Parental bonding is defined as the affectionate tie a parent feels towards their infant. These feelings generally arise prenatally and increase in intensity over the course of pregnancy, both in mothers and in fathers. Bonding is fundamental for positive parenting behaviors and optimal child development. Few quantitative studies systematically examined what factors contribute to optimal bonding in fathers during pregnancy. The aim of the present study was to examine whether a variety of risk factors (symptoms of depression and anxiety, stress, and repetitive negative thinking) and resilience factors (self-compassion and trait mindfulness facets) were associated with prenatal paternal bonding and to explore potential mechanisms. In total, 66 expectant fathers with an average age of 33.7 years completed online questionnaires during the second or third trimester of pregnancy to assess the variables of interest. Analyses demonstrated that more symptoms of depression and anxiety and elevated levels of stress were associated with poorer bonding. In addition, less repetitive negative thinking, and more self-compassion and non-judging of inner experience were indirectly related to higher levels of paternal bonding through lower levels of stress. Findings suggest that expectant fathers who suffer from psychological distress, have more repetitive, negative, or intrusive thoughts, and tend to be more judgmental and self-critical may be at risk for suboptimal prenatal bonding. Providing adequate support for fathers at risk, perhaps in the form of mindfulness-based interventions, may be essential in order to cope with the challenges of new fatherhood and facilitate the process of bonding with the unborn child.
Opmerkingen
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Parent-to-infant bonding is defined as the affectionate tie a parent feels towards their infant (Bicking Kinsey & Hupcey, 2013; Suzuki et al., 2022). As parental feelings of bonding in general arise prenatally and increase over the course of pregnancy (De Cock et al., 2016; De Cock et al., 2017; Le Bas et al., 2021; Tichelman et al., 2019; Trombetta et al., 2021), they do not necessarily require a response from the child. Although parental bonding has been associated with prenatal parenting behaviors (Lindgren, 2001) and may be enhanced by interacting with the child after birth (de Waal et al., 2023b), the core definition of bonding as used in the current study concerns parents’ affections and feelings of love towards their child, i.e., emotional bonding (i.e., emotional bonding; Brandon et al., 2009; Cuijlits et al., 2016). Qualitative research suggests that men, like women, start bonding to their child before birth, reporting positive feelings towards the fetus (e.g., de Waal et al., 2023b; Habib & Lancaster, 2010; Lagarto & Duaso, 2022; White & Jarvis, 2024). At the same time, expectant fathers can feel detached or disengaged, especially in the beginning of pregnancy (de Waal et al., 2023b; Kowlessar et al., 2015b), whereas women have described feeling a relationship with their child from the earliest stages onwards (Darvill et al., 2010). Fathers may differ in their pace and process of bonding because they are not able to engage in a pregnancy at a similar level as their partner who is carrying the baby, and encounter barriers to be involved (e.g., limited paternal leave opportunities, feeling excluded at perinatal health care appointments; de Waal et al., 2023b; Steen et al., 2012; Widarsson et al., 2015).
High levels of prenatal and postnatal maternal-infant bonding guide caregiving behaviors after birth and thereby lay a foundation for positive parenting (Alvarenga et al., 2013; Medina et al., 2021; Nakic Rados, 2021) and healthy child development (e.g., Behrendt et al., 2019; Le Bas et al., 2020; McDonald et al., 2022). Although fewer studies have addressed this association in fathers, paternal bonding was found to predict better executive functioning in toddlers (De Cock et al., 2017) and is considered valuable for children’s long-term mental health and resilience (Winston & Chicot, 2016). Moreover, higher levels of paternal postpartum bonding have been associated with more supportive and positive parenting (Marshall et al., 2022), as well as with lower levels of non-responsiveness and parent–child dysfunctional interaction (Francis et al., 2023; Nakic Rados, 2021). In turn, fathers’ positive parenting practices have been found beneficial for children’s social, emotional, and cognitive outcomes (Amodia-Bidakowska et al., 2020; Cooke et al., 2022; Robinson et al., 2021). Thus far, only few quantitative studies systematically examined what factors contribute to optimal bonding from father to fetus and, therefore, correlates of paternal bonding remain largely unexplored (Suzuki et al., 2022). Examining correlates of prenatal father-infant bonding is essential for our understanding of the emotional connection from father to child. Moreover, identifying fathers who are at risk for suboptimal bonding may provide opportunities for intervention in the earliest stages of life.
A systematic review by McNamara et al. (2019) demonstrated that psychological distress is an important risk factor for poor pre- and postnatal bonding in mothers. Reduced emotional availability, a perceived inconsistency between expectations and reality, doubts about maternal competence, and the use of inadequate coping strategies may negatively affect bonding strength in parents who suffer from psychological distress (Cuijlits et al., 2019; Hornstein et al., 2006; Staneva et al., 2015). Fewer studies are available that have addressed this association in fathers and results are mixed. In a sample of 51 fathers exposed to contextual adversity (e.g., poverty, violence), Dayton et al. (2019) found that psychological distress (depression, anxiety, and posttraumatic stress) were unrelated to levels of prenatal paternal bonding. Similarly, in a study including 93 fathers, symptoms of anxiety were not associated with both bonding intensity and quality during pregnancy (Gobel et al., 2019). In contrast, De Cock et al. (2016) found that fathers with low levels of bonding throughout the perinatal period (n = 292) reported more anxiety and parenting stress compared to fathers with intermediate or high levels of bonding. Their findings were in line with research on postpartum bonding in fathers, demonstrating that postnatal depressive symptoms were associated with impaired bonding 6 months after birth (Kerstis et al., 2016). Further, Martin and Brock (2023) found that paternal stress during pregnancy mediated the association between the quality of partner support and paternal bonding impairments, with more stress being related to more impairments in paternal bonding 6 months postpartum.
Repetitive negative thinking, as both a symptom of and transdiagnostic factor for depression and anxiety (Klemanski et al., 2017; McLaughlin & Nolen-Hoeksema, 2011; Nolen-Hoeksema et al., 2008), may also put fathers at risk for suboptimal bonding either in a direct manner or as a precursor of psychological distress (Ehring et al., 2011; Hijne et al., 2020; Spinhoven et al., 2018). Repetitive negative thinking has been defined as having repetitive negative thoughts on problems or events in an intrusive way that are difficult to disengage from (Ehring et al., 2011). In mothers, overall higher levels of repetitive negative thinking throughout the perinatal period (Schmidt et al., 2017) and more ruminative thinking during pregnancy (Muller et al., 2013) have been associated with lower levels of postpartum bonding. In a low-risk sample of 130 fathers, repetitive thinking was not significantly related to postnatal bonding. However, there was a trend in the same direction as found for mothers, with more repetitive thinking being related to lower levels of bonding (Bohne et al., 2022). Nevertheless, also fathers have described the perinatal period as an emotional rollercoaster, feeling both happy and joyful but also anxious, insecure, and worried at times (Baldwin et al., 2019; de Waal et al., 2023b; Lagarto & Duaso, 2022; White & Jarvis, 2024). Therefore, self-regulation capacities concerning thoughts and feelings may be vital for fathers’ mental health and stress levels (Raes, 2012; Sorg et al., 2012), and subsequently the process of bonding to their child.
There may also be factors that protect against suboptimal paternal bonding, either directly or through reduced psychological distress. One of those factors may be self-compassion, which refers to responding with kindness and understanding to own inadequacies, failures, and suffering, rather than being judgmental and self-critical when challenges and difficulties arise in life (Neff, 2003). A construct that is inter-related with self-compassion is mindfulness, which is defined as a mental state of awareness of the present moment and acceptance of feelings, thoughts, and physical sensations without judging or directly reacting upon them (Bishop et al., 2004; Kabat-Zinn, 2015). Although there may be some conceptual overlap between self-compassion and mindfulness, their focus differs. Whereas mindfulness concerns an accepting awareness of a wide range of experiences either pleasant, neutral, or unpleasant, self-compassion is an emotion regulatory skill in response to personal pain, failure of suffering (Boellinghaus et al., 2014; Kabat-Zinn, 2015; López et al., 2016; Neff, 2003). Moreover, both were found to uniquely predict psychological distress (López et al., 2016; Van Dam et al., 2011) and shame (Sedighimornani et al., 2019) in a general population.
The perinatal period is characterized by considerable changes in various aspects of a parent’s life (Parfitt & Ayers, 2014), as well as adapting to a new situation and mastering new skills (Kowlessar et al., 2015a; Widarsson et al., 2015). Being kind to oneself and adopting a non-judgmental attitude in this process may protect fathers against elevated levels of stress and against developing mental health problems or other negative feelings that may eventually reflect negatively on the fetus (Britton et al., 2021). Although not examined in fathers specifically, self-compassion and mindfulness have both been associated with psychological distress in a general population of men and women (Britton et al., 2021; Freudenthaler et al., 2017; Raes, 2011). Moreover, in a sample of 82 expectant parents (of whom 40 were fathers), it was found that depressive symptoms were negatively related to bonding quality in parents with lower and average levels of mindfulness, suggesting that mindfulness may buffer against risk for poor bonding quality (Hicks et al., 2018).
Alternatively, self-compassion and mindfulness skills may prevent fathers from withdrawing from the pregnancy which is important because being aware of the fetus (e.g., seeing the fetus on ultrasound scans, hearing their heartbeat, feeling its movements) has been found to considerably contribute to paternal bonding quality (e.g., de Waal et al., 2023b; Lagarto & Duaso, 2022; Poh et al., 2014). In a more direct way, such skills may be essential in managing any inconveniences and adjustments or investments that have to be made regarding the pregnancy, thereby reducing psychological distress and improve bonding. Regarding self-compassion, Mancini et al. (2023) found that higher levels of self-compassion were related to higher quality of paternal bonding, as assessed in a sample of 175 fathers of children aged 0 to 5 years. Similar results have been found in a sample of 125 mothers of infants aged between 0 and 12 months, with more self-compassion predicting less impaired bonding (Fernandes et al., 2021). Regarding mindfulness, maternal trait mindfulness facets measured during pregnancy have been associated with higher levels of bonding as assessed prenatally (De Waal et al., 2024) and 3–18 months postpartum (Brassel et al., 2020; McDonald et al., 2022). To our knowledge, no further studies are available examining these constructs in fathers.
Although the literature described above provides valuable insights in the associates of parental bonding in general, the majority of studies focused on mothers and less is known about how fathers emotionally connect to their unborn child (Suzuki et al., 2022). At the same time, fathers’ process of bonding may differ considerably from that of mothers (i.e., lower levels of bonding; Lorensen et al., 2004; Ustunsoz et al., 2010), as may the vulnerability and resilience factors that are associated with bonding. Therefore, the aim of the present study was to examine correlates of prenatal paternal bonding in a community-based sample of expectant fathers. More specifically, we investigated whether a variety of risk (i.e., symptoms of depression and anxiety, stress, and repetitive negative thinking) and resilience factors (i.e., self-compassion and trait mindfulness) were associated with prenatal paternal bonding and explored different potential mechanisms. First, we hypothesized that more symptoms of depression and anxiety, elevated levels of stress, and more repetitive negative thinking would be related to lower levels of bonding, and that higher levels of self-compassion and more trait mindfulness would be related to higher levels of paternal bonding. Second, we hypothesized that less repetitive negative thinking and higher levels of self-compassion and trait mindfulness would be indirectly related to higher levels of paternal bonding through fewer symptoms of depression and anxiety and lower levels of stress.
Methods
Participants
Participants were partners of women who participated in the Brabant Study, a large prospective cohort study examining obstetric outcomes from a biopsychosocial perspective from 12 weeks of pregnancy up until 10 weeks postpartum (for the design of the study see Meems et al., 2020). Pregnant women of 18 years and older, who had a sufficient understanding of the Dutch or English language, were eligible for participation and were informed about the study by their midwife at their first antenatal visit. Exclusion criteria were a multiple pregnancy, known endocrine disorder before pregnancy (other than thyroid function problems), diabetes type I, rheumatoid arthritis, severe psychiatric disease (schizophrenia, borderline personality disorder, or bipolar disorder), HIV, drug or alcohol addiction problems, or any other disease resulting in treatment with drugs that are potentially adverse for the fetus and need careful follow-up during the pregnancy. Data collection took place between 2018 and 2023. As of January 2023, male partners of pregnant women were also invited to participate in a study on New Life and Fatherhood, preregistered at https://aspredicted.org/HXF_12H. The present study concerns a secondary analysis on baseline data of that study. Fathers were recruited via an online advertisement sent to all women participating in the Brabant Study. Additionally, women who were pregnant at time of recruitment and who had indicated having a male partner (n = 123) were approached by a research assistant to ask if their partner would be interested in participation. Fathers could then subscribe themselves via email. To participate in the study, fathers had to have a sufficient understanding of the Dutch language. No further in- or exclusion criteria were applicable.
Recruitment took place between January and May 2023. In total, 75 fathers consented to participate, of whom seven (9%) did not complete the baseline questionnaires and two (3%) were excluded from the present study as their partners already gave birth at the time of completing these questionnaires, resulting in a final sample size of 66 (88%) participants. Participants were on average 33.68 years old (SD = 4.11, range = 21–42). The majority of participants had a bachelor’s degree or higher (n = 43, 65%) and were of Dutch origin (n = 62, 94%). All fathers were married and/or cohabiting with their partner. At time of inclusion, their partners were on average 29.92 weeks pregnant (SD = 5.13, range = 20–40), and 33 participants (50%) were first-time fathers. For 88% of the participants (n = 58), the pregnancy was intended. Eight fathers (12%) had witnessed a miscarriage prior to this pregnancy. According to the DASS cutoff scores by Lovibond and Lovibond (1995), the overall majority of fathers demonstrated normal (n = 60, 91%) or mild (n = 4, 6%) levels of stress. Scores on depressive symptoms were equal to or greater than the cutoff of 10 (Edmondson et al., 2010; Shafian et al., 2022) for only 5 fathers (8%; the threshold of 10 is derived from research on paternal postpartum depression). A power analysis in the software program G*Power (Faul et al., 2007) indicated a detectable effect size of r = 0.15 in a sample of 66 participants (power = 0.80, Cronbach’s α = 0.05). The study was approved by the Ethics Review Board of Tilburg University (reference TSB_RP757). Fathers received a financial compensation of 10 Euros for completing the baseline questionnaires. All participants gave written informed consent. This project was funded by the Herbert Simon Research Institute, Tilburg School of Social and Behavioral Sciences.
Procedures
During pregnancy and at 8 weeks and 6 and 12 months postpartum, fathers completed online questionnaires on their computer or smartphone. All questionnaires were administered in the Dutch language. The present study solely used baseline data that were collected during pregnancy. Fathers were asked to report on general demographics (age, education), pregnancy related topics (parity, pregnancy intention, previous miscarriage, gestational age), bonding, symptoms of depression and anxiety, stress, repetitive negative thinking, self-compassion, and trait mindfulness. A maximum of two reminder emails were sent to participants after not completing the questionnaires within two weeks. Materials and analysis code for this study are available by emailing the corresponding author.
Measures
Paternal bonding
Paternal bonding was assessed with the Pre- and Postnatal Bonding Scale (PPBS), a self-rated questionnaire that has been developed to measure bonding quality both before and after childbirth (Cuijlits et al., 2016). The PPBS specifically focuses on the affectionate domain and holds an advantage over other measures that assess either prenatal or postnatal maternal-infant bonding (Wittkowski et al., 2020). Although not yet validated in men, good psychometric properties have been demonstrated for both pre- and postnatal administration in women (Cuijlits et al., 2016; Wittkowski et al., 2020). No alterations were made to the instructions for administration in men. The questionnaire consists of five items that describe positive feelings of bonding (e.g., loving) and are rated on a 4-point Likert scale ranging from 0 = not at all to 3 = very much, over the past 4 weeks. The PPBS is a short questionnaire with a user-friendly set-up that solely includes positively worded items to prevent socially desirable answers (Cuijlits et al., 2016). Although this could cause a ceiling effect in some cases, preventing bonding levels to further increase, previous research also demonstrated that particularly low scores on positive feelings reflected poor bonding (Brockington et al., 2006; Cuijlits et al., 2016).’ Total scores range from 0 to 15 with higher scores indicating more positive feelings of bonding. In the current study, the PPBS demonstrated excellent reliability with Cronbach’s α = 0.90.
Depressive symptoms
Depressive symptoms were measured using the self-rated 10-item Edinburgh Postnatal Depression Scales (EDS-10; Cox et al., 1987). The EDS has been specifically designed for administration in perinatal period and is a widely used, valid, and reliable measure to assess depressive symptoms during pregnancy and in the postpartum period (Bergink et al., 2011; Cox et al., 1987), also in men (Matthey et al., 2001; Mazza et al., 2022; Rao et al., 2020; Shafian et al., 2022). Participants were asked to indicate to what extent items (e.g., ‘Things have been getting on top of me’) applied to them over the past 7 days on a 4-point scale. Instructions of the original version for women were slightly adjusted for administration in expectant fathers, i.e., ‘Because you are expecting a baby soon, we would like to know how you feel.’ Total scores range between 0 to 30 with higher scores reflecting more depressive symptoms. In the present study, reliability was good with Cronbach’s α = 0.84.
Anxiety symptoms
Symptoms of anxiety were assessed with the 10-item anxiety subscale of the Symptom Checklist-90 (SCL-90), of which the Dutch version has demonstrated good reliability and validity (Arrindell & Ettema, 1981, 2003). The SCL-90 is a user-friendly and relatively short questionnaire that has been widely used in psychological research, also in perinatal cohort studies including men (e.g., Adhikari et al., 2021; Korja et al., 2018; Kotimaki et al., 2020). Participants are asked to indicate the extent to which ten symptoms (e.g., trembling) applied to them over the past week. Items are scored on a 5-point Likert scale ranging from 1 = not at all to 5 = very much. For the analyses, scores were summed, resulting in total scores ranging from 10 to 50 with higher scores reflecting more symptoms of anxiety. Internal consistency in the current study was excellent (Cronbach’s α = 0.93).
Stress
Stress was measured with the stress subscale of the 21-item Depression, Anxiety and Stress Scale (DASS-21; De Beurs et al., 2001; Lovibond & Lovibond, 1995). Previous research found good psychometric properties of the Dutch version of the DASS-21, also for the stress subscale specifically (De Beurs et al., 2001; Wardenaar et al., 2018). The questionnaire is frequently used to assess stress in pregnant women (e.g., Bryson et al., 2021; van der Zwan et al. (2019)) and in fathers (Schulz et al., 2024), also during pregnancy (Ghaffari et al., 2022). The stress subscale holds seven items (e.g., ‘I tended to over-react to situations’) that are rated on a 4-point Likert scale (0 = did not apply to me at all to 3 = applied to me very much, or most of the time) over the past week. For the analyses, items were summed, resulting in a total score ranging between 0 and 21 with higher scores reflecting more stress. In the current study, reliability was good (Cronbach’s α = 0.87).
Repetitive negative thinking
Repetitive negative thinking was measured with the Perseverative Thinking Questionnaire (PTQ; Ehring et al., 2012). The Dutch version of the PTQ has previously been validated in a sample of 1845 men and women (Ehring et al., 2012). The self-rated questionnaire consists of 15 items (e.g., ‘My thoughts repeat themselves’) that are rated on a 5-point Likert scale ranging from 0 = never to 4 = almost always. Total scores were calculated that ranged from 0 to 60 with higher scores indicating more repetitive negative thinking. Reliability of the questionnaire in the present study was excellent (Cronbach’s α = 0.96).
Self-compassion
Self-compassion was assessed with the self-rated Self Compassion Scale Short Form (SCS-SF; Raes et al., 2011). Previous research demonstrated good psychometric properties for the Dutch version of the SCS-SF in a sample of 456 men and women (Raes et al., 2011). The questionnaire holds 12 items (e.g., ‘I try to be understanding and patient towards those aspects of my personality I don’t like’) that are rated on a 5-point Likert scale ranging from 1 = almost never to 5 = almost always. For the analyses, total scores were computed that range from 12 to 60 with higher scores reflecting more self-compassion. In the current study, Cronbach’s α = 0.88, demonstrating good internal consistency.
Trait mindfulness
Trait mindfulness was measured by means of the Three Facet Mindfulness Questionnaire Short Form (TFMQ-SF), a self-rated questionnaire with good psychometric properties that has previously been validated among pregnant women (Truijens et al., 2016). The TFMQ-SF is derived from the Five Facet Mindfulness Questionnaire (Baer et al., 2006) and its Short Form (Bohlmeijer et al., 2011). The questionnaire gives rise to three subscales, each holding four items: acting with awareness (e.g., ‘I find myself doing things without paying attention’, reversed item), non-judging (e.g., ‘I make judgments about whether my thoughts are good or bad’, reversed item), and non-reacting (e.g., ‘I watch my feelings without getting carried away by them’). Items are scored on a 5-point Likert scale ranging from 1 = never or rarely true to 5 = very often or always true. Items on each subscale were summed, resulting in scores ranging between 4 to 20 with higher scores indicating more trait mindfulness. In the present study, internal consistency was adequate to good with Cronbach’s α = 0.89, 0.76, and 0.82, for subscales acting with awareness, non-judging, and non-reacting, respectively. Based on previous research that indicated the three facets of mindfulness (acting with awareness, non-judging, and non-reacting) to be different constructs (e.g., Boekhorst et al., 2020; Lu et al., 2019), these were included as separate variables.
Covariates
Based on previous literature, several confounders were considered. First, participants were asked to report on their age (continuous variable) and education level coded as 0 = low education (vocational degree or lower) and 1 = high (bachelor’s degree or higher), since both have been related to paternal bonding (Bohne et al., 2022; Camarneiro & de Miranda Justo, 2017). In addition, based on studies that associated pregnancy-related factors to bonding (Camarneiro & de Miranda Justo, 2017; Gobel et al., 2019), fathers indicated whether their partner was pregnant with their first child (0 = primiparous, 1 = multiparous), whether the pregnancy was intended (0 = yes, 1 = no), and if the father’s partner had ever experienced a miscarriage (0 = no, 1 = yes). Finally, as feelings of bonding are known to increase in intensity over the course of pregnancy (e.g., De Cock et al., 2016; Habib & Lancaster, 2010), gestational age was considered as a covariate.
Data Analyses
Preliminary analyses were performed in Statistical Package for Social Sciences (SPSS; IBM Corp. (2023)) Version 29.0. First, missing values were inspected. Missing value rates were as follows: one value was missing for paternal bonding, depressive symptoms, anxiety symptoms, and stress (2%), two values were missing for mindfulness facets acting with awareness and non-judging (3%), three values were missing for mindfulness facet non-reacting (5%), four values were missing for repetitive negative thinking (6%), and five values were missing for self-compassion (8%). Concerning the covariates, data were complete except for parity which had two missing values (3%). A Little’s Missing Completely at Random (MCAR) Test demonstrated that missingness was completely at random (χ2 (39) = 46.22, p = 0.199). Missing values were imputed by means of multiple imputation using predictive mean matching (PMM; Little, 1988) in the package multiple imputation by chained equations (MICE; Van Buuren & Groothuis-Oudshoorn, 2011) in R version 2022.07.1. Autocorrelation function plots (Azur et al., 2011) revealed convergence of ten imputed datasets. Second, descriptive statistics were requested and bivariate correlations in the complete cases and the imputed datasets were visually inspected to investigate whether relations were similar or deviant.
Next, a series of mediation analyses were performed in R (R Core Team, 2021) using the package Lavaan (Rosseel, 2012) which provides pooled estimates for direct and indirect effects of the imputed datasets. Due to power limitations caused by a small sample size and small expected effect sizes, only variables that were significantly associated with the outcome measure at a bivariate level (p < 0.10) were included in further analyses. In addition, separate analyses were conducted that included either repetitive negative thinking, self-compassion, or one of the mindfulness facets as the independent variable, symptoms of depression, symptoms of anxiety, or stress as the mediator variable, and paternal bonding as the dependent variable. Age and education of the father, parity, pregnancy intention, previous miscarriage, and gestational age were added as covariates only if these were significantly related to the dependent variable at a bivariate level. Standardized and unstandardized regression coefficients were assessed for total, direct and indirect effects. Because bootstrapping of standard errors of the indirect effects was not possible with Lavaan, confidence intervals (95%) for indirect effects in the multiple imputed datasets were estimated using the Monte Carlo sampling method (Metropolis & Ulam, 1949).
Results
Bivariate Analyses
Descriptive statistics and bivariate correlations of study variables and significantly related covariates are presented in Table 1. Correlations were comparable for the complete cases and the imputed datasets in terms of significance and effect sizes, indicating similar associations. Paternal bonding was negatively associated with symptoms of depression, symptoms of anxiety, and stress (all medium effect sizes), as well as with repetitive negative thinking (medium effect sizes). Self-compassion and the mindfulness facet non-judging were positive related to paternal bonding (both small effect sizes). The mindfulness facets acting with awareness and non-reacting were not significantly related to paternal bonding. Further, symptoms of depression, symptoms of anxiety, and stress were positively associated with repetitive negative thinking (all large effect sizes), and negatively with self-compassion (medium effect size for depressive symptoms and stress and small effect size for anxiety symptoms). Concerning the mindfulness facets, acting with awareness was only significantly and negatively related to stress (small effect size). Non-judging was negatively correlated to symptoms of depression, symptoms of anxiety, and stress (large effect sizes), whereas non-reacting was unrelated to the psychological distress variables. Based on these correlations, the main analyses included the following independent variables: depressive symptoms, anxiety symptoms, stress, repetitive negative thinking, self-compassion, and the mindfulness facet non-judging. Concerning the covariates, parity was significantly and negatively related to paternal bonding, with first-time fathers showing higher levels of bonding. Hence, parity was included in the analyses as a covariate. Age and education of the father, pregnancy intention, previous miscarriage, and gestational age were not significantly related to paternal bonding and were therefore not included in the analyses.
Table 1
Descriptives and bivariate correlations
M
SD
1
2
3
4
5
6
7
8
9
10
1
Paternal bonding
12.34
2.90
−0.36**
−0.31*
−0.44**
−0.40**
0.31*
0.15
0.36**
0.04
−0.30*
2
Depressive symptoms
3.42
3.70
−0.34**
0.67**
0.67**
0.55**
−0.35**
−0.15
−0.52**
−0.18
0.06
3
Anxiety symptoms
11.80
2.46
−0.31**
0.68**
0.66**
0.58**
−0.27*
−0.22
−0.66**
−0.05
0.04
4
Stress
2.94
3.36
−0.43**
0.67**
0.67**
0.54**
−0.36**
−0.29**
−0.52**
−0.00
0.14
5
Repetitive negative thinking
17.03
11.41
−0.38**
0.53**
0.57**
0.53**
−0.41**
−0.25
−0.58**
−0.08
0.07
6
Self-compassion
43.23
8.55
0.31*
−0.32**
−0.23
−0.34**
−0.39**
0.23
0.36**
−0.11
−0.19
7
Acting with awareness
14.52
3.64
0.14
−0.15
−0.22
−0.29*
−0.24**
0.22**
0.43**
−0.35**
−0.02
8
Non-judging
16.80
3.31
0.33**
−0.55**
−0.66**
−0.53**
−0.57**
0.34**
0.43**
−0.22
−0.07
9
Non-reacting
12.50
5.01
0.03
−0.18**
−0.06
−0.01
−0.08
−0.08
−0.34*
−0.20**
−0.21
10
Paritya
−0.30*
0.06
0.02
0.13*
0.06
−0.20**
−0.01
−0.18**
−0.07
Descriptives are pooled over ten imputed datasets. Correlations below the diagonal are pooled over ten imputed datasets. Correlations in the complete cases are presented above the diagonal.
*p < 0.05; **p < 0.01.
aCoded as 0 = primiparous and 1 = multiparous.
Mediation Analyses
Unstandardized regression coefficients for direct effects are presented in Table 2. Standardized regression coefficients of significant mediation models are visualized in Fig. 1.
Table 2
Standardized regression coefficients from mediation analyses predicting paternal bonding
Independent variable
c-path
c’-path
ab-path
a-path
b-path
β
SE
t
p
β
SE
t
p
CI
β
SE
t
p
β
SE
t
p
Repetitive negative thinking
−0.36
0.03
−3.23
0.001
Depressive symptoms
−0.26
0.03
−2.05
0.041
−0.10
0.02
−1.61
0.107
−0.06, 0.01
0.50
0.04
4.64
<0.001
−0.19
0.10
−1.48
0.139
Anxiety symptoms
−0.28
0.03
−2.06
0.040
−0.08
0.02
−1.13
0.260
−0.06, 0.02
0.57
0.02
5.48
<0.001
−0.15
0.16
−1.10
0.271
Stress
−0.21
0.03
−1.71
0.089
−0.15
0.02
−2.22
0.027
−0.08, −0.01
0.52
0.03
4.92
<0.001
−0.29
0.11
−2.29
0.022
Self-compassion
0.25
0.04
2.20
0.028
Depressive symptoms
0.16
0.04
1.39
0.165
0.09
0.02
1.81
0.071
0.00, 0.07
−0.32
0.05
−2.82
0.005
−0.28
0.09
−2.33
0.020
Anxiety symptoms
0.19
0.04
1.65
0.098
0.06
0.02
1.47
0.143
−0.00, 0.65
−0.24
0.03
−2.01
0.044
−0.26
0.14
−2.29
0.022
Stress
0.14
0.04
1.20
0.230
0.12
0.02
2.15
0.032
0.01, 0.09
−0.34
0.05
−2.97
0.003
−0.35
0.10
−3.12
0.002
Non-judging
0.29
0.10
2.50
0.012
Depressive symptoms
0.16
0.12
1.21
0.228
0.13
0.06
1.93
0.054
−0.01, 0.26
−0.52
0.12
−4.99
<0.001
−0.25
0.10
−1.84
0.066
Anxiety symptoms
0.15
0.14
0.99
0.324
0.14
0.08
1.52
0.129
−0.05, 0.31
−0.65
0.07
−6.81
<0.001
−0.21
0.18
−1.40
0.162
Stress
0.12
0.11
0.91
0.364
0.18
0.06
2.57
0.010
0.04, 0.31
−0.52
0.11
−4.91
<0.001
−0.34
0.11
−2.71
0.007
Statistics are pooled over ten imputed datasets. Significant effects are in bold. Associations are controlled for the effect of parity on the mediator and dependent variable.
Fig. 1
Standardized regression coefficients from significant mediation models predicting paternal bonding. Note. *p < 0.05; **p < 0.01. Standardized regression coefficients are only presented for mediation models that demonstrated significant indirect effects. Associations are controlled for the effect of parity on paternal bonding
×
Repetitive negative thinking
Repetitive negative thinking was significantly and negatively associated with paternal bonding. However, there was no indirect effect through depressive symptoms. Repetitive negative thinking was positively associated with depressive symptoms, but depressive symptoms were not related to paternal bonding in this model. Further, repetitive negative thinking was negatively and directly related to paternal bonding after controlling for depressive symptoms and parity. Similarly, there was no indirect effect of repetitive negative thinking on paternal bonding via anxiety symptoms. Repetitive negative thinking was positively related to anxiety symptoms, but anxiety symptoms were not associated with paternal bonding in this model. Repetitive negative thinking was negatively and directly related to paternal bonding after controlling for anxiety symptoms and parity. For stress, a negative indirect association was found, with more repetitive negative thinking being associated with higher levels of stress and subsequently with lower levels of paternal bonding. No direct effect was found of repetitive negative thinking on paternal bonding when controlling for stress and parity, indicating a full mediation effect of stress.
Self-compassion
Self-compassion was significantly and positively associated with paternal bonding. No indirect was found through depressive symptoms, although self-compassion was negatively associated with depressive symptoms and depressive symptoms were negatively related to paternal bonding. In addition, no direct effect was found of self-compassion on paternal bonding when controlling for depressive symptoms and parity. Similarly, no indirect relation was found between self-compassion and paternal bonding through anxiety symptoms, although self-compassion was negatively associated with anxiety symptoms, and anxiety symptoms were negatively related to paternal bonding. Further, self-compassion and paternal bonding were not directly related after controlling for anxiety symptoms and parity. There was a significant indirect association between self-compassion and paternal bonding via stress, with more self-compassion being related to fewer stress and subsequently to higher levels of paternal bonding. No direct effect was found of self-compassion on paternal bonding after controlling for stress and parity, indicating a full mediation effect of stress.
Non-judging
The mindfulness facet non-judging was significantly and positively associated with paternal bonding. No indirect association was found through depressive symptoms. Non-judging was negatively associated with depressive symptoms, but depressive symptoms were not related to paternal bonding in this model. In addition, no direct effect was found of non-judging on paternal bonding when controlling for depressive symptoms and parity. Similarly, no indirect effect was found of non-judging on paternal bonding through anxiety symptoms, nor was a direct effect found when controlling for anxiety symptoms and parity. Non-judging was significantly and negatively associated with anxiety symptoms, but anxiety symptoms were not related to paternal bonding in this model. There was, however, a significant indirect association between non-judging and paternal bonding via stress, with more non-judging being related fewer stress and in turn to higher levels of paternal bonding. No direct effect was found of non-judging on paternal bonding after controlling for stress and parity, indicating a full mediation effect of stress.
Parity
In all models, parity was significantly and negatively related to paternal bonding with first-time fathers demonstrating higher levels of bonding (B = −1.63, SE = 0.64, β = −0.28, t = −2.54, p = 0.011; with slight deviations in these statistics for other models). Conducting the analyses without covariates yielded similar results.
Discussion
The present study examined whether symptoms of depression and anxiety, stress, repetitive negative thinking, self-compassion, and different facets of trait mindfulness (acting with awareness, non-judging, and non-reacting) were related to paternal bonding in expectant fathers. Additionally, it was explored whether repetitive negative thinking, self-compassion and mindfulness facets were indirectly related to paternal bonding through psychological distress. Results indicated that fathers who reported more symptoms of depression and anxiety and elevated stress also demonstrated lower levels of bonding during pregnancy. In addition, less repetitive negative thinking, more self-compassion, and higher scores on the mindfulness facet non-judging were associated with better prenatal paternal bonding, and these associations were statistically mediated by lower levels of stress, but not by fewer symptoms of depression or anxiety. The mindfulness facets acting with awareness and non-reacting were unrelated to prenatal paternal bonding.
In line with the hypotheses and previous research in mothers and fathers (e.g., Hicks et al., 2018; Mancini et al., 2023; Muller et al., 2013; Schmidt et al., 2017), we demonstrated that fathers who reported lower levels of repetitive negative thinking and more self-compassion and non-judging skills demonstrated higher levels of bonding during pregnancy. In addition, as was expected, repetitive negative thinking, self-compassion, and non-judging were indirectly associated with paternal bonding through feelings of stress. Expectant fathers who reported having more repetitive, negative, or intrusive thoughts, and fathers who tended to be more judgmental and self-critical, experienced higher levels of stress. Although not previously examined in fathers specifically, earlier studies also associated these constructs with impaired psychological distress in both men and women (Britton et al., 2021; Freudenthaler et al., 2017; Raes, 2011). Our findings suggest that a tendency to evaluate oneself harshly and not being able to disengage from disruptive and negative thoughts may enhance emotional stress. In the prenatal period specifically, regulation of thoughts and emotions and responding with kindness and understanding to one’s failures or feeling inadequate may be vital in coping with the challenges and demands of upcoming parenthood (Penner et al., 2022; Steen et al., 2012; Widarsson et al., 2015). More specifically, adopting an accepting and understanding attitude towards any uncertainties, anxieties, or concerns (e.g., with regard to future parenting competencies, unavoidable emotional and social adjustments, the growth and health of the unborn child), rather than ruminating and being self-critical, may prevent fathers from developing elevated levels of stress.
Subsequently, elevated levels of stress were associated with expectant fathers reporting fewer positive feelings towards their unborn child. These findings were in line with the hypotheses, and, in part, with previous studies that related lower levels of paternal bonding to more stress during pregnancy and after birth (De Cock et al., 2016; Martin & Brock, 2023). Poor marital relations, a lower quality of partner support, and less pregnancy involvement have all previously been associated with elevated levels of stress in expectant fathers, potentially explaining the associations found in the present study (Baldoni et al., 2020; Boyce et al., 2007; Kuljanic et al., 2016; Martin & Brock, 2023). Healthy marital relations and mutual support may be vital because expectant fathers largely depend on their partner for windows to engage with the fetus. Expectant fathers who experience a lack of support from their partner during pregnancy may feel less equipped and confident in their role as a future father, potentially reflecting negatively on their relationship with the unborn child. It may also cause them to withdraw from the pregnancy and thereby miss out on opportunities that may facilitate the process of bonding (e.g., ultrasound scans, fetal movement; de Waal et al., 2023b; Lagarto & Duaso, 2022; Poh et al., 2014).
As expected, more symptoms of anxiety and depression were associated with poor paternal bonding. These findings were in line with previous research in fathers (De Cock et al., 2016) and mothers (McNamara et al., 2019), but not with other studies that did not associate psychological problems with poor bonding (Dayton et al., 2019; Gobel et al., 2019). Although one of these studies included a sample that was smaller and differed substantially from the present study (i.e., parents who were exposed to environmental stressors such as poverty and violence; Dayton et al., 2019), insignificant associations between anxiety and bonding were also found in a sample that was more comparable to that of the present study (Gobel et al., 2019). However, the current study focused specifically on parental feelings towards the child, whereas the questionnaire that was used in the study of Gobel et al. (2019) also assessed behavioral and cognitive components of bonding that may be more strongly affected by symptoms of anxiety (i.e., thinking about what kind of child the baby will grow into, wanting to hold the baby after birth; Condon, 2015; Condon & Corkindale, 1998). Nevertheless, our results suggest that elevated levels of depression and anxiety may be a risk factor for suboptimal paternal bonding. However, effect sizes for the associations of depressive and anxiety symptoms with paternal bonding were smaller compared to that of stress and they did not statistically mediate the associations of repetitive negative thinking, self-compassion, and non-judging with paternal bonding. Possibly, fathers feel less comfortable reporting on their mental health problems, particularly in the context of participation in or with regard to fatherhood (Baldwin et al., 2019; McKenzie et al., 2022; Reupert & Maybery, 2009), whereas stress, as a reflection of being overstretched or under pressure due to circumstantial issues, may be more accepted in men. Alternatively, the prenatal period may hold various potential stressors for expectant fathers (e.g., supporting their partner, making practical preparations, providing financially; Baldwin et al., 2019; Steen et al., 2012; Widarsson et al., 2015) that may potentially explain the different associations that were found in the present study.
Against expectations, acting with awareness and non-reacting were not associated with prenatal paternal bonding, providing evidence for a potential unique effect of non-judging on bonding quality in fathers. Although the results are difficult to compare with previous research with the majority of studies only including mothers, associations have been found between other facets of mindfulness and pre- or postnatal bonding (Brassel et al., 2020; Hicks et al., 2018; McDonald et al., 2022). However, in line with findings of the present study, one other study found a positive association between prenatal non-judging skills and bonding in expectant mothers, whereas acting with awareness and non-reacting were unrelated (De Waal et al., 2024). A possible explanation for these inconsistent results may be found in the different measures that were used. The studies that related multiple mindfulness facets to bonding used the Maternal Antenatal/Postnatal Attachment Scale (Condon, 2015; Condon & Corkindale, 1998), a questionnaire that takes parental behaviors into account in the assessment of bonding (e.g., eating healthy during pregnancy, tolerate behaviors of the infant that would otherwise be perceived as irritating or frustrating). In contrast, the current study used the PPBS which focuses on the positive affectionate domain since parental emotions and feelings of love towards the infant are considered the core definition of bonding (Brandon et al., 2009; Cuijlits et al., 2016). As non-judging is an element of mindfulness that also concerns cognitions and inner evaluations, this may explain why non-judging is uniquely associated with paternal bonding here, whereas the other two facets (acting with awareness and non-reacting) have a stronger behavioral component and may therefore be more important for parental behavior or the quality of parent-child interactions (Duncan et al., 2009; McKee et al., 2018; Parent et al., 2016). Moreover, since the perinatal period can be characterized by major adjustments, new responsibilities, and feelings of uncertainty in fathers (Baldwin et al., 2019; Shorey et al., 2017), being able to refrain from self-critical and harsh evaluations and judgments may be particularly valuable (Britton et al., 2021).
The present study contributed to the literature by examining whether a variety of risk and resilience factors were associated with prenatal paternal bonding. Strengths of the study involved the inclusion of the different facets of mindfulness as separate variables in the analyses, which were demonstrated to be differently related to paternal bonding, and the mediation analyses that were conducted in order to further clarify potential pathways to optimal bonding from father to child. However, there are also some limitations. First, a small number of fathers was included in the study, resulting in a limited power to detect small effects. Second, the sample was not representative of the general population with the majority of fathers being highly educated, of Dutch origin, and married and/or cohabiting with their partner (CBS, 2021). Additionally, it is conceivable that the fathers who agreed to partake in the study were already more interested in and involved with the pregnancy, or—since the majority was recruited via their partner—that they have relatively healthy marital relationships, further limiting the generalizability of the findings to the general population. However, prevalence of depression in our current sample was comparable with the average occurrence of depression in expectant fathers as shown by a systematic review (Mazza et al., 2022) and meta-analytic research (Rao et al., 2020). Third, a relatively large number of tests was conducted to analyze the different associations and potential indirect effects in a small sample, which increased the chance of false positive findings. However, the pattern of findings is rather consistent (mediation by stress, and not by symptoms of depression or anxiety across variables), which strengthens the suggestion that results are not just due to multiple testing effects. Fourth, the cross-sectional design of the study does not enable us to draw conclusions on the directions of the associations found. It can also be that poor bonding leads to disinterest, self-criticism, or self-doubt about the upcoming role as a father, which may make fathers vulnerable for psychological distress. Fifth, along with some of the other questionnaires used in the current study, the PPBS has not previously been validated for administration in men. Sixth, although no association was found between gestational age and paternal bonding, baseline questionnaires could be completed during the second or third trimester of pregnancy, leading to a rather large window of time of assessments. Longitudinal studies including a larger and more heterogenous sample of expectant fathers are required to further examine risk en resilience factors that are associated with prenatal paternal bonding. In addition, future research may consider including the facets of mindfulness as different constructs in their analyses as results from the present study suggest that these have unique effects. Further, future studies may examine the course of paternal bonding throughout pregnancy and also include postpartum measures of bonding, to examine whether psychological distress and mindfulness related variables may be predictive for bonding to the child after birth as well.
In sum, our findings suggest that expectant fathers who suffer from psychological distress or with elevated levels of stress may be at risk for poor bonding to their unborn child. In addition, less repetitive negative thinking, and more self-compassion and non-judging skills might be protective against elevated levels of stress and subsequently benefit prenatal paternal bonding. As parental bonding is important for parenting quality (Alvarenga et al., 2013; Medina et al., 2021; Nakic Rados, 2021) and optimal child development (e.g., Behrendt et al., 2019; De Cock et al., 2017; Le Bas et al., 2020; McDonald et al., 2022), the insights of the present study, if confirmed in larger and longitudinal studies, may have some implications for clinical practice by providing windows to identify fathers who are at risk for suboptimal bonding to their child during pregnancy. Adequate support and care for expectant fathers at risk may be essential in order to cope with the challenges of becoming a father. Although scarcely examined in fathers-to-be, mindfulness-based interventions may also be helpful in supporting at-risk fathers throughout the perinatal period and subsequently benefit their mental wellbeing and process of bonding to the unborn child.
Compliance with Ethical Standards
Conflict of interest
The authors declare no competing interests.
Informed consent
All participants gave written informed consent.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
share
DELEN
Deel dit onderdeel of sectie (kopieer de link)
Optie A:
Klik op de rechtermuisknop op de link en selecteer de optie “linkadres kopiëren”
Met BSL Psychologie Totaal blijf je als professional steeds op de hoogte van de nieuwste ontwikkelingen binnen jouw vak. Met het online abonnement heb je toegang tot een groot aantal boeken, protocollen, vaktijdschriften en e-learnings op het gebied van psychologie en psychiatrie. Zo kun je op je gemak en wanneer het jou het beste uitkomt verdiepen in jouw vakgebied.