Children and Youth Services Review 51 (2015) 10–17
Contents lists available at ScienceDirect
Children and Youth Services Review journal homepage: www.elsevier.com/locate/childyouth
Exploring mental distress among immigrant mothers participating in parent training Ragnhild Bjørknes a,⁎, Marit Larsen a, Fungisai Gwanzura-Ottemöller a, John Kjøbli b a b
The Department of Health Promotion and Development, University of Bergen, Norway Norwegian Center for Child Behavioral Development, University of Oslo, Norway
a r t i c l e
i n f o
Article history: Received 11 August 2014 Received in revised form 23 December 2014 Accepted 21 January 2015 Available online 30 January 2015 Keywords: Maternal mental distress Immigrant families Parent training Randomized controlled trial Moderators
a b s t r a c t This study of ethnic minority mothers assessed the intervention effects of Parent Management Training — Oregon Model (PMTO) on maternal mental distress. Ninety-six mothers from Somalia and Pakistan and their 3-year-old to 9-year-old children were randomized with respect to enrollment in PMTO or a wait-list condition. Immigrants living in European countries report having significantly more mental distress than natives. Surprisingly, the results in this current study showed that there were low levels of mental distress at enrollment in PMTO among the sample. An analysis of covariance showed that PMTO was not effective in alleviating maternal mental distress in this sample. Ethnicity, family size and the child's age served as moderators on the relationship between enrollment in PMTO or the wait-list condition and maternal mental distress outcomes. None of the subgroup analyses were in favor of the intervention. The results emphasize the importance of research on parent training with immigrant families. © 2015 Published by Elsevier Ltd.
1. Introduction Immigrating to a new country can be a stressful life event, particularly when the immigration occurs out of necessity — either due to the pursuit of a better life or escape from war or persecution. In 2014, immigrants represented 12% of the Norwegian population, and 52% of these immigrants were from Africa, Asia, Oceania and South and Central America (excluding Australia and New Zealand; Statistics Norway, 2014). These are defined as non-western immigrants. Upon their arrival in Norway, most non-Western immigrants find themselves in an environment that features not only an alien climate but also a society and culture that is different from their own. Moreover, many nonWestern immigrants consist of either families or young people coming from cultures in which parenting and child-rearing practices differ from those in Norway (Kritz & Skivenes, 2010). Acclimatizing to their new host-country home can be difficult, as has been illustrated by a study with data from 23 European countries (including Norway) that showed that immigrants living in European countries report having significantly more depressive symptoms than natives (Missinne & Bracke, 2012). In addition, a Norwegian study by Dalgard, Thapa, Hauff, McCubbin, and Syed (2006) reported that 23.6% of immigrants from low-income countries living in Oslo had symptoms of mental distress, such as depression and anxiety. In contrast, approximately 9.8% of those born in Norway and 11.6% of the immigrants from
⁎ Corresponding author at: The Department of Health Promotion and Development, Faculty of Psychology, University of Bergen, Christiesgt. 13, 5020 Bergen, Norway. E-mail address:
[email protected] (R. Bjørknes).
http://dx.doi.org/10.1016/j.childyouth.2015.01.018 0190-7409/© 2015 Published by Elsevier Ltd.
high-income Western countries had similar symptoms (Dalgard et al., 2006). Thus the finding that a relatively large number of non-Western immigrant caregivers suffer from anxiety, depression or stress symptoms might suggest a reduction in the ability of such immigrants to parent positively and effectively. This decline in the inability to provide effective caregiving may in turn result in the children of these immigrants developing conduct problems (Barlow, Coren, & Stewart-Brown, 2002). From a theoretical perspective, this argument is supported by the social interaction learning (SIL) model (Patterson, 1982). The SIL model postulates that parenting practices have a causal effect on child behavior, whereas family situations (e.g., maternal mental distress) to a large degree influence child outcomes indirectly through parenting practices. From this perspective, immigration may be understood as a source of stress and tension that can exhaust parental resources and thereby increase levels of mental distress, disturb parenting practices and increase child conduct problems. Although there is research supporting the view that parenting practices influence child behavior (Hoeve et al., 2009), other studies suggest that the relationship may be reciprocal (Miner & Clarke-Stewart, 2008). Once child conduct problems have emerged, they may negatively influence parenting practices and parental mental health. Since, as suggested above, one of the side effects of immigration may be the development of parental mental distress, which may have knock on effects on parenting practice and child behavior, interventions aimed at reducing conduct problems in children, are highly relevant for these particular groups (Oppedal et al., 2008). Conducting such interventions as well as increasing immigrant parents' understanding of local childrearing practices may reduce both the levels of conduct problems and parental mental distress (Oppedal et al., 2008). Our previous research,
R. Bjørknes et al. / Children and Youth Services Review 51 (2015) 10–17
using the same sample as the one presented in this paper, found that the Parent Management Training — Oregon Model (PMTO) had positive effects on parenting practices and child conduct problems (Bjørknes & Manger, 2013). Therefore, in the current study, we examined whether PMTO influenced the mental distress, defined as symptoms of depression and anxiety, levels of immigrant mothers. 1.1. Parent training, maternal mental distress and the role of moderators Parent training is a collection of well-established evidence-based early intervention models that have been developed to improve parenting skills and to prevent child conduct problems (Weisz & Kazdin, 2010). In Western Europe and North America, parent training is effective in improving parenting practices and reducing child conduct problems in both ethnic majority (Furlong et al., 2012; Michelson, Davenport, Dretzke, Barlow, & Day, 2013) and minority populations (Bjørknes & Manger, 2013; Matos, Bauermeister, & Bernal, 2009; Scott et al., 2010). Parent training has also been found to have a positive effect on maternal mental distress, such as depression and anxiety (Barlow, Coren, & Stewart-Brown, 2002; DeGarmo, Patterson, & Forgatch, 2004; Hutchings, Appleton, Smith, Lane, & Nash, 2002; Kjøbli, Hukkelberg, & Ogden, 2013; Kjøbli & Ogden, 2012). However, only a few studies have examined whether parent training reduces maternal mental distress. One such study of PMTO conducted in the USA (discussed more fully in Subsection 1.2) found that the intervention indirectly reduced maternal distress as a result of more effective parenting and fewer child conduct problems (DeGarmo et al., 2004). Two recent randomized effectiveness trials conducted in Norway revealed small to moderate effects (ESs ranging from .21 to .37) on maternal distress (Kjøbli & Ogden, 2012; Kjøbli et al., 2013). Kjøbli, Nærde, Bjørnebekk, and Askeland (2013) found that low levels of maternal distress combined with high levels of child conduct problems predicted more positive outcomes for child conduct problems in the parent training group than in the control group. However, the children of parents in the parent training group with high maternal distress had more negative outcomes for their conduct problems than in the control group (Kjøbli et al., 2013). With respect to the role of moderators in parent training, two metaanalyses summarized the previous research on the predictors and moderators of child outcomes following parent training intervention (Lundahl, Risser, & Lovejoy, 2006; Reyno & McGrath, 2006). Both found that more disadvantaged families (single parents, low income, and depression) benefited less from parent training than did families with lower levels of adverse family factors. Other factors identified as possible predictors or moderators of parent training outcomes include maternal age, level of child conduct problems, education level, ethnicity, family size (Reyno & McGrath, 2006), and child gender and age (Gardner, Hutchings, Bywater, & Whitaker, 2010). Given this mixed bag of results regarding the effects of parent training on maternal mental distress and the role of moderators on outcome, more knowledge is required about when – and for which groups – parent training has an effect on maternal mental distress. Moreover, studies on parent training have predominantly been conducted with middle-class Caucasian participants, and ethnic minority families are under-researched in this field (Michelson et al., 2013). In addition, therapists working with parent training groups are concerned that clients from ethnic minority populations may not be benefitting from treatment and might thus require extra therapeutic effort (Michelson et al., 2013). To our knowledge, no published studies have examined maternal mental distress in an immigrant sample living in Norway that has participated in parent training. 1.2. Parent Management Training — Oregon Model PMTO's main goal is to teach parents how to promote a child's prosocial and competent behaviors and how to use mild and consistent
11
disciplinary strategies to stop unwanted behavior. During PMTO therapy, families are taught five main parenting skills: monitoring, problem solving, effective discipline, positive involvement, and contingent skill encouragement. Several studies suggest that PMTO strengthens positive parenting practices and decreases observed child conduct problems (Bjørknes & Manger, 2013; Forgatch, Patterson, DeGarmo, & Beldavs, 2009; Forgatch & DeGarmo, 1999; Martinez & Forgatch, 2001; Kjøbli et al., 2013; Ogden & Amlund-Hagen, 2008; Patterson, 1982). Although the focus in PMTO is not on reducing maternal depression or anxiety, studies have documented that PMTO is effective in reducing maternal mental distress (DeGarmo et al., 2004; Kjøbli et al., 2013). As discussed above, a randomized controlled trial was recently conducted to evaluate the effects of PMTO on ethnic minorities in Norway (Bjørknes & Manger, 2013). The findings from this study showed that mothers receiving PMTO significantly improved their parenting practices (i.e., decreased the use of harsh discipline and increased positive parenting) and that their children exhibited fewer conduct problems at home compared with children in the control group. Based on these results and drawing on the prior literature on mental distress and the moderators of outcomes in parent training (Gardner et al., 2010; Lundahl et al., 2006; Reyno & McGrath, 2006), the next step in this study was to investigate whether PMTO affected maternal mental distress in addition to examining potential moderators of the outcome. In the present study, PMTO was offered to Somali and Pakistani families living in Norway. We conducted secondary data analyses to assess the intervention effects and potential moderators on maternal mental distress (see Subsection 1.3). 1.3. Research questions The overarching aim of this study was to investigate maternal mental distress in ethnic minority mothers who participated in PMTO. Thus, we sought answers to the following research questions: 1) What were the levels of maternal mental distress upon enrollment into PMTO? 2) What were the post-test effects that PMTO had on maternal mental distress? 3) Which family, parenting, and child factors moderated the relationship between PMTO and maternal mental distress post-intervention? Based on previous research (Lundahl et al., 2006; Reyno & McGrath, 2006), the moderators explored included the following (a) family factors, including ethnicity, maternal education level, marital status, family size, maternal age and duration of residence in Norway; (b) parenting factors, including harsh discipline and positive parenting; and (c) child factors, including gender, age and level of conduct problems. 2. Methods 2.1. Study and participants This study was financed and conducted at The Norwegian Center for Child Behavioral Development from 2007 to 2011. The study was designed as a randomized controlled trial with pre- and post-testing. Participants were recruited from December 2007 to March 2008 (n = 118) (reported in Bjørknes, Jakobsen, & Nærde, 2011). At initiation, 96 mothers were randomized either into PMTO (n = 50) or into the wait-list condition (WLC, n = 46). Families were randomized by ethnicity (Somali or Pakistani) and within the Norwegian community in which they lived. This procedure of stratified randomization was conducted because the intervention group was divided into ethnic groups and the intervention was community based. In the PMTO group, fortysix mothers completed the intervention (drop-outs: n = 4). The overall interval between initiation and termination assessments was 43.25 weeks (SD = 5.87).
12
R. Bjørknes et al. / Children and Youth Services Review 51 (2015) 10–17
The first step in the current RCT was to conduct a qualitative pilot study that interviewed 24 former PMTO participants. Findings from this study were used to increase the accessibility and acceptability of the intervention (Lillehaug, 2007). For example the findings indicated that the parent groups should not be of mixed gender because of Islamic religious norms that require gender segregation. As a consequence, fathers were not included in the current study. Mothers with severe clinical mental health problems or with children who were mentally handicapped or who had been sexually abused were also excluded from the study. None of those recruited met the exclusion criteria. Table 1 presents the characteristics of the families and children. At initiation, the children's ages ranged from 3-years old to 9-years old (M = 5.90, SD = 1.56). The gender distribution was 63% boys and 37% girls. Mothers participating in the study were immigrants from Pakistan (59%) and Somalia (41%) who had been living in Norway from 2.74 to 31.85 years (M = 11.63, SD = 5.77). Immigrants from Pakistan predominantly came to Norway as laborers or to reunify families, and families from Somalia typically came as refugees or for family reunification. In general, the families in this sample were distressed and disadvantaged compared with the Norwegian norm. Twenty percent of the families in our sample received public financial support, compared with 2.3% at the national level. Nine percent of the women had a partor full-time job. Seven percent of the participants reported having a college or university degree, and 38% reported having finished high school. Approximately 23% of the participants were single mothers (never married, divorced, separated, or widowed). Seventy-four of the mothers that participated had difficulties understanding or speaking Norwegian. Child conduct problems were defined as any rule-breaking behaviors, including aggression, temper tantrums, stealing, noncompliance and truancy. Estimated levels of child conduct problems were based on reports from mothers and teachers. Mothers reported a mean of 99.62 (SD = 37.52, range = 36–224) on the Eyberg Child Behavior Inventory (Eyberg & Pincus, 1999) at baseline, and teachers reported a mean of 7.60 (SD = 8.69, range = 0–37) on the Teacher Report Form (Achenbach & Rescorla, 2001) at baseline. Thirty percent of the children scored within the 90th percentile (clinical range) with respect to child conduct problems based on their mothers' reports and Norwegian norms to identify such child conduct problems (Reedtz et al., 2008). There was no significant correlation between mothers' and teachers' reports (r = .15, n = 0.82, p b .19), which indicates that mothers and teachers co-identified very few children with child conduct problems (Bjørknes & Manger, 2013). Comparisons at baseline (χ2 tests and t tests) between the ethnic groups showed significant differences on the following: single mother (Somali sample n = 20, Pakistani n = 3) χ2(1, 96) = 26.92, p b .00, duration of residence in Norway (Somali sample M = 8.89, SD = 3.61, Pakistani M = 13.76, SD = 6.20) t (96) = − 4.21, p = .00, and mental distress (Somali sample M =
Table 1 Demographic information on baseline (N = 96). Categorical variables
n
%
Ethnicity (Somali) Education (Yes) Marital status (Partner) Child gender (Boy)
39 82 76 60
41.3 71.7 76.5 63.0
Continuous measures
M
SD
Maternal mental distress Harsh discipline Positive parenting Conduct problems Child age Maternal age Number of children in the family Duration of residence in Norway
7.66 30.39 67.24 52.18 5.90 33.71 4.05 11.63
3.12 10.07 10.22 8.94 1.56 5.37 1.80 5.77
6.26, SD = 2.28, Pakistani M = 8.61, SD = 3.27) t (96) = − 4.16, p = .00. There were no differences at baseline on the rest of variables in Table 1. 2.2. Procedures This study was conducted in Oslo, the capital of Norway, in areas in which there were large ethnic minority populations. These communities are situated in different parts of Oslo and are not segregated from the majority population. Participants were recruited through public services, community meetings and networks by recruitment teams. The recruitment procedures are described in greater detail elsewhere (Bjørknes et al., 2011). Extensive preliminary work translating the instruments into Urdu and Somali and training bilingual research staff was undertaken in this study. Instruments were back-translated using the method developed by Brislin (1970). Informed consent forms were translated into Urdu and Somali, and research assistants often interpreted the information verbally and collected the consent forms. Each mother selected one of her children as the target child for the study. All assessment material was available in Norwegian, Urdu and Somali, and mothers were paid $50 (300NOK) for completing each pre- and post-assessment (for a maximum total of $100). The study was approved by the Regional Committee for Medical and Health Research Ethics and the Ombudsman for Privacy in Research, Norwegian Social Science Data Services. 2.3. Intervention The intervention was largely based on the Handbook of PMTO for Somali and Pakistani Mothers (Flock & Pettersen, 2008) and was thus devised especially to implement PMTO among ethnic minority families. This manual builds on the Norwegian Handbook of Parent Management Training (Askeland, Christiansen, & Solholm, 2005) and Parenting Through Change (Forgatch, 1994). All the PMTO parent material was translated into Urdu and Somali, and culturally adapted vignettes were included in the Handbook (Flock & Pettersen, 2008). The program included 18 PMTO group sessions every second week that were 2 h in length with eight to 12 mothers. Cultural adaptations, such as ethnically homogeneous groups and the use of link workers were made to the PMTO. Group sessions were held by Norwegian therapists in Norwegian and were simultaneously translated into Urdu/Somali by the link workers. All the groups had one or two link workers participating as the bilingual assistants. They were all women of Somali or Pakistani origin who had experience working with children or families (e.g. in child care and schools) but did not necessarily have any education in this field. They were all trained in the core PMTO components prior to the intervention. Their training involved a 5-day course held by the intervention leader and therapist. As part of this intervention, link workers were monitored by a PMTO therapist before and after each PMTO session. The main responsibility of the link workers in the sessions was as translators and culturefacilitators to strengthen the relationship between the participants and the therapists. This was done as a result of the findings from the pilotstudy, as mentioned in Subsection 2.1. These contextual adaptations were systematically balanced between adherence to the original PMTO manual and structural adaptations that were made to suit the Pakistani and Somali communities. The manual described tasks for each session e.g., the therapist facilitated the role play in standardized vignettes, followed up on homework and tracked group activities. Monthly self-evaluation meetings were held to make sure therapists followed the guidelines and to support the therapists in their therapeutic work. In addition, all therapists and link workers filled out a standard checklist after sessions 3, 6, 12, and 18 for research purposes. The items were rated on a 4-point Likert scale, with higher scores indicating greater adherence. The alpha coefficient was .88 and the mean score was 3.68 (SD = 0.22). Intervention adherence reported
R. Bjørknes et al. / Children and Youth Services Review 51 (2015) 10–17
on these checklists was thus found to be high (Bjørknes & Manger, 2013). Of the 50 mothers randomly assigned to the intervention condition, all but four (8%) began the program, and the participants took part in an average of 10.75 (SD = 5.67) sessions. 2.4. Measures 2.4.1. Symptom Check List-5 (SCL-5) Maternal mental distress was assessed with the short version of the Hopkins Symptom Check List-5 (SCL-5). This instrument is selfadministered and designed to measure anxiety (2 items) and depression (3 items) over the most recent 14 days (feeling fearful, nervousness or shakiness inside, feeling hopeless about the future, feeling blue and worrying too much about things) (Tambs & Moum, 1993). The items were rated on a 4-point Likert scale. The anxiety and depression scales were highly correlated (Pearson correlation coefficient pre-test = 0.56, p = 0.01; and post-test = 0.55, p = 0.01). This questionnaire is a short form version of the SCL-25, which has been validated in Norway (Tambs & Moum, 1993). The SCL-5 had a correlation of r = 0.92 with the SCL-25 (Tambs & Moum, 1993), and it appeared that the SCL behaved similarly regarding variables such as gender, age, education, marital status and employment (Strand, Dalgard, Tambs, & Rognerud, 2003). For practical reasons, the short version with five items was used in this study. This version has a cut-off point of 2.0 (Strand et al., 2003), and a score above this clinical cut-off predicts a possible mental disorder (depression or anxiety). The Cronbach's alpha for the current sample was .87 for the pre-test and .84 for the post-test. Split into ethnic groups, the pre- and post-test Cronbach's alpha scores were .90 and .82 in the Somali group and .85 and .82 in the Pakistani group, respectively. 2.4.2. Parent Practices Interview (PPI) Parental skills were measured by the Parent Practices Interview (PPI) (Webster-Stratton, 1998), which is a questionnaire for parents of 3-year-old to 8-year-old children. In the present study, two of the scales included in the instrument were used: positive parenting (15 items, e.g., hug, compliment, praise, and reward) and harsh discipline (14 items, e.g., slap, raise voice, hit, and threaten). A 7-point Likert scale was used for all the subscales. The calculated alpha coefficients were as follows: positive parenting values of .69 (pre) and .74 (post), and harsh discipline values of .81 (pre) and .84 (post). 2.4.3. Conduct problems composite The instrument used to measure child conduct problems was the conduct problem composite (CP composite). This instrument consists of two different measures, the Eyberg Child Behavior Inventory (ECBI) intensity scale (Eyberg & Pincus, 1999) and the Parent Daily Report (PDR) (Chamberlain & Reid, 1987), both of which are scored by mothers. The ECBI measures conduct problems in 2-year-old to 16year-old children. The instrument consists of 36 items rated on a 7-point Likert scale. Alpha coefficients for the intensity scale were .93 and .94 pre- and post-test, respectively. The PDR is a conduct problem checklist designed to assess the frequency of conduct problems among young children within the past 24 h. Mothers were contacted by telephone on three consecutive days and asked to answer ‘yes’ or ‘no’ to 34 questions with respect to externalizing behaviors, such as lying, fighting and stealing. The alpha coefficients were .94 at pre-test and .95 at post-intervention assessment. The PDR results were combined with the ECBI intensity scale to create a composite variable to increase statistical power (Bjørknes & Manger, 2013), and include information from more than one measurement. Thus, we created a measure that assessed child behavior over four days to strengthen the concept conduct problem. There were moderate positive correlations between ECBI intensity and the PDR at pre- and post-intervention, r = 0.52, n = 96, p b .01 and r = 0.54, n = 83, p b .01, respectively.
13
2.5. Statistical analysis The present analyses were conducted using data from all 96 families in an intention-to-treat-sample and using the last observation carried forward for those who were lost in follow-up. In this study, the quantity of missing data replaced with the last observation carried forward was less than 15%. The expectation–maximization procedure was used to estimate and complete the missing values at the item level. Outliers were identified and inspected both at pre- and post-assessment in order to ensure that these values were within the range of scores defined by the maximum and minimum values of the scales. The 5% trimmed mean was compared to the original mean, and in all cases the differences were small, indicating that the outliers had little impact on the original mean. Therefore, outliers were not modified. All analyses were conducted using IBM SPSS Statistics 21, and post-test maternal mental distress was measured as the dependent variable. We examined the mean, SD, and frequency as well as how many mothers scored above the clinical cut-off on maternal mental distress at baseline. We then investigated how maternal mental distress was distributed between the ethnic groups and between intervention and control conditions using t-tests and chi-squared tests. Analysis of covariance (ANCOVA), with the baseline score on maternal mental distress as the covariate and the condition as the independent variable, was used to test the effects of the intervention on maternal mental distress upon program termination. All potential moderators were measured at baseline. Variables were centered to prevent multicollinearity among the interaction terms in the equation (Tabachnick & Fidell, 2001). We examined the potential moderators of the intervention effects using hierarchical multiple regressions and by conducting a separate regression for each potential moderator variable. In Step 1, the baseline maternal mental distress, intervention status and potential moderator variable were entered into the calculation. In Step 2, the interaction term (potential moderator × intervention status) was entered. Post hoc tests of the moderator effect using ANCOVA with baseline scores of maternal mental distress as the covariate were performed to further explore the interaction effect. In addition, scatterplots were performed for purposes of illustration. 3. Results 3.1. Level of maternal mental distress at intake to PMTO The average score for mental distress at the baseline was M = 7.62 (SD = 2.98) for the intervention and M = 7.70 (SD = 3.29) for the control condition, which represents a non-significant difference. The level of mental distress ranged from 5 to 19. The results showed that 20 mothers scored above the clinical cut-off for maternal mental distress at the baseline and that these mothers were evenly distributed in the intervention and control groups (10 in each). As, noted in Subsection 2.1 there was a significant difference between the two ethnic minorities in the level of maternal mental distress (p = .00) of and numbers of mothers scored above the clinical cut-off scores in this sample (p = .02): 17 were immigrants from Pakistan and three from Somalia. 3.2. Intervention outcome The mothers' mean score for post-test mental distress was M = 7.98 (SD = 3.27) for the intervention condition and M = 7.54 (SD = 2.53) for the control group. The results of an ANCOVA revealed that there were no significant effects of the intervention on maternal mental distress F(1. 95) .70, p = .40. 3.3. Moderator analysis The potential moderators included (a) family factors, including ethnicity, maternal education level, marital status, family size, maternal age
14
R. Bjørknes et al. / Children and Youth Services Review 51 (2015) 10–17
and duration of residence in Norway; (b) parenting factors, including harsh discipline and positive parenting; and (c) child factors, including gender, age and level of child conduct problems. Correlations between all potential moderators and outcome variables are shown in Table 2. To test for moderation, multiple regression analyses were conducted, see Table 3. Fig. 1 illustrates the slopes for the intervention and control groups yielded by the moderator analysis for ethnicity.
3.3.1. Family factors Ethnicity (see Fig. 1) and family size were significant moderators and interacted with intervention status to predict mental distress post-assessment. Baseline maternal mental distress, intervention status and ethnicity were entered at Step 1, explaining 17% of the variance in postassessment mental distress. After entering the interaction term (ethnicity × intervention status) in Step 2, the total variance explained by the model as a whole was 22% of the variance of maternal mental distress post-assessment. The unique interaction effect of ethnicity × intervention status explained 5% of the variance in post-assessment mental distress. R squared change = .05, F change (1, 91) = 5.65, p b .05, beta = .22, p b .05. Post hoc tests revealed that Pakistani mothers fared poorly in the intervention group, which meant that the intervention had a significant negative effect on maternal mental distress for Pakistani mothers compared with the control group F(1.57) = 3.82, p b .05. Among the Somali participants, there were no significant differences between PMTO and WLC groups postassessment. Baseline maternal mental distress, intervention status and family size were entered at Step 1, explaining 15% of the variance in postassessment mental distress. After entering the interaction term (family size × intervention status) in Step 2, the total variance explained by the model as a whole was 22% of the variance of post-assessment maternal mental distress. In other words, the unique interaction effect of family size × intervention status explained 7% of the variance in postassessment mental distress. R squared change = .07, F change (1, 88) = 7.37, p b .05, beta = .26, p b .05. Post hoc tests revealed that families with four children or more fared better in the control group than in the PMTO group F(1,50) = 4.68, p b .05. For families with three children or fewer, there were no significant differences between the PMTO and WLC groups post-assessment. There were no significant moderator effects for maternal education level, marital status, maternal age or duration of residence in Norway, which implies that these factors did not have an interaction effect on mental distress.
Table 3 Moderator analyses: regression models predicting maternal mental distress postintervention.
Potential moderator Step 1 Baseline maternal mental distress score Potential moderator Intervention Step 2 Baseline maternal mental distress score Potential moderator Intervention Interaction term
Ethnicity
Family size
Children's age
β
β
β
R^2
.27⁎⁎ .22⁎ .08 .25⁎
.17⁎⁎⁎
.37⁎⁎⁎ .03 .05
.05⁎
.34⁎⁎
.23⁎ .08 .22⁎
R^2 .15⁎⁎
.07⁎⁎
.04 .05 .26⁎⁎
R^2 .35⁎⁎⁎ .08 .06 .33⁎⁎
.13⁎⁎
.04⁎
.07 .06 −.20⁎
Note. Dependent variable = maternal mental distress score post-intervention; Step 1 covariate = maternal mental distress score at baseline. ⁎ Significant at the 0.05 level (2-tailed). ⁎⁎ Significant at the 0.01 level (2-tailed). ⁎⁎⁎ Significant at the 0.001 level (2-tailed).
3.3.2. Parenting factors There was no moderator effect at the baseline level of maternal reported harsh discipline or positive parenting, which suggests that these parenting practices did not interact with potential change in maternal mental distress due to the intervention.
3.3.3. Child factors Child gender had no moderating effect, which suggests that gender did not interact with changes in maternal mental distress after participating in the intervention. Table 3 shows that the age of the child was a significant moderator. Baseline maternal mental distress, intervention status and child age were entered at Step 1 and explained 13% of the variance in post-assessment mental distress. After entering the interaction term (child age × intervention status) in Step 2, the total variance explained by the model as a whole was 17% of the variance of postassessment maternal mental distress, and the unique interaction effect of child age × intervention status explained 4% of the variance in postassessment mental distress. R squared change = .04, F change (1, 91) = 4.28, p b .05, beta = − .20, p b .05. Post hoc tests revealed that mothers of 3-year-old to 5-year-old children in the control condition fared significantly better than the mothers in the PMTO group at post-assessment F(1,47) = 5.87, p b.05. By contrast, there were no significant differences between PMTO and WLC groups at post-assessment for mothers with 6-year-old to 9-year-old children.
Table 2 Correlation matrix between outcome variables on parent practices and child behavior at intake. Variable
1
1. Mental distress baseline 2. Mental distress post 3. Harsh discipline baseline 4. Positive parenting baseline 5. Conduct problem baseline 6. Child age 7. Maternal age 8. Family size 9. Ethnicity (Somali) 10. Education (Yes) 11. Marital status (Partner) 12. Child gender (Boy) 13. Duration of residence
– .35⁎⁎ .34⁎⁎
−.24⁎ .45⁎⁎ −.057 .20 .26⁎ −.37⁎⁎ −.22⁎ .16 .02 .18
2 – .07 −.09 .28⁎⁎ .08 −.04 .13 .32⁎⁎ −.03 .19 −.03 .21
Note. ⁎ Correlation is significant at the 0.05 level, 2-tailed. ⁎⁎ Correlation is significant at the 0.01 level, 2-tailed.
3
4
5
6
7
8
9
10
11
12
13
– −.23⁎ .48⁎⁎ .03 .01 .05 −.02 −.12 .18 −.07 −.01
– −.21⁎ .04 −.07 −.02 .00 .10 .00 −.03 −.04
– .07 −.05 .05 .14 −.20 .17 −.06 −.06
– .22⁎ .04 −.12 .17 .04 .10 .20
– .53⁎⁎ .11 .06 −.08 .00 .41⁎⁎
– .16 .23⁎ −.20 .09 .12
– .01 .02 .06 .42⁎⁎
– −.33⁎ −.01 .34⁎⁎
– .03 −.09
– .09
–
R. Bjørknes et al. / Children and Youth Services Review 51 (2015) 10–17
11 10 Pakistani PMTO 9 Pakistani Control condion
8
Somali PMTO 7 Somali Control condion
6 5 Pre
Post
Fig. 1. Intervention effects on maternal mental distress at pre- and post-assessment by ethnicity, for PMTO and control condition.
Baseline maternal mental distress, level of child conduct problems and intervention status problems at baseline were entered at Step 1 and explained 15% of the variance in post-assessment mental distress. After entering the interaction term (level of child conduct problems baseline × intervention status) in Step 2, the total variance explained by the model as a whole was 19% of the variance of maternal mental distress at post-assessment. Thus, the unique interaction effect of level of child conduct problems × intervention status explained 4% of the variance in post-test mental distress (R squared change = .04, F change (1, 91) = 5.00, p b .05, beta = −.21, p b .05). However, this significance disappeared when it was tested with ANCOVA as the post hoc analysis. There was no difference between PMTO and WLC at post-assessment for mothers with children whose scores were under the mean level of child conduct problems (Mean = 52) or above the mean level of child conduct problems. 4. Discussion The purpose of this study was to investigate maternal mental distress among 96 Somali and Pakistani immigrant mothers participating in PMTO. The results showed that there were large disparities in the level of maternal mental distress within the group of study participants at enrollment. PMTO did not have a significant effect on the levels of maternal mental distress. Ethnicity, family size and child age served as moderators on the relationship between the condition (PMTO or WLC) and maternal mental distress outcome. None of these subgroup analyses were in favor of the intervention, and importantly, PMTO had negative effects on the outcomes of the maternal mental distress of women of Pakistani origin. Strand et al. (2003) reported that factors such as gender (women), low education, unemployment and being single predict higher levels of mental distress in the Norwegian population. Immigration and refugee status are also considered to be risk factors for maternal mental distress (Dalgard & Thapa, 2007). Because most of the participants in this study are subject to one or more of these risk factors, the fact that 20% are over the clinical cut-off limit for mental distress must be considered quite low. Compared with the results from previous research on risk factors for depression and anxiety in immigrant populations (Bhui et al., 2003; Dalgard et al., 2006; Dalgard & Thapa, 2007), the sample had unexpectedly low levels of maternal mental distress. There was also a disparity in the levels of mental distress among the participants by ethnic group: 17 ethnic Pakistanis and three ethnic Somalis scored above the clinical cut-off on maternal mental distress at baseline. However, there were significant correlations between mental distress and factors such as low education, harsh discipline, and having children with higher levels of conduct problems at baseline. Compared with other studies examining samples of families with children that have conduct problems, our study sample also scored lower on mental distress than expected.
15
For example, both Hutchings et al. (2002) and DeGarmo et al. (2004) reported that approximately 50% of the mothers of children with conduct problems were clinically depressed. It is important to note that the recruitment criteria in the current study stated that parents with serious mental health problems could not participate. Even if no participant was excluded from the study, because of that criterion (Bjørknes & Manger, 2013) professionals from the public service agencies might have disqualified these potential participants without reporting it to the researchers in this study. Most of the women in this study were self-referred to the intervention (Bjørknes et al., 2011). Although this is speculation, it might be reasonable to assume that, as a self-referred group, these women may have been aware that they not only required help but also possessed the agency to actively seek help when the opportunity presented itself. These qualities may have made them less prone to mental distress because they were less likely to perceive themselves as helpless or their situation as hopeless. The results show that PMTO did not have a significant effect on maternal mental distress in this sample. However, the low level of maternal distress measured at baseline creates a floor effect, and the analysis of the effects and the moderators should be interpreted with caution. Findings from a similar study on the effects of Parent– Child Interaction Therapy for Puerto Rican preschool children (4-years old to 6-years old) with a diagnosis of attention-deficit/ hyperactivity disorder also found a surprisingly low level of clinical depression in mothers at baseline, and Matos et al. (2009) stated that this finding explained why treatment effects were not found (Matos et al., 2009). In some ways, the lack of an effect on maternal distress challenges earlier research on parent training because there have been several randomized controlled trials that have found a significantly decreased level of mental distress for mothers following parent training (Barlow et al., 2002; DeGarmo et al., 2004; Hutchings et al., 2002; Kjøbli et al., 2013; Kjøbli & Ogden, 2012). There are at least two possible explanations for why our results are inconsistent with previous studies. First, a longer time interval from intervention to follow-up may be required to show the possible effects of PMTO on mental distress. This result would be consistent with results from an earlier study showing that the effects from parent training on maternal depression are mediated by reductions in the externalization of children's behavior, which did not occur before a 30-month follow-up assessment (DeGarmo et al., 2004). A second possible reason for the missing effect might be cultural differences; most of the earlier studies were performed on Caucasian middle-class mothers (Michelson et al., 2013), and the findings from these studies are not necessarily generalizable to different ethnic groups. This hypothesis is supported by the finding that PMTO had a negative effect on the Pakistani immigrant mothers. Third, there is relatively robust evidence suggesting that cultural adaptation increases the accessibility and acceptability and could increase the effectiveness of psychosocial interventions (Chowdhary et al., 2014). We might speculate that better cultural adaptation may have reduced the potential iatrogenic effects, but unfortunately this study did not investigate cultural adaptation. Our third aim for this study was to examine whether family, parenting, and child factors moderated the relationship between PMTO and maternal mental distress post-intervention. Ten variables were tested, and three (ethnicity, family size and child age) were found to significantly moderate the relationship between the condition (PMTO or WLC) and maternal mental distress. The interaction effect between ethnicity and the condition explained an additional 5% of the variance in maternal mental distress, when the effects of the pre-score and the condition were controlled for. In other words, maternal mental distress increased after participating in PMTO for the Pakistani immigrant mothers compared with the control condition, which was not the case for the Somali sample. The significant difference between the two ethnic minorities in the baseline score could be the reason for this finding; where high level of mental distress
16
R. Bjørknes et al. / Children and Youth Services Review 51 (2015) 10–17
could be a risk factor for increased maternal mental distress after participating in PMTO. However, this finding might be explained by task overload, i.e., adding PMTO to an already strained life might have led to a higher degree of feeling helpless. Attending the PMTO training may have led these mothers to have a greater awareness of the problems with their parenting skills, which might have increased their feelings of incompetence. It might be that these women would need additional help from a program focusing specifically on depression and anxiety to show improvement. For example, Sanders and McFarland (2000) found that parent training combined with cognitive therapy for depressed mothers had better long-term effects than parent training alone. Some authors have argued that there might be different cultural explanations for the disparate findings of increases in mental distress, such as specific group norms or relationships made in the training groups, differences in how the parent training was conducted or translated, and different understandings of the survey questions (Hutchings et al., 2002). We would argue that our finding raises an important issue for research on how mental distress is understood within different cultures. First, there is the issue of what is understood by mental distress because this condition may be understood differently in different cultures, and using criteria and instruments developed in the West and tested on Western populations may not be appropriate for nonWestern populations. This difference in understanding might also be one reason for the unexpected low level at baseline. Second, the difference between the two populations at baseline, where Pakistani mothers scored higher on maternal mental distress than Somali mothers, may have other cultural explanations in addition to their immigration experience. Although this is only speculation, Pakistanis predominantly came to Norway as immigrant workers or to join family members and are more established and integrated into Norwegian society because they have been here longer. Dalgard and Thapa (2007) found that social integration had a negative effect on the mental health of non-Western immigrant women. Conversely, Somalis tend to come to Norway as refugees, typically via other countries, and have resided in Norway for a shorter time. Thus, in a sense, the Pakistanis may have been struggling longer with integration, whereas the Somalis – who have escaped war and violence and have experienced much uncertainty and many challenges – may now feel that they are settled and safe, which might lead them to feel more resilient. An alternative explanation might be that immigration out of necessity, whether caused by war or persecution, might in fact make adjustment easier because of the very lack of choice — compared with when a choice, such as electing to pursue a better life through immigration, is involved. These diverse immigration experiences may therefore represent fairly – or even very – different mindsets when addressing parenting practices and child conduct problems, which might thereby lead to various types of stress and tension and different levels of mental distress. When the effects of the pre-score and the condition were controlled for, the interaction effect between family size and condition explained an additional 7% of the variance in maternal mental distress, on the one hand, and the interaction effect between child age and the condition explained an additional 4% of the variance in maternal mental distress, on the other hand. However, a decrease in mental distress was observed for mothers in the WLC with four children or more and with 3-year-old to 5-year-old children. It is not clear why these mothers rated themselves as having less mental distress but it might be related to methodological issues, such as participant's reactivity to the study situation or it may simply be an anomaly. Poor parental practices and socioeconomic disadvantages, such as education level, marital status, and family size have frequently been associated with poor outcomes in parent training (Lundahl et al., 2006; Reyno & McGrath, 2006). None of these family factors seem to interact with changes in mental distress due to the intervention in this sample, and neither did the two parenting factors, harsh discipline and positive parenting.
4.1. Limitations There are a number of possible limitations in the current study. As noted above, because of the low level of mental distress at baseline, the non-significant effect on the level of post-assessment maternal mental distress might be due to a floor effect. In addition, this study does not include a follow-up assessment. A randomized trial with follow-up has shown that depression changes over time (30 months from the baseline) for PMTO mothers (DeGarmo et al., 2004). Based on this finding, it is clear that we do not fully understand the processes of changes in maternal mental distress; thus, we cannot draw the conclusion from this study that the intervention has no effect. Although the trial has statistical power for investigating the potential effects of the intervention, the sample size must be considered relatively small for post hoc moderator analysis. In addition, the differences between PMTO and the control condition in the significant moderator analysis are minor. Thus, these moderator analyses should be considered as exploratory, and the results interpreted with caution. However, because there is little research on parent training with immigrant families, exploratory studies are worthwhile. Another source of uncertainty in this study is the validity of all the instruments used. Although all the instruments in this study had a high reliability score and showed high psychometric quality for ethnic Norwegian samples (e.g., Reedtz et al., 2008; Tambs & Moum, 1993), the external and construct validity of these instruments in Somali and Pakistani ethnic minority groups in Norway is unknown. Furthermore, while all the instruments were translated into Urdu and Somali and the interviewers were bilingual, it might be possible that phrases used in the instruments were not culturally relevant to Pakistani and Somali respondents (Bhui et al., 2003; Syed et al., 2006). One could argue that parental distress only in extreme circumstances leads to anxiety and depression. Therefore, the measure may not have been sensitive enough to pick up effects of parental mental distress. Future research on the intervention effects of mental distress should be better grounded in the conceptual and measurement literature regarding parental distress. In addition, fathers were unfortunately not included in the study, and the results were based solely on mothers' reports; it is notable that the role of fathers has received little attention in research associated with parenting and mental distress (Goodman et al., 2011). Nevertheless, Carro, Grant, Gotlib, and Compas (1993) found that the absence of depression in the father was a protective factor for conduct problems in the child. In future studies, it will be important to test the significance of fathers' parenting practices as both mediators and moderators of the intervention effect on mental distress. 4.2. Implications To our knowledge, this study is the first to explore mental distress among Somali and Pakistani immigrants in an effectiveness trial of families receiving PMTO. The low levels of mental distress among the sample at enrollment in PMTO are possible indications that participants were resourceful despite their challenging circumstances. The findings from the main analysis in this trial indicate that PMTO has a positive effect on maternal parenting practices and child conduct problems in Norway (Bjørknes & Manger, 2013). The findings in the current study highlight the importance of having cross-cultural research when investigating parent training. An increase in maternal mental distress in the Pakistani group indicates that PMTO may actually have a negative effect on certain groups compared with the WLC. From this perspective, immigration may be understood as a source of stress and tension that might exhaust parents' resources and thereby increase levels of mental distress when attending the PMTO. This effect may in the long run disturb parenting practices and increase conduct problems in children (Hoeve et al., 2009; Miner & Clarke-Stewart, 2008). This possibility highlights the need for further research on the moderators of treatment responses regarding mental distress to determine under what conditions mental
R. Bjørknes et al. / Children and Youth Services Review 51 (2015) 10–17
distress actually decreases following PMTO — and when it does not. In addition, findings from this study suggest that more comprehensive interventions may be required for mothers with mental distress. A previous study showed that parent training combined with cognitive therapy for depressed mothers had more positive outcomes than parent training alone (Sanders & McFarland, 2000). Consistent with the foregoing and as suggested by Kjøbli et al. (2013), it may be beneficial to combine parent training with evidence-based interventions that aim directly at reducing maternal mental distress, such as cognitive therapy or Acceptance and Commitment Therapy (Biglan, Hayes, & Pistorello, 2008). In addition future research should examine the possibility of the iatrogenic effects of the intervention, particularly in terms of expanding on factors, such as measuring cultural adaptation (Chowdhary et al., 2014) that might account for such effects. References Achenbach, T. M., & Rescorla, L. A. (2001). Manual for the ASEBA school-age forms & profiles. Burlington, VT: University of Vermont, Research Center for Children, Youth, & Families. Askeland, E., Christiansen, T., & Solholm, R. (Eds.). (2005). Norwegian handbook of parent management training. Oslo, Norway: Norwegian Center for Child Behavioral Development. Barlow, J., Coren, E., & Stewart-Brown, S. (2002). Meta-analysis of the effectiveness of parenting programmes in improving maternal psychosocial health. British Journal of General Practice, 52(476), 223–233. Bhui, K., Abdi, A., Abdi, M., Pereira, S., Dualeh, M., Robertson, D., et al. (2003). Traumatic events, migration characteristics and psychiatric symptoms among Somali refugees: Preliminary communication. Social Psychiatry and Psychiatric Epidemiology, 38, 35–43. http://dx.doi.org/10.1007/s00127-003-0596-5. Biglan, A., Hayes, S. C., & Pistorello, J. (2008). Acceptance and commitment: Implications for prevention science. Prevention Science, 9(3), 139–152. http://dx.doi.org/10.1007/ s11121-008-0099-4. Bjørknes, R., Jakobsen, R., & Nærde, A. (2011). Recruiting ethnic minority groups to evidence-based parent training. Who will come and how? Children and Youth Services Review, 33, 351–357. Bjørknes, R., & Manger, T. (2013). Can parent training alter parent practice and reduce conduct problems in ethnic minority children? A randomized controlled trial. Prevention Science, 14(1), 52–63. http://dx.doi.org/10.1007/s11121-012-0299-9. Brislin, R. W. (1970). Back-translation for cross-cultural research. Journal of Cross-Cultural Psychology, 1, 185–216. Carro, M. G., Grant, K. E., Gotlib, I. H., & Compas, B. E. (1993). Postpartum depression and child development: An investigation of mothers and fathers as sources of risk and resilience. Development and Psychopathology, 5(04), 567–579. http://dx.doi.org/10. 1017/S0954579400006167. Chamberlain, P., & Reid, J. B. (1987). Parent observation and report of child symptoms. Behavioral Assessment, 9, 97–109. Chowdhary, N., Jotheeswaran, A. T., Nadkarni, A., Hollon, S. D., King, M., Jordans, M. J. D., et al. (2014). The methods and outcomes of cultural adaptations of psychological treatments for depressive disorders: A systematic review. Psychological Medicine, 44, 1131–1146. http://dx.doi.org/10.1017/S0033291713001785. Dalgard, O. S., & Thapa, S. B. (2007). Immigration, social integration and mental health in Norway, with focus on gender differences. Clinical Practice and Epidemiology in Mental Health, 3–24. http://dx.doi.org/10.1186/1745-0179-3-24. Dalgard, O. S., Thapa, S. B., Hauff, E., McCubbin, M., & Syed, H. S. (2006). Immigration, lack of control and psychological distress: Findings from the Oslo Health Study. Scandinavian Journal of Psychology, 47, 551–558. DeGarmo, D. S., Patterson, G. R., & Forgatch, M. S. (2004). How do outcomes in a specified parent training intervention maintain or wane over time? Prevention Science, 5(2), 73–89. http://dx.doi.org/10.1023/B:PREV.0000023078.30191.e0. Eyberg, S., & Pincus, D. (1999). Eyberg Child Behavior Inventory and Sutter–Eyberg Student Behavior Inventory — Revised. Professional manual. Florida: Psychological Assessment Resources, Inc. Flock, M. C., & Pettersen, R. E. (2008). Mater et magistra; handbook of PMTO for Somali and Pakistani mothers. Oslo: Norwegian Center for Child Behavioral Development. Forgatch, M. S. (1994). Parenting through change: A training manual. Eugene: Oregon Social Learning Center. Forgatch, M. S., & DeGarmo, D. S. (1999). Parenting through change: An effective prevention program for single mothers. Journal of Consulting and Clinical Psychology, 67, 711–724. Forgatch, M. S., Patterson, G. R., DeGarmo, D. S., & Beldavs, Z. G. (2009). Testing the Oregon delinquency model with 9-year follow-up of the Oregon Divorce Study. Development and Psychopathology, 21, 637–660. Furlong, M., McGilloway, S., Bywater, T., Hutchings, J., Smith, S. M., & Donnelly, M. (2012). Behavioural and cognitive–behavioural group-based parenting programmes for early-onset conduct problems in children aged 3 to 12 years. Cochrane Database of Systematic Reviews(2). http://dx.doi.org/10.1002/14651858.Cd008225.Pub2. Gardner, F., Hutchings, J., Bywater, T., & Whitaker, C. (2010). Who benefits and how does it work? Moderators and mediators of outcome in an effectiveness trial of a parenting intervention. Journal of Clinical Child & Adolescent Psychology, 39(4), 568–580. http:// dx.doi.org/10.1080/15374416.2010.486315. Goodman, S. H., Rouse, M. H., Connell, A. M., Broth, M. R., Hall, C. M., & Heyward, D. (2011). Maternal depression and child psychopathology: A meta-analytic review.
17
Clinical Child and Family Psychology Review, 14(1), 1–27. http://dx.doi.org/10.1007/ s10567-010-0080-1. Hoeve, M., Dubas, S., Eichelsheim, V., van der Laan, H., Smeenk, W., & Gerris, J. (2009). The relationship between parenting and delinquency: A meta-analysis. Journal of Abnormal Child Psychology, 37, 749–775. http://dx.doi.org/10.1007/s10802-0099310-8. Hutchings, J., Appleton, P., Smith, M., Lane, E., & Nash, S. (2002). Evaluation of two treatments for children with severe behaviour problems: Child behaviour and maternal mental health outcomes. Behavioural and Cognitive Psychotherapy, 30(3), 279–295. Kjøbli, J., Hukkelberg, S., & Ogden, T. (2013). A randomized trial of group parent training: Reducing child conduct problems in real-world settings. Behaviour Research and Therapy, 51(3), 113–121. http://dx.doi.org/10.1016/j.brat.2012.11.006. Kjøbli, J., Nærde, A., Bjørnebekk, G., & Askeland, E. (2013). Maternal mental distress influences child outcomes in brief parent training. Child and Adolescent Mental Health. http://dx.doi.org/10.1111/camh.12028. Kjøbli, J., & Ogden, T. (2012). A randomized effectiveness trial of brief parent training in primary care settings. Prevention Science, 13(6), 616–626. http://dx.doi.org/10.1007/ s11121-012-0289-y. Kritz, K., & Skivenes, M. (2010). ‘Knowing our society’ and ‘fighting against prejudices’: How child welfare workers in Norway and England perceive the challenges of minority parents. British Journal of Social Work, 40, 2634–2651. http://dx.doi.org/10.1093/ bjsw/bcq026. Lillehaug, T. (2007). Utprøving av strategier og tiltak for rekruttering av foreldre med etnisk minoritetsbakgrunn til et PMTO gruppetiltak [Investigating strategies to recruit parents with ethnic minority background to PMTO]. Norway, Oslo: Norwegian Center for Child Behavioral Development. Lundahl, B., Risser, H. J., & Lovejoy, M. C. (2006). A meta-analysis of parent training: Moderators and follow-up effects. Clinical Psychology Review, 26(1), 86–104. http:// dx.doi.org/10.1016/j.cpr.2005.07.004. Martinez, C. R., Jr., & Forgatch, M. S. (2001). Preventing problems with boys' noncompliance: Effects of a parent training intervention for divorcing mothers. Journal of Consulting and Clinical Psychology, 69, 416–428. Matos, M., Bauermeister, J. J., & Bernal, G. (2009). Parent–child interaction therapy for Puerto Rican preschool children with ADHD and behavior problems: A pilot efficacy study. Family Process, 48(2), 232–252. Michelson, D., Davenport, C., Dretzke, J., Barlow, J., & Day, C. (2013). Do evidence-based interventions work when tested in the “real world?” A systematic review and meta-analysis of parent management training for the treatment of child disruptive behavior. Clinical Child and Family Psychology Review, 16(1), 18–34. http://dx.doi. org/10.1007/s10567-013-0128-0. Miner, J., & Clarke-Stewart, K. A. (2008). Trajectories of externalizing behavior from age 2 to age 9: Relations with gender, temperament, ethnicity, parenting, and rater. Developmental Psychology, 44(3), 771–786. Missinne, S., & Bracke, S. (2012). Depressive symptoms among immigrants and ethnic minorities: A population based study in 23 European countries. Social Psychiatry and Psychiatric Epidemiology, 47, 97–109. http://dx.doi.org/10.1007/s00127-0100321-0. Ogden, T., & Amlund-Hagen, K. (2008). Treatment effectiveness of parent management training in Norway: A randomized controlled trial of children with conduct problems. Journal of Consulting and Clinical Psychology, 76(4), 607–621. Oppedal, B., Azam, G. E., Dalsøren, S. B., Hirsch, S. M., Jensen, L., Kiamanesh, P., et al. (2008). Psykososial tilpasning og psykiske problemer blant barn i innvandrerfamilier. Rapport: 14. Oslo: Folkehelseinstituttet. Patterson, G. (1982). A social learning approach. Coercive family process, vol. 3, Eugene, OR: Castalia Publishing Company. Reedtz, C., Bertelsen, B., Lurie, J., Handegård, B. H., Clifford, G., & Mørch, W. -T. (2008). Eyberg Child Behavior Inventory (ECBI): Norwegian norms to identify conduct problems in children. Scandinavian Journal of Psychology, 49, 31–38. Reyno, S. M., & McGrath, P. J. (2006). Predictors of parent training efficacy for child externalizing behavior problems — A meta-analytic review. Journal of Child Psychology and Psychiatry, 47(1), 99–111. http://dx.doi.org/10.1111/j.1469-7610. 2005.01544.x. Sanders, M. R., & McFarland, M. (2000). Treatment of depressed mothers with disruptive children: A controlled evaluation of cognitive behavioral family intervention. Behavior Therapy, 31(1), 89–112. Scott, S., O'Connor, T. G., Futh, A., Matias, C., Price, J., & Doolan, M. (2010). Impact of a parenting program in a high-risk, multi-ethnic community: The PALS trial. Journal of Child Psychology and Psychiatry, 51(12), 1331–1341. http://dx.doi.org/10.1111/j. 1469-7610.2010.02302.x. Statistics Norway (2014). Innvandrere og norskfødte med innvandrerforeldre, 1. January 2014 [Immigrants and those born in Norway with immigrant parents, 1 January 2014]. http://www.ssb.no/innvbef/ (06.06) Strand, B. H., Dalgard, O. S., Tambs, K., & Rognerud, M. (2003). Measuring the mental health status of the Norwegian population: A comparison of the instruments SCL25, SCL-10, SCL-5 and MHI-5 (SF-36). Nordic Journal of Psychiatry, 57(2), 113–118. Syed, H. R., Dalgard, O. S., Dalen, I., Claussen, B., Hussain, A., Selmer, R., et al. (2006). Psychosocial factors and distress: A comparison between ethnic Norwegians and ethnic Pakistanis in Oslo, Norway. BMC Public Health, 6. http://dx.doi.org/10.1186/14712458-6-182. Tabachnick, B. G., & Fidell, L. S. (2001). Using multivariate statistics. New York: Allyn and Bacon. Tambs, K., & Moum, T. (1993). How well can a few questionnaire items indicate anxiety and depression? Acta Psychiatrica Scandinavica, 87(5), 364–367. http://dx.doi.org/ 10.1111/j.1600-0447.1993.tb03388.x. Webster-Stratton, C. (1998). Parent practices interview. Retrieved August 15, 2007, from http://son.washington.edu/centers/parenting-clinc Weisz, J. R., & Kazdin, A. E. (2010). Evidence-based psychotherapies for children and adolescents (2nd ed.). New York: The Guilford Press.
