Breastfeeding Is Associated with Decreased Childhood Maltreatment

Abstract

Purpose:

Child maltreatment has serious implications for youth outcomes, yet its associations with early parenting practices are not fully understood. This study investigated whether breastfeeding practices are correlated with childhood maltreatment.

Materials and Methods:

Data were utilized from the National Longitudinal Study of Adolescent to Adult Health, a nationally representative and longitudinal study of adolescents. The analytic sample comprised 4,159 adolescents. The outcome variables included four subtypes of childhood maltreatment (neglect, inadequate supervision, physical abuse, and sexual abuse). The primary independent variable was breastfeeding duration. Covariates of the child, mother, and household were included in analyses. Logistic regression models were employed to predict odds of maltreatment subtypes from breastfeeding duration and covariates.

Results:

Compared with adolescents never breastfed, adolescents breastfed 9 months or longer had a reduced odds of having experienced neglect (odds ratio [OR] = 0.54; 95% confidence interval [CI] = 0.35–0.83) and sexual abuse (OR = 0.47; 95% CI = 0.24–0.93) after controlling for covariates.

Conclusions:

Breastfeeding duration is significantly associated with decreased childhood neglect and sexual abuse. Breastfeeding practices should be explored as a consideration among clinicians when assessing maltreatment risk. Further research should examine whether a causal relationship exists between breastfeeding and decreased maltreatment.

Introduction

In 2014, there were with more than 3 million investigations for child maltreatment in the United States, resulting in roughly 700,000 substantiated cases.¹ Child maltreatment comprised neglecting a child's basic needs, providing inadequate supervision, or injuring a child physically, emotionally, or sexually. The short- and long-term implications of child maltreatment are severe, which includes increased risk for failure to thrive, impairments in brain development, poorer mental health outcomes, lower academic achievement, and incarceration later in life.^2–6 Most significantly, child abuse and neglect resulted in 1,580 deaths in the United States in 2014.¹

A thorough understanding of the risk factors for child maltreatment is crucial to provide necessary services for affected children and prevent future occurrences. Presently, there are a variety of known risk factors for child maltreatment. Particularly, children under the age of 3 years and individuals with special medical needs are at an increased risk for child maltreatment.^1,7,8 Furthermore, a variety of parental characteristics are associated with neglecting or abusing one's child, in which significant associations have been found between child maltreatment with age, marital status, education, and employment.^7,8 In addition to these aforementioned demographic characteristics of parents, several parenting practices have also been linked with risk for child maltreatment. Specifically, authoritative parenting styles, parental display of warmth, and parental involvement in child's activities have all been associated with child maltreatment risk.⁹ However, the relationships between maltreatment risk and additional parenting practices have been less researched.

In particular, breastfeeding has been widely acknowledged as a beneficial parenting practice due to its associations with improved infant health and cognitive abilities.^10,11 Nonetheless, research on the relationship between breastfeeding and child maltreatment is minimal. To date, only one known study has considered the role of breastfeeding on risk for child maltreatment. Strathearn et al.¹² examined an Australian cohort of 7,223 maternal–infant pairs, finding risk for maternal neglect to be negatively associated with breastfeeding duration. Although few empirical studies have explored the relationship between breastfeeding and maltreatment, knowledge from theory and research indicates that breastfeeding may enhance maternal responsiveness, which may protect children from abuse and neglect. Specifically, research found mothers who breastfed their newborns were more responsive to their children's needs and had less depressive symptoms.^13,14 As children age, this translates into stronger parent–child relationships and enhanced maternal abilities to address her child's needs. The present research study seeks to further research findings by Strathearn et al.¹² to investigate the association between infant breastfeeding practices and future childhood maltreatment and to examine whether this relationship varies by type of maltreatment and duration of breastfeeding. We hypothesize that increased duration of breastfeeding will be associated with a decrease in child maltreatment.

Materials and Methods

Study design and sample

The research question was investigated utilizing data from Wave I and Wave III of the National Longitudinal Study of Adolescent to Adult Health (Add Health), a nationally representative and longitudinal study of >20,000 adolescents to assess social influences on adolescent health. Adolescents were selected for the study using a multistage probability sampling design; the sampling process has been described previously.¹⁵ Data were collected from a variety of sources, including interviews with the adolescents, parents, and school administrators, and biological specimens. Information was collected across four waves, with Wave I collected during 1994–1995 when participants were in grades 7–12. Wave III was collected during 2001–2002 when participants were young adults aged 18–26 years.

Adolescents were included in the analytic sample if they had valid information on all child maltreatment outcomes (N = 7,836). Among adolescents with child maltreatment information, adolescents were excluded if they did not have valid information on breastfeeding (N = 1,325). The study further limited adolescents who did not have their biological mother to complete an interview in Wave I (N = 731), as biological mothers are more likely to report reliable information on breastfeeding. Finally, adolescents with missing values on variables listed in Table 1 were excluded; the final sample size was 4,159.

Table 1.

Characteristics of the Analytic Sample (N = 4,159)

Variables	Mean (SD) or %
Maltreatment outcomes
Neglect	10.4%
Inadequate supervision	38.4%
Physical abuse	29.1%
Sexual abuse	4.6%
Child characteristics
Gender (male)	46.4%
Age (years)	15.4 (1.7)
Race
Caucasian	57.3%
African American	19.1%
Hispanic	7.1%
Asian	5.4%
Other	11.1%
Duration of breastfeeding
No breastfeeding	53.2%
1–2 months	15.2%
3–5 months	11.1%
6–8 months	7.2%
9 months or more	13.3%
Maternal characteristics
Age at child's birth (years)	25.9 (5.3)
Education
Less than high school	19.1%
High school diploma or GED	29.0%
Some college	28.7%
College degree or greater	23.1%
Household characteristics
Able to pay bills (yes)	81.4%
Welfare receipt (yes)	8.8%
Household income ($)	43,337 (49,006)

GED, General Educational Development; SD, standard deviation.

Outcome variables

Four subtypes of child maltreatment were utilized as the outcome variables. These subtypes were collected through computer-assisted self-interviewing in Wave III, when adolescents were aged 18–26 years. Participants were asked how often a parent or adult caregiver had engaged in a series of maltreatment-related events by the time the adolescent had started sixth grade. The maltreatment subtypes were categorized as (1) neglect—“not taken care of your basic needs, such as keeping you clean or providing food or clothing”; (2) inadequate supervision—“left you home alone when an adult should have been with you”; (3) physical abuse—“slapped, hit, or kicked you”; and (4) sexual abuse—“touched you in a sexual way, forced you to touch him or her in a sexual way, or forced you to have sexual relations.” Response options for each individual subtype included 1 time, 2 times, 3–5 times, 6–10 times, >10 times, or this has never happened. For this study, each subtype was dichotomized to reflect whether the adolescent had ( = 1) or had not ( = 0) experienced each maltreatment type.

Independent variable

The primary independent variable was duration of breastfeeding, collected from the parent interview during Wave I. Parents responded to how long the child was breastfed, with response options including (1) <3 months, (2) 3 months to <6 months, (3) 6 months to <9 months, (4) 9 months to <12 months, (5) 12 months to <24 months, (6) 24 months or more, (7) and he/she was not breastfed. For this study, categories 4 through 6 were collapsed into one category of 9 months or more due to small cell size.

Covariates

Three groups of covariates were adjusted for in analyses, including characteristics of the child, mother, and household. Covariates were collected during Wave I. Child-level covariates included gender, age at time of interview (measured in years), race, birthweight, receipt of special educational services, and self-rated general health. Maternal characteristics included age at child's birth (measured in years), education, employment status, self-rated general health, and marital status. Household-level covariates included parent response to whether family has enough money to pay bills, receive welfare, receive food stamps, receive aid for dependent children, household size, and household income (measured continuously in U.S. dollars).

Statistical analyses

Descriptive analyses were conducted for all outcome variables, independent variable, and covariates on the entire sample. After descriptive analyses, a logistic regression model was specified using backward selection to identify significant predictor variables that should be included in the final regression models. The outcome variable in the initial regression model was a dichotomized measure of maltreatment, indicating whether the participant had experienced any of the four subtypes of maltreatment. Breastfeeding and the aforementioned covariates were included as independent variables in the initial model, along with several additional covariates of the child, mother, and household. Using the backward selection process, the variable with the highest p-value was removed from the regression model and the multivariate analyses were run eliminating variables one at a time until all variables remaining in the model had p < 0.20. Breastfeeding duration remained in the model as the primary independent variable. The aforementioned covariates also remained in the model. Covariates that were excluded during this process consisted of child's birthweight, child's receipt of special education services, child's health, mother's health, mother's employment status, mother's marital status, family receipt of food stamps, family receipt of aid for dependent children, and household size.

Using the model specified with backward selection, four logistic regression models were conducted to examine the relationship between each maltreatment subtype with breastfeeding duration, adjusting for covariates. For all statistical analyses, weighted estimates to account for data attrition and standard errors were computed using Stata 13.1/IC (StataCorp. Stata Statistical Software: Release 13. College Station, TX: StataCorp, LP, 2013). Standard error and complex survey sampling design were adjusted by the study weights, primary sampling unit, and stratum.

Results

Descriptive results

Table 1 provides a description of the study sample. With regard to the subtypes of child maltreatment, results demonstrate that inadequate supervision and physical abuse are the most common forms of maltreatment, whereas neglect and sexual abuse occur less frequently. Particularly, nearly 38.4% affirmed that they had been left home alone on at least one occasion, whereas 29.1% of the sample had been slapped, hit, or kicked by a parent or adult caregiver at least once. Furthermore, 10.4% of the sample responded that, on at least one occasion before sixth grade, one of their basic needs had not been taken care of by a parent or adult caregiver, an indicator of neglect. Finally, 4.6% of the sample had been sexually abused by a parent or caregiver. Characteristics of the adolescents indicate that the sample comprised slightly more females (53.6%) than males (46.4%), an average age of 15 years, and primarily of Caucasian (57.3%) or African American (19.1%) race. Over half of the sample was never breastfed as infants (53.2%). Mothers of the adolescents were, on average, 26 years old when their interviewed child was born, and the majority had either received a high school diploma (29.0%) or had completed some college (28.7%). Households were typically able to pay their bills (81.4%), not receiving welfare (91.2%), and had an average income of nearly $43,000. Table 2 displays the proportion maltreated by the different lengths of breastfeeding. Children breastfed for 9 months or longer had the lowest proportion maltreated for each subtype. For example, 12% of children never breastfed experienced neglect, whereas only 6% of children breastfed 9 months or longer experienced neglect.

Table 2.

Proportion Maltreated by Duration of Breastfeeding

Breastfeeding duration	Neglected	Physical abuse	Sexual abuse	Inadequate supervision
Never	12%	29%	5%	39%
1–3 months	11%	30%	5%	42%
3–6 months	9%	30%	5%	43%
6–9 months	9%	31%	4%	35%
>9 months	6%	26%	3%	34%

Analytic results

Table 3 presents results of four logistic regression models, in which each subtype of child maltreatment was predicted by breastfeeding duration and demographic covariates. Across the four subtypes, breastfeeding duration was a significant predictor of neglect and sexual abuse during childhood. After controlling for covariates, results indicate that the odds of experiencing childhood neglect are lower if breastfed 9 months or longer compared with those never having been breastfed (odds ratio [OR] = 0.54; 95% confidence interval [CI] = 0.35–0.83). Similarly, the odds of childhood sexual abuse are lower if breastfed 9 months or longer compared with those never having been breastfed (OR = 0.47; 95% CI = 0.24–0.93). Breastfeeding duration did not significantly predict whether adolescents had experienced physical abuse or inadequate supervision during childhood.

Table 3.

Results of Logistic Regression Analyses (N = 4,159)

Variables	Neglect, OR (95% CI)	Physical abuse, OR (95% CI)	Sexual abuse, OR (95% CI)	Supervision, OR (95% CI)
Duration of breastfeeding
No breastfeeding	Reference	Reference	Reference	Reference
1–2 months	0.96 (0.69–1.34)	1.10 (0.88–1.38)	1.20 (0.75–1.93)	1.07 (0.87–1.32)
3–5 months	0.82 (0.56–1.22)	1.11 (0.86–1.43)	1.10 (0.63–1.92)	1.17 (0.92–1.48)
6–8 months	0.98 (0.61–1.58)	1.19 (0.87–1.62)	1.26 (0.64–2.50)	0.91 (0.68–1.23)
9 months or more	0.54 (0.35–0.83)^**	0.81 (0.63–1.04)	0.47 (0.24–0.93)^*	0.89 (0.71–1.12)
Child covariates
Gender
Male	Reference	Reference	Reference	Reference
Female	0.51 (0.40–0.65)^**	0.79 (0.68–0.93)^**	1.20 (0.85–1.69)	0.80 (0.70–0.93)^**
Age	0.88 (0.82–0.95)^**	0.99 (0.95–1.03)	0.94 (0.85–1.03)	0.92 (0.88–0.96)^**
Race
Caucasian	Reference	Reference	Reference	Reference
African American	1.14 (0.84–1.56)	1.02 (0.83–1.27)	0.91 (0.55–1.50)	1.06 (0.87–1.29)
Hispanic	1.22 (0.79–1.88)	1.13 (0.83–1.53)	1.00 (0.52–1.90)	1.07 (0.81–1.43)
Asian	2.00 (1.23–3.25)^**	1.71 (1.22–2.39)^**	1.74 (0.87–3.47)	1.95 (1.40–2.71)^**
Other	1.52 (1.07–2.15)^*	1.36 (1.06–1.75)^*	1.46 (0.88–2.41)	1.37 (1.09–1.74)^**
Maternal covariates
Age at child's birth	0.97 (0.95–0.99)^**	0.97 (0.96–0.99)^**	0.98 (0.95–1.01)	0.97 (0.96–0.99)^**
Education
Less than high school	Reference	Reference	Reference	Reference
High school diploma or GED	0.71 (0.52–0.97)^*	0.85 (0.68–1.07)	0.56 (0.36–0.88)^*	0.86 (0.70–1.05)
Some college	0.67 (0.49–0.92)^*	0.99 (0.79–1.23)	0.66 (0.43–1.01)	0.92 (0.74–1.13)
College degree or greater	0.63 (0.45–0.90)^*	0.99 (0.78–1.26)	0.39 (0.22–0.67)^**	0.90 (0.72–1.13)
Household covariates
Able to pay bills
No	Reference	Reference	Reference	Reference
Yes	0.65 (0.49–0.86)^**	0.82 (0.67–1.01)	0.78 (0.53–1.16)	0.85 (0.70–1.03)
Welfare receipt
No	Reference	Reference	Reference	Reference
Yes	1.69 (1.19–2.39)^**	1.10 (0.84–1.45)	2.05 (1.28–3.30)^**	1.04 (0.80–1.36)
Household income	1.00 (0.99–1.00)	1.00 (1.00–1.00)^*	0.99 (0.98–1.00)^*	0.99 (0.99–1.00)

p < 0.05; ^**p < 0.01.

CI, confidence interval; OR, odds ratio.

Discussion

Our findings suggest that breastfeeding for at least 9 months is inversely associated with child neglect and child sexual abuse. Controlling for known covariates of breastfeeding and child maltreatment, children breastfed for 9 months or longer had a reduced odds of child neglect and sexual abuse compared with children never breastfed. These findings are consistent with past research by Strathearn et al.,¹² in which breastfeeding duration was significantly associated with neglect, yet had no significant relationship with physical abuse. Our study expands upon Strathearn et al.'s¹² findings by replicating results using a nationally representative sample of American adolescents. In addition, this is the first known study that has explored the relationship between breastfeeding and risk for childhood sexual abuse.

The present findings and similar results by Strathearn et al.¹² demonstrate that a relationship does exist between breastfeeding and certain subtypes of maltreatment. Future research should investigate whether this relationship is causal. Although breastfeeding may merely be a proxy for an uncontrolled variable with childhood maltreatment, thorough understanding of the relationship between early parenting practices and future childhood maltreatment can help clinicians identify children at risk for neglect and sexual abuse. Research by Brown et al.⁹ found different patterns of risk factors to be associated with the subtypes of child maltreatment, and prevalence of maltreatment was higher as increasingly more risk factors were present. Child maltreatment assessments may be enhanced with the addition of breastfeeding duration as a risk factor.

Limitations

Several limitations of this study should be considered. First, although the sample started with >20,000 adolescents, around 80% of the sample was not included in analyses due to sample attrition or missing data. Although analyses were weighted to account for attrition, the results may still be biased by the reduced sample size. The weights may account for sample attrition with regard to demographic variables, but there may be additional differences between those included and not included in analyses that are not captured by the weighted variables. Differences between the groups may also be related to child's maltreatment experience, which, in turn, biases findings. In addition, the nonexperimental nature of the study prevents causal inferences to be made. This study is also limited by the information collected within the Add-Health. Although many covariates were considered in this study, additional covariates not collected by the Add-Health may also have a relationship with childhood maltreatment and their inclusion may have altered study findings. Particularly, we were unable to control for maternal depression and drug use. Furthermore, indicators of child maltreatment were self-reported by adolescents and may be subject to recall bias.

Conclusions

Breastfeeding for at least 9 months is associated with a significant decrease in child neglect and sexual abuse. This is consistent with previous research and provides more evidence that breastfeeding may serve as a protective factor for child maltreatment. Future research establishing a causal relationship between breastfeeding and childhood maltreatment is needed. A better understanding of the risk and protective factors for child maltreatment may enable medical practitioners to develop preventive solutions for child maltreatment.

Footnotes

Acknowledgments

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Disclosure Statement

No competing financial interests exist.

References

U.S. Department of Health and Human Services; Administration on Children and Families, Children's Bureau. Child maltreatment, 2014. Washington, DC: U.S. Department of Health and Human Services, 2016.

Mullen

, Martin

, Anderson

, et al. Childhood sexual abuse and mental health in adult life. Br J Psychiatry, 1993; 163:721–732.

Mullen

, Martin

, Anderson

, et al. The long-term impact of the physical, emotional, and sexual abuse of children: A community study. Child Abuse Negl, 1996; 20:7–21.

Springer

, Sheridan

, Kuo

, et al. Long-term physical and mental health consequences of childhood physical abuse: Results from a large population-based sample of men and women. Child Abuse Negl, 2007; 31:517–530.

Arias

. Report from the CDC. The legacy of child maltreatment: Long-term health consequences for women. J Womens Health, 2004; 13:468–473.

DePanfilis

. Child Neglect: A Guide for Prevention, Assessment, and Intervention. Washington, DC: U.S. Department of Health and Human Services, Administration for Children and Families, Administration on Children, Youth and Families, Children's Bureau, Office on Child Abuse and Neglect, 2006.

Runyan

, Wattam

, Ikeda

, et al. Child abuse and neglect by parents and other caregivers. In: World Report on Violence and Health. Krug

, Dahlberg

, Mercy

, Zhi

, Lozano

, eds. Geneva: World Health Organization, 2002, pp. 57–86. Available at www.who.int/violence_injury_prevention/violence/world_report/en/FullWRVH.pdf

Sidebotham

, Heron

, Team

. Child maltreatment in the “children of the nineties”: A cohort study of risk factors. Child Abuse Negl, 2006; 30:497–522.

Brown

, Cohen

, Johnson

, et al. A longitudinal analysis of risk factors for child maltreatment: Findings of a 17-year prospective study of officially recorded and self-reported child abuse and neglect. Child Abuse Negl, 1998; 22:1065–1078.

10.

U.S. Department of Health and Human Services. The Surgeon General's call to action to support breastfeeding. Washington, DC: U.S. Department of Health and Human Services, 2011.

11.

Huang

, Peters

, Vaughn

, et al. Breastfeeding and trajectories of children's cognitive development. Dev Sci, 2014; 17:452–461.

12.

Strathearn

, Mamun

, Najman

, et al. Does breastfeeding protect against substantiated child abuse and neglect? A 15-year cohort study. Pediatrics, 2009; 123:483–493.

13.

Pearson

, Lightman

, Evans

. The impact of breastfeeding on mothers' attentional sensitivity toward infant distress. Infant Behav Dev, 2011; 34:200–205.

14.

Kim

, Feldman

, Mayes

, et al. Breastfeeding, brain activation to own infant cry, and maternal sensitivity. J Child Psychol Psychiatry, 2011; 52:907–915.

15.

Harris

. The Add Health Study: Design and accomplishments. Chapel Hill, NC: Carolina Population Center, University of North Carolina at Chapel Hill, 2013. Available at www.cpc.unc.edu/projects/addhealth