## INTRODUCTION

Exclusive breastfeeding (EBF) is considered an effective way to provide balanced nutrition, better growth, development of the infant, prevent stunting as well as protect from infectious and chronic diseases1. Simultaneously, it decreases infant morbidity and mortality due to common illnesses2 such as diarrhea or pneumonia etc. The WHO (World Health Organization) and UNICEF (United Nations Children’s Emergency Fund) recommended exclusive breastfeeding for the first six months of age and giving no other food or drink to the infant, not even a single drop of water in this period3.

Globally, the exclusive breastfeeding rate is only 41% of infants under six months of age4. The collective target for the global rate in 2030 is 70% for exclusive breastfeeding5. Exclusive breastfeeding rate is comparatively high in Eastern and Southern Africa (55%), South Asia (54%) and lower in the Middle East (30%) and North Africa (29%)6. Moreover, the prevalence of exclusive breastfeeding among South Asian countries was: 54.9% in India, 1.4% in China, 65.2% in Nepal, 37.7% in Pakistan, and 82.3% in Sri Lanka (WHO 2018). In Bangladesh, the prevalence of exclusive breastfeeding was about 55% in children under six months of age7. This lower exclusive breastfeeding rate needs proper intervention to promote exclusive breastfeeding to achieve the targets of SDG-3 (Sustainable Development Goals) by reducing infant mortality and morbidity (minimize the neonatal mortality <12 and under-5 mortality to <25) by 20308. Worldwide, exclusive breastfeeding is considered an effective intervention for preventing infant morbidity and mortality and can prevent 1.4 million deaths among children every year2. So, neonatal and infant mortality can be reduced by increasing exclusive breastfeeding practices9.

Several factors are impeding the continuation of exclusive breastfeeding practice among lactating mothers in the world, including: 1) Geographical factors such as living area, place of delivery10; 2) Socioeconomic factors such as the education of mother11, occupation of the mother12; and 3) Health-related and individual factors such as mother’s age, mode of delivery, the gender of the child, cultural practices, support and guidance provided by health staff or doctor, etc.8 In addition, Bangladeshi mothers face several factors that hamper them from continuing exclusive breastfeeding practices resulting in increased infant and maternal morbidity and mortality rates. Previous studies report that health facilities and delivery (hospital vs home, normal vs caesarian delivery, etc.) are associated3 with the prevalence of exclusive breastfeeding practices8. However, few studies have been conducted on risk factors associated12 with the prevalence of exclusive breastfeeding practices as well as household, environmental, and cultural factors. No studies have been conducted in Bangladesh on associations among knowledge, attitudes, and some sociodemographic factors with exclusive breastfeeding practices. As a result, there is an urgent need to evaluate knowledge, attitude, and associated sociodemographic factors with exclusive breastfeeding practice. The findings of this study will be useful to governments/NGOs (Non-Governmental Organizations) to improve policy that focuses on resources and interventions to promote exclusive breastfeeding practices that help to achieve SDG-3 (Sustainable Development Goals) and reduce infant morbidity and mortality rates. Therefore, this study has been designed to evaluate maternal knowledge, attitude and factors associated with exclusive breastfeeding among mothers who had a child aged <12 months.

## METHODS

### Description of the study area

This study was conducted in the district of Dhaka at the Mother and Child Hospital. Dhaka has a population of about 12 million. Mothers from all socioeconomic status attend this hospital due to the availability of maternal and child healthcare services. Therefore, this hospital was considered as being representative of the study population.

### Study design and study population

This cross-sectional study was conducted among mothers who came to the hospital from July to December 2019. A total of 385 mothers were interviewed for the study who came to the hospital’s outpatient department to get treatment related to mother and child. The inclusion criteria were women who had children aged ≤12 months. Exclusion criteria were: the medical condition of the mother; medications which interrupted breastfeeding; and mothers who were diagnosed with severe congenital malformations; and those who had >12 months old children. These conditions were verified by using a health card which was monitored by the duty doctor at the hospital.

### Sample size estimation and procedure

The minimum required sample size was calculated by using a single population proportion formula no = Z2 p q /d2; 5% margin of error, 95% confidence intervals, and estimated to practice EBF 65% of the national rate13. This yielded a sample size of 385 with 10% non-response rate. The mothers were randomly selected from the healthcare register of the hospital and interviewed when they waited for healthcare services. If the selected mother was not interested in the study she was excluded.

### Data collection

Research assistants were recruited and trained for the fieldwork of the study and collection of data by reviewing the questionnaire through teamwork with a researcher. Structured questionnaires were used to collect data. A random sampling approach was adopted to collect information like knowledge, attitude, and some sociodemographic factors through face-to-face interviews.

### Questionnaire

The questionnaire involved some general information questions like sociodemographic information including name, age, education level, occupation, living area, economic status, religion, place, and types delivery, etc.

To ascertain maternal knowledge on exclusive breastfeeding, a set of questions was developed. Knowledge on exclusive breastfeeding was measured by asking the mother 10 questions such as benefits of exclusive breastfeeding and colostrum, approach of handling breastfeeding during pregnant condition of mother and infant with diarrhea disease, and other questions about advantages of breastfeeding both for infant and mother etc. Each correct answer was given a score of 1, while a wrong answer was given a score of 0 . The total knowledge score was then calculated and graded on a scale based on standards carried out in previous research, as follows: Good (≥7), medium (4-6) and poor (≤3)14.

In the same way, attitude toward breastfeeding was assessed using a series of questions such as exclusive breastfeeding is beneficial for infant and mother, baby can survive without water up to 6 months of age, formula milk is unsuitable for baby and not a better choice for working mothers, and breast milk is more easily digested than formula milk. If the mother agreed with this opinion she was given a score of one and 0 if she disagreed or was neutral with the opinion. The total score was then calculated and the attitude was graded as: positive (≥7), moderate (4-6) or poor (≤3)14.

### Statistical analysis

Statistical analyses were performed using IBM SPSS (version 23). Means and standard deviation (SD) were used for numerical data, whereas percentages were used for categorical data. Chi-squared test was used to assess the association of knowledge and attitude on breastfeeding with exclusive breastfeeding practices. Multiple logistic regression analyses were done to assess sociodemographic factors on exclusive breastfeeding practices. The model fitness was tested using Hosmer and Lemeshow test. The adjusted odds ratio (AOR) was calculated to evaluate the strength of the association of sociodemographic factors with exclusive breastfeeding practices at 95% CI for the significance test. A value of p<0.05 was considered statistically significant.

## RESULTS

Table 1 shows the sociodemographic and other characteristics of respondents. In this study population, more than half (54.5%) of the mothers were aged ≥24 years, with a mean (±SD) age of 26.1 (±5.3) years. In regard to education level, 17.2% respondents were illiterate, 18.7% were primary educated and the remaining 64.2% had secondary or higher level of education and most (87.1%) of the mothers were Muslims. Most of the respondents (70.9%) had a monthly family income <15000 BDT (1000 Bangladeshi Taka about US$12). About half (52.9%) of the mothers had one child and most (77.1%) of the mothers were married. Most mothers (70.9%) were housewives and more than half (60.5%) of the mothers came from a single family. About half of the mothers (55.4%) delivered their child by vaginal/normal delivery and two-thirds (68.3%) delivered their child in hospital. ##### Table 1 Sociodemographic and other characteristics of mothers in the study (N=385) CharacteristicsTotal Exclusive breastfeeding p* n (%)Yes n (%)No n (%) All participants244 (63.4)141 (36.6) Maternal age (years) ≤24210 (54.5)100 (47.6)110 (52.4)0.027 >24175 (45.5)144 (82.3)31 (17.7) Maternal education level Illiterate66 (17.2)5 (7.6)61 (92.4)<0.001 Primary72 (18.7)61 (84.7)11 (18.3) SSC or above247 (64.2)178 (72.1)69 (38.9) Maternal occupation Housewife273 (70.9)209 (76.6)64 (23.4)0.002 Service holder112 (29.1)35 (31.3)77 (68.8) Family monthly income (BDT) <15000276 (71.7)160 (57.9)116 (42.1)<0.001 ≥15000109 (28.3)84 (77.1)25 (22.9) Family type Single233 (60.5)183 (78.5)50 (21.5)<0.001 Joint152 (39.5)61 (40.1)91 (59.9) Religion Muslim335 (87.0)238 (61.8)97 (41.6) Non-muslim50 (12.9)06 (12)44 (88.0)0.268 Marital status of mother Married297 (77.1)235 (79.1)62 (20.9)<0.001 Divorced88 (22.9)9 (10.2)79 (89.8) Live birth 1 child204 (52.9)161 (78.9)43 (21.1)<0.001 2 or more children181 (47.01)83 (45.9)98 (54.1) Place of delivery Home122 (31.7)47 (38.5)75 (61.5)0.008 Hospital263 (68.3)197 (74.9)66 (25.1) Mode of delivery Normal/vaginal213 (55.3)181 (84.9)32 (15.0)<0.001 Caesarian172 (44.7)63 (36.7)109 (63.4) BDT: 1000 Bangladeshi Taka about US$12.

* Statistically significant: p<0.05 (chi-squared).

The association of maternal knowledge and attitude score on exclusive breastfeeding is shown in Table 2. Out of 384 mothers, more than half (63.4%) practiced exclusive breastfeeding. The mean (±SD) knowledge and attitude scores, out of a maximum of 10, were 7.46 ± 2.56 and 7.57 ± 2.43, respectively. More than half (58.6%) of the mothers had a high knowledge score and about 55.2% had a positive attitude. Chi-squared test found that maternal knowledge score (χ2=37.186, p<0.001) and attitude score (χ2=12.672, p<0.001) were statistically associated with exclusive breastfeeding practices among mothers.

##### Table 2

The association of maternal knowledge and attitude scores on exclusive breastfeeding

Total
Exclusive breastfeeding
Chi-squared statistics
Yes n (%)Yes (n=189) n (%)No (n=111) n (%)χ2 (df)p*
Knowledge score (out of 10)
Good225 (58.6)171 (76)54 (24)
Medium122 (31.8)56 (49.9)66 (54.1)37.186 (2)<0.001
Poor38 (9.1)17 (44.7)21 (55.3)
Attitude score (out of 10)
Positive212 (55.2)140 (66.1)72 (33.9)
Moderate139 (36.2)85 (61.2)54 (38.9)12.672 (1)0.002
Poor34 (8.9)12 (35.3)22 (64.7)

Knowledge score: good=7–10, medium=4–6, and poor=0–3. Attitude score: positive=7–10, moderate=4–6, and poor=0–3.

* Statistically significant: p<0.001 (chi-squared).

Table 3 shows the multiple logistic regression analysis used to identify some sociodemographic and other factors associated with exclusive breastfeeding practices among mothers. The adjusted regression model found the following sociodemographic factors were statistically significant with exclusive breastfeeding practice among mothers: housewife (AOR=5.84; 95% CI: 2.42-14.13, p<0.001), maternal education level of primary (AOR=8.84; 95% CI: 1.25-62.66, p<0.05) and secondary or above (AOR=7.16; 95% CI: 1.3-39.36, p<0.05), having family monthly income ≥15000 BDT (AOR=5.41; 95% CI: 1.78-16.47, p<0.001), having one child (AOR=3.96; 95% CI: 1.52-10.35, p<0.001), normal delivery (AOR=13.58; 95% CI: 5.29-35.48, p<0.001) and delivery at hospital (AOR=3.38; 95% CI: 1.24-9.23, p<0.001); these mothers were more likely to have good exclusive breastfeeding practice than their counterparts and it was statistically significant (p<0.05). Moreover, mothers from joint family (AOR=0.27; 95% CI: 0.12-0.62, p<0.05) or divorced (AOR=0.12; 95% CI: 0.03-0.58, p<0.005) were less likely to have exclusive breastfeeding practice compared to their counterparts and these factors also were statistically significant (p<0.05).

##### Table 3

Effect of sociodemographic and other characteristic on practices of exclusive breastfeeding

FactorsEBF Yes n (%)AOR (95% CI)P*
Maternal age (years)
≤24210 (54.5)0.61 (0.23–1.61)0.318
>24 (Ref.)175 (45.5)-
Maternal education level
Illiterate (Ref.)66 (17.2)-
Primary72 (18.7)8.84 (1.25–62.66)0.029
SSC or above247 (64.2)7.16 (1.3–39.36)0.024
Maternal occupation
Housewife273 (70.9)5.84 (2.42–14.13)<0.001
Service holder (Ref.)112 (29.1)-
Family monthly income (BDT)
<15000 (Ref.)276 (71.7)-
≥15000109 (28.3)5.41 (1.78–16.47)0.003
Family type
Single (Ref.)233 (60.5)-
Joint152 (39.5)0.27 (0.12–0.62)0.002
Marital status of mother
Married (Ref.)297 (77.1)-
Divorced88 (22.9)0.12 (0.03–0.58)0.008
Live birth
1 Child204 (52.9)3.96 (1.52–10.35)0.005
2 or more children (Ref.)181 (47.01)-
Place of delivery
Home (Ref.)122 (31.7)-
Hospital263 (68.3)3.38 (1.24–9.23)0.017
Mode of delivery
Normal delivery213 (55.3)13.58 (5.29–35.48)<0.001
Caesarian (Ref.)172 (44.7)-

BDT: 1000 Bangladeshi Taka about US\$12. AOR: adjusted odds ratio. EBF: exclusive breastfeeding. Multiple logistic model fitted by Hosmer and Lemeshow test (χ2=9, df=8, p=0.67).

* Statistically significant: p<0.05 (chi-squared).

## DISCUSSION

This study demonstrated that about two-thirds of mothers (63.4%) exclusive breastfed their child up to 6 months of age, which is close to the national level (65%) in Bangladesh13. The prevalence of EBF in this study was higher than other studies conducted in Bangladesh such as a study conducted among Bangladeshi mothers in a rural area which reported a prevalence of EBF of 35.9%15 and another study conducted in the rural area of Rajshahi district which reported an EBF practice rate of 34.5%16. In both studies, the rural areas lacked health facilities and a large proportion of participants was illiterate. EBF practice rate was relatively lower among illiterate mothers compared with literate mothers as found in another study17. So, health facilities and educational level were associated with lower EBF practice in that study area. In this study, the prevalence of EBF reported could be attributed to the large number of educated mothers, availability of health facilities, and presence of NGOs implementing programs aimed at improving breastfeeding and reducing child malnutrition.

This study found that a good knowledge score on breastfeeding was statistically significant with EBF practice among mothers, in accordance with previous studies3,18. A recent study conducted in Wajir, Kenya, reported that knowledge score was not associated with EBF practice, in contrast with our findings14. In that study, high knowledge was attributed to the presence of government and NGO intervention programs aimed at improving infant and young child feeding (IYCF) practices to achieve the aims of the National Nutrition Action Plan of Kenya. WHO model of IYCF behavior reports that cultural attitudes, beliefs and norms are important factors associated with infant formula feeding19. Cultural beliefs on infant feeding may result in more infant formula feeding than EBF practice. So, cultural beliefs were identified that may lead to failed EBF practices. However, a study conducted among Nigerian3 mothers found that a good knowledge score indicates better EBF practice, which was similar to our study. Furthermore, a study conducted among Jamaican mothers18 reported that the EBF practices were significantly considered by those who had adequate knowledge on breastfeeding. Good knowledge score indicates awareness of the advantages of exclusive breastfeeding and the disadvantages of formula milk/bottle feeds.

The present study revealed that a positive attitude towards breastfeeding significantly improved EBF practice among mothers, consistent with previous studies14,18. However, previous studies conducted among Jamaican mothers18 and among Kenyan mothers14 found a similarly significant relationship between EBF practices and a positive attitude of the mother on breastfeeding. Mothers who were positive toward exclusive breastfeeding agreed that breastfeeding is beneficial both for infant and mother, that the baby can survive without water up to 6 months of age, formula milk is unsuitable for baby and not as easily digested as breast milk.