Hossain and Mihrshahi  International Breastfeeding

International Breastfeeding Journal (2024) 19:35

https://doi.org/10.1186/s13006-024-00644-x Journal

RESEARCH Open Access

Effect of exclusive breastfeeding and other

infant and young child feeding practices
on childhood morbidity outcomes: associations
for infants 0–6 months in 5 South Asian
countries using Demographic and Health
Survey data
Saldana Hossain1* and Seema Mihrshahi1*

Abstract
Background Despite growing evidence of the impacts of exclusively breastfeeding infants during the first 6 months
of life on preventing childhood infections and ensuring optimal health, only a small number of studies have quanti-
fied this association in South Asia.
Methods We analyzed data from the Demographic and Health Surveys in Afghanistan (2015; n = 3462), Bangla-
desh (2017–2018; n = 1084), India (2019–2021; n = 26,101), Nepal (2022; n = 581), and Pakistan (2017–2018; n = 1,306),
including babies aged 0–6 months. Multivariate logistic regression models were used to determine the association
between exclusive breastfeeding in the last 24 h and diarrhoea, acute respiratory infections, and fever in the two
weeks before the survey. We also examined the association between other infant and young feeding indicators
and these outcomes.
Results Infants who were exclusive breastfed had decreased odds of diarrhoea in Afghanistan (AOR: 0.49, 95% CI
0.35, 0.70), India (AOR: 0.80, 95% CI 0.70, 0.91), and Nepal (AOR: 0.42, 95% CI 0.20, 0.89). Compared with infants who
were not exclusive breastfed, infants who were exclusively breastfed were less likely to have fever in Afghanistan (AOR:
0.36, 95% CI 0.26, 0.50) and India (AOR: 0.75, 95% CI 0.67, 0.84). Exclusive breastfeeding was associated with lower odds
of acute respiratory infections in Afghanistan (AOR: 0.57, 95% CI 0.39, 0.83). Early initiation of breastfeeding was pro-
tective against diarrhoea in India. Bottle feeding was a risk factor for diarrhoea in India and for fever in Afghanistan
and India. Bottle feeding was also a risk factor for acute respiratory infection in Afghanistan and India.
Conclusions Not exclusive breastfeeding is a risk factor for diarrhoea, acute respiratory infections, and fever in some
South Asian countries. These findings could have substantial implications for global and national efforts to increase

*Correspondence:
Saldana Hossain
[email protected]
Seema Mihrshahi
[email protected]
Full list of author information is available at the end of the article

© The Author(s) 2024. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which
permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the
original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or
other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line
to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory
regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this
licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecom-
mons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.
Hossain and Mihrshahi International Breastfeeding Journal (2024) 19:35 Page 2 of 18

exclusive breastfeeding rates. More support, advocacy, and action are required to boost breastfeeding rates as a cru-
cial public health measure.
Keywords Exclusive breastfeeding, South Asia, Diarrhoea, Acute respiratory infection, Fever

Background advantages of optimum breastfeeding, limited improve-

According to United Nations Children’s Fund (UNICEF), ment was seen in EBF rates between and across coun-
about 9% of all childhood deaths under the age of five tries/regions and only 44% of babies worldwide aged 0
globally in 2019 were due to diarrhoea, making it one of to 6 months were EBF from 2015 to 2020 [13]. World-
the major causes of mortality in children [1]. More than wide, breastfeeding rates are still below what is neces-
12 million children under the age of five are admitted to sary to protect children’s health. In 2012, the World
hospitals every year as a result of acute respiratory tract Health Assembly set a goal of achieving a global EBF rate
infections (ARI) [2]. Among them, acute lower respira- of at least 50% by the year 2025 [9]. In order to achieve
tory infections (ALRIs) which includes pneumonia and this target, countries need to strengthen their efforts to
bronchiolitis have emerged as one of the primary reasons increase breastfeeding rates.
for pediatric hospital admissions and in-hospital mortal- Sub-Saharan Africa and South Asia had the highest
ity of young children, particularly in low-income nations neonatal mortality rates estimated to be 27 and 24 deaths
[3]. per 1,000 live births, respectively, in 2020 [14]. UNICEF
In recent years, an increasing number of research has also reported that babies born in South Asia had a
articles have been published on the benefits of optimal nine-fold higher mortality rate than those born in high-
breastfeeding practices for the mother-infant pair [4–6]. income countries [14]. South Asia and sub-Saharan
Breastmilk has many anti-inflammatory and immuno- Africa account for the majority of deaths from diarrhoea
logical properties that safeguard babies against a range among children under the age of five [15]. Almost 90% of
of diseases [4, 7]. Exclusive breastfeeding (EBF) offers diarrhoeal and pneumonia mortality occurred in South
several known short- and long-term advantages, espe- Asia and sub-Saharan Africa [15, 16]. Due to inadequate
cially in lowering childhood morbidity and death from breastfeeding practices, insufficient access to immuniza-
respiratory and diarrhoeal illnesses [5, 6]. Findings from a tion, unsafe water and unimproved sanitation facilities,
cohort study [8] conducted in eight different low-income and lack of access to treatment, the effect of diarrhoea-
countries of 1731 babies found that EBF was protective associated morbidity and mortality on infant survival is
against the incidence of diarrhoea and ALRI. Breast- greatest among children from low-and middle-income
feeding has also shown to contribute to increased intel- countries (LMICs) [16, 17].
ligence scores and academic performance, lower risk of Rates of EBF have increased slightly around the world,
diabetes and obesity in the long term, and a decreased with South Asia being the main driver of this change
risk of breast and ovarian cancer in mothers [9]. EBF [18]. Between 2000 and 2015, this region’s rate of EBF for
for the first six months of a baby’s life is recommended babies aged 0–5 months rose from 47 to 64 percent, an
by the World Health Organization (WHO) in its policy increase of 17 percentage points [18]. Despite improve-
guidelines in order to ensure the baby’s adequate growth, ments on feeding practices in South Asia, there is sub-
development, and health [10]. At six months, the WHO stantial variation in breastfeeding practices across and
recommends introducing nutritionally rich, safe, and within countries [19]. Breastfeeding practices vary across
appropriate complementary foods while continuing to Bangladesh, India, Afghanistan, Pakistan, and Nepal,
breastfeed for at least two years [10]. with breastfeeding initiation rates ranging from 19.6% in
Approximately 18% of acute respiratory deaths, 30% of Pakistan to 59.8% in Bangladesh and EBF rates for infants
diarrhoeal deaths among children under the age of 5, and aged 0–5 months remain suboptimal, ranging from 43.3%
45% of newborn deaths from infections are attributed to in Afghanistan to 63.7% in India [20–24]. Bottle feeding
inadequate breastfeeding practices [11]. Also, around 1.4 is prevalent in children under 2, with rates as high as 48%
million under-five deaths and 10% of the disease burden in Pakistan [23]. Since 2000, India, Pakistan and Bang-
in children are attributed to sub-optimal breastfeeding, ladesh, among 15 consistently affected countries, have
particularly non-exclusive breastfeeding during the first borne a significant share of global pneumonia and diar-
six months of life [12]. According to evidence on the rhoea deaths in children under 5, with over two-thirds of
efficacy of interventions, attaining universal coverage of the mortality burden concentrated in these 15 countries,
appropriate breastfeeding may avoid 13% of deaths in highlighting the urgent need for interventions in these
children under five years old worldwide [12]. Despite the regions [25]. Socio-cultural factors such as maternal
Hossain and Mihrshahi International Breastfeeding Journal (2024) 19:35 Page 3 of 18

education, husband’s education, working status, wealth, was used. In stage one, each nation was categorized into
and family structure influence breastfeeding initiation regions, which could be either geographical regions or
practices, with variations observed across Bangladesh, political areas like provinces or states and named north,
India, Nepal, and Pakistan [26]. Evidence indicates that south, east, and west or political [30]. Populations in
infant characteristics (small infant size), healthcare fac- these subnational regions were divided into rural and
tors (poor use of antenatal services, home delivery, and urban areas of residence. A random selection of enu-
caesarean delivery), and sociocultural norms such as meration regions from the latest population census were
prelacteal feeding are all barriers to EBF [19, 27, 28]. A chosen from within these stratified zones. The likelihood
better understanding of this variability is needed to guide of each main sampling unit/cluster being chosen was set
policies and interventions to achieve the global aim of the such that it corresponded to the percentage that each
World Health Assembly on EBF by 2025 [9, 19]. cluster’s population contributed to the overall population
Several recent studies have explored the relationship [30]. In stage two, every house within the cluster were
between infant and young child feeding practices and listed, and through equal-probability systematic selec-
infant morbidity outcomes using data from the Demo- tion, an average of 25 households in a cluster were ran-
graphic and Health Surveys (DHS). For instance, a study domly chosen for an interview. Further details on tools
by Ogbo et al. [29] pooled DHS data from multiple sub- of data collection, DHS sampling techniques and house
Saharan African countries. While these studies have listing methods is published in country-specific DHS
provided valuable insights, there is still a need to ana- reports [30, 31].
lyze these relationships in diverse settings and popula- The survey data were comparable across countries and
tions. Many studies have focused on individual countries survey years due to the use of standardized measurement
or regions, limiting the generalizability of their findings. techniques and survey instruments [31]. Women aged
There is still a critical gap in the literature in South Asian 15–49 years were given a questionnaire to obtain data
countries with high diarrhoea and ARI morbidity using on household demographics, maternal and child health.
nationwide population-based data. As these populations We used data from the youngest living infants aged less
have one of the highest disease burden in the region, it is than 7 months who were living with their mother aged
crucial to examine these countries as interventions would 15–49 years. Participants with missing data on variables
have the greatest effect here [29]. Furthermore, compari- such as exclusive breastfeeding, diarrhoea, acute respira-
son of countries has the benefit of illustrating how health tory infection, and fever were excluded.
status and characteristics differs among nations in a
region to develop locally acceptable and context-specific
infant feeding-related programmes and policies [27, 29]. Outcome and exposure variables
The present study seeks to build upon existing literature The outcomes of interest were morbidity status of babies,
by providing current evidence using the latest DHS data which were measured based on any of the three morbidi-
for Afghanistan (2015), Bangladesh (2017–2018), India ties: diarrhoea, ARI, or fever. Each of these variables were
(2019–2021), Nepal (2022), and Pakistan (2017–2018) coded to indicate whether the outcome was present or
[20–24]. We assessed the relationship between EBF and not (“Yes” vs “No”). The IYCF indicators were the study’s
diarrhoea, ARI, and fever. In addition, we explored the explanatory factors, measured according to the WHO’s
association between other infant and young child feeding definitions for assessing IYCF practices [32]. The primary
(IYCF) indicators and these health outcomes. exposure variable was exclusive breastfeeding. The spe-
cific definitions of all variables are described in Table 1.

Methods
Data source, data collection, and study population Covariates
We did a secondary analysis using cross-sectional data Based on existing literature, we identified potential
from the most recent Demographic and Health Surveys confounders and other determinants of breastfeeding
(DHS) for the countries: Afghanistan (2015), Bangladesh practices in South Asia [27, 33–36]. Potential confound-
(2017–2018), India (2019–2021), Nepal (2022), and Paki- ing factors were broadly classified into individual-level
stan (2017–2018). The survey collects data on health of characteristics (infant age, mother’s age, gender), socio-
mothers and their children, nutritional outcomes, fertil- economic characteristics (maternal education, work-
ity, HIV/AIDS, immunization, and family planning (The ing status of mother), health service factors (frequency
DHS program). Within all countries, DHS follow the of antenatal care (ANC) visits, place of delivery), and
same sampling procedures to generate nationally and household-level characteristics (household wealth index,
sub-nationally (region-wide/statewide) representative place of residence, number of listed household members,
data [30]. A stratified two-stage random sampling design source of drinking water, toilet facility, cooking fuel).
Hossain and Mihrshahi International Breastfeeding Journal (2024) 19:35 Page 4 of 18

Table 1 Measurement of exposure and outcomes variables

Variable Definition

Exposure variable
EBF The percentage of babies who were fed breast milk as the only source of nutrition but allowed drops or syrups of vitamins, medi-
cines, and oral rehydration solution
To determine EBF status of babies from birth to 6 months of age, mothers were asked (i) whether the baby was still being breast-
fed; (ii) the duration of breastfeeding; and (iii) whether they had given the baby any other milk, liquids, solids, powdered milk, soft
foods, or water in the 24 h preceding the interview
Early initiation The percentage of babies within 0–6 months of age who commenced breastfeeding within 1 h of birth
of breastfeeding
Bottle feeding The percentage of babies 0–6 months of age who were fed any liquid (including breast milk) or semi-solid food from a bottle
with nipple/teat during the last 24 h
Outcome variable
Diarrhoea Defined as the passage of three or more loose or liquid stools per day. This was based on maternal recall of symptoms of the infant
in the last 2 weeks preceding the survey
ARI Mothers were asked if the infant had a cough accompanied by short, rapid breasting that was chest-related and/or have difficulty
breathing which was chest-related during the 2 weeks before the survey
Fever Mothers were asked if the infant had been ill with a fever at any time during the last 2 weeks

Type of cooking fuel was used in the regression mod- Given the dichotomous nature of the outcome variables,
els for ARI, as some papers in South Asia reported the multiple binary logistic regression models were fitted to
use of solid fuel as a risk factor for ARI [37, 38]. Solid examine the association between each exposure of inter-
fuels included charcoal, animal dung, wood, straw/ est, (1) EBF (2) early initiation of breastfeeding and (3)
grass/shrubs, agricultural crop, and kerosene, whereas bottle feeding with diarrhoea, ARI, and fever (separate
clean fuels included natural gas, electricity, biogas, and models for each outcome). All variables were entered into
liquefied petroleum gas (LPG). The type of toilet facil- the initial model and variables with a P value of < 0.10 or
ity and the source of drinking water were classified as considered to be conceptually relevant potential con-
“improved” or “unimproved” in accordance with the founders regardless of P value were retained in the final
WHO and UNICEF Joint Monitoring Programme (JMP) multivariable logistic regression model. Unadjusted and
for Water and Sanitation [39]. Type of toilet facility was adjusted odds ratios (AOR) with 95% confidence intervals
classified as ‘improved’ (included flush toilets piped to (CIs) were obtained from the regression models with sig-
the sewer system, pit latrine or septic tank, flush to an nificance at the 5% level (p < 0.05). Full regression models
unknown location, ventilated improved pit (VIP) latrine, with unadjusted ratios are detailed in Additional file 1.
composting toilet, pit latrine with slab). ‘Unimproved’ Variance inflation factor (VIF) was utilized to assess mul-
type of facility included flush not piped to sewer, hang- ticollinearity between the variables prior to running the
ing toilet/latrine, pit latrine without slab/open pit, bucket models. The VIF test indicated a lack of high multicollin-
toilet and no facility/field/bush. `Improved’ sources of earity among the variables (VIF less than five). Multiple
water were defined as a piped water into dwelling/yard/ logistic regression models were fitted to the data sepa-
plot/neighbor, public tap or standpipe, tube-well or bore- rately for each country. We applied the ‘svy’ command in
hole, protected spring and well, rainwater, cart with small all our analyses to allow for the cluster sampling design
tank, tanker truck, bottled water, filtration plant, while of DHS surveys. The forest plot of AORs and 95% CIs for
households that utilized unprotected spring and well, and each country separately as well as the unweighted over-
surface water were classified as ‘unimproved’. all estimates for all five countries combined were created
using STATA’s “metan” function.
Statistical analysis
All analyses were carried out using STATA V.17.0 (Stata- Results
Corp). Descriptive statistics of the study sample were Characteristics of the included surveys
computed as frequencies with weighted percentages for The characteristics of the women and children included
the explanatory, outcome, and control variables. Bivari- in the surveys are summarized in Table 2. The total sam-
ate analyses using Pearson’s chi-squared test was used to ple size ranged from 603 in Nepal to 26,219 in India. The
compare the prevalence of EBF among the levels of each mean age for women in the survey ranged from 24 to
covariate. All sample sizes and proportions were based 27 years old. Bangladesh had a larger number of young
on sampling weights to account for the survey design. mothers (23.3%) who were between 15- to 19-year-old
Hossain and Mihrshahi International Breastfeeding Journal (2024) 19:35 Page 5 of 18

Table 2 Selected infant, maternal, household, and IYCF characteristics among five countries
Afghanistan (2015) Bangladesh India (2019/21) Nepal (2022) Pakistan (2017/18)
(n = 3573) (2017/18) (n = 26219) (n = 603) (n = 1312)
n (%) (n = 1087) n (%) n (%) n (%)
n (%)

Infant characteristics
Infant age (months)
  0–2 1324 (37.1) 521 (47.9) 11,034 (42.1) 257 (42.6) 565 (43.1)
  3–4 1253 (35.1) 297 (27.3) 7637 (29.1) 164 (27.2) 379 (28.9)
  5–6 996 (27.9) 270 (24.8) 7547 (28.8) 182 (30.2) 368 (28.0)
Gender
  Female 1701 (47.6) 514 (47.3) 12,769 (48.7) 294 (48.8) 707 (53.9)
  Male 1872 (52.4) 573 (52.8) 13,449 (51.3) 309 (51.2) 605 (46.1)
Mother’s characteristics
Mother’s age in years
  15–19 287 (8.0) 253 (23.3) 2046 (7.8) 97 (16.0) 103 (7.8)
  20–29 2191 (61.3) 641 (59.0) 19,790 (75.5) 398 (66.1) 769 (58.6)
  30–39 904 (25.3) 184 (17.0) 4204 (16.0) 104 (17.3) 402 (30.7)
  40–49 190 (5.3) 9 (0.8) 178 (0.7) 4 (0.6) 38 (2.9)
  Mean 27.0 24.1 25.1 24.6 27.2
Maternal education
  No education 2890 (80.9) 69 (6.4) 4780 (18.2) 95 (15.7) 614 (46.8)
  Primary 325 (9.1) 305 (28.1) 2935 (11.2) 203 (33.6) 223 (17.0)
  Secondary 291 (8.2) 523 (48.1) 13,706 (52.3) 281 (46.6) 276 (21.0)
  Higher 67 (1.9) 190 (17.5) 4798 (18.3) 25 (4.1) 200 (15.2)
Mother’s employment
  Working 363 (10.2) 307 (28.2) 371 (9.2) 183 (30.4) 123 (9.4)
  Not working 3205 (89.7) 780 (71.8) 3646 (90.8) 420 (69.6) 1189 (90.6)
Socioeconomic characteristics
Type of residence
  Rural 2698 (75.5) 792 (72.8) 19,768 (75.4) 191 (31.6) 888 (67.6)
  Urban 875 (24.5) 295 (27.2) 6450 (24.6) 412 (68.4) 425 (32.4)
Household size (members)
  1–4 284 (8.0) 258 (23.7) 5176 (19.7) 132 (21.9) 144 (11.0)
  5–8 1293 (36.2) 646 (59.4) 15,926 (60.8) 360 (59.7) 533 (40.6)
  9 + 1996 (55.9) 183 (16.9) 5116 (19.5) 111 (18.5) 636 (48.4)
Wealth index
  Poorest 621 (17.4) 222 (20.4) 6292 (24.0) 119 (19.7) 300 (22.8)
  Poorer 743 (20.8) 215 (19.8) 5789 (22.1) 129 (21.3) 226 (17.2)
  Middle 818 (22.9) 226 (20.8) 5145 (19.6) 119 (19.8) 285 (21.7)
  Richer 701 (19.6) 220 (20.2) 4904 (18.7) 122 (20.3) 257 (19.6)
  Richest 690 (19.3) 204 (18.7) 4090 (15.6) 114 (18.9) 245 (18.7)
Source of drinking water
  Improved 2638 (73.9) 849 (78.1) 22,124 (84.4) 484 (80.3) 1180 (89.9)
Unimproveda
  ­ 896 (25.1) 12 (1.1) 990 (3.8) 6 (1.0) 54 (4.1)
Type of toilet facility
  Improved 1033 (28.9) 562 (51.7) 16,961 (64.7) 445 (73.7) 976 (74.3)
Unimprovedb
  ­ 2501 (70.0) 299 (27.5) 6153 (23.5) 46 (7.6) 258 (19.6)
Type of cooking fuel
  Clean 1112 (31.2) 191 (17.6) 11,069 (42.2) 195 (32.3) 527 (40.1)
  Solid 2408 (67.5) 666 (61.4) 12,017 (45.8) 296 (49.0) 706 (53.8)
Hossain and Mihrshahi International Breastfeeding Journal (2024) 19:35 Page 6 of 18

Table 2 (continued)
Afghanistan (2015) Bangladesh India (2019/21) Nepal (2022) Pakistan (2017/18)
(n = 3573) (2017/18) (n = 26219) (n = 603) (n = 1312)
n (%) (n = 1087) n (%) n (%) n (%)
n (%)

Health service use

Number of ANC visits
  None 1219 (34.2) 98 (9.0) 1363 (5.2) 20 (3.4) 127 (9.7)
  1–3 1617 (45.3) 488 (44.9) 9351 (35.7) 88.8 (14.7) 470 (35.8)
   4 or more 690 (19.4) 501 (46.1) 15,196 (58.0) 494 (81.9) 715 (54.5)
Place of delivery
  Health facility 2063 (57.8) 557 (51.3) 23,760 (90.6) 497 (82.5) 930 (71.0)
  Home 1492 (41.8) 524 (48.2)c 2407 (9.2) 94 (15.6) 378 (28.8)
IYCF indicators
Exclusive breastfeeding
  Yes 1405 (39.3) 635 (58.5) 15,326 (58.5) 306 (50.8) 556 (42.4)
  No 2100 (58.8) 451 (41.5) 10,836 (41.3) 280 (46.4) 753 (57.4)
Early initiation of breastfeeding
  Yes 1604 (44.9) 614 (56.5) 10,748 (41.0) 329 (54.5) 272 (20.7)
  No 1968 (55.1) 473 (43.5) 15,470 (59.0) 274 (45.5) 1040 (79.3)
Bottle feeding
  Yes 704 (19.7) 155 (14.2) 3433 (13.1) 107 (17.8) 507 (38.6)
  No 2844 (79.6) 930 (85.5) 22,780 (86.9) 495 (82.2) 805 (61.3)
Diarrhoea
  No 2767 (77.6) 1054 (97.0) 23,979 (91.5) 523 (86.7) 1013 (77.2)
  Yes 786 (22.0) 33 (3.0) 2219 (8.5) 78 (13.0) 299 (22.8)
Acute respiratory infection
  No 3184 (89.1) 1047 (96.3) 25,351 (96.7) 589 (97.7) 1128 (86.0)
  Yes 371 (10.4) 38 (3.5) 819 (3.1) 11 (1.8) 181 (13.8)
Fever
  No 2685 (75.3) 758 (69.7) 23,266 (88.7) 494 (81.9) 851 (64.9)
  Yes 868 (24.3) 329 (30.3) 2940 (11.2) 109 (18.1) 460 (35.1)
a
Unimproved drinking water sources includes unprotected spring and well, and surface water, and other sources
b
Unimproved toilet facilities include flush not piped to sewer, hanging toilet/latrine, pit latrine without slab/open pit, bucket toilet and no facility/field/bush and
other types
c
Delivery huts in Pakistan DHS are categorized as missing as they not considered home or facility births

compared to other countries. Meanwhile, Afghanistan relative to the other countries. More than half of women
had a larger number of mothers (5.3%) who were between lived in houses with improved drinking water source and
40 to 49 years old compared to other countries. Afghani- toilet facility in Bangladesh, India, Pakistan, and Nepal.
stan had the greatest number of women (80.9%) who did Also, Nepal (3.4%) and India (5.2%) had the lowest pro-
not attain any schooling. Across all surveys, the major- portion of mothers who did not have any antenatal care,
ity of the women (more than 69%) did not work. Nepal whereas Afghanistan had the highest (34.2%). More than
had the highest number of women living in an urban 50% of women in these South Asian countries delivered
area (68.4%), while most women in the other countries their babies in a health facility. India had the greatest
resided in rural areas. Approximately half of the women number of health facility deliveries (90.6%), while Bangla-
in Afghanistan (55.9%) and Pakistan (48.4%) had more desh had the lowest (51.3%).
than 9 people living in the house. Most women (approxi- As expected, the percentage of babies who were EBF
mately 81.3%) resided in households that used improved declined with infant age in all countries (Fig. 1) and the
sources of drinking water in South Asia. Afghanistan had prevalence of EBF was the greatest during the first two
the highest proportion of unimproved drinking water months of life, across all countries. Afghanistan had the
sources (25.1%) and unimproved sanitation facility (70%) lowest number (39.3%) of babies who were EBF relative
Hossain and Mihrshahi International Breastfeeding Journal (2024) 19:35 Page 7 of 18

Fig. 1 Prevalence of babies aged 0–6 months exclusively breastfed by age group in South Asian countries. Error bars represent 95% confidence
intervals (CI). EBF: Exclusive breastfeeding

to Bangladesh, India, and Nepal where more than 50% of fever was lower in babies who were EBF (24.9%) com-
babies were EBF. With respect to initiation of breastfeed- pared to babies who were not EBF (38.2%). However,
ing, only one fifth (20.7%) of the babies in Pakistan were there was no difference in diarrhoea and ARI prevalence
breastfed within the first hour of birth, the lowest out of among babies who were EBF and those who were not
5 South Asian countries. Furthermore, Pakistan also had EBF. In Afghanistan, diarrhoea (13.1% vs. 28.1%), ARI
the highest proportion (38.6%) of babies who were bot- (6.8% vs. 13.2%), and fever (12.6% vs. 32.3%) were less
tle fed. Pakistan also had the highest proportion of babies prevalent in babies who were EBF compared to those
experiencing diarrhoeal episodes (22.8%), ARI (13.8%), were not. Lower prevalence estimates of diarrhoea were
and fever (35.1%) in the 2 weeks before the survey, com- also observed in infants in Nepal who were EBF (7.7%),
pared to the other countries. compared to those who were not EBF (17.8%). In Paki-
stan, there was no significant difference in diarrhoea,
Prevalence of morbidity outcomes by EBF ARI, and fever prevalence among babies who were EBF
In India, the prevalence of diarrhoea in the 2 weeks and those who were not EBF (See Fig. 2).
before the survey was lower in babies who were EBF
(7.4%) compared to babies who were not EBF (10%) Association of EBF and morbidity outcomes
(See Fig. 2). Similarly, the prevalence of fever was lower In the multivariable analysis, EBF was associated with
in babies who were EBF (9.4%) compared to babies who decreased odds of having diarrhoeal disease in the
were not EBF (13.7%). There was no difference in ARI 2 weeks before the survey among babies aged 0–6 months
prevalence among babies who were EBF and those who in India (AOR: 0.80, 95% CI 0.70, 0.91), Afghanistan
were not EBF in India. In Bangladesh, the prevalence of (AOR: 0.49, 95% CI 0.35, 0.70), and Nepal (AOR: 0.42,
Hossain and Mihrshahi International Breastfeeding Journal (2024) 19:35 Page 8 of 18

Fig. 2 Prevalence of babies aged 0–6 months with morbidity outcomes (diarrhoea, acute respiratory infections, and fever) in the 2 weeks preceding
the survey by exclusive breastfeeding status in South Asian countries. Error bars represent 95% confidence intervals (CI). ARI: Acute respiratory
infection

95% CI 0.20, 0.89) compared to those who were not EBF (AOR: 0.79, 95% CI 0.69, 0.90) (Table 5). We also found
(Table 3). Infants in Afghanistan who were EBF were also that infants who were bottle fed were 1.48 times more
less likely to experience ARI compared to those who were likely to experience diarrhoea in India (AOR: 1.48, 95%
not EBF (AOR: 0.57, 95% CI 0.39, 0.83) (Table 3). EBF was CI 1.26, 1.75) compared to their counterparts (Table 5).
associated with lower odds of fever among Indian (AOR: Similarly, babies who were bottle fed were more likely
0.75, 95% CI 0.67, 0.84) and Afghani infants (AOR: 0.36, to have ARI in the 2 weeks before the survey relative to
95% CI 0.26, 0.50) (Table 4) Similar direction of effects those who were not bottle fed in India (AOR: 1.38, 95%
are also observed in Pakistan and Bangladesh, however CI 1.09, 1.75) and Afghanistan (AOR: 2.53, 95% CI 1.57,
the results are non-significant. These results are illus- 3.53) (Table 5). Infants who were bottle fed were 1.37 and
trated in Fig. 3. 2.09 times more likely to have fever in India (AOR: 1.37,
95% CI 1.17, 1.59) and Afghanistan (AOR: 2.09, 95% CI
Association of other IYCF indicators and morbidity 1.49, 2.91) compared to their counterparts (Table 6).
outcomes
In India, babies who received breastfed within the first Discussion
hour of birth had lower odds of experiencing diarrhoea Main findings
in the 2 weeks before the survey compared to those who Based on our analysis of nationally representative data-
did not receive breastfed within the first hour of birth sets in five South Asian countries, EBF is a protective
Table 3 Association of exclusive breastfeeding and various individual, household-level, and socio-demographic factors with diarrhea and acute respiratory ­infectiona in South
Asia
Diarrhoea in the last 2 weeks Acute respiratory infection in the last 2 weeks

Afghanistan Bangladesh India Nepal (2022) Pakistan Afghanistan Bangladesh India Nepal (2022)c Pakistan
(2015) (2017/18) (2019/21) (n = 581) (2017/18) (2015)b (2017/18) (2019/21) (n = 581) (2017/18)
(n = 3462) (n = 1084) (n = 26101) (n = 1306) (n = 3467) (n = 1084) (n = 26101) (n = 1306)
Variables AOR (95% CI) AOR (95% CI) AOR (95% CI) AOR (95% CI) AOR (95% CI) AOR (95% CI) AOR (95% CI) AOR (95% CI) AOR (95% CI) AOR (95% CI)

Infant age (months)

0–2 Ref Ref Ref Ref Ref Ref Ref Ref Ref
3–4 2.54 (1.70, 3.55 (1.05, 1.61 (1.37, 1.30 (0.65, 2.61) 1.19 (0.77, 1.85) 1.69 (1.02, 2.83)* 2.11 (0.78, 5.68) 1.49 (1.18, 1.89)* 2.12 (1.25, 3.59)*
3.79)** 12.05)* 1.88)**
5–6 2.95 (2.11, 4.02 (1.13, 1.81 (1.55, 1.56 (0.69, 3.56) 1.80 (1.10, 2.93)* 2.18 (1.34, 3.56)* 3.90 (1.39, 1.69 (1.32, 2.46 (1.41, 4.28)*
4.11)** 14.32)* 2.13)** 10.91)* 2.15)**
Gender
Hossain and Mihrshahi International Breastfeeding Journal

Male Ref Ref

Female 0.86 (0.76, 0.97)* 0.43 (0.20, 0.94)*
Mother’s age in years
15–19 Ref Ref Ref Ref
(2024) 19:35

20–29 0.75 (0.61, 0.93)* 0.78 (0.45, 1.35) 1.83 (0.90, 3.74) 0.80 (0.56, 1.15)
30–39 0.54 (0.42, 0.50 (0.26, 0.98)* 1.85 (0.86, 3.99) 0.71 (0.46, 1.08)
0.70)**
40–49 0.53 (0.26, 1.03) 1.35 (0.52, 3.55) 1.95 (0.74, 5.19) 0.27 (0.09, 0.84)*
Maternal education
No education Ref Ref Ref Ref Ref Ref
Primary 1.07 (0.60, 1.91) 0.29 (0.08, 1.11) 1.12 (0.49, 2.58) 2.36 (1.43, 3.91)* 1.00 (0.73, 1.37) 1.18 (0.63, 2.19)
Secondary 1.36 (0.87, 2.14) 0.30 (0.09, 1.02) 1.05 (0.42, 2.63) 1.51 (0.87, 2.59) 1.02 (0.79, 1.31) 1.52 (0.81, 2.83)
Higher 0.66 (0.27, 1.63) 0.23 (0.05, 0.99)* 0.12 (0.01, 1.15) 1.57 (0.75, 3.26) 0.84 (0.59, 1.20) 0.35 (0.14, 0.86)*
Mother’s employment
Not working Ref
Working 1.29 (0.60, 2.80)
Type of residence
Urban Ref Ref
Rural 0.86 (0.54, 1.36) 1.20 (0.99, 1.45)
Household size (members)
1–4 Ref Ref
5–8 0.57 (0.30, 1.08) 2.21 (0.94, 5.16)
9+ 0.51 (0.30, 0.87)* 3.51 (1.27, 9.67)*
Page 9 of 18
Table 3 (continued)
Diarrhoea in the last 2 weeks Acute respiratory infection in the last 2 weeks

Afghanistan Bangladesh India Nepal (2022) Pakistan Afghanistan Bangladesh India Nepal (2022)c Pakistan
(2015) (2017/18) (2019/21) (n = 581) (2017/18) (2015)b (2017/18) (2019/21) (n = 581) (2017/18)
(n = 3462) (n = 1084) (n = 26101) (n = 1306) (n = 3467) (n = 1084) (n = 26101) (n = 1306)
Variables AOR (95% CI) AOR (95% CI) AOR (95% CI) AOR (95% CI) AOR (95% CI) AOR (95% CI) AOR (95% CI) AOR (95% CI) AOR (95% CI) AOR (95% CI)

Wealth index
Poorest Ref Ref Ref Ref Ref
Poorer 1.25 (1.05, 1.49)* 1.45 (0.89, 2.35) 0.81 (0.32, 2.03) 0.97 (0.75, 1.25) 1.07 (0.57, 2.03)
Middle 1.09 (0.90, 1.31) 1.01 (0.52, 1.93) 0.59 (0.19, 1.85) 1.00 (0.75, 1.33) 0.62 (0.26, 1.49)
Richer 1.07 (0.88, 1.32) 1.33 (0.72, 2.43) 0.59 (0.20, 1.75) 0.79 (0.58, 1.07) 0.96 (0.38, 2.44)
Richest 0.87 (0.66, 1.15) 0.61 (0.29, 1.30) 0.32 (0.10, 0.95)* 0.64 (0.45, 0.91)* 1.01 (0.36, 2.83)
Source of drinking water
Improved Ref Ref
Unimproved 0.86 (0.63, 1.18) 1.36 (0.95, 1.94)
Hossain and Mihrshahi International Breastfeeding Journal

Type of toilet facility

Improved Ref Ref Ref
Unimproved 0.83 (0.59, 1.17) 1.08 (0.93, 1.25) 1.35 (0.72, 2.55)
Type of cooking fuel
(2024) 19:35

Clean
Solid
Number of ANC visits
None Ref Ref Ref
1–3 1.40 (0.96, 2.02) 2.01 (1.13, 3.56)* 2.10 (1.31, 3.37)*
4 or more 1.96 (1.31, 2.93)* 2.42 (1.21, 4.83)* 1.90 (1.19, 3.04)*
Place of delivery
Home Ref
Health facility 0.48 (0.31, 0.75)*
Exclusive breastfeeding
No Ref Ref Ref Ref Ref Ref Ref Ref Ref
Yes 0.49 (0.35, 0.65 (0.27, 1.54) 0.80 (0.70, 0.91)* 0.42 (0.20, 0.89)* 0.75 (0.51, 1.09) 0.57 (0.39, 0.83)* 0.91 (0.39, 2.09) 0.86 (0.71, 1.05) 0.70 (0.43, 1.14)
0.70)**

Only variables which met the cutoff for significance (p < 0.10) were included in the multivariable model
AOR adjusted odds ratio, Ref Reference category
Level of significance: *p < 0.05, **p < 0.001
a
Diarrhoea and acute respiratory infection symptoms in the previous 2 weeks
b
Sample sizes different in table for Afghanistan DHS as models differed by outcomes
c
Unadjusted odd ratios for EBF in Nepal not significant
Page 10 of 18
Hossain and Mihrshahi International Breastfeeding Journal (2024) 19:35 Page 11 of 18

Table 4 Association of exclusive breastfeeding and various individual, household-level, and socio-demographic factors with f­evera in
South Asia
Fever in the last 2 weeks

Afghanistan (2015)b Bangladesh (2017/18) India (2019/21) Nepal (2022) Pakistan (2017/18)
(n = 3460) (n = 1084) (n = 26101) (n = 581) (n = 1306)
Variables AOR (95% CI) AOR (95% CI) AOR (95% CI) AOR (95% CI) AOR (95% CI)

Infant age (months)

0–2 Ref Ref Ref Ref
3–4 1.52 (1.02, 2.26)* 1.60 (1.09, 2.34)* 1.84 (1.60, 2.11)** 1.56 (1.09, 2.22)*
5–6 1.83 (1.24, 2.71)* 2.11 (1.40, 3.19)** 1.95 (1.69, 2.25)** 2.29 (1.57, 3.34)**
Gender
Male Ref Ref
Female 0.75 (0.56, 0.99)* 0.90 (0.81, 1.01)
Mother’s age in years
15–19 Ref Ref
20–29 0.67 (0.55, 0.82)** 0.67 (0.38, 1.18)
30–39 0.58 (0.45, 0.74)** 0.63 (0.35, 1.13)
40–49 0.43 (0.24, 0.80)* 0.44 (0.16, 1.19)
Maternal education
No education Ref Ref
Primary 1.13 (0.93, 1.38) 1.84 (1.13, 3.01)*
Secondary 1.07 (0.91, 1.24) 2.30 (1.47, 3.61)**
Higher 0.86 (0.69, 1.08) 0.88 (0.46, 1.68)
Mother’s employment
Not working Ref Ref
Working 1.33 (0.98, 1.81) 2.02 (1.14, 3.61)*
Type of residence
Urban Ref
Rural 0.92 (0.77, 1.10)
Household size (members)
1–4
5–8
9+
Wealth index
Poorest Ref Ref
Poorer 0.87 (0.74, 1.03) 1.11 (0.65, 1.92)
Middle 0.87 (0.72, 1.05) 0.72 (0.43, 1.22)
Richer 0.65 (0.53, 0.81)** 0.87 (0.49, 1.55)
Richest 0.53 (0.41, 0.70)** 0.52 (0.27, 1.01)
Source of drinking water
Improved
Unimproved
Type of toilet facility
Improved Ref Ref
Unimproved 0.83 (0.58, 1.20) 1.18 (1.05, 1.33)*
Number of ANC visits
None Ref Ref
1–3 1.51 (1.09, 2.08)* 0.85 (0.68, 1.08)
4 or more 1.99 (1.28, 3.08)* 0.73 (0.58, 0.92)*
Place of delivery
Home Ref
Hossain and Mihrshahi International Breastfeeding Journal (2024) 19:35 Page 12 of 18

Table 4 (continued)
Fever in the last 2 weeks

Afghanistan (2015)b Bangladesh (2017/18) India (2019/21) Nepal (2022) Pakistan (2017/18)
(n = 3460) (n = 1084) (n = 26101) (n = 581) (n = 1306)
Variables AOR (95% CI) AOR (95% CI) AOR (95% CI) AOR (95% CI) AOR (95% CI)

Health facility 0.84 (0.71, 1.00)

Exclusive breastfeeding
No Ref Ref Ref Ref
Yes 0.36 (0.26, 0.50)** 0.74 (0.51, 1.06) 0.75 (0.67, 0.84)** 0.79 (0.59, 1.07)
Only variables which met the cutoff for significance (p < 0.10) were included in the multivariable model
AOR adjusted odds ratio, Ref Reference category
Level of significance: *p < 0.05, **p < 0.001
a
Fever symptoms in the previous 2 weeks
b
Sample size different between tables for Afghanistan DHS as models differed by outcomes

factor for diarrhoea in infants aged 0–6 months in and maturation are potentially aided by human milk,
Afghanistan, India, and Nepal and for fever in Afghani- which has its own immune system and a variety of solu-
stan and India. Our study also showed that infants who ble and cellular components [7]. IgA stops bacteria and
were EBF had lower odds of ARI in Afghanistan. These viruses from adhering to the mucosal epithelial cells,
findings are consistent with data from other studies in which might lead to infections [4, 5, 7]. Additionally,
LMICs which also estimated the beneficial association of there is a theory that oligosaccharides may prevent res-
EBF [29, 33–36, 40]. A large study [34] that included data piratory infections and gastrointestinal illnesses in babies
from the 2015–2016 India DHS found that EBF was pro- by preventing pathogens from attaching to their mucosa
tective against diarrhoea in babies aged 0 to 5 months at a [4, 5, 7]. The primary protein in human milk is lactofer-
national level (AOR: 0.64, 95% CI 0.57, 0.72) as well as in rin, which functions as a microbicidal agent to eradi-
the Central, Northern and Eastern regions of the country. cate viruses and bacteria [4, 5, 7]. Furthermore, infant
Another large study [29] that pooled data from the DHS weaning foods and powdered infant formula have been
in nine sub-Saharan African countries with high rates of shown to be contaminated with pathogens, putting non-
diarrhoeal morbidity reported that EBF was significantly breastfed newborns at an increased risk of exposure to
associated with decreased likelihood of diarrhoea among these contaminants [42–44]. Powdered infant formula
babies aged 0–5 months (AOR: 0.50, 95% CI 0.43, 0.57). is not sterile and has been associated with significant ill-
A cross-sectional study in Vietnam [41] suggested that ness from infections with the bacteria Cronobacter saka-
babies who were predominantly or partially breastfed had zakii and Salmonella spp [43, 45]. For example, a study
a greater likelihood of having diarrhoea in comparison to in Nigeria found that Cronobacter sakazakii was found
babies who were EBF. Individual studies in Bangladesh in 16 of the 360 powdered infant formula samples that
[33, 35] and Pakistan [36] revealed that EBF reduces the were examined, representing an average prevalence rate
risk of diarrhoea, ARI, and fever. Although our study of 4.4% [42]. Also, other sources of contamination during
found a protective effect of EBF in Bangladesh and Paki- formula milk preparation include the addition of unclean
stan where the general direction of the results was simi- water, improper handling or inadequate nipple and bottle
lar with Afghanistan, India, and Nepal, these results were cleaning [45].
non-significant. This inconsistency in findings could be Given its potential to have a significant negative
attributed to the considerable smaller sample size of the impact on health outcomes, prevalence of bottle feed-
surveys in these two countries compared to Afghanistan ing is a IYCF indicators [32]. Our study found that babies
and India. who were bottle fed were more likely to have diarrhoea
Several biological mechanisms explain why EBF may in India, ARI in Afghanistan, and fever in India and
have protective effects on infectious diseases such as Afghanistan. The similar pattern across these South
diarrhoea and ARI. Breast milk contains numerous anti- Asian countries suggests a high degree of consistency
inflammatory, antimicrobial, growth factors and bioac- in the association between bottle feeding and diarrhoea,
tive elements such as oligosaccharides, immunoglobulin ARI, and fever. This pattern was also observed in earlier
A (IgA), and lactoferrin that protect against childhood investigations in India [34] and Ethiopia [46] which have
infections [4, 5, 7]. A baby’s immunological development reported the impact of bottle feeding. The recent paper
Hossain and Mihrshahi International Breastfeeding Journal (2024) 19:35 Page 13 of 18

Fig. 3 Summary of the association between exclusive breastfeeding (EBF) and infant morbidity outcomes in 5 South Asian countries. A Association
between EBF and diarrhoea. B Association between EBF and acute respiratory infection. C Association between EBF and fever. Estimates by country
weighted according to study design; overall estimates were not weighted. * Unadjusted odd ratios for EBF and ARI and fever in Nepal in B, C. AOR:
Adjusted Odds Ratio; CI: Confidence interval
Table 5 Adjusted logistic regression models for association between other IYCF indicators and diarrhoea and acute respiratory infections in 5 South Asian countries
Diarrhoea in the last two weeksa Acute respiratory infection in the last two weeksb
IYCF indicators Afghanistan (2015) Bangladesh India (2019/21) Nepal Pakistan (2017/18) Afghanistan Bangladesh India (2019/21) Nepal (2022) Pakistan
(n = 3462) (2017/18) (n = 26101) (2022) (n = 1306) (2015) (2017/18) (n = 26101) (n = 581) (2017/18)
AOR (95% CI) (n = 1084) AOR (95% CI) (n = 581) AOR (95% CI) (n = 3462) (n = 1084) AOR (95% CI) AOR (95% CI) (n = 1306)
Hossain and Mihrshahi International Breastfeeding Journal

AOR (95% CI) AOR AOR (95% CI) AOR (95% CI) AOR (95% CI)
(95% CI)

Early initiation of breastfeeding

No - - Ref - Ref - - Ref Ref -
Yes - - 0.79 (0.69, 0.90)** - 1.40 (0.94, 2.06) - - 0.83 (0.68, 1.01) 4.21 (0.82, 21.78) -
(2024) 19:35

Bottle feeding
No Ref - Ref - - Ref - Ref - -
Yes 1.36 (0.98, 1.88) - 1.48 (1.26, 1.75)** - - 2.35 (1.57, 3.53)** - 1.38 (1.09, 1.75)* - -
Ref Reference category
Level of significance: *p < 0.05, **p < 0.001
a
ORs were adjusted for infant age, gender, mother’s age, maternal education, mother’s employment, type of residence, household size, wealth index, source of drinking water, toilet facility, number of ANC visits, place of
delivery
b
ORs were adjusted for infant age, gender, mother’s age, maternal education, mother’s employment, type of residence, household size, wealth index, source of drinking water, toilet facility, number of ANC visits, place of
delivery, and cooking fuel
- not included in multivariate analysis as p value in bivariable analysis not significant
Page 14 of 18
Hossain and Mihrshahi International Breastfeeding Journal (2024) 19:35 Page 15 of 18

Table 6 Adjusted logistic regression model for association between other IYCF indicators and fever in 5 South Asian countries
Fever in the last 2 weeks
IYCF indicators Afghanistan (2015) Bangladesh (2017/18) India (2019/21) Nepal (2022) Pakistan
(n = 3460) (n = 1084) (n = 26101) (n = 581) (2017/18)
AOR (95% CI) AOR (95% CI) AOR (95% CI) AOR (95% CI) (n = 1306)
AOR (95% CI)

Early initiation of breastfeeding

No - - - - -
Yes - - - - -
Bottle feeding
No Ref Ref Ref - -
Yes 2.09 (1.49, 2.91)** 1.34 (0.89, 2.01) 1.37 (1.17, 1.59)** - -
ORs were adjusted for infant age, gender, mother’s age, maternal education, mother’s employment, type of residence, household size, wealth index, source of drinking
water, toilet facility, number of ANC visits, and place of delivery
Ref Reference category
Level of significance: **p < 0.001
- not included in multivariate analysis as p value in bivariable analysis not significant

in Ethiopia [46] revealed that babies and young children (SDG) 2 and 3 on avoiding child deaths, attaining food
aged 0–23 months who receiving bottle feeding were 1.36 security, and improving nutrition in South Asia [50].
times more likely to experience ARI relative to those who To successfully establish a more supportive environ-
did not receive bottle feeding. Also, children who were ment for mothers who choose to breastfeed, we require
bottle fed tested positive for rotavirus infection, the most investments from governments at all levels of society
prevalent cause of diarrhoea, according to an observa- [4]. This involves providing women with the informa-
tional study [47] done in a hospital in India. This may be tion they need to make informed choices, as well as the
due to the higher risk of contamination from the water support they require from their families, communities,
used to make breastmilk substitute, bottle, teat, or milk work environments, and healthcare systems, in order to
in babies who are fed bottle milk [5]. In Peru, an investi- ensure that EBF for the first six months is possible [4].
gation [48] showed that 23% of bottles and 35% of bottle Increased protection, promotion, and support for EBF
nipples were contaminated with Escherichia coli, which would offer a cost-effective approach towards reaching
was greater than any other household item tested. the SDGs.
Another key IYCF indicator is early initiation of breast- In South Asia, various factors have an influence on
feeding, within an hour of birth. Our analysis indicated EBF. For instance, compared to boys, girls are less likely
that early initiation of breastfeeding was associated with to EBF in India [19]. Moreover, in contrast to some
lower diarrhoeal disease in India. This result was consist- low caste groups in India, the Tajik and other smaller
ent with a past study conducted on the DHS data in India four ethnic groups in Afghanistan are less likely to EBF,
[34] and sub-Saharan African nations [29]. which implies that sociocultural norms may have an
Around 27% of all deaths in children under five impact on EBF and are context specific [19]. A quali-
occurred in the South Asian region according to tative study in Pakistan in 2018 revealed that negative
UNICEF in 2020 [14]. A meta-analysis [49] undertaken attitudes towards colostrum, poor social support, influ-
estimated that babies who were predominantly breast- ence of social and family decision-makers, perceived
fed, partially breastfed, and non-breastfed had a greater inadequate milk supply, mother’s heavy workload, and
risk of death (relative risk of 1.48, 2.84, and 14.4 respec- advertising of infant formula were all factors that were
tively) relative to babies 0–5 months of age who were barriers to optimal EBF practices [51]. Therefore, these
EBF respectively in low-income countries. One of the highlight the need for context-specific and integrated
most cost-effective child survival interventions is EBF interventions tailored to each South Asian country to
during the first six months of life, which significantly support mothers in overcoming the barriers to opti-
lowers the chances of a child dying from pneumonia mal breastfeeding practices. The promotion of laws
or diarrhoea [6, 16]. Breastfeeding promotes develop- and rules governing the marketing of breast milk sub-
ment gains at all levels, from decreased illness inci- stitutes, supporting breastfeeding in the workplace and
dence to economic returns, and will be a significant households and paid maternity leave in LMICs is cru-
factor in reaching the Sustainable Development Goals cial [28].
Hossain and Mihrshahi International Breastfeeding Journal (2024) 19:35 Page 16 of 18

Strengths and limitations Asian country included to enable sufficient generaliz-

Potential limitations of the study should be taken into ability of study findings. Finally, due to the high response
consideration when interpreting the results. The statis- rate in the surveys and the use of consistent standard-
tical findings revealed significant effects for Afghanistan ized questionnaires, the study results are less likely to be
and India, which are, respectively, large sample popu- impacted by selection bias.
lations, and highly deprived environments where the
impacts of EBF are large. It is likely that the absence of Conclusion
significant results for the three nations with smaller sam- In summary, our study has shown that babies aged
ples (Bangladesh, Nepal, and Pakistan) may be due to a 0–6 months who were EBF had lower odds of diarrhoea
lack of statistical power, or Type II error. A paper pub- in Afghanistan, India, and Nepal. EBF was also associ-
lished in 1994 showed the extent of underpowered stud- ated with lower odds of fever in Afghanistan and India.
ies that led to null trials in the literature [52]. Only 36% Additionally, EBF was a protective factor against ARI in
of the null trials included in the survey had sufficient Afghanistan. Bottle feeding was a risk factor for diar-
power (80%) to identify a relative difference of 50% [52]. rhoea and ARI in some South Asian countries and early
The “absence of evidence” in these studies should not be initiation of breastfeeding was protective against diar-
regarded as “evidence of absence,” and underpowered rhoea in India. The results of this study seem to sug-
studies should be interpreted with caution [53]. gest that EBF, if expanded to universal levels may have
Given that the data were gathered via self-reports, an impact on morbidity in some South Asian countries.
recall bias may have had an impact on the findings. Nev- These findings strengthen the evidence base showing that
ertheless, we restricted the study sample to the young- EBF is a critical intervention to reduce morbidity and
est living baby who resided with the mother to lessen further advocacy efforts should be used to promote, sup-
the possible impact of recall bias. Also, our study used port and improve breastfeeding practices in South Asia.
cross-sectional data, which makes it challenging to estab-
Abbreviations
lish a causal relationship. Another important limitation ALRI Acute Lower Respiratory Infections
is that there may have been potential misclassification ANC Antenatal Care
bias, as measurement of ARI, fever, and diarrhoea was AOR Adjusted Odds Ratios
ARI Acute Respiratory Infection
determined in the two weeks before the survey. This CI Confidence interval
implies that mothers could have inaccurately stated that DHS Demographic and Health Surveys
their babies had symptoms of ARI, fever, or diarrhoea EBF Exclusive Breastfeeding
IgA Immunoglobulin A
resulting in an underestimation or overestimation of the IYCF Infant and Young Child Feeding
measure of association between EBF and other IYCF JMP Joint Monitoring Programme
indicators and morbidity outcomes. Another limitation LMIC Low-and Middle-Income Country
SDG Sustainable Development Goal
is that we were not able to adjust for all the confounding UNICEF United Nations Children’s Fund
variables such as seasonal and cultural variations which VIF Variance Inflation Factor
could have affected the association between the expo- VIP Ventilated Improved Pit
WHO World Health Organization
sure and outcomes. An important limitation to consider
is that dichotomous feeding exposures in nations where
Supplementary Information
breastfeeding is practiced differently, EBF and not EBF
The online version contains supplementary material available at https://​doi.​
have varied meanings; because of this, when comparing org/​10.​1186/​s13006-​024-​00644-x.
EBF and not EBF in nations such as Bangladesh, India,
and Nepal with high rates of both exclusive and par- Additional file 1: Table S1. Unadjusted logistic regression results for asso-
tial breastfeeding, the magnitude of the benefits of EBF ciation of exclusive breastfeeding and various individual, household-level,
and socio-demographic factors with diarrhea, acute respiratory infection,
relative to commercial milk formula or bottle feeding is and fever among infants 0-6 months in South Asia. Table S2. Unadjusted
understated. For nations like Nepal with a relatively small logistic regression results for association of other IYCF indicators and
population size in the surveys, there may be very few bot- various individual, household-level, and socio-demographic factors with
diarrhea, acute respiratory infection, and fever among infants 0-6 months
tle-fed babies in comparisons with the non-EBF group, in South Asia.
which could provide challenges for the statistical analysis.
Additionally, for similar reasons, cross-country compar-
Acknowledgements
isons of EBF and not EBF results may be inaccurate for Not applicable.
nations like Pakistan that have relatively low breastfeed-
ing rates. Authors’ contributions
SM conceptualized the study. SH planned, performed the statistical
A key strength is that our study uses the latest data analysis, interpreted the data, and wrote and revised the paper. SM critically
from nationally representative sample from each South
Hossain and Mihrshahi International Breastfeeding Journal (2024) 19:35 Page 17 of 18

appraised and edited the manuscript. All authors read and approved the final 10. World Health Organization, United Nations Children’s Fund. Planning
manuscript. guide for national implementation of the Global Strategy for Infant and
Young Child Feeding. 2007. https://​apps.​who.​int/​iris/​handle/​10665/​
Funding 43619. Accessed 1 Nov 2022.
No funding was received for the study. 11. World Health Organization. Global health risks: mortality and burden of
disease attributable to selected major risks. 2009. https://​apps.​who.​int/​
Availability of data and materials iris/​handle/​10665/​44203. Accessed 1 Nov 2022.
Datasets used for this paper are available online at: https://​dhspr​ogram.​com/​ 12. World Health Organization. Infant and young child feeding: model
data/​avail​able-​datas​ets.​cfm. chapter for textbooks for medical students and allied health profession-
als. 2009. https://​apps.​who.​int/​iris/​handle/​10665/​44117. Accessed 1 Nov
2022.
Declarations 13. World Health Organization. Infant and young child feeding. 2021. https://​
www.​who.​int/​news-​room/​fact-​sheets/​detail/​infant-​and-​young-​child-​
Ethics approval and consent to participate feedi​ng. Accessed 1 Nov 2022.
The DHS data-collection procedures were approved by the ICF International 14. United Nations Children’s Fund. Neonatal mortality. 2023. https://​data.​
institutional review board and relevant research ethics committee in all five unicef.​org/​topic/​child-​survi​val/​neona​tal-​morta​lity/. Accessed 1 Jan 2023.
countries. Participants were informed regarding the purpose of the surveys, 15. Naghavi M, Abajobir AA, Abbafati C, Abbas KM, Abd-Allah F, Abera SF,
the privacy of their answers, and that they were not required to reply to the et al. Global, regional, and national age-sex specific mortality for 264
questions if they did not feel comfortable doing so. Survey respondents pro- causes of death, 1980–2016: a systematic analysis for the Global Burden
vided written/oral informed consent. Data used in this study are taken from of Disease Study 2016. Lancet. 2017;390:1151–210. https://​doi.​org/​10.​
anonymized publicly available datasets online and permission was obtained 1016/​S0140-​6736(17)​32152-9.
from Measure DHS/ICF International to download the datasets. 16. United Nations Children’s Fund. Pneumonia and diarrhoea: tackling the
deadliest diseases for the world’s poorest children. New York: United
Consent for publication Nations Children’s Fund; 2012. https://​data.​unicef.​org/​resou​rces/​pneum​
Not applicable. onia-​and-​diarr​hoea-​tackl​ing-​the-​deadl​iest-​disea​ses-​for-​the-​worlds-​poore​
st-​child​ren/. Accessed 1 Nov 2022.
Competing interests 17. Ahs JW, Tao W, Löfgren J, Forsberg BC. Diarrheal diseases in low-and
The authors declare no competing interests. middle-income countries: incidence, prevention and management. Open
Infect Dis J. 2010;4. https://​doi.​org/​10.​2174/​18742​79301​00401​0113.
Author details 18. United Nations Children’s Fund. From the first hour of life: Making the
1
Department of Health Sciences, Faculty of Medicine, Health and Human Sci- case for improved infant and young child feeding everywhere. New York:
ences, Macquarie University, Sydney, NSW 2109, Australia. United Nations Children’s Fund; 2016. https://​data.​unicef.​org/​resou​rces/​
first-​hour-​life-​new-​report-​breas​tfeed​ing-​pract​ices/. Accessed 1 Nov 2022.
Received: 26 October 2023 Accepted: 12 May 2024 19. Benedict RK, Craig HC, Torlesse H, Stoltzfus RJ. Trends and predictors of
optimal breastfeeding among children 0–23 months, South Asia: analysis
of national survey data. Matern Child Nutr. 2018;14:e12698. https://​doi.​
org/​10.​1111/​mcn.​12698.
20. National Institute of Population Research and Training (NIPORT) and ICF.
References Bangladesh Demographic and Health Survey 2017–18. Dhaka, Bangla-
1. United Nations Children’s Fund. Diarrhoea. 2022. https://​data.​unicef.​org/​ desh, and Rockville. Maryland, USA: NIPORT and ICF; 2020.
topic/​child-​health/​diarr​hoeal-​disea​se/. Accessed 1 Nov 2022. 21. International Institute for Population Sciences (IIPS) and ICF. National
2. Nair H, Simões EA, Rudan I, Gessner BD, Azziz-Baumgartner E, Zhang JS, Family Health Survey (NFHS-5), 2019–21: India. Mumbai: IIPS; 2021.
et al. Global and regional burden of hospital admissions for severe acute 22. Central Statistics Organization (CSO), Ministry of Public Health (MoPH),
lower respiratory infections in young children in 2010: a systematic analy- and ICF. Afghanistan Demographic and Health Survey 2015. Kabul: Cen-
sis. Lancet. 2013;381:1380–90. https://​doi.​org/​10.​1016/​S0140-​6736(12)​ tral Statistics Organization; 2015. p. 2017.
61901-1. 23. National Institute of Population Studies - NIPS/Pakistan and ICF. Pakistan
3. Troeger CE, Blacker BF, Khalil IA, Zimsen SR, Albertson SB, Abate D, et al. Demographic and Health Survey 2017–18. Islamabad, Pakistan, and
Mortality, morbidity, and hospitalisations due to influenza lower respira- Rockville, Maryland, USA: NIPS and ICF; 2019.
tory tract infections, 2017: an analysis for the Global Burden of Disease 24. Ministry of Health and Population [Nepal], New ERA, and ICF. Nepal
Study 2017. Lancet Respir Med. 2019;7:69–89. https://​doi.​org/​10.​1016/​ Demographic and Health Survey 2022. Kathmandu: Ministry of Health
S2213-​2600(18)​30496-X. and Population [Nepal]; 2023.
4. Gertosio C, Meazza C, Pagani S, Bozzola M. Breastfeeding and its gamut of 25. International Vaccine Access Center (IVAC). Pneumonia and Diarrhea
benefits. Minerva Pediatr. 2015;68:201–12. Progress Report. Johns Hopkins Bloomberg School of Public Health. 2023.
5. Horta BL, Victora CG. Short-term effects of breastfeeding: a systematic https://​publi​cheal​th.​jhu.​edu/​ivac/​resou​rces/​pneum​onia-​diarr​hea-​progr​
review on the benefits of breastfeeding on diarrhoea and pneumonia ess-​repor​ts. Accessed 28 Feb 2024.
mortality. Geneva: World Health Organization; 2013. 26. Sharma IK, Byrne A. Early initiation of breastfeeding: a systematic
6. Lamberti LM, Fischer Walker CL, Noiman A, Victora C, Black RE. Breastfeed- literature review of factors and barriers in South Asia. Int Breastfeed J.
ing and the risk for diarrhea morbidity and mortality. BMC Public Health. 2016;11:17. https://​doi.​org/​10.​1186/​s13006-​016-​0076-7.
2011;11:S15. https://​doi.​org/​10.​1186/​1471-​2458-​11-​S3-​S15. 27. Dibley MJ, Roy SK, Senarath U, Patel A, Tiwari K, Agho KE, et al. Across-
7. Field CJ. The immunological components of human milk and their effect country comparisons of selected infant and young child feeding indica-
on immune development in infants. J Nutr. 2005;135:1–4. https://​doi.​org/​ tors and associated factors in four South Asian countries. Food Nutr Bull.
10.​1093/​jn/​135.1.1. 2010;31:366–79. https://​doi.​org/​10.​1177/​15648​26510​03100​224.
8. Richard SA, McCormick BJ, Seidman JC, Rasmussen Z, Kosek MN, 28. Kavle JA, LaCroix E, Dau H, Engmann C. Addressing barriers to exclusive
Rogawski ET, et al. Relationships among common illness symptoms and breast-feeding in low-and middle-income countries: a systematic review
the protective effect of breastfeeding in early childhood in MAL-ED: an and programmatic implications. Public Health Nutr. 2017;20:3120–34.
eight-country cohort study. Am J Trop Med Hyg. 2018;98:904. https://​doi.​ https://​doi.​org/​10.​1017/​S1368​98001​70025​31.
org/​10.​4269/​ajtmh.​17-​0457. 29. Ogbo FA, Agho K, Ogeleka P, Woolfenden S, Page A, Eastwood J, et al.
9. Victora CG, Bahl R, Barros AJ, França GV, Horton S, Krasevec J, et al. Breast- Infant feeding practices and diarrhoea in sub-Saharan African countries
feeding in the 21st century: epidemiology, mechanisms, and lifelong with high diarrhoea mortality. PLoS ONE. 2017;12:e0171792. https://​doi.​
effect. Lancet. 2016;387:475–90. https://​doi.​org/​10.​1016/​S0140-​6736(15)​ org/​10.​1371/​journ​al.​pone.​01717​92.
01024-7.
Hossain and Mihrshahi International Breastfeeding Journal (2024) 19:35 Page 18 of 18

30. ICF International. DHS sampling and household listing manual. 2012. transmission in Huascar. Peru Am J Epidemiol. 1989;129:785–99. https://​
https://​dhspr​ogram.​com/​publi​catio​ns/​publi​cation-​dhsm4-​dhs-​quest​ doi.​org/​10.​1093/​oxfor​djour​nals.​aje.​a1151​93.
ionna​ires-​and-​manua​ls.​cfm. Accessed 1 Aug 2022. 49. Sankar MJ, Sinha B, Chowdhury R, Bhandari N, Taneja S, Martines J, et al.
31. The DHS Program. Demographic and Health Surveys. DHS Overview. Optimal breastfeeding practices and infant and child mortality: a system-
https://​dhspr​ogram.​com/​Metho​dology/​Survey-​Types/​DHS.​cfm. Accessed atic review and meta-analysis. Acta Paediatr. 2015;104:3–13. https://​doi.​
1 Nov 2022. org/​10.​1111/​apa.​13147.
32. World Health Organization. Indicators for Assessing Infant and Young 50. United Nations. Transforming Our World: The 2030 Agenda for Sustain-
Child Feeding Practices. Definitions and measurement methods. 2021. able Development. 2015. https://​sdgs.​un.​org/​2030a​genda. Accessed 1
https://​www.​who.​int/​publi​catio​ns/i/​item/​97892​40018​389. Accessed 1 Nov 2022.
Nov 2022. 51. Zakar R, Zakar MZ, Zaheer L, Fischer F. Exploring parental perceptions
33. Abdulla F, Hossain MM, Karimuzzaman M, Ali M, Rahman A. Likelihood of and knowledge regarding breastfeeding practices in Rajanpur, Punjab
infectious diseases due to lack of exclusive breastfeeding among infants Province, Pakistan. Int Breastfeed J. 2018;13:24. https://​doi.​org/​10.​1186/​
in Bangladesh. PLoS ONE. 2022;17:e0263890. https://​doi.​org/​10.​1371/​ s13006-​018-​0171-z.
journ​al.​pone.​02638​90. 52. Moher D, Dulberg CS, Wells GA. Statistical power, sample size, and their
34. Dhami MV, Ogbo FA, Diallo TM, Olusanya BO, Goson PC, Agho KE, et al. reporting in randomized controlled trials. JAMA. 1994;272:122–4. https://​
Infant and young child feeding practices among adolescent mothers and doi.​org/​10.​1001/​jama.​1994.​03520​02004​8013.
associated factors in India. Nutrients. 2021;13:2376. https://​doi.​org/​10.​ 53. Nayak BK. Understanding the relevance of sample size calculation. Indian
3390/​nu130​72376. J Ophthalmol. 2010;58:469. https://​doi.​org/​10.​4103/​0301-​4738.​71673.
35. Khan MN, Islam MM. Effect of exclusive breastfeeding on selected
adverse health and nutritional outcomes: a nationally representa-
tive study. BMC Public Health. 2017;17:889. https://​doi.​org/​10.​1186/​ Publisher’s Note
s12889-​017-​4913-4. Springer Nature remains neutral with regard to jurisdictional claims in pub-
36. Saeed OB, Haile ZT, Chertok IA. Association between exclusive lished maps and institutional affiliations.
breastfeeding and infant health outcomes in Pakistan. J Pediatr Nurs.
2020;50:e62–8. https://​doi.​org/​10.​1016/j.​pedn.​2019.​12.​004.
37. Acharya P, Mishra SR, Berg-Beckhoff G. Solid fuel in kitchen and acute res-
piratory tract infection among under five children: evidence from Nepal
demographic and health survey 2011. J Community Health. 2015;40:515–
21. https://​doi.​org/​10.​1007/​s10900-​014-​9965-0.
38. Patel SK, Patel S, Kumar A. Effects of cooking fuel sources on the respira-
tory health of children: evidence from the Annual Health Survey, Uttar
Pradesh. India Public Health. 2019;169:59–68. https://​doi.​org/​10.​1016/j.​
puhe.​2019.​01.​003.
39. United Nations Children’s Fund. (2015). Progress on sanitation and
drinking water- 2015 update and MDG assessment. Geneva, Switzerland:
United Nations Children’s Fund. 2015. https://​www.​unicef.​org/​repor​ts/​
progr​ess-​sanit​ation-​and-​drink​ing-​water. Accessed 1 Nov 2022.
40. Nigatu D, Azage M, Motbainor A. Effect of exclusive breastfeeding ces-
sation time on childhood morbidity and adverse nutritional outcomes
in Ethiopia: analysis of the demographic and health surveys. PLoS ONE.
2019;14:e0223379. https://​doi.​org/​10.​1371/​journ​al.​pone.​02233​79.
41. Hajeebhoy N, Nguyen PH, Mannava P, Nguyen TT, Mai LT. Suboptimal
breastfeeding practices are associated with infant illness in Vietnam. Int
Breastfeed J. 2014;9:12. https://​doi.​org/​10.​1186/​1746-​4358-9-​12.
42. Enem SI, Ogbu CO, Okoli CE, Godwin E, Omeiza GK, Umeakuana PU,
et al. Detection and molecular characterization of Cronobacter sakazakii
isolated from powdered infant formula (PIF) from North Central Region,
Nigeria. Adv Microbiol. 2020;10:307–17. https://​doi.​org/​10.​4236/​aim.​2020.​
107022.
43. Losio MN, Pavoni E, Finazzi G, Agostoni C, Daminelli P, Dalzini E, et al.
Preparation of powdered infant formula: could product’s safety be
improved? J Pediatr Gastroenterol Nutr. 2018;67:543. https://​doi.​org/​10.​
1097/​MPG.​00000​00000​002100.
44. Tsai K, Simiyu S, Mumma J, Aseyo RE, Cumming O, Dreibelbis R, et al.
Enteric pathogen diversity in infant foods in low-income neighborhoods
of Kisumu, Kenya. Int J Environ Res Public Health. 2019;16:506. https://​doi.​
org/​10.​3390/​ijerp​h1603​0506.
45. Boué G, Cummins E, Guillou S, Antignac JP, Le Bizec B, Membré JM. Public
health risks and benefits associated with breast milk and infant formula
consumption. Crit Rev Food Sci Nutr. 2018;58:126–45. https://​doi.​org/​10.​
1080/​10408​398.​2016.​11381​01.
46. Ahmed KY, Page A, Arora A, Ogbo FA, Global Maternal and Child Health
Research collaboration (GloMACH). Associations between infant and
young child feeding practices and acute respiratory infection and
diarrhoea in Ethiopia: a propensity score matching approach. PloS One.
2020;15:e0230978. https://​doi.​org/​10.​1371/​journ​al.​pone.​02309​78.
47. John BM, Devgan A, Mitra B. Prevalence of rotavirus infection in children
below two years presenting with diarrhea. Med J Armed Forces India.
2014;70:116–9. https://​doi.​org/​10.​1016/j.​mjafi.​2014.​02.​008.
48. Black RE, Lopez de Romana G, Brown KH, Bravo N, Bazalar OG, Kanashiro
HC. Incidence and etiology of infantile diarrhea and major routes of