BACKGROUND. Pediatric anticipatory guidance has been associated with parenting behaviors that promote positive infant development. Maternal postpartum depression is known to negatively affect parenting and may prevent mothers from following anticipatory guidance. The effects of postpartum depression in fathers on parenting is understudied.
OBJECTIVE. Our purpose with this work was to examine the effects of maternal and paternal depression on parenting behaviors consistent with anticipatory guidance recommendations.
METHODS. The 9-month-old wave of data from a national study of children and their families, the Early Childhood Longitudinal Study, provided data on 5089 2-parent families. Depressive symptoms were measured with a short form of the Center for Epidemiologic Studies Depression Scale. Interviews with both parents provided data on parent health behaviors and parent-infant interactions. Logistic and linear regression models were used to estimate the association between depression in each parent and the parenting behaviors of interest. These models were adjusted for demographic and socioeconomic status indicators.
RESULTS. In this national sample, 14% of mothers and 10% of fathers exhibited levels of depressive symptoms on the Center for Epidemiologic Studies Depression Scale that have been associated with clinical diagnoses, confirming other findings of a high prevalence of postpartum maternal depression but highlighting that postpartum depression is a significant issue for fathers as well. Mothers who were depressed were ∼1.5 times more likely to engage in less healthy feeding and sleep practices with their infant. In both mothers and fathers, depressive symptoms were negatively associated with positive enrichment activity with the child (reading, singing songs, and telling stories).
CONCLUSIONS. Postpartum depression is a significant problem in both mothers and fathers in the United States. It is associated with undesirable parent health behaviors and fewer positive parent-infant interactions.
A now substantial body of literature documents the negative effects of parental (in most cases maternal) depression.1 Much attention has been placed on child outcomes, which have included decrements in social development, behavior problems, emotional difficulties, and a range of physical health problems.2 The long-presumed mechanism of these poor outcomes has been impaired parenting practices (eg, decreased display of emotion, involvement, and warmth), which result in disrupted parent-child relationships and consequent child difficulties. However, parenting-level mechanisms for the translation of parental depression to poor child outcomes have only recently undergone more systematic investigation.3
Anticipatory guidance for parenting infants is an area that directly addresses these mechanisms and has an established association with positive infant development. In a review article of health care services and promotion of child development, Regalado and Halfon4 report that pediatric anticipatory guidance in areas such as sleep habits, discipline, and reading effectively promotes child development in the first 3 years of life. More specifically, studies have shown that anticipatory guidance provided by pediatricians is associated with better infant sleep patterns, increased frequency of reading, and improved quality of parent-child interactions.5 The American Academy of Pediatrics anticipatory guidance recommendations for parents of children ages birth to 1 year include breastfeeding, putting the infant to sleep on its back, putting the infant to sleep awake, not putting the infant to sleep with a bottle, and engaging in positive interactive activities with the child.6 These interactive activities include reading stories and playing together, both of which have been found to promote language development, as well as social and emotional development in young children.7–9
Given the documented positive effects of anticipatory guidance on child development, it is important to examine factors that may prevent parents from following these recommended practices. Parental postpartum depression may be one such factor, because it has been linked to poor child health and development outcomes and compromised parenting practices.10 Depression is addressed variably in this literature and, although sometimes defined by a clinical diagnosis, is most often captured with self-report screening instruments on which scores above a given threshold are associated with positive clinical diagnoses.2,11 Maternal postpartum depression is prevalent, with 8% to 25% of women experiencing subclinical depressive symptomatology sometime during the first year postpartum.12 Several recent studies have found similar rates of postpartum depression in new fathers, particularly in families where the mother is also experiencing depression.13–16 However, to our knowledge, no study exists that examines paternal depression in the first year postpartum in a large, nationally representative sample. Postpartum depression in mothers has been linked with negative parenting behaviors associated with negative child outcomes.17,18 Studies have also found links between paternal depression and psychopathology in children and adolescents, but most of these studies did not focus on the postpartum period.19 Very little research to date has been conducted on the effects of paternal postpartum depression on fathers' parenting behaviors.
Several studies have examined the relationship between maternal postpartum depression and specific health-related parenting behaviors recommended by pediatricians, such as infant sleep position and breastfeeding. However, many of these studies have been limited by small samples and inconsistent definitions of health behaviors. Several studies have found that infants and young children of depressed mothers tend to have more sleep-related problems than those of nondepressed mothers.20–22 It has been suggested that these sleep problems may be a result of depressed mothers' use of ineffective sleep practices.23 In one study of mothers of 8-month-old infants, depressed mothers were more likely to sleep with their infants and nurse them to sleep, both behaviors that have been associated with later sleep difficulties.22 In addition, a study of single, low-income black women found that mothers with persistent depressive symptoms were less likely to use the back sleep position for their infants.24 To our knowledge, no study to date has examined the effects of paternal postpartum depression on sleep-related parenting behaviors in parents of infants.
The relationship between depression and breastfeeding is understudied and equivocal to date.25–27 Breastfeeding may positively influence a woman's mental state either through hormonal changes or emotional attachment. For example, hormonal fluctuations involving oxytocin and cortisol have been linked to improved mental health.28,29 Moreover, aspects of breastfeeding, such as physical proximity and touching, influence a mother's emotional attachment to her infant and may attenuate mental distress.30,31 On the other hand, several studies have found that maternal depression negatively impacts breastfeeding. Specifically, depressed mothers have been found to breastfeed for shorter durations, as well as to experience breastfeeding more negatively than nondepressed mothers.32–34 Only 1 known study has examined the impact of paternal depression on breastfeeding.35 This study found no effect of paternal depression on breastfeeding interruption; however, mothers who felt that their husbands actively supported breastfeeding had a longer duration of breastfeeding.
The existing literature suggests that depressed mothers may be less likely to follow anticipatory guidance recommendations regarding infant sleep practices and breastfeeding. In addition to these specific health behaviors, anticipatory guidance for parents of young children includes recommendations regarding positive parent-child interactions, such as reading and playing together. A large body of literature has demonstrated negative effects of maternal depression on the quality of mother-child interactions. In a meta-analysis of studies in this area,3 depressed mothers of infants and young children were found to be more irritable and hostile, to be more disengaged from their child, and to have lower rates of play and other positive social interactions with their child. Very few studies have looked at the impact of paternal depression on parenting practices related to parent-child interaction. Moreover, little research has examined the joint effects of maternal and paternal depression on parenting behaviors and on child outcomes. One notable exception is a study by Mezulis et al,36 which found that paternal depression during the postpartum period exacerbated the effects of maternal depression on later child behavior problems only if the father spent significant amounts of time caring for the child in infancy. In addition, being exposed to a nondepressed father did not buffer the effects of maternal depression even if they spent high amounts of time with their infants. In general, these findings suggest that father involvement in the postpartum period generally impacts children of depressed mothers but has little to no impact on children of nondepressed mothers. Only 1 known study has examined the relations between maternal and paternal depression and specific types of interactions, such as reading, with very young children.16 This study of parents of children ages birth to 3 years in a large national sample found that depressed mothers were less likely to play with or read to their children after controlling for social and demographic covariates. Paternal depression did not affect fathers' frequency of interactions with their children after covariates were controlled.
Overall, the current study intends to examine the extent to which postpartum depression among parents of infants influences their engagement in parenting behaviors that are consistent with anticipatory guidance recommendations and that have been associated with a stable household environment and child well-being. This study extends past research by examining the individual and interactive effects of both mothers' and fathers' depressive symptoms on their parenting practices during the postpartum period in a large, nationally representative sample. As in much of the previous literature, we define “depression” with a rating scale that characterizes individuals both in terms of symptom severity and with an empirically driven cut point, above which a respondent is likely to receive a clinical diagnosis of depression. On the basis of the findings of previous studies, we expected that mothers' and fathers' depression would negatively impact their parenting behaviors, such that parents who are depressed would be less likely to engage in positive health behaviors, as well as enrichment and play activities, with their infants than parents who are not depressed. In addition, we expected to find parenting behaviors to be the most compromised in families where both parents were depressed.
This study used data from the first wave of the Early Childhood Longitudinal Study (ECLS)-Birth Cohort (B).37 The ECLS-B is a multisource, multimethod study, conducted by the National Center for Education Statistics, designed to evaluate a range of influences on children's early development. The population from which the ECLS-B sample was extracted consisted of children born in 2001, with an oversampling of specific ethnic minority groups (eg, American Indian and Asian and Pacific Islander infants), low birth weight infants, and twins. Births were sampled within primary sampling units from the National Center for Health Statistics vital statistics system. The primary sampling units were stratified by geographical region, median household income, proportion minority population, and metro versus nonmetro area. More than 14000 births were sampled, yielding a final sample of 10688 completed parent respondent interviews at the 9-month data collection point. This represents a response rate of 76.8%.
Data were collected from mothers, resident fathers, nonresident fathers, and infants using a combination of computer-assisted personal interviews, self-administered questionnaires, and direct developmental assessments of the infants. The current study used data from the biological mother interviews and self-report questionnaires and resident father questionnaires. Because of the nature of our research question, we limited our sample to only those cases with complete mother and father mental health data. Of the 10688 completed parent interviews, 9447 had complete depression data available. Of these, we included only those cases where biological mothers were living with the target child (n = 9327) and excluded 462 cases that were duplicate data on twin infants, resulting in 8865 cases. We further narrowed down the sample to include only those biological mothers with corresponding resident father data, yielding a final sample size of 5089. Details on sample demographics and depression data appear in Table 1.
Depression for both mothers and fathers was measured by an abbreviated form of the Center for Epidemiologic Studies Depression (CES-D) Scale.38 The CES-D short form measures different depressive symptoms, including depressed affect, positive affect, somatic symptoms, psychomotor retardation, and interpersonal activity and has been used in past research studies39 and national health survey collection efforts.40 The CES-D short form contains 12 items with each item coded on a 4-point scale between 0 and 3. The range of total scores is 0 to 36, with a total score between 10 and 14 representing moderate depression and ≥15 representing severe depression. Cronbach's α for the study sample was .863 for the mothers and .862 for the fathers. Note that the term “depression” used throughout this article refers to individuals who report high levels of depressive symptomatology and is not intended to refer to individuals who are diagnosed with clinical depression.
CES-D results were used in 2 different ways in this study. For analyses using logistic regression, dichotomous depression scores for mothers and fathers were created using cut points established by the ECLS research team based on previous research. For linear regressions, continuous scores on the CES-D for mothers and fathers were used. This enabled us to examine the impact of both the likely presence or absence of depression and the influence of the degree of depressive symptoms on outcomes.
Six items from the biological mother interview were used to assess parental engagement in the following health behaviors: putting the child to sleep on its back, putting the child to bed without a bottle, putting the child to sleep awake, and breastfeeding. Mothers were asked to report the position in which they put their infant to sleep as a newborn and as a 3-month-old. Responses were combined into 1 item for sleep position and coded as putting the child to sleep on the back at both ages versus any other sleep position at either or both ages. Putting the child to bed with a bottle and putting the child to bed awake were both coded yes versus no. Breastfeeding was coded on the basis of mothers' responses to whether they had ever breastfed their infant (yes versus no).
Parent-child interactions were assessed with 7 items, reported by both mothers and fathers. Parents were asked how often in a typical week they read to their child, tell their child stories, sing songs with their child, and take their child on errands (not at all, once or twice, 3–6 times, or every day). These 4 items were dichotomized into every day versus less often. In addition, parents were asked how often in the past month they played peekaboo with their child, tickled their child, or took their child outside to walk or play (not at all, rarely, a few times a month, a few times a week, about once a day, or more than once a day). These 2 items were dichotomized into a few times a week or more versus less often. Similar methods for dichotomizing these items have been used in previous studies.41 The first 4 items have been used in the National Household Education Survey, a large, population-based survey administered every 2 years from 1991 to 2001. The last 2 items were used in the Early Head Start Research and Evaluation Project.40 All of the items except for “tickle child” were used as dependent variables in the logistic regressions. Tickle child was excluded because of lack of sufficient variability in response: 99% of mothers and fathers reported tickling their child a few times a week or more.
Because many of the items addressing parent-child interactions were similar, exploratory factor analysis42 was used to determine whether a more parsimonious approach to operationalizing parent-child interactions was feasible. The principal components method with varimax rotation was conducted on the 7 items described above for mothers and fathers separately. This revealed 2 coherent factors of parent-child interactions that were very similar for mothers and fathers. The 2-factor solution allows us to describe parent-child interactions with 2 variables rather than the original 7. This approach explains 45.84% of variance for mothers and 47.92% of variance for fathers. Factor 1 includes the following 3 items, all of which measure enrichment activities: read to child, tell stories to child, and sing songs with child. Factor 2 includes the following 4 items, which measure play activities: play peekaboo, tickle child, take child on errands, and take child outside to walk or play. Because of the logical consistency of their constituent items, we named these factors “enrichment” and “play.”
A number of demographic variables known to be associated with the outcome variables were controlled for in the analyses and operationalized as follows. Child gender was coded as male and female. Mother and father age were trichotomized into the following groups: 20 to 34 years, <20 years, and >34 years. Mother and father race were dichotomized into white and other racial backgrounds. Mother and father education were coded into 3 groups: some college or more, high school graduate or equivalent, and ≤12 years. Mother and father work status were coded into full-time employment, part-time employment, and unemployed. A composite variable measuring socioeconomic status (SES) was derived by the authors of the ECLS-B on the basis of household income, education, and occupation. This variable was coded into 3 SES groups, representing highest, middle, and lowest. Child birth weight was coded as normal or low, and birth status was coded as singleton or multiple birth. Household income was included as a continuous variable. Parity was coded as a continuous variable as the number of additional children residing in the household with the target child, which ranged from 0 to 6.
Descriptive statistics for all of the outcome variables were computed and compared for mother and father depression groups using the χ2 statistic (see Table 2). Logistic regression was used to model the relationship between maternal and paternal dichotomized depression scores and categorical outcome variables.43 Unadjusted and adjusted effects of depression were examined, and the final results are displayed in Table 3. In the initial adjusted model, all of the covariates described above were entered; however, mother and father age, father race, SES group, birth weight, and birth status (singleton versus multiple) were not significant and were dropped from additional analyses. All of the models were examined both with and without the interaction term (mother depression × father depression), and no differences in main effects were observed. Therefore, final reported results include the interaction term. The final results are reported in the form of odds ratios (ORs), allowing for ease of interpretation. All estimates are presented with 95% confidence intervals (CIs). All of the analyses were conducted using Stata 8 computer software (Stata Corp, College Station, TX),44 and adjustments were made to account for the sampling design used in the ECLS.
Following the logistic regressions described above, an additional set of analyses was conducted to examine whether the degree of maternal and paternal depression (measured continuously) might influence the degree of parent-child interaction measured by the 4 factor scores described above. Linear regressions were run on these variables to determine whether the degree of mother and father depression (measured using the continuous scale score on the CES-D) impacted the degree of parent-child interaction. These analyses used the continuous CES-D scores for mothers and fathers as the independent variables and the factor scores described above as the dependent variables. Both unadjusted and adjusted effects were examined, following the same format described above for the logistic regressions (see Table 4). All of the results discussed below are the results from the adjusted models.
Depression in Mothers and Fathers: Descriptive Statistics
On the basis of the dichotomous depression scores, 14% of mothers and 10% of fathers had moderate or severe depressive symptoms. On the continuous CES-D scores, mothers had a mean score of 4.58 (SD: 4.96) and fathers had a mean score of 3.69 (SD: 4.67). Both mothers' and fathers' continuous scores on the CES-D ranged from 0 to 36. χ2 analyses were conducted to examine simple associations among maternal and paternal depression and the outcome variables. For the purpose of these analyses, the sample was divided into the following 4 groups, on the basis of mother and father depression status using the dichotomized CES-D scores: neither parent depressed (N = 3981; 78.2%), mother only depressed (N = 586; 11.5%), father only depressed (N = 375; 7.4%), and both parents depressed (N = 147; 2.9%).
Maternal and Paternal Depression and Health Behaviors
Descriptive Statistics Based on χ2 Analyses
Significant differences were found for all of the health behaviors, with anticipatory guidance recommendations most likely to be followed when neither parent suffered depressive symptoms and least likely to be followed when both parents were depressed (see Table 2). Specifically, when both parents were depressed, the child was less likely to be put to sleep on his/her back (χ2  = 30.16; P < .001), was less likely to have ever been breastfed (χ2  = 25.39; P < .001), and was more likely to be put to bed with a bottle (χ2  = 45.74; P < .001). One exception to this pattern was found for putting the child to bed awake, which was most likely to occur when fathers only were depressed (χ2  = 9.62; P < .05).
After adjusting for covariates, several effects of parental depression on health behaviors were found in the logistic regressions (see Table 2). Depressed mothers were less likely to put their infants to sleep on their backs (OR: 1.40; P < .01), less likely to have ever breastfed their infants (OR: 1.48; P < .01), and more likely to put their infants to bed with a bottle (OR: 1.53; P < .01). Depressed fathers were more likely to put their infants to bed awake (OR: 0.66; P < .05). In addition, a significant interaction between maternal and paternal depression was found for putting the infant to bed awake, suggesting that when both parents were depressed, the infant was less likely to be put to bed awake.
Maternal and Paternal Depression and Parent-Child Interaction Behaviors
Descriptive Statistics Based on χ2 Analyses
On the basis of the χ2 analyses, significant or trend-level differences were found for 4 mother-reported parent-child interaction behaviors, with mothers being overall more engaged with their children when neither they nor their spouse suffered depressive symptoms. Interestingly, mothers were least likely to read to their child (χ2  = 19.06; P < .001), tell stories (χ2  = 15.45; P < .01), or sing songs (χ2  = 6.50; P < .10) when either they or the child's father (but not both) was depressed. As expected, mothers were least likely to play peekaboo with their infants when both parents were depressed (χ2  = 15.50; P < .01). No significant differences were found for mother-reported taking the child on errands, tickling the child, or playing outside with the child.
For father-reported enrichment/play activities, only 1 significant (play outside) and 1 trend-level (sing songs) difference was found. Fathers were most likely to play outside with their children when neither parent was depressed and least likely to do so when both parents were depressed (χ2  = 8.94; P < .05). In addition, fathers were most likely to sing songs with their infants when neither parent was depressed or when only the mother was depressed and were least likely to sing songs when both parents were depressed (χ2  = 6.66; P < .10).
After adjusting for covariates, several significant effects of maternal and paternal depression on parent-child interaction activities were found. Specifically, depressed mothers were less likely to tell their child stories every day (OR: 1.42; P < .05) and played peekaboo less often (OR: 1.57; P < .05). When mothers were depressed, fathers were less likely to sing songs with their children every day. Two significant effects of paternal depression were found: when fathers were depressed, mothers were less likely to tell stories to their child every day, and depressed fathers played outside less often with their children (OR: 1.42; P < .01). No significant interactions were found.
After adjusting for covariates, mother enrichment activities was predicted by maternal depression, as well as by a significant interaction between mother and father depression. A further analysis of this interaction revealed a stronger negative effect of maternal depression on mother enrichment activities when the father was not depressed than when the father was also depressed. Father depression predicted father enrichment activities, such that fathers who were more depressed engaged in less enrichment activities with their infants. No significant effects of maternal or paternal depression were found on mother and father play activities.
The effects of maternal and paternal depression on parenting practices related to positive child health and development has undergone very little investigation to date. The present study was among the first to examine the individual and combined effects of postpartum depression in mothers and fathers in a large, nationally representative sample of 2-caregiver families with 9-month-old children across the United States. Parenting behaviors that are emphasized in anticipatory guidance by pediatricians were the particular focus of this study. We predicted that both maternal and paternal depression would be negatively associated with parent health behaviors and positive interactions that are recommended in anticipatory guidance. Moreover, we predicted that maternal and paternal depression, together, would have a greater negative impact than just 1 parent being depressed. The findings that are discussed below are those that were observed after controlling for child gender, parental work, education, race, household income, and number of children.
The prevalence of postpartum depression in mothers (14%) reported in our study was consistent with other research and national estimates.12 Postpartum depression in fathers was strikingly high (10%) and more than twice as common than in the general adult male population in the United States.45 This finding is similar to the 1 previous national finding on this topic16 in that higher than expected rates of depression were found among fathers in the early parenting years. It adds to the body of knowledge, however, in that Lyons-Ruth et al16 included parents of children birth to age 3 years, and the current study focused on the postpartum period only. Because the first year of a child's life is particularly sensitive to parent-level influences, our current findings suggest the call for increased awareness of postpartum depression in men.
In support of the maternal depression component of our hypothesis, maternal depression had a strong association, overall, with fewer desirable health behaviors, including putting the infant to sleep in the back position less often, a lower likelihood of ever breastfeeding, and putting the child to bed with a bottle more often. These findings are consistent with past research showing that depressed mothers are less likely to engage in preventive health behaviors.17,18 In addition, these findings suggest that maternal postpartum depression may prevent mothers of infants from adhering to anticipatory guidance recommendations.
Contrary to our expectations, paternal depression was significantly associated only with a greater likelihood of the child being put to bed awake, which is consistent with anticipatory guidance for promoting good sleep habits in children. There was a similar trend for maternal depression. These counterintuitive findings (parental depression seems here to be associated with a desirable caregiving behavior) does not necessarily suggest that depressed parents attend more carefully to anticipatory guidance recommendations but may rather be an artifact of other behaviors in parents that are associated with paternal depression (eg, 1 parent puts child to bed awake to tend to the depressed parent's needs). Contrasting this, an interaction was observed between maternal and paternal depression on putting a child to bed awake, such that when both parents were depressed, the child was ∼3 times more likely to be put to bed asleep. This finding is consistent with past literature that suggests that negative effects of parental depression may be exacerbated when both parents are depressed.36
Overall, these findings suggest that higher levels of depressive symptoms in parents, particularly mothers, are associated with parenting practices that are inconsistent with anticipatory guidance recommendations and that have been linked to poorer health outcomes in children. Mothers who are distressed and experiencing the cognitive and psychosomatic symptoms of depression may be less able to attend to recommendations for even simple preventive measures. Although the general lack of significant association between paternal depression and health behaviors does not support paternal depression as a risk factor, our study is limited by the availability of only mother report for health behaviors.
In terms of the effects of parental depression on parent-child interaction, we found that maternal depression was associated with a lower likelihood of the mother regularly telling stories and playing peekaboo with the child. Maternal depression was also associated with a decreased likelihood of the father singing songs to the child every day. Paternal depression was associated with a lower likelihood of the father playing with the child outside regularly and a lower likelihood of the mother telling the child stories every day. These findings are of particular interest, because they suggest that maternal and paternal depression, in addition to negatively impacting the positive interactions of the suffering parent, have an impact on the other parent's interactions with their child.
Overall, our results suggest that where day-to-day interactions are concerned, depressed mothers and fathers engage in less positive interaction with their children, with a particular reduction in the degree of enrichment interactions, including reading, telling stories, and singing songs. These findings support the existing literature linking maternal depression to impaired parent-child interaction3 and extend the findings to include fathers as well. These findings are concerning, because they suggest that the pathway from positive parent-child interaction to developmental success46 may be jeopardized by depression in either parent. The moderation effect of paternal depression on mother enrichment behavior is unusual and has not, to our reading, been reported in previous literature. The effect is small and should be interpreted cautiously, but it is suggestive of a family compensatory model in which maternal depression maintains a negative association with enrichment when paternal depression is not pronounced, but this relationship goes to 0 when fathers are more depressed. As such, mothers “ignore” their depression to interact at baseline levels with their children when fathers might be more impaired by their own depression.
This study has several limitations. First, data from the 9-month wave of the ECLS are cross-sectional and permit inferences about association but not causation. With future waves of the ECLS birth cohort, however, prospective analysis and stronger causal assessments of the role of parental depression will become possible. In addition, this study used the CES-D to assess parental depression, a measure that does not provide an affirmed clinical diagnosis. CES-D scores reflect a range of depressive symptoms that, when above an empirically determined threshold, are strongly associated with a diagnosis of depression. Nevertheless, scores above the CES-D threshold are not fully analogous to clinically verified depression. Our measures of parenting behaviors are similarly limited in that they rely on parent self-report alone. Although parent report of behaviors has been found to correspond fairly well with observational measures, our data on health behaviors (ie, breastfeeding, sleep position, and sleep practices) only represent mother report, one possible reason why paternal depression seemed to have a largely negligible effect on these behaviors. Furthermore, the narrow range of response options given to parents may provide a generally weak measure of the association between depression and caregiving and/or parenting style. It is also important to note that depression and parenting behaviors were measured by self-report from the same source. Future studies should take these limitations into account and make use of more detailed parenting style and behavior instruments that are available47 and/or use supplementary observational or diary methods. Finally, this study did not measure whether parents actually received anticipatory guidance from their pediatricians on the relevant topics. Given the increased emphasis on anticipatory guidance in the American Academy of Pediatrics, it is reasonable to assume that most parents in the study did receive some information from their pediatricians. However, future studies are warranted that examine the relationship of parental depression to adherence to anticipatory guidance while considering amount of anticipatory guidance actually received. The overwhelming strength of this study was its careful sampling to allow for inference to the general population of the United States. Because the ECLS is an ongoing prospective study, additional waves of data will allow this study to be extended incrementally to provide a stronger picture of the role of depression in parenting behavior over time.
The primary practical application of these findings relates to identification and management of postpartum depression in both parents. Some attention (albeit insufficient) has been paid to the identification of and treatment for depression in mothers, particularly in the perinatal period. Our findings suggest that postpartum depression in both parents can interfere with the successful adherence to anticipatory guidance. Research has shown that pediatricians are increasingly delivering anticipatory guidance to parents of infants during well-child visits, although many parents report areas of unmet needs.48 Our findings suggest that postpartum depression in parents may be one factor that prevents parents from successfully applying their pediatricians' recommendations. Thus, whereas the message may be delivered to parents, it is not necessarily received. This supports the call for pediatricians to take more responsibility for the identification of depressive symptoms in mothers of young children.49–51 Training pediatricians in the diagnosis of depression may enhance their ability to recognize parental depression.49 Moreover, some research has supported the use of brief screening tools in pediatric settings in identifying mothers with depressive symptoms who may benefit from treatment.52 Considering the present findings here regarding the effect of depression on a father's parenting behavior, such efforts may bear investigation for both parents. Also, although the research on the effects of paternal depression on child outcomes is limited,19 current knowledge indicates that depressed fathers tend to have children with higher levels of physical and mental health difficulties, enhancing the urgency of catching this problem in families at the earliest opportunity.
- Accepted February 23, 2006.
- Address correspondence to James F. Paulson, PhD, Center for Pediatric Research, Eastern Virginia Medical School, 855 W Brambleton Ave, Norfolk, VA 23510. E-mail:
The authors have indicated they have no financial relationships relevant to this article to disclose.
- ↵Green M, Palfrey JS, Clark EM, et al. Bright Futures. Guidelines for Health Supervision of Infants, Children, and Adolescents. 2nd ed. Rev ed. Arlington, VA: National Center for Education in Maternal and Child Health; 2002
- ↵Bus A, Van Ijzendoorn M, Pellegrini A. Joint book reading makes for success in learning to read: A meta-analysis on intergenerational transmission of literacy. Rev Educ Res.1995;65 :1– 21
- ↵Lyons-Ruth K, Wolfe R, Lyubchik A, et al. Depressive symptoms in parents of children under age 3: sociodemographic predictors, current correlates, and associated parenting behaviors. In: Halfon N, McLearn KT, Schuster MA. Childrearing in America: Challenges Facing Parents With Young Children. New York, NY: Cambridge University Press; 2002
- ↵Leiferman JA. The effect of maternal depressive symptomatology on maternal behaviors associated with child health. Health Educ Behav.2002;29 ;596– 607
- ↵McLennan JD, Kotelchuck M. Parental prevention practices for young children in the context of maternal depression. Pediatrics.2000;105 ;1090– 1095
- ↵Hiscock H, Wake M. Infant sleep problems and postpartum depression: a community-based sample. Pediatrics.2001;107 :1317– 1322
- Taveras E, Capra A, Braveman P, Jensvold N, Escobar G, Lieu T. Clinician support and psychosocial risk factors associated with breastfeeding discontinuation. Pediatrics.2003;112 :108– 115
- ↵Falceto OG, Giugliani ERJ, Fernandes CLC. Influence of parental mental health on early termination of breast-feeding: a case-control study. J Am Board Fam Pract.2004;17 :173– 183
- ↵US Department of Education, National Center for Education Statistics. Early Childhood Longitudinal Study, Birth Cohort, 9 Month Data Collection. Washington, DC: US Department of Education, National Center for Education Statistics; 2001
- ↵Radloff LS. The CES-D Scale: a self-report depression scale for research in the general population. Appl Psych Measure.1977;1 :385– 401
- ↵McKey RH, Tarullo LB, Doan, HMFACES. The Head Start Family and Child Experiences Survey. Presented at: Biennial Meeting of the Society for Research in Child Development; April 15–18, 1999; Albuquerque, NM
- ↵Thompson B. Exploratory and Confirmatory Factor Analysis: Understanding Concepts and Applications. Washington, DC: American Psychological Association; 2004
- ↵Agresti A. An Introduction to Categorical Data Analyses. New York, NY: John Wiley & Sons, Inc; 1996
- ↵Stata Statistical Software [computer software]. Version 8.1. College Station, TX: Stata Corporation; 2004
- ↵Bavolek SJ, Keene RG. Adult-Adolescent Parenting Inventory (AAPI-2): Handbook. Asheville, NC: Family Nurturing Center; 1984
- ↵Heneghan AM, Silver EJ, Bauman LJ, Stein REK. Do pediatricians recognize mothers with depressive symptoms? Pediatrics.2000;106 ;1367– 1373
- Copyright © 2006 by the American Academy of Pediatrics