The Impact of Formula Industry Marketing on Breastfeeding Rates in Native and Migrant Mothers

Abstract

Introduction:

Lifelong benefits of breastfeeding are far-reaching. However, optimal breastfeeding practices may be negatively affected by the marketing of different forms of breast milk substitutes. In addition, whether market factors have a different impact on migrants' breastfeeding practices when compared to natives has been poorly investigated.

Objectives:

Our study's aims are (1) to assess the effect of market factors on breastfeeding discontinuation (any and exclusive), (2) comparing migrant and native women.

Methods:

A longitudinal study was conducted within baMBINO, a nationwide project on perinatal health among migrant women in Portugal. Our final sample included 1,251 migrants and 1,150 natives recruited between April 2017 and March 2019 in 32 public maternities. Cox regression analysis was performed, adjusting for important confounders, and interactions were tested.

Results:

Market factors were associated with any and exclusive breastfeeding discontinuation. Participants receiving free formula samples from a health professional were more likely to stop exclusive breastfeeding (adjusted hazard ratios [aHR] = 1.37, 95% confidence intervals [95% CI] = 1.13–1.66, p = 0.002). Reporting exposure to formula discounts was associated with discontinuation of exclusive (aHR = 1.22, 95% CI = 1.09–1.38, p = 0.001) and any breastfeeding (aHR = 1.21, 95% CI = 1.00–1.46, p = 0.046). No interactions were found between being migrant and exposure to market factors.

Conclusions:

Market factors influence discontinuation of any and exclusive breastfeeding. The impact of marketing does not differ between native and migrant women.

Introduction

Breastfeeding practices and behaviors are shaped by historical, socioeconomic, market, cultural, and individual factors that operate at multiple levels.¹ Where market factors are concerned, the commercial interests of the industry of infant formula and related products compete with public health policies aiming at breastfeeding protection, promotion, and support.^1–6 In 1981, the World Health Assembly approved the International Code of Marketing of Breast-milk Substitutes (hereafter referred to as the International Code) to regulate the commercial practices of manufacturers of baby food and related products, which include promotion of infant formula, follow-on and growing-up milks, bottle-fed complementary foods, feeding bottles and teats, and any food or beverage marketed or represented as suitable to be fed to infants aged younger than 6 months.⁷ Subsequent resolutions were endorsed⁸ due to the negative influence of industry marketing on breastfeeding initiation, exclusivity, and duration trends observed among native women across several countries.^4–6,9,10 How industry marketing impacts breastfeeding practices by migrant women and minority ethnic groups has been less studied.^11,12

Evidence shows that migrant mothers are more likely to initiate and maintain any breastfeeding^13,14 when compared to natives, although acculturation may have a detrimental effect on optimal breastfeeding practices.¹³ More acculturated mothers may perceive formula feeding as the norm in western countries and use it as a way to fit in the host country.¹⁵ Such perceptions appear to be influenced by the marketing strategies used by formula manufacturers.⁶ However, little is known about the influence of market factors on the duration of breastfeeding among migrant mothers. In Portugal, the International Code provisions adopted in 2008, and revised in 2017 (Decree-Law No. 62/2017), cover infant and follow-on formula with respect to labeling, but only cover infant formula with regard to promotion, including free samples and formula discounts.¹⁶ Our study aims to assess the effect of market factors on discontinuation of any or exclusive breastfeeding in Portugal taking into consideration multiple maternal characteristics, namely the fact that they were Portuguese native or migrants.

Materials and Methods

Study design

We carried out a longitudinal study, under the scope of baMBINO (Migrant and Perinatal Health: Barriers, Incentives and Outcomes), a national project investigating perinatal health care experience and maternal and child health disparities among migrant and native Portuguese women. The study was approved by the Ethics Committees of the Institute of Public Health of the University of Porto (Proc. No. CE14013/14th March 2014) and of every participating hospital, and the Portuguese Data Protection Authority.

Setting

The recruitment occurred in public maternity hospitals of mainland Portugal, that were responsible for 85.1% of all deliveries reported in 2017.¹⁷ Out of the 39 public hospitals with obstetric care, 32 accepted to participate in the project. In Portugal, migrants have free access to reproductive, maternal, and child health care, irrespective of their legal status.^18,19 Participants were recruited between April 2017 and March 2019 at admission for delivery.

Sample

Details about sampling methods have been described elsewhere.²⁰ Women aged 18 years or older were eligible to participate in the baMBINO study. For each migrant participant having birth, a native woman with a subsequent delivery in the same maternity unit was invited to participate. Having a stillbirth was an exclusion criterion, with the exception of mothers to twin births with one live birth.

For the purpose of this study, and as described in Figure 1, we excluded mothers (1) who never breastfed; (2) with contraindications to initiate or maintain breastfeeding; (3) with missing data on the outcome and/or exposure variables; and (4) with outliers in duration/cessation of exclusive breastfeeding (>195 days). The final sample included 2,401 participants (1,251 migrants and 1,150 natives).

FIG. 1.

Flowchart showing the selection of the study participants (N = 2,401). MFMCQ, Migrant Friendly Maternal Care Questionnaire.

Data collection

A longitudinal study was conducted from April 2017 until March 2019. Data were collected at two different time points: (1) at recruitment and (2) any time after the first month postpartum, but mostly around three postpartum months.

In each maternity unit, health professionals enrolled in the project explained the study to the women and invited them to participate. Women were given an information sheet, available in 15 different languages, containing the details of the project. After obtaining written informed consent from women willing to participate, information about sociodemographic (maternal country and date of birth, education, and marital status), obstetric and newborn characteristics, and infant feeding practices was collected.

From one month postpartum onward, computer-assisted telephonic interviews were conducted by trained researchers. A 112-item questionnaire—the Migrant Friendly Maternal Care Questionnaire (MFMCQ), was administered in participants' preferred language to investigate maternity care experience during pregnancy, delivery, and postpartum period.²¹ The project also counted on the interpreter service of the High Commission for Migration, on an as-needed basis. In addition, a set of questions on exposure to marketing of formula and related products were asked.

Outcome measures

Any breastfeeding was defined as being fed with breast milk (including milk expressed or breast milk from a donor), regardless of receiving infant formula, water, liquids, or foods other than breast milk. Our definition of exclusive breastfeeding was feeding the baby with breast milk (including milk expressed or breast milk from a donor), without any food or liquid other than breast milk, not even water, in line with the WHO definition.²²

At follow-up interview, participants were asked, “How did you feed your child in the last 24 hours?.” If they were not exclusively breastfeeding, the information about the baby's age when he/she was first fed with infant formula, any liquid or food other than breast milk was also collected. The baby's age at the time of the interview was used as a time point for discontinuation of any breastfeeding. The baby's age was also used as the time of exclusive breastfeeding discontinuation, if the time of first introduction of liquids or foods besides breast milk was not available.

Exposure variables

Our main exposure was market factors, that is the marketing of formulas and other related products carried out by the industry. Information about exposure to different types of formula marketing, within and outside the health system, were collected after completing the MFMCQ. Since the MFMCQ covers pregnancy, delivery, and postnatal period, the exposure to market factors is intended to have occurred during this time lapse. Considering the diverse cultural and socioeconomic background of our study participants, we did not make any distinction among infant formula (for infants younger than 6 months of age), follow-on milks (for infants aged 6–12 months), and toddler milks (for children aged 12 months and older). Thus, we referred to formulas using the general term “artificial baby milk,” which is the term most commonly used in Portugal.

First, mothers were asked if they, or a member of their family, had ever received from a health professional (1) a free formula sample; (2) bottles, teats, or pacifiers; (3) industry-sponsored gifts; and/or, (4) industry-sponsored brochures. In case of a positive answer, they were asked where and from whom they received those items. Then, participants were asked if, and where, they had ever seen special displays, discounts, or advertising of formula.

Covariates

The other variables included in the analysis were maternal education, defined as the highest level of educational achievement (none or primary, secondary, and tertiary education); migrant or native status, with a migrant being defined as any foreign-born woman who moved to Portugal, following the International Organization for Migration²³ maternal country of birth, according to which participants were divided into natives and mothers born in Portuguese-speaking African countries (PSAC), in Brazil and in other countries; parity (multiparous versus primiparous); gestational age (< 37 weeks versus ≥37 weeks); mode of delivery (vaginal/instrumental versus cesarean section); smoking during pregnancy (yes or no); and giving birth in a Baby-Friendly Hospital (yes or no).

Statistical analysis

Categorical data were described as proportions and compared using chi-square test. Cox regression analysis was fitted to investigate the effect of market factors on any and exclusive breastfeeding cessation. The level of significance was set at p < 0.05. Only those market factors and covariates that showed a significant association with breastfeeding duration, in univariate analysis, were included in the respective final models for any and exclusive breastfeeding discontinuation. Migration status and maternal country of birth were used in model 1 and 2, respectively, as a proxy for cultural factors. Interactions between market factors and migration were also tested. Effect estimates are expressed as crude (hazard ratio [HR]) and adjusted hazard ratios (aHR) and 95% confidence intervals (95% CI) using a Cox regression. Data were analyzed using the IBM SPSS Statistics 26.0.

Results

Table 1 shows the characteristics of study participants by exposure to market factors that were significantly associated with our outcomes. Overall, women exposed to marketing were more likely to be native. In addition, participants exposed to industry-sponsored brochures were less likely to give birth in a Baby-Friendly hospital, while those who had seen formula discounts were more highly educated than their counterparts. As to our main outcomes, significant lower rates of exclusive and any breastfeeding were found among women exposed to market factors as opposed to those who were not. Looking into the different types of exposure to industry marketing, commercial practices carried out through a health professional, namely the receipt of free formula samples, bottles, teats or pacifiers, and industry-sponsored brochures were less frequent than other forms of marketing (e.g., special displays, discounts, and ads of formula) (Table 2). When women were asked from whom and where they, or a family member, received a free formula, respectively, 74.3% and 86.9% of them gave an answer. Most reported that the health professionals were nurses (74.3% of answers), followed by doctors (20.6%) and pharmacists (4.4%). Nurses were also the most mentioned by participants when questioned about the receipt of the other items. The receipt of free samples occurred mainly in health centers (51.6% of answers), public hospitals (34.0%), and private services (8.8%), along with pharmacies (5%). In some cases, women specified also other contexts (6.3%) where it occurred, which were mainly antenatal classes. Pharmacies, drug stores, and supermarkets were the main setting of exposure to special displays and discounts. In addition, we observed that migrant women were significantly less exposed to market factors than their native counterparts. PSAC-mothers were the least exposed to all types of market factors, with the exception of industry-sponsored gifts (e.g., baby bib or spoon).

Table 1.

Comparison Between Exposed and Not Exposed to Market Factors

Characteristics	Free formula sample		p	Industry-sponsored brochure		p	Formula discounts		p
Characteristics	Not exposed (n = 2,218) n (%)	Exposed (n = 183) n (%)	p	Not exposed (n = 2,181) n (%)	Exposed (n = 220) n (%)	p	Not exposed (n = 1,523) n (%)	Exposed (n = 878) n (%)	p
Maternal education
Primary/none	561 (27.2)	45 (27.1)	0.554	563 (27.8)	43 (21.1)	0.089	410 (29.1)	196 (24.0)	0.028
Secondary	792 (38.4)	70 (42.2)		772 (38.1)	90 (44.1)		526 (37.3)	336 (41.1)
Tertiary	710 (34.4)	51 (30.7)		690 (34.1)	71 (34.8)		475 (33.7)	286 (35.0)
Migration status
Native	1,045 (47.1)	105 (57.4)	0.008	1,035 (47.5)	115 (52.3)	0.173	629 (41.3)	521 (59.3)	<0.001
Migrant	1,173 (52.9)	78 (42.6)		1,146 (52.5)	105 (47.7)		894 (58.7)	357 (40.7)
Maternal country of birth
Native	1,045 (47.1)	105 (57.4)	0.057	1,035 (47.5)	115 (52.3)	0.044	629 (41.3)	521 (59.3)	<0.001
PSAC^a	533 (24.0)	35 (19.1)		533 (24.4)	35 (15.9)		431 (28.3)	137 (15.6)
Brazil	256 (11.5)	15 (8.2)		243 (11.1)	28 (12.7)		191 (12.5)	80 (9.1)
Other countries^b	384 (17.3)	28 (15.3)		370 (17.0)	42 (19.1)		272 (17.9)	140 (15.9)
Parity
Multiparous	1,082 (51.3)	84 (47.7)	0.362	1,057 (51.1)	109 (50.2)	0.805	729 (50.6)	437 (51.8)	0.554
Primiparous	1,027 (48.7)	92 (52.3)		1,011 (48.9)	108 (49.8)		713 (49.4)	406 (48.2)
Gestational age (weeks)
≥37	2,025 (93.1)	162 (89.5)	0.067	1,986 (92.9)	201 (92.6)	0.886	1,395 (93.5)	792 (91.8)	0.117
<37	149 (6.9)	19 (10.5)		152 (7.1)	16 (7.4)		97 (6.5)	71 (8.2)
Mode of delivery
Vaginal/instrumental	1,562 (70.4)	123 (67.2)	0.361	1,539 (70.6)	146 (66.4)	0.194	1,064 (69.9)	621 (70.7)	0.655
Cesarean section	656 (29.6)	60 (32.8)		642 (29.4)	74 (33.6)		459 (30.1)	257 (29.3)
Gave birth in a BFH
No	767 (34.6)	59 (32.2)	0.522	734 (33.7)	92 (41.8)	0.015	515 (33.8)	311 (35.4)	0.425
Yes	1,451 (65.4)	124 (67.8)		1,447 (66.3)	128 (58.2)		1,008 (66.2)	567 (64.6)
Smoking during pregnancy
No	1,959 (90.1)	159 (87.8)	0.330	1,922 (89.9)	196 (90.3)	0.843	1,352 (90.8)	766 (88.5)	0.068
Yes	215 (9.9)	22 (12.2)		216 (10.1)	21 (9.7)		137 (9.2)	100 (11.5)
Exclusive BF
No	1,116 (50.3)	124 (67.8)	< 0.001	1,104 (50.6)	136 (61.8)	0.002	716 (47.0)	524 (59.7)	<0.001
Yes	1,102 (49.7)	59 (32.2)		1,077 (49.4)	84 (38.2)		807 (53.0)	354 (40.3)
Any BF
No	411 (18.5)	53 (29.0)	0.001	409 (18.8)	55 (25.0)	0.025	235 (15.4)	229 (26.1)	<0.001
Yes	1,807 (81.5)	130 (71.0)		1,772 (81.2)	165 (75.0)		1,288 (84.6)	649 (73.9)

Includes Angola, Cape Verde, Guinea-Bissau, Mozambique, and São Tomé e Príncipe.

Includes mothers born in: Europe (65.5%), most of whom were born in Eastern Europe (51.5%); Americas (13.6%); Asia (13.3%); and non-PSAC (7.5%).

BF, breastfeeding; BFH, Baby-Friendly Hospital; PSAC, Portuguese-speaking African countries.

Table 2.

Exposure to Market Factors and Breastfeeding Rates in Native and Migrant Mothers

Market factor	All sample (N = 2,401)	Native (n = 1,150) n (%)	Migrant (n = 1,251) n (%)	p	PSAC ^a (n = 568) n (%)	Brazil (n = 271) n (%)	Other countries ^b (n = 412) n (%)	p
Receipt from a health professional of
Free formula sample	183 (7.6)	105 (9.1)	78 (6.2)	0.008	35 (6.2)	15 (5.5)	28 (6.8)	0.057
Bottles/teats/pacifiers	260 (10.8)	143 (12.4)	117 (9.4)	0.015	37 (6.5)	26 (9.6)	54 (13.1)	0.001
Industry-sponsored gift	32 (1.3)	15 (1.3)	17 (1.4)	0.907	4 (0.7)	1 (0.4)	12 (2.9)	0.010
Industry-sponsored brochure	220 (9.2)	115 (10.0)	105 (8.4)	0.173	35 (6.2)	28 (10.3)	42 (10.2)	0.044
Have seen
Special displays	699 (29.1)	456 (39.7)	243 (19.4)	<0.001	86 (15.1)	45 (16.6)	112 (27.2)	<0.001
Formula discounts	878 (36.6)	521 (45.3)	357 (28.5)	<0.001	137 (24.1)	80 (29.5)	140 (34.0)	<0.001
Ads	1,073 (44.7)	603 (52.4)	470 (37.6)	<0.001	186 (32.7)	101 (37.3)	183 (44.4)	<0.001
Exclusive BF
No	1,240 (51.6)	669 (58.2)	571 (45.6)	<0.001	239 (42.1)	131 (48.3)	201 (48.8)	<0.001
Yes	1,161 (48.4)	481 (41.8)	680 (54.4)		329 (57.9)	140 (51.7)	211 (51.2)
Any BF
No	464 (19.3)	312 (27.1)	152 (12.2)	<0.001	31 (5.5)	47 (17.3)	74 (18.0)	<0.001
Yes	1,937 (80.7)	838 (72.9)	1,099 (87.8)		537 (94.5)	224 (82.7)	338 (82.0)

Includes Angola, Cape Verde, Guinea-Bissau, Mozambique, and São Tomé e Príncipe.

Includes mothers born in: Europe (65.5%), most of whom were born in Eastern Europe (51.5%); Americas (13.6%); Asia (13.3%); and non-PSAC (7.5%).

BF, breastfeeding; PSAC, Portuguese-speaking African countries.

Rates of exclusive and any breastfeeding in native and migrant mothers

The rates of exclusive and any breastfeeding were higher among migrant women (54.4% and 87.8%, respectively) than among natives (41.8% and 72.9%, respectively) (Table 2). In unadjusted Cox regression models, migrant mothers were less likely to discontinue either exclusive (HR = 0.78, 95% CI = 0.70–0.88, p < 0.001) or any breastfeeding (HR = 0.50, 95% CI = 0.41–0.61, p < 0.001), when compared to native women (Tables 3 and 4). Similar results were obtained after adjusting for market factors and other relevant covariates (Table 4). The association was stronger for discontinuation of any breastfeeding (aHR = 0.54, 95% CI = 0.44–0.66, p < 0.001) than for discontinuation of exclusive breastfeeding (aHR = 0.81, 95% CI = 0.72–0.91, p < 0.001). When compared to native Portuguese women, PSAC-born mothers had the lowest likelihood of discontinuing either exclusive (aHR = 0.75, 95% CI = 0.64–0.88, p < 0.001) or any breastfeeding (aHR = 0.25, 95% CI = 0.17–0.37, p < 0.001). No differences were found between Brazil-born mothers and native Portuguese.

Table 3.

Factors Associated with Discontinuation of Exclusive and Any Breastfeeding by Using Cox's Regression

Exposure	Exclusive BF discontinuation		Any BF discontinuation
Exposure	HR (95% CI)	p	HR (95% CI)	p
Receipt from a health professional of
Free formula sample	1.45 (1.20–1.75)	<0.001	1.22 (0.92–1.63)	0.172
Bottles/teats/pacifiers	0.92 (0.77–1.10)	0.354	0.83 (0.63–1.11)	0.216
Industry-sponsored gift	0.80 (0.51–1.26)	0.339	0.53 (0.24–1.20)	0.128
Industry-sponsored brochure	1.20 (1.00–1.43)	0.051	1.19 (0.90–1.58)	0.229
Have seen
Special displays	1.05 (0.93–1.19)	0.395	0.95 (0.78–1.16)	0.616
Formula discounts	1.30 (1.16–1.45)	<0.001	1.37 (1.14–1.64)	0.001
Ads	0.92 (0.82–1.03)	0.134	0.92 (0.77–1.10)	0.362
Maternal education
Primary/none	Reference		Reference
Secondary	0.96 (0.83–1.11)	0.581	1.06 (0.84–1.33)	0.652
Tertiary	0.82 (0.71–0.96)	0.011	0.85 (0.66–1.09)	0.198
Migration status
Native	Reference		Reference
Migrant	0.78 (0.70–0.88)	<0.001	0.50 (0.41–0.61)	<0.001
Maternal country of birth
Native	Reference		Reference
PSAC^a	0.71 (0.61–0.83)	<0.001	0.23 (0.16–0.33)	<0.001
Brazil	0.91 (0.76–1.10)	0.334	0.79 (0.58–1.07)	0.129
Other countries^b	0.81 (0.69–0.95)	0.009	0.70 (0.54–0.90)	0.006
Parity
Multiparous	Reference		Reference
Primiparous	1.13 (1.00–1.26)	0.043	1.10 (0.91–1.33)	0.315

Gestational age (weeks)
≥37	Reference		Reference
<37	1.57 (1.29–1.92)	<0.001	1.41 (1.04–1.91)	0.027
Mode of delivery
Vaginal/instrumental	Reference		Reference
Cesarean section	1.16 (1.03–1.31)	0.014	1.07 (0.87–1.30)	0.531
Gave birth in a BFH
No	Reference		Reference
Yes	0.84 (0.75–0.94)	0.003	0.77 (0.64–0.93)	0.005
Smoking during pregnancy
No	Reference		Reference
Yes	1.25 (1.05–1.49)	0.014	1.80 (1.40–2.31)	<0.001

Includes Angola, Cape Verde, Guinea-Bissau, Mozambique, and São Tomé e Príncipe.

Includes mothers born in: Europe (65.5%), most of whom were born in Eastern Europe (51.5%); Americas (13.6%); Asia (13.3%); and non-PSAC (7.5%).

BF, breastfeeding; CI, confidence intervals; HR, hazard ratio; PSAC, Portuguese-speaking African countries.

Table 4.

Adjusted Hazard Ratios for Exclusive and Any Breastfeeding Discontinuation

	N	HR (95% CI)	p	Model 1	p	Model 2	p
	N	HR (95% CI)	p	aHR^a (95% CI)	p	aHR^b (95% CI)	p
Exclusive BF discontinuation	1,240
Market factors
Formula free sample	183	1.45 (1.20–1.75)	<0.001	1.37 (1.13–1.66)	0.002	1.37 (1.13–1.66)	0.001
Industry-sponsored brochure	220	1.20 (1.00–1.43)	0.051	1.05 (0.87–1.27)	0.623	1.04 (0.87–1.26)	0.657
Formula discounts	878	1.30 (1.16–1.45)	<0.001	1.22 (1.09–1.38)	0.001	1.22 (1.09–1.37)	0.001
Migration status
Migrant	1,251	0.78 (0.70–0.88)	<0.001	0.81 (0.72–0.91)	<0.001	—
Maternal country of birth
PSAC^c	568	0.71 (0.61–0.83)	<0.001	—		0.75 (0.64–0.88)	<0.001
Brazil	271	0.91 (0.76–1.10)	0.334	—		0.92 (0.76–1.11)	0.365
Other countries^d	412	0.81 (0.69–0.95)	0.009	—		0.82 (0.70–0.96)	0.015
Any BF discontinuation	464
Market factors
Formula free sample	183	1.22 (0.92–1.63)	0.172	NA		NA
Industry-sponsored brochure	220	1.19 (0.90–1.58)	0.229	NA		NA
Formula discounts	878	1.37 (1.14–1.64)	0.001	1.21 (1.00–1.46)	0.046	1.19 (0.99–1.43)	0.072
Migration status
Migrant	1,251	0.50 (0.41–0.61)	<0.001	0.54 (0.44–0.66)	< 0.001	—
Maternal country of birth
PSAC^c	568	0.23 (1.57–0.33)	<0.001	—		0.25 (0.17–0.37)	<0.001
Brazil	271	0.79 (0.58–1.07)	0.129	—		0.81 (0.59–1.11)	0.182
Other countries^d	412	0.70 (0.54–0.90)	0.006	—		0.72 (0.56–0.93)	0.013

NA indicates that the variables with no significant association in univariate Cox Regression were not included in the model. Women exposed to each market factor were compared with non exposed ones. Native women are the reference category for migrant status and maternal country of birth.

Besides market factors, HR were adjusted for migration status (migrants versus natives) and (1) maternal education, smoking during pregnancy, gestational age, parity, mode of delivery, giving birth in a Baby-Friendly Hospital as for exclusive BF discontinuation; (2) smoking during pregnancy, gestational age, giving birth in a Baby-Friendly Hospital when assessing any BF discontinuation.

Besides market factors, HR were adjusted for maternal country of birth and (1) maternal education, smoking during pregnancy, gestational age, parity, mode of delivery, giving birth in a Baby-Friendly Hospital as for exclusive BF discontinuation; (2) smoking during pregnancy, gestational age, giving birth in a Baby-Friendly Hospital when assessing any BF discontinuation.

Includes Angola, Cape Verde, Guinea-Bissau, Mozambique, and São Tomé e Príncipe.

Includes mothers born in: Europe (65.5%), most of whom were born in Eastern Europe (51.5%); Americas (13.6%); Asia (13.3%); and non-PSAC (7.5%).

aHR, adjusted hazard ratios; CI, confidence intervals; BF, breastfeeding; HR, hazard ratio; PSAC, Portuguese-speaking African countries.

Effect of market factors on exclusive breastfeeding discontinuation

As shown in Table 3, market factors associated with discontinuation of exclusive breastfeeding on univariate Cox regression were free formula samples, industry-sponsored brochures, and formula discounts. All the remaining associated variables were included as confounders in the final models. After performing the adjusted Cox regression, free formula samples (aHR = 1.37, 95% CI = 1.13–1.66, p = 0.002) and discounts of formula (aHR = 1.22, 95% CI = 1.09–1.38, p = 0.001) were significantly associated with discontinuation of exclusive breastfeeding, while no significant association was observed with industry-sponsored brochures (See Table 4, Model 1).

Effect of market factors on any breastfeeding discontinuation

Discounts of formula constituted the only market factor associated with discontinuation of any breastfeeding in the crude model (Table 3). After adjusting for covariates that were statistically significant in the univariate analysis, discounts of formula still reduced the probability of any breastfeeding at any time (aHR = 1.21, 95% CI = 1.00–1.46, p = 0.046).

Discussion

Market factors influence the discontinuation of both exclusive and any breastfeeding, irrespective of migration status or maternal country of birth. Women receiving free formula samples from a health professional or having seen formula discounts are more likely to stop exclusive breastfeeding. Exposure to formula discounts is also a risk factor for any breastfeeding discontinuation. Thus, marketing factors impact native and migrant mothers in the same way. However, migrant women were less exposed to formula industry marketing and less likely to discontinue exclusive and any breastfeeding when compared to natives.

Lower exposure to formula marketing among PSAC-born mothers may be explained by the fact that most of the PSAC-born participants in our study gave birth at a Baby-Friendly Hospital.²⁰ This may have contributed to protect them from in-hospital distribution of industry-sponsored products. In addition, based on data collected within the baMBINO study, PSAC-born mothers are less proficient in the Portuguese language than Brazil-born mothers. Although PSAC's official language is Portuguese, a wide range of other languages are spoken by women included in this group (e.g., Creole in Cape Verde and multiple languages in Angola, Guinea-Bissau, Mozambique and São Tomé e Príncipe). Thus, it is possible that language barriers may have hindered communication with health professionals^24,25 resulting in less information being provided or understood, including information about industry-related products.

It should be noted that although migrant women have a lower risk of exclusive and any breastfeeding discontinuation than natives, migration status and maternal country of birth do not represent a protective factor against the negative influence of formula industry marketing. When present, free formula samples and formula discounts affect breastfeeding practices of migrant and native women to the same extent. Comparatively, women born in PSAC were less likely to discontinue exclusive and any breastfeeding than native Portuguese women, but no differences were observed between Brazil-born and native women. Brazilian-born mothers have been found to value the emotional bond associated with the practice of breastfeeding.²⁶ These cultural factors may have contributed to making them less permeable to the appealing messages of the formula industry, reducing their recall of exposure to marketing strategies. Further qualitative research is needed to gain a more in-depth understanding on this matter.

Previous research has found a negative effect of pre- or postnatal free formula samples provision on breastfeeding outcomes.^{4–6,27–31} Our findings support studies suggesting that the receipt of free samples increases the risk of exclusive breastfeeding discontinuation.^12,30,31 Health professionals are a trusted source of information and support concerning infant feeding. Thus, the provision of free formula samples by a health professional may be interpreted as an endorsement¹² of the use of breast milk substitutes. Furthermore, it may contrast with verbal counseling on optimal infant feeding practices, conveying mixed messages.³² In this regard, Reiff et al. suggested that hospital “modeling” of formula use may be more effective in shaping early infant feeding practices than verbal advice supportive of breastfeeding.³² However, the distribution of formula samples, which is forbidden in Baby-Friendly Hospitals in Portugal and elsewhere, may occur also in other health care settings.^28,29 This was observed in our study and in previous research leading to an increase in the risk of suboptimal breastfeeding.²⁹ Moreover, our findings suggest that the provision of free formula samples by health professionals influences native and migrant mothers in the same way.

Another key finding of our study is that formula discounts were negatively associated with any and exclusive breastfeeding, regardless of migration status and maternal country of birth. It may be that mothers who stopped any and exclusive breastfeeding prematurely are more likely to recall formula discounts, pointing out a mechanism of reverse causality. However, discounts are a price reducing strategy that may induce the purchase of formula even when not necessary.³³ In addition, the perception of an equal nutritional value to that of breast milk and the increased availability and affordability of formula in the host country have been reported by migrant women as reasons for formula feeding.^34,35 Furthermore, the relative affordability in the host country may induce the purchase of formula by women coming from settings where it represents a status symbol.³⁴ Nevertheless, our findings show that the negative effect of formula discounts span over all mothers, and not solely migrants.

It is important to point out that, in Portugal, national law forbids the provision of free samples and discounts of infant formula only (i.e., intended for infants aged 0–6 months), while the International Code covers all kinds of breast milk substitutes. However, sales promotions of follow-up formulas and toddler milks are often used to also advertise infant formula as part of the same line of products, through a mechanism of cross promotion, namely by using similar packaging, branding, and labeling.³⁶ Consequently, new mothers may not understand the difference between promotion of different types of formula, interpreting it as infant formula advertisement.³⁷ Thus, banning the discounts of all types of formulas from retailers is fundamental.

Our study is subject to some limitations. The time of exposure to market factors is unknown. Although it was not possible to identify in which reproductive stage women were more influenced by industry marketing, we were able to assess its impact on any and exclusive breastfeeding discontinuation. Data on exposure and outcome variables were collected at the same time, with the potential for reverse causality. Information about breastfeeding initiation, duration, and exclusivity intentions, which are a predictor of infant feeding behaviors, was not collected. However, our study contributes to strengthen evidence of the negative impact of formula samples provision on exclusive breastfeeding duration not only at the hospital but also in other health care settings. Moreover, we assessed the influence of different types of marketing strategies, some of which have been poorly studied so far (i.e., promotions at retailers, including special displays and price discounts). Finally, we found that being less exposed to market factors protected migrant populations from breastfeeding discontinuation. Although important, this finding may serve as an invitation to the industry to target migrant mothers and thus should be addressed with caution. Arguably, it brings awareness to health services for the need to extend special protections to these vulnerable populations because, if equally exposed to market factors, they will experience the same disadvantages observed among natives.

Conclusions

Our findings show that receiving free formula samples by a health professional is a risk factor for discontinuation of exclusive breastfeeding. In addition, formula discounts negatively influence both any and exclusive breastfeeding duration. Being migrant does not modify the association between exposure to market factors and breastfeeding duration.

Data Availability Statement

The datasets generated and analyzed during the current study are not publicly available due to a confidentiality agreement securing participants' privacy and anonymity, but they are available from the corresponding author on reasonable request.

Footnotes

Authors' Contributions

All authors contributed to the study conception and design. Material preparation, data collection, and analysis were performed by C.L. and H.B. The first draft of the article was written by C.L. H.B. and C.F. critically reviewed previous versions of the article. All authors read and approved the final article.

Acknowledgments

The authors are grateful to the participants of the study, the focal points at each hospital recruited, and the members of the baMBINO team.

Disclosure Statement

No competing financial interests exist.

Funding Information

This study is funded by FEDER funds through the Operational Programme for Competitiveness and Internationalization, and by national funds of FCT – Fundação para a Ciência e Tecnologia, under the scope of the project “Perinatal Health in Migrants: Barriers, Incentives and Outcomes” (POCI-01-0145-FEDER-016874; PTDC/DTPSAP/6384/2014), the Unidade de Investigação em Epidemiologia - Instituto de Saúde Pública da Universidade do Porto (EPIUnit) (UIDB/04750/2020), the PhD grant PD/BD/128082/2016 (C.L.) cofunded by the FCT and Human Potential Operating Program of the European Social Fund (POPH/FSE Program) and the contract DL57/2016/CP1336/CT0001 (C.F.).

References

Rollins

, Bhandari

, Hajeebhoy

, et al. Why invest, and what it will take to improve breastfeeding practices?. Lancet, 2016; 387:491–504.

Barennes

, Slesak

, Goyet

, et al. Enforcing the international code of marketing of breast-milk substitutes for better promotion of exclusive breastfeeding: Can lessons be learned?. J Hum Lact, 2016; 32:20–27.

Brady

. Marketing breast milk substitutes: Problems and perils throughout the world. Arch Dis Child, 2012; 97:529–532.

Harris

, Pomeranz

. Infant formula and toddler milk marketing: Opportunities to address harmful practices and improve young children's diets. Nutr Rev, 2020; 78:866–883.

Kaplan

, Graff

. Marketing breastfeeding–reversing corporate influence on infant feeding practices. J Urban Health, 2008; 85:486–504.

Piwoz

, Huffman

. The impact of marketing of breast-milk substitutes on WHO-recommended breastfeeding practices. Food Nutr Bull, 2015; 36:373–386.

World Health Organization (WHO). International Code of Marketing of Breast-Milk Substitutes. Geneva, Switzerland: WHO, 1981. Available at https://www.who.int/nutrition/publications/infantfeeding/bfhi-implementation-2018.pdf (accessed January 5, 2020).

World Health Organization (WHO). Nutrition. World Health Assembly resolutions and documents 33.32 (1980), 34.22 (1981), 35.26 (1982), 37.30 (1984), 39.28 (1986), 41.11 (1988), 43.3 (1990), 47.5 (1994), 49.15 (1996), 54.2 (2001), 55.25 (2002), 58.32 (2005), 59.11 (2006), 59.21 (2006), 61.20 (2008), 63.23 (2010), 65.6 (2012), 67(9) (2014), 69.9 (2016), 71.9 (2018). Available at https://www.who.int/nutrition/topics/wha_nutrition/en/ (accessed April 12, 2021).

Hastings

, Angus

, Eadie

, et al. Selling second best: How infant formula marketing works. Global Health, 2020; 16:1–12.

10.

Foss

, Southwell

. Infant feeding and the media: The relationship between Parents' Magazine content and breastfeeding, 1972–2000. Int Breastfeed J, 2006; 1:1–9.

11.

Romero-Gwynn

. Breast-feeding pattern among Indochinese immigrants in northern California. Am J Dis Child, 1989; 143:804–808.

12.

Snell

, Krantz

, Keeton

, et al. The association of formula samples given at hospital discharge with the early duration of breastfeeding. J Hum Lact, 1992; 8:67–72.

13.

Dennis

, Shiri

, Brown

, et al. Breastfeeding rates in immigrant and non-immigrant women: A systematic review and meta-analysis. Matern Child Nutr, 2019; 15:e12809.

14.

Kana

, Rodrigues

, Fonseca

, et al. Effect of maternal country of birth on breastfeeding practices: Results from Portuguese GXXI birth cohort. Int Breastfeed J, 2018; 13:15.

15.

Choudhry

, Wallace

. ‘Breast is not always best’: South Asian women's experiences of infant feeding in the UK within an acculturation framework. Matern Child Nutr, 2012; 8:72–87.

16.

Decree Law No. 62/2017. D.R. Series I. No. 112 (2017-06-09), pp. 2924–2944. Available at https://dre.pt/home/-/dre/107495709/details/maximized (accessed January 5, 2020).

17.

Instituto Nacional de Estatística. Partos (N.°) nos hospitais públicos de acesso universal e hospitais em parceria público-privada por Localização geográfica (NUTS - 2013) e Tipo de parto;anual [Births (No.) in public hospitals with universal access and hospitals in public-private partnership by geographic localization (NUTS - 2013) and type of delivery; annual]. 2017. Available at https://www.ine.pt/xportal/xmain?xpid=INE&xpgid=ine_indicadores&indOcorrCod=0008120&contexto=bd&selTab=tab2 (accessed January 5, 2020).

18.

Order No. 25360/2001. D. R. Series II. No. 286 (2001-12-12), pp. 20597–20598. Available at https://dre.pt/pesquisa/-/search/3622857/details/normal?q=Despacho+n.%C2%BA%2025360%2F2001 (accessed January 5, 2020).

19.

Directorate-General for Health. Information Bulletin No. 12/DQS/DMD (07/05/2009). Acesso dos Imigrantes ao Serviço Nacional de Saúde [Immigrants' Access to the National Health Service]. 2009. Available at https://www.dgs.pt/directrizes-da-dgs/orientacoes-e-circulares-informativas/circular-informativa-n-12dqsdmd-de-07052009.aspx (accessed January 5, 2020).

20.

Lisi

, De Freitas

, Barros

. Maternal country of birth and exclusive breastfeeding during the first in-hospital day in Portugal: The influence of the baby-friendly hospital initiative. J Hum Lact, 2021; 37:147–157.

21.

Gagnon

, DeBruyn

, Essén

, et al. Development of the Migrant Friendly Maternity Care Questionnaire (MFMCQ) for migrants to Western societies: An international Delphi consensus process. BMC Pregnancy Childbirth, 2014; 14:200.

22.

World Health Organization (WHO). Exclusive breastfeeding for optimal growth, development and health of infants. 2019. Available at https://www.who.int/elena/titles/exclusive_breastfeeding/en/ (accessed January 5, 2020).

23.

International Organization for Migration. Migrant. In Glossary on Migration (IML Series No. 34, 236). 2019. Available at https://publications.iom.int/system/files/pdf/iml_34_glossary.pdf (accessed January 5, 2020).

24.

Almeida

, Caldas

, Ayres-de-Campos

, et al. Assessing maternal healthcare inequities among migrants: A qualitative study. Cad Saude Publica, 2014; 30:333–340.

25.

De Freitas

, Massag

, Amorim

, et al. Involvement in maternal care by migrants and ethnic minorities: A narrative review. Public Health Rev, 2020; 41.

26.

Rocha

, Oliveira

, Ávila

, et al. Condicionantes da amamentação exclusiva na perspectiva materna [Conditioning factors for exclusive breastfeeding from the mother's perspective]. Cad Saude Publica, 2018; 34:e00045217.

27.

Bergevin

, Dougherty

, Kramer

. Do infant formula samples shorten the duration of breast-feeding?. Lancet, 1983; 321:1148–1151.

28.

Feldman-Winter

, Grossman

, Palaniappan

, et al. Removal of industry-sponsored formula sample packs from the hospital: Does it make a difference?. J Hum Lact, 2012; 28:380–388.

29.

Howard

, Howard

, Lawrence

, et al. Office prenatal formula advertising and its effect on breast-feeding patterns. Obstet Gynecol, 2000; 95:296–303.

30.

Rosenberg

, Eastham

, Kasehagen

, et al. Marketing infant formula through hospitals: The impact of commercial hospital discharge packs on breastfeeding. Am J Public Health, 2008; 98:290–295.

31.

Sadacharan

, Grossman

, Matlak

, et al. Hospital discharge bags and breastfeeding at 6 months: Data from the infant feeding practices study II. J Hum Lact, 2014; 30:73–79.

32.

Reiff

, Essock-Vitale

. Hospital influences on early infant-feeding practices. Pediatrics, 1985; 76:872–879.

33.

Yeong

JK.

Breaking the rules 2014 (BTR). Evidence of violations of the International Code of Marketing of Breast-milk Substitutes and subsequent resolutions compiled from January 2011 to December 2013. Penang, Malaysia: IBFAN-ICDC, 2014.

34.

Rossiter

, Yam

. Breastfeeding: How could it be enhanced? The perceptions of Vietnamese women in Sydney, Australia. J Midwifery Womens Health, 2000; 45:271–276.

35.

Vaughn

, Ireton

, Geraghty

, et al. Sociocultural influences on the determinants of breast-feeding by Latina mothers in the Cincinnati area. Fam Community Health, 2010; 33:318–328.

36.

World Health Organization (WHO). Cross-promotion of infant formula and toddler milks. Information note. 2019. Available at https://www.who.int/nutrition/publications/infantfeeding/information-note-cross-promotion-infant-formula/en/ (accessed January 5, 2020).

37.

Cattaneo

, Pani

, Carletti

, et al. Advertisements of follow-on formula and their perception by pregnant women and mothers in Italy. Arch Dis Child, 2015; 100:323–328.