The impact of maternal parenting stress and
perceived-received social support on children (aged 3 to 10) of mothers
with mental illness
Research Master Educational Sciences Thesis 2
M. H. Bregman
Dr. H. M. W. Bos, Prof. Dr. J. M. A. Hermanns, Dr. H. R. Rodenburg 01-02-2013
Preface
This study results from working as a research trainee at Context, a division of Parnassia Bavo Groep. The data belong to the larger study of researcher H. J. Wansink, who evaluates a parenting support intervention for parents with mental illness, called SOOPP. The research questions and the design of this study were developed by me, based on the available data and discussions with H. J. Wansink and H. M. W. Bos. At the start of my internship the data collection was not yet finished. I have attended the assessment of a mother and her child during a home visit and the start of the parenting support intervention in this family. I was responsible for the data input of approximately 70 participants and several accompanying decisions. For example, adding various demographic variables to the SPSS file, such as ethnicity, number of children in the family, one parent/two parent family, employed/unemployed and the variable social support. Another responsibility was the categorisation of mental health diagnoses and comorbidity which I inferred from the DSM-IV (American Psychiatric Association, 2000) and the advice from E. Hoencamp, professor of Clinical Psychology at University Leiden and director new business at Parnassia Bavo Groep.
Because I was the first person who analysed these data, discussions and decisions about the inclusion criteria, the consequences of the procedure and the measurement were challenging and took much time. The statistical analysis plan was designed by me. Thanks to the idea of professor E. Hoencamp I included the analysis of interrelations between the variables of interest. The statistical analysis, including data cleaning, was performed autonomously and tentative results were discussed with H. J. Wansink, E. Hoencamp and H. M. W. Bos.
Abstract
Objective: Children of parents with mental illness appear at risk for behavioural problems. In this respect, maternal parenting stress may have a significant impact, whereas social support is assumed to buffer this relation. Therefore, the current study investigated whether perceived-received social support moderates the relation between maternal parenting stress and children’s behavioural problems in children (aged 3 to 10) of mothers with mental illness. Method: Participants were 69 mothers from multi-ethnic backgrounds under treatment for diverse mental illnesses and one of their children who did not receive secondary mental health care. Children’s behavioural problems were measured with the Strengths and Difficulties Questionnaire; maternal parenting stress with the Parenting Daily Hassles; severity of maternal mental illness with the Clinical Global Impression Scale and perceived-received social support from parents, family, friends, neighbours and the nanny with the particularly designed Dutch Assistance and Support Questionnaire. After having investigated relations between the variables of interest and the demographics, and interrelations between the variables of interest, sequential multiple regression was performed to answer the research question. The different sources of social support were analysed separately for mothers’ reports of children’s behavioural problems. Severity of maternal mental illness was included as covariate.
Results: Mothers and teachers gave evidence of elevated children’s behavioural problems, but mothers reported more behavioural problems than teachers did. Maternal parenting stress was only related to mothers’ reports of children’s behavioural problems. When mothers received social support from friends, they reported less maternal parenting stress and less children’s behavioural problems. Perceived-received social support made a significant contribution in predicting mothers’ reports of children’s behavioural problems. However, a buffering effect of perceived-received social support from friends on the relation between maternal parenting stress and children’s behavioural problems was not found.
Discussion: Implications of the findings for targeting parenting support interventions for mothers with mental illness and for further research into this field are discussed.
Children of parents with mental illness, varying from depressive, anxiety, bipolar or eating disorders to personality or psychotic disorders (e.g., schizophrenia) are at risk for behavioural
problems, for example, hyperactivity or emotional symptoms (Beardslee, Versage, & Gladstone, 1998; Hosman, Van Doesum, & Van Santvoort, 2009; Rutter & Quinton, 1984; Siegenthaler, Munder, & Egger, 2012). On the long term these children may either develop similar disorders as their parents or different disorders (Hosman et al., 2009). This is explained by direct and indirect effects of genetic and environmental risk factors, such as parental mental illness, a hampered parent-child interaction and adverse life circumstances. There is evidence that particularly an accumulation of risk factors will be detrimental for children (Appleyard, Egeland, Van Dulmen, & Sroufe, 2005; Rutter, 1979; Sameroff, Seifer, Zax, & Barocas, 1987). For instance, growing up in an unemployed one parent family with a parent with severe mental illness.
Research focuses mainly on the impact of maternal mental illness on the offspring. As mothers with mental illness tend to perceive a high level of parenting stress (Williford, Calkins, & Keane, 2007) it is noteworthy that maternal parenting stress appears a significant environmental risk factor for children’s behavioural problems (Barry, Dunlap, Cotten, Lochman, & Wells, 2005; Benzies, Harrison, Magill-Evans, 2004; Crnic, Gaze, & Hoffman, 2005; McConnell, Breitkreuz, & Savage, 2011; Neece & Baker, 2008; Neece, Green, & Baker, 2012). There are indications for a direct impact of maternal parenting stress on children’s behavioural problems (Crnic et al., 2005; Williford et al., 2007). Besides, the reciprocal parent-child relationship suggests that children’s behavioural problems will elicit or exacerbate maternal parenting stress (Barry et al., 2005; Deater-Deckard, Smith, Ivy, & Petril, 2005; Neece et al., 2012; Solem, Christophersen, & Martinussen, 2011; Williford et al., 2007).
In order to prevent negative outcomes in children of mothers with mental illness, the ecological process model of Belsky (1984) may give insight on which concepts parenting support interventions should be targeted. According to Belsky (1984), parenting is influenced by parental mental health, child behaviour, and environmental stress and support (e.g. family structure, employment/unemployment, social network). Parental mental health is assumed to be most appropriate to alleviate the impact of environmental stress, including parenting stress, on children. It is therefore likely that a higher severity of maternal mental illness results in a higher level of maternal parenting stress and that this level will decrease when symptoms decline (Kahng, Oyserman, Bybee, & Mowbray, 2008). Nevertheless, social support from a parent’s social network is also considered essential, for example, from family, friends,
acquaintances and neighbours (Balaji et al., 2007; Belsky, 1984). Thus, when mothers with mental illness receive social support it may buffer the impact of their parenting stress on children’s behavioural problems.
Social support is distinguished into received and perceived social support. Received social support is defined by support which a person receives during a limited period of time, consisting of emotional and/or instrumental support (Meadows, 2011; Uchino, 2009). It can be viewed as an active coping strategy to adjust to stress. Received social support, at least when measured specifically, seems strongly associated with the support which is actually offered (Cohen, Lakey, Tiell, & Neeley, 2005). Perceived social support means how a person perceives the availability of social support or how satisfied she/he is about received social support (Haber, Cohen, Lucas, & Baltes, 2007; Uchino, 2009). Anyway, received and perceived social support mutually influence on another and received social support is inextricably connected to a person’s perceptions, so that it should be understood as ‘perceived-received’ social support (Barrera, 1986).
That social support from the maternal social network is considered as a protective factor for children was underlined by Barry et al. (2005). They pointed to the possible moderating function of this factor in their study about the impact of maternal parenting stress on children’s behavioural problems (boys aged 9 to 12). Likewise, the longitudinal study of Neece et al. (2012) of children aged 3 to 9, assumed that the strength of the reciprocal relation between maternal parenting stress and children’s behavioural problems would vary for different families when protective factors had been taken into account.
However, Vanderbilt-Adriance and Shaw (2008) conclude that the impact of multiple risk factors on children is difficult to buffer: For children of a high-risk population resilience or positive outcomes, resulting from interacting risk and protective factors, and coping strategies (Rutter, 2006), may be hard to reach. In such a situation probably one protective factor is not enough to lead to resilience (Vanderbilt-Adriance & Shaw, 2008). Moreover, it will be most supportive when protective factors are present on different levels of children’s environment (child, family, community). Despite their direct and indirect effects on children, protective factors on the community level, such as perceived-received social support from the maternal social network, are not yet widely studied. Even though, a better understanding of the influence of perceived-received social support on the relation between maternal parenting
stress and children’s behavioural problems would be helpful to target and implement parenting support interventions.
Therefore, the current study investigated whether perceived-received social support had a buffering effect on the relation between maternal parenting stress and children’s behavioural problems in children of mothers with mental illness. Because of the importance of perceived-received social support our hypothesis was that it would moderate this relation, independent of the extent of maternal parenting stress. We expected that the relation between maternal parenting stress and children’s behavioural problems would become less strong when mothers received social support. To strengthen our results this hypothesis was investigated twice, first by mothers’ reports and subsequently by teachers’ reports. It was not clear whether teachers’ reports would confirm the findings, because teachers are another source of informants (Kerig & Wenar, 2006) and prevalence rates for children’s behavioural problems at school age are largely dependent on the informant (see De Los Reyes & Kazdin, 2005). Besides, when reporting their children’s behavioural problems mothers with mental illness may have a negative bias (Hennigan, O‘Keefe, Noether, Rinehart, & Russell, 2006) which seems dependent on the current severity of their maternal mental illness (Mowbray, Lewandowski, Bybee, & Oyserman, 2005; Ordway, 2011). Together with the assumption that severity of maternal illness influences parenting stress, this was why we controlled for severity of mental illness.
In a nutshell, the current study may clarify whether parenting support interventions for mothers with mental illness should mainly target on alleviating parenting stress or also on increasing social support. These insights will be particularly relevant for clinical practice and will be applicable to further research.
(Insert Figure 1 about here)
Method Procedure
The current study is part of a larger study which evaluates the working mechanisms of the existing intervention Preventive Basic Care Management (PBCM) of the regional mental health care centre Parnassia Bavo Groep in The Hague. PBCM is a voluntary parenting
support intervention for parents under treatment of mental health care to prevent children’s behavioural problems (Wansink, Hosman, & Verdoold, 2010). The aims are to realize adequate parenting (sufficient daily care, support, structure, encouragement of children’s development and a low level of harsh parenting) and a balance between risk and protective factors in the family. PBCM provides case management, it assesses the needs of parents and children and initiates and coordinates professional support.
Participants of the current study are under treatment of ambulatory secondary mental health care centre PsyQ in The Hague. Our findings are based on the first measurement, which took place before the parenting support intervention PBCM started. It examined the impact of maternal parenting stress and social support on children’s behavioural problems. The medical ethical committee of mental health care (METIGG) considered approval of the study unnecessary. Inclusion criteria were: child age between 3 to 10 years; the parent is diagnosed with mental illness; the treatment of the parent takes longer than three months; living in The Hague or Zoetermeer. Because this is a preventive study of children’s behavioural problems, exclusion criteria were: the family did already take part in PBCM; the child itself did actually receive secondary mental health care; the Youth Care Office was involved for this child’s functioning; mental retardation of the child. Supplementary, a family was excluded if a measure of the dependent or independent variable was missing.
The first step of the recruitment was that all patients who met the inclusion criteria were informed about the study and its procedure by a letter, either in Dutch, Turkish or Arabic and/or by their counsellor. These concerned patients of the following departments: Depression; Anxiety; ADHD; Psychotrauma; Personality disorders; Somatic; Relation and Psyche. Secondly, counsellors asked these parents’ permission to be phoned by the primary investigator or a research assistant. Thirdly, eligible parents received a phone call of the primary investigator or a research assistant to invite them for an intake appointment at the mental health care centre. Fourthly, during the intake appointment two other inclusion criteria were checked by the primary investigator or a research assistant, namely whether parents were willing to receive a parenting support intervention and whether the family was exposed to several risk factors (e.g., adverse life circumstances). Only one child of each family who met these criteria could participate into the study. This child was chosen by random assignment. Fifthly, fathers were excluded from the current study. Sixthly, mothers whose
partner was diagnosed with mental illness and under treatment of a mental health care centre were excluded, because this is a specific risk factor.
Mothers signed written informed consent for their own and their child’s participation into the study. As part of the procedure they were explicitly informed about the possibility to withdraw from the study at any moment without consequences for their individual treatment. It was also emphasized that counsellors do not have access to the study documents. Families received a small reimbursement for their participation. One of the questionnaires was assessed immediately after the intake at the mental health care centre by the primary investigator or a research assistant. Subsequently, before the start of the parenting support intervention all families were visited at home to assess the other measures. This home visit took approximately one and a half hour. Additionally, counsellors filled in a questionnaire about the mothers’ diagnosis and its severity. Similarly, depending on the parent’s consent, teachers, day care employees or nursery school teachers were sent a questionnaire to measure children’s behavioural problems. An accompanying letter informed them that the questionnaire was used for research about the social emotional development of children in The Hague and the surrounding area and mentioned that parents had given informed consent. They were requested to send the questionnaire back to the researcher in a return envelope. A Turkish research assistant was available for Turkish mothers who did not master the Dutch language. Mothers did not use the service of the available Moroccan research assistant or the telephone translating service.
Participants
Participants were 69 mothers of multi-ethnic backgrounds. The majority (68.1%) did not have a Dutch ethnicity: 17.4% Turkish, 17.4% Moroccan, 13% Surinam, 10.1% Antillean and 10.1% other (see Table 1). The percentage participants without a Dutch ethnicity was higher than the statistics of the community of The Hague (49%), which counted Surinam, Turkish and Moroccan also as the major groups (Stoeldraijer & Loozen, 2011). Mothers were diagnosed with diverse mental illnesses, mostly depressive disorders (43.5%) and often with a comorbid disorder as well (40.6%). Their participating children were 42 boys and 27 girls. Half of these children was living in a one parent family and more than half of their mothers was unemployed.
There were 106 families recruited. Seven families withdraw after the intake or the assessment of some questionnaires, two of them explicitly expressed being overburdened. Fifteen families were excluded based on the exclusion criteria: two children were too young, (2 years old, whereas a child of nearly 3 was included); one mother was not diagnosed with mental illness; in five families only the father was under treatment of mental healthcare; in seven families both parents were diagnosed with mental illness and under treatment of mental healthcare; for 13 families questionnaires of the variables of interest were missing. Also two other families were excluded: a grandmother answered most questionnaires, because she fulfilled the parenting role due to her daughter’s mental illness; parents answered the questionnaires together.
Measures
Children’s behavioural problems. Children’s behavioural problems were reported by mothers and teachers, day care employees or nursery school teachers on the informant-rated version of the Strengths and Difficulties Questionnaire (SDQ) for children aged 4 to 16 (Goodman, 1997). The SDQ is also used for younger children, indicated by the available Spanish norms for 3 to 4 years olds (see www.sdqinfo.com). Dutch, Turkish, Arabic and English questionnaires were used. The SDQ is a short questionnaire of 25 items rated on a 3-point Likert scale from 0 = not true to 2 = certainly true (except for five items with an opposite order). The instrument consists of five scales: hyperactivity (e.g., ”easily distracted, concentration wanders”); emotional symptoms (e.g., ”many worries, often seems worried”); conduct problems (e.g., ”often lies or cheats”); peer problems (e.g., ”rather solitary, tends to play alone”) and prosocial (e.g., ”kind to younger children”). The sum of the first four scales is the total difficulties score and the sum of the prosocial scale is the prosocial behaviour score. The SDQ is a well-established instrument in studies of child behaviour worldwide, for example, in intervention studies (Vostanis, 2006). Children’s behavioural problems were evaluated with the total difficulties score. The Dutch informant-rated version indicated that this score is reliable for parents of children aged 8 to 16 (α = .81) and teachers of children aged 8 to 12 (α = .88) and valid (α = .74) compared to the Child Behavior Check List (Van Widenfelt, Goedhart, Treffers, & Goodman, 2003). Besides, the review of Stone, Otten, Engels, Vermulst, and Janssens (2010) demonstrated that the total difficulties score is reliable and valid for parents and teachers of children aged 4 to 12 and represented a moderate
inter-rater agreement between these informants (r = .44). In our study the Cronbach’s alpha showed an acceptable reliability for mothers (α = .79) and for teachers (α = .78). Clinical scores were calculated based on the advice of Goodman (1997) for studies with a high-risk population (clinical score parent report: total difficulties score ≥ 14; clinical score teacher report: total difficulties score ≥ 12).
Maternal parenting stress. Maternal parenting stress of the last three months was measured with the Parenting Daily Hassles (PDH) of Crnic and Greenberg (1990), the Dutch version of Groenendaal and Gerrits (1996) and a Turkish questionnaire translated by a professional translator (Öner, 2010). This questionnaire was first translated from Dutch into Turkish, subsequently translated back in Dutch and approved to be appropriate. Completing this short instrument takes approximately 10 minutes. The PDH contains 20 items of everyday parenting events, for example ”the kids resist or struggle with you over bed-time”. The frequency of each event is scored on a 5-point Likert scale (1= never to 5 = very often) as well as the perceived intensity (1= no hassle to 5= big hassle). The sums of these separate scales form the Frequency Scale and the Intensity Scale. We only used the Intensity Scale, because Crnic et al. (2005) underlined that perceptions of parenting stress have a larger influence than its presence. Groenendaal and Gerrits (1996) reported a good reliability of the Intensity Scale (α = .86), which is comparable with the reliability in our study (α = .87). Perceived-received social support. Mothers’ perceived-received social support from the social network outside the family, during the last three months, was investigated with the Dutch Assistance and Support Questionnaire (Vragenlijst Hulp en Ondersteuning, VHO) in an interview format. The VHO was particularly designed for the larger study to conduct a cost effectiveness analysis about received social support from persons and institutions. As our study focused on social support of the mothers direct social network we only included the question whether mothers received social support (instrumental and/or emotional) from: (1) parents or parents in law, briefly parents, (2) family, (3) friends or acquaintances, briefly friends, (4) neighbours and/or (5) the nanny. These various sources of social support were derived from the Vragenlijst opvoedingssteun (Hermanns & Vergeer, 1996) and formed five items (0 = no; 1 = yes). Each item was analyzed separately.
Severity maternal mental illness. Counsellors evaluated the severity of maternal mental illness by the severity scale of the Clinical Global Impression Scale (CGI-S) of Guy (1976). This 7-point Likert scale contains the following values: 1 = normal, not at all ill; 2 = borderline
mentally ill; 3 = mildly ill; 4 = moderately ill; 5 = markedly ill; 6 = severely ill; 7 = extremely ill. The CGI is often used in medication research and is known for its good inter reliability agreement and moderate test-retest reliability (Havenaar, Van Os, & Wiersma, 2004).
Demographic variables. Maternal mental health diagnosis, presence of a comorbid disorder, gender of the child, age of the mother and the child, family structure (one parent family/ two parent family), number of children in the family, source of income (employed, including being on sickness benefit/unemployed) and ethnicity (Dutch/Non-Dutch, major ethnical groups) were investigated to describe the participants.
Data analysis
Firstly, we investigated relations between the variables of interest and demographic variables and interrelations between the variables of interest in the total group (N = 69). Because there were only 60 teachers’ reports of children’s behavioural problems available, interrelations between this variable and the other variables of interest were investigated separately. The following tests were used: Pearson’s r, Spearman’s rho, χ2 test, Fisher’s Exact test, one-way ANOVA or Kruskal-Wallis test. As the majority of the mothers did not receive social support from the nanny, it was excluded from the analysis.
Sequential multiple regression was performed to answer the research question: Is there a moderating effect of perceived-received social support on the relation between maternal parenting stress and behavioural problems in children of mothers with mental illness. Severity of maternal mental illness was included as covariate. The methods of Tabachnick and Fidell (2007) for sequential multiple regression and the methods of Baron and Kenny (1986) and Holmbeck (2002) for testing moderational effects were performed. Severity of maternal mental illness was entered in the first step, maternal parenting stress in the second, perceived-received social support in the third and the interaction between maternal parenting stress and received social support in the fourth step. The different sources of perceived-received social support were analysed separately, only for mothers’ reports of children’s behavioural problems, because teachers’ reports turned out not to be related with maternal parenting stress. The groups who perceived that they received or did not receive social support from parents, family or friends consisted of equal numbers.
Because interactions and first-order effects in regression equations are interpreted under the condition that the other variables in the model are 0, meaningful 0-points are
advised (West, Aiken, & Krull, 1996; Whisman & McClelland, 2005). We used dummy codes for the dichotomous perceived-received social support variables. Our expectation that the relation between maternal parenting stress and behavioural problems would be less strong in children of mothers who received social support, resulted in treating this group as the base group. Consequently, presence perceived-received social support was recoded as 0 and absence perceived-received social support as 1. The continuous covariate severity of maternal mental illness and the predictor maternal parenting stress were centered in advance, so that 0 means the average score and multicollinearity was prevented (Holmbeck, 2002; West et al., 1996).
Results Descriptive statistics
Table 1 shows that the total group and the group with teachers’ reports of children’s behavioural problems were comparable on the demographics and the variables of interest. The average severity of maternal mental illness indicates that counsellors evaluated the participating mothers between moderately and markedly ill. Approximately half of these mothers perceived that they received social support from their parents, family and friends and nearly a fourth of them answered that they received social support from their neighbours. (Insert Table 1 about here)
In the total group only several demographics were related to the variables of interest. Maternal mental illness was indeed more severe in mothers with a comorbid disorder (M = 4.93; SD = 0.86) than in mothers without a comorbid disorder (M = 4.29; SD = 0.84; t = -3.05, p < .001). Besides, mothers with different diagnoses differed in their severity of maternal mental illness (χ2(7) = 14.65, p = .04).
Children’s age was positively related to mothers’ reports of children’s behavioural problems (r = .20, p = .048). That the presence of behavioural problems would increase by age was in line with our expectation, because the SDQ was originally developed for children aged 4 and over and there were ten younger children in our study. For this reason, we
expected that children’s age was also related to maternal parenting stress. It was indeed revealed that mothers reported more parenting stress for older children (r = .21, p = .04).
Conversely, mothers’ age made a difference in receiving social support. Mothers who received social support from their parents were younger (M = 33.1; SD = 5.6) than mothers who did not receive this support (M = 38.0; SD = 6.5; t = 3.34, p = .001). Similarly, mothers who received social support from their family (M = 33.7; SD = 6.7) were younger than mothers who did not receive it (M = 37.0; SD = 6.0; t = 2.16, p = .03). As a result of the relation between the age of mothers and children (r = .21, p = .04), mothers of younger children (M = 5.4; SD = 2.0) received more social support from their parents than mothers of older children (M = 6.2; SD = 1.9; t = 1.73, p = .045).
In short, whereas mothers of older children may especially need social support, due to children’s behavioural problems and their parenting stress, they less often received it from their parents than mothers of younger children.
Interrelations between variables of interest
Additionally, as illustrated in Table 2, interrelations between the variables of interest were found.
(Insert Table 2 about here)
Mothers and teachers agreed about children’s behavioural problems, but mothers reported significantly more behavioural problems for their children than teachers did. As expected, mothers with higher levels of parenting stress reported more behavioural problems for their children. Nonetheless, if mothers perceived higher levels of parenting stress teachers reported less behavioural problems for their children, although this was a nonsignificant outcome. (Insert Table 3 about here)
Table 3 depicts that perceived-received social support from friends was the only source of social support which seemed to contribute to less parenting stress. Even though, perceived-received social support from the family made a nearly significant difference in perceiving less parenting stress. In addition, mothers who received social support from their
friends suggested that their children had less behavioural problems than children of mothers who did not receive this source of social support.
(Insert Figure 2 about here)
Teachers also considered that children of mothers who did not receive social support from friends displayed more behavioural problems than children of mothers with this social support, albeit this difference was not significant. In contrast, as presented in Table 3, teachers made clear that children of mothers who received social support from the family showed more behavioural problems than children of mothers who did not receive this social support.
To summarize, mothers and teachers agreed about children’s behavioural problems, although mothers reported more behavioural problems for their children than teachers did. Apart from that, we found a positive relation between maternal parenting stress and children’s behavioural problems in mothers’ reports, but not in teachers’ reports. Mothers also reported that perceived-received social support from friends was related to less parenting stress and less children’s behavioural problems.
Buffering effect of perceived-received social support
Perceived-received social support from friends contributed to the prediction of mothers’ reports of children’s behavioural problems: R became significantly different from zero in step 3 (F = 3.32, p = .03, see Table 4).
(Insert Table 4 about here)
Beyond this, the model of step 4 with the interaction of perceived-received social support from friends and maternal parenting stress was also significant. However, compared to the model in step 3 its change of R2 was nonsignificant (ΔF = 0.67, p = .42). Thus, perceived-received social support from friends did not have a moderating effect on the relation between maternal parenting stress and mothers’ reports of children’s behavioural problems, after controlling for the severity of maternal mental illness, the predictor and the moderator. Because of this, the model of step 3 can be considered as the final model.
This final model demonstrates that mothers’ reports of behavioural problems differed significantly between children of mothers who received or did not receive social support from
friends. The regression coefficient (B = 3.49, p = .028) indicates that behavioural problems increase for children whose mothers did not receive social support from friends compared to children whose mothers received this social support, under the condition of an average severity of maternal mental illness and maternal parenting stress. The adjusted R2of this final model accounted for 9.3% of the variance in children’s behavioural problems. A post-hoc F-test for multiple regression showed that the effect size was .15 and the power .75, while using an alpha level of .05.
For the reason that the standard errors of perceived-received social support from friends were considerable high we checked whether this was due to unequal variance within the groups on the dependent variable mothers’ reports of children’s behavioural problems. Then it became clear that although there was homogeneity of variance between the groups, the variance within the groups was much larger than the variance between the groups. Despite this, it was not problematic for the outcomes: The group children whose mothers received social support from friends still differed significantly on behavioural problems from the group children whose mothers did not receive this social support (F = 8.41, p = .005). The aforementioned results of the t-test had already displayed this (see Table 3).
Apart from perceived-received social support from friends, it was not revealed that the other sources of perceived-received social support predicted mothers’ reports of children’s behavioural problems (see Appendix, Table A1, Table A2, Table A3). To conclude, in our study only perceived-received social support from friends predicted significantly less children’s behavioural problems reported by mothers. In spite of this fact, there was no indication for a buffering effect of perceived-received social support from friends on the relation between maternal parenting stress and children’s behavioural problems.
Discussion
Our multi-informant multi-ethnic preventive study of mothers with mental illness and their children confirmed that children’s behavioural problems were elevated. Both mothers and teachers gave evidence of a substantial percentage of children with clinical scores, based on cut-off points for high-risk populations (Goodman, 1997). In line with that, mothers reported certainly more behavioural problems for their children than parents of children (aged 8 to 10)
of the Dutch community population (Van Widenfelt et al., 2003). Teachers’ reports of children’s behavioural problems in our study were also slightly higher than in that study. Moreover, mothers’ reports approximated the reports of mentally ill parents for children aged 4 to 8 or 9 to 12 (Cowling, Luk, Mileshkin, & Birleson, 2004). The fact that our participating mothers reported less behavioural problems than these parents is probably because we excluded children who received secondary mental health care, whereas Cowling et al. (2004) included these children.
Children in our study can be viewed as a high-risk population, because of their mothers’ mental illness and adverse life circumstances, such as growing up in a one parent family or having an unemployed mother. However, across these individual environmental risk factors even severity of maternal mental illness was not associated with children’s behavioural problems. Likewise Wiegand-Grefe, Geers, Petermann, and Plass (2011) did not determine severity of parental mental illness a significant risk factor for children’s behavioural problems reported by parents, while using the same instrument as we did. Their arguments could count for our study as well. One of their arguments was that the influence of severity of parental mental illness on children’s behavioural problems will possibly only be found within a specific parental diagnosis. Another argument was that the influence of severity of parental mental illness on the offspring could not be derived from the Clinical Global Impression Scale, because it reflects the severity of that given moment in time, not a longer period.
Nonetheless, the environmental risk factor maternal parenting stress was related to mothers’ reports of children’s behavioural problems. Furthermore, its level was clearly higher than in mothers of five years old children studied by the designers of the instrument (Crnic & Greenberg, 1990) or in mothers of the Dutch community population of children aged 0 to 11 (Dekovic, Groenendaal, Noom, & Gerrits, 1996; Groenendaal & Gerrits, 1996). These two studies actually came up with comparable levels of maternal parenting stress. Albeit, it is not obvious if the level of maternal parenting stress in our population is a consequence of their mental illness or of other factors. This level is, for example, comparable to the level of mothers of five years old children with intellectual disability (Gerstein, Crnic, Blacher, & Baker, 2009). Moreover, our population differed from the population of these three studies on several demographics. For instance, children’s age was not the same. The study of the Dutch community population (Dekovic et al., 1996; Groenendaal & Gerrits, 1996) used a broader
age range than our study and in the study of Crnic and Greenberg (1990) or Gerstein et al. (2009) only five years old children participated. This is important to realize as children’s age plays a conceivable role in the level of parenting stress (Crnic et al., 2005; Kahng et al., 2008). We found that mothers of older children perceived more parenting stress, a finding which Kahng et al. (2008) also mentioned in their study about mothers with mental illness of children aged 4 to 16 years. In addition, although maternal parenting stress was quite stable in a study of mothers from a suburban community population, its level increased when children grew from 3 to 5 (Crnic et al., 2005). Besides, a striking difference between our study and Groenendaal and Gerrits (1996) or Crnic and Greenberg (1990) was that the large majority of their participants came from two parent families. Their studies did also not consist of a multi-ethnic population, but of merely ”White” families or only of families with a Dutch multi-ethnicity (Crnic & Greenberg, 1990; Dekovic et al., 1996; Vergeer & Hermanns, 1996).
It is remarkable that there was only a relation between maternal parenting stress and mothers’ reports of children’s behavioural problems. Contrastingly, another study of a high-risk multi-ethnic population, of children of an average age of four years, revealed a direct relation between parenting stress and teachers’ reports of children’s internalizing or externalizing behavioural problems (Anthony et al., 2005). One of their explanations was that adverse life circumstances may exacerbate both parenting stress and behavioural problems. Interestingly, Barry et al. (2005) in their study about boys (aged 9 to 12) who attended an intervention project to treat aggression also referred to adverse life circumstances to explain their outcomes. Despite that, maternal parenting stress was only related to mothers’ reports of children’s behavioural problems and not to teachers’ reports, like in our study. The reason why our results did not point to the influence of adverse life circumstances may have been that unmeasured environmental risk factors, such as income or maternal education, are responsible for that. Another reason may be that there was not much variance in our high-risk population. However, it can be viewed as a positive outcome that there was no relation between these mothers’ maternal parenting stress and teachers’ reports of children’s behavioural problems. This seems especially true because positive outcomes may be hard to reach for children of a high-risk population (Vanderbilt-Adriance & Shaw, 2008).
With this respect it is noteworthy that there was a weak agreement between mothers’ and teachers’ reports of children’s behavioural problems and that mothers reported more behavioural problems than teachers did. In general, research, including studies of multi-ethnic
and clinical populations, did not prove more than a modest agreement between parents’ and teachers’ reports (Achenbach, McConaughy, & Howell, 1987; De Los Reyes & Kazdin, 2005; Stone et al., 2010). Parents usually report more behavioural problems for their children than teachers do (De Los Reyes & Kazdin, 2005; Van der Ende, Verhulst, & Tiemeier, 2012). This disagreement is explained by the context and the perceptions of the informants (Berg-Nielsen, Solheim, Belsky, & Wichstrom, 2012; Van der Ende et al., 2012). Different settings (at home or in school) influence which behaviour is observed and how it is evaluated. Regarding perceptions, research often refers to ‘the distortion-hypothesis’ for parents with mental illness, particularly for mothers with depression and/or anxiety disorders (Berg-Nielsen et al., 2012; De Los Reyes & Kazdin, 2005; Hennigan et al., 2006), namely a tendency to overreport children’s behavioural problems due to negative maternal perceptions. Because of the large percentage of participants with a depressive disorder this explanation has to be taken into account. Moreover, it is strengthened by indications that parents with a higher stress level evaluate their children’s behaviour more often as problematic (Deater-Deckard et al., 2005; De Los Reyes & Kazdin, 2005).
Furthermore, a valuable approach to explain disagreement between parents’ and teachers’ reports of children’s behavioural problems was addressed by De Los Reyes and Kazdin (2005) with their Attribution Bias Context (ABC) Model. An important element of this model is the way memory processes work when informants assess a child’s behaviour. Informants easily retrieve global information about the child, particularly when general questionnaires are answered. Owing to the fact that all informants are prone to bias, the emphasis of this model is not on who is the reliable and valid informant, but on whether and why informants perceive a child’s behaviour as problematic in a specific context. For example, when the teacher-child relation was conflicted teachers indicated more behavioural problems than parents did (Berg-Nielsen et al., 2012). Supplementary, De Los Reyes and Kazdin (2005) argued that the goal of the behavioural assessment is of influence. If an informant is convinced that a child needs treatment she/he will more likely report elevated behavioural problems. It may be well that this partly clarifies why the participating mothers reported more behavioural problems for their children than teachers did. Mothers were already willing to receive the parenting support intervention PBCM, whereas teachers were not informed about this aspect of the study or that only families with a mother with mental illness were studied.
The significance of our results is that perceived-received social support from friends turned out essential for mothers with mental illness. Mothers who received this source of social support perceived less parenting stress and reported less behavioural problems for their children. Thus, mothers’ social support from friends appears to be related to better circumstances for children. Mothers’ social support from friends offers children various opportunities, namely to receive social support from these network members and to learn social behaviour (McConnell et al., 2011; Riley et al., 2008). These are valuable aspects, since modelling social behaviour may be impaired in parents with a high stress level (Neece & Baker, 2008). In addition to this, Nasser and Overholser (2005) estimated that social support from friends was related to less severe depressive symptoms three months later in persons under treatment for a depressive disorder. Unfortunately, only half of the mothers in our study perceived that they received social support from friends. The amount of social support they received may yet have been larger, hence mothers with a depressive disorder appear to perceive received social support differently (Nasser & Overholser, 2005). Nevertheless, mothers who lacked social support from friends reported on average nearly clinical behavioural problems for their children.
There could be several reasons why social support from parents, family or neighbours was not significantly related to positive outcomes. Firstly, only a fourth of the participants received social support from neighbours. Secondly, social support which a person receives may not always be adequately attuned to a person’s specific needs or perceived as supportive (Uchino, 2009). This is especially the case in relationships marked by conflict or ambivalence (Holt-Lunstad, Uchino, Smith, & Hicks, 2007). Therefore, findings about received social support and health outcomes are often inconsistent. Thirdly, these inconsistent findings may be associated with the specific source of social support. For instance, Krause and Rook (2003) suggested that conflicts with family are more complicated than with friends, because these relationships are less voluntary. Apart from that, we assume that social support from friends may require more active coping strategies. Fourthly, that perceived-received social support from the family was related to increased behavioural problems in teachers’ reports will warrant another interpretation than that these social network members added to the burden of the family. Logically, the family like teachers, observed behavioural problems in these children and as a matter of course supported these mothers like the family also supported
young mothers. This argument is plausible, given the nearly significant negative relation between perceived-received social support from the family and maternal parenting stress.
Despite the hopeful expectations of social support, we could not yet demonstrate a buffering effect on the relation between maternal parenting stress and mothers’ reports of children’s behavioural problems. The moderation model with the moderator perceived-received social support from friends was significant, but did not improve the prediction of children’s behavioural problems compared to the model without the interaction term. This former model explained already more than nine percentage of the variance and had a medium effect size (Cohen, 1988). Various methodological limitations need to be considered to explain why the buffering effect was not found. The first point is that there was only a weak relation between maternal parenting stress and mothers’ reports of children’s behavioural problems. It disappeared when the severity of maternal mental illness was included in the model. Together with the aforementioned finding that maternal parenting stress was not related to teachers’ reports of children’s behavioural problems this leads to the question whether maternal parenting stress in itself might be a really good predictor of children’s behavioural problems. Because parenting stress seems inextricably related to parenting and evidence shows that parenting stress mediates the effect of parental depression on parenting future research should also include parenting into the model (Abidin, 1992; Rodenburg, Meijer, Dekovic, & Aldenkamp, 2007). The following point is that each source of perceived-received social support was assessed with only one question. The fact that we did not use a reliable social support instrument reduced the power of the multiple regression analysis while a good power is particularly important in detecting interaction effects (Whisman & McClelland, 2005). However, the assessment of an extended social support instrument could have overburdened our population. Lastly, the explained variance of the final model emphasized that more factors have to be studied to add to our understanding about behavioural problems in children of mothers with mental illness.
Our study did not infer a significant influence of ethnicity, neither for Dutch/Non-Dutch nor for the major ethnical groups, but these were rather small groups. The fact that we studied a multi-ethnic population could indeed have influenced our results. Perceived parenting stress and perceived-received social support are assumed to be connected to a person her/his cultural background (Berger Cardoso, Padilla, & Sampson, 2010; Lazarus & Folkman, 1984). This concerns values which are considered worthwhile in a certain
environment, their impact on the level of parenting stress and the use of social support. It is also true that parenting stress may increase when there is not a good fit between a person and her/his broader environment. Nevertheless, the influence of ethnicity is a complex issue. Berger Cardoso et al. (2010) illustrated this when applying the model of Belsky (1984) to mothers of diverse ethnical backgrounds. They concluded that various factors did not have an equal effect on these mothers’ parenting stress levels. For example, only for the Mexican American mothers depression was not related to a higher level of parenting stress. Mixed findings about ethnicity and parents’ or teachers’ reports of children’s behavioural problems stress that ethnicity is interconnected with multiple factors (Stevens et al., 2003; Vollebergh et al., 2005; Zwirs et al., 2011; Zwirs et al., 2007). Due to migration parents may perceive more parenting stress and therefore report more internalising behavioural problems for their children (Vollebergh et al., 2005). Alternatively, parents with a specific ethnical background may less likely report children’s behavioural problems (Stevens et al., 2003). Further, teachers’ perception of migrant children may influence their evaluation (Vollebergh et al., 2005; Zwirs et al., 2011) and a low social economical status may have more influence than ethnicity (Zwirs et al., 2007).
In conclusion, to prevent negative outcomes in children of mothers with mental illness early intervention is warranted. The behaviour of young schoolchildren has still a good chance to change and larger problems and costs can be prevented (Efstratopoulou, Janssen, & Simons, 2012). Therefore, we advise counsellors of mothers with mental illness to pay attention to their parenting stress and social support and to provide appropriate parenting support interventions for high-risk families which decrease parenting stress and improve social support (Johnson Silver, Heneghan, Bauman, & Stein, 2006). Our results emphasize that these parenting support interventions should especially be targeted on improving social support from friends. Group interventions could be a good option to activate coping strategies to receive social support, to share struggles with mothers in the same kind of situation and to provide cultural sensitive parenting support (Boyd, Diamond, & Bourjolly, 2006; Riley et al., 2008). Parenting support interventions should also directly alleviate parenting stress to have a positive impact on a mother’s perception of her child’s behaviour and decrease children’s behavioural problems. The improvement of adverse life circumstances should indeed also be one of the aims of parenting support interventions.
Finally, to attune interventions adequately further research should go beyond the dominant discourse of risk and protective factors and focus on the varying experiences of mothers with mental illness and their children (McConnell Gladstone, Boydell, & McKeever, 2006). Qualitative studies highlight that mothers with mental illness are frequently confronted with the idea in society of ‘not being a good enough mother’ when being diagnosed with mental illness (Blegen, Hummelvoll, & Severinsson, 2010; Davies & Allen, 2007). On the one hand, as a consequence these mothers may be highly concerned about their children’s mental health (Blegen et al., 2010). On the other hand, this stigma may cause an obstacle to seek parenting support interventions (Blegen et al., 2010; Corrigan, 2004; Davies & Allen, 2007) and children may keep silent about their home situation (Blegen et al., 2010; McConell et al., 2006). The problematic juxtaposition of a mother with mental illness becomes even clearer by the statement of Corrigan (2004) that stigmatization often leads to less self-efficacy. This means that a mother with mental illness will be less convinced to be competent to reach her goals in a certain context (e.g., parenting) (Bandura, 1977). Further research about the impact of maternal parenting stress and social support on children of mothers with mental illness should therefore include mothers’ self-efficacy. This research should also explicitly assess perceived social support, because received social support may decrease self-efficacy when its timing or quantity does not match with a person’s needs (Martire & Schulz, 2007; Uchino, 2009). While parenting should be included in these studies, a careful consideration about how to measure it is necessary as its relation with parenting stress and children’s behavioural problems is not yet totally clear (Crnic et al., 2005; Deater-Deckard et al., 2005; Williford et al., 2007).
In sum, to receive more insights into the working mechanisms of parenting stress and social support, families with maternal mental illness should in-depthly be studied over a longer period of time with a combination of quantitative and qualitative methods.
Acknowledgements
We wish to thank the Netherlands Organisation for Health Research and Development (ZonMw), which financially supported this study and Parnassia Bavo Groep for the opportunity to work as a research trainee under supervision of researcher H. J. Wansink. We
are also thankful for the advice of H. M. W. Bos, J. M. A. Hermanns and H. R. Rodenburg of the University of Amsterdam.
References
Abidin, R. R. (1992). The determinants of parenting behaviour. Journal of Clinical Child Psychology, 21(4), 407-412.
Achenbach, T. M., McConaughy, S. H., & Howell, C. T. (1987). Child/adolescent behavioural and emotional problems: Implications of cross-informant correlations for situational specificity. Psychological Bulletin, 101(2), 213-232.
American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: Author.
Anthony, L. G., Anthony, B. J., Glanville, D. N., Naiman, D. Q., Waanders, C., & Shaffer, S. (2005). The relationships between parenting stress, parenting behaviour and preschoolers’ social competence and behaviour problems in the classroom. Infant and Child
Development, 14(2), 133-154.
Appleyard, K., Egeland, B., Van Dulmen, M. H. M., & Sroufe, L. A. (2005). When more is not better: The role of cumulative risk in child behavior outcomes. Journal of Child Psychology and Psychiatry, 46(3), 235-245.
Balaji, A. B., Claussen, A. H., Smith, D. C., Visser, S. N., Johnson Morales, M., & Perou, R. (2007). Social support networks and maternal mental health and well-being. Journal of Women’s Health, 16(10), 1386-1396.
Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioural change. Psychological Review, 84(2), 191-215.
Baron, R. M., & Kenny, D. A. (1986).The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51(6), 1173-1182.
Barrera, M. (1986). Distinctions between social support concepts, measures, and models. American Journal of Community Psychology, 14(4), 413-445.
Barry, T. D., Dunlap, S. T., Cotten, S. J., Lochman, J. E., & Wells, K. C. (2005). The
influence of maternal stress and distress on disruptive behavior problems in boys. Journal of the American Academy of Child & Adolescent Psychiatry, 44(3), 265-273.
Beardslee, W. R., Versage, E. M., & Gladstone, T. R. G. (1998). Children of affectively ill parents: A review of the past 10 years. Journal of the American Academy of Child & Adolescent Psychiatry, 37(11), 1134-1141.
Belsky, J. (1984). The determinants of parenting: A process model. Child Development, 55(1), 83-96.
Benzies, K. M., Harrison, M. J., & Magill-Evans, J. (2004). Parenting stress, marital quality, and child behavior problems at age 7 years. Public Health Nursing, 21(2), 111-121. Berger Cardoso, J., Padilla, Y. C., & Sampson, M. (2010). Racial and ethnic variation in the
predictors of maternal parenting stress. Journal of Social Service Research, 36(5), 429-444.
Berg-Nielsen, T. S., Solheim, E., Belsky, J., & Wichstrom, L. (2012). Preschoolers’ psychosocial problems: In the eyes of the beholder? Adding teacher characteristics as determinants of discrepant parent-teacher reports. Child Psychiatry & Human
Development, 43(3), 393-413.
Blegen, N. E., Hummelvoll, J. K., & Severinsson, E. (2010). Mothers with mental health problems: A systematic review. Nursing & Health Sciences, 12(4), 519-528.
Boyd, R. C., Diamond, G. S., & Bourjolly, J. N. (2006). Developing a family-based
depression prevention program in urban community mental health clinics: A qualitative investigation. Family Process, 45(2), 187-203.
Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Lawrence Erlbaum Associates.
Cohen, J. L., Lakey, B., Tiell, K., & Neeley, L. C. (2005). Recipient-provider agreement on enacted support, perceived support, and provider personality. Psychological Assessment, 17(3), 375-378.
Corrigan, P. (2004). How stigma interferes with mental health care. American psychologist, 59(7), 614-625.
Cowling, V., Luk, E. S. L., Mileshkin, C., & Birleson, P. (2004). Children of adults with severe mental illness: Mental health, help seeking and service use. The Psychiatrist, 28, 43-46.
Crnic, K. A., Gaze, C., & Hoffman, C. (2005). Cumulative parenting stress across the
preschool period: Relations to maternal parenting and child behaviour at age 5. Infant and Child Development, 14(2), 117-132.
Crnic, K. A., & Greenberg, M. T. (1990). Minor parenting stresses with young children. Child Development, 61(5), 1628–1637.
Davies, B., & Allen, D. (2007). Integrating ’mental illness’ and ’motherhood’: The positive use of surveillance by health professionals. A qualitative study. International Journal of Nursing Studies, 44(3), 365-376.
Deater-Deckard, K., Smith, J., Ivy, L., & Petril, S. A. (2005). Differential perceptions of and feelings about sibling children: Implications for research on parenting stress. Infant and Child Development, 14(2), 211-225.
Dekovic, M., Groenendaal, J. H. A., Noom, M. J., & Gerrits, L. A. W. (1996). Theoretisch kader en opzet van het onderzoek. [Theoretical context and design of the research]. In J. Rispens, J. M. A. Hermanns, & W. H. J. Meeus (Eds.), Opvoeden in Nederland (pp. 6-27). Assen: van Gorcum.
De Los Reyes, A., & Kazdin, A. E. (2005). Informant discrepancies in the assessment of childhood psychopathology: A critical review, theoretical framework, and
recommendations for further study. Psychological Bulletin, 131(4), 483-509.
Efstratopoulou, M., Janssen, R., & Simons, J. (2012). Agreement among physical educators, teachers and parents on children’s behaviours: A multitrait-multimethod design approach. Research in Developmental Disabilities, 33(5), 1343-1351.
Gerstein, E. D., Crnic, K. A., Blacher, J., & Baker, B. L. (2009). Resilience and the course of daily parenting stress in families of young children with intellectual disabilities. Journal of Intellectual Disability Research, 53(12), 981-997.
Goodman, R. (1997). The Strengths and Difficulties Questionnaire: A research note. Journal of Child Psychology and Psychiatry, 38(5), 581-586.
Groenendaal, J. H. A., & Gerrits, L. A. W. (1996). Dagelijkse Beslommeringen Lijst (DBL) [Daily Hassles Questionnaire]. Utrecht: University Utrecht.
Guy, W. (1976). Clinical Global Impressions. ECDEU assessment manual for
Haber, M. G., Cohen, J. L., Lucas, T., & Baltes, B. B. (2007). The relationship between self-reported received and perceived social support: A meta-analytic review. American Journal of Community Psychology, 39(1-2), 133-144.
Havenaar, J. M., Van Os, J., & Wiersma, D. (2004). Algemene meetinstrumenten in de psychiatrische praktijk. [General measurement instruments for psychiatric practise]. Tijdschrift voor psychiatrie, 46(10), 647-651.
Hennigan, K. M., O’Keefe, M., Noether, C. D., Rinehart, D. J., & Russell, L. A. (2006). Through a mother’s eyes: Sources of bias when mothers with co-occurring disorders assess their children. The Journal of Behavioral Health Services and Research, 33(1), 87-104. Hermanns, J. M. A., & Vergeer, M. M. (1996). Opvoeding en opvoedingsondersteuning.
[Parenting and parenting support]. In J. Rispens, J. M. A. Hermanns, & W. H. J. Meeus (Eds.), Opvoeden in Nederland (pp. 134-157). Assen: Van Gorcum.
Holmbeck, G. N. (2002). Post-hoc probing of significant moderational and mediational effects in studies of pediatric populations. Journal of Pediatric Psychology, 27(1), 87-96.
Holt-Lunstad, J., Uchino, B. N., Smith, T. W., & Hicks, A. (2007). On the importance of relationship quality: The impact of ambivalence in friendships on cardiovascular functioning. Annals of Behavioral Medicine, 33(3), 278-290.
Hosman, C. M. H., Van Doesum, K. T. M., & Van Santvoort, F. (2009). Prevention
of emotional problems and psychiatric risks in children of parents with a mental illness in the Netherlands: I. The scientific basis to a comprehensive approach. Australian e-Journal for the Advancement in Mental Health, 8(3), 250-263.
Johnson Silver, E., Heneghan, A. M., & Bauman, L. J. (2006). The relationship of depressive symptoms to parenting competence and social support in inner-city mothers of young children. Maternal and Child Health Journal, 10(1), 105-112.
Kahng, S. K., Oyserman, D., Bybee, D., & Mowbray, C. (2008). Mothers with serious mental illness: When symptoms decline does parenting improve? Journal of Family Psychology, 22(1), 162-166.
Kerig, P. K., & Wenar, C. (2006). Developmental psychopathology: From infancy through adolescence (5th ed.). New York: The McGraw-Hill Companies.
Krause, N., & Rook, K. S. (2003). Negative interaction in late life: Issues in the stability and generalizability of conflict across relationships. Journal of Gerontology: Psychological Sciences, 58(2), 88-99.
Lazarus, R. S., & Folkman, S. (1984). Stress, appraisal, and coping. New York: Springer Publishing Company, Inc.
Martire, L. M., & Schulz, R. (2007). Involving family in psychosocial interventions for chronic illness. Current Directions in Psychological Science, 16(2), 90-94.
McConnell, D., Breitkreuz, R., & Savage, A. (2011). From financial hardship to child
difficulties: Main and moderating effects of perceived social support. Child: Care, Health and Development, 37(5), 679-691.
McConnell Gladstone, B., Boydell, K. M., & McKeever, P. (2006). Recasting research into children’s experiences of parental mental illness: Beyond risk and resilience. Social Science & Medicine, 62(10), 2540-2550.
Meadows, S. O. (2011). The association between perceptions of social support and maternal mental health: A cumulative perspective. Journal of Family Issues, 32(2), 181-208.
Mowbray, C. T., Lewandowski, L., Bybee, D., & Oyserman, D. (2005). Relationship between maternal clinical factors and mother-reported child problems. Community Mental Health Journal, 41(6), 687-704.
Nasser, E. H., & Overholser, J. C. (2005). Recovery from major depression: The role of support from family, friends, and spiritual beliefs. Acta Psychiatrica Scandinavica, 111(2), 125-132.
Neece, C., & Baker, B. (2008). Predicting maternal parenting stress in middle childhood: The roles of child intellectual status, behaviour problems and social skills. Journal of
Intellectual Disability Research, 52(12), 1114-1128.
Neece, C. L., Green, S. A., & Baker, B. L. (2012). Parenting stress and child behavior problems: A transactional relationship across time. American Journal on Intellectual and Developmental Disabilities, 117(1), 48-66.
Öner, O. (2010). Dagelijkse beslommeringen. [Daily Hassles]. The Hague: Öner Vertaaldiensten B.V. Den Haag.
Ordway, M. R. (2011). Depressed mothers as informants on child behavior: Methodological issues. Research in Nursing & Health, 34(6), 520-532.
Riley, A. W., Valdez, C. R., Barrueco, S., Mills, C., Beardslee, W., Sandler, I., & Rawal, P. (2008). Development of a family-based program to reduce risk and promote resilience among families affected by maternal depression: Theoretical basis and program description. Clinical Child and Family Psychology Review, 11(1-2), 12-29.
Rodenburg, R., Meijer, A. M., Deković, M., & Aldenkamp, A. P. (2007). Parents of children with enduring epilepsy: Predictors of parenting stress and parenting. Epilepsy & Behavior, 11(2), 197-207.
Rutter, M. (1979). Protective factors in children’s responses to stress and disadvantage. In M. W. Kent & J. E. Rolf (Eds.), Primary prevention of psychopathology, Volume 3: Social competence in children (pp. 49–74). Hanover, NH: University Press of New England. Rutter, M. (2006). Implications of resilience concepts for scientific understanding. Annals of
the New York Academy of Sciences, 1094, 1-12.
Rutter, M., & Quinton, D. (1984). Parental psychiatric disorder: Effects on children. Psychological Medicine, 14(4), 853-880.
Sameroff, A., Seifer, R., Zax, M., & Barocas, R. (1987). Early indicators of developmental risk: Rochester longitudinal study. Schizophrenia Bulletin, 13(3), 383–394.
Siegenthaler, E., Munder, T., & Egger, M. (2012). Effect of preventive interventions in mentally ill parents on the mental health of the offspring: Systematic review and meta-analysis. Journal of the American Academy of Child & Adolescent Psychiatry, 51(1), 8-17. Solem, M-B., Christophersen, K-A., & Martinussen, M. (2011). Predicting parenting stress:
Children’s behavioural problems and parents’ coping. Infant and Child Development, 20(2), 162-180.
Stevens, G. W. J. M., Pels, T., Bengi-Arslan, L., Verhulst, F. C., Vollebergh, W. A. M., & Crijnen, A. A. M. (2003). Parent, teacher and self-reported problem behavior in the Netherlands: Comparing Moroccan immigrant with Dutch and with Turkish immigrant children and adolescents. Social Psychiatry and Psychiatric Epidemiology, 38(10), 576-585.
Stoeldraijer, L., & Loozen, S. (2011). Den Haag telt half miljoen inwoners. [The Hague counts a half million inhabitants]. Retrieved from Central Bureau of Statistics, website: http://www.cbs.nl/nl-NL/menu/themas/bevolking/publicaties/artikelen/archief/2011/2011-3470-wm.htm
Stone, L. L., Otten, R., Engels, R. C. M. E., Vermulst, A. A., & Janssens, J. M. A. M. (2010). Psychometric properties of the parent and teacher versions of the Strengths and Difficulties Questionnaire for 4- to 12-years-olds: A review. Clinical Child and Family Psychology Review, 13(3), 254-274.
Tabachnick, B. G., & Fidell, L. S. (Eds.). (2007). Using Multivariate Statistics (5th ed.). Boston: Pearson Education, Inc.
Uchino, B. N. (2009). Understanding the links between social support and physical health: A life-span perspective with emphasis on the separability of perceived and received support. Perspectives on Psychological Science, 4(3), 236-255.
Unitat d’Epidemiologia i de Diagnòstic en Psicopatologia del Desenvolupament. (2011). Spanish norms for the Strengths & Difficulties Questionnaire (SDQ) for 3 and 4-year-old schoolchildren: Parents’ and teachers’ versions. Retrieved from the Strengths and
Difficulties Questionnaire website: http://www.sdqinfo.com/Spanish_norms_for_3-4_year-old.pdf
Vanderbilt-Adriance, E., & Shaw, D. S. (2008). Conceptualizing and re-evaluating resilience across levels of risk, time, and domains of competence. Clinical Child and Family
Psychology Review, 11(1-2), 30-58.
Van der Ende, J., Verhulst, F. C., & Tiemeier, H. (2012). Agreement of informants on emotional and behavioural problems from childhood to adulthood. Psychological Assessment, 24(2), 293-300.
Van Widenfelt, B. M., Goedhart, A. W., Treffers, P. D. A., & Goodman, R. (2003). Dutch version of the Strengths and Difficulties Questionnaire (SDQ). European Child & Adolescent Psychiatry, 12(6), 281-289.
Vergeer, M. M., & Hermanns, J. M. A. (1996). De dagelijkse omgeving waarbinnen kinderen en jeugdigen opgevoed worden. [The everyday environment in which children and youth are educated]. In J. Rispens, J. M. A. Hermanns, & W. H. J. Meeus (Eds.), Opvoeden in
Nederland (pp. 28-40). Assen: van Gorcum.
Vollebergh, W. A. M, Ten Have, M., Dekovic, M., Oosterwegel, A., Pels, T., Veenstra, R.,... Verhulst, F. (2005). Mental health in immigrant children in the Netherlands. Social Psychiatry and Psychiatric Epidemiology, 40(6), 489-496.
Vostanis, P. (2006). Strengths and Difficulties Questionnaire: Research and clinical applications. Current Opinion in Psychiatry, 19(4), 367-372.
Wansink, H. J., Hosman, C. M. H., & Verdoold, C. J. (2010). Basiszorg, een handleiding: Preventieve zorgcoördinatie voor ouders met psychiatrische problemen. [Basic Care Management manual: Preventive basic care management for parents with psychiatric problems]. The Hague: Parnassia Bavo Groep, ZorgService BV, afdeling Preventie.
West, S. G., Aiken, L. S., & Krull, J. L. (1996). Experimental personality designs: Analyzing categorical by continuous variable interactions. Journal of Personality, 64(1), 1-48. Whisman, M. A., & McClelland, G. H. (2005). Designing, testing, and interpreting
interactions and moderator effects in family research. Journal of Family Psychology, 19(1), 111-120.
Wiegand-Grefe, S., Geers, P., Petermann, F., & Plass, A. (2011). Kinder psychisch kranker Eltern: Merkmale elterlicher psychiatrischer Erkrankung und Gesundheit der Kinder aus Elternsicht [Children of mentally ill parents: The impact of parental psychiatric diagnosis, comorbidity, severity and chronicity on the well-being of children]. Fortschritte der Neurologie Psychiatrie, 79(1), 32-40.
Williford, A. P., Calkins, S. D., & Keane, S. P. (2007). Predicting change in parenting stress across early childhood: Child and maternal factors. Journal of Abnormal Child
Psychology, 35(2), 251-263.
Zwirs, B., Burger, H., Schulpen, T., Vermulst, A. A., HiraSing, R. A., & Buitelaar, J. (2011). Teacher ratings of children’s behavior problems and functional impairment across gender and ethnicity: Construct equivalence of the Strengths and Difficulties Questionnaire. Journal of Cross-Cultural Psychology, 42(3), 466-481.
Zwirs, B. W. C., Burger, H., Schulpen, T. W. J., Wiznitzer, M., Fedder, H., & Buitelaar, J. K. (2007). Prevalence of psychiatric disorders among children of different ethnic origin. Journal of Abnormal Child Psychology, 35(4), 556-566.
Figure 1. Expected moderation model of the effect of mothers’ perceived-received social support on the relation between maternal parenting stress and children’s behavioural problems, after controlling for the severity of maternal mental illness.
Table 1
Variable M (SD) Range n % M (SD) Range n %
Age mothers 35.42 (6.50) 24-52 35.53 (6.56) 24-52 Age children 5.75 (2.00) 2-10 5.70 (1.93) 2-10 Gender Boys 42 60.9 35 58.3 Girls 27 39.1 25 41.7 Family structure
One parent family 36 52.2 33 55.0
Two parent family 33 47.8 27 45.0
Number of children in the family 2.16 (1.01) 1-5 2.07 (0.96) 1-5
Source of income Unemployed 41 59.4 34 56.7 Employed 28 40.6 26 43.3 Ethnical background Dutch 22 31.9 20 33.3 Turkish 12 17.4 10 16.7 M oroccan 12 17.4 10 16.7 Surinam 9 13.0 7 11.7 Antillean 7 10.1 7 11.7 Other 7 10.1 6 10.0
M aternal mental illness
Depressive disorder 30 43.5 27 45.0
Bipolar disorder 5 7.2 4 6.7
Posttraumatic stress disorder 12 17.4 9 15.0
Anxiety disorder 8 11.6 6 10.0 Attention disorder 3 4.3 3 5.0 Eating disorder 4 5.8 4 6.7 Somatoform disorder 2 2.9 2 3.3 Personality disorder 5 7.2 5 8.3 Comorbidity 28 40.6 23 38.3
Severity maternal mental illness 4.55 (0.90) 2-6 4.53 (0.89) 2-6
M aternal parenting stress 50.54 (15.06) 21-97 51.52 (15.19) 21-97
Perceived-received social support
Parents 36 52.2 32 53.3
Family 33 47.8 28 46.7
Friends 31 44.9 29 48.3
Neighbours 17 24.6 15 25.0
Nanny 7 10.1 6 10.0
Children's behavioural problems
M others' reports 11.70 (6.27) 0-29 11.82 (5.98) 2-29
Clinical scores (≥ 14) 23 33.3 21 35.0
Teachers' reports 7.50 (5.29) 0-21
Clinical scores (≥ 12) 15 25.0
Note. aN = 68, due to missing values.
Descriptives of the Total Group and the Group with Teachers' Reports of the Strenghts and Difficulties Questionnaire (SDQ) Total group (N = 69) Group with SDQ teachers (N = 60)
