|Home | About | Journals | Submit | Contact Us | Français|
The present study investigates the effectiveness of a short-term day treatment program for children with a primary diagnosis of a Disruptive Behaviour Disorder (DBD) using best-practice treatment strategies. This study, using a wait list control, compared children’s admission and discharge test scores on standardized measures of behaviour and functioning, as rated by parents.
A repeated measures MANOVA was used to evaluate symptom change.
The treatment group was found to have improved significantly more than the waitlist group on parent reports of externalizing behaviour, aggression, social problems, hyperactivity and parent stress.
Children with DBD who attended a short-term day treatment program using best-practice treatment strategies showed significant improvement in their behaviour at home. The current study lends support to the idea that severe DBD can be treated using multi-modal, intensive and evidence-based treatment techniques.
Étudier l’efficacité d’un traitement en hôpital de jour à court terme des enfants qui ont reçu un diagnostic primaire de troubles sévères du comportement (TSC). Le traitement se base sur les meilleures pratiques. L’étude utilise une liste d’attente comme contrôle et compare les notes des enfants en début et en fin de traitement au rapport des parents sur le comportement et le fonctionnement des enfants.
L’évolution des symptômes a été mesurée à plusieurs reprises par analyse MANOVA.
Les résultats du groupe traité étaient significativement meilleurs que ceux de la liste d’attente et du rapport des parents sur l’extériorisation, l’agression, les problèmes sociaux, l’hyperactivité et le stress parental.
Le comportement à la maison des enfants qui suivent une TCC à court terme en hôpital de jour est en nette amélioration. Cette étude confirme que le traitement des troubles sévères du comportement peut être multimodal, intensif et factuel.
Disruptive Behaviour Disorders (DBD) are one of the most frequently diagnosed psychiatric disorders in childhood (Kazdin et al., 1994) with up to 10% of children having disruptive behaviour problems (Offord, Boyle & Racine, 1989). Onset of externalizing disorders in childhood has been shown to be highly predictive of future behaviour problems and to be costly to society (Fergusson et al., 2005; van Bokhoven et al., 2006). These societal costs typically stem from involvement with the criminal justice system, need for education specialists, and residential placements (Scott et al., 2001). As such, DBDs represent a very significant mental health concern.
In general, there is a continuum of treatment services for DBD ranging from low to high intensity. Typically, day treatment programs are conceptualized as any program that falls “in the middle of the continuum of care between inpatient and outpatient treatment” (Topp, 1991). Studies have compared day treatment programs to both inpatient and out-patient treatment options (e.g., van Bokhoven et al., 2005) and found that when compared to residential treatment, day treatment was shown to be equally effective in reducing problem behaviours. As such, day treatment programs are often seen as a positive alternative to residential placement due to the fact that they are less costly, less restrictive, and maintain a child’s contact with their home environment (Erker et al, 1993; Whitemore et al., 2003).
When children had comparable levels of behavioural difficulty, Grizenko and associates (1993) found that, compared to outpatient treatment, day treatment led to more improvement in behaviour, social skills, and family functioning. In addition, research indicates that clinicians refer children to day treatment programs when they have higher levels of psychopathology and family difficulties which are not as easily served in an outpatient setting, indicating that referrals are appropriately falling along the treatment continuum, with less severe DBD being treated in outpatient services, and more severe cases being referred to day treatment (McDermott et al, 2002; Ware et al., 2001).
Although there are some positive comparisons to residential and outpatient treatment, Farmer and associates (2002) noted that there is a lack of research on the multi-modal treatment of DBD in children between the ages of 6–12, making it difficult to draw conclusions about efficacy in this age range. Looking at day treatment programs specifically, some programs have been shown to significantly reduce disruptive behaviors and promote successful reintegration to communities and schools (Grizenko et al., 1993). However, other studies (Bennett et al., 2001; McCarthy et al., 2006) found no significant change in externalizing problems, although other difficulties (i.e., poor peer relations) were reduced. Zimet and associates (1980) actually found that children had higher levels of aggression at discharge than they had at admission to a two-year day treatment program, although some clinical improvement occurred post-discharge.
Many factors may help explain the inconsistencies in the research on day treatment programs (Kotsopoulos et al., 1996). First, treatment duration times vary for almost all programs (Kiser et al, 1995). These durations range from programs with 20–26 treatment days (McCarthy et al., 2006) to the many programs that last for more than a year (e.g., Gabel et al., 1990). The average length of stay in a child/adolescent day treatment program is 143 treatment days (Milin et al., 2000).
Treatment modalities also differ between programs. Evidence from review and meta-analytic studies indicates that cognitive behavioural strategies, parent management training, psychopharmacological treatment, and behavioural strategies are effective components of all treatment programs for DBD, including day treatment (Brestan and Eyberg, 1998; Pappadopulos et al., 2006). Furthermore, programs combining these interventions show the most success in treating severe DBD (Burke et al., 2002; Connor et al., 2006). Many of the day treatment programs that have been studied do not include all of these evidence-based treatment components. Evidence-based interventions in day treatment programs have been found to have larger effects than “usual or eclectic” approaches (Farmer et al., 2002). In fact, Frick (2001) cautioned that treatment for children with conduct disorder should be implemented with careful consideration due to the potential harmful effects of using therapy tools that are not evidence based.
Another difficulty with previous studies of day treatment programs is that they are not always limited to children with DBD as a primary diagnosis. In fact, in one evaluation of a partial hospitalization program (Robinson, 2000) only 59% of children had a primary diagnosis of a DBD. This is problematic as some research indicates that treatment gains in day treatment programs are less likely for children with externalizing difficulties, with depression found to be a very strong predictor of behavioural improvement (Grizenko and Sayegh, 1990).
One of the difficulties with almost all previous studies of Day Treatment programs (e.g., McCarthy et al., 2006, Whitemore et al., 2003) is that they do not include a control group (Erker, Searight et al., 1993). As noted by Milin and associates (2000), this is typically due to the difficulties in establishing a comparable control group and the ethical considerations that preclude “no-treatment” control groups. A waiting list control group would be the obvious choice, but waiting lists can be short for day treatment programs due to the severity level and high needs of their referrals. In one study of Day Treatment outcomes with a wait list control (Grizenko et al., 1993), day treatment produced much greater change in deviant behaviors, self-perception and school reintegration. However, when compared to the control group, there was no change in peer relations or family functioning. In the many studies with no control group, it could be argued that the treatment effects are a result of developmental maturation. In addition, it is not possible to say how much regression to the mean or a placebo effect could account for some of the improvement seen.
The present study investigates the effectiveness of a short-term day treatment program for children with a primary diagnosis of DBD using best-practice treatment strategies. Using a wait-list control, we compared children’s admission and discharge test scores on standardized measures of behaviour and functioning, as rated by the parent.
Outcome data were obtained for 40 children for whom there was pre- and post-treatment symptom measures from their parent. The children attended the Child and Family Day Treatment program between 2002–2005. This sample was comprised of 32 boys (80%) and 8 girls (20%). The mean age of the sample was 10.51 years (SD = 1.86 years), with a range of 5 to 13 years. Twenty-nine of the children had ADHD and comorbid ODD, 9 had ODD only and 2 had ADHD only. Secondary diagnoses were most commonly learning disabilities and borderline intellectual functioning. Fifty percent of the sample (n = 20) attended the treatment program two days a week for an average of 16-weeks (average of 32 days), while the remainder of the sample (50%; n = 20) attended the treatment program beginning four days a week with a transition down to one day a week (average of 40 treatment days). This difference was due to a management led programming change. There was no difference on the measures between the two and four-day a week groups at point of admission. Mothers completed the measures most frequently (88%; n = 35).
A waitlist control group was formed of children who were referred to the Day Treatment program but were currently waiting for space to become available for treatment. The Day Treatment waitlist is not prioritized due to severity, due to the high needs of all children on the list. They are admitted based on referral date. Therefore, children on the wait list would be expected to have the same severity of behaviour as children who are admitted to the program. Seventeen children were in the waitlist control group with measures completed by a parent at the point of referral and 4-months after the point of referral. This sample was comprised of 16 boys (94%). The mean age of the sample was 10.12, (SD = 1.73 years), with a range of 7 to 12.7 years. There were no significant differences between the treatment and waitlist groups on demographic variables.
The Child and Family Day Treatment (CFDT) Service at the IWK Health Centre in Halifax, Nova Scotia provides assessment and management for children with severe DBD, ages five to 12 years old, and their families. These children are not able to manage their daily life in their family, school or community and require more support than can be provided on an outpatient basis, but do not yet require a residential setting. More specifically, this day treatment program is for children who are in need of additional support to manage their behaviour, assess medications, return to full-time school attendance, review diagnoses, and improve self-esteem and social competence. The treatment team consists of Youth Care Workers, a Psychologist, a Psychiatrist, an Occupational Therapist, a Registered Nurse, a Social Worker, and a Teacher.
The treatment program is based on best practice parameters for working with children with DBD. As part of the program, parents are required to be involved in the treatment process: attending parenting groups, completing daily program sheets, and attending scheduled meetings. Behavioural parental strategies are implemented at home and home visits are conducted to determine areas of strength and difficulty. In addition, daily contact between the program and the school occurs, and behavioural strategies are recommended in the schools. The day treatment program employs a Cognitive-Behavioural approach, using a token economy, and skill building groups with a focus on social skills training, anger management, processing of school difficulties, hygiene and relaxation training, to name a few. The parent group also follows a Cognitive-Behavioural model with skills for reinforcing positive behaviours, giving good instructions, applying appropriate consequences for negative behaviours, coping with parenting stress, and having fun with children. Parents are taught about ADHD and techniques for dealing with hyperactive, impulsive and inattentive behaviour. In addition, the team Psychiatrist prescribes or modifies medications when needed. Final DSM diagnosis for all children is made by a joint decision of the team Psychiatrist and Psychologist based on clinical interviews with parent and child, questionnaires and staff observations.
Parents completed identical standardized assessment packages on admission and discharge. All questionnaires were then scored by a research assistant who was blind to the clinical status of the participant and had no clinical contact with the family or the child. This research received ethics approval from the IWK Health Centre Ethics Board and written consent was obtained from parents for participation in the study.
The Child Behaviour Checklist (CBCL) is a highly recognized measure, with well-established psychometric properties, designed to obtain parents’ reports of the child’s current level of functioning (Achenbach, 1991). Social Problems, Aggressive Behaviour, and the Externalizing scales were used.
The Conners’ Parent Rating Scale Revised: Short Form (CPRS-R:S) is a 27-item scale designed to measure levels of oppositional Behaviour, cognitive problems/inattention and hyperactivity (Conners’, 1997). The ADHD scale was used.
The Parenting Stress Index (PSI) is designed to identify parent and child characteristics that contribute to parenting stress (Abidin, 1995). The Child and Parent stress scales were used. The PSI is able to detect changes in the level of stress as a result of intervention making it a good tool of choice for program evaluation (Abidin, 1997).
The Eyberg Child Behaviour Index is designed to assess behaviours associated with conduct disorders in childhood by measuring the number of difficult behaviour problems and the frequency with which they occur (Eyberg and Pincus, 1999). The Intensity scale is a measure of the severity of conduct problems as rated by parents.
Data analyses were designed to assess whether differences in outcomes between the wait list and the treatment group were evident. To test for these treatment effects, a repeated-measures MANOVA was calculated. Further, independent sample t-tests were calculated to compare reports of behavioural functioning at discharge between waitlist and treatment groups. Paired sample t-tests were calculated to examine treatment effects for treatment and wait-list groups separately. Cohen’s d was used to examine both between and within group effect sizes.
All results were analyzed using SPSS for Windows.
Group Comparisons: Test Scores. Means and standard deviations for the groups’ test scores are regrouped in Table 1. Cohen’s d reflects large effect sizes across all measures for the treatment group except for the parent stress measure (d = .39).
Behavioural Functioning. Parents reported on child’s behaviour functioning at admission and discharge. At intake both groups of children (waitlist and treatment groups) had scores in the clinically significant range. At discharge, the children’s scores in the treatment group had moved out of the clinical range. To test for treatment effects, a repeated-measures MANOVA was calculated with group as the independent variable (Treatment vs. Waitlist) and CBCL externalizing total score, aggressive subscale, social problem subscale, Conners’ hyperactivity index, and the Eyberg intensity score as dependent variables. The treatment group was found to have improved significantly more than the waitlist group on the combination of these variables, using Hotelling’s T2 criterion, F = 2.60, df = 5,40, p = .04. Univariate F statistics showed significant treatment effects for each of the variables taken separately, F (externalizing) = 11.91, df = 1,44, p = .001; F (aggression) = 13.88, df = 1,44, p = .001; F (social problems) = 26.35, df = 1,44, p < .001; F (hyperactivity) = 21.90, df = 1,44, p < .001; F (intensity) = 49.57, df = 1,44, p < .001. Group (Treatment vs. Waitlist) by Time (admission vs. discharge) F was significant, Hotelling’s T2 criterion, F = 3.33, p = .013. Univariate F statistics showed significant change for three of the variables between the point of admission to discharge, F (externalizing) = 8.92, df = 1,44, p = .005; F (aggression) = 6.51, df = 1,44, p = .014; F (intensity) = 13.72, df = 1,44, p < .001 and nonsignificant on the remaining two domains of functioning, F (social problems) = 1.17, df = 1,44, p = .286; and F (hyperactivity) = 3.20, df = 1,44, p = .08. Table 2 displays the between group (treatment vs. waitlist) effect sizes. Cohen’s d reflects large effect sizes between the treatment and waitlist groups at the point of discharge for the child’s aggressive, externalizing, and intensity of behaviour measures (d = .79 to 10.01) and small effect on the parents report of stress related to parenting their children (d = .23 to .38). There was no difference in report of children’s behaviour on the social problems scale.
Independent sample t-tests were calculated to compare reports of behavioural functioning at discharge, between the waitlist and treatment groups. The treatment group had significantly lower scores than the waitlist group on the aggressive behaviour measure, t (53) = 2.61, p = .012, the externalizing measure, t (53) = 3.41, p = .001, and the intensity of the behaviour, t (53) = 2.54, p = .014. There were no differences between the waitlist and treatment groups on the social problems, hyperactivity, and parent stress measures (child and parent).
Parent Stress. Parents rated their feeling of stress in relation to parenting their child referred for treatment (child) and stress in relation to their parenting role (parent) at admission and discharge. Paired sample t-tests were calculated to examine treatment effects for the two groups (Treatment and Waitlist) independently. For the treatment group, significant reductions in parent report of child-related stress, t (33) = 5.76, p <.001 and parent stress was reported, t (33) = 2.27, p = .03; however no significant change was found for the waitlist group on either measure. Independent sample t-tests were calculated to compare the two groups (Treatment and Waitlist) at discharge. There were no significant differences in parents’ report of change between the treatment and waitlist groups at discharge (see Table 2).
Results of the present study indicate that, compared with a wait list control group, children with DBD who attend a short-term day treatment program showed significant improvement in their behaviour at home. At the time of discharge, children’s scores on measures of externalizing behaviour and social behaviour were in the non-clinical range. The parent’s level of stress regarding their child was also reduced to non-clinical levels. This study is a significant improvement over the large majority of studies in the day-treatment literature that do not include a control group. When the wait list group and treatment group were directly compared at discharge, the treatment group showed less aggression, externalizing behaviour and behavioral intensity. This is an important finding due to the fact that it is much more difficult to achieve significant improvement in externalizing behaviour when research takes place in a clinical setting, as opposed to a university laboratory setting (Farmer et al., 2002). The large effect sizes in the present study suggest optimism about treating severe DBD in the “real world”.
In the present study, all children had a diagnosis of a DBD, ensuring that the large treatment effects were not due to the inclusion of children with disorders that might be more responsive to treatment. The use of psychometrically sound pre- and posttest measures is also a strength of the current study.
Effective treatments for DBD are important for the child, his or her family, and for society as a whole. The current study lends support to the idea that severe DBD can be treated using multi-modal, intensive and evidence-based treatment techniques. A significant limitation of the current study is that it is not yet clear if the significant behavioural gains will be maintained over time, as the children were not assessed at a post-discharge time point. It will be very important to further investigate long-term effectiveness, due to the fact that children with DBD are at high risk for major adolescent and adult dysfunction (Scott et al., 2001).
In future studies, if long-term improvement is noted, it would suggest that day treatment programs are a less costly treatment option, as compared to residential care, with good clinical outcomes. In addition, the fact that the present program was relatively short-term may lead to the suggestion that longer programs could be shortened, making them more cost effective and allowing for the treatment of a larger number of children and families. This could have a direct impact on health care costs, and the number of children who have to wait for treatment.
A Nova Scotia Health Research Foundation Grant and an IWK Health Centre research grant funded this work. We thank all of the children, parents and staff who participated in this study.