OBJECTIVES. Maternal depression is an established risk for adverse child development. Two thirds of clinically significant depressive symptoms occur in mothers reporting an infant sleep problem. We aimed to determine the long-term effects of a behavioral intervention for infant sleep problems on maternal depression and parenting style, as well as on child mental health and sleep, when the children reached 2 years of age.
METHODS. We conducted a cluster-randomized trial in well-child centers across 6 government areas of Melbourne, Australia. Participants included 328 mothers reporting an infant sleep problem at 7 months, drawn from a population sample (N = 739) recruited at 4 months. We compared the usual well-child care (n = 154) versus a brief behavior-modification program designed to improve infant sleep (n = 174) delivered by well-child nurses at ages 8 to 10 months and measured maternal depression symptoms (Edinburgh Postnatal Depression Scale); parenting practices (Parent Behavior Checklist); child mental health (Child Behavior Checklist); and maternal report of a sleep problem (yes or no).
RESULTS. At 2 years, mothers in the intervention group were less likely than control mothers to report clinical depression symptoms: 15.4% vs 26.4% (Edinburgh Postnatal Depression Scale community cut point) and 4.2% vs 13.2% (Edinburgh Postnatal Depression Scale clinical cut point). Neither parenting style nor child mental health differed markedly between the intervention and control groups. A total of 27.3% of children in the intervention group versus 32.6% of control children had a sleep problem.
CONCLUSIONS. The sleep intervention in infancy resulted in sustained positive effects on maternal depression symptoms and found no evidence of longer-term adverse effects on either mothers' parenting practices or children's mental health. This intervention demonstrated the capacity of a functioning primary care system to deliver effective, universally offered secondary prevention.
Infant sleep problems have major implications for health care systems because they contribute substantially to long-term maternal depression, parenting stress, and child behavior problems.1 The prevalence of sleep problems drops rapidly after the first year of life from ∼45% to ∼15% by 2 years of age, after which the prevalence remains fairly constant into the early school years.1,2 The sheer prevalence of Australian infant sleep problems imposes a substantial population burden in the first year of life, similar to other developed countries.2,3
Maternal depression and anxiety are consistently associated with infant sleep problems, even after controlling for known depression risk factors.4,5 Postnatal depression disproportionately affects mothers whose infants suffer sleep problems, with two thirds of all mothers with Edinburgh Postnatal Depression Scale (EPDS) scores in the clinically significant range also reporting an infant sleep problem at 6 to 12 months of age.6 Postnatal depression short- and long-term sequelae include relationship breakdown, insecure mother-child attachment, child cognitive, behavior and emotional problems, and (in rarer cases) child abuse.7,8 Many mothers with postnatal depression are reluctant to accept this diagnosis, medication, or therapy.9 However, effectively managing infant sleep problems significantly reduces maternal depression symptoms, and this approach is highly acceptable to mothers6,10 and cost-effective.6 Thus, according to the Mrazek and Haggerty11 model for mental health promotion, management of infant sleep problems could be viewed as a “selective” preventive intervention for some mothers with postnatal depression, whereby the intervention (ie, infant sleep intervention) is “targeted to a subgroup of the population whose risk of developing mental disorders is significantly higher than average” (ie, mothers who report an infant sleep problem).
Effective management of infant sleep problems involves behavioral strategies, and concerns have been raised about their potential for adverse impacts in areas of infant brain development, insecure mother-child relationship, and later child mental health problems (eg, anxiety).12 To date no evidence exists to support such concerns, whereas substantial evidence has accumulated for the short-term benefit of behavioral sleep strategies.12 Yet, these concerns have led to community reluctance to manage infant sleep problems using behavioral strategies.
In 2001, findings were published from our small efficacy trial of a behavioral intervention designed to reduce sleep problems in infants aged 6 to 12 months.4 At age 3 to 4 years, we found that graduated extinction in infancy did not adversely affect later child behavior, maternal depression, or family functioning.13 This study did not examine effects on the mother-child relationship. In 2004–2005, we conducted a large effectiveness trial in which well-child nurses were trained to deliver the same brief behavioral sleep intervention to infants aged 8 months.6 When infants turned both 10 and 12 months of age, mothers in the intervention group reported fewer infant sleep problems, suffered fewer depression symptoms, and were less likely to have sought other professional help for their infant's problem than control mothers.6
We now report on the long-term impacts of the intervention when the children reached the age of 2 years in this large effectiveness trial. We aimed to determine whether the positive impacts on maternal depression and child sleep were sustained to this age. On the basis of our own and others' studies,1,3 we expected that most sleep problems would have resolved by age 2 years in both the intervention and control groups. We also aimed to determine whether there was any evidence of negative impacts of the infant sleep intervention on mothers' parenting practices or relationship quality with her child and 2-year child mental health.
The trial was conducted in greater Melbourne, Victoria, Australia (population: 3.4 million). Melbourne has 31 local government areas, which can be ranked according to a census-based geographical index of relative disadvantage.14 Six local government areas were selected to provide a broad sociodemographic spread (2 in the lowest, 2 in the middle, and 2 in the highest tertile for disadvantage).
Ninety-one percent of infants born in the state of Victoria attend the free key health visit offered at 4 months of age by their assigned Maternal and Child Health (MCH) nurse.15 All of the MCH nurses in the 6 local government areas participated. MCH nurses consecutively invited mothers of 4-month-old infants attending in October or November 2003 to take part in the Infant Sleep Study. Infants born before 32 weeks' gestation and mothers with insufficient English to complete questionnaires were excluded.
When infants turned 7 months, MCH centers (clusters) were allocated to intervention or control arms using computer-generated random numbers by an independent statistician unaware of the MCH identifiers, thus ensuring allocation concealment. Clusters were stratified according to local government area and ranked according to size of case population; within each stratum, the largest cluster was randomly allocated and subsequent clusters alternately allocated to avoid a marked imbalance in cluster sizes between trial arms. Mothers who reported a sleep problem in a questionnaire administered at 7 months composed the study sample.
Intervention nurses attended 2 structured 2.5-hour training sessions conducted by Dr Hiscock (pediatrician) and Dr Bayer (child psychologist), supported by a written manual. Training incorporated didactic teaching, written information, role-play sessions, and trouble-shooting common problems (such as partner conflict over sleep management). Nurses were trained regarding how to instruct families in the use of the 2 widely practiced behavioral interventions, namely graduated extinction (“controlled crying”) and adult fading (“camping out”).16–18 Two different structured behavioral interventions were taught, because Dr Hiscock's clinical experience suggested that different families prefer different sleep management approaches. In graduated extinction, parents respond to their infant's cry at increasing time intervals (ie, 2, 4, 6, 8, to 10 minutes), allowing an infant to learn to fall asleep by him or herself. Adult fading is an even more graduated approach to managing infant sleep problems, in which a parent sits with their infant until they fall asleep and slowly removes their parental presence over 2 to 3 weeks.
At the first consultation (the infant's well-child, 8 month visit), MCH nurses used a structured consultation sheet to elicit the nature of the sleep problem, identify solutions, and develop a written individualized sleep management plan with the mother on the basis of graduated extinction or adult fading. Nurses also recorded on the consultation sheet the number and duration of consultations. Two handouts discussed normal sleep patterns in infants aged 6 to 12 months and sleep associations (ie, rocking an infant off to sleep) and their causal role in sleep problems. Handouts on managing problem overnight feeding (ie, reducing volume or time spent feeding over a week) and pacifiers (ie, removal or teaching infant to replace his or her own pacifier) were also tailored into the individualized sleep plan as needed.
Mothers completed a written questionnaire when their child turned 10 months, 12 months, and 2 years old (ie, 17 months after random assignment). The primary outcome of maternal depression was measured by the EPDS (cut point for depression in the community is a score of >9 and in clinical samples is >12).19,20 Mothers' parenting practices were measured with the Parent Behavior Checklist, which yields standardized t scores for both harsh discipline and nurturing (with mean of 50 and SD of 10).21 Child mental health was assessed with the Child Behavior Checklist for children aged 18 months to 5 years, which yields standardized t scores for externalizing and internalizing behavior problems.22 Mothers also reported the presence of current child sleep problems (“Over the last 2 weeks, has your child's sleep generally been a problem for you?” yes or no). In 2 large community studies, maternal report of an infant sleep problem is a strong predictor of frequent and prolonged night wakings and difficulty settling to sleep.4,23 Potential confounders measured were mothers' rating of temperament on the single-item 5-point Global Infant Temperament Scale24 and overall confidence in the competence of their nurse's health advice on a 4-point ordinal study-designed scale.
Mothers in the intervention group reported on the quality of their relationship with their child as a result of the earlier sleep intervention. Mothers were asked, “Think about the strategies your MCH nurse recommended (eg, leaving your child to settle to sleep by themselves even if it involved short periods of your child crying). Do you think these strategies have changed your relationship with your child at all?” Mothers responded on a study-designed 10-point visual analog scale ranging from 0, “relationship worse,” to 10, “relationship better.”
The project was approved by the ethics in human research committee of Melbourne's Royal Children's Hospital (EHRC23067C). The trial was conducted in accordance with the Consolidated Standards of Reporting Trials statement for cluster-randomized trials,25 and all of the mothers gave written, informed consent.
Sleep was the primary short-term outcome of the trial, and, therefore, the sample size was based on detecting a difference of 20% between the proportions of mothers reporting infant sleep problems at each of the 10- and 12-month follow-ups (reported by 70% of the control group and 50% of the intervention group in our previous efficacy trial).10 An individually randomized trial would require 103 infants in each arm to have 80% power at the 5% level of significance. This sample size was inflated by a design effect of 1.2 to 124 infants per trial arm to allow for correlation between responses within the same cluster (ie, MCH center),26 with an expected average cluster size of 11 (ie, number of eligible mothers attending the center) and intracluster correlation coefficient of 0.02.
The trial groups were analyzed as randomized, applying the intention-to-treat principle.25 Because of the nature of the intervention, it was not possible to blind either the nurses or the mothers. Allocation concealment at random assignment and blinding of the data entry process took place. Outcomes were compared between the intervention and control groups, adjusting for potential confounders selected a priori on the basis of our previous research6,10 (see Table 3 footnote). To allow for clustering, quantitative outcomes were analyzed using random-effects linear regression fitted using full information maximum likelihood estimation,27 and dichotomous outcomes were analyzed using marginal logistic regression models fitted using generalized estimating equations with information sandwich estimates of SEs.28 An exchangeable correlation matrix was specified for the generalized estimating equation analyses. Ordinary logistic regression was used to analyze the sleep problem outcome, because this had a negative intracluster correlation coefficient. Confidence intervals from analyses of quantitative outcomes were validated using the bootstrap method.29 Analyses were conducted by using Stata 9.2 (Stata Corp, College Station, TX).
At 4 months, 739 infants were recruited and mothers of 695 infants (94%) completed the 7-month questionnaire. Of these, 328 mothers reported an infant sleep problem and participated in the intervention trial. In accordance with how their MCH center was randomized, 174 were allocated to the intervention group and 154 mothers to the control group (see Fig 1). In this “real-world” effectiveness trial, 100 of 174 intervention families took up the offer of help from their nurse. Of these 100 families, 53 families chose graduated extinction, 7 families chose adult fading, and for 40 families, nurses indicated simple strategies, such as the development of a positive bedtime routine or management of pacifiers overnight (n = 20), or nurses did not note their choice of behavioral technique (n = 20).
Characteristics of the participants at baseline are presented in Table 1, showing that the 2 arms were balanced. Mothers in the intervention group attended their nurse specifically for sleep advice for an average of 1.52 visits (recorded by nurses); their first visit lasted on average 25 minutes, and follow-up visits lasted on average 19 minutes. In the control arm, 34 of 154 mothers also reported receiving some help from their nurse for their infant's sleep problem (none of these nurses had received the structured training but still provided care to families, as per usual care), and these mothers averaged 1.32 visits per family.
At 2 years, 31 intervention (18%) and 25 control (16%) families did not return the questionnaire. There was no marked relationship between the chosen confounding variables and loss to follow-up in the intervention arm; but in the control arm, mothers of female children and those from a disadvantaged background and/or with lower levels of education were disproportionately more likely to be lost to follow-up (Table 2). However, none of these demographics was strongly associated with the baseline EPDS dichotomous measures (all P values were >.3), so we did not adjust for them in the regression analyses.
Table 3 shows the main outcome comparisons at 2 years of age. Mothers in the intervention group were significantly less likely to report clinical levels of depression symptoms than control mothers (EPDS >9: 15.4% vs 26.4%; adjusted odds ratio [OR]: 0.41 [95% confidence interval (CI): 0.20 to 0.86]; P = .02), and the mean depression scores were also lower for intervention compared with control mothers (EPDS score: 5.50 vs 6.72; adjusted mean difference: −1.47 [95% CI: −2.42 to −0.51]; P = .003). Even when applying the more stringent clinical cut point of EPDS of ≥13, mothers in the intervention group remained substantially less likely to report clinical levels of depression symptoms than control mothers (4.2% vs 13.2%; adjusted OR: 0.20 [95% CI: 0.07 to 0.60]; P = .004). Neither parenting practices (harsh discipline and nurturing) nor child mental health (externalizing and internalizing behavior problems) differed markedly between the intervention and control groups. Both intervention and control group T distributions for parenting and child mental health were comparable to community standardized norms (mean: 50 [SD: 10]).21,22
At 2 years, 27.3% (39 of 143) of mothers in the intervention group reported child sleep problems compared with 32.6% (42 of 129) of control mothers (adjusted OR: 0.83 [95% CI: 0.48 to 1.43]; P = .49). A smaller proportion of intervention rather than control mothers reported sleep problems that persisted throughout the entire study, that is, at all 3 of the 10-, 12-, and 24-month assessment times (11.2% [16 of 142] vs 21.7% [28 of 129]; adjusted OR: 0.51 [95% CI: 0.25 to 1.03]; P = .06).
Fifty-five mothers in the intervention group reported on how the infant sleep management strategies had affected the relationship quality with their child. Mothers strongly endorsed both behavioral interventions, with 84% (46 of 55) rating these as having had a positive effect on their relationship with their child. On the 10-point scale (on which 0 = very negative effect and 10 = very positive effect on relationship), the median score was 6.2 (interquartile range: 5.0–8.3), and the lowest rating was 4.8.
This is the first effectiveness randomized trial to report on long-term effects of behavioral interventions for infant sleep problems, delivered systematically at the population level. Sixteen months after delivery (when the child was 2 years old), the intervention still had a beneficial impact in reducing maternal depression symptoms, with the odds of reporting depression symptoms 59% lower for intervention relative to control group mothers. This was achieved although only 57% of mothers in the intervention group had chosen to take up the sleep intervention in infancy.
The intervention did not have a long-term impact (positive or negative) on parenting practices or child mental health. The majority of mothers in the intervention group reported that the sleep intervention in infancy had a positive effect on their relationship with their child. Sleep problems had largely resolved in both groups by 2 years, which is consistent with previous literature.1,3 Furthermore, only half as many mothers in the intervention group as control mothers reported sleep problems at all of the follow-up points (10, 12, and 24 months). Reduction in the prevalence of persistent sleep problems could at least partly explain the lasting reduction in maternal depression symptoms, because we have shown previously that it is persistent, rather than transient, sleep problems that predict poorer maternal mental health by the time the child is aged 2 years.1 It may also be possible that improved maternal mood and sleep in late infancy led to improvement in maternal parenting abilities and subsequently more positive mother-child interactions. In turn, this could have lead to lasting improvements in maternal confidence and mood at child age 2 years. The effectiveness trial was conducted in a “real world” setting, conformed to the rigorous standards of the Consolidated Standards of Reporting Trials statement, and used validated outcome measures for maternal depression, parenting practices, and child mental health. All of the clusters and 83% of families were followed up. Families who completed the follow-up did not differ from those that did not in terms of maternal depression scores or infant sleep severity. Thus, results are likely to generalize to English-speaking families and mothers with varying degrees of depression symptoms.
The study had some limitations. First, all of the outcomes were parent report measures, which may have biased results in the intervention group toward more favorable responses, because they could not be blinded. In our previous short-term follow-up at 12 months, however, mothers in the intervention group reported poorer physical health than control mothers, suggesting that response bias of mothers in the intervention group is unlikely.6 Second, sleep problem status and mother-child relationship were measured by maternal perception of a problem. Maternal report of a sleep problem, however, is a reliable indicator of sleep patterns in infants, as validated against overnight infrared video recording, actigraphy, and more detailed maternal ratings of shorter sleep hours, more frequent and prolonged night wakings, and longer settling times.4,18,23 A third study limitation was that very few mothers reported extreme depression scores, so findings cannot be generalized to mothers with severe depression or postpartum psychosis. Also, formal child attachment classification was not used in this large effectiveness trial, because its intensive direct observation protocol virtually rules out its use in population studies such as this. Finally, because not all of the mothers in the intervention group received the trial intervention, it might be surmised that simply talking to their nurse led to improvement in maternal mood. However, we did not offer, or have time to offer, any additional training or counseling in active listening strategies, which have sometimes been shown to be effective for postnatal depression.30 In addition, the mother's visits were short and highly structured and as such there was little time for the nurse to engage in any discussions other than those around infant sleep.
The long-term effects of this infant sleep intervention on maternal depression seem to outstrip those of other interventions, including those focusing specifically on maternal mood. Sixty-eight percent of high depression scores in the baseline survey at 7 months occurred in mothers who also report an infant sleep problem.6 Mothers who report an infant sleep problem, therefore, represent a readily identifiable “at-risk” subgroup for depression. Managing infant sleep represents a feasible, acceptable, low-intensity, and cost-effective preventive intervention approach for maternal depression.10 Because this approach has the potential to reach two thirds of the population experiencing depressive symptoms, it represents a major public health gain. In a systematic review of psychosocial and psychological interventions for preventing postnatal depression in the first year postpartum (when our intervention was delivered), the authors concluded that psychosocial intervention was as effective as standard care.31 Only 1 small randomized, controlled trial has examined the effects of antidepressant medication on postnatal depression and found that fluoxetine was as effective as cognitive behavioral therapy in reducing depression.32 The American Academy of Pediatrics has indicated concern about the use of antidepressant medications in breastfeeding mothers because of unknown effects on the developing child. Many breastfeeding mothers themselves report reluctance to take antidepressant medication.30 Thus, our effective approach to reducing postnatal depression involving infant sleep management may prove more acceptable to mothers in the first instance, because it does not stipulate antidepressant medication.
A brief behavioral intervention for infant sleep problems in the second 6 months of life delivered at a population level had a lasting impact on reducing maternal depression symptoms at 2 years. This demonstrates the capacity of a functioning well-child system to deliver effective, universally offered secondary prevention to reduce an important problem affecting a large proportion of the population. This approach could equally be delivered through other systematic primary care models, such as pediatric practices. Managing infant sleep problems should not be the primary clinical approach to postnatal depression, but at a population level it may be a very important component.
This project was funded by the Pratt Foundation. The salaries of Drs Hiscock and Ukoumunne are funded by the National Health and Medical Research Council Capacity Building Grant. Dr Bayer's salary is funded by the Colin Dodds Postdoctoral Fellowship from the Australian Rotary Health Research Fund.
We thank the Maternal and Child Health nurses and families of the cities of Bayside, Darebin, Hobson's Bay, Manningham, and Monash and the Shire of Yarra Ranges who took part in this study.
- Accepted May 7, 2008.
- Address correspondence to Harriet Hiscock, MD, Royal Children's Hospital, Centre for Community Child Health, Flemington Road, Parkville, Victoria 3052, Australia. E-mail:
The authors have indicated they have no financial relationships relevant to this article to disclose.
Dr Hiscock had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis; Drs Hiscock and Wake were involved in the conception and design of the study and obtained funding; Drs Hiscock, and Bayer, and Ms Hampton were involved with acquisition of study data and, together with Drs Ukoumunne and Wake, were responsible for analysis and interpretation of the data; Dr Ukoumunne performed the statistical analysis with assistance from Ms Hampton; and Dr Hiscock drafted the article with critical revision from Drs Ukoumunne, Wake, Bayer, and Hampton.
This trial has been registered as Current Controlled Trials ISRCTN 48752250, registered November 2004.
What's Known on This Subject
Up to two thirds of women who report symptoms of postnatal depression also report infant sleep problems. In the short term, brief behavioral strategies can improve infant sleep problems and associated maternal depression symptoms. The long-term effect of these strategies is unknown.
What This Study Adds
Brief behavioral strategies, designed to improve infant sleep, have sustained positive effects on maternal depression symptoms at a child age of 2 years. Managing infant sleep represents a feasible, acceptable, low-intensity, and cost-effective preventive intervention approach for maternal depression.
- ↵Wake M, Morton-Allen E, Poulakis Z, Hiscock H, Gallagher S, Oberklaid F. Prevalence, stability, and outcomes of cry-fuss and sleep problems in the first 2 years of life: prospective community-based study. Pediatrics.2006;117 (3):836– 842
- ↵Hiscock H, Wake M. Infant sleep problems and postnatal depression: a community-based study. Pediatrics.2001;107 (6):1317– 1322
- ↵Hiscock H, Bayer J, Gold L, Hampton A, Ukoumunne OC, Wake M. Improving infant sleep and maternal mental health: a cluster randomised trial. Arch Dis Child.2007;92 (11):952– 958
- ↵Buist A. Child abuse, post partum depression and parenting difficulties: a literature review of associations. Aust NZ J Psychiatry.1998;32 (3):370– 378
- ↵Buist A, Bilszton J. Beyondblue postnatal depression research program report. Volume 1. Melbourne, Australia: beyondblue, 2006. Available at: www.beyondblue.org.au/index.aspx?link_id=4.665&tmp=FileDownload&fid=348. Accessed December 21,2006
- ↵Hiscock H, Wake M. Randomized controlled trial of a behavioral infant sleep intervention to improve infant sleep and maternal mood. BMJ.2002;324 (7345):1062– 1065
- ↵Mrazek PJ, Haggerty RJ. Reducing Risks for Mental Disorders: Frontiers for Preventative Intervention Research. Washington, DC: National Academy Press;1994
- ↵Murray L, Ramchandani P. Might prevention be better than cure? Arch Dis Child.2007;92 (11):943– 944
- ↵Lam P, Hiscock H, Wake M. Outcomes of infant sleep problems: a longitudinal study of sleep, behavior, and maternal well-being. Pediatrics.2006;111 (3). Available at: www.pediatrics.org/cgi/content/full/111/3/e203
- ↵Australian Bureau of Statistics. Census basic community profile and snapshot. Canberra, ACT: Australian Bureau of Statistics. Available at: www.abs.gov.au/AUSSTATS/abs@.nsf/allprimarymainfeatures/C53F24374E92E8D3CA2573F0000DA0BE?opendocument. Accessed July 7,2008
- ↵Department of Human Services. Maternal and child health statewide data report 2002–2003. Victoria: Department of Human Services. Available at: www.cyf.vic.gov.au/_data/assets/pdf_file/0009/16992/mch_statewide_2002.pdf. Accessed July 7th2008
- ↵Ferber R. Solve Your Child's Sleep Problems. New York, NY: Simon & Schuster Inc;1986
- ↵Cox JL, Holden JM, Sagovsky R. Detection of postnatal depression: development of a 10-item Edinburgh Postnatal Depression Scale. Br J Psychiatry.1987;150 (6):782– 786
- ↵Murray L, Carothers AD. The validation of the Edinburgh Post-natal Depression Scale on a community sample. Br J Psychiatry.1990;157 (2):288– 290
- ↵Achenbach T, Rescorla L. Child Behavior Checklist for Ages 1 1/2–5. Burlington, VT: University of Vermont, ASEBA;2000
- ↵Sadeh A. A brief screening questionnaire for infant sleep problems: validation and findings for an Internet sample. Pediatrics.2004;113 (6). Available at: www.pediatrics.org/cgi/content/full/113/6/e570
- ↵Campbell MK, Elbourne DR, Altman DG. CONSORT statement: extension to cluster randomized trials. BMJ.2004;328 (7441):702– 708
- ↵Donner A, Klar N. Design and Analysis of Cluster Randomisation Trials in Health Research. London, United Kingdom: Arnold;2000
- ↵Goldstein H. Multilevel Statistical Models. 2nd ed. London, United Kingdom: Arnold;1995
- ↵Hanley JA, Negassa A, Edwardes MD, Forrester JE. Statistical analysis of correlated data using generalized estimating equations: an orientation. Am J Epidemiol.2003;157 (4):634– 675
- ↵Davison AC, Hinkley DV. Bootstrap Methods and Their Application. Cambridge, United Kingdom: Cambridge University Press;1997
- ↵Appleby L, Warner R, Whitton A, Faragher B. A controlled study of fluoxetine and cognitive-behavioral counselling in the treatment of postnatal depression. BMJ.1997;314 (7085):932– 936
- ↵American Academy of Pediatrics Committee on Drugs. The transfer of drugs and other chemicals into human milk. Pediatrics.2001;108 (3):776– 789
- Copyright © 2008 by the American Academy of Pediatrics