The Association Between Breastfeeding and Length of Hospital Stay Among Infants Diagnosed with Neonatal Abstinence Syndrome: A Population-Based Study of In-Hospital Births

Abstract

Objective:

The incidence of neonatal abstinence syndrome (NAS), a drug withdraw syndrome mainly associated with intrauterine opioid exposure, has increased considerably in the United States since 2000. Nonpharmacological options, including breastfeeding, may be effective at improving outcomes in this population. The objective of this population-based study was to examine the association between breastfeeding and length of hospital stay among infants diagnosed with NAS.

Methods:

This was a retrospective cohort study of singleton in-hospital births to resident mothers in Pennsylvania. Hospital discharge data from births occurring between 2012 through 2014 were linked with corresponding birth certificate data. International Classification of Diseases, Ninth Revision, Clinical Modification discharge diagnosis codes were used to identify NAS neonates (N = 3,725). Breastfeeding at discharge was used to determine breastfeeding status. Infant and maternal characteristics were compared by breastfeeding status and the association between breastfeeding and infant length of hospitalization was assessed.

Results:

Less than one-half of infants diagnosed with NAS were breastfed at discharge. Significant differences in infant birth weight and gestational age, and maternal education, marital status, prenatal care, smoking, and insurance status were found by breastfeeding status. A significant inverse relationship existed between breastfeeding and hospital length of stay for infants diagnosed with NAS. Specifically, length of hospitalization was reduced by 9.4% in the breastfed group compared to the nonbreastfed group.

Conclusion:

Breastfeeding may be beneficial for infants diagnosed with NAS by shortening the length of hospital stay. Future prospective studies are warranted to further examine the benefits of breastfeeding and other nonpharmaceutical interventions in NAS populations.

Introduction

Opioid use has become increasingly common among pregnant women in the United States. Between 2000 and 2009, the number of delivering mothers diagnosed as dependent on or using opiates at the time of delivery increased almost five-fold, from 1.19 to 5.63 per 1,000 hospital births per year.¹ During this same time period, there was an 11.9% annual increase in rates of opioid use among pregnancy-related hospitalizations.² Among pregnant admissions to drug treatment centers, the proportion reporting any prescription opioid abuse increased from 2% to 28% from 1992 to 2012,³ while the reporting of prescription opioids as the primary substance of abuse similarly increased from 1% to 19%.³

A common sequela of intrauterine opioid exposure is neonatal abstinence syndrome (NAS), a condition of the neonate characterized by hyperactivity of the central and autonomic nervous systems.⁴ An increase in the incidence of NAS in the United States has paralleled the increase of opioid use during pregnancy. Specifically, from 2009 to 2012, the incidence of NAS increased from 3.4 to 5.8 per 1,000 hospital births nationwide.⁵ From 2004 to 2013, the rate of neonatal intensive care unit (NICU) admissions for NAS increased from 7 to 27 cases per 1,000 admissions.⁶

The increase in NAS incidence has important economic and maternal–infant implications. Hospital charges for NAS in 2012 were ∼1.4 billion dollars and more than 81% of NAS newborns were billed to state Medicaid programs.⁵ Overall treatment costs for newborns with NAS are higher than infants without NAS,⁷ primarily due to longer hospital stays.^1,5,8 As extended hospitalizations also increase maternal-infant physical separation, which can limit bonding opportunities,⁹ strategies that help to reduce prolonged stays are warranted in the NAS population.

Optimal treatment for NAS has yet to be established, however, pharmaceuticals, including morphine and methadone, are commonly used to manage moderate to severe NAS. Nonpharmacological options, including breastfeeding, may also be effective at improving outcomes in NAS populations. Little methadone or buprenorphine is present in breast milk^10,11 and the American College of Obstetrics and Gynecologists¹² and the American Academy of Pediatrics¹³ support breastfeeding among opioid-dependent women if the women are enrolled in substance abuse treatment and there are no contraindications to breastfeeding. Benefits to breastfeeding for substance using mothers include protecting against relapse and stress.¹⁴ While breastfeeding is well-established as beneficial for all infants,¹⁵ evidence is growing that breastfeeding and breast milk have specific benefits for the NAS population by preventing or limiting treatment or reducing the length of hospital stay.^16–20 However, much of this literature assessing the relationship between infant feeding methods and NAS-related outcomes originates from small and/or single-site studies, or studies restricted to women in methadone treatment programs. Thus, the objective of this population-based study was to examine the association between breastfeeding and length of hospital stay in a large sample of infants diagnosed with NAS.

Materials and Methods

Study design

This was a retrospective cohort study of singleton births to resident mothers in Pennsylvania between January 1, 2012 and December 31, 2014, which utilized state-based registry data. Hospital discharge data from in-hospital births were linked with corresponding birth certificate data using multiple fields with similar characteristics. A total of 20 matching iterations were performed using variables such as child date of birth, gender, race, ethnicity, zip code, facility/hospital number, gestation, and birth weight. Only one-to-one matches were deemed successful. The final analytic dataset was aggregate and void of personal identifiers. The Institutional Review Board at Nemours approved this study.

Neonatal abstinence syndrome

Analyses were restricted to infants with an International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) discharge diagnosis code of 779.5 (drug withdrawal syndrome in a newborn) in any of the 18 discharge diagnosis fields. Infants born less than 34 weeks of gestation were excluded to control for possible iatrogenic NAS.¹

Infant and maternal characteristics

Infant characteristics assessed included gestational age in weeks (continuous; term [≥37 weeks], late preterm [34–36 weeks]), birth weight in grams (continuous; low birth weight [<2,500 g], normal birth weight [≥2,500 g]), and admission to an NICU during a portion of the hospitalization (yes, no). From the birth certificate, the question “Is the infant being breastfed at discharge?” (Yes/No) was used to determine breastfeeding status. Infant length of hospital stay was defined as number of hospitalization days and originated from the discharge data file. Maternal characteristics assessed included the following: age in years (continuous), race (black, white, other/unknown), ethnicity (non-Hispanic, Hispanic), marital status (married, not married), highest level of education (high school or less, greater than high school), smoking during pregnancy (yes, no), history of at least one prenatal care visit (yes, no), Medicaid enrollment during pregnancy (yes, no), and use of the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) during pregnancy (yes, no).

Statistical analyses

Infant and maternal characteristics were described among the entire study population. Chi-square tests were used to compare characteristics by breastfeeding status. Log transformation was performed to normalize the distribution of length of hospital stay. Simple linear regression models were used to assess the unadjusted effect of breastfeeding and other infant and maternal characteristics on infant length of hospital stay. Variables significant in these models (p ≤ 0.10) were adjusted for in a multiple linear regression model, assessing the association between breastfeeding and length of stay. This model was further adjusted for infant comorbidities that have been reported to be common among infants with NAS and have been associated with hospital length of stay, including respiratory complications, feeding difficulties, and seizures.^1,8 ICD-9-CM diagnostic codes were used to identify these conditions. Results from the regression models were reported as parameter estimates and standard errors and p-values. The exponent of the parameter estimate for the breastfeeding variable was interpreted as the change in length of stay among breastfed infants compared to nonbreastfed infants. Finally, nearly 8% of our sample was missing information regarding prenatal care, thus, linear regression models including and excluding history of prenatal care were constructed. All statistical analyses were performed using SAS version 9.4 (SAS Institute, Cary, NC) and data were regarded as statistically significant at p < 0.05.

Results

Descriptive characteristics of the entire study population are presented in Table 1. A total of 3,725 NAS cases were identified during the study years. The majority of these infants were born term (87.5%) with a normal birth weight (83.9%) to White (90.0%), non-Hispanic (96.6%) unmarried (79.8%) mothers who received some prenatal care (95.7%) and Medicaid (68.4%) and smoked during pregnancy (75.9%). Nearly forty percent of the infants were admitted to the NICU during the hospitalization. Five percent (n = 181) of the study population was missing breastfeeding data. Less than half (44.5%) of all NAS infants were being breastfed at discharge.

Table 1.

Population Characteristics of Infants Diagnosed with Neonatal Abstinence Syndrome by Breastfeeding Status

	All NAS infants (N = 3,725), N (%)	Breastfed NAS infants^a (n = 1,576), n (%)	Nonbreastfed NAS infants^a (n = 1,968), n (%)	p^b
Infant characteristics
Median hospital length of stay^c,d	11 (17)	10 (14)	12 (17)	—
Mean hospital length of stay^c	15.5	14.6	15.9	—
Breastfed at discharge
Yes	44.5	—	—	—
No	55.5	—	—
NICU admission
Yes	40.8	39.3	41.6	0.17
No	59.2	60.7	58.4
Birth weight
Mean (SD)^e	3056.8 (738.6)	3118.8 (766.0)	3009.6 (730.3)
Low (<2,500 g)	16.1	13.1	18.4	<0.0001
Normal (≥2,500 g)	83.9	86.9	81.6
Gestational age
Mean (SD)^f	38.5 (1.6)	38.6 (1.5)	38.4 (1.7)
Late preterm (34–36 weeks)	12.5	10.4	13.8	0.002
Term (≥37 weeks)	87.5	89.6	86.2
Maternal characteristics
Age (years)
<20	2.2	2.4	2.2	0.22
20–24	25.8	27.4	24.9
25–34	61.9	60.9	62.1
35+	10.0	9.3	10.8
Race
Black	5.7	5.5	5.8	0.44
White	90.0	89.9	90.5
Other/Unknown	4.3	4.6	3.7
Ethnicity
Hispanic	3.4	3.5	3.3	0.73
Non-Hispanic	96.6	96.5	96.7
Education level
≤High school	61.8	55.1	67.4	<0.0001
>High school	38.2	44.9	32.6
Marital status
Married	20.2	25.2	16.9	<0.0001
Not married	79.8	74.8	83.1
Source of payment^g
Medicaid	68.4	66.6	72.6	0.0001
Other	31.6	33.4	27.4
Prenatal care
Yes	95.7	98.8	94.0	<0.0001
No	4.3	1.2	6.0
Smoking during pregnancy
Yes	75.9	70.1	81.0	<0.0001
No	24.1	29.9	19.0
WIC recipient
Yes	56.8	57.4	56.7	0.67
No	43.2	42.6	43.3

Missing values ranged from to <1% for marital status to 7.7% for history of prenatal care.

Breastfeeding status at discharge.

Chi-square p-value of breastfed versus nonbreastfed NAS infants.

Infant hospital length of stay in days.

Median (interquartile range) number of days.

Mean (standard deviation) birth weight in grams.

Mean (standard deviation) gestational age in weeks.

Principal source of payment of delivery.

NAS, neonatal abstinence syndrome; NICU, neonatal intensive care unit; SD, standard deviation; WIC, the Special Supplemental Nutrition Program for Women, Infants, and Children.

Significant differences in infant and maternal characteristics were found between infants who were breastfed and those who were not breastfed. Specifically, infants who were breastfed were significantly more likely to have a normal birth weight (86.9% versus 81.6%, p < 0.0001) and be born term (89.6% versus 86.2%, p < 0.002) than infants who were not breastfed. They were also significantly more likely to have mothers who had greater than a high school education (44.9% versus 32.6%, p < 0.0001), were married (25.2% versus 16.9%, p < 0.0001), and had a history of prenatal care (98.8% versus 94.0%, p < 0.0001). Breastfed infants were significantly less likely to have mothers who smoked (70.1% versus 81.0%, p < 0.0001) or received Medicaid (66.6% versus 72.6%, p = 0.0001) compared to nonbreastfed infants. No differences in NICU admission or maternal age, race, ethnicity, or WIC use were found.

Median length of stay for all NAS infants was 11 days (interquartile range 5–22). Median length of stay was 10 days (interquartile range 5–19) for breastfed infants and 12 days (interquartile range 5–22) for nonbreastfed infants. Results from the linear regression analyses are presented in Table 2. There was a significant inverse relationship between breastfeeding and length of stay (B = −0.085, p = 0.008). This relationship remained significant after adjusting for birth year, hospital, NICU admission, mode of delivery, birth weight, infant comorbidities, and maternal age, race, marital status, Medicaid, and prenatal smoking (B = −0.060, p = 0.05). After controlling for all covariates, length of hospital stay was reduced by 9.4% in the breastfed group compared to the nonbreastfed group (results not shown).

Table 2.

Associations with Hospital Length of Stay for Infants Diagnosed with Neonatal Abstinence Syndrome for Listed Factors

	Unadjusted model ^a		Adjusted model 1 ^b		Adjusted model 2 ^c
	B (SE)	p	B (SE)	p	B (SE)	p
Breastfed at discharge	−0.085 (0.032)	0.008	−0.062 (0.029)	0.03	−0.060 (0.031)	0.05
Birth year			−0.039 (0.018)	0.07	−0.038 (0.019)	0.04
Hospital			−0.0009 (0.00001)	<0.0001	−0.00009 (0.00001)	<0.0001
NICU admission			0.71 (0.030)	<0.0001	0.67 (0.03)	<0.0001
Method of delivery			−0.02 (0.031)	0.41	−0.015 (0.033)	0.64
Birth weight					0.00003 (0.00003)	0.33
Maternal age					0.0002 (0.003)	0.94
Maternal race					0.057 (0.029)	0.05
Marital status					−0.10 (0.040)	0.01
Maternal Medicaid recipient					−0.09 (0.0234)	0.005
Maternal prenatal smoking					0.029 (0.037)	0.43
Respiratory complications					0.18 (0.04)	<0.001
Seizures					−0.57 (0.25)	0.02
Feeding difficulties					0.09 (0.23)	0.69

Log transformed length of stay was used in the regression models.

Unadjusted model (n = 3,544).

Adjusted model 1 (n = 3,538) adjusted for birth year, hospital, NICU admission, and method of delivery.

Adjusted model 2 (n = 3,326) adjusted for variables in model 1 plus birth weight, maternal age, race, marital status, Medicaid and prenatal smoking, and infant comorbidities (i.e., respiratory complications, seizures, and feeding difficulties).

SE, standard error.

Discussion

In the United States., the increased incidence of NAS has paralleled the increased use of prescription opioids among pregnant women. Nonpharmacological options, such as breastfeeding, may be effective at improving outcomes in the population of infants with NAS. In this study, we investigated the variation in length of hospitalization following birth by infant feeding method at discharge and found significant differences by breastfeeding status. Our results suggest that NAS infants who are breastfed have a significantly shorter hospitalization than nonbreastfed NAS infants, even after controlling for differences in maternal and infant characteristics.

Our findings are consistent with the findings from previous, smaller studies of women undergoing drug treatment, which reported an inverse relationship between breastfeeding and length of hospital stay and other adverse outcomes among NAS infants. In a U.S. study of 86 infants exposed to maternal methadone or buprenorphine in utero for at least 30 days, breastfed infants experienced a significantly shorter length of stay compared to nonbreastfed infants (15.8 versus 27.4 days, p < 0.001).¹⁸ Among infants born to opioid-dependent women in Maine, Pritham et al. found a shorter length of stay among breastfed infants than formula fed.¹⁷ Similar findings have been reported outside the United States. In a cohort of 124 children born to women in opioid management therapy in Norway, duration of NAS treatment was shorter for breastfed than nonbreastfed infants (28.6 versus 46.7 days, p < 0.05).¹⁹ Abdel-Latif et al. reported that among 190 infants born to drug-dependent mothers in New South Wales, breast milk significantly reduced the severity of NAS and reduced the length of hospital stay.¹⁶ The length of stay among infants who were breastfed was 14.7 days compared to 19.1 days for nonbreastfed infants (p = 0.049).¹⁶ To our knowledge, ours is the first population-based study using state-based administrative data to examine the impact of breastfeeding on a large sample of infants with NAS. While the reduction we found was not as great as that previously reported in other studies, the nearly 10% length of stay reduction for infants who were breastfed does represent an opportunity for significant cost savings. Specifically, a shortened hospitalization may equate to potential savings of more than $3,000 per inpatient treatment day.¹

The overall breastfeeding rate in our study population was well below the national rate. Only 44% of infants in our cohort were breastfed compared to 79% of U.S. newborns.²¹ Lower rates of breastfeeding among NAS infants are not unexpected and could be due to higher NICU admission rates and/or the physical symptoms more commonly found in this population,⁵ which could make breastfeeding difficult. National estimates of breastfeeding among women with NAS infants are unknown, but rates among opioid-dependent women eligible to breastfeed have been reported to be low.²² Based on our current findings, designing and implementing targeted breastfeeding promotion activities to increase the overall breastfeeding rate among pregnant women at risk for having an infant with NAS could be worthwhile.

We also found some significant differences in the characteristics of breastfed and nonbreastfed infants. Again, this finding is not surprising given that a variety of sociodemographic factors, including educational attainment, Medicaid utilization, and income,^23,24 have been shown to be associated with infant feeding method in the general postpartum population. It is likely that we did not see differences in other factors associated with breastfeeding, such as race and ethnicity,²³ given the homogeneity of the overall study population. Psychosocial, clinical, and hospital environmental factors can also influence a woman's choice of infant feeding method. However, information pertaining to the role that these measures play in breastfeeding behaviors among mothers with NAS infants is limited. Our findings do offer some insight into which women are choosing to breastfeed, but future studies are needed to describe potentially modifiable nonsociodemographic breastfeeding determinants among this population.

It is the act of breastfeeding, rather than the actual breast milk, which likely impacts NAS infants. Other nonpharmacological interventions used to address NAS-related symptoms both compliment and support the act of breastfeeding itself, such as skin to skin contact²⁵ and kangaroo care.²⁶ Research shows that rooming-in, the practice of allowing close uninterrupted postpartum contact between the mother–infant dyad, can positively affect NAS infants.²⁷ Thus, it is possible that the association we found between breastfeeding and improved outcome was due to the combination of interventions that positively impacted the infant. However, data on other types of nonpharmacological interventions the infants may have been exposed to in the hospital setting were not available.

This study adds to the limited, although growing, literature related to the benefits of breastfeeding for infants with NAS and has several strengths. First, a major strength of this study is the use of a large, population-level data source that has timely and complete information from nearly all in-hospital births in the state for a 3-year period. To our knowledge, no previous study has utilized this administrative database to examine NAS. Second, our use of a standardized data collection tool and ICD-9 codes to identify our study population will allow for replication and validation of our findings. Third, this information is urgently needed to reduce the burden on newborn infants affected by a rapidly growing epidemic.

Despite the strengths of this study, there are potential limitations to consider when interpreting our results. Some inaccuracies in the classification of NAS may be expected due to the use of admission diagnoses and the variability in assessment of NAS. However, findings from a recent study suggested that potential misclassification of NAS was likely to be small.²⁸ Patrick et al. attempted to establish the accuracy of administrative coding for NAS by conducting a chart review of 228 randomly selected cases and noncases of NAS.²⁸ Using a standard definition of NAS as reference, ICD-9-CM–based identification yielded an 88.1% sensitivity and 97.0% specificity.²⁸ Therefore, we are confident that ICD-9 codes are an acceptable method for identifying NAS cases.

Another limitation is the lack of information regarding both in-utero exposure and postnatal treatment for NAS. Substance(s), timing, and amount of last maternal use can impact the clinical presentation of NAS. This study also did not identify drugs used to treat maternal opioid dependence, although treatment may influence length of stay. While methadone is considered standard treatment for opioid dependence during pregnancy, a randomized double-blind trial found that infants who had prenatal exposure to buprenorphine required a significantly shorter hospital stay than did infants with prenatal exposure to methadone.²⁹ Similarly, we were unable to differentiate cases resultant from maternal opioid abuse and the legitimate use of an opioid prescription. However, the objective of our study was to describe the effect of breastfeeding on all NAS infants. Researchers may want to consider these issues when planning future studies.

Lastly, while the use of a standardized data collection tool in our study will allow for replication and validation of our findings, there was a lack of specific data about breastfeeding practices. Only one question was used to assess the breastfeeding status of the infant at discharge and it did not differentiate between exclusive breastfeeding and mixed feeding of breast milk and formula. Neither did the data include information on duration of breastfeeding beyond discharge. Nevertheless, initiation of breastfeeding is a strong predictor of continuation of breastfeeding beyond discharge.³⁰ In addition, a recent National Vital Statistics Report evaluating the quality of selected medical and health data from the 2003 revision of the birth certificate by comparing birth certificate data with data abstracted from hospital medical records found high sensitivity and a low false discovery rate for breastfeeding.³¹

Conclusion

Findings from this population-based study support the notion that breastfeeding is beneficial for infants diagnosed with NAS. Prospective research into the development and evaluation of breastfeeding interventions targeting NAS infants and their mothers that address other related outcomes such as disease severity is warranted. Moreover, while breastfeeding may be beneficial to this population, individual, clinical, and hospital environmental determinants can influence breastfeeding intention and initiation; so identifying these would be key in planning for such interventions.

Footnotes

Acknowledgments

This research was funded by Nemours. The authors would like to thank Jobayer Hossain (Nemours) and Okan U. Elci. Much of our data were from the Pennsylvania Healthcare Cost Containment Council (PHC4). PHC4 is an independent state agency responsible for addressing the problem of escalating health costs, ensuring the quality of healthcare, and increasing access to healthcare for all citizens regardless of ability to pay. PHC4 has provided data to this entity in an effort to further PHC4's mission of educating the public and containing healthcare costs in Pennsylvania. PHC4, its agents, and staff have made no representation, guarantee, or warranty, express or implied, that the data—financial, patient, payer, and physician-specific information—provided to this entity, are error-free, or that the use of the data will avoid differences of opinion or interpretation. This analysis was not prepared by PHC4. This analysis was done by the authors of this article. PHC4, its agents, and staff bear no responsibility or liability for the results of the analysis. Data were also supplied by the Bureau of Health Statistics & Registries, Pennsylvania Department of Health, Harrisburg, Pennsylvania. The Pennsylvania Department of Health specifically disclaims responsibility for any analyses, interpretations, or conclusions.

Disclosure Statement

No competing financial interests exist.

References

Patrick

, Schumacher

, Benneyworth

, et al. Neonatal abstinence syndrome and associated health care expenditures: United States, 2000–2009. JAMA, 2012; 307:1934–1940.

Salihu

, Mogos

, Salinas-Miranda

, et al. National trends in maternal use of opioid drugs among pregnancy-related hospitalizations in the United States, 1998 to 2009. Am J Perinatol, 2015; 32:289–298.

Martin

, Longinaker

, Mishka

. Recent trends in treatment admissions for prescription opioid abuse during pregnancy. J Subst Abuse Treat, 2015; 48:37–42.

Center for Substance Abuse Treatment. Medication assisted treatment for opioid addiction during pregnancy. SAHMSA/CSAT treatment improvement protocols. Rockville, MD: Substance Abuse and Mental Health Services Administration, 2008. Available at www.ncbi.nlm.nih.gov/books/NBK26113 (accessed May 12, 2016).

Patrick

, Davis

, Lehmann

, et al. Increasing incidence and geographic distribution of neonatal abstinence syndrome: United States 2009 to 2012. J Perinatol, 2015; 35:650–655.

Tolia

, Patrick

, Bennett

, et al. Increasing incidence of the neonatal abstinence syndrome in U.S. neonatal ICUs. N Engl J Med, 2015; 372:2118–2126.

Roussos-Ross

, Reisfield

, Elliot

, et al. Opioid use in pregnant women and the increase in neonatal abstinence syndrome: What is the cost?. J Addict Med, 2015; 9:222–225.

Stabler

, Long

, Chertok

, et al. Neonatal abstinence syndrome in West Virginia substate regions, 2007–2013. J Rural Health, 2016 [Epub ahead of print]; DOI: 10.1111/jrh.12174.

Norr

, Roberts

, Freese

. Early postpartum rooming-in and maternal attachment behaviors in a group of medically indigent primiparas. J Nurse Midwifery, 1989; 34:85–91.

10.

Jansson

, Choo

, Harrow

, et al. Concentrations of methadone in breast milk and plasma in the immediate perinatal period. J Hum Lact, 2007; 23:184–190.

11.

Lindemalm

, Nydert

, Svensson

, et al. Transfer of buprenorphine into breast milk and calculation of infant drug dose. J Hum Lact, 2009; 25:199–205.

12.

American College of Obstetricians and Gynecologists (ACOG) Committee on Health Care for Underserved Women, American Society of Addiction Medicine. ACOG Committee Opinion No. 524: Opioid abuse, dependence, and addiction in pregnancy. Obstet Gynecol, 2012; 119:1070–1076.

13.

American Academy of Pediatrics. Policy statement. Breastfeeding and the use of human milk. Pediatrics, 2012; 129:e827–e841.

14.

Tops

, Koole

, H

, et al. Why social attachment and oxytocin protect against addiction and stress: Insights from the dynamics between ventral and dorsal corticostriatal systems. Pharmacol Biochem Behav, 2014; 119:39–48.

15.

, Chung

, Raman

, et al. Breastfeeding and Maternal and Infant Health Outcomes in Developed Countries. Evidence Report/TechnologyAssessment No. 153 (Prepared by Tufts-New England Medical Center Evidence-based Practice Center, under Contract No. 290-02-0022). AHRQ Publication No. 07-E007. Rockville, MD: Agency for Healthcare Research and Quality, April 2007.

16.

Abdel-Latif

, Pinner

, Clews

, et al. Effects of breast milk on the severity and outcome of neonatal abstinence syndrome among infants of drug-dependent mothers. Pediatrics, 2006; 117:e1163–e1169.

17.

Pritham

, Paul

, Hayes

. Opioid dependency in pregnancy and length of stay for neonatal abstinence syndrome. J Obstet Gynecol Neonatal Nurs, 2012; 41:180–190.

18.

Wachman

, Hayes

, Brown

, et al. Association of OPRM1 and COMT single-nucleotide polymorphisms with hospital length of stay and treatment of neonatal abstinence syndrome. JAMA, 2013; 309:1821–1827.

19.

Welle-Strand

, Skurtveit

, Jansson

, et al. Breastfeeding reduces the need for withdrawal treatment in opioid-exposed infants. Acta Paediatr, 2013; 102:1060–1066.

20.

Isemann

, Meinzen-Derr

, Akinbi

. Maternal and neonatal factors impacting response to methadone therapy in infants treated for neonatal abstinence syndrome. J Perinatol, 2011; 31:25–29.

21.

Centers of Disease Control and Prevention. Breastfeeding Report Card, 2014. Available at www.cdc.gov/breastfeeding/data/reportcard.htm (accessed May 12, 2016).

22.

Wachman

, Byun

, Philipp

. Breastfeeding rates among mothers of infants with neonatal abstinence syndrome. Breastfed Med, 2010; 5:159–164.

23.

Phares

, Morrow

, Lansky

, et al. Surveillance for disparities in maternal health-related behaviors—selected states, Pregnancy Risk Assessment Monitoring System (PRAMS), 2000–2001. MMWR Surveill Summ, 2004; 53:1–13.

24.

Beck

, Morrow

, Lipscomb

, et al. Prevalence of selected maternal behaviors and experiences, Pregnancy Risk Assessment Monitoring System (PRAMS), 1999. MMWR Surveill Summ, 2002; 51:1–27.

25.

Hiles

. An Evidence-Based Intervention for Promoting Sleep in Infants Experiencing Neonatal Abstinence Syndrome (NAS) Due to Maternal Methadone Use. Philadelphia, PA: National Association of Clinical Nurse Specialists, 2011.

26.

Conde-Agudelo

, Belizan

, Diaz-Rossello

. Kangaroo mother care to reduce morbidity and mortality in low birthweight infants. Cochrane Database Syst Rev, 2011:CD002771.

27.

Newman

, Davies

, Dow

, et al. Rooming-in care for infants of opioid-dependent mothers: Implementation and evaluation at a tertiary care hospital. Can Fam Physician, 2015; 61:e555–e561.

28.

Patrick

, Dudley

, Martin

, et al. Prescription opioid epidemic and infant outcomes. Pediatrics, 2015; 135:842–850.

29.

Jones

, Kaltenbach

, Heil

, et al. Neonatal abstinence syndrome after methadone or buprenorphine exposure. N Engl J Med, 2010; 363:2320–2331.

30.

Donath

, Amir

. Relationship between prenatal infant feeding intention and initiation and duration of breastfeeding: A cohort study. Acta Paediatr, 2003; 92:352–356.

31.

Martin

, Wilson

, Osterman

, et al. Assessing the quality of medical and health data from the 2003 birth certificate revision: Results from two states. Natl Vital Stat Rep, 2013; 62:1–19.