[Change of protocol in the verbal autopsy method and measure of malaria mortality in rural areas in Senegal]. / Changement de protocole dans la méthode d'autopsie verbale et mesure de la mortalité palustre en milieu rural sénégalais.

In rural Senegal, three populations have been followed up since several decades and the malaria mortality trend has been observed since the mid-1980s. However, limits are associated with the verbal autopsy method used to determine causes of death, especially deaths due to malaria. A change in protocol occurred in recent years in two of these three sites with the involvement of two physicians (instead of only one) in the diagnosis. The objective is here to measure its potential impact on diagnosis of malaria deaths. Five hundred and fifteen diagnoses reported on child deaths occurred from 2000 to 2005 have been analysed. We have identified, on the basis of a multinomial logistic regression model, factors affecting the determination of malaria deaths among the characteristics of the child, the death, the illness and its symptoms, and we also took into account method factors. Factors related to the method do not play on the malaria diagnosis. This result insures the continuity of the series on malaria mortality statistics since 2000 in the two sites despite changes in the method. However, the new protocol leads to vanish possibility of having deaths among ill-defined or unknown causes. In the African context of absence of health statistics, data obtained through the verbal autopsy method in demographic surveillance system can provide accurate information in the epidemiological field, even regarding malaria.

Mobility and the spread of human immunodeficiency virus into rural areas of West Africa.

BACKGROUND: In eastern and southern Africa, the human immunodeficiency virus (HIV) epidemic appeared first in urban centres and then spread to rural areas. Its overall prevalence is lower in West Africa, with the highest levels still found in cities. Rural areas are also threatened, however, because of the population's high mobility. We conducted a study in three different communities with contrasting infection levels to understand the epidemiology of HIV infection in rural West Africa. METHOD: A comparative cross-sectional study using a standardized questionnaire and biological tests was conducted among samples in two rural communities of Senegal (Niakhar and Bandafassi, 866 and 952 adults, respectively) and a rural community of Guinea-Bissau (Caio, 1416 adults). We compared the distribution of population characteristics and analysed risk factors for HIV infection in Caio at the individual level. RESULTS: The level of HIV infection was very low in Niakhar (0.3%) and Bandafassi (0.0%), but 10.5% of the adults in Caio were infected, mostly with HIV type 2 (HIV-2). Mobility was very prevalent in all sites. Short-term mobility was found to be a risk factor for HIV infection among men in Caio (adjusted odds ratio (aOR) = 2.06; 95% CI: 1.06-3.99). Women from Caio who reported casual sex in a city during the past 12 months were much more likely to be infected with HIV (aOR = 5.61 95% CI: 1.56-20.15). Short-term mobility was associated with risk behaviours at all sites. CONCLUSIONS: Mobility appears to be a key factor for HIV spread in rural areas of West Africa, because population movement enables the virus to disseminate and also because of the particularly risky behaviours of those who are mobile. More prevention efforts should be directed at migrants from rural areas who travel to cities with substantial levels of HIV infection.

[Mortality from snake bites, wild and domestic animal bites and arthropod stings in the savannah zone of eastern Senegal]. / La mortalité par les morsures de serpent, d'animaux sauvages et domestiques et les piqûres d'arthropodes en zone de savane soudanienne du Sénégal oriental.

From 1976 to 1999, we conducted a prospective study of overall and cause-specific mortality among the population of 42 villages of south-eastern Senegal. Of 4,228 deaths registered during this period, 26 were brought on by snakebites, 4 by invertebrate stings and 8 by other wild or domestic animals. The average annual mortality rate from snakebite was 14 deaths per 100,000 population. Among persons aged 1 year or more, 0.9% (26/2,880) of deaths were caused by snakebite and this cause represented 28% (26/94) of the total number of deaths by accident. We also investigated the snake fauna of the area. Of 1,280 snakes belonging to 34 species that were collected, one-third were dangerous and the proportion of Viperidae, Elapidae and Atractaspididae was 23%, 11% and 0.6%, respectively. The saw-scaled viper Echis ocellatus was the most abundant species (13.6%). Other venomous species were Causus maculatus (6.5%), Naja katiensis (5.5%), Bitis arietans (2.7%), Elapsoidea trapei (2.4%), Naja nigricollis (1.2%), Naja melanoleuca (1.1%), Atractaspis aterrima (0.4%), Dendroaspis polylepis (0.3%) and Naja haje (0.1%).

High mortality from snakebite in south-eastern Senegal.

Over 24 years, from 1976 to 1999, we conducted a prospective study of overall and cause-specific mortality among the population of 42 villages of south-eastern Senegal. Of 4228 deaths registered during this period, 26 were caused by snakebite, 4 by invertebrate stings and 8 by other wild or domestic animals. The average annual mortality rate from snakebite was 14 deaths per 100,000 population. Among persons aged > or = 1 year, 0.9% (26/2880) of deaths were caused by snakebite and this cause represented 28% (26/94) of total deaths by accidents. We also investigated the snake fauna of the area. Of 1280 snakes belonging to 34 species that were collected, one-third were dangerous and the proportion of Viperidae, Elapidae and Atractaspidae was 23%, 11% and 0.6%, respectively. The saw-scaled viper Echis ocellatus was the most abundant species (13.6%). Other venomous species were Causus maculatus (6.5%), Naja katiensis (5.5%), Bitis arietans (2.7%), Elapsoidea trapei (2.4%), Naja nigricollis (1.2%), Naja melanoleuca (1.1%), Atractaspis aterrima (0.4%), Dendroaspis polylepis (0.3%) and Naja haje (0.1%).

Comparative analysis of more than 3000 sequences reveals the existence of two pseudoknots in area V4 of eukaryotic small subunit ribosomal RNA.

The secondary structure of V4, the largest variable area of eukaryotic small subunit ribosomal RNA, was re-examined by comparative analysis of 3253 nucleotide sequences distributed over the animal, plant and fungal kingdoms and a diverse set of protist taxa. An extensive search for compensating base pair substitutions and for base covariation revealed that in most eukaryotes the secondary structure of the area consists of 11 helices and includes two pseudoknots. In one of the pseudoknots, exchange of base pairs between the two stems seems to occur, and covariation analysis points to the presence of a base triple. The area also contains three potential insertion points where additional hairpins or branched structures are present in a number of taxa scattered throughout the eukaryotic domain.

Religion and protective behaviours towards AIDS in rural Senegal.

OBJECTIVES: To describe the association between religion and factors related to sexually transmitted diseases (STD)/AIDS in a country where religious leaders were involved early in prevention. DESIGN: A cross-sectional study conducted in a rural area in central Senegal. METHODS: Questionnaire-based interviews of a random sample of 858 adults from the general population aged 15-59 years and in-depth interviews of four religious leaders and 50 people. RESULTS: Seventy-six per cent of the respondents were Muslim, 24% Catholic, 1% Animist and 0.2% Protestant. A total of 86% of men and 87% of women reported religion to be very important to them. Important prevention-related variables were inversely associated with the importance of religion. Men who considered religion to be very important were less likely to cite AIDS as a major health problem [odds ratio (OR) 0.4, P = 0.008] and were less likely to feel at risk of getting HIV (OR 0.5, P = 0.0005). Women who considered religion to be very important were less likely to report an intention to change to protect themselves from AIDS (OR 0.2, P = 0.0001), less likely to report having discussed AIDS with others (OR 0.4, P = 0.01) and much more likely to feel at risk of getting HIV (OR 9.3, P = 10(-4)). Individuals who considered religion to be very important were not more likely to report intending to or actually having become faithful to protect themselves from AIDS. CONCLUSION: These findings stress the need to intensify the involvement of religious authorities in HIV/STD prevention at the local level.

Peripheral markers of serotonergic and noradrenergic function in post-pubertal, caucasian males with autistic disorder.

Some studies have suggested that disorders in the peripheral and central metabolism of serotonin (5-HT) and noradrenaline may play a role in the pathophysiology of autistic disorder. This study examines serotonergic and noradrenergic markers in a study group of 13 male, post-pubertal, caucasian autistic patients (age 12-18 y; I.Q. > 55) and 13 matched volunteers. [3H]-paroxetine binding Kd values were significantly higher in patients with autism than in healthy volunteers. Plasma concentrations of tryptophan, the precursor of 5-HT, were significantly lower in autistic patients than in healthy volunteers. There were no significant differences between autistic and normal children in the serum concentrations of 5-HT, or the 24-hr urinary excretion of 5-hydroxy-indoleacetic acid (5-HIAA), adrenaline, noradrenaline, and dopamine. There were no significant differences in [3H]-rauwolscine binding Bmax or Kd values, or in the serum concentrations of tyrosine, the precursor of noradrenaline, between both study groups. There were highly significant positive correlations between age and 24-hr urinary excretion of 5-HIAA and serum tryptophan. The results suggest that: 1) serotonergic disturbances, such as defects in the 5-HT transporter system and lowered plasma tryptophan, may play a role in the pathophysiology of autism; 2) autism is not associated with alterations in the noradrenergic system; and 3) the metabolism of serotonin in humans undergoes significant changes between the ages of 12 and 18 years.

Preventive attitudes towards the threat of AIDS: process and determinants in rural Senegal.

This report is intended to identify the determinants of preventive attitudes and thereby enable the further elaboration of effective prevention programs in a country where the HIV epidemic has remained controlled. Across-sectional survey using a standardized questionnaire was conducted among a sample of 866 adults of the general population of the rural area of Niakhar, Senegal. The analysis identified psychosocial determinants related to preventive attitudes and the influence of sociodemographic characteristics and of sources of AIDS information on these determinants. Psychosocial determinants of preventive attitudes differed according to gender: Among men, personal risk perception was associated with preventive attitudes. Among women, level of AIDS-related knowledge, communication about AIDS, and optimism about the future were associated with preventive attitudes. Sociodemographic factors and sources of information were similar for men and women: Preventive attitudes were adopted primarily by young, educated subjects who had been exposed to urban life. Religion appeared to be a major obstacle to preventive attitudes. Preventive messages broadcast on radio or television may be determinative. Our results call for a strengthened collaboration with religious leaders and a focus on risk perception to rationalize preventive actions. We also encourage the targeting of vulnerable populations, such as women in rural areas, by developing preventive messages to be delivered through national media (radio, television) and by staff from health centers, which most women visit, at least for prenatal care.

The two-factorial symptom structure of post-traumatic stress disorder: depression-avoidance and arousal-anxiety.

UNLABELLED: The first part of this study showed that the DSM-III-R symptom structure of post-traumatic stress disorder (PTSD), i.e. criteria B (reexperience), C (avoidance-numbing), and D (arousal), and, consequently the diagnosis of PTSD, could not be validated in fire and car-accident victims. The aims of this study were to: (i) determine the factors as well as their structure in the symptoms of PTSD; and (ii) develop a new classification or typology of PTSD. Exploratory and confirmatory factor analyses and cluster analyses were employed to: (i) examine the factors in PTSD symptomatology; and (ii) find and validate adequate diagnostic criteria for PTSD. The Composite International Diagnostic Interview (CIDI), PTSD Module, was used between 7 and 9 months after the traumatic event in a study group of 185 victims of two different traumatic events, i.e. 130 fire and 55 car-accident victims. Our findings support the existence of two factors, i.e. a first labeled 'depression-avoidance (DAV) dimension', as it contains items reminiscent of depression and avoidance, and a second labeled 'the anxiety-arousal (AA) dimension', as it contains symptoms reminiscent of anxiety and increased arousal. Cluster analysis yielded two clusters, i.e. a cluster of subjects with PTSD cases and another with non-cases. Our PTSD algorithm was significantly less restrictive than the DSM-III-R diagnosis of PTSD. There are only quantitative, but no qualitative, differences between the cluster analytically derived classes. IN CONCLUSION: PTSD is not a well-delineated clinical entity, as there is a clinical continuum from PTSD non-cases to cases with less and more severe DAV and AA symptoms. It is more appropriate to express PTSD in terms of general severity of PTSD and severity of the DAV and AA dimensions.

Epidemiologic and phenomenological aspects of post-traumatic stress disorder: DSM-III-R diagnosis and diagnostic criteria not validated.

The aim of this cohort study was: (i) to validate the diagnostic criteria for post-traumatic stress disorder (PTSD) of the DSM-III-R; and (ii) to examine the incidence rate of PTSD in a study population exposed to two different traumatic events, i.e. a fire in a hotel ball-room and a multiple collision car-crash on a Belgian highway. One hundred and eighty-five victims (130 fire and 55 car accident victims) were assessed between 7 and 9 months after the traumatic event using the Composite International Diagnostic Interview (CIDI), PTSD Module, a fully structured diagnostic interview for the assessment of PTSD according to DSM-III-R criteria. Twenty-three percent of the study population met DSM-III-R criteria for PTSD. By means of unsupervised and supervised multivariate statistical analyses we were unable to validate the three-factorial structure, i.e. criteria B, C and D, of the DSM-III-R PTSD diagnosis. The latter relies heavily on the C diagnostic criteria, which appear to be too restrictive. Women were more likely to develop symptoms of reexperience (B) and arousal (D) than men. There was a significantly higher incidence of criteria B, C and D, but not of PTSD, in fire than in car-accident victims. Between 42 and 57% of the victims developed the first PTSD symptoms on the day of the trauma; within the next week these incidence rates increased to 77.1, 57.8 and 73.5% for criteria B, C and D, respectively. In conclusion, this study was unable to demonstrate the validity of the diagnostic criteria for PTSD according to DSM-III-R. The present cohort study has defined a number of factors that may predict new occurrences of PTSD symptoms after a traumatic event, i.e. gender, type of trauma and time delay between the trauma and the assessment of the diagnostic criteria.

Risk behaviours and AIDS knowledge in a rural community of Senegal: relationship with sources of AIDS information.

BACKGROUND: The objective of this paper is to describe sources of information on HIV/AIDS and their relationship with AIDS-related knowledge and sexual behaviour in a rural area of south Senegal. METHODS: A cross-sectional study using a standardized questionnaire was administered in 1994 by local interviewers to 240 men and 242 women aged 15-59 years, randomly selected from the general population. RESULTS: Sources of HIV/AIDS information most frequently cited were radio for men (61% of men) and the local health centre for women (52% of women). Among men, citing radio as a source of information was associated with an improved overall AIDS-related knowledge (a seven-questions based average score was 4.30 for men citing radio acquired information and 5.90 for men not citing radio acquired information; P < 10(-4)) and was associated with a smaller number of casual sexual partners in the 12 months preceding the interview (1.94 versus 1.48; P = 0.04). Women citing the local health centre as a source of HIV/AIDS information had a better perception of condom use and more often felt threatened by HIV/AIDS, but did not declare a significantly different number of casual sex partners in the 12 months preceding the interview. Television as a source was cited by 42% of men and 33% of women and was associated with an increased AIDS-knowledge score for men, with a smaller number of casual sex partners for women and with better perception of condoms for men. CONCLUSION: Because of its large spread and impact, radio appears to be an efficient way to reduce risk-taking behaviour among men. In addition, it is a very convenient way to reach people with high mobility such as male seasonal migrants. For women, attendance at health centres for maternity purposes is an opportunity to receive prevention messages. Finally, numerous men and women have had the opportunity to watch television when they are in towns during the migration period. This method seems to deliver effective messages.

PIP: Local interviewers administered a questionnaire in 1994 to 240 men and 242 women aged 15-59 years in the Ziguinchor region of southwest Senegal, randomly selected from the general population, in an investigation of respondents' sources of HIV/AIDS-related information and their relationship with AIDS-related knowledge and sex behavior. The main sources of HIV/AIDS-related information cited by respondents were radio for men (61%), the local health center for women (52%), and television for 42% of men and 33% of women. Information from school was cited by 16% of men and 5% of women. Among men, citing radio as a source of information was associated with an improved overall AIDS-related knowledge and fewer casual sex partners in the 12 months before the interview. Women citing the local health center as a source of HIV/AIDS information had a better perception of condom use and more often felt threatened by HIV/AIDS, but declared no significantly different number of casual sex partners in the 12 months preceding the interview. Exposure to television was associated with a higher AIDS knowledge score for men, fewer casual sex partners for women, and a better attitude about condoms among men.

Impact of chloroquine resistance on malaria mortality.

Over 12 years, from 1984 to 1995, we conducted a prospective study of overall and malaria specific mortality among three rural populations in the Sahel, savanna and forest areas of Senegal. The emergence of chloroquine resistance has been associated with a dramatic increase in malaria mortality in each of the studied populations. After the emergence of chloroquine resistance, the risk of malaria death among children 0-9 years old in the three populations was multiplied by 2.1, 2.5 and 5.5, respectively. This is the first study to document malaria mortality at the community level in Africa before and after the emergence of chloroquine resistance. Findings suggest that the spread of chloroquine resistance has had a dramatic impact on the level of malaria mortality in most epidemiological contexts in tropical Africa.

Improvement in AIDS knowledge, perceptions and risk behaviours over a short period in a rural community of Senegal.

The objective of this paper is to assess knowledge, perceptions and behavioural changes in response to AIDS in a rural community in the south of Senegal, comparing 2 cross-sectional surveys using standardized questionnaires and performed in 1990-1992 and 1994. An AIDS-related knowledge score was built using 4 questions about routes of HIV transmission, ranging from 0 to 4. The score increased between the 2 surveys from 1.6 to 2.1 for men (P = 0.006) and from 0.8 to 2.6 for women (P < 10(-4)). The proportion of those who responded 'I don't know' to the 4 questions dealing with routes of AIDS transmission decreased from 24% to 14% on the average for men and from 66% to 20% on the average for women. The proportion of men who declared casual sex partners in the past 12 months decreased from 39% to 21% (P = 0.01). However, the proportion remained stable for women (from 15% to 18%). These results show that despite a relatively low level of HIV infection (0.8% of all adults), AIDS-related knowledge increased and at-risk behaviour decreased in a rural area of west Africa.

[Results of a preliminary study on the various factors of HIV infection and sexually transmitted disease in rural west Africa]. / Résultats d'une étude préliminaire sur les facteurs de variation de l'infection par le VIH et les maladies sexuellement transmissibles en zone rurale d'Afrique de l'Ouest.

A pilot study has been carried out in two Senegalese rural communities in January-February 1996 in order to evaluate the feasibility of a study that is being launched in several rural sites in West Africa. This study aims to shed light on the factors of HIV infection and other sexually transmitted diseases. A sample of 100 adults were asked to answer a questionnaire on at risk behaviors for these infections. Few of them have refused to answer the questionnaire: 9% (6/66) in a first site and 3% (1/34) in the other site. Interviewers', remarks made after this pilot study showed that sex and age of interviewer and interviewee must be matched and that they must know each other so as to establish a relation of confidence. Participation to the biological part of the study may prove to be poor since, in each two communities, 27% and 40% of the interviewees declared they would refuse to provide a blood sample if asked. Finally, preliminary results suggest that at risk sexual behaviors and motivation of casual sex may be different between the two communities. The present pilot study shows that the use of our questionnaire is possible among rural communities in sub-Saharan Africa. However, it will be necessary to further explain the importance of the biological part of the study to achieve a satisfactory participation rate.

Knowledge, attitudes and perception of AIDS in rural Senegal: relationship to sexual behaviour and behaviour change.

OBJECTIVES: To describe the determinants of 'at risk' sexual behaviour and perception of AIDS-related prevention messages in rural Africa. SETTING: A rural area in Southern Senegal. DESIGN: Cross-sectional study using a standardized questionnaire administered by local interviewers to 240 men and 242 women aged 15-59 years, randomly selected among the general population. RESULTS: Twenty-eight per cent of the sexually active men and 27% of the sexually active women declared at least one casual sexual partner in the 12 months preceding the interview. Among these, 27% of men and 30% of women declared having used a condom in most acts of casual intercourse. Seasonal migrants and divorced or widowed women were more likely to declare casual sex. Causal sex was motivated by material needs for 66% of the women who experienced it, and those of the women who reported casual sexual intercourse were less likely to feel at risk of AIDS [odds ratio (OR), 3.9; P = 0.01] and were more optimistic about their future (OR, 3.6; P = 0.03). For men, the motivations explaining a change in sexual behaviour in order to avoid HIV infection included the perception of AIDS as a health problem (OR, 11; P = 0.004), the perception of the disease as serious (OR, 5.4; P = 0.001) and the feeling of personal risk of becoming HIV-infected (OR, 3.2; P = 0.02). Perceived skill in changing one's behaviour was strongly associated with declaration of past behaviour change for both men and women (men: OR, 3.4; P = 0.02; women: OR, 6.3; P = 0.0001). CONCLUSION: Men and women exhibit two different patterns regarding their behaviour and perception towards AIDS. Material needs appear to be of importance for women, whereas perception of a real threat lead men to adopt protective behaviours. In the very area of this study, widowed and divorced women as well as male seasonal migrants are particularly exposed to HIV infection. They are characterized by a higher risk behaviour, a low rate of condom use and seldom declared any protective measures to avoid HIV infection.

PIP: 240 men and 242 women, randomly selected, aged 15-59 years were questioned by local interviewers about their sexual behavior in this study on the determinants of risky sexual behavior and the perception of AIDS-related prevention messages in rural Africa. 28% of the sexually active men and 27% of the sexually active women reported having at least one casual sex partner during the 12 months preceding the interview. Among these, 27% of men and 30% of women reported having used a condom in most acts of sexual intercourse with casual partners. Male seasonal migrant workers and widowed and divorced women were more likely to report engaging in casual sex. These individuals in the area of study are particularly at risk for exposure to HIV infection. Casual sex was motivated by material needs among 66% of women who experienced it. Women who reported engaging in such behavior were less likely to feel at risk of AIDS. The perception of a real threat of personal vulnerability to contracting HIV lead men to adopt protective behaviors.

A study of sexual behavior change in rural Senegal.

This study describes recent sociodemographic and behavioral changes in a rural community in Senegal and investigates how these changes may have modified the risk of HIV/STD transmission. The authors combine findings from two surveys: (a) a marriage survey among all adults of a rural community of approximately 7,000 inhabitants; and (b) a sexual practices survey conducted among a sample of the same population (165 men and 165 women) between 1989 and 1994. During the last 30 years--between 1955-64 and 1985-92--age at first marriage increased in the study area from 25 to 32 years for men and from 21 to 24 years for women. At the same time, age at first sexual intercourse decreased by approximately 10 years on average for men. As expected, this has led to an increase in premarital sexual activity, and we have documented both an increase in the percentage of individuals declaring sexual relations before marriage and, for men, an increase in the number of sexual partners before marriage. Multipartnership is more frequent among men than women (OR=4.4), among those who had their first intercourse at an early age (OR=3.3), among rural migrants (OR=2.7), and among those who had a good knowledge of AIDS. These findings show how recent changes in marriage, sexual behaviors, and seasonal migration have combined to increase the risk of HIV/STD infection within the community. Male rural migrants and their partners are particularly exposed.

Role of immunizations in the recent decline in childhood mortality and the changes in the female/male mortality ratio in rural Senegal.

In early 1987, immunizations were introduced to an isolated area of eastern Senegal where there had previously been no regular immunizations. Since immunizations were the only change introduced in the area during this period, this allowed the authors to study the impact of immunizations of mortality in different age groups and the possible interaction with sex differences in mortality. They compared mortality rates for the 6 years before and the 6 years after the introduction of immunization. Neonatal mortality declined 31% (95% confidence interval (CI) 17 to 43); between 1 and 8 months of age, the reduction was 20% (95% CI -2 to 37); and between 9 and 59 months of age, mortality declined 48% (95% CI 39 to 56). Excluding acute measles deaths, the reduction was 16% (95% CI -8 to 35) between 1 and 8 months of age and 32% (95% CI 20 to 43) between 9 and 59 months of age. The decline was stronger in villages that maintained high coverage after the initial national campaign, whereas mortality increased again in the villages where the coverage declined. Since the reduction in mortality was most marked after 9 months of age, measles immunization is likely to have been the most important vaccination. Both female and male mortality declined but not equally quickly. The reduction in mortality in the neonatal period was significantly greater in males than in females, resulting in an increase in the female/male mortality ratio from 0.64 (95% CI 0.50 to 0.83) to 0.96 (95% CI 0.71 to 1.30), p = 0.04. After 9 months of age, the reduction in mortality was somewhat greater in females than in males, resulting in a decrease in the female/male mortality ratio from 1.04 (95% CI 0.85 to 1.28) to 0.79 (95% CI 0.62 to 1.02), p = 0.10.

Lower mortality for female-female twins than male-male and male-female twins in rural Senegal.

Twins have been registered prospectively for 12-22 years in 42 small villages in the Bandafassi area of Eastern Senegal. We studied 98 pairs of twins to test whether twins in opposite-sex pairs have higher postneonatal mortality than same-sex twins. Neonatal mortality for twins was 41.3%; mortality for infants and for children under age 5 years was 53.0% and 66.8%, respectively. Neonatal mortality was identical for same-sex and opposite-sex twin pairs, but much higher for boys than girls [relative risk = 1.8; 95% confidence interval (CI) = 1.2-2.6]. There was clustering of double neonatal deaths for all types of twins. In the postneonatal period, female-female twins had lower mortality than other twin types. Twins had higher post-neonatal mortality as long as the co-twin was alive [mortality rate ratio (MR) = 2.6; 95% CI = 1.0-6.7]. Girls had excess mortality when the co-twin was of the opposite sex (MR = 4.3; 95% CI = 1.2-15.3), whereas there was no difference for boys. In conclusion, contact with a co-twin of the opposite sex increased child mortality for female twins. Our data are not sufficient to determine whether this difference is specific for girls or applies to boys as well.