Pregnancy and the postpartum period is a time of increased vulnerability for retention of excess body fat in women. Breastfeeding (BF) has been shown to have many health benefits for both mother and baby; however, its role in postpartum weight management is unclear. Our aim was to systematically review and critically appraise the literature published to date in relation to the impact of BF on postpartum weight change, weight retention and maternal body composition. Electronic literature searches were carried out using MEDLINE, EMBASE, PubMed, Web of Science, BIOSIS, CINAHL and British Nursing Index. The search covered publications up to 12 June 2012 and included observational studies (prospective and retrospective) carried out in BF mothers (either exclusively or as a subgroup), who were 2 years postpartum and with a body mass index (BMI) >18.5 kg m–2, with an outcome measure of change in weight (including weight retention) and/or body composition. Thirty-seven prospective studies and eight retrospective studies were identified that met the selection criteria; studies were stratified according to study design and outcome measure. Overall, studies were heterogeneous, particularly in relation to sample size, measurement time points and in the classification of BF and postpartum weight change. The majority of studies reported little or no association between BF and weight change (n=27, 63%) or change in body composition (n=16, 89%), although this seemed to depend on the measurement time points and BF intensity. However, of the five studies that were considered to be of high methodological quality, four studies demonstrated a positive association between BF and weight change. This systematic review highlights the difficulties of examining the association between BF and weight management in observational research. Although the available evidence challenges the widely held belief that BF promotes weight loss, more robust studies are needed to reliably assess the impact of BF on postpartum weight management.
Obesity is a major public health problem throughout Europe, with increasing prevalence in women of childbearing age. The Health Survey for England1 reported 44% of UK women aged 25–34 years and 58% of women aged 35–44 years as overweight or obese. Pregnancy and the postpartum period is a time of increased vulnerability to weight gain and body composition changes in women. Although most women have a desire to return to their pre-pregnancy weight following childbirth,2 very few achieve this goal.3, 4, 5 Average estimates of postpartum weight retention range from 0.5 to 3.0 kg;6 however, 14–20% of women are 5 kg heavier at 6–18 months postpartum than they were before becoming pregnant.7 Excessive postpartum weight retention is also recognized as a strong contributing factor to the future development of overweight and obesity6, 8, 9 and is indirectly associated with obesity-related illnesses, such as type 2 diabetes and cardiovascular disease.10 Furthermore, the excess weight gained in one pregnancy can have a cumulative effect on weight gain in subsequent pregnancies,6 thus amplifying the trajectory of weight gain and risk of obesity in a woman’s lifetime.
Little is known about the trajectory of postpartum weight change, particularly in relation to breastfeeding (BF). Although BF is associated with health benefits for both mother and baby,11, 12 its role in postpartum weight management remains unclear. Theoretically, BF should promote weight loss due to the increased energy cost of lactation.13 However, given the many factors that influence postpartum weight change, such as socio-economic status,14 ethnicity,15, 16 pre-pregnancy weight,16, 17, 18, 19 parity,7 gestational weight gain (GWG)7, 20 and lifestyle,16, 21, 22 the evidence regarding the specific role of BF warrants careful review. Indeed, a recent report by the National Institute for Health and Clinical Excellence23 highlighted the lack of evidence regarding the role of BF in weight management, with a key research need highlighted to determine whether BF can help women to achieve a healthy weight.
With obesity currently regarded a public health problem post pregnancy, a clearer understanding of the role of BF in postpartum weight management is required. The primary objective of this systematic review was to critically appraise the literature published to date regarding the relationship between BF and postpartum weight change. The secondary objective was to examine the association between BF and changes in measures of body composition.
Electronic literature searches were conducted using MEDLINE and EMBASE, using the Ovid Interface, PubMed, Web of Science, BIOSIS, CINAHL and British Nursing Index. The search terms were: ‘postpartum or postpartum period’, ‘post-partum’, ‘puerperium’, ‘postpartal’, ‘post natal’, ‘postnatal’, ‘post pregnancy’, ‘postpregnancy’, combined with ‘weight’, ‘body mass index’, ‘body weight’, ‘weight change’, ‘weight loss’, ‘BMI’, ‘body composition’, ‘weight gain’, ‘overweight’ and ‘weight management’, which were then combined with ‘mother’ or ‘women’. Papers containing the terms ‘BF’ and ‘lactation’ were then identified by manual selection from this broader search. The search, which was conducted on 12 June 2012, was restricted to English language and humans. An example of one of the search strategies is provided in Supplementary Appendix 1.
The literature search formed the basis for two systematic reviews: one on observational studies (as described here) and the second on randomized controlled trials (reported elsewhere).
For the current review, the following selection criteria were used: Inclusion criteria were studies conducted in BF mothers (either exclusively or as a subgroup); women 2 years postpartum; and observational (prospective/retrospective) studies.
Exclusion criteria were studies conducted exclusively in non-BF women; no non-BF comparison or control group; did not examine infant feeding method in relation to outcome; did not examine postpartum weight change, weight retention or body composition; conducted in developing countries; included underweight women (BMI <18.5 kg m−2) only; cross-sectional and intervention studies; examined alternative behavioral therapies, for example, hypnotherapy; non-English language papers; conducted exclusively in women receiving specific weight loss medical or drug treatment; conducted exclusively in women with pre-existing clinical conditions, for example, type 1, type 2 or gestational diabetes mellitus.
The search results generated from each database were imported into the RefWorks bibliographic management program to systematically sort the papers. One reviewer (CEN) screened the titles and abstracts of the articles to establish whether studies were suitable for inclusion. Bone measurements were not included as a measure of body composition. Full-text versions of the remaining papers were then screened to identify relevant studies. Any uncertainties regarding specific studies were resolved through discussion with a second reviewer (JVW). Bibliographies of retrieved articles were also reviewed for additional relevant citations.
Eligible papers were categorized into two subgroups according to the primary and secondary objectives of the review: observational studies examining the association between (i) BF and maternal weight change; and (ii) BF and maternal body composition.
Owing to the heterogeneous nature of the studies, a narrative approach was used to appraise the studies. Information extracted from each study included the following: authors; year of publication; country; study design; infant feeding categories; method used to assess BF status; sample size; age of participants; time of measurements/duration of follow-up; whether weight or body composition was measured as a primary or secondary outcome; whether or not BF intensity or duration was examined; method used to measure weight or body composition; weight change and/or body composition results; statistical methods used in analyses and adjustment for covariates.
The study selection process is summarized in Figure 1. The combined database searches identified 18 306 potential studies. Following the initial screening of titles, 646 papers were retained. After reading the abstracts and/or full papers, a further 595 papers were rejected, leaving a total of 51 full papers. Three additional papers were included (two recently published papers and one paper obtained from a bibliography of a retrieved article), which resulted in a total of 54 papers being retained. This included 45 observational studies and 9 intervention studies. The current review focuses on the observational studies only.
Of the 45 observational studies identified, 37 were prospective studies17, 18, 21, 22, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56 and 8 were retrospective studies.15, 16, 19, 57, 58, 59, 60, 61 Tables 1 and 2 summarize each study in relation to weight and body composition outcomes, respectively. Further details regarding the location of each study, the age of participants, specific study findings, statistical methods and covariates included in the analyses are provided in Supplementary Appendix 2.
Of the studies included, 27 (19 prospective, 8 retrospective) examined the association between BF and weight change,15, 16, 19, 21, 22, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 57, 58, 59, 60, 61 16 (all prospective) examined both weight change and body composition17, 18, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54 and 2 (both prospective) examined only the body composition.55, 56 Studies were mainly conducted in the United States (n=32), with other studies conducted in the United Kingdom (n=2), Denmark (n=2), Japan (n=1), Taiwan (n=1), Hong Kong (n=1), The Netherlands (n=1), Canada (n=2), Sweden (n=1) and Australia (n=2). Sample size varied widely between studies, ranging from 1141 to 36 030 women.24 In the 43 studies that examined weight change, 30 carried out serial weight measurements,17, 18, 19, 21, 24, 25, 26, 27, 28, 30, 31, 32, 33, 35, 36, 37, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 53, 54, 58 with measurement time points ranging from 8 days to 2 years postpartum: 10 studies only examined weight change between 0 and 3 months postpartum,18, 22, 29, 38, 40, 44, 47, 50, 51, 59 12 studies examined weight change until 3–6 months postpartum,15, 17, 36, 37, 41, 42, 43, 48, 57, 58, 60, 61 16 studies examined weight change until 6–12 months postpartum16, 19, 21, 25, 26, 27, 30, 31, 32, 33, 34, 46, 49, 52, 53, 54 and 5 studies continued measurements beyond 12 months.24, 28, 35, 39, 45 Loss to follow-up rates varied substantially among the studies (ranging from 0%27, 29, 44, 49 to 86.8%46), although for many studies loss to follow-up rates were not clearly reported. Reasons for losses to follow-up were only reported in three studies and included pregnancy, moving house, returning to work, infant death, effort in participating and lack of time.31, 46, 47 Weight change was a primary outcome measure in 26 studies. Body weight was either measured objectively16, 17, 18, 19, 22, 28, 29, 31, 32, 33, 34, 35, 36, 37, 38, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 58, 59 by self-report15, 21, 24, 26, 30, 39, 52, 53, 57, 60, 61 or by a combination of both objective and self-report methods depending on the time of measurement.25 Objective methods included clinical/hospital records, scales, beam balance, bioelectrical impedance and total body fat analyser. Two studies did not report the method used to measure weight.51, 54 Only one study examined the influence of BF on weight change according to the pre-pregnancy BMI category.24 Method of infant feeding was assessed using recall methods in all of the retrospective studies and in 13 (35%) of the prospective studies.21, 24, 25, 26, 28, 30, 31, 33, 34, 35, 39, 51, 52
Relationship between BF and postpartum weight change in prospective studies
Of the 35 prospective studies that examined weight change, 21 reported no significant relationship between BF and weight change.18, 21, 22, 27, 29, 30, 35, 36, 37, 39, 40, 42, 43, 44, 46, 47, 49, 50, 51, 53, 54 The remaining 14 of the 35 studies17, 24, 25, 26, 28, 31, 32, 33, 34, 38, 41, 45, 48, 52 reported a direct relationship between BF and weight change. The associations observed in some studies depended on when the measurements were carried out,17, 28, 34, 45, 48 BF duration,31, 32, 52 BF exclusivity31 and pre-pregnancy BMI.24 Adjustment for potential confounding factors also varied between studies: within the 14 studies that reported a direct association between BF and weight change, 6 studies adjusted for pre-pregnancy weight or BMI,17, 24, 31, 33, 34, 45 7 for GWG24, 28, 31, 33, 34, 45, 48 and 5 for parity.17, 24, 31, 33, 45 A limited number of studies adjusted for energy expenditure or exercise, food intake, smoking, marital status, education, income/occupation, time of measurement and infant sex. Six of the 14 studies did not adjust for any confounding factors, rather their findings were based on bivariate analyses/correlation analyses or t-tests.25, 26, 32, 38, 41, 52 Of the 21 studies that reported no significant association between BF and weight change, only 5 adjusted their analyses for pre-pregnancy weight or BMI,30, 35, 40, 43, 49 4 adjusted for GWG27, 30, 35, 54 and 4 adjusted for parity.30, 35, 37, 50
In relation to the method used to assess postpartum weight, the majority of studies (11 out of 14),17, 25, 28, 31, 32, 33, 34, 38, 41, 45, 48 which reported a direct relationship between BF and weight change measured weight objectively rather than using self-report. Objective measurements of weight were also used in the majority of the studies (15 out of 21), which showed no relationship between BF and weight change.18, 22, 27, 35, 36, 37, 40, 42, 43, 44, 46, 47, 49, 50, 53 Almost half of the studies (10 out of 21) in which no associations between BF and weight change were observed tended to have small sample sizes (<60 participants)18, 40, 44, 47, 49, 50, 53 or short duration of follow-up (3 months).18, 22, 29, 40, 44, 47, 50, 51 In terms of overall quality, only five studies objectively measured postpartum weight change, as well as adjusting for key covariates, that is, pre-pregnancy BMI,31, 33, 34, 35, 45 GWG31, 33, 34, 35, 45 and parity.31, 33, 35, 45 All of these studies, except one,35 reported a direct association between BF and weight change. In all of these studies, weight change was examined over a period of 12 months or longer.
Prospective studies that examined weight change between months 1–3 postpartum
Of the 22 prospective studies that examined weight change in the early postpartum months (that is, months 1–3), the majority (73%) found no significant between-group differences in weight loss, or no association between BF and weight change.17, 18, 22, 28, 29, 36, 37, 40, 41, 42, 43, 44, 45, 50, 51, 53 Contrary to expectations, the study by Chou et al.44 observed that at 12 weeks postpartum, non-BF mothers had lost more weight than BF mothers; however, this study did not adjust for any confounding factors and the findings may reflect the fact that the non-BF mothers had lower energy intakes and lower pre-pregnancy weight compared with BF mothers. Brewer et al.17 examined weight loss at months 3 and 6, and reported that most weight loss occurred in the first 3 months regardless of the feeding method. Another study, which examined BMI rather than weight, found that lactation score (that is, higher score representing greater BF) was inversely correlated with BMI at 1.5 months postpartum.32 Kramer et al.,48 adjusting for GWG only, also noted greater weight loss at 1 month in BF and combined feeding (CF, that is, formula and breast) women compared with formula feeding (FF) women, with no differences evident at 3 or 6 months.
Prospective studies that examined weight change between months 3–6 postpartum
The relationship between BF and weight change at 3–6 months varied, with 13 out of 20 studies reporting no between-group differences in weight loss or no association between BF and weight change,27, 30, 34, 35, 36, 37, 39, 42, 43, 46, 48, 49, 54 and 7 out of 20 studies reporting greater weight loss in BF women and/or CF women compared with that in non-BF women.17, 24, 31, 33, 41, 45, 52 Baker et al.24 found that BF was associated with reduced weight retention, except in women with a pre-pregnancy BMI 35 kg m−2. Although Bradshaw and Pfeiffer41 observed that BF women lost weight consistently between weeks 4 and 28 postpartum, with greater total weight loss than that of FF women, this study did not adjust for any confounding factors. However, other studies which reported greater weight loss in women who breast-fed or combination-fed compared to women who formula-fed did adjust for a variety of confounding factors,17, 24, 31, 33, 45 including adjustment for energy intake and energy expenditure.17
Prospective studies that examined weight change for more than 6 months postpartum
Fifteen studies examined weight change beyond 6 months postpartum.21, 24, 25, 26, 27, 28, 30, 31, 33, 34, 35, 39, 45, 46, 49 Results were again inconsistent with eight studies, showing no association between BF and weight change.21, 27, 30, 33, 35, 39, 46, 49 In contrast, the other seven studies consistently showed a positive association between BF at 12 months or more and weight change.24, 25, 26, 28, 31, 34, 45 In one of the largest studies by Baker et al.,24 BF was independently associated with a reduction in weight retention at 18 months postpartum, but this association was only observed in women who had exclusively breast-fed for 6 months and had continued BF for 12 months.
Prospective studies that examined the association between BF duration and/or intensity and weight change
Of the 35 prospective studies, 14 studies also examined weight change in relation to both BF intensity and duration,15, 17, 24, 26, 27, 28, 31, 32, 33, 34, 35, 45, 48, 58 6 studies in relation to BF duration only25, 30, 46, 49, 52, 60 and 6 studies in relation to BF intensity only.18, 21, 22, 47, 51, 54 Three studies reported that women who breast-fed for a longer period and more intensively (that is, 5–6 months of exclusive or almost exclusive BF) lost weight more rapidly between 3 and 6 months17, 33, 45 and retained less weight by 12 months28, 39, 45 than women who formula-fed or breast-fed for a shorter duration or less intensely; although some of these studies were based on small sample sizes and/or recruited highly educated women, or women of mid-high socio-economic status, thus limiting the generalizability of the findings.17, 45 Some of these studies also constructed a scoring system in order to estimate the BF status. In a study of 1423 Swedish women, Ohlin and Rossner33 assigned a lactation score based on every full month of full or partial BF (4 points and 2 points, respectively). Women with a higher lactation score, that is, a greater BF duration and intensity, lost significantly more weight between 2.5 and 6 months postpartum than women with a lower lactation score. However, the relationship was weak and no significant between-group differences were evident in total weight loss between 2.5 and 12 months. Another study that used a similar scoring method34 found no association between the BF score and weight retention, but reported that women who breast-fed until 1 year postpartum retained less weight than non-BF women, whereas BF at all other time points (6 weeks, 6 months) was not related to weight retention. Manning-Dalton and Allen18 similarly found no correlation between percent lactation and weight change at 12 weeks postpartum, although their scoring system was based on a more complex formula that incorporated an estimate of energy intake obtained from breast milk. A more recent study by Baker et al.24 assigned 1 point per week and 0.5 points per week for full BF and partial BF, respectively, until 12 months postpartum, and then 0.5 points per week for continuation of any BF beyond 12 months, therefore, enabling the consideration of BF as a continuous variable and the examination of the total contributions of full and partial BF. The authors reported that exclusive BF for 6 months and to any extent for 12 months was independently associated with decreased weight retention at 6 months postpartum, irrespective of pre-pregnancy BMI, and for every point increase in BF they reported a decrease in weight retention of 0.01–0.04 kg.24 Weight measurements were based on self-report and fully BF was defined as exclusive and almost exclusive, that is, additional vitamins, minerals and water were allowed. Janney et al.28 constructed four BF patterns to reflect both intensity and duration, and reported that although both BF and non-BF women lost weight between 0.5 and 18 months postpartum, BF women lost weight at a faster rate and achieved their pre-pregnancy weight ∼6 months earlier than FF women. Despite their observation that BF duration was an independent predictor of weight retention, the magnitude of the association was relatively small. Rather, the pattern of weight retention observed was also strongly influenced by other factors, including GWG, age and marital status. Specifically, greater weight retention was observed in women who were older, single or had greater GWG.
One of the most recent studies by Martin et al.31 found that BF duration was more important than the type of BF, that BF duration was inversely associated with weight retention and that every additional week of any form of BF between 0 and 12 months postpartum was associated with a 0.04 kg decrease in weight retention. When they split their analysis according to the World Health Organisation62 BF categories (exclusive, predominant, complementary or non-BF), they found no association between feeding method and weight retention. This study adjusted for a wide range of confounding factors, but not for dietary intake or physical activity. Similarly, Gould Rothberg et al.25 reported that every week of continued BF resulted in an additional 1.5 lbs weight loss, although these findings were based on preliminary bivariate analyses. Generalizing results is also difficult, as participants were young and predominantly African American and Hispanic women with low socio-economic background. Another recent study by Ostbye et al.35 reported that longer duration of BF was associated with greater weight loss; however, significance was lost after adjusting for other confounding factors, including education, parity, baseline weight, marital status, GWG, junk food intake and physical activity. The study found that eating less healthy foods and being less physically active were the main contributors to postpartum weight retention. In contrast, Dugdale and Eaton Evans46 observed no association between BF duration and weight change at 12 months postpartum. Likewise, a more recent study by Lyu et al.30 reported no association between BF duration and weight retention at either 6 or 12 months. Although this study adjusted for a wide range of confounding factors, weight measurements were based on self-report, and the education level and socio-economic status of participants were narrowly distributed, which therefore makes it difficult to generalize findings to other populations. Despite the inconsistencies, the overall evidence indicates that BF duration may be important for weight change and BF may need to be continued for longer than 6 months to have an appreciable influence on weight change.
Relationship between BF and postpartum weight change in retrospective studies
Unlike the prospective studies, the majority of the retrospective studies (six out of eight studies) found no association between BF and weight change,15, 16, 19, 57, 59, 61 although in five of these studies15, 16, 57, 59, 61 weight was only examined at one time point. Although one of the largest and more recent studies by Krause et al.58 found no association between BF status and weight retention at 3 months postpartum, weight retention was lower in mothers who combination-fed or breast-fed compared with those who formula-fed at 6 months, after adjusting for race, education, pre-pregnancy BMI and GWG. Slotkin and Herbold60 also reported that BF for any duration between 0 and 6 months was associated with weight loss, with longer duration associated with greater weight loss, although the analyses were not adjusted for confounding factors. Furthermore, participants in that study were highly educated women with high income and, therefore, it is difficult to generalize findings to other populations. Finally, four of the studies15, 57, 60, 61 relied on self-reported measures of weight with BF status obtained many months after the birth of the baby.
Other factors influencing postpartum weight change
Many of the studies, particularly those that found little or no association between BF and weight change, highlighted that other factors were more important contributors to postpartum weight change. GWG was most frequently cited as being a strong contributor to weight retention.15, 22, 24, 28, 30, 31, 33, 37, 43, 57, 58 Other factors that were less consistently associated with weight retention included age,17, 28, 33 parity,16, 17, 28, 31, 36, 45, 52 pre-pregnancy weight,16, 17, 18, 19 demographic factors,15, 16, 35, 36, 58 lifestyle factors, such as physical activity and dietary intake,16, 18, 21, 30, 34, 35 smoking27, 33 and marital status.28 One study also reported lack of sleep as being a strong contributing factor to postpartum weight retention.26 There were no notable differences in the findings of studies conducted across different socio-economic groups.
Heterogeneity in the definition of BF and weight change among observational studies
There was considerable variation in the definitions of infant feeding status within both prospective and retrospective studies. In 16 of the studies, infant feeding status was broadly categorized into two groups: ‘BF’/‘lactating’ and ‘non-BF’/‘non-lactating’, with many studies defining BF as ‘any’ BF.19, 36, 37, 38, 42, 43, 44, 59, 60, 61 In two other studies,52, 57 the term ‘BF’ encompassed both full and partial BF. Another study defined lactation as ‘>8 lactation events/day with the addition of <200 ml per day of formula milk’ and non-lactating as ‘women who fed their infant >200 ml per day of formula milk’.50 Fourteen studies categorized feeding method into three groups: ‘exclusive’ or ‘fully’, or ‘predominantly BF’; ‘combined or mixed feeding’; and either ‘FF’ and/or ‘no BF’.15, 16, 17, 22, 27, 32, 33, 34, 35, 40, 41, 48, 51, 58 Within many studies, ‘exclusive or fully BF’ was also defined in a variety of ways, including exclusive BF until 6 months;26, 41 breast milk as a measurement of infant’s energy intake;28 breast milk and up to 120 ml per day of other milk;45 breast milk with the inclusion of vitamins/minerals and/or water;24 and BF with no introduction of formula, other milk or fluids.47 Only one recently published study31 used the infant feeding categories (exclusive, predominant, complementary or no BF) as recommended by the World Health Organisation.62 Six studies constructed a lactation score as a composite estimate of lactation status, although, as mentioned previously, the methods used to calculate scores varied among the studies.18, 24, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45
The definitions used to describe changes in weight varied within each of the observational studies. Seventeen out of 35 studies measured weight change as the difference between pre-pregnancy weight and the weight recorded at the time of the postpartum measurement.15, 16, 21, 22, 24, 25, 26, 28, 30, 31, 33, 37, 39, 51, 57, 58, 61 In all of these studies, pre-pregnancy weight was self-reported. Eighteen out of 35 studies measured change as the difference in weight between the early postpartum baseline measurement and follow-up measurement.17, 18, 27, 29, 35, 38, 40, 41, 43, 44, 45, 46, 47, 48, 49, 53, 54, 60 Other studies compared postpartum follow-up weight with the last recorded weight before delivery,42, 59 the weight recorded during early pregnancy52 or the weight at the first antenatal hospital visit.34, 36 The study by Potter et al.19 calculated weight loss as the weight recorded at the last prenatal visit minus the infants birth weight, minus the weight recorded at the postpartum measurement time point. Of the 17 studies, which based postpartum weight change on pre-pregnancy weight, 10 studies reported no association between BF and weight change, and 7 studies reported a positive association between BF and weight change. Of the 18 studies, which based weight change on postpartum measurements, 14 studies reported no association between BF and weight change, whereas only four studies reported a positive association between BF and weight change.
Prospective studies that examined the relationship between BF and change in body composition
Eighteen prospective studies investigated the influence of BF on body composition (Table 2). Of these, nine studies used skinfold thickness measurements to assess body composition. Eight out of the 18 studies also examined lean body mass, fat-free mass and/or fat mass using a variety of methods, including dual-energy X-ray absorptiometry, urinary creatinine, bioelectrical impedance, whole-body potassium and skinfold thickness measurements. A limited number of studies also measured percentage body fat, body density, body volume, body protein, total body water and body circumferences.
Of the 18 studies, only two reported a greater influence of BF on body composition compared with FF or CF.41, 51 Sidebottom et al.51 reported lower skinfold thicknesses in exclusively BF women compared with FF or CF women. Body composition change was only examined until 6 weeks postpartum, and only in healthy well-educated women, which therefore makes it difficult to generalize results. Bradshaw and Pfeiffer41 also reported a significant decrease in the sum of skinfolds in BF women, but this was only observed at 1 month postpartum in a small sample (n=11) of primiparous women. Both studies41, 51 failed to adjust for key confounding factors.
Thirteen of the 18 studies reported no significant influence of BF on body composition compared with other feeding methods.17, 18, 42, 43, 44, 45, 46, 47, 48, 49, 52, 53, 55 The majority of these studies were of small sample size (<60 participants).17, 18, 44, 47, 48, 49, 53, 55 Only five of these studies adjusted their analyses for pre-pregnancy weight or BMI,17, 43, 45, 47, 49 two for GWG45, 48 and two for parity.17, 45 Two studies reported greater influences of FF on body composition (change in fat mass) compared with BF,50, 54 although the study by Ota et al.50 was based on a small sample size (n=49), with body composition measurements only performed at 1 month postpartum. Finally, a more recent study by Moller et al.56 observed that fat mass decreased at a faster rate in women who breast-fed for <4 months compared with women who breast-fed for >4 months; however, there was no between-group difference observed in lean tissue mass.
In 8 of the 18 studies that examined both weight loss and body composition, the associations between BF and body composition closely reflected those observed between BF and weight loss, that is, no association.18, 42, 43, 44, 46, 47, 49, 53 Similar to the associations between BF and weight loss, the associations between body composition and BF also seemed to depend on the time of postpartum measurement,17, 41, 44, 45, 47, 48, 49, 51, 53, 54, 56 and also the duration and exclusivity of BF. In some studies, associations between BF and body composition changes also appeared to be site-specific, particularly in relation to skinfold measurements. Three studies reported either no change44, 45 or an increase17 in biceps or triceps skinfold, whereas three studies reported a decrease in suprailiac and/or subscapular skinfolds.17, 41, 51
The aim of this review was to systematically examine the evidence to date regarding the role of BF in postpartum weight change. The overall findings highlight that there is currently insufficient evidence to suggest that BF is directly associated with postpartum weight change. Although the majority (63%) of the observational studies reviewed found no significant associations between BF and postpartum weight change, or no significant differences in weight change between BF and non-BF women, it is difficult to draw any firm conclusions, as many of the associations observed depended on the time at which the postpartum measurements were carried out. Of the studies that did show a positive influence of BF on weight loss, the associations tended to be relatively weak and were often confounded by other factors, such as GWG, age and pre-pregnancy weight. Associations also appeared to be dependent on the duration and intensity of BF. Interestingly, four out of the five studies that were of higher methodological quality in terms of weight measurements and adjustment for key covariates demonstrated a positive association between BF and weight change.
In relation to BF duration, it appeared that for the majority of studies (73%), BF for <3 months had little or no influence on weight change, whereas there was some evidence to suggest that BF, if continued for >6 months, may have a positive influence on weight change, but again this was not supported by all of the studies, and in many studies the associations were only observed in women who continued BF until 12 months postpartum.
BF intensity also appeared to be a strong determining factor for weight change, with longer duration of exclusive BF, compared with CF, associated with greater weight loss, but again this was not consistent among all of the studies. Furthermore, in two of the studies that observed a positive association between exclusive BF for 6 months and weight loss, the authors concluded that the findings were not strong enough to warrant emphasis on BF as a means of minimizing postpartum weight retention28 or as a predictor of weight management.33 This review demonstrates the uncertainty regarding the optimal duration and intensity of BF in order to achieve and sustain weight loss, thus highlighting a need for more well-designed prospective studies. Furthermore, if the findings of this review are confirmed, the practicality and reality of continuing BF until 12 months is likely to be an issue for many women, particularly in the United Kingdom, where only 34% of BF mothers continue until 6 months postpartum.63
The findings regarding body composition change and BF status were similar, with the majority (89%) of studies reporting little or no association, again with time of measurement, duration and exclusivity of BF appearing to influence findings. Comparing results was also difficult due to variation in methods used to measure body composition. Site-specific differences were also evident in the body composition changes; however, this most likely reflects the changes in fat mobilization during lactation, the magnitude of which can vary between women.64
The strength of this review lies in the clarity and robustness of its methods. Broad search terms and multiple bibliographic databases were used in the searches to capture as many relevant papers as possible, and a robust systematic approach was used to select the final papers. However, as with any systematic review the results are determined by the strengths and limitations of the included studies. The study design and methods used within the studies were, for the majority, clearly defined. However, failure to draw any firm conclusions regarding the influence of BF on weight change may reflect the multifactorial influences affecting postpartum weight loss and the methodological challenges faced when examining this specific population group. As highlighted in many of the studies reviewed, other factors such as pre-pregnancy weight or GWG may have greater influence on postpartum weight change than BF. However, although these are key contributing factors it is important to acknowledge that these factors apply to both BF and non-BF women. What was noticeably absent from the studies reviewed was the lack of consideration given to contextual factors that are unique to BF women and that are likely to impact on their ability to lose weight. The individual needs of BF women are likely to be considerably different to those of non-BF women. In addition to adjusting to the needs of a newborn baby and recovering from childbirths, BF women often experience added pressures, such as problems in establishing feeding routines and infant sleeping patterns, which, in turn, may affect the mother’s quality and duration of sleep, and psychological health. Any combination of these factors is likely to impact on the extent of weight loss. Indeed, although there was some evidence to suggest that BF for at least 6 months may promote weight loss, this will not apply to all BF women. As highlighted by Lovelady,5 some women will also need to increase their exercise levels or restrict their dietary intake in order to achieve this weight loss. As increased energy needs of BF mothers may counteract the effect of BF on weight loss and the demand of BF may limit the amount of physical activity being undertaken, it is paramount that changes in dietary intakes and physical activity patterns are explored. Furthermore, it may not be enough to just consider the impact of BF duration and intensity on weight change, the frequency of feeding may be just as important, particularly as babies differ in their energy requirements and hunger needs. The spacing of pregnancies may also be a contributing factor in that closely spaced pregnancies can result in cumulative weight gain. In summary, there needs to be a greater understanding of why women do or do not lose weight when BF.
The methodological rigour of many of the studies is questionable. There was considerable heterogeneity observed in terms of sample size, duration of follow-up and methods used to measure weight, body composition and infant feeding status, thus making comparisons difficult. Failure to adjust for key covariates, including pre-pregnancy BMI, GWG and parity, was also evident in many of the studies reviewed. In addition, in many studies weight and body composition changes were not the primary outcome measures and, therefore, results should be interpreted with caution. Selection bias was evident, with many studies conducted in highly educated women or women from high socio-economic groups, making it difficult to generalize results. Inconsistencies among findings may also reflect the variation in measures of BF duration and intensity, and, more specifically, the lack of clarity in the definition of BF. Many studies broadly defined BF as ‘any’ BF. Only one recent study31 used the ‘gold’ standard World Health Organisation62 infant feeding categories to define the pattern of BF. In many studies, particularly those with small sample sizes, categorization of infant feeding status is likely to have resulted in loss of statistical power, thus influencing the findings. The term ‘exclusively’ BF encompassed an array of definitions and there was considerable variation in scoring methods used within some studies as a composite measure of BF intensity and duration.24, 28, 33 The use of arbitrary scoring systems is also problematic, as it is a surrogate measure of BF intensity that may have resulted in non-differential misclassification of BF intensity or duration. This misclassification could have impacted on the overall findings by diluting the association between BF and weight change, and is likely to have biased findings towards the null. Misreporting of infant feeding status may also have occurred, as all of the retrospective studies and more than a third of the prospective studies relied on recall methods for assessing infant feeding status. Similar inconsistencies were evident in relation to the definition of weight change. In particular, all of the studies that examined weight change as the difference between postpartum weight and pre-pregnancy weight relied on self-reported pre-pregnancy weight, and were therefore susceptible to recall bias. Misreporting of weight can lead to incorrect BMI classification and to inaccurate calculation of weight change, and therefore may not reflect the true trajectory of weight change between pregnancy and postpartum. Indeed, it has been shown that women have a tendency to underestimate their pre-pregnancy weight,33 with the extent of underestimation depending on their weight. Misreporting weight by 1–5 kg can lead to an overestimation in postpartum weight change by more than 100–200%.65 To improve the comparability among studies, there is a need for consistency in the definition of postpartum weight change, as well as greater clarity and uniformity in the definition of BF. We recommend the future adoption of the World Health Organisation62 infant feeding categories of exclusive, predominant, complementary or no BF.
There was also considerable variation in statistical analyses used within studies, and in some studies conclusions were based on observations that were not always statistically significant59 or only approaching significance.47 Many studies also combined repeated measurements into an overall measure of weight change, thus failing to examine patterns of weight loss at different time points.17, 18, 19, 33, 41, 46
Finally, within the structure of the review it is possible that relevant studies were not identified, although this is unlikely, as several bibliographic databases were searched and broad search terms were used. The narrative approach of this review also meant that a meta-analysis could not be performed; nonetheless, the review provides a valuable insight into the studies to date, and the findings should be useful in guiding the development of future studies.
This review highlights the difficulties that arise when examining the association between BF and weight management in observational research. Weight and body composition changes are highly variable in BF mothers; however, the majority of the studies reviewed here failed to convey the extent of individual variability and the reasons surrounding it. It is therefore unclear why some BF women lose weight while others do not, and this needs to be addressed in future studies. Although multiple factors influence postpartum weight, there is a need to explore factors unique to BF women. BF mothers are heterogeneous in terms of lifestyle, family structure, social support, sleep quality, psychological health, rate of recovery from birth, hormonal changes and their readiness to lose weight. These contextual factors mean that what works for one BF mother may not necessarily work for another. BF simply represents one possible component within weight management strategies, but this cannot be considered as a sole means to losing postpartum weight. Nonetheless, even if BF encourages weight loss in a third or half of the women, then that is still a positive outcome. Furthermore, in the absence of weight loss, previous research has shown that BF may confer benefits to women through mechanisms other than weight loss.66
The review findings raise questions about how to clearly communicate messages about weight management and BF to women. Although women are often told by health professionals that BF will help them lose weight, this message will not necessarily be true for many women and, therefore, health messages need to be carefully construed to reflect this. Furthermore, if the results of this review were confirmed and women need to breast feed more intensively and for a longer duration in order to achieve weight loss, there is uncertainty as to how women would respond to this challenge. Although some women may be encouraged to breast feed for a longer duration if it is going to have a positive outcome, there are others who may be more likely to give up BF or not try BF at all, particularly if the challenge seems impossible. Ultimately, however, such communication is premature in the absence of a more robust evidence base.
As pre-pregnancy weight and GWG were frequently cited as strong contributing factors to postpartum weight retention, observational studies should commence pre-conception with continued monitoring into the postpartum period, to capture the true trajectory of weight change. In addition to being a pivotal point for weight change, the prenatal and early postnatal period is an opportune time to intervene with weight management programs when women are still in regular contact with health professionals, community midwives and health visitors, and are more motivated to embrace weight management advice.
The findings undoubtedly challenge the common belief portrayed across scientific literature that BF promotes weight loss. Overall, there is insufficient evidence to suggest that BF promotes greater postpartum weight loss compared with other methods of feeding. Although BF may help some women lose weight, it cannot be generalized across all BF women. This may simply reflect the fact that there are a variety of other reasons why BF women do not lose weight. As highlighted by Lederman,67 BF should be promoted for its health benefits for both mother and child, but should not be solely relied upon as a way for women to compensate for excessive GWG or to increase postpartum weight loss. Given the multifactorial nature of postpartum weight change, it is difficult to establish what factors have the greatest influence on weight retention and what areas need to be targeted in preventative measures. Although there was some evidence that longer duration of BF may be beneficial for encouraging weight loss, this will need to be more rigorously tested. Nonetheless, it is encouraging that the studies of high methodological quality observed a positive association between BF and weight change. However, there is now an urgent need for more robust studies to reliably assess the direct impact of BF on postpartum weight management and to explore further the reasons why not all BF women lose weight.
We acknowledge Ohaka Ugomma Chijioke who assisted with the literature searches and data extraction. This work was supported by a grant from the Department of Employment and Learning Northern Ireland.
CEN conducted the literature searches, extracted the data and wrote the first draft of the manuscript. JVW assisted in study selection and data extraction. All of the authors contributed to the writing, editing and reading of the manuscript. CEN had final responsibility for the final content of the paper.
About this article
Supplementary Information accompanies this paper on International Journal of Obesity website (http://www.nature.com/ijo)
Journal of Human Lactation (2019)