Bulletin of the World Health Organization

Breastfeeding policy: a globally comparative analysis

Jody Heymann a, Amy Raub a & Alison Earle b

a. Fielding School of Public Health, University of California, 650 Charles E Young Dr S, Los Angeles, CA, 90095, United States of America.
b. Brandeis University, Waltham, United States of America.

Correspondence to Jody Heymann (e-mail: jody.heymann@ph.ucla.edu).

(Submitted: 27 June 2012 – Revised version received: 12 September 2012 – Accepted: 28 January 2013 – Published online: 18 April 2013.)

Bulletin of the World Health Organization 2013;91:398-406. doi: http://dx.doi.org/10.2471/BLT.12.109363

Introduction

Breastfeeding lowers the risk of diarrhoeal disease by four- to 14-fold1 and the risk of respiratory illness by fivefold.2 Although the absolute benefits are greater in settings of poverty, poor nutrition and poor hygiene, where baseline disease rates are higher, the relative risk of these illnesses is significantly reduced by breastfeeding in high-income settings as well.35 Breast milk also markedly improves nutritional status in infants. Since malnutrition contributes to half of all infant deaths,6 breastfeeding helps to reduce infant mortality. Studies around the world in affluent and poor nations alike have shown a 1.5- to five-fold decrease in mortality among breastfed infants. 710 Breastfeeding has also been associated with lower rates of chronic diseases such as diabetes 11,12 and inflammatory bowel disease13,14 and with improved neurocognitive development. 1518

Breastfeeding is beneficial to the health of both women and infants. Women who breastfeed have longer intervals between births and, as a result, a lower risk of maternal morbidity and mortality, as well as lower rates of breast cancer rates before menopause and potentially lower risks of ovarian cancer, osteoporosis and coronary heart disease.3,19 As a result, the World Health Organization (WHO) recommends exclusive breastfeeding for at least 6 months.20 Nonetheless, the rates of breastfeeding vary substantially around the world; the rate of exclusive breastfeeding among infants under 6 months of age ranges from 1% to 89%.21 One of the most common reasons that women stop breastfeeding is that they need to return to work.2225 According to World Bank figures on the female share of the labour force, between 1960 and 2009 this share increased from 32% to 46% in the United States of America, from 25% to 47% in Canada, and from 21% to 41% in Latin America and the Caribbean. In sub-Saharan Africa, East Asia and the Pacific, as well as in Europe and Central Asia, women already made up at least 40% of the labour force in 1960 and this proportion remained the same or increased over subsequent decades. In the Middle East and North Africa females comprised a smaller share of the labour force, but this share increased from 21% to 25% between 1960 and 2009.26,27

Working does not necessarily have to lead to lower rates of breastfeeding.28 The quantity and nutritional quality of breast milk are not undermined by maternal work or activity, including vigorous exercise, and there is no indication that working women are less interested in breastfeeding than non-working women.28,29 Rather, it is the difficulty of continuing to breastfeed under the conditions experienced when they return to work that women most often cite as the reasons for supplementary feeding or for weaning infants.22 A woman’s ability to breastfeed is markedly reduced when she returns to work if breastfeeding breaks are not available, if quality infant care near her workplace is inaccessible or unaffordable, and if no facilities are available for pumping or storing milk.30

Given that circumstances play a major role in whether women breastfeed after returning to work, it makes sense to ask whether providing breastfeeding breaks from work might not increase the number of women who breastfeed for the recommended 6 months. Legislation guaranteeing breastfeeding breaks could substantially improve working mothers’ ability to continue to breastfeed. However, it might not make a substantial difference if the legislation covers a small fraction of the labour force, if breaks are too short for women to be able to pump milk or breastfeed, if infants are far from workplaces and locations for storing pumped milk are not available, or if legislation is not enforced.

This study analyses how many countries guarantee mothers breastfeeding breaks, how long the daily breaks are, and how often the breaks are guaranteed for enough months so that women can breastfeed for the minimum 6 months recommended by WHO for breastfeeding. We then conduct the first analysis of how labour policies affect breastfeeding rates around the world. Understanding the relationships between national policy and breastfeeding rates is critically important because of the substantial health benefits of breastfeeding to infant and mother, in addition to the facts that the majority of pregnant women are now employed and that a sizable proportion of women prefer to breastfeed.31,32

Methods

In this study, we examined the number of countries that guarantee mothers breastfeeding breaks, the time allowed daily for breaks and the months of breaks granted. We conducted multivariate regression analyses to test the association between national policy on breastfeeding breaks and national rates of exclusive breastfeeding among women with children less than 6 months of age. We controlled for other national-level factors that could also influence the likelihood of a woman in a given country initiating and sustaining breastfeeding for a minimum of 6 months, as recommended by WHO.

Data sources

To obtain current, detailed information on national policies on breastfeeding breaks in the workplace, we collected and analysed information on breastfeeding policy in all Member States of the United Nations (UN). Our primary data source was original national labour legislation. Our research team reviewed all national labour legislation collected by the International Labour Organization (ILO) and available in their NATLEX database in original languages and in translations. While this comprehensive review of primary source legislation provided us with the most reliable information available on legal rights to breastfeeding breaks, the legislation has some limitations. The ILO repository relies on countries to provide updated legislation. Additionally, we did not capture any subnational variation in breastfeeding break policies or policies that were regulated at an industry level or for subsets of the labour force through collective bargaining. To supplement the data available through the ILO and ensure that we were using the most up-to-date information available, we reviewed legislation on line through country web sites, the World Bank’s “Doing Business” law library, and the Lexadin World Law Guide legislation database, as well as in hard copy through Harvard University, McGill University and the ILO headquarters library.33,34 Finally, to search for any additional information we also investigated secondary sources, such as reports on national policies, information compiled by global organizations, including the World Alliance for Breastfeeding Action, and the maternity protection component of the International Labour Organization’s Database of Conditions of Work and Employment Laws. These secondary sources were selected after a comprehensive computerized search of the academic and grey literature for comparative databases.35,36 Based on all of these sources, we developed a database with information on breastfeeding break policies for 182 of the 193 countries that were Member States of the United Nations at the time the database was last updated, in March 2012.

Nations vary in how they structure their laws on breaks for breastfeeding at the workplace. We had therefore gathered information on whether countries guarantee women the ability to take breastfeeding breaks at work, whether these breaks are paid, how much time is allowed daily for breaks, and for how many months after giving birth women are entitled to take the breaks. National policies also vary in terms of how the breaks can be scheduled. For example, some policies indicate the total time allotted for breaks without restriction on how the breaks are taken, whereas others specify whether each break has to be taken at once or can be split into multiple, shorter breaks. In some countries, legislation specifies the number of minutes of break allowed for a certain number of hours worked (e.g. 30 minutes over a four-hour period). In others, legislation stipulates that breaks can be accumulated and taken at the start or the end of the day, which shortens the number of hours per day that a woman works.

Analyses conducted

We conducted descriptive analyses of our data on breastfeeding breaks in the workplace to provide a global picture of working women’s ability to breastfeed their infants. Most commonly, the law specifies for how long women can take breastfeeding breaks in terms of the age of the child, but sometimes it does so in terms of the time transpired after a woman returns to work, in which case we added this to the total maternal leave available to new mothers since this is guaranteed time off. In this article, when breastfeeding breaks vary based on the employee’s tenure, hours worked or the number of young children, we report the minimum granted.

We used multivariate regression to test for an association between the presence of national legislation guaranteeing breastfeeding breaks at the workplace and breastfeeding rates. Ordinary least squares regression analyses were conducted. We conducted a stepwise regression analysis in which we first examined the effect of breastfeeding break policies, controlling only for national resources and level of urbanization (measures found in other research studies to be associated with breastfeeding rates).37,38 This model is referred to as the more “parsimonious” one. The sample for this model was the 116 countries with data on all three independent variables and the outcome variable – the rate of exclusive breastfeeding in infants aged less than 6 months. Female literacy rate, a proxy for degree of access to information, was then added. This model is referred to as the “full” model. The sample for the “full” model was the 108 countries for which data on all four independent variables and the outcome variable were available. All analyses included controls for the year of breastfeeding rate data available, since data for every country were not available for the same point in time and substantial efforts have been made internationally to promote exclusive breastfeeding rates as part of the Millennium Development Goals. Policies allowing women to take breaks from work to breastfeed affect employed women directly; they may also affect women outside the labour force indirectly through changing norms. To the extent that the laws have a direct effect, one would expect a greater impact where a greater number of women are in the paid labour force. To test this, we examined whether the policies on breastfeeding breaks in the workplace are associated with higher exclusive breastfeeding rates in countries with a higher female share of the labour force than in countries where females represent a smaller share. This was done by analysing the extent to which the breastfeeding outcomes are explained by the interaction between the guarantee of breastfeeding breaks and the national female share of the labour force. Again, first we estimated a “parsimonious” model in which we used national resources and level of urbanization as controls. For this model the sample size was the 115 countries with data on all included variables. We then estimated a “full” model that also included the female literacy rate as a control. For this model the sample size was 108 countries.

Variables

Outcome examined

The outcome variable in this study was the percentage of children aged less than 6 months who were exclusively breastfed in each country. The data were obtained from WHO’s Global Data Bank on Infant and Young Child Feeding,39 which contains internationally comparable data derived primarily from Demographic and Health Surveys, Multiple Cluster Indicator Surveys and Ministries of Health. Because of the nature of some of these sources, data on breastfeeding rates were more frequently available for lower-income countries. These rates were provided for 129 nations for at least one year between 2000 and 2011. When multiple years of data on breastfeeding rates were available, we used the most recent rate.

Key independent variables

Paid breastfeeding breaks

For every country for which we had labour policy information, we constructed a binary indicator variable and set it to 1 if national legislation required employers to offer paid breastfeeding breaks until the child was at least 6 months old, and to 0 in the absence of such legislation. For every nation included in our analysis, we verified that the legislation had been in place for at least one year before the year for which we obtained data on exclusive breastfeeding rates.

Per capita gross domestic product

From the World Bank’s World Development Indicators Online we obtained the gross domestic product (GDP) of each country, measured in purchasing-power-parity-adjusted constant 2005 international dollars, and used it as an indicator of national economic resource level.40 For analysis, we used GDP from the year for which exclusive breastfeeding rates were available. A log transformation of per capita GDP was used instead of a linear term to allow for the common finding that changes in income at the lower end of the income spectrum have a larger impact on breastfeeding rates than changes in wealth at the higher end of the income spectrum.

Female share of the labour force

The female share of the labour force is the percentage of females among members of the labour force. A rate of 50% would indicate that females and males make up an equal share of the labour force. We preferred this measure to the female labour force participation rate – i.e. the percentage of females active in the labour force – because it obviates the need to adjust for differences in cross-country employment rates. Our data were taken from the World Bank’s World Development Indicators Online.40 For analysis, we used female share of the labour force from the most recent year for which exclusive breastfeeding rates were available.

Female literacy rate

The female literacy rate was defined as the percentage of literate females among all females aged 15 years or older. We obtained the data from the World Bank’s World Development Indicators Online.40 We used the female literacy rate as a proxy for access to information (since information is disseminated in writing as well as orally). The influence of literacy on breastfeeding depends on the nature of the information made available to women. If substantial public health information on breastfeeding is provided in written form, literate women will have greater access to information on the benefits of breastfeeding. If advertising for formula is conducted in written form, then higher literacy rates may be associated with greater exposure to marketing and lower rates of breastfeeding. Countries’ female literacy rates are not available for every year. For analysis, we used the female literacy rate available for the most recent year for which data on exclusive breastfeeding rates for each country were available, or for the preceding year. For higher-income countries for which data were unavailable, we used a female literacy rate of 99%, consistent with the female literacy rate assumed by the United Nations Development Programme for such countries in constructing the Human Development Index.41

Urban percentage of the population

We included the percentage of the population living in an urban area as a control in the models because other researchers have found the level of urbanization to be negatively associated with breastfeeding rates.42,43 We extracted data from the World Bank’s World Development Indicators Online.40 For analysis, we used the urbanization rate for the year for which exclusive breastfeeding rates were available. Table 1 summarizes the variables included in the analyses.

Results

Of the 182 nations that had data for 2012 on the existence of a national policy for breastfeeding breaks in the workplace, 45 countries (25%) have no policy in place. A policy guaranteeing paid breastfeeding breaks is in place in 130 countries (71%) and seven countries (4%) have policies guaranteeing unpaid breaks. There is significant variation in the length of time for which working mothers have access to breastfeeding breaks (Table 2 and Fig. 1). Most nations’ policies specify how long employers must guarantee breastfeeding breaks in terms of the age of the child, rather than according to how long the employee has been back at work. Only three countries (Bhutan, San Marino and Swaziland) provide for breastfeeding breaks for less than 6 months after an infant’s birth. Working mothers are guaranteed a break most commonly until the infant is one year old. This is true of 41 countries. In 32 countries, the legislation does not specify the child’s age.

Fig. 1. World map showing period of time after giving birth during which national policy guarantees breaks from work for breastfeeding women, 2012
Fig. 1. World map showing period of time after giving birth during which national policy guarantees breaks from work for breastfeeding women, 2012
Source: World Policy Analysis Centre, adult labour database.

The time allowed daily for breastfeeding breaks was specified in the policies of 111 of the 137 countries that had a policy. Most countries’ policies provide for a daily total of one hour during work for breastfeeding breaks. In 30 countries, breastfeeding breaks can be accumulated to shorten the work day. In 22 countries, the legislation specifies over how many hours the break can be spread; for example, if 30 minutes are allotted in total, the legislation might specify that the employee can take a single 30-minute break or two 15-minute breaks as long as the time is taken within a 3- or 4-hour window.

In multivariate models, national policies guaranteeing paid breastfeeding breaks at least until the child turns 6 months old were associated with significantly higher rates of exclusive breastfeeding. This was observed even when controlling for countries’ GDP per capita, percentage of the population living in an urban area, female literacy rate and year of exclusive breastfeeding data. The guarantee of paid breastfeeding breaks until the infant was at least 6 months of age was associated with an increase of 8.86 percentage points in the rate of exclusive breastfeeding of infants under 6 months of age (P < 0.05) in the full model (Model 2 in Table 3).

As expected, our findings show that national policies guaranteeing breastfeeding breaks in the workplace are associated with a higher increase in exclusive breastfeeding in countries where the share of females in the labour force is higher (Table 4).44,45 The coefficient for the interaction of our policy indicator with the female share of the labour force is positive and statistically significant at the P < 0.05 level in the more parsimonious model as well as in the full model. Controlling for access to information did not affect the size or significance of the estimated coefficient. Based on the full model estimates, at the average female share of the labour force for the countries in the sample on which this model was run, which was 41%, a national policy guaranteeing breastfeeding breaks was associated with an increase of 7.7 percentage points in the rate of exclusive breastfeeding of infants less than 6 months of age. Fig. 2 shows the predicted rates of exclusive breastfeeding during the first 6 months of life for countries with and without a national policy guaranteeing breastfeeding breaks, by income group, with predictions based on the average value of each independent variable in the model.

Fig. 2. Predicted exclusive breastfeeding rates for typical countries, by income and breastfeeding break policy
Fig. 2. Predicted exclusive breastfeeding rates for typical countries, by income and breastfeeding break policy
a For children less than 6 months of age.

Discussion

Although most countries guarantee breastfeeding breaks to working women, at least 45 of them had still failed to do so as of 2012. We found that the existence of a national policy guaranteeing breastfeeding breaks until an infant is at least 6 months old was associated with significantly higher rates of exclusive breastfeeding. This was true even after controlling for national GDP, female literacy rate and percentage of the population living in urban areas.

Because this study is cross-sectional, it can demonstrate association but not causation. The true effect size could differ from the estimated one if pre-existing breastfeeding rates influence the likelihood of countries passing legislation guaranteeing breastfeeding breaks. Longitudinal studies are needed to address this question. Nonetheless, the presence of clear mechanisms by which guaranteed breastfeeding breaks would contribute to the higher rates of exclusive breastfeeding suggest that the relationship is causal. If these results continue to be borne out, countries that do not yet have legislation that guarantees breaks for breastfeeding in the workplace may benefit greatly by passing such legislation. Any workplace that can provide a break for working adults to have a meal should be able to provide a break for working mothers to breastfeed. Our previous research has also demonstrated that countries can be economically competitive while providing these breaks.46 Yet without breastfeeding breaks, women who return to work less than 6 months after giving birth may not be able to breastfeed for as long as recommended to protect their infants’ health and their own. Policies that guarantee working women breastfeeding breaks for at least 6 months after giving birth increase the probability of exclusive breastfeeding for the recommended period.

Policies that provide for breastfeeding breaks at the workplace are in line with international agreements that protect the rights of children and women. The Convention on the Rights of the Child has been ratified by 190 countries. Article 24 of the Convention stipulates that “States Parties recognize the right of the child to the enjoyment of the highest attainable standard of health” and that “States Parties shall pursue full implementation of this right and, in particular, shall take appropriate measures: To diminish infant and child mortality.” The Convention goes on to refer specifically to the importance of breastfeeding. International conventions focusing on women also call for legislation seeking to ensure equity at work for mothers. The Convention on the Elimination of All Forms of Discrimination against Women (CEDAW) requires that “States Parties shall take all appropriate measures to eliminate discrimination against women in the field of employment in order to ensure, on a basis of equality of men and women, the same rights.” CEDAW goes on to specifically call on all signatory countries “to prevent discrimination against women on the grounds of marriage or maternity and to ensure their effective right to work.”

The longer breastfeeding lasts, the greater its nutritional benefits and the greater the protection it confers against diarrhoeal disease. Moreover, a dose–response effect has been observed between breastfeeding duration and neurocognitive outcomes in children.11,12 Future research should examine the impact of policies guaranteeing breaks from work for breastfeeding children beyond the first 6 months of life and subsequent health outcomes for women and infants.

Little is known about the impact of legislation on breastfeeding breaks for women in the informal economy. Women who work independently, such as those who sell goods in marketplaces, may be able to bring their infants to work with them to feed. However, many women in the informal economy work for an employer. Some countries are starting to cover jobs previously considered part of the informal economy with labour legislation, to the extent possible. Extending policies on breastfeeding breaks to the informal sector should be readily feasible, since small employers can provide these breaks. Moreover, even in cases in which legislation does not apply to workers in the informal economy, it can help set norms and guidelines for all employers. Because many of the world's poorest women work in the informal economy, future studies should examine the extent to which these women are covered by breastfeeding policies in different countries.


Acknowledgements

The authors are immensely grateful to Denise Maines for research and staff assistance. We are indebted to Kip Brown and Danielle Foley, who helped build this database for the World Policy Analysis Centre at the Institute for Health and Social Policy at McGill University.

Funding:

This research would not have been possible without support for building the global policy database from the Ford Foundation and the Canada Foundation for Innovation.

Competing interests:

None declared.

References

Share