Breastfeeding in the First Hours of Life Protects Against Pacifier Use: A Birth Cohort Study

Abstract

Objective:

To investigate risk factors for pacifier use in the first year of life.

Materials and Methods:

A prospective cohort study was conducted with children enrolled at birth in Porto Alegre, Southern Brazil, whose mothers underwent prenatal care at primary care units. Soon after the birth of the children, data were collected on anthropometrics, type of childbirth, and time until breastfeeding on the first day of life. At 6 and 12 months of age, data were collected on breastfeeding practices and whether the child had used a pacifier in the previous 6 months. Statistical analysis involved the use of Poisson regression with robust variance.

Results:

The incidence of pacifier use in the first year of life was 60% (317/532). The multivariable analysis showed that pacifier use in the first year of life was 33% higher when the mother was younger than 18 years of age (relative risk [RR] = 1.33; confidence interval [95% CI]: 1.01–1.76). Infants who breastfed in the first 30 minutes after birth had a 25% lower risk of pacifier use in the first year of life (RR = 0.75; 95% CI: 0.60–0.94), and those who breastfed between 30 minutes and 6 hours after birth had an 18% lower risk (RR = 0.82; 95% CI: 0.69–0.97) compared to those who took longer to begin breastfeeding or did not breastfeed.

Conclusions for Practice:

Breastfeeding soon after being born protected against pacifier use in the first year of life. This finding suggests pathways to improve child health, especially in the prenatal period and with an emphasis on pregnant adolescents.

Introduction

Pacifier use is highly prevalent in children throughout the world, but with considerable variability depending on the community evaluated.^1–3 In general, studies show that more than half of mothers offer a pacifier to their children at some point in the first year of life, with a predominance of pacifier use in the first month of life.^1,4–6

Evidence in the literature indicates pacifier use to be a risk factor for several general and oral health outcomes in the short and medium terms, such as the premature interruption of exclusive and total breastfeeding,^1,5,7–10 acute middle ear infection,^11,12 fungal infection,¹³ intestinal problems,¹⁴ childhood overweight and obesity,¹⁵ anterior open bite, accentuated overjet, and posterior crossbite in the primary dentition,^10,16,17 and traumatic dental injury in the primary dentition.¹⁸ However, pacifier use is reported to reduce the risk of sudden infant death syndrome, particularly when the child is placed for sleep.¹⁹ Regarding the long term, an epidemiological cohort study conducted in Southern Brazil described a strong association between prolonged pacifier use in childhood and the habit of smoking in adolescence and early adulthood.²⁰

The high prevalence and negative effects stemming from this practice indicate the need to establish effective interventions to diminish pacifier use. However, predictors for the occurrence of this habit are not yet clear, with few cohort studies available. To date, only one randomized clinical trial investigated the effect of an intervention on the frequency of pacifier use,²¹ and four longitudinal studies estimated risk factors for this habit.^6,14,22,23 It is generally recognized that the early introduction of a pacifier contributes to the interruption of breastfeeding,^2,4,5 but this relation could be bidirectional, that is, late initiation of breastfeeding increases the risk of pacifier use. However, no studies have performed a prospective evaluation of whether breastfeeding in the first hours of life contributes to a reduction in pacifier use in the first year of life.

The aim of the present study was to investigate risk factors for pacifier use in the first year of life in a birth cohort in the city of Porto Alegre, Brazil, with an emphasis on the effect of breastfeeding in the first hours after childbirth.

Materials and Methods

Study design and participants

The present prospective cohort study was performed as a secondary analysis of data collected during a cluster randomized clinical trial (clinicaltrials.gov NCT00635453),²⁴ conducted in the city of Porto Alegre (Southern Brazil), which has a population of 1.41 million and a fluoridated water supply (0.7 ppm F). The larger study included pregnant women recruited from 20 primary care units in the city. Units with at least 100 appointments with newborns per month and not involved in any other nutritional counseling program were considered eligible. The original study investigated the effect of a health professional training program to counsel recent mothers with regards to the impact of healthy feeding practices on different child health outcomes. The intervention only reduced the occurrence of dental caries in children whose mothers always visited the sample primary care unit or reported that the primary care unit was their main source of counseling.²⁴

At the onset of the study, 736 pregnant women (exclusion criterion: HIV+) were invited to participate in the study, 715 of whom consented. Team of researchers who had undergone training and calibration exercises performed the interviews during pregnancy and after childbirth, as well as at 6 months and 12 months of life of the children. The sample was composed of 619 and 532 children at 6 and 12 months of age, respectively.

The number of participants recruited and included in the study determined the sample size of the cohort study. A calculation was performed to determine whether this number (n = 532) was sufficient to estimate the effect of exposure (interruption of breastfeeding) on pacifier use in the first year of life. Considering a confidence interval (95% CI), 95% power, 1:7 ratio of exposed to nonexposed, and frequencies of 89% and 58% of pacifier use among exposed and nonexposed infants, respectively,²¹ the minimum sample size would be n = 233. Thus, even with the addition of 20% for the multivariable analysis, the actual sample size (n = 532) was sufficient for the purposes of the study.

Data collection

Trained field researchers collected the sociodemographic and behavioral data at baseline with the aid of structured questionnaires. The following data were collected: mother's age at the birth of the child, mother's schooling (complete years of study), family structure (nuclear [living with both parents] versus non-nuclear [living with one or neither parent]), child's birth order, family income (in Brazilian currency [R$] and categorized in quartiles using the Brazilian monthly minimum wage as reference), and whether anyone in the home was a smoker. After childbirth, the following data were collected: anthropometrics (birth weight and length), type of birth, and time (in hours) to the onset of breastfeeding on the first day of life (categorized as up to 30 minutes, from 30 minutes to 6 hours, and >6 hours or no breastfeeding on the first day).

At 6 and 12 months of the children's lives, data were collected on breastfeeding practices and other eating practices for the larger study and whether the child had used a pacifier. The outcome in the present study was the use of a pacifier at any time in the first year of life (data collected at 12 months).

Data analysis

The SPSS program, version 20.0, was used for the statistical analysis. The frequency of pacifier use was compared according to the sociodemographic, anthropometric, and behavioral characteristics using the chi-squared test. Poisson regression with robust variance was then performed, with the calculation of relative risk (RR) and respective CIs. A hierarchical model with three levels was used for the multivariable analysis²⁵: (1) demographic and socioeconomic variables (child's sex, mother's age at the birth of the child, mother's schooling, income per capita, family structure, and whether the infant was the first child); (2) birth-related variables (type of birth, birth weight, and length); and (3) behavioral variables (smoker in the home and time until onset of breastfeeding in first hours of life). All variables were maintained in the multivariable model on each level, independent of the significance level after the multivariable adjustment. The final model estimated RRs with adjustments for variables on the same or higher level. “Group” (intervention or control) was maintained as an adjustment variable on all levels.

Ethical aspects

This study received approval from the Human Research Ethics Committee of the Federal University of Health Sciences of Porto Alegre (certificate number: 471-07). The mothers signed a statement of informed consent agreeing to the participation of the children. Children with dental caries, anemia, or obesity were referred to a local primary care unit.

Results

Table 1 displays the demographic, anthropometric, and socioeconomic characteristics of the families at baseline and when the children were 12 months of age, demonstrating similarities between the two groups. Among the 715 children that began the study, 183 (25.6%) were lost to follow-up. The final sample was composed of 532 children, with a predominantly low socioeconomic status: 85.1% with an income per capita less than or equal to the Brazilian monthly minimum wage, 78.3% categorized in class C or lower, and a mean of 8.6 years of mother's schooling. No significant differences were found between the children analyzed and those lost to follow-up in terms of mother's schooling (p = 0.232), income (p = 0.460), pregestational body mass index (p = 0.139), or mother's age at the birth of the child (p = 0.230).

Table 1.

Characteristics of Participants of the Cohort

Variables	Initial cohort (N = 715)		12-Month sample (N = 532)
Variables	(%)	Mean (SD)	(%)	Mean (SD)
Mother's schooling (years)		8.5 (2.7)		8.6 (2.8)
Pregestational BMI		24.6 (4.8)		24.4 (4.7)
Income per capita ≤BMMW	(82.4)		(85.1)
Social class ≤C	(79.8)		(78.3)
Mother's age at birth of child		25.4 (6.7)		25.7 (6.8)
Child's sex (male)	(52.4)		(52.4)
Familiar structure (nuclear)	(52.0)		(49.8)
First child	(44.6)		(45.9)

BMI, body mass index; BMMW, Brazilian monthly minimum wage.

The incidence of pacifier use in the first year of life was 60% (317/532), and all these children began the habit in the first 6 months of life. Pacifier use was significantly greater among boys (p = 0.038), children of mothers <18 years of age (p = 0.042), and children who began breastfeeding >6 hours after being born or did not breastfeed (p = 0.028). Pacifier use was not associated with socioeconomic characteristics of the families or the anthropometric characteristics of the children (Table 2).

Table 2.

Pacifier Use According to the Independent Variables

Variables	N (%)	Pacifier use	p
Variables	N (%)	N (%)
All	532 (100.0)	317 (59.6)
Sex
Male	279 (52.4)	178 (63.8)	0.038
Female	253 (47.6)	139 (54.9)	0.038
Mother's age at birth of child
<18 years	54 (10.2)	36 (66.7)	0.042^a
18 to 34 years	389 (73.1)	236 (60.7)
≥35 years	89 (16.7)	45 (50.6)
Mother's schooling
<8 years	159 (29.9)	99 (62.3)	0.411
≥8 years	373 (70.1)	218 (58.4)	0.411
Income per capita
First quartile	127 (24.6)	81 (63.8)	0.933^a
Second quartile	145 (28.0)	79 (54.5)
Third quartile	118 (22.8)	68 (57.6)
Fourth quartile	127 (24.6)	80 (63.0)
Family structure
Nuclear	265 (49.8)	154 (58.1)	0.490
Non-nuclear	267 (50.2)	163 (61.0)	0.490
First child
No	288 (54.1)	169 (58.7)	0.644
Yes	244 (45.9)	148 (60.7)	0.644
Type of childbirth
Normal	365 (68.9)	214 (58.6)	0.489
Cesarean	165 (31.1)	102 (61.8)	0.489
Birth weight
<2,500 g	36 (6.8)	22 (61.1)	0.854
≥2,500 g	492 (93.2)	293 (59.6)	0.854
Length at birth
<48 cm	127 (24.2)	76 (59.8)	0.967
≥48 cm	398 (75.8)	239 (60.1)	0.967
First breastfeeding after birth
<30 minutes	117 (22.2)	63 (53.8)	0.028^a
30 minutes to 6 hours	334 (63.5)	195 (58.4)
>6 hours	75 (14.3)	53 (70.7)
Smoker in home
No	299 (56.2)	176 (58.9)	0.700
Yes	233 (43.8)	141 (60.5)	0.700

Chi-squared test for linear trend.

Table 3 displays the crude and adjusted effect measures for pacifier use. The multivariable analysis showed that the risk of pacifier use in the first year of life was 16% greater among boys compared to girls (RR = 1.16; 95% CI: 1.01–1.34) and 33% higher when the mother was younger than 18 years of age compared to mothers aged 35 years or older (RR = 1.33; 95% CI: 1.01–1.76). Breastfeeding in the first hours of life was a protection factor against pacifier use: infants who breastfed in the first 30 minutes after birth had a 25% lower risk of pacifier use in the first year of life (RR = 0.75; 95% CI: 0.60–0.94), and those who breastfed between 30 minutes and 6 hours after birth had an 18% lower risk (RR = 0.82; 95% CI: 0.69–0.97) compared to those who took longer to begin breastfeeding or who did not breastfeed. None of the anthropometric variables at the birth of the child or socioeconomic variables was associated with the outcome after the adjustment for confounding factors.

Table 3.

Univariable and Multivariable Regression: Relative Risk and Respective 95% Confidence Intervals for Pacifier Use in First Year of Life According to Independent Variables

Variables	Crude model			Adjusted model^a
Variables	RR	95% CI	p	RR	95% CI	p
Demographic and socioeconomic variables
Sex
Male	1.16	(1.01–1.34)	0.040	1.16	(1.01–1.34)	0.039
Female	1.00		0.040	1.00		0.039
Mother's age at birth of child
<18 years	1.32	(1.00–1.74)	0.050	1.33	(1.01–1.76)	0.045
18 to 34 years	1.20	(0.96–1.45)	0.105	1.19	(0.95–1.48)	0.128
≥35 years	1.00			1.00
Mother's schooling
<8 years	1.06	(0.92–1.24)	0.403	1.06	(0.89–1.26)	0.532
≥8 years	1.00		0.403	1.00		0.532
Income per capita
First quartile	1.01	(0.84–1.22)	0.896	0.95	(0.76–1.19)	0.668
Second quartile	0.86	(0.71–1.06)	0.154	0.83	(0.67–1.02)	0.079
Third quartile	0.91	(0.75–1.12)	0.393	0.88	(0.72–1.08)	0.222
Fourth quartile	1.00			1.00
Family structure
Nuclear	1.00		0.491	1.00		0.973
Non-nuclear	1.05	(0.91–1.21)	0.491	1.00	(0.86–1.16)	0.973
First child
No	0.97	(0.84–1.11)	0.643	1.03	(0.87–1.22)	0.736
Yes	1.00		0.643	1.00		0.736
Variables at birth
Type of childbirth
Normal	1.00		0.482	1.00		0.438
Cesarean	1.05	(0.91–1.22)	0.482	1.06	(0.91–1.23)	0.438
Birth weight
<2,500 g	1.03	(0.78–1.34)	0.852	1.10	(0.82–1.46)	0.535
≥2,500 g	1.00		0.852	1.00		0.535
Length at birth
<48 cm	1.00	(0.85–1.17)	0.967	1.01	(0.83–1.22)	0.918
≥48 cm	1.00		0.967	1.00		0.918
Behavioral variables
Smoker in home
No	1.00		0.723	1.00		0.723
Yes	1.03	(0.89–1.18)	0.723	0.97	(0.84–1.13)	0.723
First breastfeeding after birth
<30 minutes	0.76	(0.61–0.95)	0.017	0.75	(0.60–0.94)	0.014
30 minutes to 6 hours	0.83	(0.70–0.98)	0.029	0.82	(0.69–0.97)	0.024
>6 hours	1.00			1.00

Multivariable estimates adjusted for variables on the same or a higher level.

CI, confidence interval; RR, relative risk.

As an additional analysis, we investigated if the relationship between breastfeeding and pacifier use also occurred in the opposite direction, that is, if early introduction of a pacifier (first month of life) was associated with shorter breastfeeding duration. The results showed that the prevalence of exclusive breastfeeding at 4 months of age was, respectively, 14.4% (29/201) and 32.9% (107/305) in children who used and those who did not use a pacifier in the first month of life (p < 0.001). Furthermore, the prevalence of total breastfeeding at 6 months of age was, respectively, 38.8% (78/201) and 72.6% (236/325) in children who did not use and those who used a pacifier in the first month of life (p < 0.001).

Discussion

The finding that 6 out of 10 children used a pacifier in the first year of life is consistent with data reported in Brazil and other countries,^14,17,21,22 indicating that most children are exposed to the consequences of this habit to general and oral health, which are widely described in the literature.^1,5,7,16,20 These data underscore the importance of pacifier prevention strategies based on knowledge of the network of causality.

The most important finding in the present cohort study was that beginning breastfeeding in the first hours of life and especially in the first 30 minutes of life is a protection factor against pacifier use in the first year of life. To the best of our knowledge, this is the first prospective study to describe such an association.

Although the association between a shorter breastfeeding time and pacifier use has been described in cross-sectional studies,^26,27 this design does not enable establishing causality or the direction of the association. Moreover, the focus of previous longitudinal studies was to evaluate the association in the opposite direction, with the majority concluding that the early introduction of a pacifier is a risk factor for the premature interruption of breastfeeding in the subsequent months.^4,5 Other researchers who have evaluated this direction of the association suggest that a pacifier is only a marker of the difficulty or low motivation of the mother with regards to breastfeeding.^4,28

The introduction of a pacifier generally occurs in the first or second month of life,²¹ making it difficult for traditionally collected breastfeeding variables—months of exclusive or total breastfeeding—to represent adequate exposure for pacifier use. In the present study, the determination of the time in hours to the onset of breastfeeding enabled obtaining information on exposure before a pacifier could be offered to the infant. Thus, reverse causality—at least with regards to the first category evaluated (first 30 minutes)—could not have occurred, ensuring that exposure preceded the outcome, which is one of the principles of causality. Moreover, a dose–response effect, although tenuous, was found, which is another principle that suggests causality. Interestingly, the additional analysis showed that, in this sample, the association occurred in both directions.

The World Health Organization recommends that breastfeeding should begin in the first hour after childbirth,²⁹ considering the proneness of newborns to seek the nipple at this time, as well as the potential benefits for a longer breastfeeding time³⁰ and reduction in neonatal mortality.³¹ Although the mechanism by which breastfeeding soon after childbirth reduces pacifier use was not investigated in the present study, it is plausible that the exposure contributes positively to the mother–infant bond.³² Moreover, there is evidence that skin-to-skin contact in the first 2 hours of life, including breastfeeding, contributes decisively to a better temperament of the child and a better future mother–child relationship in the first year of the child's life.³³ It is generally recognized that the pacifier is offered to an infant due to its calming effect, avoiding or diminishing crying and agitation.^1,26,28 If this role is previously played by breastfeeding, it is possible that there is less need to offer a pacifier. In some families, this situation may continue for weeks after the birth of the child and pacifier use is not introduced.

The greater risk of pacifier use among children of mothers <18 years of age had been described in children in the city of São Luís, northeastern Brazil,²³ and Bristol, England.¹⁴ It is possible that adolescent girls are less prepared for motherhood and have less emotional strength to deal with the complex issues of managing an infant,^34,35 turning to this device because they believe that it is the only effective method for calming a child in moments of crying.⁶ The difference in the use of a pacifier between boys and girls has been reported in different publications, always with a greater frequency among boys.^6,14,22 It is possible that the way parents deal with difficulties involving infants differs between sexes. Differences in infant behavior in the first months of life hardly explain the difference found between sexes. Some authors have raised the hypothesis that the greater frequency of crying among boys may be the reason why pacifiers are offered more often to this sex.²²

Divergent results are found in the literature regarding the effect of socioeconomic level on pacifier use. Some studies describe a greater risk among children with a low socioeconomic status¹⁴; others describe a greater risk among those with a higher socioeconomic status²³; and others have found no association.²¹ In the present study, no association was found between pacifier use and socioeconomic status, suggesting that this habit affects all social layers of the community.

Interestingly, the anthropometric characteristics of the children at birth, including low birth weight, did not exert an influence on pacifier use. As only one child in the sample had a very low birth weight (<1,500 g), it is not possible to analyze the effect of this variable, which was previously identified as a risk factor among children in northeastern Brazil.²³

A randomized clinical trial with mother–infant pairs in Southern Brazil demonstrated that counseling on healthy eating practices in the first year of life, including the promotion of exclusive breastfeeding, seems to contribute to a reduction in pacifier use.²¹ An observational study conducted in northeastern Brazil showed that the probability of using an artificial nipple (pacifier or feeding bottle) was lower among children whose mothers received counseling on breastfeeding during prenatal and postnatal care.³⁶ One of the clinical implications of the present study is the potential for prenatal counseling for the promotion of breastfeeding in the first hours after childbirth to be a simple more objective intervention with no additional cost for a reduction in pacifier use with the potential to reduce the negative health outcomes associated with this habit.

Psychosocial variables, such as a lower level of mother's trust in breastfeeding and postpartum depression, were, respectively, considered risk factors for greater pacifier use in studies conducted in New Zealand and Brazil.^21,22 Future studies could contribute more to the understanding of this issue if maternal psychosocial characteristics that exert an influence on inequalities in children's health outcomes, such as perceived seriousness, fatalisms, locus of control, and self-efficacy, were collected in the prenatal period.³⁷

The present study has limitations that should be considered, particularly the considerable proportion of dropouts between baseline and follow-up. However, the possibility of selection bias is low, as the characteristics of the initial cohort and follow-up sample were virtually the same. A certain degree of imprecision may also have occurred with regards to the time until the first breastfeeding, which was the main exposure of the study. However, it is unlikely that an occasional information bias regarding this variable influenced the results, considering the small amount of time between the practice and the data collection. The children in the present study were born in public hospitals and were followed up through the public health care system, making them representative of the population that should be the target of public policies.

Conclusion

The results of the present study indicate that the introduction of breastfeeding in the first hours of life protects against pacifier use in the first year of life. Moreover, infants of adolescent mothers and male infants were at greater risk of pacifier use. Taken together, these findings suggest pathways for improving cost-effective strategies aimed at the promotion of child health, especially in the prenatal period and with an emphasis on pregnant adolescents.

Footnotes

Disclosure Statement

No competing financial interests exist.

Funding Information

This work was supported by the Coordination for the Improvement of Higher Level Education Personnel (CAPES), Brazil.

References

Mauch

, Scott

, Magarey

, et al. Predictors of and reasons for pacifier use in first-time mothers: An observational study. BMC Pediatr, 2012; 12:7.

Nelson

EAS

, Yu

L-MM

, Williams

, et al. International child care practices study: Breastfeeding and pacifier use. J Hum Lact, 2005; 21:289–295.

Ministry of Health of Brazil. II research of breastfeeding predominance in Brazilian capitals and Federal District. Brasília: Ministry of Health, 2009.

Victora

, Behague

, Barros

, et al. Pacifier use and short breastfeeding duration: Cause, consequence, or coincidence?. Pediatrics, 1997; 99:445–453.

Feldens

, Vitolo

, Rauber

, et al. Risk factors for discontinuing breastfeeding in Southern Brazil: A survival analysis. Matern Child Health J, 2012; 16:1257–1265.

Soares

, Giugliani

ERJ

, Braun

, et al. Pacifier use and its relationship with early weaning in infants born at a Child-Friendly Hospital. J Pediatr (Rio J), 2003; 79:309–316.

Buccini G dos

, Pérez-Escamilla

, Paulino

, et al. Pacifier use and interruption of exclusive breastfeeding: Systematic review and meta-analysis. Matern Child Nutr, 2017; 13:e12384.

Karabulut

, Yalçın

, Özdemir-Geyik

, et al. Effect of pacifier use on exclusive and any breastfeeding: A meta-analysis. Turk J Pediatr, 2009; 51:35–43.

Buccini G dos

, Pérez-Escamilla

, D'Aquino Benicio

, et al. Exclusive breastfeeding changes in Brazil attributable to pacifier use. PLoS One, 2018; 13:e0208261.

10.

Rocha

, Verga

, Sipsma

, et al. Pacifier use and breastfeeding: A qualitative study of postpartum mothers. Breastfeed Med, 2020; 15:24–28.

11.

Rovers

, Numans

, Langenbach

, et al. Is pacifier use a risk factor for acute otitis media? A dynamic cohort study. Fam Pract, 2008; 25:233–236.

12.

Niemelä

, Pihakari

, Pokka

, et al. Pacifier as a risk factor for acute otitis media: A randomized, controlled trial of parental counseling. Pediatrics, 2000; 106:483–488.

13.

Mattos-Graner

, de Moraes

, Rontani

, et al. Relation of oral yeast infection in Brazilian infants and use of a pacifier. ASDC J Dent Child, 2001; 68:33–36, 10.

14.

North

, Fleming

, Golding

. Pacifier use and morbidity in the first six months of life. Pediatrics, 1999; 103:e34.

15.

Hohman

, Savage

, Birch

, et al. Pacifier use and early life weight outcomes in the intervention nurses start infants growing on healthy trajectories study. Child Obes, 2018; 14:58–66.

16.

Schmid

, Kugler

, Nalabothu

, et al. The effect of pacifier sucking on orofacial structures: A systematic literature review. Prog Orthod, 2018; 19:8.

17.

da Costa

, Shqair

, Azevedo

, et al. Pacifier use modifies the association between breastfeeding and malocclusion: A cross-sectional study. Braz Oral Res, 2018; 32:e101.

18.

Kramer

, Feldens

, Bruch

, et al. Clarifying the effect of behavioral and clinical factors on traumatic dental injuries in childhood: A hierarchical approach. Dent Traumatol, 2015; 31:177–183.

19.

Hauck

, Omojokun

, Siadaty

. Do pacifiers reduce the risk of sudden infant death syndrome? A meta-analysis. Pediatrics, 2005; 116:e716–e723.

20.

Ferreira

, Rosa

, Antunes

JLF

, et al. Prolonged pacifier use during infancy and smoking initiation in adolescence: Evidence from a Historical Cohort Study. Eur Addict Res, 2015; 21:33–38.

21.

Feldens

, Ardenghi

, Cruz

, et al. Advising mothers about breastfeeding and weaning reduced pacifier use in the first year of life: A randomized trial. Community Dent Oral Epidemiol, 2013; 41:317–326.

22.

Vogel

, Hutchison

, Mitchell

. The impact of pacifier use on breastfeeding: A prospective cohort study. J Paediatr Child Health, 2001; 37:58–63.

23.

Martins

RFM

, Lima

AASJ

, Ribeiro

CCC

, et al. Lower birthweight, shorter breastfeeding and lack of primary health care contributed to higher pacifier use in a birth cohort. Acta Paediatr Int J Paediatr, 2018; 107:1650–1651.

24.

Chaffee

, Feldens

, Vítolo

. Cluster-randomized trial of infant nutrition training for caries prevention. J Dent Res, 2013; 92:S29–S36.

25.

Victora

, Huttly

, Fuchs

, et al. The role of conceptual frameworks in epidemiological analysis: A hierarchical approach. Int J Epidemiol, 1997; 26:224–227.

26.

Buccini G dos

, Benicio

MHDA

, Venancio

. Determinants of using pacifier and bottle feeding. Rev Saude Publica, 2014; 48:571–582.

27.

Carcavalli

, Martins

, Rocha

, et al. Preterm birth, pacifier use and breastfeeding: Is there a relationship?. Braz Dent J, 2018; 29:388–394.

28.

Kramer

, Barr

, Dagenais

, et al. Pacifier use, early weaning, and cry/fuss behavior: A randomized controlled trial. J Am Med Assoc, 2001; 286:322–326.

29.

WHO

UNICEF

. Implementation Guidance: Protecting, Promoting, and Supporting Breastfeeding in Facilities Providing Maternity and Newborn Services—The Revised Baby-Friendly Hospital Initiative 2018. Geneva: World Health Organization, 2018.

30.

Esteves

TMB

, Daumas

, de Oliveira

MIC

, et al. Factors associated to breastfeeding in the first hour of life: Systematic review. Rev Saude Publica, 2014; 48:697–708.

31.

Phukan

, Ranjan

, Dwivedi

. Impact of timing of breastfeeding initiation on neonatal mortality in India. Int Breastfeed J, 2018; 13:27.

32.

Widström

A-M

, Wahlberg

, Matthiesen

A-S

, et al. Short-term effects of early suckling and touch of the nipple on maternal behaviour. Early Hum Dev, 1990; 21:153–163.

33.

Bystrova

, Ivanova

, Edhborg

, et al. Early contact versus separation: Effects on mother-infant interaction one year later. Birth, 2009; 36:97–109.

34.

Sipsma

, Jones

, Nickel

. Hospital practices to promote breastfeeding: The effect of maternal age. Birth, 2017; 44:272–280.

35.

Muelbert

, Giugliani

ERJ

. Factors associated with the maintenance of breastfeeding for 6, 12, and 24 months in adolescent mothers. BMC Public Health, 2018; 18:675.

36.

Batista

CLC

, Rodrigues

, Ribeiro

, et al. Nutritive and non-nutritive sucking patterns associated with pacifier use and bottle-feeding in full-term infants. Early Hum Dev, 2019; 132:18–23.

37.

Harris

, Pennington

, Whitehead

. Preventive dental visiting: A critical interpretive synthesis of theory explaining how inequalities arise. Community Dent Oral Epidemiol, 2017; 45:120–134.