Determinants of Initiation and Exclusivity of Breastfeeding in Al Hassa,Saudi Arabia

Abstract

Aims:

The objectives of this cross-sectional study were to define the possible determinants of early initiation and exclusivity of breastfeeding and to assess knowledge towards breastfeeding among Saudi mothers in Al Hassa, Saudi Arabia.

Subjects and Methods:

Six hundred forty-one Saudi mothers with singleton infants approximately 24 months old attending well-baby clinics at four urban and six rural primary health care centers were selected through the multistage sampling method. Eligible mothers were invited to a personal interview using pretested questionnaires to gather data regarding sociodemographics, health-related variables, breastfeeding initiation, and current breastfeeding practices and to assess mothers' knowledge about breastfeeding.

Results:

Breastfeeding was initiated by 77.8% of mothers within the first 24 hours of childbirth. Exclusive breastfeeding at birth was reported in 76.1%, which declined to 32.9% and 12.2% at the age of 2 and 6 months, respectively. Increased maternal age, multiparity (three or more children), and vaginal delivery were significant positive predictors for early breastfeeding initiation as revealed by stepwise logistic forward regression. Rural, less-educated, low-income multiparous mothers were more likely to exclusively breastfed their infants as revealed by multivariate logistic regression. Irrespective of educational status, surveyed mothers demonstrated several misconceptions towards breastfeeding. Furthermore, early initiations and exclusivity were significantly influenced by sociodemographics, especially maternal educational and employment status.

Conclusions:

The rate of initiation and exclusivity of breastfeeding in Al Hassa is far below the World Health Organization recommendations.

Introduction

Breastfeeding is well recognized as the most appropriate mode for infant feeding. Providing several short- and long-term benefits for both infants and mothers, it is closely related to optimal physical and psychomotor development and chronic disease prevention.^1,2 For several reasons, breastfeeding epidemiology is crucial for an understanding of the health outcomes shaped by the nutrition transition currently underway in many developing countries.³

Exclusive breastfeeding for the first 6 months of life and continued breastfeeding up to 2 years of age or beyond are recommended by the World Health Organization (WHO)¹ and other health authorities.⁴ The patterns of breastfeeding showed variation when comparing developed and developing countries. For instance, 38% of all infants ≤5 months of age in the developed world and half of infants in many developing countries are exclusively breastfed for the first 6 months.⁵ Exclusive breastfeeding rates in Sub-Saharan Africa and East Asia/Pacific demonstrated an upward trend in the last decade from 24% to 32%⁵ and from 27% to 32%,⁶ respectively. Surprisingly, the rates of exclusive breastfeeding remained roughly constant or declined in the Middle East and North African region.⁶

In Saudi Arabia there has been a considerable change in the pattern of breastfeeding in recent decades due to population transition as a result of advancements in socioeconomic status.⁷ Breastfeeding in Saudi Arabia has been customary;⁸ its duration used to exceed the age of 24 months, and solid food would be introduced as late as 12–18 months and complementary to breastfeeding.⁹ Studies from Saudi Arabia have recorded a progressive decline in breastfeeding practice and duration, especially among young mothers in urban areas,^10–12 with early introduction of bottle feeding^12,13 and earlier introduction of solid foods.¹⁴

The initiation and duration of breastfeeding depend on several determinants, namely, sociodemographic, psychosocial, biomedical, and healthcare-related factors, community attributes, and public policy;¹⁵ however, the importance of these factors varies across countries and over time. Few studies^14–16 have addressed factors related to initiation and exclusivity of breastfeeding in Saudi Arabia with inherent faults either in samples studied of purely urban mothers¹⁴ or using different cutoffs for exclusive breastfeeding.¹⁶ There are many factors that influence breastfeeding duration: Some investigators have suggested that mothers' knowledge, attitudes, and support were stronger determinants of breastfeeding duration along with sociodemographic and biological factors.¹⁷ Studies that provide information regarding Saudi mothers' knowledge towards breastfeeding are scarce. The objectives of this study were to define possible determinants of early initiation and exclusivity of breastfeeding and to assess knowledge about different aspects of breastfeeding among Saudi mothers in Al Hassa, Saudi Arabia.

Subjects and Methods

Setting and design

A cross-sectional descriptive study was carried out in Al Hassa Governorate located in the Eastern Province of Saudi Arabia, 450 km from Riyadh. Al Hassa is populated with about 1 million Saudis and consists of three regions: urban, occupied by about 60% of the population; rural, composed of six major villages and occupied by 35% of the population; and Hegar areas, "Bedouin scattered communities," populated by the remaining 5%. Primary healthcare services are provided through 54 centers in Al Hassa and serving around 85% of the population.

Sampling

Infants 24 months old brought to the Well-Baby Clinics at primary healthcare centers (PHCs) accompanied with their mothers were targeted in both urban and rural areas (the Hegar areas were excluded for the sake of convenience because of difficulties in transportation). Considering the total registered infants 24 months old at Al Hassa PHCs (n = 15,993 in 2008, according to the local Health Directorate), with a prevalence of exclusive breastfeeding of 40%,^10–13 and assuming the worst acceptable prevalence of 36%, with an alpha level of 0.05, the total sample size should include 556 infants. Adding a potential non-response rate of 20%, the final sample size would be around 723 subjects. The proportionate sampling method was applied in relation to urban/rural distribution using a suitable sampling fraction. Two main urban areas (Hofuf and Mubaraz) and six major rural villages have been identified. Two PHCs from each urban area and one from each village were randomly selected using an updated PHC list. Subjects were selected according to the following inclusion criteria: Saudi nationality; mothers of singleton infants (multiple births were excluded); infants full term at birth with no congenital anomalies to interfere with feeding; and the informer should be the biological mother.

Data collection

Eligible mother–infant pairs were approached personally and invited to an individualized interview after proper orientation. The interviews were carried out by trained female Arabic-speaking nurses recruited at each health center using a pretested questionnaire to gather data regarding:

Sociodemographics: age in years, residence, marital status, age at marriage, educational and occupational status, family income in Riyals, parity, and age at current childbirth.

Detailed inquiries regarding current breastfeeding practices: time of initiation of breastfeeding in hours after childbirth, its duration in months, duration of exclusive breastfeeding, time of introduction of formula (if any), and detailed inquires about time of introduction and types of fluids, solids, and semisolids foods given.

Number of antenatal care visits, facilities providing the service, modes and place of childbirth and hospitalization and its duration (if any), sex of neonate, postpartum use of contraceptives (types and duration), and history of chronic illnesses (if any) and its nature. Some data were ascertained through reviewing of maternal and infants' available health records.

Knowledge about different aspects of breastfeeding practices through using 14 closed-ended questions, adopted from the available literature,^18–20 to assess maternal knowledge regarding benefits, misconceptions, and practices of breastfeeding and infant weaning. They were formulated as closed-ended questions with true/false and multiple-option formats. The following scoring system was applied: Those with correct responses received one point, while those with wrong responses and those who did not know received nil.

Interviews were carried out at the conclusion of the visit of the mothers and their infants to the PHC.

Definitions of breastfeeding patterns

Breastfeeding definitions used in this study were according to the infant feeding recommendations of the 2001 WHO Expert Consultation⁴ and the 55^th World Health Assembly.²¹

Exclusive breastfeeding: Infant receives only breastmilk without any additional food or drink, not even water, except for syrups and drops contain vitamins and minerals and medications in the first 6 months.

Partial breastfeeding: Breastfeeding the infant and supplementing his or her diet with other fluids or foods such as prelacteals, non-human milk, and solid and semisolid foods.

Pilot testing

Field pretesting was carried out through interviewing attendees for routine well-baby care at a nearby PHC (49 subjects beyond sample size) to ensure proper administration and reliability of the instrument. The inter-rater reliability was 0.81, and the knowledge part demonstrated reliability (Cronbach's alpha of 0.74).

Data analysis

Of the 723 mother–infant pairs eligible, 667 agreed to participate, for a response rate of 92.3%. Data were analyzed using SPSS version 16.0 (SPSS Inc., Chicago, IL). Data collection forms missing more than two items were discarded (n = 26). Therefore the total number of valid questionnaire forms was 641. Categorical variables were expressed in proportions and a percentage; χ² and Fisher's Exact tests were applied as appropriate. Continuous variables were expressed using mean, median, and SD; the t test was used for comparison. Knowledge scores were also expressed in mean, median, and SD; the Mann–Whitney and t test of significance were applied for comparison. Models of multivariate stepwise (forward method) and binary logistic regression analyses were generated to define the possible determinants of breastfeeding initiation and exclusivity (dependent variables), by inclusions of significant independent variables found at univariate analysis. Age and parity were entered as continuous variables, while others were entered as a dichotomy. Confidence intervals of 95.0% and p value of <0.05 were applied.

Ethical considerations

Permissions from the local Health Authorities as well as our institution were obtained after approval of the study protocol and data collection tools. Participants were provided with full orientation, with an emphasis on the right not to participate. Informed consent forms were obtained from those agreed to participate.

Results

Sample characteristics

The age of the mothers surveyed ranged from 16 to 44 (mean ± SD, 28.9 ± 4.8) years, parity ranged from one to 11 (mean ± SD, 3.9 ± 2.1) with a median of 4.0, and the age of the infants surveyed ranged from 23 to 26 months.

Table 1 demonstrates the sociodemographic characteristics of the surveyed mothers. Three hundred thirty-six (52.4%) had less than a secondary education; illiteracy was found among 24.8%, whereas 13.6% had university degrees. Of the mothers surveyed, 30.9% were employed, mostly in the governmental sector as teachers, technicians, or accountants or in clerical jobs. Postpartum contraceptive usage was reported by 76.6%: oral hormonal contraceptives in 68.2%, intrauterine device in 10.8%, and other methods, including coitus interruptus, safe period, and condoms, in 21%. Cesarean section as a mode of delivery was found in 12.3%. Hospitals as places for childbirth were mentioned in 97% of cases; 81.9% of cases delivered at hospitals had a median length of stay of 2 days and in the remaining 113 cases (18.1%) a median of 10 days. Chronic disease conditions were encountered in 29 mothers: Gestational diabetes in eight, bronchial asthma in 12, epilepsy in two, sickle cell disease in three, pregnancy-induced hypertension in four, and anemia in one.

Table 1.

Sociodemographic and Health-Related Characteristics of the Sample Surveyed in Relation to Breastfeeding Initiation

	No. (%)
		Early initiation of breastfeeding
Variable	Total (n = 641)	Yes (n = 499)	No (n = 142)^a	Univariate analysis [OR (95% CI)], p value
Age (years) (mean ± SD)	28.9 ± 4.8	29.8 ± 5.1	27.9 ± 3.7	0.001^**
Residence
Urban	403 (62.9)	297(59.5)	106 (74.6)	Reference
Rural	238 (37.1)	202 (40.5)	36 (25.4)	2.0 (1.29–3.11)^**
Education status
<Secondary	336 (52.4)	285 (57.1)	51 (35.9)	Reference
≥Secondary	305 (47.6)	214 (42.9)	91 (64.1)	0.42 (0.28–0.63)^**
Marital status
Married	613 (95.6)	481 (96.4)	132 (93.0)	Reference
Divorced/widowed	28 (4.4)	18 (3.6)	10 (7.0)	0.49 (0.12–1.18)
Occupational status
Working	198 (30.9)	141 (28.3)	57 (40.1)	Reference
Housewife	443 (69.1)	358 (71.7)	85 (59.9)	1.71 (1.13–2.56)^*
Parity
<3	311 (48.5)	219 (43.9)	92 (64.8)	Reference
≥3	330 (51.5)	280 (56.1)	50 (35.2)	2.35 (1.57–3.53)^**
Family income (in Saudi Riyals):
<6,000	332 (51.8)	289 (57.9)	43 (30.3)	Reference
≥6,000	309 (48.2)	210 (42.1)	99 (69.7)	0.34 (0.22–0.52)^**
Mode of delivery
Vaginal	562 (87.7)	473 (94.8)	89 (62.7)	Reference
Cesarean section	79 (12.3)	26 (5.2)	53 (37.3)	0.09 (0.05–0.16)^**
Place of delivery
Hospital	622 (97.0)	488 (97.8)	134 (94.4)	Reference
Home	19 (3.0)	11 (2.2)	8 (5.6)	0.38 (0.14–1.05)^†
Age at childbirth (years):
<20	101 (15.8)	53 (10.6)	48 (33.8)	Reference
20–<30	283 (44.1)	232 (46.5)	51 (35.9)	1.55 (1.04–2.32)^*
≥30	257 (40.1)	214 (42.9)	43 (30.3)	1.71 (1.13–2.60)^*
Gender of infant
Male	328 (51.2)	251 (50.3)	77 (54.2)	Reference
Female	313 (48.8)	248 (49.7)	65 (45.8)	1.17 (0.79–1.73)
Antenatal visit frequency
<9	173 (27.0)	131 (26.3)	42 (29.6)	Reference
≥9	468 (73.0)	368 (73.7)	100 (70.4)	1.18 (0.77–1.82)
Chronic maternal illness
No	612 (95.5)	482 (96.6)	130 (91.5)	Reference
Yes	29 (4.5)	17 (3.4)	12 (8.5)	0.30 (0.13–0.69)^**

Early initiation of breastfeeding was defined as within 24 hours of delivery.

Includes never breastfed and late initiators of several days.

p < 0.05, ^**p < 0.001.

†

Exact confidence intervals.

CI, confidence interval; OR, odds ratio.

Breastfeeding initiation

The total of non-breastfeeders and late initiators amounted to 142 (22.2%). Of the 641 mothers, 58 (9.0%) had never breastfed their infants. For those who never breastfed, low milk flow was the main reason (n = 33), followed by cesarean section and sedation (n = 13) and maternal chronic disease condition (n = 12).

Eighty-four (13.1%) women had initiated breastfeeding several days after delivery; the main stated reasons for late breastfeeding initiation were pain following delivery (n = 18), inability to sit properly to breastfeed due to episiotomies (n = 12), sedation (n = 14), intake of drugs (n = 11), weakness and fatigue (n = 9), no or low milk flow (n = 23), and a combination of several reasons (n = 20). For late initiators, sugar water, herbal tea, and formula feeding were used as alternatives.

Breastfeeding was initiated within the first 24 hours for 499 (77.8%) neonates: 56 (11.2%) within 1 hour after birth, 141 (28.3%) from 1 to <6 hours, 218 (43.7%) from 6 to 24 hours, and 84 (13.1%) ≥24 hours. The mean onset of breastfeeding initiation was 15.9 ± 6.8 (median, 14) hours. Among the 499 early breastfeeding initiators, in 284 (56.9%) the initiation was solely by breastfeeding, in 142 the initiation was by breastfeeding along with bottle feeding, while 73 used sugared water and herbal tea with breastfeeding.

Table 1 also demonstrates univariate analysis of breastfeeding initiation in relation to sociodemographics and health-related variables. Older maternal age, multiparity (three or more children), rural residence, and being a housewife were significantly associated with early breastfeeding initiation, while higher educational status (secondary or higher), being employed, having a higher income (≥6,000 Saudi Riyals), cesarean section delivery, and chronic maternal conditions were significant negative predictors.

Table 2 gives the results of multivariate stepwise logistic regression (forward method) analysis of predictors for early breastfeeding initiation. The final model shows that low parity (fewer than three children) and cesarean section were negative predictors, while increased maternal age was positively associated with early breastfeeding initiation.

Table 2.

Multivariate Stepwise Logistic Regression (Forward Method) for Determinants of Breastfeeding Initiation Among the Mothers Surveyed

Model, predictor	β	OR (95% CI)	p value^*	% predicted	Model χ ²	Model p value
Model 1
Maternal age	0.231	2.26 (1.10–4.66)	0.000	17.7	63.4	0.031
Constant	01.16	3.819	0.000
Model 2
Education (≥secondary)	−0.214	0.81 (0.73–0.89)	0.002	41.6	26.59	0.022
Maternal age	0.313	2.04 (1.28–3.30)	0.005
Constant	1.269	3.55
Model 3
Parity (<3)	−0.415	0.66 (0.49–0.96)	0.003	64.6	33.98	0.020
Family income (>6,000 SR)	−0.380	0.68 (0.49–0.88)	0.010
Education (≥secondary)	−0.213	0.82 (0.74–0.98)	0.022
Maternal age	0.205	1.99 (1.06–3.73)	0.033
Constant	0.673	1.961
Model 4
Mode of delivery (cesarean section)	−0.664	0.52 (0.35–0.76)	0.001	68.4	39.88	0.028
Parity (<3)	−0.641	0.66 (0.49–0.88)	0.005
Education (≥secondary)	−0.412	0.81 (0.73–0.90)	0.010
Maternal age	0.219	1.88 (1.13–2.42)	0.029
Residence (rural)	0.198	1.61 (1.03–3.43)	0.033
Constant	0.772	2.165
Final model
Mode of delivery (cesarean section)	−0.218	0.71 (0.53–0.94)	0.004	70.4	49.24	0.003
Parity (<3)	−0.411	0.56 (0.31–0.98)	0.010
Maternal age	0.564	2.24 (1.37–3.68)	0.016
Constant	1.171	3.235

Reference groups include: maternal educational status (<secondary); parity (three or more children); family income (>6000 Saudi Riyals [SR]); mode of delivery (vaginal); and residence (rural).

Significance at p < 0.05.

β, β coefficient.

Patterns and duration of breastfeeding

Table 3 displays the status of breastfeeding as revealed by the interviewed mothers in relation to the infant's age. Formula feeding reaches its peak in the second month (37.9%). A rapid decline occurs in exclusivity of breastfeeding from 66.5% at birth to 19.2% at 4 months and to 12.2% by 6 months of age (confidence intervals = 9.9–14.9). Median breastfeeding duration was 6.0 (mean, 8.5 ± 7.4) months and was longer among low-educated mothers (median of 8.0 [10.8 ± 7.9] vs. a median of 6.0 [7.5 ± 7.0] months for those with secondary education or higher, p = 0.001). Rural mothers breastfed longer compared to urban mothers (9.6 ± 7.6 vs. 7.9 ± 7.3 months, p = 0.007). Also, low-income and older mothers (≥30 years) breastfed their infants for longer duration (9.3 ± 7.4 [p = 0.001] and 8.7 ± 7.2 [p = 0.002] months, respectively).

Table 3.

Breastfeeding Patterns of Infants Surveyed (Through 24 Months of Age) in Relation to Their Age in Months

	Number (%) [95% CI]
	Breastfeeding status (total n = 641)
Infant's age	Exclusive	Partial	Never breastfed	Total breastfeeding
Birth	426 (66.5) [62.7–70.0]	73 (11.4) [9.2–14.1]	142 (22.2) [19.1–25.5]	499 (77.8) [74.5–80.9]
1 month	488 (76.1) [72.2–79.3]	95 (14.8) [12.3–17.8]	58 (9.0) [7.1–11.5]	583 (91.0) [88.5–92.9]
2 months	211 (32.9) [29.4–36.6]	276 (43.1) [39.3–46.9]	154 (24.0) [20.9–27.5]	487 (76.0) [72.5–79.1]
4 months	123 (19.2) [16.3–22.4]	312 (48.7) [44.8–58.5]	206 (32.1) [28.6–35.8]	435 (67.9) [64.2–71.4]
6 months	78 (12.2) [9.9–14.9]	319 (49.8) [45.9–53.6]	244 (38.0) [34.4–41.9]	397 (61.9) [58.1–65.6]
9 months	53 (8.3) [6.4–10.7]	251 (39.2) [35.5–43.0]	337 (52.5) [48.7–56.4]	304 (47.4) [43.6–51.3]
12 months	—	218 (34.0) [30.4–37.8]	423 (66.0) [62.2–69.6]	218 (34.0) [30.4–37.8]
>12 months	—	117 (18.3) [15.5–21.4]	524 (81.7) [78.6–84.5]	117 (18.3) [15.5–21.4]

Partial breastfeeding was defined as breastfeeding the infant but supplementing his or her diet with other fluids or foods. Exclusive breastfeeding was defined as the infant was receiving only breastmilk without any additional food or drink.

Correlates of exclusive breastfeeding

Table 4 displays the possible independent predictors for exclusive breastfeeding among the sample surveyed. Older maternal age, multiparity (three or more children), rural residence, and being a housewife were significant positive correlates for exclusive breastfeeding as revealed by univariate analysis, whereas higher education, higher family income, oral hormonal contraceptives, late initiation of breastfeeding, and chronic maternal illnesses were significantly associated with nonexclusivity of breastfeeding. Multivariate regression analysis revealed that rural residence, older age, being a housewife, and multiparity were positively correlated with exclusive breastfeeding.

Table 4.

Multivariate Logistic Regression Analysis of Sociodemographic and Health-Related Determinants of Exclusive Breastfeeding Among the Mothers Surveyed

	Exclusive breastfeeding [n (%)]			Multivariable regression analysis
Independent variable	Yes (n = 78)	No (n = 563)	Univariate analysis OR (95% CI), p value	β	OR (95% CI)	p value
Age (years) (mean ± SD)	29.4 ± 6.7	27.8 ± 3.9	0.001^*	0.141	1.14 (1.03–1.23)	0.034^*
Residence
Urban	31 (39.7)	372 (66.0)	Reference
Rural	47 (60.3)	191 (34.0)	2.95 (1.77–4.94)^**	0.371	1.74 (1.28–2.36)	0.008^*
Educational status
<Secondary	63 (80.8)	273 (48.5)	Reference		Reference
≥Secondary	15 (19.2)	290 (51.5)	0.22 (0.12–0.42)^**	−0.488	0.61 (0.39–0.97)	0.026^*
Occupational status
Working	11 (14.1)	187 (33.2)	Reference		Reference
Housewife	67 (85.9)	376 (66.8)	3.03 (1.51–6.22)^**	0.444	1.84 (1.31–2.57)	0.010^*
Number of children (mean ± SD)	4.26 ± 2.4	3.32 ± 1.9	0.001^**	0.146	1.15 (1.01–1.30)	0.035^*
Family income
<6,000 SR	65 (83.3)	267 (47.4)	Reference		Reference
≥6,000 SR	13 (16.7)	296 (52.6)	0.18 (0.09–0.35)^**	−0.385	0.53 (0.31–0.91)	0.020^*
Mode of delivery
Vaginal	61 (78.2)	491 (87.2)	Reference
Cesarean section	7 (21.8)	72 (12.8)	0.78 (0.31–1.80)^†
Oral hormonal contraception
No	51 (65.4)	255 (45.3)	Reference		Reference
Yes	27 (34.6)	308 (54.7)	0.44 (0.26–0.74)^**	−0.053	0.95 (0.62–1.44)	0.071
Breastfeeding initiation
≤6 hours	49 (62.8)	279 (49.6)	Reference		Reference
>6 hours	29 (37.2)	284 (50.4)	0.58 (0.35–0.97)^*	−0.128	0.84 (0.57–1.24)	0.121
Infant's gender
Male	41 (52.6)	287 (51.0)	Reference
Female	37 (47.4)	276 (49.0)	0.94 (0.57–1.55)
Antenatal care visits
<9	21 (26.9)	152 (27.0)	Reference
≥9	57 (73.1)	411 (73.0)	1.00 (0.57–1.77)
Maternal chronic illness
Yes	9 (11.5)	20 (3.6)	Reference		Reference
No	69 (88.5)	543 (96.4)	0.28 (0.12–0.70)^**^†	−0.041	0.92 (0.71–1.19)	0.163

For multivariate logistic regression analysis, constant = −2.899, χ² = 38.151, p = 0.0001, percentage predicted = 73.8%.

p < 0.05, ^**p < 0.001.

†

Exact confidence intervals.

Breastfeeding knowledge

Table 5 demonstrates responses of the mothers surveyed towards the 14 knowledge items. The total knowledge scores (out of 14 points) were 8.8 ± 1.4 (median, 8). Knowledge scores were higher among those with secondary education or higher (10.2 ± 1.6 vs. 8.4 ± 1.3 for less than secondary education, p = 0.011). Multiparous mothers (three or more children) scored lower compared to those with lower parity (9.68 ± 1.10 vs. 8.1 ± 1.0, p = 0.003).

Table 5.

Responses of the Mothers Surveyed to Knowledge-Related Items Towards Breastfeeding (n = 641)

	Number (%) of responses
Knowledge item	Correct	Incorrect	Do not know
Breastfeeding should start immediately after delivery. (True)	219 (34.2)	306 (47.7)	116 (18.1)
Breastfeeding causes maternal obesity. (False)	179 (27.9)	267 (41.7)	195 (30.4)
Breastfeeding spoils the shape of breasts. (False)	213 (33.2)	301 (47.0)	127 (19.8)
Breastfeeding protects women from breast cancer. (True)	360 (56.2)	119 (18.6)	162 (25.3)
Breastfeeding should be stopped once pregnancy occurs. (False)	188 (29.3)	386 (60.2)	67 (10.5)
Breastfeeding should continue even if the infant has diarrhea. (True)	297 (46.3)	185 (28.9)	159 (24.8)
Breastfeeding is inappropriate for working mothers. (False)	162 (25.3)	417 (65.0)	62 (9.7)
Breastfeeding decreases infant's respiratory infections. (True)	268 (41.8)	171 (26.7)	202 (31.5)
Breastfeeding duration should be for ___ months. (2 years) (multiple options)	463 (72.2)	113 (17.6)	65 (10.1)
Juices and other fluids should be given at the 3^rd month of infant's age. (False)	309 (48.2)	207 (32.3)	125 (19.5)
Colostrum should be discarded before initiating breastfeeding. (False)	322 (50.3)	163 (25.4)	156 (24.3)
Breastfeeding should be given ___ frequency/day vs. ___. (On demand) (multiple options)	462 (72.1)	101 (15.8)	78 (12.2)
Exclusive breastfeeding definition? (For 6 months) (multiple options)	283 (44.1)	219 (34.2)	139 (21.7)
Minced food supplements should be given before the 4^th month. (False)	244 (38.1)	291 (45.4)	106 (16.5)

Correct answers are given in bold type in parentheses after the statement.

Several misconceptions towards breastfeeding were found: Of the mothers surveyed, 34.2% knew that breastfeeding should be started immediately after birth, 41.7% and 47.0% believed that breastfeeding causes obesity and spoils the breast's shape, respectively, over 60.0% believed that breastfeeding should be stopped once pregnancy occurs, 28.9% of mothers agreed that breastfeeding should be ceased if diarrhea occurs, 32.3% stated that fluids should be introduced beginning in the third month, 28.0% were unable to mention that breastfeeding should be given on demand, and 55.9% failed to define the exact duration of exclusive breastfeeding.

Discussion

In this study, out of the 641 mothers surveyed, 77.8% had initiated breastfeeding within 24 hours of childbirth, of which 11.2% were within 1 hour, 28.3% from 1 to 6 hours, and 43.7% from 6 to 24 hours after childbirth. Exclusive breastfeeding as a mode of infant nutrition accounted for 76.1% at birth with an abrupt decline at 2 months of age to 32.9%, and those who exclusively breastfed were found to be only 19.2% and 12.2% at 4 and 6 months, respectively. The WHO has recommended that neonates should be breastfed immediately or within half an hour after birth, with exclusivity during the first 4–6 months of life.^4,22 Our figures of breastfeeding initiation are lower than those reported by previous studies carried out in Saudi Arabia¹⁰ and neighboring Arab countries:^17,23 in Jeddah¹⁰ 23.2% of mothers initiated breastfeeding within the first hour, in Lebanon¹⁷ 18.3% of mothers breastfed their infants within half an hour, and in Kuwait²³ 39.0% of mothers initiated breastfeeding within the first hour after birth. Previous studies from Saudi Arabia^10,12–14 have reported higher rates of breastfeeding initiation: For instance, Al Mouzan et al.¹¹ reported that breastfeeding was initiated in 91.6% of newborns, with a rate of 98.9% in Riyadh¹² and 94.0% in Jeddah.¹⁰ Our figures are close to those reported from Kuwait,²³ where the rate of initiation was reported to be 79%.

Several factors were found to interplay in the process of breastfeeding initiation.^15,17 Our study delineates that some sociodemographic predictors—maternal age and parity—and health-related factors—like cesarean section—may influence early initiation of breastfeeding.

These results are consistent with those obtained from Lebanon,¹⁷ where it was reported that cesarean section and hospital-related factors significantly influenced breastfeeding initiation. Previous breastfeeding experience and maternal age were independent significant predictors for breastfeeding intention and feeding choice in a similar study.²⁴

Furthermore, Trussel et al.²⁵ found that children of women with higher parity tended to breastfeed the longest; those with seven or more children were twice as likely to breastfed their children as women with firstborn children.

In Saudi Arabia, the prevalence of exclusive breastfeeding is not precisely known, while considering our findings and others,^11,12 the prevalence of exclusive breastfeeding is far from WHO recommendations, which have called for exclusive breastfeeding for the first 6 months.⁴

Rates of exclusive breastfeeding in this study are close to those reported from other Arab countries: rates of 10.1% and 12% in Lebanon¹⁷ and Kuwait,²³ respectively, at 6 months of age. These figures as well as ours are low compared to those reported from the developed world; for example, at 4–6 months, in Luxembourg 54% of mothers exclusively breastfed their newborns, in The Netherlands 37%, in Japan 41%,¹⁵ and in Austria46%.²⁶

The median breastfeeding duration in this study was 6.0 (mean ± SD, 8.5 ± 7.4) months and was longer among low-educated, rural, low-income mothers and those ≥30 years old.

The duration of breastfeeding among our sample is shorter compared to figures previously reported from a community-based survey¹¹ from Saudi Arabia where the duration of breastfeeding was 12.5 months; working mothers breastfed their children less than nonworking mothers. Another study carried out in Riyadh¹² reported a mean duration of breastfeeding of 8.5 ± 6.2 months.

Consistent to our results, previous studies^27,28 have found that urban working mothers in developing countries tend to breastfed for shorter intervals. Grummer-Strawn²⁹ has reported that the odds of breastfeeding for a child whose mother had had no education was twofold higher than those of a child whose mother had had at least 7 years of education even after controlling of other confounding factors. Also, children belonging to higher socioeconomic strata were substantially less likely to be breastfed. The previous finding is consistent to our results where those belonged to families with higher income (>6,000 Saudi Riyals) had a lower likelihood for breastfeeding initiation and exclusivity. Several studies^25,30,31 from developing countries have shown that within countries, breastfeeding is more prolonged among rural and less or non-educated women rather than women who reside in urban areas and with higher levels of formal education. Countries with a higher proportion of their population living in rural areas present more extended breastfeeding compared with more urbanized nations.^30,31

Furthermore, in developing countries demonstrating population transition with increasing urbanization, women achieving higher levels of formal education and more working outside of their households are expected to witness a decrease in the practice and duration of breastfeeding across time,^30,31 a scenario that is applicable to the Saudi Arabian community where women account for 55% of university graduates and the urban population represents 82% of the total with a rate of urbanization equals to 2.5% annual rate of change (for the years 2005–2010) and ranked 39^th in the global rank of urbanization in the year 2009.³²

In addition, there is a change in the status of Saudi women in response to the socioeconomic advancement with more educational and employment opportunities; for instance, the male to female ratio for students at universities changed considerably over a period of a few years from one to over two women for every man with a dramatic increase in literacy among females in a very short period.³³

Also, there is a change in the roles of women in contemporary Gulf societies. Women's traditional monorole of marriage and mothering has changed to multirole models and more outdoor socialization; they choose to pursuit higher education and careers and are less accepting of having their roles restricted to motherhood. More schooling is associated with shorter breastfeeding;^30,31 however, this differential is decreasing over time in some developing and most developed countries, where the direction of this association is already reversed. Women in the United States and other industrialized nations with higher levels of education have improved breastfeeding outcomes compared to their less educated counterparts.^25,30,31

Our study indicates that positive predictors for exclusive breastfeeding include older maternal age, parity (three or more children), rural residence, and being a housewife, while higher educational levels and economic status were negatively associated with exclusive breastfeeding.

Several studies^14,17,23 carried out in the region have reported similar results where low maternal educational level and rural residence were the most important factors for maintenance of an ideal breastfeeding pattern, while higher maternal education correlated with a shorter period of exclusive breastfeeding.

These changes in the trend of breastfeeding in developing countries witnessing population and nutrition transition can be explained by increasing urbanization, improving education, increase in contraceptive use, and changing pattern of childbearing, all contributing to the trend of shorter breastfeeding in these countries^30,31,34 and placing proportionately more children into those groups that breastfeed least.³¹

This study and others^10,35 have shown that maternal employment is a risk factor for nonexclusivity and early breastfeeding cessation among Saudi women, contrary to patterns found in developed countries where breastfeeding is positively related to the socioeconomic status, namely, household income, higher educational status, and maternal employment; this relation is reversed in developing countries.^30,31

In Saudi Arabia, women may work outside the home in settings where they do not have contact with unrelated men: In girls' schools and the women's sections of universities, social work and development programs for women, banks that cater to female clients, medicine and nursing for women, television and radio programming, and computer and library work. Significant social implications that act as barriers to breastfeeding for employed women in Saudi Arabia and the Gulf countries include embarrassment at breastfeeding before others, even of the same gender,³⁶ fears of the evil eye³⁷ (superstitious fears of envy of the lactating woman with inflicting injury or bad luck, including refusal of breastfeeding, cessation of milk flow, or disease for the nursing infant), lack of special facilities such as lactation rooms, inconvenience, and isolation.³⁸

Furthermore, in Saudi society, breastfeeding in public is considered a taboo, and it is prohibited,³⁹ with the lack of family support that can overshadow the unquestionable benefits of breastfeeding.³⁸ Additional constraint for employed women includes the relatively short maternity leave (about 10 weeks in the governmental sector), which may force Saudi women to hire foreign maids or nannies who often do much of the work of child rearing and feeding.^37,39 Breastfeeding sometimes is rejected for not being modern, especially among those of higher socioeconomic status.^34,36 For breastfeeding interventions to be successful, public perceptions and societal norms that shape the women's decisions to initiate and continue breastfeeding should be explored.^18,20

Our results demonstrate the prevalence and acceptance of several misconceptions regarding breastfeeding. Urban employed mothers with higher educational levels demonstrated higher knowledge compared to those with low levels of education, rural residence, and housewives, yet this knowledge was not interpreted in terms of a higher rate of initiation and breastfeeding exclusivity.

In Iraq, Abdul Ameer et al.¹⁸ found that illiterate mothers and those with informal or unknown education lacked appropriate knowledge compared to urban women in almost all parameters studied except for frequency of breastfeeding. Lack of such knowledge may result in early introduction of supplements prior to 6 months of age with subsequent breastfeeding cessation.¹⁸ Among our participants, 28.9% and 60.2% stated that breastfeeding should be stopped in the case of the baby's diarrhea or with the occurrence of pregnancy, respectively. Bella and Dabal¹⁹ in their study found that 60% of their female college participants believed that breastfeeding should be stopped immediately once pregnancy occurs. Another misconception is the false belief that breastfeeding will adversely spoil the shape of the breast and will cause obesity; a similar result was found by the previously mentioned study,¹⁹ where 33% of participants believed that breastfeeding will spoil their figure. This misconception is probably widely promoted in response to the changing role of Saudi women in an urbanized modern society with more opportunities for higher education and employment.

Aggressive steps taken to protect and promote breastfeeding in some developing countries have been documented to slow down negative behaviors and health outcomes associated with nutrition transition, including the obesity epidemic. Popkin el al.⁴⁰ showed that in Honduras these improvements are likely to be explained at least in part by massive, well-planned, well-executed national breastfeeding policies and promotion program. In Saudi Arabia as well in other Gulf countries the proposed policies to promote breastfeeding may include expanding awareness of the benefits of breastfeeding to include a larger sample of the community through social clubs and the curricula of high schools and universities. Breastfeeding awareness needs to be supported via peer counseling at the crucial period during breastfeeding along with allocating comfortable rooms for mothers to breastfeed in private and to support breastfeeding in public, especially at work places, hospitals, and other facilities; this may improve social acceptance of breastfeeding. Maternity leave needs to be reconsidered as women's participation in the Saudi workforce is increasing. Finally, a tax on or an increase in the price of formula milk would lead to an increase in breastfeeding by ensuring it as a feasible option.^36,39

Study Limitations

The study design was cross-sectional with inherent limitations basically in the form of recall bias. The sample included PHC attendees and did not include those received similar care at other health facilities, including the private sector, who may have different socioeconomic status, which might imply different patterns and determinants of breastfeeding. The study findings merely convey associations rather than inferences because of the study design adopted; a prospective cohort design would be more appropriate.

Conclusions

Early breastfeeding initiation and breastfeeding exclusivity among our sample were determined by sociodemographic factors, especially educational and employment status: Educated, employed, and high-income mothers were less likely to initiate and maintain breastfeeding despite their relatively higher level of knowledge. The reported rate of breastfeeding initiation and exclusivity are far lower compared to the current WHO recommendations. Irrespective of maternal educational status, many misconceptions are prevalent regarding breastfeeding practices.

Footnotes

Disclosure Statement

No competing financial interests exist.

References

Effect of breastfeeding on infant and child mortality due to infectious diseases in less developed countries. A pooled analysis. WHO Collaborative Study Team on the Role of Breastfeeding on the Prevention of Infant Mortality. Lancet 2000;355:451–455. Erratum in Lancet, 2000; 355:1104.

Binns

, Lee

, Scott

. The fetal origins of disease hypothesis: Public health implications for the Asia-Pacific region. Asia Pac J Public Health, 2001; 13:68–73.

Yusuf

, Reddy

, Ounpuu

et al. Global burden of cardiovascular diseases. Part 1: General considerations, the epidemiologic transition, risk factors and impact of urbanization. Circulation, 2001; 104:2746–2753.

World Health Organization. Report of the Expert Consultation on the Optimal Duration of Exclusive Breastfeeding. World Health Organization: Geneva, 2001; 28–30.

Koosha

, Hashemifesharaki

, Mousavinasab

. Breastfeeding pattern and factors determining exclusive breast-feeding. Singapore Med J, 2008; 49:1002–1006.

UNICEF. Desirable rate of exclusive breastfeeding, still a distant goal in most Arab world [Press centre. News note] www.unicef.org/media/media_27854.html. 2010 June 10.

Al-Ayed

, Qureshi

. Breastfeeding practices in urban Riyadh. J Trop Pediatr, 1998; 346:131–117.

Al Othaimeen

, Sawaya

, Tannous

et al. A nutrition survey of infants and preschool children. Saudi Med J, 1988; 9:40–48.

Haque

. Feeding pattern of children under two years of age in Riyadh, Saudi Arabia. Ann Trop Paediatr, 1983; 3:129–132.

10.

Shawky

, Abalkhail

. Maternal factors associated with the duration of breast feeding in Jeddah, Saudi Arabia. Pediatr Perinatal Epidemiol, 2003; 17:91–96.

11.

El Mouzan

, Al Omar

, Al Salloum

et al. Trends in infant nutrition in Saudi Arabia: Compliance with WHO recommendations. Ann Saudi Med, 2009; 29:20–23.

12.

Al-Jassir

, El-Bashir

, Mouziddin

. Surveillance of infant feeding practices in Riyadh city. Ann Saudi Med, 2004; 24:136–140.

13.

Ogbeide

, Siddiqui

, Al Khalifa

et al. Breastfeeding in a Saudi Arabian community: Profile of parents and influencing factors. Saudi Med J, 2004; 25:580–584.

14.

Al Hreashy

, Tamim

, Al Baz

et al. Patterns of breastfeeding practice during the first 6 months of life in Saudi Arabia. Saudi Med J, 2008; 29:427–431.

15.

Yngve

, Sjostrom

. Breastfeeding determinants and a suggested frame of action in Europe. Public Health Nutr, 2001; 4:729–739.

16.

Madani

, Al-Nowaisser

ARA

, Khashoggi

. Breastfeeding patterns in Saudi Arabia. Ecole Food Nutr, 1994; 31:239–245.

17.

Batal

, Boulghoujin

, Abdallah

et al. Breastfeeding and feeding practices in a developing country: A national survey in Lebanon. Public Health Nutr, 2006; 9:313–319.

18.

Abdul Ameer

, Al-Hadi

, Abdulla

. Knowledge, attitudes and practice of Iraqi mothers and family child-bearing women regarding breastfeeding. Eastern Medit Health J, 2008; 14:1003–1014.

19.

Bella

, Dabal

. Misconceptions about breastfeeding among Saudi female college students. Ann Saudi Med, 1998; 18:69–72.

20.

Al-Jassir

, El-Bashir

, Moizuddin

et al. Infant feeding in Saudi Arabia: Mothers' attitudes and practices. Eastern Medit Health J, 2006; 12:6–13.

21.

The 55th World Health Assembly. Infant Feeding RecommendationsPublication A67/INF.DOC./8.Switzerland: World Health Organization Geneva, 2002.

22.

World Health Organization. Global Strategy for Infant and Young Child Feeding. The Optimal Duration of Exclusive BreastfeedingPublication A54/INF.DOC./4.World Health Organization: Geneva, 2001. www.who.int/childadolescenthealth/New_Publications/NUTRITION/gs_iycf.pdf. 2009 December 12.

23.

Al-Fadli

, Masoud

, Al-Jasem

. Breastfeeding among children less than two years old in Ahmadi region, Kuwait. Kuwait Med J, 2002; 34:281–285.

24.

McInnes

, Love

, Stone

. Independent predictors of breastfeeding intention in a disadvantaged population of pregnant women. BMC Public Health, 2001; 1:10.

25.

Trussel

, Grummer-Strawn

, Rodriguez

et al. Trends and differentials in breastfeeding behaviour: Evidence from the WFS and DHS. Population Studies, 1992; 46:285–307.

26.

Yngve

, Sjostrom

. Breastfeeding in countries of the European Union and EFTA: Current and proposed recommendations, rationale, prevalence, duration and trends. Public Health Nutr, 2001; 4:631–645.

27.

Leung

, Tam

, Hung

et al. Socio-demographic and atopic factors affecting breastfeeding intention in Chinese mothers. J Clin Psychiatry, 2003; 64:966–968.

28.

Dubois

, Girard

. Social determinants of initiation, duration and exclusivity at the population level: The results of the Longitudinal Study of Child Development in Quebec (ELDEQ 1998–2002) Can J Public Health, 2003; 94:300–305.

29.

Grummer-Strawn

. The effect of changes in population characteristics on breastfeeding trends in fifteen developing countries. Int J Epidemiol, 1996; 25:94–102.

30.

Pérez-Escamilla

. Breastfeeding and the nutritional transition in the Latin American and Caribbean Region: A success story? Cad Saude Publica, 2003; 19,Suppl 1:S119–S127.

31.

Monteiro

, Conde

, Popkin

. Independent effects of income and education on the risk of obesity in the Brazilian adult population. J Nutr, 2001; 131:881S–886S.

32.

Central Intelligence Agency. World Fact Book. www.cia.gov/library/publications/theworldfactbook/fields/2212.html. 2010 June 15.

33.

Ghubash

, Hamdi

, Bebbingtopn

. The Dubai Community Psychiatric Survey: I. Prevalence and socio-demographic correlates. Social Psychiatry Psychiatr Epidemiol, 1992; 37:53–61.

34.

Al-Sahab

, Tamim

, Mumtaz

et al. Predictors of breastfeeding in a developing country: Results of a prospective cohort study. Public Health Nutr, 2008; 11:1350–1356.

35.

Fida

, Al-Aama

. Pattern of infant feeding at a university hospital in Western Saudi Arabia. Saudi Med J, 2003; 24:725–729.

36.

Osman

, El Zein

, Wick

. Cultural beliefs that may discourage breastfeeding among Lebanese women: A qualitative analysis. Int Breastfeed J, 2009; 4–12.

37.

Khassawneh

, Khader

, Amarin

et al. Knowledge, attitude and practice of breastfeeding in the north of Jordan: A cross-sectional study. Int Breastfeed J, 2006; 1:17.

38.

Al Sinani

. Breastfeeding in Oman—the way forward [abstract] Oman Med J, 2008. www.omjournal.org/ReviewArticle/Fulltext/200810/BreastfeedingOmaThe.html. 2010 June 1.

39.

Musaiger

. Food habits and nutritional status of the family in the Arabian Gulf. Arab Journal of Food and Nutrition. Bahrain Centre for Studies and Research: Bahrain, 2001; 40.

40.

Popkin

, Canahuati

, Bailey

et al. An evaluation of a national breastfeeding promotion programme in Honduras. J Biomed Sci, 1991; 23:5–21.