Consistent findings about the inverse association of social network level with coronary heart disease mortality and morbidity suggest the importance of investigating biological pathways of association. Differences in plasma fibrinogen level were investigated among middle-aged men with weak and strong structural and functional social network ties. Men with low scores in the adequacy of social participation variable (structural) had higher mean values of plasma fibrinogen than those with high scores. The difference remained after adjustment for age, smoking and cardiovascular health status and after possible modifying factors were taken into account, but did not remain significant after allowing for physical fitness. Men with high scores in overall support (functional) had higher plasma fibrinogen levels compared to the men with low scores. This difference persisted after age and cardiovascular health status were taken into account but was explained by smoking. The data suggest that smoking and cardiorespiratory fitness are important mediating or modifying factors between structural and functional aspects of social network ties and plasma fibrinogen.

From the data of the 1989 Bangladesh Fertility Survey, aggregate deaths reported at ages 0-12 and 13-60 months are used to estimate infant and child mortality. Multivariate analysis shows that preceding birth interval length, followed by survival status of the immediately preceding child, are the most important factors associated with differential infant and child mortality risks; sex of the index child and mother's and father's education are also significant. Demographic factors are influential during infancy as well as childhood, but social factors, particularly mother's and father's education, now emerge as significant predictors of infant mortality risks. This indicates a change in the role of socioeconomic factors, since the earlier Bangladesh Fertility Survey in 1975.

It is widely believed that people can remember the age at which they first had sexual intercourse. Questions about age at onset are routinely asked in population sexual behaviour surveys and in clinical history-taking. However, there are limited test–retest data, especially with regard to individual differences in unreliable recall. In this study, telephone interviews and follow-ups an average of 15 months later were conducted with 570 non-virgin subjects aged between 28 and 73 years. Test–retest correlations for recalled age at first intercourse were 0·85 for females and 0·91 for males. Consistency was slightly lower among older people and women with a history of sexual abuse. There were no significant associations between consistency of recall and measures of personality, educational background or history of alcohol dependence and depression.

The marital structure of Italians living in Boston, Massachusetts, in the period 1880–1920 was studied in order to explore the integration process in the urban context. The study analyses endogamy and inbreeding, using data on 15,579 marriages from the parish books of the three Italian parishes of Boston. Endogamic rates are very high and increased in time, ranging from 93·9% to 97·3%. This correlated with the growth of the Italian community and the decline of the biased sex ratio. One parish, Our Lady of Pompeii in the South End, displays lower endogamic rates because of the reduced and scattered population attending it. The rate of consanguineous (2·33%) and isonymous (6·38%) marriages, and the coefficients of inbreeding, alpha; (0·98×10−3) and Ft (0·0159), are similar to those of north Italian populations, and lower than those for south and insular Italy. The parish of Our Lady of Pompeii shows consistently higher values than the other two parishes. Marriages between first and second cousins are the main cause of the above values in each parish. Consanguineous marriages and inbreeding increased over time, from the 1890s, and this is in general agreement, although slightly delayed, with the Italian trend.

A study of urinary schistosomiasis in Umueze-Anam, Anambra State, Nigeria, showed a Schistosoma haematobium infection of 26% (85) among school children with no significant difference by sex except when age as a variable is introduced. Eleven percent (37) of the 333 children were positive for haematuria; all these 37 children lived within 1·0 km of the water sources. Of the 85 infected children, swimming and laundering accounted for 65% and 48% of all water contact activities, for boys and girls respectively. One-third of the 230 adults interviewed believed haematuria to be a venereal disease and 20% thought it was a sign of maturity. Individual perception of causation and seriousness of haematuria differed by level of education and by sex. Less than 2% of the respondents knew that snails transmitted the disease. The effects of social restrictions on the epidemiology of infection is discussed.

The relationships between selected demographic and socioeconomic variables and current use status of contraception were examined using logistic regression technique. Information on current contraceptive use was provided by 3156 out of 4488 currently married women aged 15–49 interviewed in the 1988 Ghana Demographic and Health Survey. Respondents' approval of family planning emerged as the most important predictor of current contraceptive use, followed by discussion of family planning with partner and level of education. As a policy measure, information, education and communication programmes on family planning should be intensified, particularly in rural areas. Female education, at least up to secondary level, should be given top priority.

A survey of consanguineous marriages in Japan was conducted on 1 September 1983, by questionnaires. The total number of couples surveyed was 9225. They were chosen from six widely different areas and the inbreeding coefficients from isonymy and pedigrees were estimated for each area. Random inbreeding remained constant with the marriage year whereas total (F) and non-random (Fn) inbreeding from isonymy and inbreeding from pedigrees (α) decreased with the marriage year in each area. Estimates of genetic microdifferentiation from surnames were obtained; estimates of the reduced variance indicated low microdifferentiation (0·000014-0·000081) in five areas and a higher value in Fukue City (0·0004). The ratio of FST (the average within-group a priori kinship) to α-value ranged from 0·42 in Kawanishi City to 3·9 in Minobu-Cho.

To examine the association of environmental factors (including housing) with respiratory infections and diarrhoea in children under 5 years of age, a cross-sectional study of 403 families was conducted in a squatter settlement of Karachi. In the 2-week period before the survey prevalence of diarrhoea and respiratory infections was 14·4% and 15·0% respectively. The factors significantly associated (p>0·05) with diarrhoea in households in the multivariate analysis were: number of children under 5, regular cleaning of sewers, storage of water in small utensils and cooking inside a one room house. Five households risk factors were identified for respiratory infections: the number of children under 5, poor housing structure, increased household density, humidity inside the house less than the humidity outside house, and frequent cooking inside the house. These results suggest that children under 5 years of age in lower income areas are at additional risk to health because of poor environmental conditions.

This paper outlines the demographic characteristics of the Baiga tribe, one of the most primitive of the aboriginal tribal groups of Central India. The Baiga population has grown steadily since the first anthropological study of the tribe in the 1930s. Age at menarche, age at marriage, breast-feeding, and time interval between marriage and first conception are natural. There are more females than males. Sub-tribe endogamy is common; consanguineous marriage is favoured (34% of marriages are between first cousins) and marital distance is low (mean 7·1 km). Though the mean menarcheal age is high (15·2 years), mean age at first marriage is low (16·6 years), giving a long reproductive period.

Multifactorial analyses of data from representative British and German national contraception surveys were used to examine the principal demographic determinants of contraceptive use by women. Contraceptive use appeared to be determined mainly by reference to ‘reproductive status’ (the combined impact of age, marital status, parity and future child wish). Women who were postponing pregnancies were using oral contraceptives, whereas those who wanted no more children relied more on intrauterine devices or sterilisation. Differences between the countries suggested that the choice of contraceptive method was influenced by health care policy, the organisation of the relevant services and differential provider preferences. The contraceptive method used was also related to having occasional rather than steady sexual partners (more condom use), lower educational level (less oral contraceptive use) and frequent church attendance (greater use of condoms and periodic abstinence). Contraception decisions appeared to follow a fixed pattern, based more on a couple's demographic situation (reproductive status, country, educational level and religious beliefs) than on the characteristics of the contraceptive methods. This resulted in an unnecessarily restricted choice of methods.

Illegitimacy in a historic, single community at Penrith, Cumbria (1557–1812), has been studied using aggregative analysis, family reconstitution and time series analysis. This population was living under extreme conditions of hardship. Long, medium and short wavelength cycles in the rate of illegitimacy have been identified by time series analysis; each represents a different response to social and economic pressures. In a complex interaction of events, the peaks of the cycles in wheat prices were associated with rises in adult mortality which promoted an influx of migrants and a concomitant rise in illegitimacy. The association between immigration and illegitimacy was particularly noticeable after the mortality crises of the late sixteenth and early seventeenth centuries. Children of immigrant families also tended to produce illegitimate offspring. Native and immigrant families responded differently to extrinsic fluctuations, and variations in their reproductive behaviour were probably related to access to resources.

Parental investment theory postulates that where physical condition varies significantly then birth sex ratio will be correlated with social status. Application of this theory to man remains contentious. This study examines physique, wealth, and social status in relation to the sex of live births. It reports a female-biased sex ratio in high social and economic status Nepalese. Close consanguineous marriage, intended to conserve landed wealth within related lineages, and increased female work burdens accompanying larger farm size, are proximate factors which may underlie this finding. A differential payback hypothesis is one way of explaining this pattern.

This paper proposes a causal model of sexual activity among a randomly selected sample of 305 Junior secondary school girls in Zambia. The results indicate that liberal sexual attitudes influence romantic involvement with boys. Emotional involvement is likely to result in sexual activity. Traditional courtship forms are slowly being replaced by modern patterns of courtship behaviour. Policy and programme implications are discussed.

Using data from the 1988 Ghana Demographic and Health Survey, this study examines couples' demographic and socioeconomic characteristics in the context of their attitudes towards family planning, and the impact of these factors on the use of contraceptives. The characteristics of the husbands and their influence on wives' behaviour illustrate the role of intra-household relations between men and women and their effect on fertility-related behaviour in patriarchal African societies.

Data on patterns of marriage, differential fertility and mortality were collected from 211 Kotia women residing in Visakhapatnam district of Andhra Pradesh, India. Consanguineous marriages made up just over a quarter of the total, and of these, father's sister's daughter (FSD) were more common than mother's brother's daughter (MBD). The mean inbreeding coefficient for the sample (F) was 0·0172. Women in consanguineous marriages had a lower mean number of total conceptions, live births and living offspring (net fertility) than women in non-consanguineous marriages. Significant heterogeneity was found in the means of living offspring for FSD, MBD and non-consanguineous couples, but not for conceptions and live births.

Psychological determinants of contraceptive use were investigated in Great Britain and Germany, using national data obtained in 1992. It was hypothesised that current contraceptive use among sexually active, fertile women aged 15-45 was related to their attitude towards the various contraceptive methods, social influences, perceptions of being able to use a method correctly and consistently, a correct estimation of fertility, and communication with their partner. Effects of age and country were also taken into account.

The attitude of respondents towards the various contraceptive methods was ambivalent and no method was seen as ideal. On medical methods (OCs, IUDs and sterilisation) many respondents expressed doubts as to their safety for health. Social influences most frequently concerned the use of OCs. Respondents considered themselves able to use oral contraceptives correctly, but expressed general fear about intrauterine devices and sterilisation, and many women believed they were not able to use condoms and periodic abstinence consistently.

Multifactorial analyses revealed that current contraceptive use was principally determined by social influences, attitude and self-efficacy with respect to medical methods. Age and country, and, for use of unreliable methods, fertility awareness also played a role. Communication with the partner was less relevant.

Contraceptive choice (and the use of non-medical methods) depended greatly on encouragement to use and being in favour of medical methods. A lack of social support for use of medical methods and a negative attitude towards them was related to higher use rates of condoms, periodic abstinence, withdrawal and reliance on ‘luck’. In the case of withdrawal and/or no method, underestimation of fertility played an additional role.

Contraceptive choice appears to be determined more by a general like or dislike of medical methods rather than on a weighing of the merits of individual available methods.

Retrospective and prospective data collected in Cameroon were used to reassess hypotheses about how infant and early childhood mortality is affected by birth spacing and breast-feeding. These data show that: (a) a short preceding birth interval is detrimental for child survival in the first 4 months of life; (b) full and partial breast-feeding have direct protective effects on child survival in the first 4-6 months of life, with the effects of the former stronger than those of the latter; (c) early subsequent conception significantly increases mortality risks in the first 16 months of life of the index child. These findings are robust to various controls, e.g. study design, data defects, child's health conditions at/around birth, postnatal maternal and child recurrent illnesses, patterns of utilisation of health care services, and immunisation status of the child.

The Cerdanya valley in the eastern Pyrenees has a physical unity into which a political frontier has been imposed to divide it. The social and cultural repercussions of this Franco-Spanish border have created obstacles to marriage which are not due to topography. Choice of month of marriage is under cultural control and the study of seasonality in marriages recorded in the registers of all the Cerdan parishes on both sides of the border demonstrated differences over time and between French and Spanish sectors. It is suggested that these changes demonstrate the process of distancing of the two populations. Cluster and correspondence analysis showed progressive differentiation of the seasonality patterns of the French and Spanish Cerdans despite the geographic unity of the valley. Sociocultural factors are presumed responsible.

Patterns of reproduction associated with extramarital conception are examined using data on non-marital births, marital births occurring during less than 8 months after marriage, and spontaneous and induced abortions experienced by unmarried women. Trends in the incidence and demographic outcomes of conceptions resulting from extramarital coitus are analysed by means of age-specific probabilities of becoming pregnant outside marriage; and of terminating such a pregnancy by abortion, by legitimating it through marriage before confinement, or by having a baby while remaining unmarried. Substantial increases in the proportion of extramaritally conceived pregnancies leading to non-marital births are detected for the period since the late 1980s, and ascribed mainly to rising levels of unmarried cohabitation. The demographic effects of the post-1989 transition from state to market economy are discussed.

Classification of the boys in 48 families of three or more brothers according to their Moray House Test (VIQ) scores awarded at age 11±1½ years in 1947–75 confirms the postulated existence of a total of nine male phenotypes for this X-linked trait in the score range <70 to 140 points. The phenotypic means lie close to the sequence 69(8)133. Recombination is shown to occur. An additive effect of three alleles at each of two X-chromosome loci is the most likely simple scheme. An iterative approach matching the observed mixtures of proposed genotypes in the families with probabilities for variable levels of homozygosity, non-crossover/crossover heterozygosity, and maternal meiotic recombination yields a best goodness of fit where this last is 20%.