Factors associated with duration of breastfeeding in women giving birth for the first time

The longer a woman breastfeeds her infant, the greater the benefits for both her own and her child’s health. Breastfed infants have decreased risk of infections, including gastrointestinal diseases, sepsis, wheezing respiratory tract infections, necrotizing enterocolitis, meningitis, and urinary tract infections [1‐4]. Duration of being breastfed also is positively associated with intelligence later in life [5], as well as decreased prevalence of hypertension, hyperlipidemia, cardiovascular disease, diabetes and non-alcoholic fatty liver disease [6‐9]. Indeed, excess deaths of women due to myocardial infarction, breast cancer, and diabetes have been attributed to lack of or a shorter duration of breastfeeding [10].

So powerful are the overall health benefits of breastfeeding that the World Health Organization’s fifth Global Nutrition Target for 2025 is to increase the rate of exclusive breastfeeding in the first 6 months to at least 50% of all mothers [11]. As of 2017, however, only 25.6% of US women were exclusively breastfeeding at 6 months, with only 58.3% doing any breastfeeding at 6 months [12]. Previous successful breastfeeding is a factor strongly associated with duration of breastfeeding for a subsequent child [13, 14]. However, for women giving birth for the first time who have not had the opportunity to breastfeed previously, it is important to determine other factors associated with breastfeeding duration to inform potentially effective interventions that could be used to improve that rate. Prior reviews found that not smoking, having a vaginal delivery, high maternal educational attainment, and specific breastfeeding education were associated with higher rates of breastfeeding continuation [13]. These factors might inform multilevel strategies to optimize conditions so that all individuals who wanted to breastfeed would be able to breastfeed for as long as they desired. Additionally, there is a gap in the literature as most previous work on breastfeeding duration did not include validated psychosocial measures as covariates. As much of the previous work is not generalizable to US populations, and given demographic trends in the United States, contemporary understanding of these factors in a large representative diverse cohort is needed.

Therefore, the objective of this study was to examine maternal (both demographic and psychosocial) and pregnancy factors associated with breastfeeding for at least 6 months in a cohort of women giving birth for the first time.

For women giving birth for the first time in a large and racially and ethnically diverse U.S. cohort, 89.8% breastfed their infant, with 58.2% breastfeeding for at least 6 months. Being older, having a lower BMI, not smoking in the month prior to delivery, and being at 100–200% of the federal poverty level (compared to those at < 100%) were associated with higher odds of exclusive breastfeeding for at least 6 months. While rates of breastfeeding and breastfeeding for at least 6 months differed across some racial and ethnic groups, the groups did not have significantly different odds of exclusive breastfeeding for at least 6 months.

Our findings are consistent with other reports noting that age and weight are associated with breastfeeding duration [26, 27]. We also found that women who smoked proximate to delivery were less likely to breastfeed. This finding was also noted among individuals in a Spanish birth cohort, in whom smoking was associated with a more than two-fold higher rate of formula feeding and shorter breastfeeding duration [28]. Previous studies found that other social markers (such as lower education attainment and not attending prenatal classes), as well as physiologic factors such as delayed onset of lactation, inadequate milk production, nipple pain, latching problems and lack of social support are associated with early discontinuation of breastfeeding [29, 30]. Attending prenatal classes and a previous successful breastfeeding experience have been associated with longer duration of breastfeeding [11, 30].

We found that Hispanic women were more likely to initiate breastfeeding, but that these women were less likely to breastfeed for at least 6 months. This has also been noted in a few other studies [31‐33]. Level of acculturation and whether women speak primarily English or Spanish may impact breastfeeding duration in Hispanic women [33, 34]. Spanish-speaking women may represent a group who could benefit from culturally-sensitive, family-level interventions and work place policies to increase breastfeeding duration [33, 35]. While non-Hispanic Black women giving birth for the first time in the cohort had lower odds of initiating breastfeeding, those who did breastfeed had similar odds of exclusively breastfeeding for at least 6 months. Thus, improved initial uptake of breastfeeding for non-Hispanic Black mothers is important.

Multiple reports have addressed some aspects of the relationship of maternal psychosocial measures and mental health on breastfeeding duration. As in our cohort, other smaller cohorts noted that higher anxiety scores are independently associated with shorter breastfeeding duration [36‐39]. Social support was also positively associated with breastfeeding duration [27]. Other studies that found a negative association of depression and breastfeeding highlighted the importance of a pre-pregnancy diagnosis of depression [40]. In the current cohort, a history of mental health disorders was not significantly associated with breastfeeding duration. Only a history of bipolar disorder was associated with lower rates of breastfeeding, but with so few women endorsing that history (n = 79), these findings should be interpreted with caution as it was not included in the regression model.

Breastfeeding benefits are numerous for both the mother and infant. Benefits to women who breastfeed include increased caloric expenditure in the postpartum period and lower risks for cardiovascular disease, type 2 diabetes, and breast cancer [2, 6, 9]. Breastfed infants have decreased long-term risks of childhood cancers and many other positive health benefits [41‐43]. Formula costs approximately $1200–$1500 USD per year per infant which places extra financial burden on parents and community resources [44]. These individual medical and nonmedical benefits translate into significant population-level benefits given the nearly 4 million births that occur yearly in the US [10, 45].

Investigators and health agencies have called for identifying modifiable factors that can be targeted to help improve breastfeeding duration. The need for improvements is underscored by the findings in the current study: the rate of exclusive breastfeeding for at least 6 months (24.7%) was well below the 2025 WHO goal of 50% [11]. The findings in our population are slightly lower than other published rates in the US [12]. This may be because our population only included women giving birth for the first time. Educational campaigns and community-level support have been proposed to help improve breastfeeding rates, particularly among underserved populations [11, 46]. Identifying the women most at risk for not initiating any breastfeeding may be an important approach to help target intervention strategies as well. A comprehensive approach of support includes not only education and engagement, but also policies that foster and protect a woman’s right, choice, and opportunity to breastfeed [11, 47, 48].

Our study was limited in that we did not ask women for a specific date or number of weeks that they breastfed. The first interval contact was no earlier than 6 months after delivery, allowing for women who had breastfed for at least 6 months to note that outcome. As we asked the women to answer the question after breastfeeding may have stopped, it is possible that their responses may have been affected by recall or social-desirability bias. Yet, it has been shown that most women have accurate recall of duration of breastfeeding their first baby [49, 50], and accordingly we do not believe this type of bias is likely to influence our results systematically. We also did not ask participants when they returned to work, which may influence breastfeeding longevity [51, 52]. We also did not address other reasons for discontinuation, such as perception of milk supply, in the current study. We were unable to collect information on contraception or other medications being used in the postpartum timeframe. As a multicenter, diverse cohort, our findings are generalizable to similar populations of individuals giving birth for the first time.

Designing multilevel public health strategies and policies to help optimize prepregnancy health and education regarding breastfeeding, along with support in the early postpartum period, may help improve the rate of exclusive breastfeeding for at least 6 months in order to reach the 2025 Global Nutrition Target of at least 50% of mothers. These programs may include smoking cessation programs, more frequent contact with lactation specialists in the early postpartum period, community-based peer support for women in at-risk racial groups or with high anxiety, and effective workplace policies to support breastfeeding for women returning to work [53‐56]. Supporting women who begin breastfeeding but who may stop before 6 months could be a particular area of focus, particularly for Hispanic women as seen in this cohort. These strategies and interventions, which help to ensure women who want to breastfeed are able to continue, should be tested in prospective cohorts as the benefits of breastfeeding extend to the infant, the mother, and society.

In conclusion, in this cohort of women giving birth for the first time, exclusive breastfeeding for at least 6 months was associated with older age, lower BMI, and not smoking in the month before delivery.