Knowledge, attitude, practice and determinants of exclusive breastfeeding among women in Ethiopia: Systematic review and meta-analysis

Abstract

Objective

This study aimed to assess the pooled estimate of Ethiopia's women's knowledge, attitudes, practices, and determinants of exclusive breastfeeding.

Methods

PubMed, Google Scholar, Scopus, Science Direct, and Addis Ababa University online library were searched. Data were extracted using Microsoft Excel and analyzed using STATA statistical software (v. 14). Publication bias was checked by forest plot, Begg's rank test, and Egger's regression test. To look for heterogeneity, I² was computed, and an overall estimated analysis was carried out. Subgroup analysis was done by region, study setting, and publication. The pooled odds ratio for associated factors was also computed.

Results

Out of 995 studies reviewed, 33 met our eligibility criteria and were included in this meta-analysis study. The total number of study participants was 13,397. The pooled prevalence of good knowledge, positive attitude, and poor practice of kangaroo mother care were 74.2% (95% CI: 62.9–85.4), 77.2% (95%CI: 68.3–86.0), and 58.3% (95% CI: 49.9–66.6), respectively. In sub-group analysis, the highest prevalence of knowledge was in institution-based studies (78.3%) and unpublished studies (76.3%). A positive attitude was also highest in institutional-based studies (81%). The highest practice prevalence was in Afar (68%) and the lowest was in Addis Ababa (34.6%).

Women who had a secondary level of education (AOR = 3.3; 95%CI: 1.8–6.0) were a housewife (AOR = 3.1; 95%CI: 2.1–4.7), delivered vaginally (AOR = 2.0; 95%CI: 1.4–2.9), health facility delivery (AOR = 3.3; 95%CI: 2.1–5.1) and attending antenatal care were predictors of exclusive breastfeeding.

Conclusion

Although women have good knowledge and positive attitude toward exclusive breastfeeding, there is a significant gap in exclusive breastfeeding practice. Maternal education, occupation, vaginal delivery, institutional delivery, and antenatal care visit were predictors of exclusive breastfeeding. It is recommended to strengthen maternal and child health services.

1. Introduction

Breastfeeding is a baby's natural first food, giving them the energy and minerals they need during their first six months [1,2]. Exclusive breastfeeding is a crucial public health method for improving mother and child health by lowering healthcare costs and reducing infant morbidity and mortality [3]. Breastfeeding practices can save 1.4 million deaths in children under the age of five each year [4]. Additionally, breastfeeding strengthens the mother-child relationship, which lowers the risk of childhood diseases like sepsis, meningitis [5,6], pneumonia, diabetes mellitus, and diarrhea [7,8]. Breastfeeding increases sensory and cognitive growth in children and is associated with improved IQ scores [9,10]. It also accelerates weight loss, prevents pregnancy, and lowers bleeding, postpartum depression, ovarian, and endometrial cancer in women [7,11].

The World Health Organization and UNICEF recommend all mothers exclusively breastfeed their infants for the first six months [12]. When an infant receives no other food or drink during the first six months of life except vitamins, minerals, and medications, this is referred to as exclusively breastfeeding [13]. For the first six months of life, 38% of babies worldwide were exclusively breastfed [14]. According to research done in Western and Sub-Saharan African nations, rates of exclusive breastfeeding ranged from 13.0% in Côte d'Ivoire to 58.0% in Togo and 45.2% in sub-Saharan nations, respectively [15,16].

Literatures showed that barriers to breastfeeding included cultural norms, lack of maternity leave services [[17], [18], [19]], education and healthcare access, and cultural attitudes [20,21]. Important modifiable factors for children's health, growth, and development include mothers' knowledge, attitudes, and practices around newborn feeding [[22], [23], [24], [25], [26]]. As a result, a high degree of maternal education and a positive attitude has a significant impact on the process of exclusive breastfeeding [[27], [28], [29], [30]]. A recent study [31,32] found that women with more knowledge of exclusive breastfeeding and a positive attitude were more likely to practice exclusive breastfeeding than their counterparts. Furthermore, there is a strong correlation between infant mortality and non-exclusive breastfeeding. There are 1.24 million infant deaths within the first six months because non-exclusive breastfeeding accounts for over 96% of all infant deaths. Additionally, baby fatalities from pneumonia and diarrhea were associated with two- and four-fold greater risks, respectively, compared to breastfeeding exclusively [33]. Nutritional therapies like exclusive breastfeeding may be able to prevent these deaths [34].

Even though Ethiopia's government adopted the World Health Organization's recommendation to exclusively breastfeed infants younger than six months [35], the country still has a high infant mortality and morbidity rate. In Ethiopia, the prevalence of exclusive breastfeeding was also between 22.8% [36] and 90.8% [37] in terms of knowledge, 56.7% [38] to 90% [39] in terms of practice, and 14.9%–96.3% [40] in terms of attitude. Also noted was the discrepancy between studies regarding key factors. There is no complete analysis of Ethiopia's knowledge, attitudes, and practices on exclusive breastfeeding. Therefore, the present study sought to determine the pooled prevalence of knowledge, attitude, and practice of exclusive breastfeeding and associated factors in Ethiopia. The goal was to provide fundamental data for policymakers, clinicians, and other stakeholders to help develop appropriate strategies and interventions for the control and management of exclusive breastfeeding.

2. Methods

2.1. Reporting

This systematic review and meta-analysis study was conducted to determine the pooled prevalence of knowledge attitude practice and its determinants of exclusive breastfeeding in Ethiopia using the standard PRISMA checklist guideline [41] (Supplementary file 1). This systematic review and meta-analysis study was not registered under Prospero, but we checked that any author had not registered it yet.

2.2. Search strategy

International online databases (Pub Med, Science Direct, Scopus, and Google Scholar) were used to search for articles on the prevalence of exclusive breastfeeding knowledge, attitude, and practice. We also retrieved gray literature from Addis Ababa University's online research institutional repository. The search string was established using “AND” and “OR” Boolean operators. The following core search terms and phrases with Boolean operators were used to search related articles: ((((((((((Knowledge) OR (“Knowledge” OR “Awareness")) AND Attitude) OR (“Attitude” OR “Perception")) AND Practice) OR (“Practice” OR “Practice management” OR “Breastfeeding practice")) AND Determinants) OR (“Determinants” OR “Factors” OR “Predictors")) AND Exclusive breastfeeding) OR (“Exclusive breastfeeding” OR “Breastfeeding” OR” Breast")) AND Ethiopia. Search terms were based on PICO principles to retrieve relevant articles through the databases mentioned above. The search period was from April 1/2022 to May 10/2022.

2.3. Outcome measurement

Exclusive breastfeeding knowledge: Mothers who responded to ≥70% of the knowledge-related questions were considered to have a good level of knowledge, while those who responded to less than 70% were considered to have a poor level of knowledge [42].

Exclusive breastfeeding Attitude: Mothers who answered ≥70% of the mean value of the attitude-related questions were considered to exhibit a positive attitude towards exclusive breastfeeding whereas those who answered below the mean value of the attitude-related questions were considered to exhibit a negative attitude [42].

Exclusive breastfeeding practice: Mothers who responded to ≥70% of the practice-related questions were categorized as demonstrating good practice, whereas those respondents who answered <50% of the questions were considered to demonstrate poor practice [42].

2.4. Inclusion and exclusion criteria

Studies that reported the prevalence of knowledge, attitude, and practice of exclusive breastfeeding women as study participants, only English language articles, both published and unpublished studies that had full text available for search and took place in Ethiopia were included in this meta-analysis. Those studies that reported duplicated sources, qualitative studies from developed countries, and articles without full text available were excluded from this systematic review and meta-analysis.

2.5. Quality assessment

Two authors (NA and KD) independently appraised the standard of the studies using the Joanna Briggs Institute (JBI) standardized quality appraisal checklist [43]. The disagreement raised during the quality assessment was resolved through a discussion led by the third author (GB). Finally, the argument was solved and reached an agreement. The critical analysis checklist has eight parameters with yes, no, unclear, and not applicable options. The parameters involve the following questions:

• (1)Where were the criteria for inclusion in the sample clearly defined?
• (2)Were the study subjects and, therefore, the setting described in detail?
• (3)Was the exposure measured result validly and reliably?
• (4)Were the main objective and standard criteria used to measure the event?
• (5)Were confounding factors identified?
• (6)Were strategies to affect confounding factors stated?
• (7)Were the results measured indeed and dependably? And (8) Was the statistical analysis suitable? Studies were considered low risk when they scored 50% and above on the quality assessment indicators, as reported in a supplementary file (Supplementary file 2).

2.6. Risk of bias assessment

Hoy et al. [44] developed a bias assessment tool, which consists of 10 items that assess four domains of bias and internal and external validity, two authors (NA and BB) independently evaluated included articles for risk of bias. Any disagreement raised during the risk of bias assessment was resolved through a discussion led by the third author (NE). Finally, the argument was solved and reached an agreement. The first four items (items 1–4) evaluate the presence of selection bias, non-response bias, and external validity. The other six items (items 5–10) assess the presence of measuring bias, analysis-related bias, and internal validity. Therefore, studies that received 'yes' for eight or more of the ten questions were classified as ‘low risk of bias.’ If studies that received 'yes' for six to seven of the ten questions were classified as ‘moderate risk’ whereas studies that received 'yes' for five or fewer of the ten questions were classified as ‘high risk’ as reported in a supplementary file (Supplementary file 3).

2.7. Data extraction

For data extraction and analysis, STATA version 11 Software and Microsoft Excel spreadsheet from 2016 were used, respectively. A standardized Joanna Briggs Institute data extraction format was used by two authors (NA and KD) to independently extract all pertinent data. The disagreement raised during data extraction was resolved through a discussion led by the third author (LT). Finally, the argument was solved and reached an agreement. The data automation tool was not used due to this study's absence of the paper form (manual data). The name of the first author, year of publication, study region, study setting, study design, prevalence of knowledge, attitude, practice, sample size, and quality of each paper was extracted.

2.8. Data analysis

The data were exported to STATA software version 14 for analysis after being extracted from all pertinent findings in a Microsoft Excel spreadsheet. A 95% confidence interval was used to calculate the pooled prevalence of knowledge, attitude, and practice. The funnel plot and more objective Begg and Egger's regression tests were used to check for publication bias, with a P value of 0.05 indicating potential bias. The Cochrane Q statistic was used to assess whether there was between-study heterogeneity. This study-to-study heterogeneity was measured using the I² statistic, where values of 0, 25, 50, and 75% indicated low, medium, and high levels of heterogeneity, respectively. Using a random-effects model for analysis to determine the overall prevalence of heterogeneity, a forest plot was used to visually assess the presence of heterogeneity. A forest plot was utilized to visually analyze the presence of heterogeneity, which was presented at a high level. By study context, study design, and publication status, subgroup analysis was carried out (Published versus Unpublished). Sensitivity analysis was carried out to determine the impact of a single study on the estimated prevalence from the meta-analysis as a whole. Text, tables, and graphics were used to present the study's results.

3. Results

3.1. Search findings and study characteristics

Using a search strategy, 995 publications about the knowledge, attitude, practice, and factors influencing exclusive breastfeeding in Ethiopia were found using online search engines such PubMed, Scopus, Google Scholar, Science direct, and online research repository home. There were 791 items left after duplicates were excluded. Following a review of the complete titles and abstracts for the remaining 791 papers, 582 studies were removed. As a result, 209 full-text studies were evaluated for eligibility, and 176 papers were further discarded for various reasons from consideration. Finally, 33 articles were included as criteria for this systematic review and meta-analysis study (Fig. 1).

Fig. 1.

PRISMA flow chart displays the article selection process for knowledge, attitude and practice of exclusive Breastfeeding in Ethiopia.

The included studies were published from 2000 to 2021. All included studies employed by cross-sectional study design. Of these, twenty-six studies were community-based, whereas seven were institutional-based cross-sectional studies. Eleven studies were conducted in Southern Nations Nationalists and Peoples Region [29,38,39,[57], [58], [59], [60], [61], [62], [63], [64]], eight in Amhara [[46], [47], [48], [49], [50], [51],56,70], six in Oromia [36,37,45,[65], [66], [67]], three in Addis Ababa [35,52,53], two in Tigray [40,69], two in Afar [54,68] and one in Dire Dawa [55]. The sample sizes ranged from 118 to 819. The prevalence of knowledge, attitude, and practice of exclusive breastfeeding ranged from 22.8 to 94.3, 56.7 to 90, and 14.9 to 96.3, respectively. All studies were assessed by using Joanna Briggs Institute (JBI) quality appraisal checklist and yielded low risk (Table 1).

Table 1.

Characteristics of the included studies in the systematic review and meta-analysis for the prevalence of Knowledge, Attitude and practice of exclusive breastfeeding Ethiopia.

Author/Year	Region	Dioscreptor	Study setting	Study Design	Sample size	Knowledage	Attitude	Practice	quality
Habtamu Gebresenbet/2020[45]	Oromia	Urban	Community	cross-sectional	150	73.94%	63.99	69.69	Low-risk
Kassaw Eeshtu/2015[38]	SNNPR	Rural	Institutional	Cross-sectional	436	NR	56.7	48.2	Low- risk
Bayissa ZB et.al/2015[36]	Oromia	Rural	Community	Cross-sectional	371	90.8	NR	82.2	Low- risk
Admasu and Cione/2016[46]	Amhara	Urban	Community	Cross-sectional	543	87.3	88.95	48.97	Low -risk
Mazengia et.al/2020 [47]	Amhara	Rural	Institutional	Cross-sectional	429	92.1	NR	38.5	Low-risk
Alamirew et.al /2017[70]	Amhara	urban	Institutional	Cross-sectional	384	69.8	76	NR	Low-risk
Ballo T.H/2016[35]	Addis Ababa	urban	Institutional	Cross-sectional	375	22.8	NR	30.5	Low-risk
Teklebrehan Tema/2000 [37]	Oromia	Urban	Community	Cross-sectional	150	87	90	25	Low-risk
Embaye MH /Un-pub[69]	Tigray	Rural	Institutional	Cross-sectional	239	84.4	NR	NR	Low-risk
G.Hadgu et.al/2016[40]	Tigray	Urban	Community	Cross-sectional	253	65.4	63.4	96.3	Low-risk
Tsedek et.al/2014[67]	Oromia	Urban n	Community	Cross-sectional	220	91.8	87.3	43.8	Low-risk
Gelaw Mitiku/Un-pub [56]	Amhara	rural	institutional	Cross-sectional	398	65.1	NR	78.1	Low-risk
Tadele et.al/2016[63]	SNNPR	Urban	Community	Cross-sectional	314	34.7	89.5	26.4	Low-risk
Wana et.al/2017[64]	SNNPR	Urban	Community	Cross-sectional	351	94.3	NR	56.1	Low-risk
Tesfa Getanew Woldie et.al/2014[50]	Amhara	Rural	Community	Cross-sectional	819	NR	NR	47.3	Low-risk
Setegn et.al/2012[65]	Oromo	Rural	Community	Cross-sectional	608	NR	NR	71.3	Low-risk
Kelay T/2017[60]	SNNPR	Rural	Community	Cross-sectional	421	NR	NR	64.8	Low –risk
Gedion Asnake et.al/2019[59]	SNNPR	Urban	Community	Cross-sectional	403	NR	NR	64.8	Low-risk
Arage & H.Gedamu/2016[48]	Amhara	Urban	Community	Cross-sectional	453	NR	NR	70.8	Low-risk
Lewam Mebratu et.al/2020[61]	SNNPR	Rural	Institutional	Cross-sectional	209	NR	NR	81.6	Low-risk
Laykewold Elyas et.al/2017[52]	Addis Ababa	urban	institutional	Cross-sectional	380	NR	NR	44.2	Low-risk
Ermancho B et.al/2021[39]	SNNPR	Rural	Community	Cross-sectional	591	NR	NR	14.9	Low-risk
Bekere A/2014[66]	Oromia	Urban	Community	Cross-sectional	118	NR	NR	72	Low-risk
Reddy S/2016[62]	SNNPR	Rural	Community	Cross-sectional	347	NR	NR	57.6	Low-risk
Abay Sefene et.al/2013[51]	Amhara	Urban	Community	Cross –sectional	159	NR	NR	49.1	Low-risk
Sisay A/2021[58]	SNNP	Rural	Community	Cross-sectional	630	NR	NR	76	Low-risk
Desalew A/2020[55]	Dire dawa	Urban	Community	Cross-sectional	704	NR	NR	81.1	Low-risk
Lenja et.al/2016[29]	SNNPR	Rural	Community	Cross-sectional	396	NR	NR	78	Low-risk
Liben et.al/2016[54]	Afar	Urban	Community	Cross-sectional	333	NR	NR	81.1	Low-risk
Shiferaw et.al/2015[53]	Addis Ababa	Rural	Institutional	Cross-sectional	648	NR	NR	29.1	Low-risk
Anjulo and Haile/2018[57]	SNNPR	Urban	Community	Cross-sectional	300	NR	NR	53.9	Low-risk
Tsgaye et.al/2019[68]	Afar	Rural	Community	Cross-sectional	631	NR	NR	55	Low-risk
Asfaw et.al/2015[49]	Amhara	Rural	Community	Cross-sectional	634	NR	NR	68.6	Low-risk

4. Meta-analysis

4.1. Level of knowledge, attitude, and practice of exclusive breastfeeding in Ethiopia

The pooled prevalence of the knowledge about, attitudes towards, and practice of exclusive breastfeeding in Ethiopia are presented by the forest plots in Fig. 2, Fig. 3, Fig. 4. A random-effect model showed that the pooled good knowledge level was 74.2% (95% CI: 62.9–85.4; I² = 99.2%). The overall estimated positive attitude towards exclusive breastfeeding mother care was 77.2% (95%CI: 68.3–86.0; I² = 947.1%), while the pooled estimate of the poor practice of exclusive breastfeeding was 58.25% (95% CI: 49.92–66.57; I2 = 99.2%).

Fig. 2.

Forest plot of knowledge about exclusive breastfeeding. The height of the diamond is the overall effect size (74.17%), while the width is the 95% confidence interval (62.93–82.41). The y-axis shows the standard error of each study while the x-axis shows the estimate of effect size of the each study. The vertical line denotes the no effect. The box represents the effect size of each study and the line across the box is confidence interval of each study.

Fig. 3.

Forest plot of attitude towards exclusive breastfeeding. The diamond represents the summary point estimate (77.2%) and the horizontal extremity of the diamond is the 95% confidence interval (68.3–86.0). The standard error is plotted at the y-axis and the effect size plotted at x-axis. The squares represent the effect estimate of the individual studies and the horizontal lines indicate the confidence interval; the dimension of the square reflects the weight of each study.

Fig. 4.

Forest plot of practice of exclusive breastfeeding. The height of the diamond is the overall effect size (58.3%), while the width is the 95% confidence interval (49.9–66.6). The y-axis shows the standard error of each study, while the x-axis shows the estimate of the effect size of each study. The vertical line denotes no effect. The square represents the effect size of each study and the line across the square is confidence interval of each study.

4.2. Sub-group analysis

The sub-group analysis based on study region, setting, and publication status showed that good knowledge was found at 86.6% in Oromia, 78.3% in institutional-based studies, and 76.8% in unpublished studies, respectively. For attitude and practice, 82.6% of the participants had a positive attitude toward exclusive breastfeeding, but 34.6% of respondents in Afar showed poor practice toward exclusive breastfeeding. The poor practice was higher among community-based studies (60.6%) compared to institution-based studies (50%) (Supplementary Table 1).

4.3. Leave–one–out sensitivity analysis

A leave-one-out sensitivity analysis was carried out to detect each study's effect on the overall prevalence of a good level of knowledge about, a positive attitude towards, and a poor level of practice of exclusive breastfeeding among postnatal women by excluding one study at a time. The results showed that the excluded study leads to a significant change in the overall prevalence of a good level of knowledge and positive attitude. In the sensitivity analysis, both Ballo T.H et al. and Wana et al. showed an impact on the pooled level of good knowledge, while Teklebrehan Tema and Kassaw Eshetu showed an impact on the level of positive attitude towards exclusive breastfeeding (Supplementary Table 2).

4.4. Publication bias

The presence of publication bias was checked using funnel plot visualization and Egger's and Begg's regression tests (P < 0.05). The Egger and Begg tests both revealed no statistical evidence of publication bias for a good level of knowledge (P = 0.083 and P = 0.240, respectively (supplementary Fig. 1). There was also no statistical evidence of publication bias for a positive attitude in terms of the Egger (P = 0.082) and Begg (p = 0.083) tests (supplementary Fig. 2). The results of the Begg (P = 0.54) and Egger (P = 0.153) tests showed no evidence of statistical evidence of publication bias for the level of poor practice of exclusive breastfeeding (supplementary Fig. 3).

4.5. Factors associated with exclusive breastfeeding in Ethiopia

In this study, women having a vaginal birth, having sufficient knowledge of EBF, health facility delivery, attending antenatal care service, being a housewife in occupation, and having a secondary maternal education level were determinants of exclusive breastfeeding.

The odds of exclusive breastfeeding among vaginally delivered women (AOR = 2.0; 95%CI: 1.4–2.9), I² = 0.0%, P = 0.000) were two times more likely than those delivered by cesarean section. The finding of our study showed that the odds of exclusive breastfeeding among housewives (AOR = 3.1; 95%CI: 2.1–4.7), I² = 4.2%, P = 0.000) were three times more than government-employed women. This study showed that the prevalence of exclusive breastfeeding in the secondary education level of women (AOR = 3.3; 95%CI: 1.8–6.0), I² = 0.0, P = 0.000) was 3.3 times more than women who had informal education. The odds of exclusive breastfeeding who gave birth in a health facility (AOR = 3.3; 95%CI: 2.1–5.1). I² = 14.9%, P = 0.000) were 3.3 times more likely than home delivery. Our finding showed that the prevalence of exclusive breastfeeding among women who attended antenatal care service (AOR = 2.7; 95%CI: 1.7–4.5), I² = 42.6%, P = 0.000) were 2.7 times more likely than in women who had no antenatal care service (Table 2).

Table 2.

Pooled odd ratio for the association of variables with exclusive breastfeeding

Determinants (ref. no)	Number of studies	OR(95%CI)	P-value	I2 (%)
Antenatal care visit [37,39,52,56 ]	4	2.735(1.659-4.509)	0.000	42.6
Health facility delivery [[39], [49]]	2	3.282 (2.125-5.070)	0.000	14.9
Vaginal delivery [[53], [54]]	2	2.016(1,397-2.910)	0.000	0.0
Being housewife [[37], [49], [55]]	3	3.905(2.053-4.666)	0.000	4.2
Secondary level of education [[39], [52]]	2	1.80(1.22-2.66)	0.000	42.6

5. Discussion

This systematic review and meta-analysis study investigated the pooled estimate of knowledge, attitudes, practices (KAP), and determinants of exclusive breastfeeding among Ethiopian women. The key findings were that the national pooled estimate of KAP and exclusive breastfeeding practice were significantly related to the maternal educational status, antenatal care visit, delivery in a health facility, occupation, and vaginal delivery. As a result, the pooled prevalence of good knowledge about exclusive breastfeeding in this study was 74.17% (95%CI: 62.93–85.41), which is consistent with a study done in Nigeria (71.3%) [71]. The findings of the present study are higher than those of studies conducted in Malaysia reported 44%–55%) [72] and Nigeria (31%) [73]. On the other hand, this finding is lower than a systematic review study conducted in East Africa, which estimated knowledge of exclusive breastfeeding to be 84.4%, the right time to start complementary feeding to be 81% [74], and in Bhutan (98%) [75]. The knowledge gaps may explain these differences regarding exclusive breastfeeding among the study participants in each country and the level of awareness and accessibility of the maternal and child health information system.

This study found that the pooled level of positive attitudes towards exclusive breastfeeding was 77.2% (95%CI: 68.320–86.022), which is in line with studies done in Nigeria (75.6%) [76]. The current study has higher findings than the study in Nigeria (54%) [71]. By contrast, this study's finding is lower than the systematic review and meta-analysis study done in East Africa, in which the preference for exclusively breastfeeding a baby for the first six months was 91.61% [72]. The reason could be when the study was done and inadequate knowledge and awareness of the benefits of exclusive breastfeeding during the first six months.

The pooled prevalence of exclusive breastfeeding in Ethiopia was 58.25% (95%CI: 49.92–66.57) in the current study. This finding is congruent with the previous meta-analysis studies done in Ethiopia (59.3% and 60.42%) [77,78]. The possible explanation for the above similarity could be due to socio-demographics, methodologies, and the character of each study included in both reviews with the present study.

The result of our study is higher than a systematic review and meta-analysis study done in Iran (49.1%) [79], 27 Sub-Saharan African countries (36%) [80], Demographic and Health Survey of Madagascar (48.8%) [81], in Tanzania (22.9%) [82] and in Ghana (27.7%) [83]. On the contrary, the finding of the present study is lower than the national prevalence of Peru (68.4%) [84] and Nepal (66.3%) [84]. The discrepancy could be due to methodological differences such as data analysis and sampling of study participants, variation in infant and maternal socio-demographic characteristics, and health service utilization.

In our study, mothers who had antenatal care visits were 2.7 times more likely to practice exclusive breastfeeding practice than women who hadn't received antenatal care services. This finding is in agreement with the studies conducted in Ethiopia [78] and Rawalpindi [85]. This could be that since women received antenatal care services obtaining good nutritional advice from health providers, intern positively affects exclusive breastfeeding.

Women who gave birth at health institutions were 3.3 times more likely to practice exclusive breastfeeding than women who gave birth at home. This is in line with studies done in Ethiopia [78] and Tanzania [86]. This might be due to women delivered at health facilities having a golden opportunity to receive postnatal counseling.

Our findings showed that women who gave birth through the vaginal were two times more likely to practice exclusive breastfeeding than women who gave birth through cesarean section. This is congruent with a meta-analysis study done in Iran [79]. This might be because women delivered through cesarean section associated with uterine rupture; bleeding, infection, and more extended hospitalization reduce exclusive breastfeeding rates, as supported by a previous meta-analysis study [87].

Women working as housewives were three times more likely to practice exclusive breastfeeding than employed mothers. This finding is similar to a meta-analysis study done in Iran [79]. This is justified by employed mothers suffering from lack of time and fatigue and may get difficulties in exclusive breastfeeding [88]. Lastly, congruent with a pooled demography and health survey study done in nine Sub-Saharan African countries [89], our meta-analysis showed that women with secondary educational status were 3.3 more likely to practice exclusive breastfeeding than women with no formal education. This is because as the educational status of women increases, maternal and child health concerns increase, which in turn affects exclusive breastfeeding positively.

This study utilized a random-effect model to manage a significant variation that resulted in between-study heterogeneity. We assessed leave-one-out sensitivity, and the results show that every study had a significant impact on the pooled good level of knowledge, positive attitude, and poor level of exclusive breastfeeding. We assessed the possible variability source via sub-group analysis using the study settings, regions, and publications (published versus unpublished studies). The high heterogeneity might be due to differences in the sample populations, paper qualities, or socio-cultural, ethnic, and regional differences.

5.1. Limitations and strengths

This study has some limitations, just as other meta-analyses. First, the study protocol was not registered in the prospective international register of systematic reviews (PROSPERO). Second, only English-language published articles were included. Third, in this survey, 70% were used as a threshold for knowledge, attitude, and practice. A KAP survey can be used to measure progress towards acquiring these specific concepts, but it cannot be used to investigate the culturally specific knowledge that communities have (such as indigenous knowledge about food systems and various definitions of nutrition-related issues). For lengthy periods, attitudes change. According to the circumstances of the interview, responses may vary. It is impossible to rely on attitudes to represent consistent opinions. Moreover, Rather than using objective measurements, a KAP survey relies on self-reported comments. The surveyor's abilities as well as the respondent's judgment, cooperation, and recall can affect the results. What is said and what is done may not always line up. As a result, caution should be taken while interpreting our findings. Fourth, all of the included studies were cross-sectional, which could cause other confounding factors to alter the outcome variable. Finally, our meta-analysis only included papers that were published in seven distinct regions across the country. Due to the small number of papers included, the areas might therefore be underrepresented.

There is also some strength. The study's findings are significant to address the EBF gaps and carefully provide evidence for the need for immediate adjustments in certain areas. Pediatricians and other healthcare professionals can utilize these findings as an initial point to fill in these gaps. To improve EBF practice, the workforce should encourage mothers to attend antenatal care appointments. It also demonstrates the significance of educational strategies in enhancing and correcting mothers' knowledge, attitudes, and practices of exclusive breastfeeding, as these practices negatively affect developing children by exposing them to malnutrition and refusing them access to all of the breastfeeding's advantages.

6. Conclusion

In conclusion, our study demonstrates that the prevalence of knowledge and attitude toward exclusive breastfeeding was good, but there is a gap in practice. Besides, the pooled prevalence of knowledge, attitudes, and practices differed based on the study settings, regions, and publications status. Maternal educational status, antenatal care service, health facility delivery, being a housewife, and vaginal births were predictors of exclusive breastfeeding. Accordingly, it is better to increase the quality of antenatal care service, institutional delivery, and promotion of vaginal birth among reproductive-age women.

Availability of data and materials

All relevant data are within the Manuscript and its Supporting Information files.

Ethics approval and consent to participants

Not applicable because no primary data were collected.

Authors' contributions

NA conceptualized the study: NA, KD, and LT contributed during data extraction and analysis: NA, GB, and NE wrote result interpretation: NA, NA, and NE Prepared the first draft: NA, KD, and LT contributed during the conceptualization and interpretation of results and substantial revision: NA, KD, NE, GB, NA, and LT. Revised and finalized the final draft manuscript. All the authors read and approved the final version of the manuscript.

Funding

The authors received no specific funding for this work.

Consent for publication

Not applicable.

Declaration of competing interest

The authors declared that they have no competing interests

Appendix A. Supplementary data

The following is/are the supplementary data to this article: