patterns and for the change from one to the other in some countries.

The magnitude of seasonality correlates positively with latitude: it is suggested that this is partially consequent on variation in luminosity.

This analysis investigates the determinants of contraceptive discontinuation in six developing countries, using data from Phase I surveys of the DHS programme. Cumulative probabilities of discontinuation at 24 months for reasons other than the desire for another child were examined. By this time, typically about 40% of couples have stopped use and most are subsequently at risk of an unwanted conception. Discontinuation of IUD use was found to be less common than for other methods, partly perhaps because cessation of use requires a deliberate decision to have the device removed.

The most important results are negative ones. Neither the schooling of couples nor their type of residence exerted appreciable influence on discontinuation. The policy and programme implications are discussed. Prior use of a method, fertility preferences and the related demographic factors of age and family size emerged as pervasive predictors of discontinuation.