Indmedica Home | About Indmedica | Medical Jobs | Advertise On Indmedica
Search Indmedica Web
Indmedica - India's premier medical portal

Indian Journal of Community Medicine

Breastfeeding Duration in Bangladesh and Factors Associated With It

Author(s): M.S. Giashuddin, M. Kabir*

Vol. 28, No. 1 (2003-01 - 2003-03)

Deptt. of Epidemiology and Biostatistics, Institute of Child and Mother Health, Matuail, Dhaka, Bangladesh *Deptt. of Statistics, University of Jahangirnagar, Savar, Dhaka, Bangladesh

Abstract:

Research question: What is the duration of breastfeeding in Bangladesh and what are the factors affecting it?

Objective: To assess the duration of breastfeeding among women in Bangladesh and socio-economic and demographic factors affecting it.

Design and Setting: Bangladesh Demographic and Health Survey 1999-2000 data were used for this study

Participants: The study included mothers of 6,939 children from six geographical divisions.

Methods: Information was collected from the women on socio-demographic characteristics and current breastfeeding status of their last and next to last child. The risk of complete weaning at different ages was estimated by life table survival method Independent effects of each of the study variables were estimated by using Cox's regression model.

Results: The median duration of breastfeeding was 30 months. Life table analysis showed complete risk of weaning children was found at age 3 months (1.3%), 6 months (1.1 %) and 12 months (4.3%) during the first year of l i f e. Cox regression analysis revealed that women who had lived in rural area were less likely to terminate breastfeeding than the women who lived in urban area (OR = 0 .78; 95% CI = 0. 70-0 .86). Women who had completed at least ten years education were more likely to stop breastfeeding than their counterparts (OR = 1 .32, 95% CI = 1.19-1.46). The women who receive antenatal care by the health professional had lower risk of terminating breastfeeding than the women who did not receive antenatal care (OR = 0 .57; 95% CI = 0 .51-0 .63).

Conclusion: Future breastfeeding programmes should give special attention to the mothers with higher education, those who lived in urban areas and did not receive antenatal care, since they breastfed relatively shorter periods of time.

Keywords: Breastfeeding, Maternal education, Bangladesh, Hazard model

Introduction:

Breastfeeding is the unique source of nutrition that plays an important role in the growth, development and survival of infants. The benefits of breastfeeding, especially exclusive breastfeeding, are well established 1,2, particularly in poorer environments where the early introduction of other milk is of particular concern because of the risk of contamination and over dilution of milk leading to increased risks of morbidity and under nutrition 2. Breastfeeding is promoted internationally as the proffered method of feeding infant s up to 4-6 months and continued up to 2 years with the addition of home cooked food 3-5 . The duration of breast feeding largely determines how much this form of feeding has a favourable influence on infectious diseases prevention, child development and child survival. Study of factors determining duration of breastfeeding is important because the success of nutrition programmes depend on identification of factors susceptible to intervention that can serve as the basis for designing and implementing such programmes. Breastfeeding is universal in Bangladesh.

Almost 97% of mothers breastfed their children for some period of time 6,7. The proportion of Bangladeshi women initiating breastfeeding is relatively low 6-8. The duration of breastfeeding has frequently been associated with socio-economic and demographic factors such as maternal age, education etc 9,12. Breastfeeding duration tends to increase with parity of mothers 9,13.

Material and Methods:

The primary source of data for the present study was Bangladesh Demographic and Health Survey (BDHS) 1999-2000, which collected information throughout the nation during November 1999 and March 2000. BDHS employed a two-stage probability sample design to select respondents. At the first stage the primary sampling units chosen from the Integrated Multi-purpose Master Sample (IMPS) created by Bangladesh Bureau of Statistics. A total of 341 primary sampling units were used for the BDHS survey (99 units in urban areas and 242 in rural areas)14 . It was a nationally representative sample. The study population consisted of 10,544 ever-married women between ages 10-49 years. Eligible mothers of 6,939 children were interviewed and information was collected on socio-demographic characteristics and current breastfeeding status of their last and next to last child. Data regarding duration of breastfeeding among children born over the preceding four years were gathered through interviews with study mothers. Although breastfeeding rarely continues more than four years, only mothers with breastfed 48 months old and their last-born breastfed child were selected for this study.

The duration of breastfeeding was defined as the child's age (in months) at the time of complete weaning, regardless of when consumption of other food began. For the children who had died or still being breastfed at the time of the survey, duration of total breastfeeding was considered equal to child's age at death or at the time of the survey. The risk of complete weaning-at each month of age (the probability that a child being breastfed at the beginning of a given month would be completely weaned during that month) and the cumulative risk of complete weaning was estimated by life table survival method. Survival analysis provides a good understanding of breastfeeding behaviour over time. It is used due to the presence of censored data. Data from those cases where mothers continued breastfeeding beyond the end of the survey period or children died before the survey referred censored data. The median duration and trimean of total breastfeeding were used as the summary measure.

The factors affecting duration of breastfeeding were investigated in a regression using Cox's proportional hazards model15 . This model permits joint estimation of the effects of independent variables on the "hazard" (the risk of stop breastfeeding), rather than the duration itself and can be used to analyse data that contain censored observations. The assumption of proportionality of categorical variables was tested by calculating the log minus -log of the survival function of each explanatory variable and plotting it against log-time. In the model, completed data were available for 5700 of the 6939 breastfeeding mothers. The degree of statistical significance of the variables included in the model was calculated by using Wald's test 16 . Statistical Package for Social Science (SPSS for Windows 10.0) were used to analyse the data 17.

Results:

Mean age of the respondents in this study was 25. 72-6.43 years. About 17% children lived in urban areas and the rest of them lived in rural areas. Among the mothers of the children 47% of had no formal education, whereas, only 24% mothers had secondary and higher education. More than 51 % children were male. Almost 72% mothers did not receive prenatal care.

Table I: Life table estimate of the risk of weaning at different age.

Age
Interval
(month)
Conditional risk
of complete
weaning (%)
Cumulative
risk of
complete weaning
95% CI
0-1 2.66 - -
1-2 1.28 2.6 2.2 - 3.0
2-3 0.78 3.9 3.4 - 4.4
3-4 1.34 4.6 4.1 - 5.1
4-5 0.39 5.9 5.3 - 6.5
5-6 0.67 6.2 5.7 - 6.9
6-7 1.10 6.9 6.3 - 7.5
7-8 0.31 7.9 7.3 - 8.6
8-9 0.49 8.2 7.5 - 8.9
9-10 0.42 8.6 8.0 - 9.3
10-11 0.39 9.06 8.3 - 9.7
11-12 0.06 9.4 8.7 - 10.1
12-13 4.33 9.4 8.7 - 10.2
13-14 0.16 13.3 12.5 - 14.2
14-15 0.28 13.5 12.7 - 14.3
15-16 0.59 13.7 12.9 - 14.6
16-17 0.51 14.2 13.4 - 15.1
17-18 0.13 14.7 13.8 - 15.5
18-19 8.87 4.8 13.9 - 15.7
19-20 0.3 22.3 21.3 - 23.4
20-21 1.23 22.6 21.6 - 23.6
21-22 0.29 23.5 22.5 - 24.6
22-23 0.66 23.7 22.7 - 24.8
23-24 0.38 24.2 23.2 - 25.3
24-25 25.74 24.5 23.5 - 25.6
25-26 0.56 43.9 42.7 - 45.2
26-27 0.81 44.3 43.1 - 45.5
27-28 0.80 44.7 43.3 - 45.9
28-29 1.05 45.1 43.9 - 46.4
29-30 0.69 45.7 44.3 - 47.0
30-31 24.21 46.1 44.9 - 47.3
31-32 0.27 59.2 58.9 - 60.3
32-33 1.73 59.3 59.3 - 60.5
33-34 0.40 59.8 59.8 - 61.0
34-35 0.84 59.9 59.9 - 63.1
35-36 0.55 60.3 59.1 - 61.4
36-37 37.94 60.5 60.5 - 63.7
37-38 2.03 75.5 76.4 - 76.5
38-49 2.04 75.7 75.6 - 76.7
39-40 1.77 76.2 75.2 - 77.2
40-41 2.24 76.6 75.6 - 77.2
41-42 0.31 77.1 76.1 - 78.1
42-43 6.86 77.2 76.2 - 78.2
43-44 1.76 78.6 77.7 - 79.7
44-45 2.79 78.9 77.3 - 79.8
45-46 2.09 79.1 76.3 - 79.8
46-47 0.59 79.7 78.7 - 81.1
47-48 30.97 80.1 79.3 - 81.1
48-49 - 80.2.2 79.3 - 81.2
49-50 6.1 86.3 85.5 - 87.2
50-51 1.57 86.7 85.5 - 87.2
51-52 5.88 87.2 85.3 - 88.0
52-53 - 87.4 86.5 - 88.2
53-54 3.28 88.3 87.3 - 88.9
54-55 - 88.1 87.3 - 88.9
55-56 6.67 88.5 87.7 - 89.3

The cumulative risk OR = Odds Ratio; CI = Confidence Interval; TBA = Trained birth attendant.

The median duration of breastfeeding in Bangladesh was 30 months (Table II). The results revealed that urban mothers breastfed for relatively shorter duration, i.e. tri mean 28.5 months, compared with rural women. The respondents who had less than 10 years of schooling were found to breastfeed on an average 31.5 months, which decreased to 29.5 months for mothers who had secondary and higher education. Mother's age at the birth of child had impact on the duration of breastfeeding. The women who received antenatal care breastfed for longer duration (Median 36 months and Tri mean 37 months) than the women who did not receive antenatal care.

Table II: Life table estimates of median duration of Breastfeeding (MBDF) ande risk of shorter duration breastfeeding by selected socio-econimic and demographic characteristics (BDHS 1999-2000)

Characteristics MBDF
(months)
Tri
mean
OR 95% CI
Residence
Urban 30 28.50 1.00  
Rural 30 31.50 0.78 0.70 - 0.86
Maternal Education
<10 yrs 30 31.50 1.00  
>10yrs 30 29.50 1.32 1.19 - 1.46
Child's Sex
Male 30 30.5 1.00  
Female 30 30.25 0.99 0.91 - 1.07
Maternal age
≤25 years 30 29.75 1.00  
≥25 years 30 30.75 0.93 0.87 - 1.01
Access to Mass Media
no 30 30.5 1.00  
yes 30 29.5 1.05 0.95 - 1.16
Antenatal Care
no 30 28.5 1.00  
health professional 36 36.00 0.57 0.51 - 0.63
TBA 36 37.50 0.44 0.29 - 0.65
Division
Banishal 30 30.00 0.67 0.57 - 0.80

Socio-economic and demographic factors associated with breastfeeding duration are given in Table II. Multivariate analysis showed that women who had lived in rural area were 22% less likely to terminate breastfeeding than the women in urban area. The hazard (risk of stop breastfeeding) increased with increasing maternal education. Women who had completed the higher (.... 10 years) education were 1.32 time more likely to stop breastfeeding than the women who had less than 10 years of schooling. Female children had the lower risk of stop breastfeeding than the male children. Women with older age (>25 years) had lower risk of stop breastfeeding as compared to the women with younger age. The women who received antenatal care by the health professionals had 43% lower risk of terminating breastfeeding than the women who did not receive antenatal care. The women of Dhaka, Barshal, Rajshahi and Khulna division were less likely to stop breastfeeding than the women of Sylhet division.

Discussion:

The estimated median duration of 30 months found in this study was almost same as reported (30.5 months) by Bangladesh Demographic and Health Survey 1999-2000 14. The difference might be attributable as this study used life table survival analysis to estimate median duration of breastfeeding. Duration of breastfeeding has declined in last decade 14 . Mannan 18 obtained average duration of breastfeeding in Bangladesh for surviving children as 28.2 months which is close to this study. As compared to South Asian countries (India, Pakistan and Srilanka) where average duration of breastfeeding was 18.4 months, 21.8 months and 23.2 months respectively breastfeeding duration is relatively longer in Bangladesh m19-21 . Highest conditional risk of complete weaning was found at 3, 6, and 12 months of child age. Another study by Bautista 22 revealed similar results.

Mother's age at birth had the negative association with duration of breastfeeding. Women with more than 25 years of age had lower risk of stopping breastfeeding than younger ages. Ahamed 23 found that the average duration of breastfeeding increased with mother's age in Bangladesh. A significant association was found between place of residence and duration of breastfeeding. Urban children were more likely to be weaned at any time as compared with rural children. In Southeast Asia, women in urban areas breastfed less than rural areas 24. A similar tendency was found among the mothers in some other studies 22,25 .

The study found the duration of total breastfeeding to be shorter among mothers having secondary and higher education than among those with less or not educated mothers. In developing countries, higher education is associated with shorter duration of breastfeeding 26. On the other hand, in the industrialized countries duration of breastfeeding appears to increase with the maternal educational level 25 . The result revealed that risk of weaning was slightly lower for the female than male. In developing countries, breastfeeding duration did not appear to vary by the sex of child 18,22,26 . The shorter duration was associated with antenatal care. The women who had visited to doctors, nurses or trained birth attendant for antenatal care during pregnancy were more likely to breastfed than the women who did not visit for antenatal check-up. In this study significant difference in breastfeeding was observed in administrative divisions. Similar results were also found by other studies 6,18 .

The study results indicated that breastfeeding duration in Bangladesh is higher than the other South Asian countries. Since male children had breastfed shorter than the female children, breastfeeding promotion programmes should address the mothers that male infants have higher nutritional need than the female infants. In addition, the women with higher education, those who lived in urban areas and did not receive antenatal care tend to breastfeed their children for relatively shorter periods, so breastfeeding promotion programme in Bangladesh should be designed focussing on them.

References:

  1. Victora CG, Smith PG, Vaughan JP. Evidence for protection by breastfeeding against infant deaths from infectious diseases in Brazil. Lancet 1987; 2: 319-22.
  2. Huttly SRA, Morris SS, Pisani V. Prevention of diarrhoea in young children in developing countries. Bulletin of World Health Organisation 1997; 75: 163-74.
  3. World Health Organisation. The quantity and quality of breast milk. Report of the WHO collaborative study on breastfeeding. Geneva: World Health Organisation, 1985.
  4. Talukder MQK. Bangladesh Campaign for the Protection and Promotion of- Breastfeeding. Bangladesh J Child Health 1992; 16: 25-31.
  5. UNICEF. The Innocenti Declaration on the protection, promotion and support breastfeeding. New York: UNICEF, 1990.
  6. National Institute of Population Research and Training (NIPORT), Mitra and Associates, Macro International Inc. Bangladesh Demogrpahic and Health Survey 1996-97. Calverton, Maryland: NIPORT, Mitra and Associates and macro International, 1997.
  7. Ahmed S, Parveen SD, Islam A. Infant feeding practices in rural Bangladesh: Policy implication. J Trop Paediatrics 1999; 45: 37-41.
  8. Bangladesh Breastfeeding Foundation (BBF). Situation of Breastfeeding in Bangladesh. BBF Research Bulletin 2001; 7(2): 4-5.
  9. Nolan L, Goel V. Socio-demographic factors related to breastfeeding in Ontario: results from Ontario Health Survey. Can J Public Health 1995; 86: 309-12.
  10. Ford K, Labbok M. Who is breastfeeding? Implications of associated social and biomedical variables for the research on the consequences of method of infant feeding. Am J Clin Nutr 1990; 52: 451-6.
  11. Quarles A, Williams PD, Hoyle DA, Brimeyer M, Williams AR. Mother intention, age, education and duration and management of breastfeeding. Maternal-Child Nurs J 1995; 22: 102-8.
  12. Michwelsen KF, Larsen PS, Thomson BL, Samuelson G. The Copenhagen cohort study on infant nutrition and growth: duration of breastfeeding and influencing factors. Acta Paediatr 1994; 83: 565-71.
  13. Piper S, Parks PL. Predicting the duration of lactation: evidence from a national survey. Birth 1996; 23: 7-12.
  14. National Institute of Population Research and Training (NIPORT), Mitra and Associates, ORC Macro. Bangladesh Demographic and Health Survey 1999-2000. Calverton, Maryland: NIPORT, Mitra and Associates and ORC Macro, 2001.
  15. Lawless JF. Statistical Model and Methods for Life Time Data. New York: John Wiley and Sons, 1982; 343-400.
  16. Clayton D, Hills M. Statistical Models in Epidemiology. Oxford: Oxford University Press 1993;23-109.
  17. SPSS Inc. SPSS for windows version 10.0 Chicago: SPSS Inc., 1999.
  18. Mannan RH, Islam MN. Breastfeeding in Bangladesh: patterns and impact on fertility. Asia-Pacific Population Journal. 1995; 10(4): 23-38.
  19. Rajaretnam T. Trend and differentials in breastfeeding and amenorrhea durations in a rural area in south India. Demography India 1994; 23(1-2): 83-95.
  20. Page HJ, Lesthaeghe RJ, Shah IH. Illustrative analysis: Breastfeeding in Pakistan. WFS Scientific Reports No. 37. Netherlands: International Statistical Institute, 1982.
  21. Mahler K. Women breastfeeding infants longer in many developing countries. International Family Planning Perspectives 1996; 22(3): 134-5.
  22. Bautista LE. Duration of maternal breastfeeding in the Dominican Republic. Pan Am J Pub Health 1997; 1(2): 104-11.
  23. Ahamed MM. Breastfeeding in Bangladesh. Journal of Biosocial Science 1986; 18(4): 425-34.
  24. Williamson NE. Breastfeeding trends and patterns. International Journal of Gynaecology and Obstetrics 1989;(Suppl 1): 145-52.
  25. Vestermark V, Hogdall CK, Plenov G, Brich M, Toftager-Larsen K. The duration of breastfeeding: a longitudinal prospective study in Denmark. Scand J Soc Med 1991; 19: 105-9. 26. Grummer-Strawn LM. The effect of changes in population characteristics on breastfeeding: trends in fifteen developing countries. International Journal of Epidemiology 1996; 25(1): 94-102.
Access free medical resources from Wiley-Blackwell now!

About Indmedica - Conditions of Usage - Advertise On Indmedica - Contact Us

Copyright © 2005 Indmedica