The Journal of Nutrition Vol. 128 No. 10 October 1998, pp. 1672-1680

Maternal Nutritional Status Is Inversely Associated with Lactational Amenorrhea in Sub-Saharan Africa: Results from Demographic and Health Surveys II and III^1,2,3,4,5

Yu-Kuei Peng, Virginia Hight-Laukaran^*, Anne E. Peterson, and Rafael Pérez-Escamilla⁶

Department of Nutritional Sciences, University of Connecticut, Storrs, CT 06269-4017; ^* John Snow, Inc., Arlington, VA 22209; and  Breastfeeding and MCH Division, Institute for Reproductive Health, Department of Obstetrics and Gynecology, Georgetown University Medical Center, Washington, DC 20007

	ABSTRACT

Abstract Introduction Results Discussion References

Breast-feeding is positively associated with the duration of postpartum amenorrhea; thus it is a major determinant of fertility in countries where effective contraceptive methods are not widely available. The objective of these analyses was to examine the association between maternal nutritional status and lactational amenorrhea (LA) among breast-feeding women. Women who were not pregnant, who were breast-feeding, who were not using hormonal contraceptives and who had a child 2 y old at the time of survey were included in the analyses. Multivariate logistic regression was used to examine this association with the use of Demographic and Health Surveys data collected in seven Sub-Saharan African countries between 1990 and 1994. Analyses were adjusted for nine confounders, including breast-feeding behavior, child nutritional status and child age. Within-country analyses consistently showed the trend of low maternal body mass index (BMI) associated with a higher likelihood of being amenorrheic. Pooled analyses (n = 9839) were performed using two child age groups (<9 mo and 9-24 mo). The <9 mo pooled analysis showed that women with BMI < 18.5 kg/m² (odds ratio; 95% confidence interval: 1.6; 1.2-2.3) were more likely to remain amenorrheic at the time of the survey than their "better-nourished" counterparts. The 9-24 mo pooled analysis showed that the differential probability of being amenorrheic between undernourished and "better-nourished" women increased with time postpartum, where the adjusted difference in the median duration of amenorrhea associated with maternal nutritional status among breast-feeding women was 1.4 mo. These results suggest that maternal nutritional status plays an independent role in the return of ovulation after delivery.

	INTRODUCTION

Abstract Introduction Results Discussion References

Lactational amenorrhea (LA),⁷ the interval between birth of a child and subsequent return of menses, is an important regulator of fertility in developing countries. Breast-feeding, in particular exclusive breast-feeding, has a significant contraceptive effect by postponing the return of ovulation after the birth of a child (Habicht et al. 1985, Labbok and Laukaran 1994, Zohoori and Popkin 1996).

The effects of nutrition on ovarian function were investigated by Frisch and co-workers (Frisch et al. 1973, Frisch and McArthur 1974) in the 1970s and resulted in the formulation of the "critical body composition hypothesis." Frisch suggested that a minimal amount of fat as percentage of body weight is necessary for attaining menarche and for maintaining ovarian cycles. However, the effect of maternal nutritional status on ovarian function remains controversial. Although some researchers claim that maternal nutritional status plays a role in determining the length of postpartum amenorrhea (Delgado et al. 1982, Huffman et al. 1987, Popkin et al. 1993, Tracer 1996), others have concluded that this association is biologically insignificant (Bongaarts 1980, Diaz et al. 1988, Kurz et al. 1993). In addition, it has also been postulated that child nutritional status may influence the duration of lactational amenorrhea (Kurz et al. 1993). To our knowledge, the association of maternal nutritional status with lactational amenorrhea independent of child nutritional status has not been systematically explored using nationally representative data.

Thus, the objective of this study is to examine the independent association of maternal nutritional status with lactational amenorrhea among breast-feeding women using seven nationally representative Demographic and Health Surveys collected in Sub-Saharan Africa.

	CONCEPTUAL FRAMEWORK

A conceptual model showing determinants of lactational amenorrhea is presented in Figure 1. The duration of postpartum infertility is closely related to the plasma concentration of prolactin and the subsequent changes in the release pattern of gonadotropic hormones (Campino et al. 1994, Delvoye et al. 1977, Genazzani et al. 1991, Huffman 1983, Lunn et al. 1984). Breast-feeding behavioral characteristics, such as nursing frequency, intensity and duration, have been found to affect levels of sex hormones (Labbok et al. 1991, Labbok and Laukaran 1994, Perez et al. 1992, Zohoori and Popkin 1996). High breast-feeding frequency is associated with hyperprolactinemia, which in turn is associated with a prolonged duration of amenorrhea (Delvoye et al. 1977). Maternal body fat reserves could alter the maternal hormonal profile and affect the duration of lactational amenorrhea (Frisch et al. 1973). An interaction between maternal nutritional status and breast-feeding behavioral characteristics on lactational amenorrhea is also plausible (i.e., the influence of level of body fat on lactational amenorrhea may be modified by nursing frequency). Child nutritional status may affect lactational amenorrhea through breast-feeding behavioral characteristics (Huffman 1983, Kurz et al. 1993) and maternal hormonal profiles (Delgado et al. 1979). It is possible that after 6 mo postpartum, "better-nourished" breast-fed children are those who are more likely to be supplemented with weaning foods and thus likely to be consuming less breast milk and causing less stimulation of their mothers' breasts (Kurz et al. 1993). Supplements given to breast-fed infants decrease the amount of time infants spend breast-feeding and the likelihood of their mothers remaining amenorrheic (Dewey et al. 1997, Kurz et al. 1993). Child gender might influence breast-feeding behavioral characteristics (Pérez-Escamilla et al. 1997). Nursing behavior would likely vary with child age. Possible interactions exist between child age and maternal nutritional status (i.e., differences in the risk of remaining amenorrheic between undernourished and better-nourished women might vary with time postpartum) and between child gender and child nutritional status (i.e., child gender modifies the relationship between child nutritional status and lactational amenorrhea, or child nutritional status modifies the influence of child gender on lactational amenorrhea). Women of higher parity have an increased likelihood of remaining amenorrheic (Zohoori and Popkin 1996). In addition, demographic and socioeconomic status indicators, such as maternal age, place of residence, availability of a refrigerator and maternal education, influence breast-feeding behavioral characteristics, parity, and maternal and child nutritional status (Adair et al. 1993, Bisgrove et al. 1989, Gross 1991, Huffman et al. 1983).

View larger version (23K): [in this window] [in a new window]   Fig 1. Conceptual framework of the determinants of lactational amenorrhea among breast-feeding women. Lines with circles depict potential interactions between the determinants.

	DATA SETS AND STATISTICAL ANALYSES

Demographic and health surveys.  The Demographic and Health Surveys (DHS) program assists developing countries to conduct national population and health surveys. The program is funded primarily by the United States Agency for International Development (USAID) and is administered by Macro International, in Calverton, MD (Institute for Resource Development/MACRO International 1991). The DHS surveys target women of reproductive age (i.e., 15-49 y of age in most surveys) and are designed to generate nationally representative data on infant feeding practices, population and family planning, and various maternal and child health outcomes (Institute for Resource Development/MACRO International 1990). Sampling frameworks of DHS are based on complex, stratified sampling designs.

The seven countries included in this paper were selected from Sub-Saharan Africa (Table 1). Data from these nations were collected between 1990 and 1994 (DHS II and III). The number of women sampled ranged from 4562 in Ghana to 7540 in Kenya. The subsample included in the analyses was based on women who were not pregnant, who were breast-feeding, who were not using hormonal contraceptives and who had a child 2 y old at the time of survey (n = 9839). Women using hormonal contraceptives (n = 686, 7% of sample) were excluded from the analyses because of the induction of bleeding caused by this type of contraceptives (Laukaran and Winikoff 1985). The proportion of breast-feeding women who remained amenorrheic declined monotonically with time postpartum (Fig. 2) and stabilized after 24 mo postpartum. About 72% of breast-feeding women had resumed menstruation at 24 mo postpartum. Thus, only women with a child 24 mo old were included in the analyses.

View larger version (28K): [in this window] [in a new window]   Fig 2. The percentage of women who were amenorrheic by time postpartum, based on unweighted pooled data from seven Sub-Saharan African countries. Total sample refers to non-pregnant women who have a child 36 mo old and who were not using hormonal contraceptives (n = 16,383). Results for the subsample of women who were breast-feeding (n = 12,298) and women who were no longer breast-feeding (n = 4085) are also presented. The percentage of women who were amenorrheic does not equal 100% at mo 0 postpartum because this time point includes 0 to <1 mo postpartum. Only 2% of women reported bleeding at this time point.

Variables. 

Dependent variable.  At the time of the survey, women were asked if their menstruation had returned since the birth of their youngest child. A dichotomous variable was used to define lactational amenorrhea. Women who reported not having resumed menstruation after the delivery of the last child were coded as amenorrheic or non-amenorrheic if otherwise.

Independent variables.  The main independent variables included in the analyses were maternal body mass index (BMI) and child weight-for-age (wt-for-age) Z-score. Maternal BMI was recoded as a dichotomous variable using a cut-off value of 18.5 kg/m² (James et al., 1988). Child wt-for-age Z-score was also recoded as a dichotomous variable using a cut-off Z-score of 2. A variable for country of origin was entered as an independent variable in the pooled analyses described below.

The covariates entered in the analyses were classified as infant feeding, socioeconomic and demographic. The infant feeding variable included in the analyses was breast-feeding frequency (self-reported total nursing frequency during the 24 h preceding the interview categorized as 1-4 times in 24 h, 5-7 times in 24 h, and 8 times in 24 h). More categories of breast-feeding frequency were explored in the analyses, but all groups 8 times in 24 h were similarly associated with lactational amenorrhea. The socioeconomic and demographic variables incorporated in the analyses were as follows: place of residence (urban or rural), availability of refrigerator in the household (yes or no), maternal education (none or 1 y), child's sex (male or female), parity (number of children), child's age (mo) and maternal age (y).

Statistical analyses.  The Statistical Program for the Social Sciences (SPSS) was used to conduct all statistical analyses (SPSS 1996). A two-tailed P value  0.05 was used as the criterion for statistical significance. Descriptive statistics for each country were generated by using survey sampling weights. Multivariate logistic regression was used with lactational amenorrhea entered as the dependent variable. All independent variables and covariates described above were entered as the independent variables for each within-country analyses. Although maternal BMI and child wt-for-age were significantly correlated, the magnitudes of the correlation were very small (Spearman correlation = 0.075 for <9 mo and 0.119 for 9-24 mo). Thus, multicollinearity was unlikely to occur when both nutritional status variables were included in the model. In addition to the within-country analyses, pooled analyses were performed by combining the data from all seven countries and adding the country identifier variable in the logistic regression model. Pooled analyses were divided into two child age groups, <9 mo (n = 4653) and 9-24 mo (n = 5186). The 9-mo cut-off point was selected to test for the association between maternal nutritional status and lactational amenorrhea in the following two periods of time: 1) when breast milk is likely to be the main source of nourishment, and 2) when complementary foods play a more important role in the child's diet. This cut-off point was used because in Sub-Saharan Africa, during the first 9 mo of life, a large proportion of infants are either exclusively or fully breast-fed (i.e., breast milk and water only), or receive only token amounts of non-breast milks or semisolid foods (Labbok et al. 1997). In the pooled analyses, each covariate was individually tested for a potential interaction with maternal and child nutritional status. Only significant interactions (P value  0.05) were specified in the final multivariate model. Descriptive analyses for the pooled data were weighted to adjust for the differences in sample size across countries. The sample size weight was created from the equation: 1/(7·[n_C/n_T]), where n_C is the sample size for each country and n_T is the sample size for the pooled data. Multivariate analyses were performed with and without the use of sample size weight. Because results were similar with or without the use of sample size weight, only the unweighted analyses are presented for the pooled analyses.

	RESULTS

Abstract Introduction Results Discussion References

Descriptive information.  Table 2 shows the socioeconomic, demographic and biomedical characteristics for the whole sample of "ever-married" women and the subsample of non-pregnant women who were not using hormonal contraceptives and breast-feeding a child 24 mo. Women were, on average, in their late twenties and had a parity ranging from three to four children. As expected, they were of low socioeconomic status and had a low level of education.

Maternal and child anthropometry.  Maternal BMI and the percentage of women who were undernourished are presented in Table 3. The pooled data showed that the mean maternal BMI in the seven countries was 21.8 kg/m² (SD = 3.5) and that 13% of these mothers were undernourished.

Child malnutrition was highly prevalent in these Sub-Saharan African countries (Table 4). Pooled data results showed that overall, the average child wt-for-age Z-score was 1.2 (SD = 1.4), and that 28% of the children were underweight (i.e., wt-for-age Z-score < 2). The mean height-for-age (ht-for-age) Z-score was 1.2 (SD = 1.5) with 29% of children classified as stunted, as defined by a ht-for-age Z-score < 2. The mean weight-for-height (wt-for-ht) Z-score was 0.6 (SD = 1.3), which translated into a wasting (i.e., wt-for-ht Z-score < 2) prevalence of 11%.

Nutritional status and lactational amenorrhea. 

Within-country analyses.  The within-country multivariate logistic regression analyses demonstrated a consistent pattern of association between maternal nutritional status and lactational amenorrhea across all seven Sub-Saharan African countries (Table 5). This relationship was significant in two countries and marginally significant in three countries. In the remaining two countries, the relationship was in the expected direction but was not significant.

Pooled analyses.  For the <9 mo pooled analysis, breast-feeding frequency was positively associated with lactational amenorrhea, although the result was significant only for the breast-feeding frequency of 5-7 times in 24 h (Table 6). Living in urban areas was inversely associated with amenorrhea compared with living in rural areas. Lack of formal education in mothers and the absence of a refrigerator in the household were positively associated with amenorrhea. Women with boys were slightly more likely to remain amenorrheic than women with girls, although the result was not significant. Women of higher parity were more likely to be amenorrheic. As expected, child age was inversely associated with lactational amenorrhea. Mothers with BMI < 18.5 kg/m² were significantly more likely to remain amenorrheic. Child wt-for-age Z-score was inversely associated with amenorrhea. There were no significant interactions between each covariate and maternal or child nutritional status in this age group.

For the 9-24 mo pooled analysis, breast-feeding frequency was positively associated with lactational amenorrhea (Table 6). Living in urban areas was inversely associated with amenorrhea compared with living in rural areas. Lack of formal education in mothers and the absence of a refrigerator in the household were positively associated with amenorrhea. Women of higher parity were more likely to be amenorrheic. Child age was inversely associated with lactational amenorrhea. There was a significant interaction between child wt-for-age and child gender. Women with undernourished children tended to be more likely to remain amenorrheic than women with better-nourished children, although the result was not significant. Women with boys were slightly more likely to remain amenorrheic than women with girls. The combination of child nutritional status and gender acted synergistically on the likelihood of lactational amenorrhea. Women with undernourished boys were significantly more likely to be amenorrheic than women with better nourished girls. Another significant interaction was found between maternal BMI and child age. To understand the meaning of the interaction, mean predicted probabilities of being amenorrheic were calculated by time postpartum and by maternal nutritional status using case-specific predicted probabilities (Fig. 3). As illustrated in the figure, undernourished women were more likely to be amenorrheic than better-nourished women. The magnitude of this difference increased with time postpartum, particularly after 17-mo postpartum. The median duration of lactational amenorrhea for undernourished women was 16.5-mo postpartum, compared with 15.1-mo postpartum for their better-nourished counterparts. Thus, the adjusted difference in the median duration of amenorrhea associated with maternal nutritional status among breast-feeding women was 1.4 mo.

View larger version (18K): [in this window] [in a new window]   Fig 3. Predicted probability of remaining amenorrheic among breast-feeding women by months postpartum and by maternal nutritional status. Results of 9-24 mo pooled analysis, after controlling for all other variables in the multiple logistic regression. Undernourished women had higher probability of remaining amenorrheic than better-nourished women, and the magnitude of probability differences increased with time postpartum. The median duration of lactational amenorrhea for undernourished women was 16.5-mo postpartum, compared with 15.1-mo postpartum for their better-nourished counterparts. Thus, the adjusted difference in the median duration of amenorrhea associated with maternal nutritional status among breast-feeding women was 1.4 mo.

	DISCUSSION

Abstract Introduction Results Discussion References

These analyses support previous findings suggesting that breast-feeding practices affect the likelihood of amenorrhea. Mothers who breast-feed their children more frequently were less likely to have resumed menstruation. Parity was found to be positively associated with the likelihood of remaining amenorrheic. Socioeconomic status was found to be inversely associated with lactational amenorrhea. Living in an urban area, higher level of maternal education and the availability of a refrigerator were all associated with a lower likelihood of remaining amenorrheic. It is possible that women of low socioeconomic status have less access to food for their child. Thus, they may breast-feed their child more frequently or spend more time per day doing it, and as a result have a higher likelihood of remaining amenorrheic. As expected, we could not fully break down the effect of socioeconomic status on all intermediate and proximal determinants explaining lactational amenorrhea. The fact that socioeconomic status remained significantly associated with lactational amenorrhea after controlling for the more proximate determinants of amenorrhea suggests that there are additional unmeasured factors that affect lactational amenorrhea. It is important for future studies to measure factors such as breast-feeding duration (i.e., minutes breast-feeding per day) and intensity to identify additional intermediate variables that mediate the association between socioeconomic status and lactational amenorrhea.

Mothers who were undernourished were found to have a higher likelihood of remaining amenorrheic even after controlling for nursing frequency, child nutritional status, and socioeconomic and demographic covariates. Although the results were significant in only two countries, this trend was highly consistent across all seven countries included in the analyses. The pooled analysis for infants < 9 mo indicated that undernourished women were 1.6 times more likely to be amenorrheic compared with better-nourished women. One possible explanation is that women with low BMI have less body fat to support the return of menses after giving birth. Poorly nourished women may experience greater inhibition of the ovulatory hormones than better-nourished women, given the same amount of sucking, and thus are more likely to remain amenorrheic (Kurz et al. 1993). Another possibility is that undernourished women produce less milk per nursing episode (Delgado et al. 1982, Lunn et al. 1984), and their children need to suck longer or more intensely than children of better-nourished mothers to obtain the amount of milk that they require. This increase in sucking frequency or intensity might be associated with the increase in plasma prolactin level and thus increase the likelihood of being amenorrheic (Loudon et al. 1983). The effect of low maternal nutritional status on lactation amenorrhea was also observed in the 9-24 mo pooled analysis, in which the differential probability of remaining amenorrheic between undernourished and better-nourished women increased over time (Fig. 3). Prolonged lactation has an effect on maternal energy reserves and causes maternal weight loss. It is possible that prolonged lactation in women decreases the amount of body fat that is needed to resume menstruation, and the reduction of body fat accumulates over time. Therefore, the probability of undernourished women remaining amenorrheic is likely to be greater and increase over time compared with better-nourished women.

The specific biological mechanisms that explain our findings have not been fully elucidated (Dewey 1996). Recent studies, however, are improving our understanding of how maternal nutritional status may influence postpartum amenorrhea. A recent study conducted with ovariectomized rats found that unrestricted access to food, after 50% food restriction during pregnancy, increased maternal weight and shortened the duration of the postpartum anestrus period compared with the group whose diet was restricted by 50% throughout the study (i.e., pregnancy and lactation) (Gournis et al. 1997). Unrestricted access to food during lactation was associated with higher levels of luteinizing hormone and follicle stimulating hormone. Behavioral measurements indicated that pups of supplemented dams suckled less intensively and the dams spent more time away from the pups. In this seminal study, it was not possible to separate the influence of maternal body composition from behaviors leading to less sucking on the metabolic/physiologic changes determining the duration of the anestrus period. It has been suggested that changes in sucking behavior are more likely than maternal nutritional status per se to influence the duration of postpartum amenorrhea (Loudon et al. 1983). In addition to human epidemiologic studies, there is accumulating evidence strongly suggesting that maternal nutritional status does play an independent role in the return of fertility. Leptin, a protein hormone released from adipocytes, appears to play an important role in reproductive performance (Frübeck 1997). Injecting leptin-deficient mice with this hormone restores fertility (Chehab et al. 1996) and hastens the onset of reproductive function consistent with changes in luteinizing hormone and 17-estradiol levels (Chehab et al. 1997). Because leptin is produced by adipocytes, it may prove to be a crucial link between maternal nutritional status and the duration of postpartum amenorrhea. Findings from a study conducted with female adolescent patients diagnosed with anorexia nervosa indicated that at first contact, their leptin plasma level was below that of matched controls. The subsequent weight improvements that occurred after treatment were positively correlated with circulating leptin levels (Heberbrand et al. 1997). Mean serum leptin concentration during the first 4 wk of treatment of anorexic females was positively correlated with follicle stimulating hormone, luteinizing hormone and estradiol levels (Kopp et al. 1997). This research group also found that low leptin levels predicted amenorrhea among anorexic patients (Kopp et al. 1997). It is important that future studies examine the role that leptin plays in female reproductive function in developing country settings. These studies may help improve the understanding of the physiologic mechanisms underlying the influence of maternal undernutrition on lactational amenorrhea under conditions of chronic energy undernutrition and/or negative energy balance.

Mothers who had children who were underweight also were found to be more likely to remain amenorrheic. The effect of child nutritional status on lactational amenorrhea was observed independently of maternal nutritional status. The <9 mo pooled analysis showed that women with undernourished children were 1.4 times more likely to be amenorrheic compared with women who had better-nourished children. It is possible that poorly nourished children are those not receiving adequate weaning foods after 6 mo of age and are more likely to be breast-feeding more intensively (Dewey et al. 1997, Kurz et al. 1993). Child gender did not significantly affect the probability of remaining amenorrheic. No significant interaction between child nutritional status and the covariates was detected in this age group.

In the older age group, the combination of child gender and nutritional status was found to have a synergistic effect on lactational amenorrhea. Women with undernourished boys were 1.2 times more likely to remain amenorrheic compared with women who had better-nourished girls. It is possible that undernourished boys are breast-fed more frequently or for longer bouts in relation to better-nourished girls.

Kurz et al. (1993) found a significant negative association between maternal nutritional status, as assessed by triceps skinfold thickness, and the duration of lactational amenorrhea. When infant supplementation was accounted for, however, this association became only marginally significant and, according to the authors, of little biological importance. In our analyses, better-nourished women were significantly less likely to remain amenorrheic even after controlling for breast-feeding frequency and child nutritional status. The difference in the adjusted median duration of amenorrhea between undernourished and better-nourished women in the 9-24 mo subsample was 1.4 mo compared with the 0.5 mo found by Kurz et al. (1993). The magnitude of the effect of maternal nutritional status on amenorrhea was similar to the one found in the subsample of younger children. Thus, our results suggest that maternal nutritional status has a biologically significant independent influence on postpartum amenorrhea. On the one hand, it is important to note that this is a cross-sectional study, whereas the study by Kurz et al. was an experimental nutritional supplementation longitudinal study that could have provided more precise estimates. Other differences between the two studies, such as population characteristics and maternal nutritional status, also could have contributed to the differences in the findings. For example, it is possible that the degree of chronic energy deficiency may be more pronounced in Sub-Saharan Africa than in Guatemala.

The directionality and magnitude of the association between maternal nutritional status and lactational amenorrhea was highly consistent across countries. This finding confirmed our a priori expectation that maternal nutritional status has a direct biologically mediated influence on amenorrhea, and that this association could be detected after controlling for socioeconomic, demographic and biocultural confounders, including infant feeding behaviors (proxied by nursing frequency in our analyses). Subsequent pooling of the data was very useful for "summarizing" the directionality, magnitude and significance of this relationship. The pooled analyses also controlled for a number of confounders to adjust results for cross-country socioeconomic, demographic and biocultural variations. The fact that the "country" effect was significant suggests that there are unmeasured variables that differ among countries and influence amenorrhea. For example, it is likely that infant feeding behaviors such as nursing frequency, duration (time spent breast-feeding per day) and intensity varied from country to country.

In summary, our analyses confirmed that breast-feeding frequency is positively associated with lactational amenorrhea. After controlling for child nutritional status, breast-feeding frequency, and other potential socioeconomic and demographic confounders, maternal nutritional status was independently and inversely associated with lactational amenorrhea.

	FOOTNOTES

Manuscript received 28 July 1997. Initial reviews completed 11 September 1997. Revision accepted 5 June 1998.

	LITERATURE CITED

Abstract Introduction Results Discussion References

• Adair L. S., Popkin B. M., Guilkey D. K. The duration of breastfeeding: how is it affected by biological, sociodemographic, health sector, and food industry factors. Demography 1993; 30:63-80[Medline]
• Bisgrove E. Z., Popkin B. M., Barba C. Infant feeding in the Philippines: a cluster analysis approach. Ecol. Food Nutr. 1989; 23:75-90
• Bongaarts, J. (1980) Does malnutrition affect fecundity? A summary of Evidence. Science (Washington, DC) 208: 564-569.
• Campino C., Ampuero S., Díaz S., Serón-Ferré M. Prolactin bioactivity and the duration of lactational amenorrhea. J. Clin. Endocrinol. Metab. 1994; 79:970-974[Abstract]
• Chehab F. F., Lim M. E., Lu R. Correction of the sterility defect in homozygous obese female mice by treatment with the human recombinant leptin. Nat. Genet. 1996; 12:318-320[Medline]
• Chehab, F. F., Mounzih, K., Lu, R. & Lim, M. E. (1997) Early onset of reproductive function in normal female mice treated with leptin. Science (Washington, DC) 275: 88-90.
• Delgado H., Brineman E., Lechtig A., Bongaarts J., Martorell R., Klein R. E. Effect of maternal nutritional status and infant supplementation during lactation on postpartum amenorrhea. Am. J. Obstet. Gynecol. 1979; 135:303-307[Medline]
• Delgado H. L., Martorell R., Klein R. E. Nutrition, lactation, and birth interval components in rural Guatemala. Am. J. Clin. Nutr. 1982; 35:1468-1476[Abstract/Free Full Text]
• Delvoye P., Demaegd M., Delogne-Desnoeck J., Robyn C. The influence of the frequency of nursing and of previous lactation experience on serum prolactin in lactating mothers. J. Biosoc. Sci. 1977; 9:447-451[Medline]
• Dewey K. G. Does maternal supplementation shorten the duration of lactational amenorrhea? Am. J. Clin. Nutr. 1996; 64:377-378[Free Full Text]
• Dewey K. G., Cohen R. J., Rivera L. L., Canahuati J., Brown K. H. Effects of age at introduction of complementary foods to breastfed infants on duration of lactational amenorrhea in Honduran women. Am. J. Clin. Nutr. 1997; 65:1403-1409[Abstract/Free Full Text]
• Diaz S., Rodriguez G., Marshall G. Breastfeeding pattern and the duration of lactational amenorrhea in urban Chilean women. Contraception 1988; 35:37-51
• Frisch R. E., McArthur J. W. Menstrual cycles: fatness as a determinant of minimum weight for height necessary for their maintenance or onset. Science (Wash. DC) 1974; 185:949-951[Abstract/Free Full Text]
• Frisch R. E., Revelle R., Cook S. Components of the critical weight at menarche and at initiation of the adolescent growth spurt: estimated total water, lean body mass, and fat. Hum. Biol. 1973; 48:469-483
• Frübeck G. Leptin involvement in reproductive performance. J. Nutr. 1997; 127:1533[Free Full Text]
• Genazzani A. D., Petraglia F., Benatti R., Montanini V., Algeri I., Volpe A., Genazzani A. R. Luteinizing hormone (LH) secretory burst duration is independent from LH, prolactin, or gonadal steroid plasma levels in amenorrheic women. J. Clin. Endocrinol. Metab. 1991; 72:1220-1225[Abstract]
• Gournis E., McGuire M. K., Rasmussen K. M. Food supplementation during lactation shortens anestrus and elevates gonadotropins in rats. J. Nutr. 1997; 127:785-790[Abstract/Free Full Text]
• Gross B. A. Is the lactational amenorrhea method a part of natural family planning? Biology and policy. Am. J. Obstet. Gynecol. 1991; 165:2014-2019[Medline]
• Habicht J. P., Da Vanzo J., Butz W. P., Meyers L. The contraceptive role of breastfeeding. Popul. Stud. 1985; 39:213-232
• Heberbrand J., Blum W. F., Barth N., Coners H., Englaro P., Juul A., Ziegler A., Warnke A., Rascher W., Remschmidt H. Leptin levels in patients with anorexia nervosa are reduced in the acute stage and elevated upon short-term weight restoration. Mol. Psychiatry 1997; 2:330-334[Medline]
• Huffman S. L. Maternal and child nutritional status: its association with the risk of pregnancy. Soc. Sci. Med. 1983; 17:1529-1540
• Huffman S. L., Ford K., Allen H. A., Streble P. Nutrition and fertility in Bangladesh: breastfeeding and postpartum amenorrhea. Popul. Stud. 1987; 41:447-462
• Institute for Resource Development/MACRO International (1990) An Assessment of Demographic and Health Surveys-I: Data Quality. Calverton, MD.
• Institute for Resource Development/MACRO International (1991) Proceedings of the Demographic and Health Surveys world conference, August 5-7, Washington, DC. Calverton, MD.
• James W.P.T., Ferro-Luzzi A., Waterlow J. C. Definition of chronic energy deficiency in adults. Eur. J. Clin. Nutr. 1988; 42:969-981[Medline]
• Kopp W., Blum W. F., von Prittwitz S., Ziegler A., Lubbert H., Emons G., Herzog W., Herpertz S., Deter H. C., Remschmidt H., Heberbrand J. Low leptin levels predict amenorrhea in underweight and eating disordered females. Mol. Psychiatry 1997; 2:335-340[Medline]
• Kurz K. M., Habicht J. P., Rasmussen K. M., Schwager S. J. Effects of maternal nutritional status and maternal energy supplementation on length of postpartum amenorrhea among Guatemalan women. Am. J. Clin. Nutr. 1993; 58:636-640[Abstract/Free Full Text]
• Labbok, M. H. & Laukaran, V. H. (1994) Breastfeeding and family planning. In: Gynecology and Obstetrics, Vol. 6 (Dilts, P. V. & Sciarra, J. J., eds.), pp. 1-13. J. B. Lippincott, Philadelphia, PA.
• Labbok, M. H., Pérez-Escamilla, R., Peterson, A. E. & Coly, S. (1997) Breastfeeding and Child Spacing Country Profiles, pp. 7-50. Institute for Reproductive Health, Washington, DC.
• Labbok, M., Stallings, R., Shah, F., Perez, A., Klaus, H., Jacobson, M. & Muruthi, T. (1991) Ovulation method used during breastfeeding: is there increased risk of unplanned pregnancy? Am. J. Obst. Gyn. 165 (suppl. 2): 2031-2036.
• Laukaran V. H., Winikoff B. Contraceptive use, amenorrhea, and breastfeeding in postpartum women. Stud. Fam. Plann. 1985; 16:293-301[Medline]
• Loudon A.S.I., McNeilly A. S., Milne J. A. Nutrition and lactational control of fertility in red deer. Nature (London) 1983; 302:145-147[Medline]
• Lunn P. G., Austin S., Prentice A. M., Whitehead R. G. The effect of improved nutrition on plasma prolactin concentrations and postpartum infertility in lactating Gambian women. Am. J. Clin. Nutr. 1984; 39:227-235[Abstract/Free Full Text]
• Perez A., Labbok M. H., Queenan J. T. Clinical study of the lactational amenorrhea method of family planning. Lancet 1992; 339:968-970[Medline]
• Pérez-Escamilla, R., Peterson, A. E., Laukaran, V. H. & Labbok, M. H. (1997) Nursing frequency is positively associated with lactational amenorrhea in developing countries: results from DHS II and III. Poster presentation at Experimental Biology Conference, April 5-9, 1997, New Orleans, LA. FASEB J. 11: A395 (abs.).
• Popkin B. M., Guilkey D. K., Akin J. S., Adair L. S., Udry J. R. Nutrition, lactation, and birth spacing in Filipino women. Demography 1993; 30:333-352[Medline]
• SPSS Inc. (1996) SPSS for Windows, Release 7.0, Chicago, IL.
• Tracer D. P. Lactation, nutrition, and postpartum amenorrhea in lowland Papua New Guinea. Hum. Biol. 1996; 68:277-292[Medline]
• Zohoori N., Popkin B. M. Longitudinal analysis of the effects of infant-feeding practices on postpartum amenorrhea. Demography 1996; 33:167-180[Medline]

0022-3166/98 $3.00 ©1998 American Society for Nutritional Sciences