OUP user menu

Impact of working status on breastfeeding in Singapore
Evidence from the National Breastfeeding Survey 2001

Gary Ong, Mabel Yap, Foo Ling Li, Tai Bee Choo
DOI: http://dx.doi.org/10.1093/eurpub/cki030 424-430 First published online: 19 July 2005

Abstract

Background: This study uses data from the 2001 Singapore National Breastfeeding Survey to examine factors, including working status, associated with breastfeeding duration. Methods: All women who delivered in the eight hospitals with obstetric services in Singapore from 1 April to 31 May 2001 were invited to participate in a survey on infant feeding and nutrition at 2 and 6 months postpartum. A total of 2149 respondents were interviewed for variables that were known or suspected to be associated with breastfeeding initiation and duration. Cox proportional hazards model was used to determine the associated effect of working status on breastfeeding duration. Kaplan–Meier estimate and survival curves were compared between working and non-working mothers. Results: Working status had no effect on initiation of breastfeeding, but had an effect on breastfeeding duration. The median breastfeeding duration for non-working and working mothers was 9 weeks and 8 weeks, respectively. This difference was significant by log rank test [hazard ratio (HR) = 1.27, 95% confidence interval (CI) 1.14–1.41, P value <0.001]. About 31% of non-working mothers breastfed for up to 6 months as compared to 20% of working mothers. Working mothers were more likely to stop breastfeeding than non-working mothers (HR = 1.61, 95% CI 1.43–1.85, P value = 0.001) after adjusting for potential confounders. The most important reason for working mothers stopping breastfeeding between 2 and 6 months was attributable to work. Conclusion: More breastfeeding-friendly initiatives need to be put in place at workplaces to encourage working mothers to continue breastfeeding upon returning to work.

  • breastfeeding duration
  • exclusive breastfeeding
  • motivating factors
  • working status

Females constituted approximately 40% of the economically active resident Singapore workforce in the year 2000. The labour force participation rate in the year 2000 for married women was 49.2%.1 The duration of maternity leave in Singapore is 56 days (2 months). In addition, working mothers may choose to take childcare leave after their maternity leave ends. However, there is no fixed policy for childcare leave and it varies from company to company.

Various foreign and local studies have shown varying effects of work status on the initiation and duration of breastfeeding. Most studies showed that there was no association between maternal employment and decision to breastfeed.24 The intention to return to work was observed to be associated with a decreased likelihood of initiating breastfeeding.4,5 Whilst some studies found a correlation between maternal employment and breastfeeding duration,6,7 others did not.8,9

The World Health Organization Expert Consultation on the Optimal Duration of Exclusive Breastfeeding recommended in 2001 that infants should be exclusively breastfed during the first 6 months of life, instead of the previous recommendation of 4–6 months, and that they should continue to receive breast milk throughout the remainder of the first year and during the second year of life.10,11

While it is often perceived that a significant proportion of mothers in Singapore stop breastfeeding by the end of the 2 months maternity leave in preparation to return to work, no study has been conducted locally to verify if returning to work following 2 months of maternity leave had any effect on breastfeeding duration between working and non-working mothers.

The aim of this study was to examine the impact of working status on initiation and duration of breastfeeding.

Methods

Study design

The National Breastfeeding Survey was conducted by the Health Promotion Board, Singapore in 2001. The study involved 2149 Singaporean female residents (citizens and permanent residents) and is the first comprehensive study on breastfeeding attitudes and practices conducted at the national level.

Prior to designing the questionnaire for the survey, a qualitative study involving a focus group discussion was conducted among mothers to ascertain the factors perceived to be important in determining breastfeeding duration. The National Breastfeeding Survey questionnaire was then designed and the survey was conducted in two phases. During Phase 1, mothers who agreed to participate in the survey were contacted by phone 2 months after delivery for the first interview. Mothers who were breastfeeding at 2 months after delivery were invited to participate in Phase 2 of the survey whereby a second interview was conducted by phone at 6 months after delivery. Nurses were specially trained to conduct these telephone interviews in order to minimize any interviewer bias. Care was taken not to influence or prompt the mother's answers. Mothers were free to respond to the questions and were not presented with a structured list of common answers. Information was obtained regarding maternal socioeconomic characteristics (age, ethnic group, educational level, working status, household income), biomedical factors (number of children, type of delivery, sex and birth weight of infant, previous breastfeeding experience), knowledge and attitudes (benefits of breastfeeding, motivating factors for breastfeeding, reasons for stopping breastfeeding). The duration for each telephone interview was approximately 20 minutes and was conducted either in English or an appropriate mother tongue.

Variables that were known or suspected to be associated with breastfeeding initiation and duration, including sociodemographic, biomedical and motivating factors were collected in Phase 1 of the survey. Other variables collected included reasons for stopping before 2 months, current feeding practices and motivating factors for breastfeeding. Phase 2 of the survey was designed to follow up on information obtained in Phase 1, in particular for the current feeding practices and reasons for stopping between 2 and 6 months.

Incidence and duration of breastfeeding described by working and non-working mothers were compared. Sociodemographic and biomedical variables, including ethnic group, maternal age, educational level, previous breastfeeding experience, working status, number of children, monthly household income and motivating factors, were also evaluated in relation to the duration of breastfeeding.

Eligibility criteria for study population

All women who delivered in the eight restructured and private hospitals in Singapore with obstetric services during the recruitment period from 1 April 2001 to 31 May 2001 were given invitation letters by the ward nurses to participate in the Survey. Potential respondents were not told that this was a survey specifically on breastfeeding, but rather it was one on infant feeding and nutrition so as to reduce the possibility of bias. A total of 6724 women were recorded by the Singapore Registry of Births and Deaths to have delivered during the period of the study, of which 2212 women agreed to participate in the Survey. The 62 non-residents and one respondent whose nationality was not recorded were excluded from this analysis, thus giving us a study population of n = 2149.

Data analysis

Data from the survey was analysed using Statistical Package for Social Sciences (SPSS) version 11.0.12 χ2 test was used to compare differences in proportions among the comparison groups. For categorical variables such as educational level that was ranked on an ordinal scale, χ2 trend test was applied.

The Kaplan–Meier method for event-time data was used in the analysis of the status and duration of breastfeeding. The duration of breastfeeding was calculated as the difference in time between the start and end of breastfeeding. This was censored at the termination of breastfeeding or at 6 months, the conclusion of the study, if breastfeeding continued beyond this period. A comparison of the breastfeeding survival curves was made between working and non-working mothers using the log-rank test statistic.

The Cox proportional hazards model was used to determine factors that were significantly associated with the duration of breastfeeding. Potential confounders identified at the univariate analyses included ethnic group, age, educational level, working status, number of children and household income. The hazard ratio (HR) for breastfeeding and its associated 95% confidence interval (CI) was calculated for the sociodemographic and motivating factors associated with breastfeeding.

Previous breastfeeding experience could not be considered in the analysis due to the large number of first-time mothers, who were coded as ‘not applicable’. We also could not include 44 subjects whose household income was coded as ‘confidential’. The level of significance was set at 5% unless otherwise stated.

During Phase 1 of the survey, which was conducted at the end of 2 months, the breastfeeding duration of 1 out of 2032 mothers was not recorded. She was given a duration of 1 month for analysis purposes. Similarly, during Phase 2 of the survey, conducted at the end of 6 months, 55 out of 1074 mothers were lost to follow-up. As these mothers had breastfed for at least 2 months duration, they were given a duration of 2 months for analysis purposes.

The term ‘breastfeeding’ refers to all types of breastfeeding according to the definition used by the World Health Organisation.10,11,13 In this study, as long as the mother had given her breast milk to her child, she would be considered to have initiated breastfeeding.

Results

Demographic and social characteristics of respondents

The sociodemographic characteristics of the respondents are shown in table 1. The median age of the mothers was 31 years 4 months. Approximately 44% of mothers were first-time mothers and the majority (99%) gave birth to one child. Close to two-thirds of our study population were working mothers.

View this table:
Table 1

Description of profile of respondents

Characteristics of subjectsAll respondents (n = 2149)
SOCIODEMOGRAPHIC FACTORS
    Mother's age (years)
        Median31.25
        Range13.4–47.1
    Ethnic group
        Chinese1363 (63.4)
        Malay577 (26.8)
        Indian158 (7.4)
        Other51 (2.4)
    Working status
        Working1387 (64.5)
        Not working762 (35.5)
    Educational level
        Primary and below364 (16.9)
        Secondary (GCE ‘N’/‘O’ Level)789 (36.7)
        Post-secondary (GCE ‘A’ Level/ Diploma)479 (22.3)
        Tertiary (Degree)517 (24.1)
    Monthly household income
        Confidential44 (2.0)
        <S$2000356 (16.6)
        S$2000–S$3999701 (32.6)
        S$4000–S$5999487 (22.7)
        S$6000–S$7999216 (10.0)
        S$8000–S$9999150 (7.0)
        ≥S$10,000195 (9.1)
BIOMEDICAL FACTORS
    Number of children (including child/children from this pregnancy)
        1945 (44.0)
        2716 (33.3)
        3336 (15.6)
        ≥4152 (7.1)
    Type of delivery
        Normal vaginal delivery1424 (66.3)
        Assisted vaginal delivery146 (6.8)
        Caesarean delivery579 (26.9)
    Sex of infant
        Male1154 (53.2)
        Female1017 (46.8)
    Singleton2128 (99.0)
    Birth Weight (g)
        Mean3091
        SD478
    Initiated breastfeeding for this pregnancy
        Initiated breastfeeding2032 (94.6)
        Did not initiate breastfeeding117 (5.4)
    Breastfed previous childrena
        Yes1012 (47.1)
        No182 (8.5)
        Not applicable954 (44.4)
PSYCHOSOCIAL FACTORSb
    Intention to breastfeed more than 2 months1217 (59.9)
    Intention to breastfeed more than 6 months833 (41.0)
    As long as possible/As long as there is milk/As long as baby wants/No planning427 (21.0)
    Husband encourages breastfeeding220 (11.2)c
  • Figures in parenthesis denote percentages

  • a: Previous breastfeeding experience was not recorded for one respondent (n = 1)

  • b: Figures includes only those who initiated breastfeeding (n = 2032)

  • c: Figure excludes mothers whose exact duration of breastfeeding were not known (n = 56) and those whose findings of whether their husband encouraged breastfeeding were not recorded (n = 3)

Relating working status with initiation and duration of breastfeeding

Working status had no effect on the initiation of breastfeeding (table 2), but it had a significant effect on breastfeeding duration (table 2 and figure 1).

Figure 1

Cumulative probability of continuing to breastfeed by postpartum working status (Singapore National Breastfeeding Survey)

View this table:
Table 2

Percentage of mothers who breastfeed by working status

Not working (n = 762)Working (n = 1387)P value
Initiated breastfeeding
    Yes711 (93.3)1321 (95.2)0.73
    No51 (6.7)66 (4.8)
Breastfed for >2 months
    Yes326 (44.9)557 (40.7)0.07
    No400 (55.1)810 (59.3)
Breastfed for ≥6 months
    Yes200 (27.5)257 (18.8)0.001
    No526 (72.5)1110 (81.2)
  • Figures in parenthesis denote percentages

A higher proportion of non-working than working mothers breastfed for more than 2 months. However, the univariate analysis suggested that working status did not have a significant effect on the duration of breastfeeding for more than 2 months (table 2).

As mother's educational level was a potential confounder on the effect of working status on breastfeeding status, stratification was done for both groups according to this factor (table 3). It was observed that for non-working mothers, a higher proportion of mothers with higher educational levels (post-secondary and tertiary) were breastfeeding for more than 2 months as compared to mothers with lower educational level (secondary and below). For working mothers, a higher proportion of those with a tertiary education were breastfeeding for more than 2 months.

View this table:
Table 3

Duration of breastfeeding by working status and educational level

Breastfeeding forP value
≤2 months (n = 1210)>2 months (n = 883)
Non-working mothers
    Primary and below158 (71.2)64 (28.8)<0.001
    Secondary168 (57.7)123 (42.3)
    Post-secondary57 (44.2)72 (55.8)
    Tertiary17 (20.2)67 (79.8)
Working mothers
    Primary and below106 (84.1)20 (15.9)<0.001
    Secondary333 (69.7)145 (30.3)
    Post-secondary210 (61.0)134 (39.0)
    Tertiary161 (38.4)258 (61.6)
  • Figures in parenthesis denote percentages

For the post-secondary group, a greater proportion of those who were not working managed to breastfeed for longer than 2 months. On the other hand, a greater proportion of those who were working breastfed for less than or equal to 2 months.

Although working status was not found to have a significant effect on duration of breastfeeding beyond 2 months duration on univariate analysis, a significant difference in the duration of breastfeeding between working and non-working mothers can be seen after stratification by educational levels (table 3) and from their respective survival curves.

Figure 1 displays the cumulative probability of breastfeeding for mothers who initiated breastfeeding and continued to breastfeed during the first 6 months after delivery by working status. It shows that the proportion of mothers who breastfed dropped steeply during the first 4 weeks following delivery. Thereafter, the decrease was more gradual with the proportion of working mothers who breastfed falling below the proportion of non-working mothers who breastfed at around 6 weeks postpartum. This suggests that mothers who returned to work began weaning approximately 2 weeks prior to starting work.

A further drop in cumulative probability of breastfeeding among working mothers took place at 8 weeks postpartum just prior to the return of mothers to work. The median breastfeeding duration was 9 weeks for non-working mothers and 8 weeks for working mothers. The difference in proportion of mothers in the two groups who were breastfeeding became more marked over time, especially from the 8 week point onwards. About 31% (95% CI 27.7–34.8%) of non-working mothers breastfed for up to 6 months as compared to 20% (95% CI 17.8–22.5%) of working mothers. The log rank test showed that the difference was significant (HR = 1.27, 95% CI 1.14–1.41, P value <0.001).

This was further affirmed when we carried out a univariate analysis to assess the effect of working status on the breastfeeding duration of 6 months or longer (table 2). It showed a higher tendency for working mothers to stop breastfeeding after 6 months as compared with non-working mothers.

Factors associated with breastfeeding duration

Potential confounders identified by the Cox proportional hazards model to influence breastfeeding duration are shown in table 4. Young mothers (age <19 years old) tended to breastfeed for 2 months or less. Less educated mothers were more likely than their more highly educated counterparts to breastfeed for 2 months or less. Mothers from households with a monthly income of less than S$4000 were also significantly more likely to breastfeed for a shorter duration.

View this table:
Table 4

Associated factors for breastfeeding duration of ≤2 months by sociodemographic variables and motivating factors

VariableHazard ratio95% CIP value
SOCIODEMOGRAPHIC AND BIOMEDICAL FACTORS
    Mother's ethnic group0.001
        Chinese1.00
        Malay0.630.55 – 0.720.001
        Indian0.730.59 – 0.900.003
        Other0.700.48 – 1.010.057
    Age of mother0.001
        ≤19 years1.661.05–2.630.031
        20–29 years1.160.89 – 1.510.288
        30–39 years1.010.78 – 1.300.945
        40–49 years1.00
    Mother's working status
        Working1.611.43–1.850.001
        Not working (housewife/homemaker)1.00
    Mother's educational level0.001
        Primary and below2.872.30–3.570.001
        Secondary2.241.87–2.680.001
        Post–secondary1.731.46–2.050.001
        Tertiary1.00
    Number of children, including child from this delivery0.023
        1–21.00
        3–40.820.71–0.940.005
        5 or more0.860.57–1.300.461
    Monthly household income0.018
        <S$20001.601.11–2.140.002
        S$2000–S$39991.451.14–1.850.003
        S$4000–S$59991.220.97–1.540.090
        S$6000–S$79991.321.03–1.700.027
        S$8000–S$99991.120.85–1.460.432
        ≥S$10,0001.00
MOTIVATING FACTORS
    Breastfeeding is convenient
        No1.591.30–1.950.001
        Yes1.00
    Baby prefers breast milk to formula milk
        No1.351.00–1.820.049
        Yes1.00
    Breastfeeding fosters better bonding with the baby
        No1.201.01–1.360.006
        Yes1.00
    Encouragement from husband to breastfeed
        No1.201.01–1.420.039
        Yes1.00

Working mothers were 1.6 times more likely than non-working mothers to breastfeed for 2 months or less (HR = 1.61, 95% CI 1.43–1.85, P value = 0.001), after adjusting for potential confounders using Cox proportional hazards model.

Significant motivating factors include convenience of breastfeeding, better bonding through breastfeeding, husband's encouragement to breastfeed and baby's preference for breast milk.

Reasons for stopping breastfeeding between 2 and 6 months

The important reasons for working mothers stopping breastfeeding between 2 and 6 months were due to work-related factors (48.4%), insufficient breast milk (27.0%) and baby preferring formula milk (12.5%). In contrast, insufficient breast milk (43.1%), baby preferring formula milk (16.4%) and lack of help with care of baby, other children or household chores (12.9%) were the main reasons for non-working mothers stopping breastfeeding between 2 and 6 months. Important work-related reasons included the need to return to work, facilities at the workplace being not conducive for breastfeeding and demands of work interfering with breastfeeding. It must also be noted that factors unrelated to employment, such as insufficient breast milk, baby preferring formula milk also influenced the duration of breastfeeding following a mother's return to work.

Discussion

The National Breastfeeding Survey 2001 is the first comprehensive study on breastfeeding attitudes and practices conducted at the national level. Previous studies were of a much smaller scale and were usually conducted on a hospital institutional basis.

Of the local studies that examined the effect of working status on breastfeeding, Chen's6 study was the most comprehensive. However, she only compared the differences in breastfeeding prevalence at 1 month postpartum between working and non-working mothers. As there was no previous study on whether returning to work following the 56 days maternity leave had any effect on breastfeeding duration between working and non-working mothers, research into this area was warranted.

Our study showed that duration of breastfeeding was related to the mother's working status. The median breastfeeding duration was 9 weeks for non-working mothers and 8 weeks for working mothers. Approximately 31% of non-working mothers breastfed for up to 6 months as compared to 20% of working mothers. This is in agreement with Chen's6 study that working mothers breastfed for shorter duration than non-working mothers. In contrast, the studies of Chao8 and Chung9 did not show any correlation between maternal employment and breastfeeding duration.

Breastfeeding can still continue even with the resumption of employment. Approximately 41% of working mothers continued to breastfeed after the end of their maternity leave and 20% of them managed to breastfeed for up to 6 months. However, non-working mothers were shown to be more likely to breastfeed their infants for a longer period of time than working mothers.

These results do not necessarily mean that returning to work leads to a shorter duration of breastfeeding. Additional factors such as weaning in preparation to return to work, maternal fatigue, difficulty in juggling the demands of work and breastfeeding may also contribute to the shorter duration of breastfeeding after returning to work. Work-related reasons are the main reasons for mothers stopping breastfeeding between 2 months and 6 months postpartum. There is a need to make work places more friendly towards breastfeeding and expressing of breast milk.

Cox regression could be employed for identifying factors that influenced duration of breastfeeding. It is a powerful tool for analysis of time-to-event data such as the duration of breastfeeding as it effectively uses data available at all time points. Good follow-up was achieved in Phase 2 of the survey at 6 months after delivery. Out of 1074 mothers who continued breastfeeding beyond 2 months, only 55 were not contactable.

The limitations of this study are as follows:

1. Possible responder biases. All mothers who delivered during the 2 months recruitment period were invited to participate in a study on infant feeding and nutrition. The response rate was 32.9%. The reasons for refusal to participate were not known, but might include mothers who did not place importance on infant feeding and nutrition. In order to determine if our sample was representative of the women who delivered in the year 2001, we compared the sociodemographic factors (median age, ethnic group, educational level, working status) of mothers for live-births in the year 2001 using data obtained from the Singapore Registry of Births and Deaths14 and that of our sample (table 5). While the age profile of the survey respondents is similar to the total population of mothers, a slight difference was observed in terms of ethnic group and educational level. Mothers with primary education and below were under-represented while Malay mothers and working mothers tended to be over-represented in our study. There is however no reason to assume selection bias.

View this table:
Table 5

Live-births by sociodemographic variables of mother

VariableTotal populationStudy population
Age of mother (years)
    Median30.731.25
    Range(15.0–49.0)(13.4–47.1)
No of live-births in
year 2001 n (%)the Survey n (%)
Mother's ethnic group
    Chinese27 034 (65.2)1378 (63.5)
    Malay8537 (20.6)584 (26.9)
    Indian3404 (8.2)159 (7.3)
    Other2476 (6.0)51 (2.3)
Mother's educational level
    Unknown23 (0.1)
    Primary and below9109 (22.0)369 (17.0)
    Secondary13 636 (32.9)795 (36.6)
    Post-secondary9720 (23.4)485 (22.3)
    Tertiary8940 (21.6)523 (24.1)
Mother's working status
    Working23 822 (57.5)1401 (64.5)
    Not working17 629 (42.5)771 (35.5)
Total41 451 (100.0)2172 (100.0)
  • Figures in parenthesis denote percentages

2. Possible social desirability bias. As this was an interviewer-administered questionnaire, some mothers might have felt compelled to provide the ‘right’ or ‘socially acceptable’ answer. Efforts were made to minimise this by allowing mothers to answer freely and interviewers not prompting them for any answers.

3. Working status was not subdivided into part-time and full-time work. The 56 days of maternity leave was also assumed to start following delivery although women can opt to start their maternity leave 14 days prior to the expected date of delivery. Not all mothers may have taken maternity leave for 2 months following delivery. However, the number of mothers who took their maternity leave prior to delivery or shortened their maternity leave to return back to work earlier is expected to be very small as most mothers would want to spend time with their newborn.

Recommendations

Work-related reasons were the main reasons for mothers stopping breastfeeding between 2 and 6 months postpartum. There is a need to make work places more friendly towards breastfeeding and expressing of breast milk.

The aim of breast milk expression in the workplace is to achieve highest possible prolactin levels, most efficient emptying of the breasts and greatest milk volume possible. Achieving these goals will help mothers maintain optimal milk supply after they return to work and experience longer periods of separation from their child.15

Work-related obstacles to breastfeeding and expression of breast milk among employed mothers are as follows:15,16

  • Lack of on-site child care arrangements and childcare facilities that support breastfeeding.

  • Inability to return to work with a flexible schedule.

  • Lack of information regarding benefits of breastfeeding on part of employers.

  • Passive role of company doctors (doctors appointed by the mother's employer for medical care) in providing advice on breastfeeding.

Time, space and support were identified as essential elements for a ‘mother-friendly work environment’ by the World Alliance for Breastfeeding Action. More highly educated women, who are more likely to be professionals, are more likely to be successful in obtaining these elements in their workplace as they have more influence and control over their working environment.4 In addition, time for breastfeeding can be provided through flexible and reduced hours, nursing breaks or sufficient break time to express breast milk when mothers return to work. The Alliance also recommends providing workplace childcare facilities as a way to ensure space for breastfeeding by allowing mothers to work with their infants in close proximity. A breastfeeding mother should be provided with a place in the work environment where she can express breast milk in comfort and privacy if she has to be separated from her infant.

Areas for improvement in order to increase the prevalence of breastfeeding women among working women include providing a more conducive work environment, facilities that support breastfeeding (e.g. lactation programmes, private room on site for breastfeeding or expression of breast milk, equipment to express and store breast milk), having more flexible working conditions such as part-time or flexi-time work, working from home and job sharing, providing more information regarding the benefits of breastfeeding to the mother, child, company doctor and the company itself and through legislation.

As part of Singapore's efforts to boost the country's birth rate, the duration of maternity leave is being reviewed with paid maternity leave being extended for up to 6 months. If the duration of maternity leave were to be extended to 6 months, it would enable more working mothers to exclusively breastfeed their child for the first 6 months of life.

Conclusion

Breastfeeding and employment are not incompatible. Although working does not prevent a mother's initiation of breastfeeding, the results show that it limits the duration of breastfeeding. The short duration of maternity leave, separation of mother and infant, inadequate facilities at workplaces for breastfeeding or expression of breast milk and inflexible hours may contribute to the reduced duration of breastfeeding by working mothers. Working conditions for female employees who are breastfeeding should be improved so that they may be encouraged to breastfeed and work at the same time.

Returning to work is associated with a shorter duration of breastfeeding. Mothers who have better control over their work environment, e.g. higher educated in management positions (predisposing factors) are the most likely to breastfeed after returning to work and have a longer duration of breastfeeding. Similarly, mothers provided with enabling factors that facilitate expression of breast milk would also be motivated to breastfeed following their return to the workplace. Doctors and, more importantly, nurses and midwives must also play their part, not only in educating mothers on the benefits of breastfeeding and teaching them the necessary skills to do so (enabling factors), but also in periodically reminding them of the importance of breastfeeding in the early stages following delivery (reinforcing factors).

Key points

  • Our study used data from the national breastfeeding survey which examined factors, including working status, associated with breastfeeding in Singapore.

  • Working status had no effect on the initiation of breastfeeding, but had an effect on breastfeeding duration.

  • About 31% of non-working mothers breastfed for up to 6 months as compared to 20% of working mothers.

  • Work-related factors was the the main reason for mothers in Singapore stopping breastfeeding between 2 and 6 months postpartum.

  • There is a need to make work places in Singapore more friendly towards breastfeeding and expressing of breast milk.

Acknowledgments

We thank Dr Lam Sian Lian, CEO of Health Promotion Board, for allowing us to use data from the Health Promotion Board's National Breastfeeding Survey 2001 and Professor Lee Hin Peng of the Department of Community, Occupational and Family Medicine, National University of Singapore for the guidance given during the preparation of this paper.

References

View Abstract