Regression analysis of "current status" life tables on duration of breastfeeding in Sri Lanka

Provided that women report the dates of their children's births with reasonable accuracy, it is possible to derive good estimates of the duration of breastfeeding from women's breastfeeding status at the time of the interview. This paper illustrates the application of conventional regression techniques to the analysis of breastfeeding rates derived in this manner. Construction of current status rates is explained and a comparison between open interval, closed interval, and current status breastfeeding life tables is presented, indicating the extent of bias to which tables of the former types are open. Birth-weighted rates are used for WFS data from Sri Lanka; the variables entered into the regression equation include parity, educational level, residence, work experience since marriage and use of contraception since the birth. Contraception is not found to influence net breastfeeding rates in the 1st interval (1-16 months), although it is about as prevalent as in later intervals. The positive coefficients at intervals beyond the 1st also imply that contraceptive use is not a substitute for lactation in Sri Lanka or not a predominant one. Lifetime urban residence is associated with short durations of breastfeeding, and being an urban migrant is associated with intermediate durations relative to those of rural women. The effects of residence on breastfeeding are especially pronounced in the 1st interval. By parity as by contraception, differences in breastfeeding rates are not significant at short durations but become so with time as lower parity women reach pregnancy. Patterns by age are similar, but less sharp. Middle school attendance and work at home are both strongly associated with with lactation behavior, the former negatively and the latter to about an equal degree positively. Working outside the home seems not to influence breastfeeding to any great extent. In the multiple attribute regressions, middle schooling depresses breastfeeding durations about as strongly as lifetime urban residence relative to women who are rural and uneducated. Being at parity 5 or above or working at home almost as sharply increases durations. The use of contraception has slightly smaller positive effects. The patterns suggest that as Sri Lankan women become increasingly well educated and urban, and as family sizes decline, durations of breastfeeding will decline. Equally important however, even among better educated urban wives, breastfeeding continues longer than is typical of western countries.     

Kin influences on fertility in Thailand: Effects and mechanisms

It has been suggested that human mothers are cooperative breeders, as they need help from others to successfully raise offspring. Studies working under this framework have found correlations between the presence of kin and both child survival and female fertility rates. This study seeks to understand the proximate mechanisms by which kin influence fertility using data from the 1987 Thailand Demographic and Health Survey (DHS), a nationally representative sample of 6775 women. Kin influence is measured by the length of time couples live with the husband's or wife's parents after marriage. Event history analysis, multilevel modeling and structural equation modeling are used to investigate both fertility outcomes and potential pathways through which postnuptial residence may influence fertility outcomes, including employment status, maternal and child outcomes, contraceptive use, breastfeeding duration, and age at marriage. We show that living virilocally (with husband's kin after marriage) increases total fertility by shortening time from marriage to first birth, and increasing the likelihood of progression to each subsequent birth. These effects are mediated through correlations between virilocal residence and earlier age at marriage as well as delayed initiation of contraceptive use. We find no influence of husband's kin on maternal or child outcomes. Living uxorilocally (with wife's kin after marriage) also reduces age at marriage, shortens time from marriage to first birth and (marginally) improves child survivorship, but has no effect on other child and maternal outcomes or progression to subsequent births and results in a similar number of living children as women living neolocally.     

Sociodemographic attributes of Iranian wives who reported unwanted pregnancies

In a 1989 Family Planning Study in Iran, 40 percent of the married women of reproductive age reported that their last or current pregnancies were unwanted and unintended. This finding is consistent with the results obtained from a number of studies undertaken in the countries of North Africa and the Middle East. Although the phenomenon of unwanted pregnancy is a significant topic in the population studies, it has received very little attention. This paper shows the differences between two groups: group A, those married women who reported that their last or current pregnancies were wanted, and group B, those married women who reported that their last or current pregnancies were unwanted. The findings of this study clearly show significant differences between these two groups in regard to some key sociodemographic attributes: wife/husband's education, actual and desired fertility, wife's current age, past and present practice of contraceptive methods, and extent of satisfaction with family planning services are among the attributes differentiating these two groups. Our examination of these variables suggests that group B had higher parity, fertility, less desire for more children, less use of contraceptive techniques, and less satisfaction with the efficiency of the contraceptive techniques than group A. The relationship between education and wanted/unwanted pregnancies is mixed. The urban women who wanted pregnancies and were 25 years old or older were more educated than those who did not want pregnancies. On the contrary, the rural women whose pregnancies were not wanted were more educated than those who wanted pregnancies regardless of age, parity, and locality differences. Furthermore, the urban/rural women with different levels of parities who did not want to be pregnant were more educated than the urban/rural women who wanted pregnancies. Finally, the standardized regression coefficients, obtained in logistic regression, reveal that among urban women the desire for more children and parity are the first and second most significant independent variables differentiating between group A and group B. Among rural women, living children and the desire for more children were the first and second most important variables differentiating between group A and group B.     

Cessation of childbearing in the absence of contraception in Sri Lanka

In data from the Sri Lanka Fertility Survey, 1975, the cessation of childbearing is examined among women who have never used contraception. The sample consisted of 6810 currently or previously married women, 57% of whom reported that they had never used contraception. Cessation of childbearing is studied according to age and marriage duration. The variables analyzed are age at last birth, proportions infertile during the last 5 or 7 years, and the infertile open interval. The duration of breastfeeding is taken into account where necessary, and the contraceptive users and nonusers are compared where appropriate. Non-users tend to cease childbearing early, and therefore are infertile for longer periods during their marriages. It is probably age of the 1st child that influences decisions on future fertility. Among women aged 45-49 who married before age 20 and continued in their 1st marriage, mean age at last live birth in non-users, was 34.5 years, about 2 years earlier than in those who had used contraception. Non-users who married at any age below 30 years cease childbearing well below age 40. The proportion not currently pregnant and infertile over the past 5 years increases with marriage duration among the fertile non-users in each age group. When age at last birth and the duration of breastfeeding in the open interval are taken into account and the reference period is increased to 7 years, the period of infertility increases with marriage duration among nonpregnant non-users below age 45. The proportion of women who were currently not pregnant and had remained infertile over the past 7 years is higher among the older non-users whose 1st child was born more than 10 years ago.     

Sociodemographic attributes of Iranian wives who reported unwanted pregnancies

Abstract

In a 1989 Family Planning Study in Iran, 40 percent of the married women of reproductive age reported that their last or current pregnancies were unwanted and unintended. This finding is consistent with the results obtained from a number of studies undertaken in the countries of North Africa and the Middle East. Although the phenomenon of unwanted pregnancy is a significant topic in the population studies, it has received very little attention. This paper shows the differences between two groups: group A, those married women who reported that their last or current pregnancies were wanted, and group B, those married women who reported that their last or current pregnancies were unwanted. The findings of this study clearly show significant differences between these two groups in regard to some key sociodemographic attributes: wife/husband's education, actual and desired fertility, wife's current age, past and present practice of contraceptive methods, and extent of satisfaction with family planning services are among the attributes differentiating these two groups. Our examination of these variables suggests that group B had higher parity, fertility, less desire for more children, less use of contraceptive techniques, and less satisfaction with the efficiency of the contraceptive techniques than group A. The relationship between education and wanted/unwanted pregnancies is mixed. The urban women who wanted‐pregnancies and were 25 years old or older were more educated than those who did not want pregnancies. On the contrary, the rural women whose pregnancies were not wanted were more educated than those who wanted pregnancies regardless of age, parity, and locality differences. Furthermore, the urban/rural women with different levels of parities who did not want to be pregnant were more educated than the urban/rural women who wanted pregnancies. Finally, the standardized regression coefficients, obtained in logistic regression, reveal that among urban women the desire for more children and parity are the first and second most significant independent variables differentiating between group A and group B. Among rural women, living children and the desire for more children were the first and second most important variables differentiating between group A and group B.     

Interruption of breastfeeding by child death and pregnancy

This paper estimates the mean monthly losses and proportionate interruptions of breastfeeding intervals due to child death and pregnancy. The paper uses a microanalytic model with stochastic risks for the basic processes of human reproduction. The model results show that a high proportion of women, depending upon mortality level and length of breastfeeding, have their breastfeeding interrupted by either child death or pregnancy. The results of this work suggest the need for caution in interpreting observed durations of breastfeeding. Child death and pregnancy causes bias in the reported length of breastfeeding, and this bias needs to be taken into account in statistical analysis.     

Women's status and infant mortality in rural Colombia

This paper investigates the effects of maternal demographic characteristics and social and economic statuses on infant mortality in rural Colombia. Demographic characteristics include the age of the mother, parity and length of preceding interbirth interval, and sex of infant. Measures of women's status at the time of birth include education, wage labor and occupation, economic stratum, place of residence, and whether the mother is living with a husband. The life history data for the study (involving 4,928 births) were collected in 1986 from a representative sample of two cohorts of women resident in rural central Colombia. Overall differentials in infant mortality by measures of women's status are small and are in good part associated with the differing reproductive behaviors of the women and variations in breastfeeding practices. The sharp declines in infant mortality recorded in rural Colombia in recent years appear less related to improved status of women than to reductions in fertility that enhance infant survivorship and to public health interventions shared by all segments of the population.     

Interruption of breastfeeding by child death and pregnancy

Abstract

This paper estimates the mean monthly losses and proportionate interruptions of breastfeeding intervals due to child death and pregnancy. The paper uses a microanalytic model with stochastic risks for the basic processes of human reproduction. The model results show that a high proportion of women, depending upon mortality level and length of breastfeeding, have their breastfeeding interrupted by either child death or pregnancy. The results of this work suggest the need for caution in interpreting observed durations of breastfeeding. Child death and pregnancy causes bias in the reported length of breastfeeding, and this bias needs to be taken into account in statistical analysis.     

Pregnancy Prevention and Condom Use Practices among HIV-Infected Women on Antiretroviral Therapy Seeking Family Planning in Lilongwe,Malawi

Background

Programs for integration of family planning into HIV care must recognize current practices and desires among clients to appropriately target and tailor interventions. We sought to evaluate fertility intentions, unintended pregnancy, contraceptive and condom use among a cohort of HIV-infected women seeking family planning services within an antiretroviral therapy (ART) clinic.

Methods

200 women completed an interviewer-administered questionnaire during enrollment into a prospective contraceptive study at the Lighthouse Clinic, an HIV/ART clinic in Lilongwe, Malawi, between August and December 2010.

Results

Most women (95%) did not desire future pregnancy. Prior reported unintended pregnancy rates were high (69% unplanned and 61% unhappy with timing of last pregnancy). Condom use was inconsistent, even among couples with discordant HIV status, with lack of use often attributed to partner’s refusal. Higher education, older age, lower parity and having an HIV negative partner were factors associated with consistent condom usage.

Discussion

High rates of unintended pregnancy among these women underscore the need for integ rating family planning, sexually transmitted infection (STI) prevention, and HIV services. Contraceptive access and use, including condoms, must be improved with specific efforts to enlist partner support. Messages regarding the importance of condom usage in conjunction with more effective modern contraceptive methods for both infection and pregnancy prevention must continue to be reinforced over the course of ongoing ART treatment.     

Birth spacing in Pakistan

Life table analysis was applied to data from the 1975 Pakistan Fertility Survey to identify child spacing differentials between population subgroups. Women in urban areas had shorter birth intervals than their rural counterparts from parities 1-6; only after parity 7 was this differential reversed. Similarly, women with some education had shorter birth intervals at the earlier parities than uneducated women. While overall family size is relatively homogeneous in Pakistan, women of more modern backgrounds seem to space their children more closely than traditional women. Age at marriage appears to play an important role not only in determining the length of the 1st interval, but also that of subsequent intervals. An unexpected finding was that ever users of contraception had distinctly more rapid spacing of their births than never users. The median interval to 1st birth was shortest in North West Frontier Province, but similar in Punjab and Sind. Multiple classification analysis revealed that some differentials in child spacing by education, residence, and province persisted even after other variables were controlled. Cohort of mother had an independent effect, with younger cohorts having shorter birth intervals. However, the variable that had the strongest effect on length of interval (aside from the 1st interval) was breastfeeding duration. It is likely that increasing urbanization and improved levels of education among women will lead to high levels of marital fertility associated with shorter birth intervals. Even though these trends tend to increase the age at marriage, they are associated with shorter durations of breastfeeding. In the longer term, greater use of contraception among women in the modern sector may partially counteract the fertility increasing effect of reduced birth intervals.     

Medically Complex Pregnancies and Early Breastfeeding Behaviors: A Retrospective Analysis

Background

Breastfeeding is beneficial for women and infants, and medical contraindications are rare. Prenatal and labor-related complications may hinder breastfeeding, but supportive hospital practices may encourage women who intend to breastfeed. We measured the relationship between having a complex pregnancy (entering pregnancy with hypertension, diabetes, or obesity) and early infant feeding, accounting for breastfeeding intentions and supportive hospital practices.

Methods

We performed a retrospective analysis of data from a nationally-representative survey of women who gave birth in 2011–2012 in a US hospital (N = 2400). We used logistic regression to examine the relationship between pregnancy complexity and breastfeeding. Self-reported prepregnancy diabetes or hypertension, gestational diabetes, or obesity indicated a complex pregnancy. The outcome was feeding status 1 week postpartum; any breastfeeding was evaluated among women intending to breastfeed (N = 1990), and exclusive breastfeeding among women who intended to exclusively breastfeed (N = 1418). We also tested whether breastfeeding intentions or supportive hospital practices mediated the relationship between pregnancy complexity and infant feeding status.

Results

More than 33% of women had a complex pregnancy; these women had 30% lower odds of intending to breastfeed (AOR = 0.71; 95% CI, 0.52–0.98). Rates of intention to exclusively breastfeed were similar for women with and without complex pregnancies. Women who intended to breastfeed had similar rates of any breastfeeding 1 week postpartum regardless of pregnancy complexity, but complexity was associated with >30% lower odds of exclusive breastfeeding 1 week among women who intended to exclusively breastfeed (AOR = 0.68; 95% CI, 0.47–0.98). Supportive hospital practices were strongly associated with higher odds of any or exclusive breastfeeding 1 week postpartum (AOR = 4.03; 95% CI, 1.81–8.94; and AOR = 2.68; 95% CI, 1.70–4.23, respectively).

Conclusions

Improving clinical and hospital support for women with complex pregnancies may increase breastfeeding rates and the benefits of breastfeeding for women and infants.     

Women's status and infant mortality in rural Colombia

Abstract

This paper investigates the effects of maternal demographic characteristics and social and economic statuses on infant mortality in rural Colombia. Demographic characteristics include the age of the mother, parity and length of preceding interbirth interval, and sex of infant. Measures of women's status at the time of birth include education, wage labor and occupation, economic stratum, place of residence, and whether the mother is living with a husband. The life history data for the study (involving 4,928 births) were collected in 1986 from a representative sample of two cohorts of women resident in rural central Colombia. Overall diflFerentials in infant mortality by measures of women's status are small and are in good part associated with the differing reproductive behaviors of the women and variations in breastfeeding practices. The sharp declines in infant mortality recorded in rural Colombia in recent years appear less related to improved status of women than to reductions in fertility that enhance infant survivorship and to public health interventions shared by all segments of the population.     

A note on the determinants of breastfeeding durations in an African country

Abstract

This paper utilizes data from the 1977–78 Kenya Fertility Survey, one component of the World Fertility Survey, to analyze the determinants of breastfeeding durations for women 15 to 50 years old who had their last‐but‐one live birth between 3 and 15 years prior to the interview. Comparisons are made with the findings from the World Fertility Surveys in eight other developing countries in Asia and Latin America. Findings indicate that literacy, urban residence, secondary school education, and modern employment reduce the duration of breastfeeding in Kenya. In addition, the subgroups of women who appear to be curtailing breastfeeding are growing in proportional size or are composed of women who may be innovators or leaders. A continuation of this pattern into the future may increase levels of infant morbidity and mortality and, in the absence of increased modern contraceptive practice, may increase the societal level of fertility.     

Determinants of breastfeeding initiation among mothers in Sydney,Australia: findings from a birth cohort study

Background

Breastfeeding has short-term and long-term benefits for both the infant and the mother. The objective of this study was to identify the incidence of breastfeeding initiation among women in South Western Sydney, and the factors associated with the initiation of breastfeeding.

Methods

Child and Family Health Nurses recruited mother-infant dyads (n = 1035) to the Healthy Smiles Healthy Kids birth study in South Western Sydney, an ethnically and socio-economically diverse area, at the first post-natal home visit. A sample of 935 women completed a structured, interviewer-administered questionnaire at 8 weeks. Multivariate logistic regression analysis was used to identify those factors independently associated with the initiation of breastfeeding.

Results

In total, 92% of women (n = 860) commenced breastfeeding in hospital. Women who completed a university degree were more likely to initiate breastfeeding compared to those who did not complete high school (AOR = 7.16, 95% CI 2.73, 18.79). Vietnamese women had lower odds of breastfeeding initiation compared to Australian born women (AOR = 0.34. 95% CI 0.13, 0.87). Women who had more than one child were less likely to breastfeed than those who had one child (AOR = 0.38, 95% CI 0.19, 0.79). Women who gave birth via a caesarean section were less likely to breastfeed their baby compared to those who had a vaginal delivery (AOR = 0.27, 95% CI 0.14, 0.52). Women who drank alcohol during pregnancy had 72% lower odds to breastfeed compared to those who did not drink alcohol during pregnancy (AOR = 0.28, 95% CI 0.11, 0.71). Women who reported that their partner preferred breastfeeding were more likely to initiate breastfeeding (AOR = 11.77, 95% CI 5.73, 24.15) and women who had chosen to breastfeed before pregnancy had more than 2.5 times the odds of breastfeeding their baby compared to those women who made their decision either during pregnancy or after labour (AOR = 2.80, 95% CI 1.31, 5.97).

Conclusions

Women with lower levels of education, who consume alcohol during pregnancy, have more than one child, and make infant feeding decision after becoming pregnant, and those born in Vietnam should be targeted when implementing breastfeeding promotion programs. Further, women who deliver by caesarean section require additional breastfeeding support post-delivery and it is important to include fathers in breastfeeding related decisions and encourage them to participate in antenatal programs.

     

A note on the determinants of breastfeeding durations in an African country

This paper utilizes data from the 1977-78 Kenya Fertility Survey, 1 component of the World Fertility Survey, to analyze the determinants of breastfeeding durations for women 15 to 50 years old who had their last-but-1 live birth between 3 and 15 years prior to the interview. Comparisons are made with the findings fro m the World Fertility Surveys in 8 other developing countries in Asia and Latin America. Findings indicate that literacy, urban residence, secondary school education, and modern employment reduce the duration of breastfeeding in Kenya. In addition, the subgroups of women who appear to be curtailing breastfeeding are growing in proportional size or are composed of women may be innovators or leaders. A continuation of this pattern into the future may increase levels of infant morbidity and mortality and, in the absence of increased modern contraceptive practice, may increase the societal level of fertility. The death of the infant curtails the period of breastfeeding. Although there is a pronounced preference for male children in Kenya, this preference does not lead to differential durations of breastfeeding by sex of child. About 10% of Kenyan women used contraception in the last closed interval. Parity and age explain less than 1% of the variation in duration of breastfeeding in Kenya. Kenyan women are among the least likely to have attended secondary school, to have worked since marriage, and to have used modern contraception. The most traditional groups of Kenyan women, those who are Muslim or who are in polygamous unions, breastfeed for the longest durations. The Kenya Fertility Survey suggests that the subgroup of women with some secondary school education is growing considerably. Kikuyu women may be regarded as innovators in many respects. In addition to having shorter breastfeeding durations, they are the least likely to be in polygamous unions or to want more children, and they are the most likely to be using contraception.     

Maternal age at last birth in Egypt

To assess the effects of women's education, residence, and marital experience on their age at the birth of their last child, a proportional hazards regression model was applied to 1980 Egyptian Fertility Survey (EFS) data. The detailed data include the date of birth of each child for every women interviewed, and the woman's date of birth and age at interview. Age at last birth was examined by regression analysis on birth history and socioeconomic information. 4 hypotheses were tested: women who are well educated have a greater probability of ending childbearing earlier than women with less education; women in rural areas have a higher probability of having their last child at older ages than urban women; marital disruption without remarriage lowers the probability of older maternal age at last birth; and marital disruption with remarriage increases the probability that a woman stops reproducing at an older age. The overall chi-square indicates a significant regression. All coefficients were significant, except the coefficient for women with intact 1st marriages. Women with more education had a greater probability of ending childbearing earlier than women with less education. Rural women tended to have their last children at ages significantly older than overall age at last birth. Current residence in urban areas had the opposite effect. The coefficient for those with intact 1st marriages was insignificant, meaning that the mean age at last birth for this group of women was not much different from the overall mean. Remarried women tended to end childbearing at ages significantly older than the overall average age at last birth, suggesting that these women tended to have children by their new husbands. Those with dissolved 1st marriages who had not remarried had a higher probability of ending childbearing earlier than did older women. Marriage age and final parity had highly significant negative coefficients; as marriage age and number of children born increased, so did the "survival" time or the age at last birth. Results from the hazards model indicate that the effects were as anticipated. The median age at last birth for the total sample of women aged 45-49 was 45-49 years. The median age at last birth was about 2 years older for rural compared to urban women. Illiterate women had the oldest median age at last birth of the education groups. There was little differences between median ages at last birth for women with intact 1st marriages and those whose 1st unions were dissolved and who had remarried. The median age at last birth increased with final parity.     

Resumption of menstruation among urban post-partum women in Bangladesh.

To investigate the return of menstruation among postpartum women in Dacca, a random sample of 360 women (1-12 months postpartum) was selected from 4 Dacca hospitals for a cross-sectional study; another random sample of 200 women ( 1 month postpartum at time of interview) was selected from the same hospitals for prospective follow up. Many of the women were from rural areas who left the city after delivering their babies; most were under 25 years old at the time of interview. Results of the cross sectional study showed that 63% started menstruating by the 6th postpartum month; 10% were still amenorrhoeic by the 12th postpartum month; 46% were breastfeeding their infants by the 4th month; 90% resumed sexual relations, and 48% were using contraceptives, by the 6th month. Results of the prospective study showed that slightly more than half of the women started menstruating by the end of the 5th postpartum month; almost half stopped breastfeeding by the 5th month, 88.8% were having sexual relations with their husbands, and about 50.5% were on some form of birth control. The findings suggest a much shorter duration of lactational amenorrhea than has been claimed for a rural area of Bangladesh; they also suggest that urban women start menstruating earlier because of their primary reliance on supplementary feeding. The contraceptive effectiveness, as well as health benefits to infants, of prolonged and intensive lactation should be promoted to urban mothers.     

Son preference and its effects on Korean lactation practices

Using retrospective data from Korea, multivariate log-linear analyses were used to determine if the sex of a child influences the probability of it being breastfed, or if it influences the number of months he/she was breastfed. In Korea the likelihood of a woman not breastfeeding a child is very small. The fact that a woman is without living sons increased the average odds of not breastfeeding by 1.55 times. The absence of living sons has the strongest effect of any of the predictor variables used. The highest predicted conditional odds of not breastfeeding are for women with female infants, who live in rural areas, have less than 3 children of which none are boys, and who do not use oral contraceptives. The lowest predicted conditional odds of a woman not breastfeeding are for a city dweller, with 3 or more children of which at least 1 is a boy, who uses the pill, and whose infant is a boy. Although oral contraceptive use is significant for deciding how long to breastfeed, the direction of its effect is not what was expected had it affected lactation physiologically. Apparently, social factors are operative which make pill users more likely to breast feed than non-users. 1 possible explanation for this is that Korean women are aware of the fertility-suppressing effects of lactation and are using breastfeeding as a deliberate form of birth control. The fact that the effect of pill use becomes progressively stronger the longer that women breastfeed lends further support to this explanation. The effect of sex of child is not significant in determining the duration he/she will be breastfed. This study indicates support for the contention that Korean women are aware of the contraceptive effects of lactation, and that their decisions of breastfeed are affected by their preference for sons.     

Birth spacing,breastfeeding, and early child mortality in a traditional Indian society: A hazards model analysis

Abstract

There are few studies of the interrelationships among breastfeeding, child spacing, and child mortality in traditional societies that incorporate extensive controls for social and demographic characteristics of the mother and child. In this paper, we investigate the impact of breastfeeding and the length of the preceding birth interval on early child mortality (defined as a death in the first two years of life) using data from a traditional society of India. Multivariate hazards models are used to analyze the data. Most prior analyses related the impact of breastfeeding duration to the duration of child survivability by taking breastfeeding as a fixed covariate. The present study has a methodological focus in the sense that breastfeeding information from retrospective survey data is treated as a time‐dependent covariate both as a status variate as well as a duration—with empirical findings compared across the two specifications. The effects of postpartum amenorrhoea and various other demographic and socioeconomic characteristics of mother and child are also studied. The results suggest that breastfeeding duration has a strong impact in reducing the relative risk of early child mortality; but it does not explain the effect of the length of the preceding birth interval on early child mortality.     

Brazilian mothers' beliefs, attitudes and practices related to child weight status and early feeding within the context of nutrition transition

With the rapid pace of the nutrition transition worldwide, understanding influences of child feeding practices within a context characterized by the co-existence of overweight and undernutrition in the same population has increasing importance. This qualitative study describes Brazilian mothers' child feeding practices and their perceptions of their association with child weight status and explores the role of socioeconomic, cultural and organizational factors on these relationships. Forty-one women enrolled in the Family Health/Community Health Workers Programme were selected from rural, urban, coastal and indigenous areas in Ceara State, north-east Brazil, to participate in four focus group discussions. Content analysis identified fourteen emergent themes showing mothers' child feeding practices in this setting were influenced by economic resources, mothers' immediate social support networks (e.g. neighbours and family members) and participation in nutrition assistance programmes. Child malnutrition was the most common nutritional concern; nevertheless, mothers were aware of the negative health consequences of obesity but misunderstood its causes (e.g. foods filled with fat would make a person fat; others thought that birth control pills and stimulants given to children were causes of obesity); several reported their own struggles with overweight. Food assistance programmes emerged as an important influence on children's dietary adequacy, especially among mothers describing dire economic situations. The findings have implications for targeting food assistance as well as health and nutrition education strategies in low-income families undergoing the nutrition transition in north-east Brazil.