The Relationship of Previous Breastfeeding Experiences and Factors Affecting Breastfeeding Rates: A Follow-Up Study

Abstract

Background/Objective:

Many studies have focused on the effects of previous breastfeeding experience (PBE) on subsequent breastfeeding, but few have explored their specific relationships. To explain the relationship between PBE and subsequent breastfeeding behavior based on a follow-up study.

Materials and Methods:

After delivery, 394 participants who had no PBE completed a demographic questionnaire, breastfeeding knowledge questionnaire, the breastfeeding self-efficacy short-form scale (BSES-SF), and the Iowa infant feeding attitudes scale (IIFAS). Multiparas with PBE also completed the maternal breastfeeding evaluation scale (MBFES) in addition to the aforementioned four questionnaires. On the 42nd day after delivery, participants completed the breastfeeding experience scale (BES) through social networking platforms (QQ, WeChat: both are Chinese social medias). At 4 and 6 months postpartum, researchers contacted participants by phone or a social network regarding their exclusive and partial breastfeeding experiences.

Results:

In this study, exclusive breastfeeding rates were 58.6% and 30.5% at 4 and 6 months. PBE affected breastfeeding attitudes (p < 0.05), self-efficacy (p < 0.01), and difficulties (p < 0.05). Breastfeeding knowledge, attitude, self-efficacy, and difficulties were relevant to exclusive and partial breastfeeding at 4 and 6 months (p < 0.05). Multivariate logistic regression analysis showed that compared with women without PBE, the probability of exclusive breastfeeding of multiparas with PBE at 4 and 6 months increased by 275% and 369%, respectively.

Conclusions:

The rate of breastfeeding remains low among Chinese women, but PBE is associated with a higher probability of breastfeeding at 4 and 6 months postpartum. Multiparas, especially those having PBE were more likely to breastfeed for an extended period based on their knowledge, attitude, self-efficacy, and ability to manage difficulties.

Introduction

Breast milk is the best food for infants. Among the several benefits associated with breastfeeding, it can reduce the risk of infectious illness and being overweight and obese in infants and children.¹ Breastfeeding also has been reported to reduce the risk of breast and ovarian cancer in women's later life.²

The World Health Organization (WHO)³ suggested that breastfeeding should begin within 1 hour after birth and continue exclusively for 6 months without additional water, other liquids, or solids, and then continue to 2 years of age. WHO also sets the goal of achieving an exclusive breastfeeding rate of 50% for all infants at 6 months of age by 2025. However, at present, only about 45.0% of infants worldwide are breastfed within the first hour after birth (United Nations International Children's Fund, UNICEF, 2018).⁴ In 2011–2017 ∼ 40% of infants between 0 and 6 months of age were exclusively breastfed (UNICEF, 2018).⁴

In China, the State Council (2011) declared that the prevalence of infants' exclusive breastfeeding at the age of 6 months should be increased to >50% by 2020. However, in recent years, no national rates of exclusive breastfeeding have been reported. The latest data indicated that only 20.8% of infants were breastfed exclusively at 6 months in 2013.⁵ Breastfeeding rates in diverse regions and at various periods differed greatly. Setting aside the effects of regional customs and habits, it appears that other factors having major impact on breastfeeding status should be investigated.

The association between parity and breastfeeding status was inconsistent in many studies. Compared with primiparas, multiparas with two or three children were 0.68 times more likely to breastfeed, whereas multiparas with four or more children were 2.05 times more likely to breastfeed.⁶ In the study by Imsiragic et al.,⁷ multiparas with two children were 2.12 times more likely to breastfeed compared with primiparas; however, mothers with three or more children had the same likelihood to exclusively breastfeed to 6–8 weeks compared with primiparas. Saffari et al.⁸ reported that primiparas are 1.35 times more likely to maintain breastfeeding.

The inconsistency in research findings made these researchers aware that previous breastfeeding experience (PBE) did affect the rate of breastfeeding a subsequent child. There were multiparas with no PBE who were less likely to initiate breastfeeding and to breastfeed for a shorter period.⁹ Until now, PBE has not been adequately studied. Existing literature suggested that women with PBE were more likely to initiate breastfeeding and have longer breastfeeding duration compared with women without PBE, no matter whether they were primiparas or multiparas.¹⁰ Mortazavi et al. stated that mothers without PBE were 1.6 times more likely to introduce water-based fluids and then terminate exclusive breastfeeding than those women with PBE.¹¹

However, if mothers had a negative PBE, there was no significant difference with breastfeeding duration between mothers with or without PBE.¹² If mothers encountered difficulty during prior breastfeeding such as suck or latch problems, the odds of initiating breastfeeding in the subsequent child would be lower (odds ratio 0.15, 95% confidence interval [CI] 0.04–0.56) than those mothers who did not have such experience, because mothers with negative PBE may anticipate the same experience in breastfeeding a second child.⁸

The impact of PBE on subsequent breastfeeding outcomes has not been fully explored. Current evidence indicates that PBE is the most important factor in predicting and influencing mothers' intentions to breastfeed their subsequent child.¹³ On the one hand, Palmer¹⁴ found that prior negative breastfeeding history counteracted maternal subsequent breastfeeding intention. On the other hand, the positive PBE could help enhance confidence and self-efficacy to breastfeed a consequent child.¹⁵ Palmer's study showed that women who had breastfed a previous child for any length of time had significantly higher motivation toward breastfeeding than women who had not breastfed before.¹⁴ However, Gianni et al. reported that if mothers had shorter breastfeeding duration in the past, they may have less confidence and lower self-efficacy toward breastfeeding.¹⁶

With the increasing number of multiparas in China after the enactment of the universal two-child policy, the status of breastfeeding based on women's PBE necessitates further investigation. Therefore, this study explains the specific relationship between PBE and subsequent breastfeeding behaviors to offer new insight for future breastfeeding promotion.

Methods

Study setting

Participants were recruited after their delivery of an infant while hospitalized in the obstetric units of three large university-affiliated hospitals in a major city in central China.

Study design and participants

A prospective cohort study was conducted from December 2017 to December 2018. Purposive sampling was used, followed by telephone contact or through social media (QQ, WeChat, both are Chinese social medias) at 42 days, 4 and 6 months postpartum.

Inclusion criteria were (1) minimum age of 20 years, (2) term delivery, (3) single birth, (4) can read and understand Mandarin, and (5) able to access social networking platforms (QQ, WeChat). Exclusion criteria were (1) serious complications at delivery (hemorrhage, eclampsia) and (2) concurrent chronic illness (hypertension, diabetes, HIV, liver diseases, thyroid diseases, infection, leukemia or other blood disorders, and mental disorders). In addition, if the mother had an infant of low birth weight (<2.5 kg), low Apgar score (six scores or below), requiring care in the neonatal intensive care unit or having health complications that influenced feeding (systemic illness or anomaly), these were also considered as exclusion.

Sample

The sample size was calculated using PASS by considering the following assumptions: P₁ (breastfeeding duration of 3–6 months for women in the PBE group) = 0.414, P₂ (breastfeeding duration of 3–6 months for women in the non-PBE group) = 0.586, sample allocation ratio is 2, 20% attrition rate, and two-side α = 0.05, β = 0.10. The sample size needed was calculated to be 394.

Instruments

Demographic questionnaire

This questionnaire was designed by the researchers based on a review of the literature. Data obtained included age, education, household income, occupation, prepregnancy body mass index, parity, breastfeeding experience, delivery mode, breastfeeding initiation time, and newborn gender.

Breastfeeding Knowledge Questionnaire

This questionnaire includes 18 items and was designed by Ouyang et al.¹⁷ It is a closed-ended questionnaire with true or false responses and is used to assess breastfeeding knowledge. The internal reliability with Cronbach's α was 0.93.¹⁷ In this study, a true response is scored as 1 and a false response is scored as 0 with total scores ranging from 0 to 18.

Breastfeeding Experience Scale

The breastfeeding experience scale (BES) was designed by Wambach.¹⁸ This scale contains two parts and only the first part was used in this study. It consists of 18 items and is used to measure the degree of common breastfeeding difficulties encountered in the early postpartum period. A 5-point Likert scale is used to evaluate the severity of breastfeeding difficulties. A response of “not at all” is scored as 1 and “unbearable” is scored as 5. The total score ranges from 18 to 90, with a higher score representing increased problem severity. The Cronbach's α was 0.76 during the early development of the BES and 0.77 at 6 weeks after modification.¹⁹ In the study by Mortazavi et al.,¹¹ Cronbach's alpha coefficient for the BES was 0.83. The Cronbach's α of this scale at 42 days postpartum was 0.829 in this study.

Maternal Breastfeeding Evaluation Scale (MBFES)

This scale was developed by Leff et al.²⁰ and is used to evaluate the quality of the maternal breastfeeding experience. It consists of 30 items that are scored by a 5-point Likert scale, with 1 indicating strong disagreement and 5 indicating strong agreement. Of these 30 items, 19 items are positive statements related to breastfeeding and are graded accordingly, whereas 11 items are negative statements and are graded reversely.

Breastfeeding Self-Efficacy Short-Form Scale

The Breastfeeding Self-Efficacy Scale was designed by Dennis and Faux in 1999 and revised in 2003.²¹ It measures a mother's view of self-efficacy with breastfeeding, that is, her belief in her ability to breastfeed. This study used the short form scale, which was translated into Chinese by Ip et al.,²² and the Cronbach's α was 0.94.

Iowa Infant Feeding Attitudes Scale

This scale is used to assess women's attitudes toward breastfeeding and was designed by De la Mora and Russell.²³ It uses a 5-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). There are 17 items, 9 items are positive statements related to breastfeeding and are graded accordingly, whereas 8 items are negative statements and are graded by subtracting from 6. Total scores range from 17 to 85. The higher the score, the more positive the mother's attitude is toward breastfeeding. The Cronbach's α ranged from 0.68 to 0.86. The Chinese version has been translated by Chen et al. and has been considered reliable and valid.²⁴

Data collection

Data collection was divided into four periods. The first period occurred at the time of delivery/hospitalization and participants completed questionnaires before discharge. The purpose of the questionnaire was explained and informed consent was obtained along with phone numbers and/or social media accounts. Primiparas and multiparas without PBE completed a demographic questionnaire, Breastfeeding Knowledge Questionnaire, the breastfeeding self-efficacy short-form scale (BSES-SF) and the Iowa infant feeding attitudes scale (IIFAS). Multiparas with PBE also completed the maternal breastfeeding evaluation scale (MBFES) in addition to the aforementioned four questionnaires. On average, 15 minutes are required for primiparas to complete the questionnaires and 20 minutes for multiparas.

The second data collection period occurred at 42 days postpartum. At that time, all participants were contacted by social media (QQ, WeChat) and completed the BES. The third and fourth data collections occurred at 4 and 6 months postpartum. Participants were asked about exclusive or any breastfeeding activity.

All questionnaires were uploaded to software called WenJuanXing. Once participants were added as “friends” on this software account, they received a link to the questionnaires. Participants were asked to identify themselves using abbreviations of their name or their social account nicknames when completing a questionnaire. All questionnaires collection was supervised on the spot by the principal researcher. Each questionnaire was then coded by the researchers before data analysis. Completed questionnaires were stored on the software account, which was password protected and only the researchers had access to this account.

Breastfeeding duration

“Breastfeeding duration” was measured based on participants' self-reported answers to the question “How do you feed your baby since the last follow up? Answers were: (1) breast milk only; (2) formula only; and (3) breast milk and formula; and (4) other.”

Previous breastfeeding experience

“Previous breastfeeding experience” was measured based on participants' self-reported responses to the question “Have you ever breastfed your previous child (children)? Answers were: (1) yes; or (2) no.” “How long did you breastfeed your previous child?” followed the first question if the answer was clearly “yes.” Data on satisfaction with PBE were also measured using the MBFES.

Data analysis

Data were analyzed using SPSS version 19.0. Descriptive statistics were used to describe participants' characteristics and the questionnaire variables. Categorical variables are presented with frequency distributions, whereas continuous variables are presented using means and standard deviations. t Test and analysis of variance (ANOVA) were then performed to test the association between PBE and breastfeeding self-efficacy, attitude, knowledge, and difficulties. Bivariate associations between breastfeeding status at 4 and 6 months and characteristics of participants were evaluated using Pearson's chi-square test. Bivariate associations between breastfeeding status at 4 and 6 months and breastfeeding self-efficacy, attitude, knowledge, difficulties, and satisfaction were analyzed using t test and ANOVA. Multivariate logistic regression model was conducted to test the determinants of breastfeeding status at 4 and 6 months. Adjusted odds ratios and 95% CIs were calculated. p-Value was considered significant if it is <0.05.

Ethical considerations

This study was approved by the Research Ethics Committee of the School of Health Science, Wuhan University and the university-affiliated hospitals. All participants were fully informed about the purpose and process of the study and that participation was voluntary and participants could withdraw at any time without prejudice. Written information about the study was provided and informed written consent was obtained. Participants were told that all information would remain anonymous and confidential and never used for any purpose other than this study.

Results

Demographic variables of participants

There were 259 primiparas (65.7%) and 135 multiparas (34.3%) in the study. Out of 394 participants, 109 (27.7%) had PBE. Results are shown in Table 1.

Table 1.

Demographic Variables of Participants (N = 394)

Characteristics	N	%
Age (years)
18–24	27	6.9
25–34	312	79.2
≥35	55	14
Education
Middle school	17	4.3
High school	55	14
University	261	66.2
Graduate	61	15.5
Household income (monthly, in RMB)
<5,000	38	9.6
5,000–10,000	164	41.6
10,000–20,000	140	35.5
≥20,000	52	13.2
Occupation
Yes	304	77.2
No	90	22.8
Prepregnancy BMI
Underweight	69	17.5
Normal weight	294	74.6
Overweight	30	7.6
Obese	1	0.3
Parity
Primiparas	259	65.7
Multiparas	135	34.3
PBE
Yes (multiparas)	109	27.7
No (multiparas)	26	6.6
No (primiparas)	259	65.7
Delivery mode
Vaginal	147	37.3
Cesarean	247	62.7
Having been breastfed as infants
Yes	334	84.8
No	41	10.4
Unknown	19	4.8
Breastfeeding initiation time (after delivery)
1 Hour	81	20.6
1–24 Hours	141	35.8
2 Days	81	20.6
3 Days	76	19.3
4 Days	15	3.8
Gender of newborn
Male	211	53.6
Female	183	46.4
EBF to 4 months	231	58.6
EBF to 6 months	120	30.5
ABF to 4 months	138	35
ABF to 6 months	202	51.3

ABF, any breastfeeding; BMI, body mass index; EBF, exclusive breastfeeding; PBE, previous breastfeeding experience.

Descriptive analyses of breastfeeding-related scales

All 135 multiparas completed the MBFES. Then 240 participants completed the BSE on the 42nd day after delivery. Subscales scores of the MBFES and BES are shown. The mean scores of breastfeeding knowledge, IIFAS, BSES-SF, MBFES, and BES s were 14.77, 46.51, 60.57, 114.05, and 33.95, respectively. Descriptive data of these five scales are presented in Table 2.

Table 2.

Descriptive Data of Scales

Variables	Sample size	M ± SD
Breastfeeding knowledge	394	14.77 ± 2.20
IIFAS	394	46.51 ± 11.77
BSES-SF	394	60.57 ± 5.82
MBFES	135	114.05 ± 11.44
Maternal enjoyment	135	59.95 ± 7.63
Infant satisfaction	135	29.83 ± 4.00
Maternal lifestyle	135	24.26 ± 4.12
BES	240	33.95 ± 8.32
Breast concerns	240	4.83 ± 1.60
Process concerns	240	11.29 ± 3.08
Mechanic concerns	240	8.25 ± 3.19
Milk insufficiency concerns	240	6.10 ± 2.65
Social concerns	240	3.46 ± 1.61

BES, breastfeeding experience scale; BSES-SF, breastfeeding self-efficacy short-form scale; IIFAS, Iowa infant feeding attitudes scale; MBFES, maternal breastfeeding evaluation scale; SD, standard deviation.

The association between PBE and breastfeeding self-efficacy, attitude, knowledge, and difficulties

According to the ANOVA, the F value of PBE was 0.34 (p = 0.713) in breastfeeding knowledge, 0.31 (p < 0.001) in IIFAS, 5.62 (p = 0.004) in BSES-SF, 4.45 (p = 0.013) in BSE, respectively.

Multiple comparison was then used to explore the specific difference among multiparas with PBE, multiparas without PBE and primiparas in breastfeeding attitude, self-efficacy, and difficulties. There were no significant differences between multiparas without PBE and primiparas in breastfeeding self-efficacy and difficulties. However, there were statistical differences between multiparas with PBE, multiparas without PBE and primiparas in breastfeeding attitude (Table 3).

Table 3.

Multiple Comparison of the Association Between Previous Breastfeeding Experience and Breastfeeding Attitude, Self-Efficacy, and Problems

Scales	PBE		p
BSES-SF
	Yes (multiparas)	No (multiparas)	0.006^a
	Yes (multiparas)	No (primiparas)	0.006^a
	No (multiparas)	No (primiparas)	0.166
IIFAS
	Yes (multiparas)	No (multiparas)	0.001^b
	Yes (multiparas)	No (primiparas)	0.001^b
	No (multiparas)	No (primiparas)	0.026^c
BES
	Yes (multiparas)	No (multiparas)	0.029^c
	Yes (multiparas)	No (primiparas)	0.005^a
	No (multiparas)	No (primiparas)	0.548

p < 0.01; ^bp < 0.001; ^cp < 0.05.

The association between characteristics of participants and breastfeeding at 4 and 6 months

Parity was significantly associated with exclusive breastfeeding at 4 and 6 months (p = 0.002; p = 0.001); PBE was significantly associated with exclusive and any breastfeeding at 4 and 6 months (both p < 0.001); breastfeeding initiation time was significantly associated with any breastfeeding at 4 and 6 months (p = 0.037; p = 0.001).

The association between breastfeeding self-efficacy, attitude, knowledge, difficulties and satisfaction, and breastfeeding status at 4 and 6 months

For exclusive breastfeeding, the self-efficacy and difficulties were significant at 4 months but only self-efficacy was significant at 6 months. For any breastfeeding, the five variables were significant at 4 months but only three at 6 months (Table 4).

Table 4.

Association Between Breastfeeding Self-Efficacy, Attitude, Knowledge, Problems, and Satisfaction and Breastfeeding Status at 4 and 6 Months

Scales	4 Months				6 Months
	EBF		ABF		EBF		ABF
	t	p	t	p	T	p	t	p
Breastfeeding knowledge	−1.16	0.247	−2.00	0.046^a	−1.48	0.141	−1.82	0.070
IIFAS	−0.95	0.342	−2.80	0.005^b	−1.66	0.098	−2.88	0.004^b
BSES-SF	−9.56	0.000^c	−7.27	0.000^c	−4.19	0.000^c	−67	0.000^c
MBFES	−0.90	0.371	−2.12	0.036^a	−0.44	0.660	−0.80	0.428
BES	6.30	0.000^c	6.18	0.000^c	5.39	0.000	7.31	0.000^c

p < 0.05; ^bp < 0.01; ^cp < 0.001.

Multivariate logistic regression of the association between variables and breastfeeding at 4 and 6 months

The factors included in the regression were selected when they had a significant effect on breastfeeding status at 4 and 6 months in chi-square tests, t test, and ANOVA. No matter exclusive or any breastfeeding, multiparas with PBE were more likely to breastfeed for 4 and 6 months based on higher self-efficacy and ability to manage difficulties (Table 5).

Table 5.

Multivariate Logistic Regression of the Association Between Variables and Breastfeeding at 4 and 6 Months

Variables	4 Months, OR (95% CI)		6 Months, OR (95% CI)
Variables	EBF	ABF	EBF	ABF
IIFAS	—	0.92 (0.81–1.05)	1.04 (0.97–1.11)	1.02 (0.94–1.11)
BSES-SF	1.10 (1.07–1.12)^a	1.08 (1.04–1.13)^a	1.07 (1.05–1.10)^a	1.10 (1.07–1.13)^a
Parity
Multiparas	—	1	1	—
Primiparas	—	1.66 (0.64–4.32)	1.66 (0.65–4.21)	—
Breastfeeding experience
No (multiparas)	1	1	1	1
Yes (multiparas)	3.75 (1.18–11.95)^a	4.71 (1.65–13.41)^a	4.69 (1.68–13.08)^a	4.10 (1.73–9.75)^a
No (primiparas)	1.45 (0.50–4.25)	1.66 (0.64–4.32)	1.66 (0.65–4.21)	1.73 (0.45–4.87)
Breastfeeding initiation time after delivery
1 Hour after delivery	—	1	—	1
1–24 Hours	—	1.73 (0.91–3.30)	—	0.92 (0.41–2.08)
2 Days	—	1.45 (0.70–2.99)	—	1.46 (0.54–3.95)
3 Days	—	1.16 (0.55–2.44)	—	0.30 (0.13–0.69)^a
4 Days	—	1.00 (0.27–3.73)	—	0.56 (0.14–2.28)
BES	0.93 (0.89–0.97)^a	0.82 (0.74–0.91)^a	0.87 (0.81–0.93)^a	0.86 (0.81–0.92)^a

p < 0.05.

CI, confidence interval; OR, odds ratio.

Discussion

Breastfeeding rates in this study

According to the results of this study, exclusive breastfeeding rates were 58.6% and 30.5% at 4 and 6 months; and any breastfeeding rates were 35.0% and 51.3% at 4 and 6 months. The latest available large survey on breastfeeding in China was conducted in 2015, and reported that exclusive breastfeeding rates at 4 and 6 months were 45.4% and 3.1% and any breastfeeding rates at 4 and 6 months were 38.1% and 69.7%.²⁵ The high rates of exclusive breastfeeding in this study might be attributed to the education level and economic status of the participants as well as their access to health care professionals. They were also recruited from large university-affiliated hospitals that promote a baby-friendly approach to care. Despite this setting, the exclusive breastfeeding rate at 6 months in this study does not meet the recommended rate. Factors that influence breastfeeding status will be discussed as follows.

The direct relationship between PBE, parity, and the duration of breastfeeding

PBE was an important factor influencing breastfeeding duration according to the current findings, which was similar to Wagner et al.,⁹ but some differences also existed. Wagner et al.⁹ reported that multiparas who had no PBE were less likely to maintain breastfeeding than primiparas. However, this study did not find any significant difference between multiparas without PBE and primiparas in breastfeeding duration at 4 and 6 months. Although many studies inferred that multiparas would have more breastfeeding difficulties with the increasing demands of caring for additional children, actual breastfeeding outcomes were better than primiparas. Hence, PBE played a more important role than social influences in predicting subsequent feeding decisions and behavior.²⁶ It was reported that the reasons why mothers without PBE chose to breastfeed might be persuaded by health care professionals²⁷ or the avoidance of feeling guilt and/or shame if they did not breastfeed.²⁸ Hackman et al. demonstrated that multiparas with PBE would have different subsequent breastfeeding experiences compared with primiparas, which reflected in several aspects: primiparas intending shorter breastfeeding duration, delaying breastfeeding initiation, being less likely to feed at least eight times in the first 24 hours, and having more breastfeeding difficulties during their maternity stay.²⁹ Therefore, there may be a direct relationship between PBE and exclusive breastfeeding status.

Apart from prior breastfeeding experience, Ahmed et al.³⁰ found that multiparas were more likely to continue breastfeeding to 6 months than primiparas, which was the same as found in this study. It was reported that primiparas might be overly optimistic about their breastfeeding goals and did not understand the challenges that might be concerned with breastfeeding, especially those who reported >6 months of planned duration.²⁹

The indirect relationship between PBE and the duration of breastfeeding

In this study, multiparas and primiparas without PBE do not have significant differences in self-efficacy and difficulties in breastfeeding. But multiparas with PBE had higher breastfeeding self-efficacy scores, breastfeeding attitude scores, and lower breastfeeding difficulties scores. This was also shown in other studies. One study reported that breastfeeding self-efficacy was greater in women with PBE.³¹ In addition, this study found an important positive effect of breastfeeding self-efficacy scores on continuing exclusive breastfeeding for at least 6 months, which was also uncovered in other studies.³² According to the results of this study, the odds of continuation of exclusive and any breastfeeding at 4 and 6 months were increased by 7.4–9.8% for every 1-unit increment in the BSES-SF score, which was slightly higher than the Husin et al.'s study.³² This difference may be associated with the characteristics of the population that was studied. Our study population is from mainland of China, but their study was carried out in Malaysia. Participants in these two studies are currently residing in a different cultural context. Beyond that, Kronborg et al.²⁶ reported that the association between PBE and subsequent exclusive/any breastfeeding could be interpreted as 48%/27% by intention and self-efficacy toward later breastfeeding. Therefore, self-efficacy may be an indirect factor between PBE and exclusive breastfeeding status.

As for attitude, multiparas with PBE had higher breastfeeding attitude scores. This finding is consistent with an earlier study.³³ One study reported that PBE explained a large part of breastfeeding continuation by improving breastfeeding self-efficacy, attitudes, and subjective norms.³³ One possible interpretation is the continuum of breastfeeding behavior, which may form customary beliefs. In addition, this study has found a significant association between breastfeeding attitude and continuing any breastfeeding for at least 6 months. There is controversial evidence in this regard. In Chen et al.'s study,³⁴ higher scores on the IIFAS in simplified Chinese are associated with breastfeeding initiation and duration in Chinese populations. On the contrary, in Bartle and Harvey's study,³³ it is notable that despite strong positive attitudes to breastfeeding, these were not associated with intentions or behavior. One possible reason is different measurements. Differences among different tools used in the assessment of the same variable are not new. Chen et al.'s and our studies used the IIFAS to measure attitudes. But the tool source of Bartle and Harvey's study is based on the theory of reasoned action. Clearly, more research is required to solve this controversial issue. Therefore, in our study, breastfeeding attitude may be another indirect factor between PBE and exclusive breastfeeding status.

As for breastfeeding difficulties, on the one hand, multiparas with PBE had lower breastfeeding difficulties scores. That may be because breastfeeding is a learning process. When multiparas with PBE was a first-time mother, they might have encountered many difficulties in breastfeeding. But difficulties encountered during the early breastfeeding experience may have been overcome once they mastered problem-solving skills. That is why PBE is so important. On the other hand, this study discovered that the higher the breastfeeding difficulty scores, the shorter the exclusive breastfeeding duration. Since the scores of the BES can help with evaluating the severity levels of breastfeeding difficulties, it is hard to know which kind of difficulty is associated with exclusive breastfeeding status. Breastfeeding difficulties associated with a higher risk of nonexclusive breastfeeding, which underline the importance of continued tailored professional breastfeeding support.¹⁶ Although most mothers would encounter difficulties during breastfeeding, they also persisted in breastfeeding for the recommended period. This might be that they attached greater importance to the satisfaction gained in breastfeeding by mothers and infants and their reflection on maternal identity. Therefore, in our study, breastfeeding difficulties may also be an indirect factor between PBE and exclusive breastfeeding status.

In summary, PBE can influence the breastfeeding outcomes indirectly through self-efficacy, attitude, and difficulties. For women without PBE, improving their self-efficacy, attitude, and solving their breastfeeding difficulties may be complementary approaches to promote breastfeeding.

The mean score on the BSES-SF was 60.57 in this study. One study reported that the mean scores of BSES-SF ranged from 46.48 to 47.3, lowering than scores in this study.³⁵ The mean score on the IIFAS was 46.51 in this study. One study found that the mean scores of IIFAS ranged from 57.70 to 64.67, which were much higher than this study.³⁶ Different composition of studied subjects and different location of the studies may be the reason for difference. Although all the populations are Chinese, economic development levels and groups of people are different in Wuhan and Guangzhou.

The five most serious difficulties indicated by participants at 42 days after delivery were sore nipple(s), leaking breasts, feeling very tired, or fatigued, worrying that baby was not getting enough milk and concern about not having enough breast milk. These results differed from other study that found that the most common breastfeeding difficulties after delivery were cracked and inverted nipples, breast mastitis, engorgement, and abscess.²⁷ This may be because we used different methods compared with the previous study,²⁷ in which qualitative approach was used to capture a person's lived experience. Our quantitative study has larger sample size and explained the relationships better.

The relationship between MBFES, knowledge, and any breastfeeding status

The mean score on the MBFES was 114.05 in this study, which meant multiparas with PBE had a high level of satisfaction in previous breastfeeding. Our study found a significant correlation between maternal breastfeeding satisfaction and any breastfeeding status at 4 months. Similar to the current findings, one study indicated that a high maternal breastfeeding satisfaction score was a positive factor in breastfeeding duration.³⁷ Our study also adds weight on persuasion of findings of Moimaz et al.'s study, who indicated that the positive PBE could help enhance confidence and self-efficacy to breastfeed a consequent child.¹⁵

Prior studies showed that higher scores on breastfeeding knowledge were positively associated with longer exclusive breastfeeding duration,³⁸ but in our study, we just found that a significant correlation between knowledge and any breastfeeding status at 6 months. The scores on breastfeeding knowledge in this study were high because participants were well educated and they could access to breastfeeding information on the Internet, classes, or books. Nevertheless, lack of practical skills of breastfeeding or misunderstanding the exact definition of exclusive breastfeeding might influence exclusive breastfeeding. This could be interpreted by a recent study, which demonstrated a gap between breastfeeding knowledge and practice.³⁹ As reported, some breastfeeding mothers offered water and other fluid to infants during breastfeeding and regarded as exclusive breastfeeding.⁴⁰ These suggest that translating knowledge into practice is also crucial in promoting breastfeeding.

Implications for future research

The findings in this study suggest that future interventions would be more efficient by considering PBE than other conventional factors. This is especially important for mothers who did not breastfeed a previous child. Breastfeeding interventions should focus on understanding the experience and be adapted to this new experience. Large baby-friendly hospitals can expand their support for postpartum women to include breastfeeding support groups, offering lactation consultation services, and using multiparas with PBE as peer leaders to work with primiparas. For women without PBE, improving their self-efficacy, attitude, and solving their breastfeeding difficulties may be complementary approaches to promote breastfeeding. Moreover, “breastfeeding online support group” on WeChat, QQ can be used to teaching breastfeeding practical skills and describing breastfeeding realistically. This group then becomes a model for promoting breastfeeding.

Conclusions

The rate of breastfeeding remains low among Chinese women, but PBE is associated with a higher probability of breastfeeding at 4 and 6 months postpartum. Multiparas, especially those having PBE, were more likely to breastfeed for an extended period based on their attitude, self-efficacy, and ability to manage difficulties. In addition, multiparas and primiparas without PBE do not have significant differences in self-efficacy and difficulties in breastfeeding.

Footnotes

Authors' Contributions

J.-Y.L. was in charge of methodology, software, formal analysis, writing original draft, reviewing, and editing; Y.H. took the responsibility of conceptualization, investigation, methodology, and formal analysis; H.-Q.L. took up investigation; J.X. was in charge of validation and resources; L.L. was responsible for resources and reviewing; S.R.R. performed reviewing and editing; Y.-Q.O. was in charge of conceptualization, project administration, reviewing, and editing.

Acknowledgments

We are especially grateful to all who participated in this study. This study could not have been possible without their assistance.

Disclosure Statement

No competing financial interests exist.

Funding Information

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

References

Rouw

, von Gartzen

, Weissenborn

. The importance of breastfeeding for the infant. Bundesgesundheitsblatt Gesundheitsforschung Gesundheitsschutz, 2018; 61:945–951.

Victora

, Bahl

, Barros

, et al. Breastfeeding in the 21st century: Epidemiology, mechanisms, and lifelong effect. Lancet, 2016; 387:475–490.

World Health Organization (WHO). Infant and young child feeding. 2017. Available at www.who.int/mediacentre/factsheets/fs342/en (accessed January 1, 2020).

United Nations Children's

Fund

, Division of Data Research and

Policy

. Global UNICEF global databases: Infant and young child feeding, New York. 2018. Available at https://data.unicef.org/topic/nutrition/infant-and-young-child-feeding (accessed January 1, 2020).

Yang

, Lai

, Yu

, et al. Breastfeeding rates in China: A cross-sectional survey and estimate of benefits of improvement. Lancet, 2016; 388:S47.

Khanal

, Lee

, Karkee

, et al. Prevalence and factors associated with prelacteal feeding in Western Nepal. Women Birth, 2016; 29:12–17.

Imsiragic

, Begic

, Sarajlic

, et al. Predictors of exclusive breastfeeding 6–9 weeks after delivery: A prospective cohort study. Public mental health perspective. Psychiatr Danub, 2016; 28:395–403.

Saffari

, Pakpour

, Chen

. Factors influencing exclusive breastfeeding among Iranian mothers: A longitudinal population-based study. Health Promot Perspect, 2017; 7:34–41.

Wagner

, Kersuzan

, Gojard

, et al. Breastfeeding initiation and duration in France: The importance of intergenerational and previous maternal breastfeeding experiences—Results from the nationwide ELFE study. Midwifery, 2019; 69:67–75.

10.

Huang

, Ouyang

Y-Q

, Redding

. Previous breastfeeding experience and its influence on breastfeeding outcomes in subsequent births: A systematic review. Women Birth, 2019; 32:303–309.

11.

Mortazavi

, Mousavi

, Chaman

, et al. Validation of the Breastfeeding Experience Scale in a sample of Iranian mothers. Int J Pediatr, 2014; 2014:608657.

12.

Gallagher

Sub-optimal breastfeeding experience and its effects on women, factors affecting initiation and duration of breastfeeding in Ireland [PhD]. Trinity College, Dublin, 2012.

13.

Crippa

, Colombo

, Morniroli

, et al. Do a few weeks matter? Late preterm infants and breastfeeding issues. Nutrients, 2019; 11:312.

14.

Palmer

. Previous breastfeeding difficulties: An existential breastfeeding trauma with two intertwined pathways for future breastfeeding-fear and longing. Int J Qual Stud Health Well-being, 2019; 14:1588034.

15.

Moimaz

SAS

, Rocha

, Garbin

CAS

, et al. Factors affecting intention to breastfeed of a group of Brazilian childbearing women. Women Birth, 2017; 30:e119–e124.

16.

Gianni

, Bettinelli

, Manfra

, et al. Breastfeeding difficulties and risk for early breastfeeding cessation. Nutrients, 2019; 11:2266.

17.

Ouyang

, Xu

, Zhang

. Survey on breastfeeding among Chinese female physicians and nurses. Nurs Health Sci, 2012; 14:298–303.

18.

Wambach

. Breastfeeding intention and outcome: A test of the theory of planned behavior. Res Nurs Health, 1997; 20:51–59.

19.

Wambach

. Lactation mastitis: A descriptive study of the experience. J Hum Lact, 2003; 19:24–34.

20.

Leff

, Jefferis

, Gagne

. The development of the Maternal Breastfeeding Evaluation Scale. J Hum Lact, 1994; 10:105–111.

21.

Dennis

. The breastfeeding self-efficacy scale: Psychometric assessment of the short form. J Obstet Gynecol Neonatal Nurs, 2003; 32:734–744.

22.

, Gao

, Choi

, et al. The Short Form of the Breastfeeding Self-Efficacy Scale as a prognostic factor of exclusive breastfeeding among Mandarin-speaking Chinese mothers. J Hum Lact, 2016; 32:711–720.

23.

De la Mora

, Russell

. The Iowa Infant Feeding Attitude Scale: Analysis of reliability and validity. J Appl Soc Psychol, 1999; 29:2362–2380.

24.

Chen

, Binns

, Liu

, et al. Attitudes towards breastfeeding - the Iowa Infant Feeding Attitude Scale in Chinese mothers living in China and Australia. Asia Pac J Clin Nutr, 2013; 22:266–269.

25.

, Li

, Zhang

, et al. National survey showed that Chinese city children under two years of age had similar feeding patterns to developed countries. Acta Paediatr, 2018; 107:1555–1561.

26.

Kronborg

, Foverskov

, Vaeth

, et al. The role of intention and self-efficacy on the association between breastfeeding of first and second child, a Danish cohort study. BMC Pregnancy Childbirth, 2018; 18:454.

27.

Dietrich Leurer

, Misskey

. “Be positive as well as realistic”: A qualitative description analysis of information gaps experienced by breastfeeding mothers. Int Breastfeed J, 2015; 10:10.

28.

Williams

, Webber

, Pell

, et al. “Nobody knows, or seems to know how rheumatology and breastfeeding works”: Women's experiences of breastfeeding whilst managing a long-term limiting condition—A qualitative visual methods study. Midwifery, 2019; 78:91–96.

29.

Hackman

, Schaefer

, Beiler

, et al. Breastfeeding outcome comparison by parity. Breastfeed Med, 2015; 10:156–162.

30.

Ahmed

, Talha

, Khalid

, et al. Breastfeeding and weaning: Practices in urban slums of Southern Punjab, Pakistan. Cureus, 2018; 10:e2189.

31.

Ngo

LTH

, Chou

, Gau

, et al. Breastfeeding self-efficacy and related factors in postpartum Vietnamese women. Midwifery, 2019; 70:84–91.

32.

Husin

, Isa

, Ariffin

, et al. The Malay version of antenatal and postnatal breastfeeding self-efficacy scale-short form: Reliability and validity assessment. Malaysian J Public Heal Med, 2017; 17:62–69.

33.

Bartle

, Harvey

. Explaining infant feeding: The role of previous personal and vicarious experience on attitudes, subjective norms, self-efficacy, and breastfeeding outcomes. Br J Health Psychol, 2017; 22:763–785.

34.

Chen

, Binns

, Liu

, et al. Attitudes towards breastfeeding—The Iowa Infant Feeding Attitude Scale in Chinese mothers living in China and Australia. Asia Pac J Clin Nutr, 2013; 22:266–269.

35.

Sun

, Chen

, Yin

, et al. Why Chinese mothers stop breastfeeding: Mothers' self-reported reasons for stopping during the first six months. J Child Health Care, 2017; 21:353–363.

36.

Hamze

, Mao

, Reifsnider

. Knowledge and attitudes towards breastfeeding practices: A cross-sectional survey of postnatal mothers in China. Midwifery, 2019; 74:68–75.

37.

Ding

, Luo

. Effect of individualized intervention on postpartum breast-feeding behavior after cesarean section. Sichuan Da Xue Xue Bao Yi Xue Ban, 2019; 50:609–614.

38.

Zielińska

, Sobczak

, Hamułka

. Breastfeeding knowledge and exclusive breastfeeding of infants in first six months of life. Rocz Panstw Zakl Hig, 2017; 68:51–59.

39.

Mundagowa

, Chadambuka

, Chimberengwa

, et al. Determinants of exclusive breastfeeding among mothers of infants aged 6 to 12 months in Gwanda District, Zimbabwe. Int Breastfeed J, 2019; 14:30.

40.

Park

, Ryu

. Effects of breastfeeding interventions on breastfeeding rates at 1, 3 and 6 months postpartum: A systematic review and meta-analysis. J Korean Acad Nurs, 2017; 47:713–730.