Master Thesis
Parental Bonding, adult attachment and its
cross-sex effects on anxiety and depression: a
retrospective approach
L.M. Roetman S1908030
Clinical Psychology
Supervised by M.L.J. Kullberg Department of Clinical Psychology Universiteit Leiden
2
Table of Contents
Abstract ... 3
Introduction ... 4
Parental Bonding and Anxiety ... 4
Parental Bonding and Depression ... 5
The Cross-Sex Component... 6
Adult Attachment ... 7
The Current Study ... 8
Methods ... 8
Participants & study design ... 8
Procedure ... 9
Measures ... 9
Statistical analysis... 11
Results ... 12
Demographics and Clinical Characteristics ... 12
Parental bonding and anxiety: mediation by adult attachment ... 13
Parental bonding and depression: mediation by adult attachment ... 14
Maternal and Paternal Bonding and Anxiety Symptoms: Moderation by Subject Sex... 16
Maternal and Paternal Bonding and Depression Symptoms: Moderation by Subject Sex .. 16
Discussion ... 17
Parental bonding, adult attachment style and levels of anxiety and depression ... 18
Cross-sex effects in parental bonding and levels of anxiety and depression ... 19
Strengths and Limitations ... 21
Clinical implications ... 22
Directions for future research ... 24
3
Abstract
Background: Research has repeatedly linked suboptimal parental bonding to adult anxiety and
adult depression. However, no studies have evaluated the mediating effect of adult attachment behaviours. Furthermore, to this date, very few studies have included both parental sex and subject sex and assessed potential differences between these unique relationships. Method:
This study attempts to cover this research gap by 1) building mediation models for anxiety
and depression, incorporating adult attachment as a mediator, and 2) building moderation models for anxiety and depression, with subject sex as moderator. Subjects were 2069 participants of the NESDA cohort study at the 9-year follow-up point. Results: The association between parental bonding and depression and anxiety symptoms was partially mediated by insecure adult attachment: people with suboptimal bonding were more likely to be insecurely attached as an adult, and consequently experienced more anxiety and depression. A significant interaction was only found for maternal bonding and subject sex, for anxiety only: females reported more anxiety symptoms compared to males after having suboptimal maternal bonding experiences. Conclusion: Adult attachment style and associated attachment behaviours partly explain why suboptimal bonding leads to higher scores on anxiety and depression. For women in particular, suboptimal maternal bonding experiences affect their anxiety levels in adulthood – unfavourable bonding experiences with the mother are thus more harmful for women, than they are for men. Our findings underline the importance of prevention and intervention strategies and once again highlight the detrimental effect of a negative parent-offspring relationship across the lifespan.
Key words: parental bonding, parental bonding instrument, adult attachment, anxiety, depression, cross-sex, gender
4
Introduction
Two types of mental disorders with the highest lifetime prevalence are mood and anxiety disorders (3.3% – 21.4% and 4.8% – 31.0% respectively; Kessler et al., 2007). Research has shown that both types inflict a large burden on communities worldwide (Murray & Lopez, 1996), and furthermore demonstrated that adequate treatment for these disorders is frequently lacking (Bijl et al., 2003). One of the aspects that is known to play a fundamental role in the potential development of mood and anxiety disorders is the family setting and, in particular, the parents (Tam & Yeoh, 2008). The quality and type of the parent-child interaction is called parental bonding and is a well-investigated topic within the field of mental health care, especially due to the previously mentioned societal burden and high prevalence (Tam & Yeoh, 2008). Elements that are important for this bond to be considered strong are presence of parental warmth, affection and involvement, and additionally the encouragement of autonomy and self-sufficiency (Rikhye et al., 2008).
Intertwined with parental bonding is attachment and attachment theory. This theory assumes that the way in which an infant organises its behaviour towards the mother influences the way in which it organises behaviour towards all other elements in its environment (Ainsworth, 1979). These early attachment experiences form a template, a so-called ‘internal working model’, that guides the infant’s future behaviour (Bowly, 1969). This internal working model influences the way a child approaches social interactions, impacts its emotional functioning and affects its level of cognitive capacities (Bowly, 1969). Over the course of the past decades, research has indicated that within child attachment theory four attachment styles can be distinguished: the secure attachment style, the ambivalent attachment style, the avoidant attachment style and the disorganised attachment style (Ainsworth, 1979; Main & Solomon, 1990). These attachment styles are known to be relatively stable from infancy to adulthood. (Waters, Hamilton & Weinfield, 2000).
Parental Bonding and Anxiety
Some types of attachment or specific parent-child interaction patterns have been pointed out as possible risk factors in mental dysfunction (Alonso et al., 2018). One of the impairments that is frequently linked to these impaired interaction patterns is adult anxiety (Alonso et al., 2018; Jinyao et al., 2012). Suboptimal parental bonding has been found to associate with
5
anxiety symptoms (Carter et al., 2001; Yap et al., 2014), generalised fear and fear of dying (Meites, Ingram, & Siegle, 2012). Yap and colleagues (2014) also found that substandard bonding experiences during childhood have an effect on anxiety symptoms in adolescents. Similar results were found by Khalid et al. (2018), though the effect was only present in the case of maternal bonding. Other studies confirm this aforementioned link between suboptimal maternal bonding experiences and anxiety symptoms in children and adolescents (Breinholst, Esbjørn & Reinholdt-Dunne, 2015; Van der Bruggen et al., 2010), nevertheless the conclusion is ambiguous, as similar effects for paternal bonding are found in other studies (Möller et al., 2013; Möller et al., 2016). Bögels and Phares (2008) also found that high levels of optimal paternal bonding had a general positive effect on internalising disorders and specifically anxiety. In an elderly sample, overall poor parental quality was associated with ever having received an anxiety diagnosis, but only for males (Burns et al., 2018). Sometimes, an age difference was found: in younger children, suboptimal maternal bonding was associated with anxiety, whereas for adolescents, suboptimal paternal bonding held this association (Verhoeven, Bögels & van der Bruggen, 2012). Finally, emotional- and physical maltreatment and neglect are strongly associated with anxiety symptoms later in life (Gallo et al., 2008). To summarise: some discrepancy regarding the effect of suboptimal parental bonding and its effect on anxiety symptoms exists, though it seems that especially the father seems to play an important role in reducing anxiety symptomatology.
Parental Bonding and Depression
A second form of psychopathology that is often associated with poor parental bonding is adult depression (Haaga et al., 2002). Del Barrio and colleagues (2016) found in their longitudinal study that suboptimal parental bonding was significantly related to the child’s risk to develop depression, one and two years later. McLeod, Weisz & Wood (2007) also reported a strong association between low scores on parental bonding and depressive symptomatology in children. Interestingly, the Del Barrio et al. (2016) study found a larger effect of maternal bonding on reducing depressive symptoms compared to paternal bonding. This parental sex effect is not found in every study; Burbach and colleagues (1986) for example, concluded in their literature review that suboptimal maternal bonding is related to lower levels of depressive symptomatology. Also Khalid et al. (2018) found no difference between suboptimal paternal and maternal bonding and its ability to predict depressive symptoms in an
6
adolescent population. An interesting effect related to the subject sex is found in an elderly population, where a larger effect of substandard paternal bonding and current symptomatology was witnessed, but in males only (Burns et al., 2018). The same study reports a clear link between poor parental quality and ever and current depression occurrence in an elderly population (Burns et al., 2018).
An overall good paternal bond is also found to be predictive of general quality of life in adulthood (Rikhye et al., 2002). Remarkably, de Minzi (2006) found in her study that in some cases, extreme scores indicating superb bonding could be associated with presence of depression in adolescents. Other studies, however, found a reverse effect and link these high, optimal scores to lower levels of depression (Yap et al., 2014). Literature about the impact of parental emotional- and physical abuse and neglect is clear in its conclusions: they are found to be clear predictors of depression in later life (Weich et al., 2009). To conclude, there is no general agreement when it comes to the quality of the parental bond and its association with depressive symptomatology in adults. Whereas some studies find an effect of the maternal bond in children, others find no difference or highlight the importance of the paternal bond.
The Cross-Sex Component
Following the previous two paragraphs it is evident that some interesting patterns regarding the influence of maternal and paternal bonding emerge, such as the larger buffer effect of maternal bonding on the development of depressive symptomatology in children, and the ambiguous sex-pattern when it comes to suboptimal parental bonding and anxiety in adolescents. Parent sex is thus certainly a factor that is incorporated in the topic of parental bonding and the development of psychopathology. Some studies took the sex of the offspring into account, but almost never a cross-sex approach -where differences and resemblances between the four distinct parent-child dyads (father-son, father-daughter, mother-son, mother daughter) is evaluated- is implemented. One of the few studies that incorporated sex of both the parent and child was an experimental study into child trait anxiety by Van der Bruggen, Bögels & Zeilst (2010). They found suboptimal parental bonding behaviours to have a larger effect on girls’ trait anxiety compared to boys’ trait anxiety. A second study incorporating all four dyadic relationships investigated effects of parental bonding behaviours on observed anxiety symptoms in pre-schoolers (McShane & Hastings, 2009). They found that internalising problems could be predicted from suboptimal maternal and paternal bonding
7
experiences, with the association for boys to be stronger than for girls. Optimal paternal bonding behaviour was associated with a decrease in internalising problems in girls only. Two previously mentioned cross-sex effecs are found in an elderly sample, where males only experienced increased depression symptoms in the case of suboptimal paternal bonding, and males only had increased chance on ever receiving an anxiety diagnoses after suboptimal parenting behaviours from either one of the parent (Burns et al., 2018).
Albeit the existence of research incorporating all four parent-child dyads, the question rises whether the relationships in the four dyads are actually different. A meta-analysis by Russell & Saebel (1997) attempted to investigate this, focusing on various aspects of the relationship itself (e.g. child’s acceptance of parent, parental warmth to the child, parent-child disagreements) and on child outcomes as a consequence of the interaction (for example witnessed effects on self-esteem). There was limited evidence for the ‘four distinct dyads hypothesis’, though, there was legitimate evidence for differences in separate, individual dyads. This could be a significant difference between two dyads (father – son, vs. father – daughter), dyad pairs (same-sex vs. cross-sex) or one dyad vs. the other three.
Adult Attachment
Although it is apparent that suboptimal bonding experiences affect adult psychopathology, it is not completely clear yet what role adult attachment plays. Some studies have indeed found that secure adult attachment is associated with general well-being (Kafetsios & Sideridis, 2006) and lower levels of depression and anxiety (Priel & Shamai, 1995), but did not incorporate a measure of parental bonding experience during childhood. Other studies have found similar results, but based on an adolescent sample (Muris et al., 2001). One study included both child- and adult attachment and found that subjects with secure adult attachment, but insecure child attachment, scored higher on depression compared to adults with insecure attachment patters during both time points (Pearson et al., 1994). The current study will therefore evaluate the role of adult attachment by incorporating this factor as an element in the analysis.
8
The Current Study
Research into specific parental sex and cross-sex effects has thus yielded some interesting results, showing that the relationships and evaluation of the relationships is different. Though, literature shows some discrepancies regarding the specific maternal and paternal bonding experiences and its effects on anxiety and depression symptoms. In addition to that, the number of studies investigating cross-sex differences is very limited. This study therefore attempts to shed a new light on these ambiguous results and ventures to add knowledge about the cross-sex effects to the existing literature. Third, most studies that have evaluated parental bonding and its effect on anxiety and depression outcomes use a child/adolescent population. By means of this study, we attempt to also capture this research gap by focusing on presence of depression and anxiety symptoms in adults aged 18 – 85. Fourth, the role of adult attachment is not yet understood – no research investigating adult symptomatology and parental bonding has incorporated a measure of adult attachment. We want to find out whether secure adult attachment plays a mediating role on adult symptomatology.
We hypothesise that subjects with optimal parental bonding will have lower levels of depression and anxiety symptomatology. Additionally, we will incorporate secure adult attachment as a potential mediator, and consequently hypothesise that subjects with secure adult attachment patterns will score lower on depression and anxiety symptomatology compared to subjects with suboptimal adult attachment patterns. In order to evaluate the cross-sex effects, a second model will be built, in which we look at the moderating role of subject sex. Based on our literature review, we hypothesise that male subjects with suboptimal maternal and paternal bonding will experience more anxiety compared to female subjects with suboptimal maternal and paternal bonding; no other cross-sex differences are expected. We furthermore hypothesise that male subjects with suboptimal paternal bonding will experience more depressive symptomatology compared to female subjects with suboptimal paternal bonding; no other cross-sex differences on depression are expected.
Methods Participants & study design
The subjects in this study are 2069 participants who are part of NESDA (NEderlandse Studie naar Depressie en Angst – translated: Netherlands Study of Depression and Anxiety), a
9
longitudinal cohort study investigating potential vulnerabilities, risk factors and course in the development of anxiety and depressive symptomatology (Penninx et al., 2008). The subjects that are part of this study are recruited from the general population, primary care practices and specialised mental health care institutions in the regions of Amsterdam and Leiden, and in the provinces of Groningen, Drenthe and Friesland. The original baseline sample consisted out of 2981 participants; the current study sample consists of participants who were part of the 9-year follow up assessment since the PBI was only conducted at this assessment wave.
Procedure
Information on psychopathology and childhood experiences, i.e. parental bonding, is collected through the administration of a self-report questionnaire and interviews. During this clinical visit, also data regarding physical health, health care utilisation and personal characteristics is registered.
Measures
Parental bonding In order to assess parental bonding, a 16-item version of the Parental Bonding Instrument (Parker, Tupling, & Brown, 1979) is utilised. This instrument is a self-report questionnaire, in which the subject must retrospectively score both the mother and father (separately) on whether various parenting behaviours and attitudes were present during their childhood. Each item was scored ‘a lot’ (1), ‘somewhat’ (2), ‘a little’ (3) or ‘not at all’ (4). Some items from the questionnaire were reverse keyed. A total sum score was calculated for the mother and father separately, as well as a sum score of mother and father combined. The latter was calculated by averaging the mother’s and father’s score for each question, and summing these averages. Scores could thus range from 16 to 64. High scores indicate suboptimal bonding, whereas low scores are indicative of a good parental bond. In the current sample the internal consistency of the PBI is good (maternal bonding: α = .88; paternal bonding: α = .88).
Anxiety symptoms Anxiety severity is measured using the Beck Anxiety Inventory (Beck et al., 1988). This is a 21-item self-report measure where the subjects has to report how often he/she was bothered by common anxiety symptoms during the past week. Items could be scored as ‘a lot’ (1), ‘somewhat’ (2), ‘a little’ (3) or ‘not at all’ (4). A total sum score was
10
calculated, with high scores meaning that many of anxiety symptoms had been present in the past week, and low scores meaning that one had experienced few anxiety symptoms in the past week. Scores could range from 16 to 64. The internal consistency was found to be excellent (α = .98).
Depressive symptoms Depression is measured with the Inventory of Depressive Symptomatology (Rush et al., 1986). This self-report measure is 30-item questionnaire assessing common complaints associated with depression in the past week. Participants were asked to score each item 0, 1, 2 or 3. For example on the item “Falling asleep”, the following answers were possible: ‘I never take longer than 30 minutes to fall asleep’ (0), ‘I take at least 30 minutes to fall asleep, less than half the time’ (1), ‘I take at least 30 minutes to fall asleep, more than half the time’ (2) or ‘I take more than 60 minutes to fall asleep, more than half the time’ (3). The scores could range from 0 until 90. High scores thus indicate that participants experienced a lot of symptoms associated with depression, and low scores indicate that participants experience relatively few symptoms associated with depression. The internal consistency was found to be excellent (α = .98).
Adult attachment Adult attachment is measured using a shortened version of the Experiences in Close Relationships Scale (Brennan, Clark & Shaver, 1998). The subject rates to what extent each of the 12 items applies to him/herself. Each item was scored between 1 and 7, with ‘strongly disagree’ (1), ‘disagree’ (2), ‘slightly disagree’ (3), ‘neutral/mixed’ (4), ‘slightly agree’ (5), ‘agree’ (6) ‘strongly agree’ (7). Some items from the questionnaire were reverse keyed. To get a final attachment score, the average was calculated for each participant, with high scores indicating optimal adult attachment experiences, and lower scores indicating less optimal adult attachment experiences. Final attachment scores could thus range from 1 until 7. The internal consistency for adult attachment was found to be acceptable (α = .76).
Lifetime occurrence of psychopathology For both anxiety and depression, lifetime occurrence is assessed through the administration of the Composite International Diagnostic Interview (CIDI) version 2.1 (World Health Organisation, 1997). Participants are assessed on whether they have reached the diagnostic criteria for depression, dysthymia or anxiety as
11
extracted from the DSM-IV (American Psychiatric Association, 1994) and ICD-10 (World Health Organisation, 1994). Participants were categorised as ‘lifetime depression/dysthymia’ (1), ‘lifetime anxiety’ (2), ‘lifetime comorbid depression/dysthymia and anxiety’ (3) or ‘healthy’ (4).
Statistical analysis
As a first step in the analysis the participants with no father- or mother figure present during childhood or participants who did not return the PBI questionnaire were excluded (N = 194). The assumptions for regression analysis were checked and found to be adequate. After exclusion of previously mentioned cases, the number of missing values left was substantially low (<10%) and was therefore neglected (Schafer & Graham, 2002). In the instance where scores were computed through summing or averaging the items, participants were incorporated only in the case all questions for that item were answered. After that, correlations between all study variables were calculated, i.e. maternal bonding score, paternal bonding score, anxiety score, depression score, adult attachment score, sex, age and education.
With respect to our research questions regarding the mediating effect of adult attachment, a total of six models were built. In order to give the reader a more substantial understanding of the concept of mediation analysis, a small explanation on the mechanism will be given first. Mediation entails that a predictor variable affects the outcome variable indirectly through a more direct cause, the mediator variable. It explains why the predictor and outcome variable are related. In Figure 1, a schematic representation of the mediation models as utilised in this paper is presented. X is the predictor variable (Parental Bonding), with mediator
M (Adult Attachment) and outcome
variable Y (Anxiety or Depression). The coefficient that is associated with the effects of the predictor on the mediator is presented as a, with b representing the coefficient that quantifies the effect of the mediator on the outcome variable. The total indirect effect is a × b. c’ is the direct effect of X on Y, when the mediator M is incorporated Figure 1 Illustration of the mediation model as used in this
12
in the analysis. Full mediation is confirmed if the introduction of te mediator would drop the direct effect c’ to zero. If introducing M reduces the direct effect but the effect is > 0 then partial mediation is confirmed. Z1, Z2, and Z3 represent the confounding variables that affect
the outcome variable but are not associated with the mediator or predictor variable.
As a first step in our analysis a model with total parental bonding score as a predictor and adult attachment as a mediator was created for anxiety and depression separately. As we found significant results, we decided to run two separate analyses for anxiety and depression, one with maternal bonding as predictor and one with paternal bonding as predictor. ‘Highest level of education achieved’, ‘age’ and ‘sex’ are used as covariates in all analyses. Mediation effects were assessed using model 4 provided by the PROCESS v3.3 macro by Andrew F. Hayes (Hayes, 2017) for SPSS version 26.0 (IBM Corp, 2019).
Third, in order to assess the effect of sex, a moderation model was built. In the first step of the analysis, the centred scores of the mothers and fathers were entered in the analysis as predictors, as well as the interaction terms (maternal bonding*subject sex and paternal bonding*subject sex). In the second step, the covariates ‘Highest level of education achieved’ and ‘Age’ were entered to control for confounding. All analyses were executed using SPSS version 26.0 (IBM Corp, 2019).
Results Demographics and Clinical Characteristics
In our sample we found mean score M = 31.30 (SD = 9.5) for maternal bonding M = 31.96 (SD = 9.27) for paternal bonding. Mean anxiety and depression scores were respectively 7.65 (SD = 8.41) and 14.77 (SD = 11.71). Of all participants, 15.5% (N = 290) was lifetime depressed, 9.5% (N = 178) had a lifetime anxiety disorder, 54.8% (N = 1028) had comorbid anxiety and depression and 20.2% (N = 379) had no lifetime anxiety or depressive disorder. During the past month, 15.5% (N = 291) of the participants had an anxiety diagnoses, whereas 11.0% (N = 207) reported a depression diagnosis. The adult attachment scores as measured by the ECR had a sample mean of 2.90 (SD = 0.84). An overview of all mean scores of our study variables and their correlations can be found in Table 1.
13
Parental bonding and anxiety: mediation by adult attachment
A regression analysis was executed and indicated a significant positive relationship between suboptimal bonding and anxiety (F(1, 1864) = 158.05, p < .001, R2 = 0.078) and suboptimal
bonding and depression (F(1, 1869) = 302.60, p < .001, R2 = 0.139). In other words:
suboptimal bonding scores thus predict presence of more symptoms associated with depression and anxiety. After we confirmed this simple linear regression, we continued with our mediation and moderation analyses.
Table 1. Descriptives and correlations between the study variables (N = 1875)
Variable name M (SD) Range Sex Age Educ. MB PB Dep Anx.
Sex 1.66 (.47) 1 - 2
Age 51.12 (13.15) 26 - 75 -.07**
Education 5.97 (2.02) 1 - 9 .015 -.25**
Maternal bonding score 31.30 (9.50) 16 - 64 .035 .18** -.08** Paternal bonding score 31.96 (9.27) 16 - 62 .012 .17** -.08** .56** Depression score Anxiety score Attachment score 7.65 (8.41) 14.77 (11.71) 2.90 (.84) 0 – 63 0 – 69 1.08 – 5.25 .055* .073** 0.05* .10** .034 .12** -.18** -.18** -.11** .34** .26** .39** .33** .24** .37** .78** .46** .34** Note. * = p < .05, ** = p < .01, ** = p < .001
As we wanted to test whether anxiety could be better explained through the effect of suboptimal parental bonding on adult attachment and adult attachment on anxiety, we decided to execute mediation analysis. We used step-by-step regression analyses to test for the potential mediating effect. Results indicated that parental bonding was a significantpredictor of adult attachment (b = 0.04, t(1738) = 18.96, p < .001), and that adult attachment was a significant predictor of anxiety (b = 2.46, t(1737) = 10.33, p < .001). With adult attachment incorporated into the model, we still find a significant effect of parental bonding on anxiety (b = 0.17, t(1737) = 6.94, p < .001), indicating that the mediation is partial. The coefficients can be found in Figure 2. The indirect effect was tested and yielded that it was significant (b = 0.10, SE = 0.01, 95% CI = [0.08, 0.13]). A Sobel’s test yielded significance of our mediation effect (z = 9.10, p < .001). Our results thus show that people with suboptimal bonding experiences during childhood have less optimal adult attachment styles, and consequently score higher op anxiety symptoms.
As we were interested in potential differences between father and mother, separate analyses were executed once the first model with a combined parental bonding score yielded a
14
significant mediation effect. In the case of maternal bonding, we found a significant effect of maternal bonding on adult attachment (b = 0.03, t(1715) = 16.59, p < .001) and a significant effect of adult attachment on anxiety (b = 2.60, t(1714) = 11.16, p < .001). The final step of the regression analysis revealed that, when incorporating adult attachment in our model, maternal bonding is a significant predictor of anxiety (b = 0.13, t(1714) = 6.34, p < .001), again proving partial mediation. Similar results were found in our third analysis: the coefficient between paternal bonding and adult attachment was significant (b = 0.03, t(1697) = 15.81, p < .001), just as the coefficient between adult attachment and anxiety (b = 2.71, t(1696) = 11.55, p < .001). Paternal bonding is a significant predictor of anxiety with adult attachment in our model (b = 0.12, t(1696) = 5.41, p < .001), indicating partial mediation. Evaluating the indirect effects, we find a significance for maternal bonding (b = 0.09, SE = 0.01, 95% CI [0.07, 0.10]) and paternal bonding (b = 0.09, SE = 0.010, 95% CI [0.07, 0.11]). For each of the two analyses a Sobel’s test was performed, yielding significance for both (z = 9.23, p < .001 and z = 9.26, p < .001, respectively for maternal and paternal bonding). Our results thus show that people with suboptimal maternal and paternal bonding experiences generally have suboptimal attachment styles in adulthood, and experience more anxiety symptoms. These mediation effects can be found in Table 2.
Table 2. Associations between the independent variables (X), moderator variable (M) and dependent variables (Y)
Effect of X on M (a) Effect of M on Y (b) Direct effect of X on Y (c’) Indirect effect (a × b) (95% CI) Total effect (c) Anxiety Maternal bonding 0.03*** 2.60*** 0.13*** 0.08 (0.07, 0.10) 0.22*** Paternal bonding 0.03*** 2.71*** 0.12*** 0.09 (0.07, 0.11) 0.20*** Parental bonding 0.04*** 2.46*** 0.17*** 0.10 (0.08, 0.13) 0.27*** Depression Maternal bonding 0.03*** 5.01*** 0.22*** 0.17 (0.14, 0.19) 0.38*** Paternal bonding 0.03*** 5.09*** 0.22*** 0.17 (0.14, 0.19) 0.39*** Parental bonding 0.04*** 4.74*** 0.30*** 0.20 (0.17, 0.23) 0.50*** Note. * = p < .05, ** = p < .01, *** = p < .001
Parental bonding and depression: mediation by adult attachment
Similarly to our previous anxiety analysis, mediation analysis was used to investigate whether adult attachment mediates the effect of parental bonding on depression. Results indicated that parental bonding was a significant predictor of adult attachment (b = 0.04, t(1742) = 18.99, p
15
< .001), and that adult attachment was a significant predictor of depression (b = 4.74, t(1741) = 15.18, p < .001). Incorporating adult attachment in our model, we still find a significant effect of parental bonding on anxiety (b = 0.30, t(1741) = 9.30, p < .001), indicating that the mediation is partial. In Figure 2 the coefficients are presented. We tested whether the indirect was significant, which was confirmed (b = 0.20, SE = 0.017, 95% CI [0.17, 0.23]). For this analysis our Sobel’s test was significant (z = 11.90, p < .001). Aforementioned results indicate that people with suboptimal parental bonding experiences during childhood, have less optimal adult attachment styles, and experience more depressive symptomatology.
Also, for depression, we executed separate analyses for maternal bonding and paternal bonding. We found a significant effect of maternal bonding and adult attachment (b = 0.03,
t(1718) = 16.62, p < .001) and a significant effect of adult attachment on depression (b = 5.01, t(1717) = 16.21, p < .001). In
our model where adult
attachment is incorporated,
maternal bonding is still a significant predictor of depression (b = 0.22, t(1717) = 7.91, p < .001), proving partial mediation. Finally in our third model, we found paternal bonding to be a significant predictor of adult attachment (b = 0.03, t(1701) = 15.81, p < .001), and adult attachment to be a significant predictor of depression (b = 5.09, t(1700) = 16.58, p < .001). Paternal
bonding is also a significant predictor of depression when adult attachment is part of our model (b = 0.22, t(1700) = 16.58, p < .001), thus indicating partial mediation. When looking at the indirect effects, we find a significance for maternal bonding (b = 0.17, SE = 0.02, 95% CI [0.14, 0.19]) and paternal bonding (b = 0.17, SE = 0.02, 95% CI [0.14, 0.19]). The Sobel’s test found significance for maternal bonding (z = 11.55, p < .001) and paternal bonding (z = Figure 2 Regression coefficients for the relationship between parental bonding
(total scores) and anxiety and the relationship between parental bonding (total scores) and depression, as mediated by adult attachment. * = p < .05, ** = p < .01, *** = p < .001
16
11.30, p < .001). People with suboptimal bonding experiences with their mother or father will thus have suboptimal adult attachment behaviours and experience more symptoms associated with depression. These mediation effects can be found in Table 2.
Maternal and Paternal Bonding and Anxiety Symptoms: Moderation by Subject Sex In our analysis we wanted to evaluate whether experienced psychopathology differed by being male or being female. We thus incorporated subject sex as a moderator in our model in order to evaluate its effect on parental bonding and anxiety.
In our model for anxiety, we observed that the covariates improved the model fit (age and level of education): ΔR2 = 0.03, F(2, 1790) = 25.88, p < .001, meaning that with increasing age and higher level of education achieved, anxiety scores were lower. In the full model, a significant interaction of subject sex and maternal bonding was found (t(1790) = 2.08, p = 0.037). This means that when looking at the relationship between maternal bonding and anxiety, it matters whether you are male or female. Being female with suboptimal bonding experiences was associated with higher levels of anxiety compared to males. We found no significant effect of paternal bonding and subject sex (t(1790) = -0.994, p = 320). This means that with respect to suboptimal paternal bonding experiences, levels of anxiety are the same for males and females. Figure 3 and Table 3 present the regression coefficients, standard errors and p-values for this moderation model for anxiety.
Figure 3 Regression coefficients for the moderation model
for anxiety. *= p < .05, ** = p < .01, *** = p < .001
Figure 4 Regression coefficients for the moderation model for
depression. * = p < .05, ** = p < .01, *** = p < .001
Maternal and Paternal Bonding and Depression Symptoms: Moderation by Subject Sex Also, for the link between bonding and depression, sex was investigated as potential moderator. The full model explained a significant increase in variance compared to the model
17
without the covariates (age and highest level of education achieved): ΔR2 = 0.02, F(2, 1794) =
22.54, p < .001. Education was found to be a significant covariate (p < .001). This indicates that with increasing level of education achieved, depression scores are lower. Contrary to our anxiety analysis, age was not a significant covariate in the model (p = 0.906), indicating that age has no effect on the depression scores. No significance regarding the interaction terms was found (p = .799 for the maternal bonding and subject sex interaction and p = .961 for paternal bonding and subject sex interaction). This indicates that for presence of depressive symptoms, neither for maternal bonding nor for paternal bonding it matters whether you are male or female. Figure 4 and Table 4 present the regression coefficients, standard errors and
p-values for this moderation model for depression.
Table 3 Variables and their coefficients, standard errors
and p-values for anxiety.
Variables B SE p
Maternal bonding .07 .04 .109
Paternal bonding .17 .05 .000***
Sex 1.13 .39 .004**
Maternal bonding * sex. .11 .05 .037* Paternal bonding * sex. -.05 .05 .320
Age -.04 .02 .015*
Education -.67 .09 .000***
Table 4 Variables and their coefficients, standard errors
and p-values for depression.
Variables B SE p
Maternal bonding .25 .06 .000***
Paternal bonding .26 .06 .000***
Sex 1.12 .527 .034*
Maternal bonding * sex. .02 .071 .799 Paternal bonding * sex. -.004 .073 .961
Age -.002 .020 .906
Education -.83 .126 .000*
* = p < .05, ** = p < .01, *** = p < .001 * = p < .05, ** = p < .01, *** = p < .001
Discussion
To our knowledge this is the first study exploring the relationship between parental bonding and the mediating effect of adult attachment on levels of anxiety and depression. Furthermore, this study is one of the first in its kind that assesses potential cross-sex effects in the parental bonding domain, and one of the first to examine an adult-age clinical cohort. We will discuss and interpret the results and additionally provide some suggestions and explanations of why the results could look like they do. Some limitations as well as some alterations and suggestions for future research will be presented. Finally, clinical implications of our study will be discussed.
18
Parental bonding, adult attachment style and levels of anxiety and depression
Our results confirm what has been found in the majority of research: people who had suboptimal bonding experiences with their father or mother during childhood, show higher levels of adult depression and anxiety (see for example Rikhye et al., 2008; Khalid et al., 2018). Secondly, we assessed the role of adult attachment in this relationship. We found that depression and anxiety were not only predicted by parental bonding, but also partly by adult attachment style. As hypothesised, we found that secure adult attachment was linked to lower levels of depression and anxiety symptomatology compared to individuals with insecure adult attachment patterns. Insecure adult attachment styles thus partly explain the relationship between suboptimal parental bonding experiences and psychopathology in adulthood. Bifulco and colleagues (2006) found a similar mediating role of adult attachment style in their study, where they attempted to predict the onset of new episodes of depression and anxiety in participants who had suffered childhood neglect and childhood abuse. They found that insecure adult attachment (partially) mediated the relationship between childhood adversity and depression and anxiety. Similarly to our results, adverse parenting experiences during childhood were found to lead to less optimal adult attachment behaviours and a consequent increased chance of experiencing a new episode of depression or anxiety.
The mediating effect of adult attachment style is thus an explanation of why suboptimal bonding is related to increased levels of anxiety and depression in adulthood. The mechanism behind this relationship can be explained through the ‘lifespan approach’ of attachment theory: early parental experiences form a template that guides the child’s future social interactions (Bowly, 1969). If these parenting experiences are unfavourable or inconsistent, children will start to evaluate themselves as unworthy and incompetent, and approach the world as if it is hostile and unsafe (Bretherton, 1992). This specific type of self-evaluation and way to approach the world will consequently shape the way in which one behaves towards others during adulthood: in an insecure fashion (Bretherton, 1992). This finding is confirmed by for example Lopez and colleagues (2000), who reported that people with insecure adult attachment styles reflect upon their early parental bond as less warm, more conflicted, more controlling and more invasive compared to people with secure adult attachment styles. The latter group would describe their early parental experiences as warm, positive and supportive. These securely attached adults possess a positive sense of self and a positive model of others:
19
they are comfortable with intimacy in their close relationships (Bartholomew & Shaver, 1998), consequently are socially competent and experience rewarding social interactions where emotions and feelings of distress can be safely shared (Wei, Russell, & Zakalik, 2005). These aspects are found to have a negative effect on the presence of depression (Wei, Russell, & Zakalik, 2005) and anxiety (La Greca & Harrison, 2005), as satisfying interpersonal relationships, emotional intimacy and emotional connectedness stimulate acceptance of the self by the self and validation of the self by others (Wei, Russell, & Zakalik, 2005). It is thus clear that suboptimal bonding experiences are related to insecure adult attachment behaviours, which consequently explain increased levels of anxiety and depression.
Cross-sex effects in parental bonding and levels of anxiety and depression
Our hypotheses regarding the cross-sex effects were not confirmed. We expected males with suboptimal parental bonding to experience more adult anxiety compared to females with suboptimal parental bonding, and we expected males with suboptimal paternal bonding to have more symptoms associated with depression compared to females with suboptimal paternal bonding. In fact, we found that females experience more adult anxiety compared to males after suboptimal maternal bonding experiences in childhood: no similar patterns for paternal bonding were found. We found no sex differences in depression outcomes, hence it does not matter whether you are male or female, or whether suboptimal bonding experiences were with the father or the mother, for experienced levels of depression.
This result thus highlights the importance of positive maternal bonding experiences for girls in particular. This importance of a good mother-offspring relationship has been previously confirmed by several studies (Khalid et al., 2018; Van der Bruggen et al., 2010), though they did not conclude that within this suboptimal maternal bonding experience, females experienced more complaints than males. This type of ‘gendered pathway’ into young adult symptomatology, however, was found in a study by Seiffge-Krenke & Persike (2017). The researchers analysed the importance of perceived support from fathers, mothers and peers (NRI; Furman & Buhrmester, 1985) in terms of companionship, nurturance and affection. The instrument that was utilised in their study has some overlap with the constructs measured by the PBI. They conclude that at age 23, females who received more support by mothers had lower stress levels and less symptomatology compared to males (Seiffge-Krenke & Persike, 2017).
20
A hypothesised explanation for this specific mother-daughter interaction effect is that parents differ in their parenting behaviours towards girls and boys, and that also a difference between mothers and fathers regarding boys versus girls can be expected (Möller et al., 2013). In female college students, positive perceptions of their mothers were related levels of anxiety, depression and self-esteem, whereas in the case of fathers only the negative perceptions were related to levels of depression (Renk et al., 2006). The positive and negative perceptions concerned concepts like involvement, autonomy and warmth (POPS; Phares & Renk, 1998), showing some overlap with the PBI. Similar to our findings, suboptimal experiences with the mother during childhood led to increased levels of both anxiety and depression, with a reversed effect for optimal maternal bonding experiences. When evaluating the father-daughter relationship and its effect on anxiety symptomatology, no significant relationship was found: only depression could be predicted from negative paternal perception. A possible explanation for this discrepancy is that mothers tend to spend more time with their children and/or adolescents, and consequently have a larger caretaking role than fathers (Möller et al., 2013; Renk et al., 2006). As especially is the case for daughters, mothers thus have a more influential role which could affect depression and anxiety outcomes, as well as levels of self-esteem (Renk et al., 2006).
The absence of an effect of a suboptimal father-son bond on heightened depression outcomes could possibly be explained by the time-period factor. Our study included participants aged 26 to 75, whereas the Burns et al. (2010) study (on which the hypothesis is mainly based) included participants aged 60 to 64. Fifty to sixty years ago, fathers generally played a small role in children’s upbringing, whereas in our age-cohort a large proportion of the participants will have had a more active father figure (Möller et al., 2013). If the father was not very present in general, suboptimal paternal bonding experiences are more salient and stand out more, without many opportunities for these unfavourable experiences to be countered by multiple other positive bonding experiences. Furthermore, if we dissect the suboptimal father behaviour, it becomes clear that especially overprotectiveness negatively affected depression outcomes in males (Burns et al., 2010). As protectiveness for sons is in general lower compared to protectiveness for daughters (Khalid et al., 2018), and especially from the father (Möller et al., 2013), this salient overprotection in the elderly age group is a potential explanation of the mechanism behind a suboptimal father-son bond and increased depression
21
symptomatology. Consequently, the more contemporary tendency for a father to be actively present in child rearing could explain the absence of this effect in our study.
Strengths and Limitations
As stated before, this study is unique in the sense that it assesses the relationship between parental bonding and psychopathology in a very diverse adult-age cohort with a considerably large sample size (N = 2069). In addition to that, we differentiated between the four distinct parent-offspring dyads and evaluated how these could differently affect anxiety and depression outcomes. The latter could simultaneously also be considered one of the limitations that should be taken into consideration when interpreting the results. Although we were of course interested in the discrepancies between the four different dyads, parents generally function as an interacting team when raising children (Majdandžić et al., 2012). It could thus be of importance to examine bonding behaviours in a co-parental context. An example can be found in the case of paternal affection. Paternal affection is generally associated with good mental health outcomes for the children. The opposite, however, is the case when the father shows high levels of affection, but the mother shows low levels of affection. In these cases an increasing number of family problems (such as emotional problems or household conflicts) were witnessed, mediating child anxiety and depression (Jorm et al., 2003). The large discrepancy in levels of affection is related to more family problems, which could explain why children in these cases in fact experience more anxiety and depression. Here, it is important to note that one can speculate about order of appearance of these phenomena. Second, our study uses a retrospective approach, making it possible that current mood or other (more recent) interactions with the parent could have biased the reports. This potential bias, however, is generally small, even when a large change in current mood state occurrs (Lewinsohn & Rosenbaum, 1987). Third, the high percentage of present psychopathology in our sample is most probably due to the sampling method. As a substantial part of our sample was generated from primary care practices and specialised mental health care institutions, it is likely that a larger proportion of our sample already experienced some psychiatric problems. It is possible that the effects we witnessed are stronger than they would be in the general population, causing problems regarding the generalisability of our results. Finally, this research has focused on suboptimal bonding as an all-encompassing, standalone concept. Within the parental bonding domain, it is common practice to also differentiate
22
between types of bonding (see for example Parker, 1990). It was decided to implement continuous scaling ranging from suboptimal to optimal bonding, as differentiating between types of bonding would be beyond the scope of this research study. It could be valuable to make this distinction in future studies and evaluate whether different effects are witnessed. Clinical implications
This study confirmed the generally assumed relationship between suboptimal parental bonding in childhood, adult anxiety and depression. Moreover, it also confirmed the understudied role of adult attachment in this relationship: insecure adult attachment style partly explains the relationship between suboptimal parental bonding experiences and increased levels of psychopathology. We found that for women in particular, suboptimal maternal bonding experiences affect their anxiety levels in adulthood – suboptimal bonding experiences are thus more detrimental for women, than they are for men. Our study shows that not only conspicuous adverse childhood experiences like physical or sexual abuse affect adult symptomatology (Weich et al., 2009; Gallo et al., 2008), also less obvious types of unfavourable parenting behaviours have a negative effect. These results indicate the importance of early intervention and prevention strategies for suboptimal parenting, especially when such a suboptimal mother-daughter bond is recognised.
Prevention programs for at-risk adolescents are often already in place, for example when one or both of the parents suffers psychiatric problems (De Richtlijn Kinderen van Ouders met Psychische Problemen (KOPP) – translated: Guideline regarding Children of Parents with Mental Illness), or in the case of parental divorce (McClain et al., 2010). However, long-term programs that promote optimal and effective parenting in the general population are uncommonly witnessed (Spoth et al., 2004), even though research has proven that early intervention and prevention strategies in general have long lasting effects (Sandler et al., 2011). Systematically educating parents about the potential consequences of unfavourable and favourable parenting behaviours would thus be advisable. In practise, this could be implemented within a kindergarten or (primary) school setting, or for infants and toddlers at so-called consultation clinics (in Dutch: Consultatiebureaus). Already after a seven-week intervention program, with a total of seven two-hour sessions for families and elementary school students, long term effects are visible (Redmond et al., 2009). Long-term positive treatment effects (such as increased level of parenting competence and a decrease in the
23
child’s disruptive behaviour) are also found in families of 3-year olds after completing a ten-session supported intervention program, or even after a self-directed intervention program (Sanders et al., 2007). It is thus apparent that even low-key, non-invasive prevention programs could have a positive and long-lasting effect, which would be especially beneficial to girls. An additional argument for prevention and intervention strategies can be found in the so-called ‘intergenerational transmission’ of parental bonding (Mannarini et al., 2018). As children are known to adopt similar parenting behaviours and styles as their parents (Serbin & Karp, 2003), implementing above-mentioned prevention and intervention strategies could break the dysfunctional parenting circle and hereby foster intergenerational transmission of effective, positive parenting behaviours (Chen, & Kaplan, 2001).
Elucidated earlier is the hypothesised mechanism between unfavourable, suboptimal parenting behaviours, insecure adult attachment style and psychopathology (Bowly, 1969; Bretherton, 1992; Ravitz et al., 2010). Research has taught us that especially childhood experiences are important in the development of adult attachment styles (Fraley & Roisman, 2019). The above-suggested prevention strategies could thus also indirectly play a role on psychopathology through their influence on the development of secure adult attachment behaviours. Especially because early unfavourable experiences do not have to be deterministic and can be balanced out by multiple other positive bonding experiences (Fraley & Roisman, 2019), as promoted by prevention programs.
Although childhood experiences in particular are important for the development of adult attachment styles (Fraley & Roisman, 2019), it is certainly possible to cultivate more secure attachment behaviours in adulthood. As securely attached adults generally have a positive sense of self (Bartholomew & Shaver, 1998), this could potentially be one of the factors in treatment to focus on: increasing one’s self-esteem in order promote more secure attachment behaviours. Studies investigating alterations in adult attachment style have also indicated that a change towards more secure attachment behaviours is related to an increase in self-esteem (Cozzarelli et al., 2003). A change towards more attachment security was positively related to well-being and negatively related to feelings of distress (Cozzarelli et al., 2003). A second factor that is often associated with attachment security in adulthood is perceived social support (Kafetsios & Sideridis, 2006). People who became more secure in their attachment behaviours also showed improvements in their perceptions of social support and well-being
24
(Cozzarelli et al., 2003). Social support is theorised to mirror a general sense of how social relations are evaluated (Sarason, Pierce & Sarason, 1990), or as Bartholomew & Shaver (1998) write: securely attached adults carry a positive model of others. In treatment, intervention strategies supporting a client in developing or expanding meaningful relationships, in order to improve the model they hold of others, could be a point of focus. Third, coping strategies have found to be related to attachment security and well-being as well (Zhang & Labouvie-Vief, 2004). Securely attached individuals are able to adaptively approach problems, for example by evaluating possible solutions or seeking social support, whereas insecure attached individuals tend to withdraw (Mahmoud, Staten, Hall, & Lennie, 2012). Treatment could possibly focus on practicing constructive coping strategies, which would improve secure attachment behaviours. To sum up: insecure attachment styles explain part of the relationship between parental bonding and psychopathology. By applying treatment interventions focusing on fostering secure attachment behaviours, this part of the relationship could be altered and possibly lower levels of experienced psychopathology. Directions for future research
It would be advisable for future studies to make sure to incorporate both subject- and parent sex as part of their study design in order to be able to properly understand the effects. An interesting topic for further research would be to investigate the potential mechanism behind the witnessed interaction effect: what makes this mother-daughter bond unique, such that daughters experience heightened levels of anxiety after suboptimal maternal bonding? A previously mentioned limitation is simultaneously a suggestion for future research: as parental bonding is now measured through retrospective reports, it would be interesting to be able to examine prospective data on the topic. A longitudinal study could assess current parental bonding experiences, and at follow-up points measure both adult attachment behaviours and experienced psychopathology. At follow up points, retrospective evaluation of parental bonding could even be investigated, to make a comparison with parental bonding as experienced at baseline.
As we found that adult attachment behaviours partly explain why suboptimal bonding leads to higher scores on anxiety and depression, it would be advised for future parental bonding studies to also include this concept. Some potential mechanisms behind suboptimal bonding, insecure adult attachment and presence of anxiety and depression have been briefly touched
25
upon. Self-esteem (Cozzarelli et al., 2003), social support (Kafetsios & Sideridis, 2006) and coping (Zhang & Labouvie-Vief, 2004) have in this paper been mentioned as potential factors underlying the process. Research into the exact mechanisms and potential other underlying elements could be of value, also with regards to potential intervention strategies targeting these aspects.
26
References
A. F. Hayes (2017) PROCESS v3.3 macro, retrieved from http://processmacro.
org/download.html.
About NESDA. (2019, June 1st). Retrieved from https://www.nesda.nl/nesda-english/.
Ainsworth, M. S. (1979). Infant–mother attachment. American psychologist, 34(10), 932. Alonso, Y., Fernández, J., Fontanil, Y., Ezama, E., & Gimeno, A. (2018). Contextual determinants of psychopathology. The singularity of attachment as a predictor of mental dysfunction. Psychiatry research, 261, 338-343.
American Psychiatric Association (APA). Diagnostic and Statistical Manual of Mental Disorders 4th Ed. Washington DC, 1994.
Bartholomew, K., & Shaver, P. R. (1998). Methods of assessing adult attachment. Attachment
theory and close relationships, 1998, 25-45.
Beck, A. T., Epstein, N., Brown, G., & Steer, R. A. (1988). An inventory for measuring clinical anxiety: psychometric properties. Journal of consulting and clinical psychology,
56(6), 893.
Bifulco, A., Kwon, J., Jacobs, C., Moran, P. M., Bunn, A., & Beer, N. (2006). Adult attachment style as mediator between childhood neglect/abuse and adult depression and anxiety. Social psychiatry and psychiatric epidemiology, 41(10), 796-805.
Bijl, R. V., de Graaf, R., Hiripi, E., Kessler, R. C., Kohn, R., Offord, D. R., ... & Wittchen, H.U. (2003). The prevalence of treated and untreated mental disorders in five countries.
Health affairs, 22(3), 122-133.
Bögels, S. M., & Brechman-Toussaint, M. L. (2006). Family issues in child anxiety: Attachment, family functioning, parental rearing and beliefs. Clinical psychology review,
26(7), 834-856.
Bögels, S., & Phares, V. (2008). Fathers' role in the etiology, prevention and treatment of child anxiety: A review and new model. Clinical psychology review, 28(4), 539-558.
27
Bowlby, J. (1969), Attachment and loss, Vol. 1: Attachment. New York: Basic Books.
Breinholst, S., Esbjørn, B. H., & Reinholdt-Dunne, M. L. (2015). Effects of attachment and rearing behavior on anxiety in normal developing youth: A mediational study. Personality
and Individual Differences, 81, 155-161.
Brennan, K.; Clark, C.; Shaver, P. (1998). Self-report measures of adult romantic attachment. In J. Simpson and W. Rholes, Attachment Theory and Close Relationships. New York: Guilford Press.
Bretherton, I. (1992). The origins of attachment theory: John Bowlby and Mary Ainsworth.
Developmental psychology, 28(5), 759.
Burbach, D. J., & Borduin, C. M. (1986). Parent-child relations and the etiology of depression: A review of methods and findings. Clinical Psychology Review, 6(2), 133-153. Burns, R. A., Loh, V., Byles, J. E., & Kendig, H. L. (2018). The impact of childhood parental quality on mental health outcomes in older adults. Aging & mental health, 22(6), 819-825. Carter, M. M., Sbrocco, T., Lewis, E. L., & Friedman, E. K. (2001). Parental bonding and anxiety: Differences between African American and European American college students.
Journal of Anxiety Disorders, 15(6), 555-569.
Chen, Z. Y., & Kaplan, H. B. (2001). Intergenerational transmission of constructive parenting. Journal of Marriage and Family, 63(1), 17-31.
Cox, B. J., Enns, M. W., & Clara, I. P. (2000). The Parental Bonding Instrument: confirmatory evidence for a three-factor model in a psychiatric clinical sample and in the National Comorbidity Survey. Social Psychiatry and Psychiatric Epidemiology, 35(8), 353-357.
Cozzarelli, C., Karafa, J. A., Collins, N. L., & Tagler, M. J. (2003). Stability and change in adult attachment styles: associations with personal vulnerabilities, life events. and global construals of self and others. Journal of Social and Clinical Psychology, 22(3), 315-346.
28
de Minzi, M. C. R. (2006). Loneliness and depression in middle and late childhood: The relationship to attachment and parental styles. The Journal of Genetic Psychology, 167(2), 189-210.
del Barrio, V., Holgado-Tello, F. P., & Carrasco, M. A. (2016). Concurrent and longitudinal effects of maternal and paternal warmth on depression symptoms in children and adolescents.
Psychiatry research, 242, 75-81.
Fraley, R. C., & Roisman, G. I. (2019). The development of adult attachment styles: four lessons. Current opinion in psychology, 25, 26-30.
Furman, W., & Buhrmester, D. (1985). Children’s perception of the personal relationships in their social networks. Developmental Psychology, 21, 1016–1024.
Gallo, E. A. G., Munhoz, T. N., de Mola, C. L., & Murray, J. (2018). Gender differences in the effects of childhood maltreatment on adult depression and anxiety: a systematic review and meta-analysis. Child abuse & neglect, 79, 107-114.
Haaga, D. A., Yarmus, M., Hubbard, S., Brody, C., Solomon, A., Kirk, L., & Chamberlain, J. (2002). Mood dependency of self-rated attachment style. Cognitive Therapy and Research,
26(1), 57-71.
Hayes, A. F. (2017). Introduction to mediation, moderation, and conditional process analysis: A regression-based approach. Guilford Publications.
IBM Corp. Released 2019. IBM SPSS Statistics for Windows, Version 26.0. Armonk, NY: IBM Corp.
Jinyao, Y., Xiongzhao, Z., Auerbach, R. P., Gardiner, C. K., Lin, C., Yuping, W., & Shuqiao, Y. (2012). Insecure attachment as a predictor of depressive and anxious symptomology.
Depression and anxiety, 29(9), 789-796.
Jorm, A. F., Dear, K. B. G., Rodgers, B., & Christensen, H. (2003). Interaction between mother's and father's affection as a risk factor for anxiety and depression symptoms. Social
29
Kafetsios, K., & Sideridis, G. D. (2006). Attachment, social support and well-being in young and older adults. Journal of health psychology, 11(6), 863-875.
Kendler, K. S. (1996). Parenting: A genetic-epidemiologic perspective. The American journal
of psychiatry, 153(1), 11.
Kessler, R. C., Angermeyer, M., Anthony, J. C., De Graaf, R. O. N., Demyttenaere, K., Gasquet, I., ... & Kawakami, N. (2007). Lifetime prevalence and age-of-onset distributions of mental disorders in the World Health Organization's World Mental Health Survey Initiative.
World psychiatry, 6(3), 168.
Khalid, A., Qadir, F., Chan, S. W., & Schwannauer, M. (2018). Parental bonding and adolescents’ depressive and anxious symptoms in Pakistan. Journal of affective disorders,
228, 60-67.
KOPP. (2019, june 18th). Retrieved from http://richtlijnenjeugdhulp.nl/kopp/.
La Greca, A. M., & Harrison, H. M. (2005). Adolescent peer relations, friendships, and romantic relationships: Do they predict social anxiety and depression?. Journal of clinical
child and adolescent psychology, 34(1), 49-61.
Lam, L.T., Rai, A. & Lam, M.K. (2019). Attachment problems in childhood and the development of anxiety in adolescents: A systematic review of longitudinal and prospective studies. Mental Health & Prevention, 14, in progress.
Lewinsohn, P. M., & Rosenbaum, M. (1987). Recall of parental behavior by acute depressives, remitted depressives, and nondepressives. Journal of Personality and Social
Psychology, 52(3), 611.
Lopez, F. G., Melendez, M. C., & Rice, K. G. (2000). Parental divorce, parent–child bonds, and adult attachment orientations among college students: A comparison of three racial/ethnic groups. Journal of Counseling Psychology, 47(2), 177.
Mahmoud, J. S. R., Staten, R. T., Hall, L. A., & Lennie, T. A. (2012). The relationship among young adult college students’ depression, anxiety, stress, demographics, life satisfaction, and coping styles. Issues in mental health nursing, 33(3), 149-156.
30
Main, M. & Solomon, J. (1990). Procedures for identifying infants as
disorganized/disoriented during the Ainsworth Strange Situation. In Greenberg, M., Cicchetti, D. & Cummings, E.M. (Eds.), Attachment in the preschool years: theory, research, and
intervention (p. 121-82). Chicago: University of Chicago Press.
Majdandžić, M., de Vente, W., Feinberg, M. E., Aktar, E., & Bögels, S. M. (2012). Bidirectional associations between coparenting relations and family member anxiety: A review and conceptual model. Clinical child and family psychology review, 15(1), 28-42. Mannarini, S., Balottin, L., Palmieri, A., & Carotenuto, F. (2018). Emotion regulation and parental bonding in families of adolescents with internalizing and externalizing symptoms.
Frontiers in psychology, 9, 1-9.
McClain, D. B., Wolchik, S. A., Winslow, E., Tein, J. Y., Sandler, I. N., & Millsap, R. E. (2010). Developmental cascade effects of the New Beginnings Program on adolescent adaptation outcomes. Development and Psychopathology, 22(4), 771-784.
McLeod, B. D., Weisz, J. R., & Wood, J. J. (2007). Examining the association between parenting and childhood depression: A meta-analysis. Clinical psychology review, 27(8), 986-1003.
McShane, K. E., & Hastings, P. D. (2009). The New Friends Vignettes: Measuring parental psychological control that confers risk for anxious adjustment in preschoolers. International
Journal of Behavioral Development, 33(6), 481-495.
Meites, T. M., Ingram, R. E., & Siegle, G. J. (2012). Unique and shared aspects of affective symptomatology: The role of parental bonding in depression and anxiety symptom profiles.
Cognitive therapy and research, 36(3), 173-181.
Möller, E. L., Majdandžić, M., de Vente, W., & Bögels, S. M. (2013). The evolutionary basis of sex differences in parenting and its relationship with child anxiety in Western societies.
31
Möller, E. L., Nikolić, M., Majdandžić, M., & Bögels, S. M. (2016). Associations between maternal and paternal parenting behaviors, anxiety and its precursors in early childhood: A meta-analysis. Clinical Psychology Review, 45, 17-33.
Muris, P., Meesters, C., van Melick, M., & Zwambag, L. (2001). Self-reported attachment style, attachment quality, and symptoms of anxiety and depression in young adolescents.
Personality and Individual Differences, 30(5), 809-818.
Murray, C. J., Lopez, A. D., & World Health Organization. (1996). The global burden of disease: a comprehensive assessment of mortality and disability from diseases, injuries, and risk factors in 1990 and projected to 2020: summary.
Parker, G. (1990). The parental bonding instrument. Social psychiatry and psychiatric
epidemiology, 25(6), 281-282.
Parker, G., Tupling, H. & Brown, L. B. (1979). A parental bonding instrument. British
Journal of Medical Psychology; 52, 1–10.
Pearson, J. L., Cohn, D. A., Cowan, P. A., & Cowan, C. P. (1994). Earned-and continuous-security in adult attachment: Relation to depressive symptomatology and parenting style.
Development and psychopathology, 6(2), 359-373.
Penninx, B. W., Beekman, A. T., Smit, J. H., Zitman, F. G., Nolen, W. A., Spinhoven, P., ... & Van Der Meer, K. (2008). The Netherlands Study of Depression and Anxiety (NESDA): rationale, objectives and methods. International journal of methods in psychiatric research,
17(3), 121-140.
Phares, V., & Renk, K. (1998). Perceptions of parents: A measure of adolescents' feelings about their parents. Journal of Marriage and the Family, 646-659.
Priel, B., & Shamai, D. (1995). Attachment style and perceived social support: Effects on affect regulation. Personality and individual differences, 19(2), 235-241.
Ravitz, P., Maunder, R., Hunter, J., Sthankiya, B., & Lancee, W. (2010). Adult attachment measures: A 25-year review. Journal of psychosomatic research, 69(4), 419-432.
32
Redmond, C., Spoth, R. L., Shin, C., Schainker, L. M., Greenberg, M. T., & Feinberg, M. (2009). Long-term protective factor outcomes of evidence-based interventions implemented by community teams through a community–university partnership. The Journal of Primary
Prevention, 30(5), 513-530.
Renk, K., McKinney, C., Klein, J., & Oliveros, A. (2006). Childhood discipline, perceptions of parents, and current functioning in female college students. Journal of Adolescence, 29(1), 73-88.
Rikhye, K., Tyrka, A. R., Kelly, M. M., Gagne Jr, G. G., Mello, A. F., Mello, M. F., ... & Carpenter, L. L. (2008). Interplay between childhood maltreatment, parental bonding, and gender effects: Impact on quality of life. Child Abuse & Neglect, 32(1), 19-34.
Rush, A. J., Giles, D. E., Schlesser, M. A., Fulton, C. L., Weissenburger, J. E., & Burns, C. T.
(1986). The Inventory of Depressive Symptomatology (IDS): Preliminary findings. Psychiatry Research, 18, 65–87.
Russell, A., & Saebel, J. (1997). Mother–son, mother–daughter, father–son, and father– daughter: Are they distinct relationships?. Developmental Review, 17(2), 111-147.
Salkind, N. J. (Ed.). (2010). Encyclopedia of research design (Vol. 3). Sage.
Sanders, M. R., Bor, W., & Morawska, A. (2007). Maintenance of treatment gains: a comparison of enhanced, standard, and self-directed Triple P-Positive Parenting Program.
Journal of abnormal child psychology, 35(6), 983.
Sandler, I. N., Schoenfelder, E. N., Wolchik, S. A., & MacKinnon, D. P. (2011). Long-term impact of prevention programs to promote effective parenting: Lasting effects but uncertain processes. Annual review of psychology, 62, 299-329.
Sarason, I. G., Pierce, G. R., & Sarason, B. R. (1990). Social support and interactional processes: A triadic hypothesis. Journal of Social and Personal Relationships, 7(4), 495-506. Schafer, J. L., & Graham, J. W. (2002). Missing data: our view of the state of the art.
