Introduction

For most parents, pregnancy involves a psychologically intense internal process of emotional preparedness for parenting (Vreeswijk et al. 2014). For mothers, this process is stimulated, at least in part, by the physiologic transformations her body is experiencing (Slade et al. 2009). There is a growing body of research suggesting that fathers undergo a period of psychological preparation for parenting during the prenatal period as well (Vreeswijk et al. 2014). Pregnancy is, in fact, a developmentally critical time for both the parent and the baby, in part because it forms the foundation of the parent-infant relationship that ultimately supports healthy infant development during the postnatal period (Schore 2001). During pregnancy, parents form a mental and an emotional connection to their unborn child, which often begins once they see an ultrasound imagine of the baby, hear the baby’s heartbeat, and feel and see the movements of the fetus inside the mother’s womb (Heidrich and Cranley 1989; Walsh et al. 2014). This connection, or relationship, is referred to as the prenatal bond, and it plays an important role in the ways in which parents engage and interact with their infants once they are born (Parfitt et al. 2014).

Symptoms of parental psychopathology during pregnancy can disrupt the healthy development of the prenatal bond (Lefkovics et al. 2014) resulting in detrimental effects for both the parent and the baby (Alhusen et al. 2013). Specifically, depression symptoms, which are reported with relative frequency during pregnancy in expectant mothers and to a lesser degree in expectant fathers, can be particularly unfavorable harmful to the developing prenatal bond (Kim and Swain 2007; Murray et al. 1996). Expectant parents suffering from depression sometimes find it challenging to sense a connection with their unborn baby, may feel inadequate or unsure about their ability to care for their baby, experience excessive guilt about the various aspects of the pregnancy, or spend a significant amount of time thinking about possible negative events that could harm the baby. Through relationship mechanisms, these symptoms may ultimately affect the socio-emotional development of the infant (Junge et al. 2017).

Populations that experience increased stress due to factors such as living below the federal poverty line (Bolton et al. 1998), racial minority status (Rich-Edwards et al. 2006), and community and interpersonal violence exposure (Kendall-Tackett 2007) are found to be especially at risk for experiencing depression during pregnancy. Nevertheless, not all people who are exposed to stressful life events or depression may have difficulty forming an emotional and mental bond with their unborn baby. The question then becomes what resiliency factors may protect the prenatal bond in expectant parents experiencing risk? One potential factor is the ability to be mindful. Mindfulness is defined as the ability to intentionally shift awareness to the present moment in a non-judgmental and compassionate manner (Kabat-Zinn 1982). Higher levels of mindfulness have been associated with lower levels of depression symptomology and improvement of overall well-being in depressed adults (Segal et al. 2002), as well as in perinatal and postpartum samples (Dimidjian et al. 2015; Duncan and Bardacke 2010). Therefore, mindfulness may buffer the negative effects of depression in expectant parents, thereby protecting the emerging parent-infant relationship. The current study tested this hypothesis in a sample of expectant mothers and fathers who were considered at-risk for having parenting difficulties due to previous or ongoing exposure to contextual risk as a result of living in poverty.

Attachment theory provides a framework for understanding the role of the parent-child relationship in child development. It argues that parents develop internal representations of their children based on both their own childhood relational histories as well as current contextual factors (e.g., relationship with romantic partner, psychopathology) and that these representations serve to guide their parenting behaviors (Dayton et al. 2010). This relationship begins during pregnancy, continues to develop over time, and becomes the foundation on which children’s socio-emotional development is built. For mothers, the development and quality of the maternal-fetal relationship are linked to maternal postnatal feelings about the baby (Benoit et al. 1997), the quality of postnatal parenting behaviors (Dayton et al. 2010), infant attachment security (Fonagy et al. 1991), and infant developmental outcomes (Alhusen et al. 2013). Although less is known about paternal-fetal relationship development, there is new evidence suggesting that antenatal relationship formation for fathers is associated with postnatal father-infant relationship quality as well (Condon et al. 2013). These findings point to similar relationship development processes for mothers and fathers during the prenatal period and to similar infant outcomes associated with these processes.

Parents who experience greater ecological risks, such as lack of adequate prenatal or postnatal healthcare, experiencing community and interpersonal violence, and identifying as a racial minority, may be more likely to develop depression during the perinatal period (Kim and Swain 2007). Further, infants raised in high-risk environments are also more likely to experience difficulty forming secure relationships with their primary caregivers and therefore have more socio-emotional and behavioral difficulties than those in low-risk environments (Bakermans-Kranenburg et al. 2004). Therefore, understanding the potential resiliency factors in parents who raise their children in high-stress environments is imperative to improving both parenting and child outcomes.

Rates of depression in pregnancy range from 10 to 30% for expectant mothers (Ashley et al. 2016) and 8 to 16% for expectant fathers (Paulson and Bazemore 2010) with increased rates for parents from low socioeconomic backgrounds (Jallo et al. 2015). The clinical presentation of prenatal depression is often associated with persistence of symptoms during the postnatal period (Baker et al. 2015). Parents who suffer from mood disorders may have fewer positive feelings toward their infants. They may struggle with finding ways to be emotionally available to their infant, effectively mirroring their infant’s emotional states, and regulating their own negative emotions during times of distress (Woody et al. 2016). Further, for mothers, antenatal and postpartum depression (PPD) place the mother, infant, and the maternal-infant relationship at psychosocial and medical risk for poor outcomes, including preterm labor, low-birth weight, and increased risk of NICU admittance, which, additionally may disrupt the bond between the mother and infant (Field et al. 2004; Marcus 2009). Less research has been conducted using samples of fathers, but it is known that perinatal depression in fathers confers similar risks of poor father-infant relationship outcomes and poorer postnatal parenting quality (Letourneau et al. 2012). Further, antenatal and PPD in fathers contributes to negative behavioral outcomes in children, such as later conduct problems (Ramchandani et al. 2008). In this way, parental depression may represent one mechanism accounting for the intergenerational transmission of emotion regulation disruptions; infants of depressed parents may grow into children who struggle with managing their own emotions, resulting in increased rates of child behavior problems (Raskin et al. 2016).

Recent interventions for individuals with mood disorders have begun to integrate mindfulness practices into contemporary treatment modalities, and a growing body of evidence has supported the effectiveness of mindfulness-based interventions on depression both outside of (Segal et al. 2013) and during pregnancy (Dimidjian et al. 2015). Mindfulness is conceptualized as a dispositional (trait) characteristic and as a temporary state of mind that can increase through mindfulness practices (Baer et al. 2006). Little is known about the links between quality of the prenatal bond and mindfulness. However, dispositional mindfulness is related to better emotion regulation ability (Desrosiers et al. 2013), greater adult attachment security (Goodall et al. 2012; Melen et al. 2016), and lower depression symptoms (Dixon and Overall 2016). Caldwell and Shaver (2013), for example, found that adults who felt anxious and insecure in their romantic relationships also reported lower levels of dispositional mindfulness. Furthermore, trait mindfulness moderated the relationship of adult attachment and overall well-being by reducing the effect of insecure adult attachment on depression and anxiety symptoms (Davis et al. 2016). Additionally, research has demonstrated intergenerational transmission processes of attachment (Benoit and Parker 1994). Therefore, trait mindfulness may help increase the security of a parent’s attachment state of mind and thereby strengthen their ability to bond with their unborn infant.

The current study evaluated whether increased trait mindfulness in expectant parents supported the healthy development of a strong bond with the unborn infant. The primary aim of the study was to examine dispositional mindfulness as a possible moderator of the links between depression symptoms and quality of prenatal bonding in a sample of predominantly low-income, urban expectant mothers and fathers. We hypothesized that parents with high levels of depression symptoms and low mindfulness would have low levels of prenatal bonding quality, whereas parents with higher depression symptoms and high dispositional mindfulness would have higher levels of prenatal bonding quality.

Method

Participants

Participants for this study were drawn from a larger longitudinal study examining the influence of bio-psycho-social-spiritual factors on early parenting processes in pregnancy. Inclusion criteria were adult status, English fluency, a medically uncomplicated singleton pregnancy, and the participation of both biological parents; however, parents were not required to be in a sustained romantic relationship with each other. A total of 102 parents from a metropolitan/urban area participated in the study; however, the mindfulness questionnaire was added after the study commenced, so only a portion of the parents are included in this analysis (N = 82, 42 mothers and 40 fathers) aged 18 to 48 (M = 27.2, SD = 6.5). Demographics of the sample are presented in Table 1.

Table 1 Demographics (N = 82)
Full size table

Procedure

The research protocol was approved by the University’s Institutional Review Board and deception was not used. Parents were recruited from obstetrics clinics, social services agencies, and local community establishments using recruitment fliers and through online advertisements on the university website. Interested parents were screened for inclusion and exclusion criteria. Eligible parents were invited to the research laboratory, completed informed consent, and were told they were participating in a study on the influence of early experiences on the development of parenting thoughts and behaviors during pregnancy. Parents were offered a choice of a one-time or two-time visit totaling 5–6 h. The visit included qualitative interviews, quantitative questionnaires, and biobehavioral data collection as part of a larger study. Mothers and fathers completed all aspects of the research protocol separately, even though they attended the visit together. They did not complete the protocol in the same sequence and the quantitative questionnaires used in the current analysis took approximately 40 min to complete. As one parent completed the biobehavioral and qualitative portions of the protocol, the other parent completed a battery of self-report questionnaires and vice versa. Parents were given the option to complete the self-report questionnaires either at home or in-person at the laboratory visit. Most parents chose to complete the protocol during one visit and complete the questionnaires during the visit. This paper reports findings exclusively from the four self-report questionnaires described below. Parents were provided monetary compensation for their participation.

Measures

Demographics

Demographic information was collected from each participant. This included participant age, race, gender, income level, relationship status, and education.

Depression

Edinburgh Postpartum Depression Scale (EPDS; Cox et al. 1987) is a 10-item self-report questionnaire that assesses levels of depression symptoms over the past 2 weeks during the perinatal period in mothers and fathers (Madsen and Juhl 2007; Perren et al. 2005). Items highlight cognitive and affective symptoms of depression (e.g., “I have blamed myself unnecessarily when things go wrong” and “I have been able to laugh and see the funny side of things”) while de-emphasizing somatic symptoms. Individual items were scored 0–3 on a Likert scale and sum scores ranging from 0 to 30 are calculated. The EPDS is not a diagnostic tool; however, in community settings, a score of greater than nine is suggestive of depression symptoms that may indicate clinical levels of depression. Validation studies find reliability to be consistently high at .80 (Eberhard-Gran et al. 2001). This study’s Cronbach’s alpha was calculated as .83.

Maternal Quality of Attachment

Maternal Antenatal Attachment Scale (MAAS; Condon 1993) is a 19-item scale that taps maternal feelings, behaviors, and attitudes toward the fetus, mainly over the past 2 weeks. The questionnaire consists of the following: (1) the 10-item quality of attachment subscale which asks questions such as “Over the past two weeks I think of the developing baby mostly as...” and “from a real little person inside me with special characteristics to a thing not really alive” and (2) the 8-item time spent in attachment mode subscale which taps the intensity of the parent’s preoccupation with the fetus and asks questions such as “Over the past two weeks I have found myself talking to my baby when I am alone...”. The MAAS is scored on a 5-point Likert-scale: 1 (low attachment) to 5 (high attachment). Global attachment total scores ranging from 19 to 95 were calculated. The quality of attachment subscale was used for these analyses. Both subscales have been found to be negatively correlated with depression symptoms (Condon and Corkindale 1997) in (non-pregnant) adults. Internal consistencies of the overall scale and subscales were good, ranging from α = .69 to α = .84 (Schmidt et al. 2016; van Bussel et al. 2010). The Cronbach’s alpha for the quality of attachment subscale in this study was calculated as .81.

Paternal Quality of Attachment

Paternal Antenatal Attachment Scale (PAAS; Condon 1993) is a 16-item self-report questionnaire geared toward fathers and is similar to the MAAS but with the absence of three questions that relate the experience of the baby being inside the mother’s body. It focuses on feelings, attitudes, behaviors, and thoughts toward the unborn baby. Many of the questions require the respondent to select their answer based mainly upon their experiences within the previous 2 weeks. Like the MAAS, a total score and two subscales can be scored: quality of attachment and time spent in attachment mode. PAAS items were scored on a 5-point Likert scale: 1 (low attachment) to 5 (high attachment). The quality of attachment subscale was used for this analysis. Global attachment total scores range from 16 to 80. Internal consistency and reliability are acceptable at α = .80 (Condon 1993). The Cronbach’s alpha in this study was calculated as .66 for the paternal quality of attachment subscale.

The MAAS and the PAAS were combined for use in the analytic models. However, because the total scores for mothers and fathers have a different maximum (95 versus 80, respectively), participant scores on the MAAS and PAAS quality of attachment scales were standardized and the z-score was used so that data from both mothers and fathers could be analyzed together.

Dispositional Mindfulness

Five Facets of Mindfulness Questionnaire (FFMQ; Baer et al. 2006) is a 39-item self-report measure based on a factor analytic study of five independently developed mindfulness questionnaires. Analysis of these five mindfulness questionnaires yielded five factors that capture core aspects of mindfulness: (1) observing, (2) describing, (3) acting with awareness, (4) non-judging of inner experience, and (5) non-reacting to inner experience. Items were rated on a 5-point Likert scale ranging from 1 (never or very rarely true) to 5 (very often or always true). The scales may be combined for an overall level of dispositional mindfulness with a sum score ranging from 39 to 195. Prior work has established that the subscales demonstrate good internal consistency ranging from .75 (non-reactivity) to .91 (describing; Baer et al. 2006). The Cronbach’s alpha for the full scale for this study was .84.

Data Analyses

The probability of Mahalanobis distance was used to identify outliers (p < .001) and revealed that there were no outliers in the dataset. A post hoc power analysis for linear regression was conducted in G-POWER to determine that there was sufficient power for the proposed analyses using the sample size of 82, an alpha of 0.05, and a medium effect size (F2 = 0.15). The resulting power for the current analyses was 0.88 which is above the recommendation of 0.80 (Faul et al. 2008).

Distributional properties of the study variables were evaluated via histogram, normality plot, and diagnostic hypothesis tests of normality. Descriptive analysis of each variable (by gender and combined) is displayed in Table 2. Note that the only significant difference by gender is the mindfulness subscale of observing one’s own thoughts (t75.51 = 2.23, p = .04). First, a bivariate (Pearson’s r) correlational analysis was performed to evaluate the direct associations of study variables. Results are presented in Table 3. Then, a second bivariate correlation analysis was performed to evaluate the direct association between demographic variables (age, gender, ethnicity, income, education, parity) and the dependent variable, prenatal bonding, to determine if they should be controlled for in the primary analysis. The effects of depression symptoms on prenatal attachment, with dispositional mindfulness as a moderator, were analyzed using PROCESS macro (model 1) for a simple slope analysis (Hayes 2013). PROCESS uses a path analysis framework and utilizes bootstrapping to assess indirect effects. The use of bootstrapping is recommended in the literature since it does not require assumptions to be made about the shape of the sampling distribution (Preacher and Hayes 2004; Preacher et al. 2007). The estimated indirect effects were tested from a 95% level of confidence and a series of 10,000 bootstrap re-samples (Preacher et al. 2007) in which the sampling distribution of the conditional effect of depression symptoms on quality of prenatal attachment is not assumed to be normal. The model was run with quality of prenatal attachment as the dependent variable, parental depression symptoms as the independent variable, and dispositional mindfulness as the moderator. The interaction term was centered to reduce multicollinearity in accordance with Aiken et al. (1991) recommendations. The Johnson-Neyman technique was applied to identify regions in the range of the moderator variable (dispositional mindfulness) where the effect of the focal predictor (depression symptoms) on the outcome (quality of parental attachment) was statistically significant or not.

Table 2 Descriptive analysis of expectant parents who completed the questionnaires for dispositional mindfulness, depressive symptoms, and maternal/paternal quality of attachment (N = 82)
Full size table
Table 3 Correlation of depressive symptoms, dispositional mindfulness (total and subscale scores), and quality of attachment in expectant parents in a metropolitan Midwest city (N = 82)
Full size table

Results

Demographic findings are reported in Table 1. All variables met the assumptions of normality, linearity, and homoscedasticity. Correlations are reported in Table 3. Nearly a quarter (24.4%) of participants endorsed levels of depression symptomology greater than 9 and 14.6% endorsed levels of clinical depression indicated by scores of 12 or greater, which is typically seen in pregnant women with clinical depression (Murray & Cox, 1990). These findings are also similar to what is reported in lower socioeconomic populations (Bennett et al. 2004; Holzman et al. 2006; Yonkers et al. 2001). An ANOVA test revealed no differences in depression levels by gender. The average MAAS quality of attachment score was 46.23 (range 28–50) and PAAS quality of attachment score was 29.87 (range 23–36).

Depression symptom sum scores were negatively correlated with four of the mindfulness subscale scores: ability to describe one’s own thoughts, acting with awareness, ability to not judge one’s own thoughts, and ability to not react to one’s own thoughts. Depression symptoms were also negatively correlated with total mindfulness score and the overall quality of attachment to their unborn child. The parents’ quality of attachment was positively correlated with total dispositional mindfulness score and the mindfulness subscales of acting with awareness, ability be non-judging and non-reacting to one’s own thoughts. Parental age, gender, ethnicity/race, income, education, and parity were not significantly associated with prenatal attachment and were therefore not included in the primary analysis as control variables.

Results of the PROCESS analysis are depicted in Table 4; the moderation model is shown in Fig. 1. Higher depression symptoms were significantly associated with lower quality of prenatal bonding. Conversely, higher levels of dispositional mindfulness were significantly associated with higher quality of prenatal bonding. In line with our hypothesis, the interaction between depression symptoms and dispositional mindfulness emerged as a significant predictor of quality of prenatal bonding. Figure 2 illustrates the interaction by depicting the regression lines of the relation between quality of prenatal bonding and depression symptoms at high, medium, and low (+ 1 SD, mean, − 1 SD) scores of the trait mindfulness sum score. Decreases in the slope of the regression line with increasing mindfulness scores show that the relation between prenatal bonding and depression symptoms becomes weaker at average and high levels of trait mindfulness. It is important to note that parents grouped into the high depression symptoms category in the moderation analysis reported EPDS scores of 12.4 or above. Cox et al. (1987) considers EPDS total scores between 12 and 13 or higher to identify probable risk for major depression disorder. Using the Johnson-Neyman technique, we found that the conditional effect of current depression on quality of prenatal bonding transitioned to non-significance at a mindfulness value of 1.045, b = − .05, SE = .03, t = − 1.99, p = .05, 95% CIs: − .10–.00, at the 46th percentile of the distribution in our sample, with the relation between depression symptoms and quality of prenatal bonding scores significant at dispositional mindfulness scores below this threshold.

Table 4 The effects of depressive symptoms on prenatal attachment, with mindfulness as the moderator
Full size table
Fig. 1
figure 1

Moderation model: dispositional mindfulness moderates depressive symptoms on prenatal bonding in expectant parents

Full size image
Fig. 2
figure 2

Dispositional mindfulness in expectant mothers and fathers moderates the relationship between depression symptoms (Edinburgh Postpartum Depression Scale) and quality of attachment to their unborn baby (Maternal/Paternal Attachment Scale, mean-centered). Number of participants in each group: low depression group, n = 13; average depression group, n = 56; high depression group, n = 13

Full size image

Discussion

This study adds to our understanding of how trait mindfulness may protect against depression symptoms and improve parental relationship formation with the unborn infant during pregnancy. As predicted, dispositional mindfulness moderated the relationship between parental depression symptoms and quality of prenatal bonding. Increased levels of dispositional mindfulness were association with lesser depression symptomatology and prenatal bonding in this sample. Further, for the expectant parents who experienced high dispositional mindfulness, high levels of depression symptoms were not significantly associated with the quality of prenatal bonding. Essentially, dispositional mindfulness buffered the effects of depression on prenatal bonding. It may be that when a parent experiences depression, their ability to be present and reflective about their budding relationship with their unborn baby is reduced. However, if they inherently exhibit higher levels of mindfulness, they may be able to pay attention to their growing baby in a more present way, even in the face of some depression symptoms.

It is important to note that the parents in the high depression symptoms group reported scores of 12 and greater, which is often associated with a major depression diagnosis, and these parents reported levels of mindfulness significantly below the mean of this sample. Growing literature has established an inverse relationship between dispositional mindfulness and depression symptoms (Sugiura and Sugiura 2015); however, a mechanistic understanding of depression symptoms and dispositional mindfulness is not fully understood, especially in diverse, low-SES populations. It may be that it is not possible to experience both high levels of mindfulness and high symptoms of depression concurrently and that increasing trait mindfulness may innately protect individuals from depression, even without participating in mindfulness practices. If this were true, high trait mindfulness may be one explanation for why some individuals are resilient to depression even when faced with many contextual challenges.

Practicing mindfulness is linked to improved emotion regulation (Arch and Craske 2006), and mindful yoga is connected to an improvement in prenatal bonding with the fetus (Muzik et al. 2012). Prior work has established the likelihood that increasing levels of mindfulness through mindfulness-based interventions may improve well-being in relationships; however, it is not known whether improving mindfulness through intervention may influence the parents’ developing relationship with their unborn baby. In the current study, we measured the dispositional (trait) level of mindfulness and not the ability to evoke mindful attributes (state) and found that dispositional mindfulness did, in fact, moderate the relationship between prenatal bonding and parental depression. Kiken et al. (2015) found that increasing state mindfulness through practice also increased trait mindfulness, which contributed to a more mindful and less distraught disposition. Taken together, these findings support the development of future studies that examine the use of interventions designed to improve mindfulness in pregnancy. When used with groups of expectant parents, interventions, such as Mindfulness-Based Childbirth and Parenting Education (Bardacke 2012; Duncan and Bardacke 2010) or Mindfulness-Based Cognitive Therapy for Postpartum Depression (Dimidjian et al. 2015), for example, may reduce the negative influence of depression on the parent-infant relationship and on later child outcomes. Both interventions have adapted the well-validated Mindfulness-Based Stress Reduction curriculum for pregnancy and include specific practices that include noticing and observing the baby in utero.

Mindfulness-based interventions may be an effective mechanism by which to teach expectant parents to increase their state mindfulness, henceforth, decreasing the influence of higher levels of depression symptoms on bonding. In fact, previous research (Shapiro et al. 2011) found that mindfulness-based stress reduction, while more effective among individuals with higher levels of trait mindfulness, still accounted for significant state change among individuals with low trait mindfulness. Additionally, the ability to be mindful may aid parents in their ability to pay attention to their unborn baby in a more reflective way which may allow for increased feelings of bonding during pregnancy.

These findings are important as early bonding relationships set the stage for a range of developmental outcomes across a child’s life and are a significant protective factor among at-risk populations (Goyal et al. 2014; Huth-Bocks et al. 2004). Further, research has overwhelmingly identified the negative effects of depression on the development of early attachment relationships (Lefkovics et al. 2014), which poses a greater risk among populations identified with higher rates of depression. Research suggests that lowering depressive symptoms in pregnancy may initiate a more positive trajectory for increased parental bonding and improved infant development (Weikum et al. 2013). This study highlights the importance of mindfulness for expectant parents suffering from depression and suggests that utilizing mindfulness skills during pregnancy may improve the parent-infant relationship despite the presence of psychopathology. This may be especially important for parents from socioeconomically disadvantaged backgrounds as they are often faced with significantly higher levels of chronic stress, traumatic life events, and lack of resources and support (Creel et al. 2013; Frodl and O'Keane 2013).

Previous research on early attachment relationships and the impact of depression on early bonding relationships has focused almost exclusively on maternal infant bonding to the exclusion of fathers. One strength of this study is that we included both mothers and fathers, finding similar levels of depression and that mindfulness buffered depression for both parents. Further, traditionally, research on both mindfulness and prenatal bonding relationships has focused extensively on Caucasian, middle-class research participants. This study drew from a much more diverse sample including low-income, and ethnically diverse families, many of whom are exposed to toxic levels of stress and tend to experience higher rates of depression. These families, in particular, may benefit from mindfulness-based interventions to improve the likelihood of developing healthy prenatal parent-infant relationships.

Limitations

Limitations of the current study include the use of a small sample of convenience, the reliance on self-report measurement of study constructs, and the use of cross-sectional data. Ordering effects were partially controlled by alternating the order of questionnaires with the physiologic and qualitative portions of the protocol; however, this was not fully controlled as some participants chose to complete the questionnaires at home, and some chose to complete the entire protocol over two visits instead of one. There was also the potential of fatigue of the participants since the visits sometimes lasted as long as 6 h. Furthermore, differing results may be found using different mindfulness, depression, and bonding scales. Lastly, as the expectant mother and father share the commonality of the fetus, the potential for clustered findings exists. Future research may address this by examining partner effects of mindfulness on prenatal bonding.

Despite these limitations, our results suggest that interventions and/or workshops that increase mindfulness among expectant parents may simultaneously decrease both the development of postpartum depression and/or buffer the impact of depression on the developing parent-infant relationship. Mindfulness-based interventions with expectant mothers and fathers are recommended, especially for parents who are at higher risk for developing depression. Additionally, although a number of studies have identified the usefulness of mindfulness interventions in increasing state mindfulness and decreasing many mood-related symptoms, much less is known about the impact of intervention on trait mindfulness. Future research examines the relationship between state and trait mindfulness and the impact of intervention of these characteristics.

Future research also examines the efficacy of mindfulness interventions across multiple domains of parenting including actual observed parenting behaviors, longitudinal infant behavior, and development and associations of prenatal bonding with current child-parent attachment classifications. Further, evaluating the effects of mindfulness on parent-infant bonding in the presence of other psychological conditions, such as anxiety disorders or substance abuse, would help determine whether mindfulness skills buffer the effects of other psychological disorders on the parent-infant attachment relationship. More importantly, these associations should be evaluated from a diversity sensitive lens, taking into account culture, race/ethnicity, generational differences, and gender in both parents and children. For example, investigations examine whether mindfulness is experienced and expressed differently based on the cultural background of the parent, or among mothers versus fathers will contribute to our understanding of mindfulness within diverse samples of parents. Lastly, identifying other variables, such as overall family functioning or the co-parenting relationship, that impact the associations between mindfulness and bonding will be essential in our efforts to further tailor treatment services for parents from diverse socio-cultural backgrounds. Future research may also consider investigating mindfulness and its effects on prenatal bonding in different domains of depression (i.e., somatic, mood). This may be especially useful when considering the somatic effects of pregnancy on women. Much research to date is not mechanistic, and, therefore, a study that examines the effect of decreasing depression in pregnancy through mindfulness interventions and the effect of this on fetal programming, parental bonding, and infant behavior and development, would broaden our knowledge of the causal pathways influencing antenatal depression across the lifespan.

The results of this study highlight the importance of mindfulness as a moderator in the relationship between depression symptoms and prenatal bonding in a group of diverse, mainly low-SES, expectant parents. Mindfulness was found, as hypothesized, to be a significant protective factor against depression symptoms on bonding. Additional research conducted with different samples will provide further evidence for external validity and will improve generalizability. Further studies are indicated that explore longitudinal effects of improving mindfulness in pregnancy, especially in depressed parents, on infant behavior and development.