دخالت پدر در درمان سیستمیک چندگانه: بررسی تاثیرات بر نتایج نوجوان و افسردگی مادر

Abstract The association between paternal involvement in therapy, adolescent outcomes and maternal depression was examined within the context of Multisystemic Therapy (MST), an empirically supported, family- and community-based treatment for antisocial adolescents. Ninety-nine families were recruited from five mental health agencies providing MST. We compared families with paternal involvement in therapy (PIT) to families with no paternal involvement in therapy (NPIT) in pre-post change in adolescents’ externalizing and internalizing behaviours and also in maternal depression. There was a significant reduction in both groups in externalizing and internalizing behaviours. However, the magnitude of improvement was significantly greater for the PIT families. Both groups saw a significant reduction in maternal depression but no significant group differences were found. Results suggest that if possible, paternal figures should be encouraged to actively participate in therapy, as adolescents outcomes are enhanced when mothers and paternal figures participate in MST together.

Results Preliminary analyses We first compared our NPIT and PIT groups on the demographic variables using independent samples t-tests, chi-square analyses, or Fisher’s exact tests as appropriate. All tests were two-tailed. As expected, the groups differed significantly in marital status, x2 (3, N = 99) = 66.35, p < .0001, household size, t (97) = −3.63, p < .001, and income, x2 (7, N = 96) = 31.39, p < .001, with the PIT group more likely to be married (versus separated or divorced, single or widowed) and thus having more members in the household and greater family income. The groups also differed in the extent to which the respondents’ spouse or partner was involved in the child’s life (versus not involved, whether or not the spouse/partner shared a home with the child). In the PIT group, 56 of 57 respondents partners were involved in the child’s life, while in the NPIT group only 14 of 42 respondents’ partners were involved, p < .0001. The groups did not differ significantly on any of the other demographic variables, including child age, gender and ethnicity; parent age, and education. The two groups were roughly equally distributed across the various agencies and the specific therapists treating the families, and did not differ in the time period during which they entered the study and received therapy (i.e., whether they entered the study earlier or later). Furthermore, the NPIT and PIT families did not differ significantly in their mean TAM-R scores (M = 4.29, SE = .40; M = 4.27, SE = .37, respectively). Thus, the two groups received treatment that was equivalent in degree of adherence to MST principles. Active participation in therapy For each of the measures of interest (BCFPI Externalizing and Internalizing, CAFAS and CES-D), we conducted a mixed 2 (Time; pre-, post-treatment) by 2 (Group; PIT, NPIT) analysis of variance (ANOVA), with Time as the within-subjects variable and Group as the between- subjects variable. Table 2 shows the means and standard deviations for each measure. Table 2. Pre- and post-treatment means and standard deviations of all outcome variables for families with NPIT and PIT. Scale NPIT PIT Pre Post Pre Post BCFPI Externalizing 76.48 (10.59) 70.97 (13.58) 76.66 (11.38) 65.10 (12.43) BCFPI Internalizing 65.53 (13.44) 63.49 (13.23) 64.91 (15.25) 56.46 (12.16) CAFAS Externalizing 76.76 (26.25) 36.97 (30.77) 61.20 (24.21) 21.43 (21.01) CES-D 3.57 (2.87) 2.50 (2.45) 3.23(2.81) 1.84 (2.66) Note. Standard deviations are in parentheses next to mean score. BCFPI = Brief Child and Family Phone Interview; CAFAS = Child and Adolescent Functional Assessment Scale; CES-D = Center for Epidemiologic Studies Depression Scale. Table options The ANOVA results for the BCFPI Externalizing and Internalizing scales were analogous. There were significant main effects of Time (Externalizing: F(1, 97) = 46.58, p < .0001; Internalizing: F(1, 94) = 15.93, p < .0001), indicating improvement from pre- to post-treatment. These findings were qualified by significant Time by Group interactions (Externalizing: F(1, 97) = 5.84, p < .02; Internalizing: F(1, 94) = 5.66, p < .02). The means for the two groups did not differ significantly at pre-treatment. However, the means for the PIT group were significantly lower than the means for the NPIT group at post-treatment (p < .03 for Externalizing and p < .01for Internalizing). Although both groups improved significantly over time, the magnitude of improvement was greater for the PIT families compared to the NPIT families. In order to ensure that our results were particular to paternal involvement in therapy and not just about fathers’ presence in the home or marital status, we conducted additional analyses. We conducted the same ANOVA substituting marital status (married or not married) or amount of involvement of the mother’s spouse or partner in the child’s life (involved or not involved) for the Group variable. No significant differences in outcomes were found based on marital status or extent of involvement in the household. Alternatively, treatment efficaciousness may be measured by assessing the degree of clinically significant improvement. We operationalized clinically significant improvement on the BCFPI externalizing and internalizing measures as a decrease in five points, or half of a standard deviation (Lewis et al., 2008). To determine if our groups met this criterion after treatment, we conducted separate one-sample t-tests comparing the mean change scores for each group (NPIT and PIT) to five. For both measures, whereas the magnitude of change was not significantly different from five for the NPIT families, it was significantly greater than five for the PIT families (Externalizing: t (56) = 3.97, p < .0001; Internalizing: t (54) = 2.09, p < .04). Thus, the NPIT group met the criteria for clinically significant improvement while the PIT group exceeded it. Another way to think about the effectiveness of treatment is in terms of whether or not the adolescents’ level of functioning met borderline clinical or clinical levels at pre- and at post- treatment. The conventional cutoff for borderline clinical-level functioning for the BCFPI externalizing and internalizing scales is 65 (Cunningham et al., 2006). Accordingly, participants were categorized as borderline clinical/clinically impaired if they scored 65 or higher. The categorization was done for pre- and post-treatment, for each scale separately. Chi-square analyses revealed that, for both measures, the two groups did not differ significantly in the proportion of adolescents categorized as borderline clinical/clinical at the outset of therapy. On the externalizing scale, 91% of the NPIT group met or exceeded the cutoff score of 65 at pre-treatment, while 86% of the PIT group met or exceeded this level. A smaller proportion of youth met or exceeded the borderline clinical/clinical cutoff on the internalizing scale, with 52% of the NPIT group and 51% of the PIT group considered borderline clinical/clinically impaired at the outset of therapy. However, only the PIT group saw a significant reduction in the total number of borderline clinical/clinical cases from pre- to post-therapy (p < .02 and p < .003 for Externalizing and Internalizing for the PIT group; Fisher’s exact test). The BCFPI outcome scores were reported by parents, so reporting biases may have influenced our results. To corroborate our findings, we examined therapist-rated externalizing behaviour as well. The ANOVA comparing PIT and NPIT families on CAFAS scores showed main effects of group (F(1, 80) = 9.61, p < .003) and time (F(1, 80) = 191.26, p < .0001). In general, CAFAS externalizing scores improved significantly from pre- to post-treatment, but the NPIT families remained significantly more impaired than the PIT families across time. Both groups exceeded the widely accepted criterion for improvement of a decrease of at least 20 points (t (32) = 3.61, p < .001 for the NPIT group and t (48) = 6.47, p < .0001 for the PIT group). We conducted Pearson product moment correlations to assess the relationship between parent-reported and therapist-reported externalizing scores using pre-, post-, and pre- to post-treatment change scores. All tests were two-tailed. At pre-treatment, parent-reported scores were not significantly correlated with therapist-reported scores; however, by post-treatment, externalizing scores from the two reporting sources were found to be positively correlated, r = .35, n = 99, p = .001. Analysis of pre- to post-treatment change scores revealed a positive correlation between the parent-reported BCFPI externalizing and therapist-reported CAFAS externalizing scores, r = .24, n = 99, p = .03. Finally, we examined the influence of paternal involvement in therapy on maternal depressive symptoms. The only significant finding from the ANOVA for the CES-D was a significant main effect of time (F(1, 97) = 17.57, p < .0001), indicating that both groups saw a significant reduction in parent depression from pre- to post-treatment. Although differences between PIT and NPIT groups were in the hypothesized direction, no significant differences by group were found. For the shortened version of the CES-D (used here), it has been suggested that a score of four or more reliably indicates clinical levels of depression ( Santor & Coyne, 1997). We therefore categorized families according to this criterion at pre- and also at post-treatment. The proportion of depressed mothers did not differ significantly between the NPIT and PIT groups at the start of treatment (43% and 44% for the NPIT and PIT families, respectively). However, both groups saw significant reductions from pre- to post-treatment in the number of mothers with scores in the clinical range (p < .004 and p < .007 for the NPIT and PIT families, respectively; Fisher’s exact test).