Parenting Intervention for Prevention of Behavioral Problems in Elementary School-Age Filipino-American Children: A Pilot Study in Churches

Abstract

Objective

This study aims to test an evidence-based parenting program offered in churches among Filipino-American parents and estimate effect size for a fully powered trial.

Methods

Twenty-two parents of children aged 6 to 12 years were randomly assigned to either an intervention or a waiting-list control group. Parents’ perceptions of child behavior, parenting practices, and parenting stress were obtained at baseline. Parents in the experimental group attended The Incredible Years School Age Program, which consisted of 12 weekly 2-hour sessions. A follow-up assessment was performed after the intervention and 12 weeks later. The intervention was subsequently repeated with the control group. Satisfaction was assessed with a 40-item measure. Analysis of covariance was used to compare the intervention group postintervention versus the control group. Paired t-tests compared mean parenting practices, parenting stress, and child behavior outcomes. Satisfaction was assessed descriptively.

Results

Twenty-two parents completed all assessments and the intervention. Analysis of variance comparing intervention and control groups with repeated measures (pre- and post-test measures) revealed that the program has a positive impact on parenting stress, parenting practices (physical punishment, positive verbal discipline), and parent’s perception of their child’s behavior (internalizing symptoms, externalizing symptoms, and number of problematic behaviors). Analyses of all participants comparing pre- and post intervention revealed improvements in parenting stress, positive verbal discipline, and child externalizing and total problem behaviors. Families reported high satisfaction with the content and format of the intervention.

Conclusion

Results support the feasibility of providing an evidence-based parenting program to Filipino parents in churches to prevent future behavioral health problems.

Filipinos are the second largest Asian subgroup in the United States with the largest number living in Los Angeles County.¹ Compared with whites and other Asian subgroups, Filipino youth have a disproportionately heavy burden of behavioral health problems, including higher depressive symptoms, suicidal ideation, and substance use.^2,3 Filipino youth also have significant mental health risk factors, including parents with high levels of unmet mental health needs and exposure to harsh discipline.^4,5 Despite these behavioral health challenges, Filipino youth have low rates of mental health care and preventive care utilization.^3,6,7 Given the disparity between mental health needs and service utilization among Filipinos, research aimed at describing factors that influence help-seeking is growing.⁸

Cultural mistrust may significantly limit mental health help-seeking attitudes among Filipinos, contributing to a mental health services utilization rate that is approximately one-third of the expected utilization.^8,9 Other factors contributing to underutilization include stigma,⁴ avoidance of shame to self or family; and reliance on religious and indigenous folk healing resources.¹⁰ An estimated 65% of Filipinos in the United States are Catholic,¹¹ making collaboration with this faith community a potentially effective approach to decreasing cultural mistrust and increasing utilization of preventive mental health interventions.¹²

The prevention of behavioral health problems in youth is a national priority.¹³ Children’s challenging behaviors can elicit coercive, increasingly harsh, or avoidant parenting responses,^14,15 which inadvertently can reinforce the problem behaviors in children that the parents are trying to decrease. This dynamic is also bidirectional in that coercive parenting can elicit problem behaviors in children. Parent training programs have been shown to alter parents’ behavior and presumably in response, children’s behavior.^16,17 Although efficacious parent training programs exist, participation is typically low, especially among low-income urban minority populations.¹⁸

The Incredible Years Parent Training Program (IYP)¹⁹ is a social cognitive theory–based intervention that is designated as a Blueprints Model Program for its demonstrated effectiveness in preventing early-onset conduct problems.²⁰ A large scale meta-analysis of behavioral and cognitive-behavioral group parenting programs for children aged 3 to 12 years concluded that there is strong evidence that such programs are cost-effective interventions for improving parenting practices, parental mental health, and significant child conduct problems.²¹ IYP was used in 9 of the 13 studies that met the stringent criteria to be selected for this meta-analysis.

Among Asian-American groups, a study of IYP with Korean mothers²² divided in 2 subgroups by level of acculturation resulted in increases in positive discipline among both subgroups and a decrease in harsh discipline in the less acculturated group. Employing a wait list randomized control design, a study with a referred high-risk sample of Chinese immigrant parents provided preliminary evidence of efficacy in reducing clinical levels of child behavior problems.²³ Our decision to use IYP in this study is based on its demonstrated prevention, treatment, and cost effectiveness, as well as its cultural adaptability. The aim of this study was to conduct a preliminary assessment of the efficacy and feasibility of IYP when offered in a community faith-based setting to Filipino parents.

METHODS

We adopted the Evidence-Based Community/Partnership Model proposed by Wells and colleagues as an approach to bringing evidence-based interventions to a community setting.²⁴ The adopted approach included a needs assessment conducted from 2009 to 2010, which identified that offering an evidence-based parenting program as a prevention program in churches would be a culturally congruent strategy to prevent Filipino adolescent mental health disparities.¹² Then, negotiation among community stakeholders representative of the target population (i.e., pastor and associate pastor of partnering churches, catechism coordinator, church parishioners who were also parents and grandparents, and Filipino community–based organization leaders) and researchers (i.e., first and second authors) took place to arrive at a set of shared goals that included linkage with an evidence-based solution as a first step. This process of negotiation was conducted by the first author through the use of a community forum to increase awareness regarding the problem and a series of meetings with key stakeholders within 2 predominantly Filipino churches. Strategies were developed to reduce stigma and reluctance to participate in research; and to overcome barriers to making time available to participate in a 12-week parenting group.

A community partnership was formed through the creation of a community advisory board (CAB), led by the Filipino associate pastor of 2 Catholic churches in Los Angeles and the first author, a Filipino pediatrician. During several planning meetings, the CAB provided feedback regarding recruitment method and what aspects of the parent training intervention should be highlighted during recruitment. For example, the CAB recommended that the first author make announcements about the study at the end of mass and that the program should be described to parents as a prevention program aimed at promoting the social and academic success of their children. A plan was developed to offer the parent training group as a workshop that would take place concurrently with catechism classes on Saturday mornings. Overall, this community engagement process spanned 3 years from a needs assessment in 2009 to obtaining funding for the current project in 2011.

Participants

Based on CAB feedback, the study was presented to parents as a prevention program called The Filipino Family Initiative to reduce the anticipated cultural barriers to participation that might be associated with having a child with behavior problems or attending “parent training” that might imply something wrong with the participants’ parenting.²⁵ Parents were recruited from September 2011 to January 2012 from two partnering churches, including parents of children already enrolled in catechism, its associated school, and community-based organizations using announcements by the first author after masses, letters of invitation, fliers, and word-of-mouth.

Figure 1 summarizes the enrollment, randomization, follow-up, and data analysis for all study participants. Thirty of 60 parents who were eligible agreed to participate, resulting in a 50% refusal rate. Two parents had to discontinue the study because they were not available to participate in the workshop at the time offered. Twenty-eight parents were randomly assigned. During the course of the parent training groups, 6 parents discontinued participation, yielding a retention rate of 79%. The total number of parents completing the groups and the total analyzed was 22, including 18 mothers, 2 fathers, and 2 grandmothers who were legal guardians.

Figure 1.

Enrollment, randomization, follow-up, and data analysis.

Measures

Participants completed a demographic questionnaire, assessing family structure, income, and education levels, and some health information about the child and family. They also completed the Child Behavior Checklist for Ages 6 to 18 (CBCL)²⁶; the Parenting Stress Index-Short Form (PSI-SF)²⁷; The Parenting Practices Interview (PPI)²⁸; and the Parent Satisfaction Questionnaire (PSQ).²⁹ The CBCL is a commonly used 118-item parent questionnaire for rating emotional and behavior problems in school age children and was administered by computer pre- and postintervention. Internal consistency has been excellent for the internalizing, externalizing, and total problems subscales (Cronbach’s alpha ranging from 0.90 to 0.97 in normed samples).²⁶ The PSI-SF²⁷ is a 36-item questionnaire that measures parenting stress related to parent-child problem areas in parents of children aged 1 month to 12 years. The Short Form version used yields a Total Stress Score and has 3 subscales. The Cronbach’s alpha for this study sample is 0.891 for the Total Stress Score, 0.838 for the parental distress score, 0.777 for the parent-child dysfunctional interaction score, and 0.880 for the difficult child score. The PPI is a 73-item questionnaire adapted from the Oregon Social Learning Center’s Discipline questionnaire.²⁸ This study focused on 3 subscales: appropriate discipline (e.g., use of time out for aggression); positive verbal discipline (e.g., use of praise to reinforce positive behaviors); and physical punishment (e.g., spanking). The Cronbach’s alpha for this study sample is 0.71 for appropriate discipline, 0.72 for positive verbal discipline, and 0.80 for physical punishment. The PSQ is a 40-item Likert type scale that measures parent satisfaction regarding the overall program, teaching format, specific parenting techniques, evaluation of group leaders, and feelings about the parent group.²⁹ This measure was created by the developer of the Incredible Years and has been used in multiple previous studies evaluating the program.^22,30 The Cronbach’s alpha for this study sample is 0.955. The above measures were the original assessments and have been validated among multiple US ethnic groups.

Research Design

A randomized controlled trial design was used to provide data about the efficacy and feasibility of using the Incredible Years Parent Training Program with Filipino parents in a faith-based setting. Our a priori power analysis suggested a sample of 36 participants based on our goals to understand the strength of relationships between intervention group and outcomes and to provide preliminary estimates of effect sizes and variance of the intervention for further study with this population. We were able to recruit 28 participants during the study period, and 22 of these participants completed the study.

Procedures

All the procedures were reviewed and approved by the Institutional Review Board at our institution. Informed consent was obtained from all study participants. Once all parents completed the pre-intervention assessment, the families were randomly assigned to the intervention group (Incredible Years Parent Training Program (IYP) immediately) or the control group (3-month wait-list), using a computer-generated randomization list performed by author (SS) and 1:1 allocation ratio. Baseline assessments were conducted by the first author who was blind to condition because randomization occurred after the baseline assessment. Time 2 and 3 assessments were conducted by trained research assistants who were blind to the study condition and independent of the intervention team. The IYP was offered in 2 waves in the winter/early spring and summer of 2012. Each wave consisted of two 12-week groups meeting on Saturdays; 1 in the morning and 1 in the afternoon. Parents missing a group were offered make-up sessions before the start of the group the following week or at another time during the week. Make-up sessions consisted of viewing 2 to 3 key video vignettes, a review of the parenting principles developed during the missed session, and at least 1 role play of the skills illustrated. After the 12-week intervention, focus groups to determine acceptability of the program were conducted; the methods and results are described in a separate article.²⁵ Participating parents attended the 12-week parent training workshop for 2 hours each week. The workshops were facilitated by the first author who is a pediatrician and the second author who is a senior psychologist and certified Incredible Years agency mentor both of whom completed the 3-day authorized training in the intervention. Also in attendance were research assistants whose first language is Tagalog. They served to interpret for the parents when needed. To maintain fidelity when implementing the program, after each session group leaders completed a collaborative process checklist to ensure features that make the program effective and a checklist for each week that includes key content that should be covered.

The parent training workshop consisted of the Incredible Years School Age BASIC Parent Program³¹ for parents of children ages 6 to 12 years (See Supplemental Digital Content 1, Table S1, http://links.lww.com/JDBP/A113 for a list of topics). The social cognitive theory–based program shows vignettes of real parents from diverse backgrounds modeling parenting skills in effective and ineffective ways. IYP is designed to provide a culturally sensitive structure through principles embedded in the program such as respecting and affirming cultural differences (e.g., group determining its own “ground rules”), enhancing parent engagement by exploring, understanding, and addressing possible cultural barriers to intervention content (e.g., barriers to praise), and helping parents apply strategies to their goals. The parents in the group are treated as the “experts” in their children’s behavior, whereas the group leaders, through leading the discussion with key questions and brainstorming the benefits and barriers of specific techniques, help the parents develop principles that the parents can use to achieve their self-identified short- and long-term goals for themselves and their children. These principles are restated in terms of the benefit for the child. Parents then practice the techniques related to the principles in the role of the parent or the child. Group leaders coach the role plays to be fun experiences for the participants and then debrief the group about what the participants did well and how it felt to play the role of the parent or the child.

Through this process, the parents are able to develop empathy for their children and a self-reflective stance regarding their own parenting, as well as confidence to go home and practice the skills with their own children. The process also affords parents the opportunity to openly discuss their own cultural values related to a particular skill and brainstorm with the group how to implement the skill in a culturally congruent way that helps them achieve their goals for their children. For example, the program’s focus on play and using this special time to model behavior and improve the parent-child relationship serves as an opportunity to pass on cultural values of respeto, or respect and kapwa, a sense of togetherness or connectedness to each other.

Description of Assessment Protocol

Participants completed all outcome measures 3 times (Time I, II, and III). Participants completed the assessments in person at a convenient location using computer-assisted survey software. Although there were differences in time to completion based on computer literacy, trained research assistants were always available to assist with the computer, and thus we were able to capture the data needed in the study. The preintervention assessment, which included demographic variables took about 60 minutes to complete and the postintervention assessment took 45 minutes to complete. All the measures were administered at baseline, 3 months, and 6 months. Participants received $50 for each assessment completed and $20 for their participation in a focus group. They also received up to $100 for their participation in the Incredible Years. Of note, we implemented a standard operating procedure for positive responses to survey items that may raise concern about abuse or neglect (i.e., have you hit or slapped your child in the past? Has your child been left alone without adult supervision?). If a parent answered yes to any of the items, subsequent questions were asked to determine whether a report to Child Protective Services was warranted. None of the participants required such a report. Quality assurance of data entry procedures involved double entry for manual entry, typical data cleaning and procedures, and screening for data out of range and missingness.

Statistical Analysis

The data were analyzed using SPSS 22.³² All the variables were normally distributed with the exception of the Child Behavior Checklist (CBCL) (internalizing, externalizing, and total problems subscales). Owing to non-normal distribution, nonparametric tests were conducted for these variables. First, to describe the treatment effect, we pooled the intervention and control groups and conducted pre-post paired t tests and Wilcoxon signed-rank tests and calculated effect size using Cohen’s d.³³ Because 6 parents withdrew from the study before completion, analyses were conducted to maximize the sample size for the immediate postintervention analysis.

Next, differences in study variables between the intervention and control group were compared using 2-group analysis of covariance, in which pre-intervention scores serve as covariates to control for baseline functioning. The primary hypotheses were that the intervention group would improve more than the control group on 4 primary outcome variables (e.g., parenting stress, appropriate discipline, positive verbal discipline, and physical punishment). Three secondary child behavior outcome variables included the internalizing, externalizing, and total problems subscales on the CBCL.

RESULTS

Preintervention Group Equivalence and Descriptive Data

Table 1 presents demographic variables for Groups A and B. Parents were predominantly first-generation immigrants who were born in the Philippines (86%), and nearly all the identified children in the study (91%) were born in the United States.

Table 1.

Summary of Baseline Characteristics

	Intervention (n = 12)			Wait-list Control (n = 10)
Variable	n	Mean or %	SD	n	Mean or %	SD
Parents sex
Male	1	8.3		1	10
Female	11	91.7		9	90
Parent’s age, yr		41.33	5.85		47.30	12.28
Child’s age at intake, yr
6	2	16.7		1	10
7	2	16.7		1	10
8	3	25		4	40
9	3	25		1	10
10	1	8.3		0	0
11	1	8.3		2	20
12	0	0		1	10
Child’s sex
Female	7	58.3		5	50
Male	5	41.7		5	50
Parent’s education
Grade 0–8	1	8.3		0	0
Grade 9–12	1	8.3		0	0
Some college	2	16.7		1	10
College graduate	7	58.3		8	80
Postcollege graduate	1	8.3		1	1
Birthplace of parent
Philippines	10	83		9	90
United States	2	16.7		1	10
Birthplace of child
Philippines	0	0		1	10
United States	12	100		9	90

There were no significant differences in above variables between Intervention and Wait-list control groups.

Pre-post Comparison for All Participants Using Paired t Test

Table 2 shows pre-post comparisons for all participants using paired t tests results. After the parenting program, parents had significant improvements in parenting stress, increased use of positive verbal discipline, decreased physical punishment, and decreased child externalizing and total child problem behaviors. Non-parametric analysis methods (e.g., Wilcoxon signed-rank tests) were used because of non-normal distribution of Child Behavior Checklist variables.

Table 2.

Mean or Medians and SDs for Parent and Child Outcomes

Scale	Pre Mean/ Median	SD	Post Mean/Median	SD	Significance Test	Effect Size
Parental Stress Index	60.27	14.9	51.14	11.32	t(21) = 2.855**	−0.615
Appropriate discipline	4.66	0.88	4.94	0.69	t(21) = −.746	0.162
Positive verbal discipline	5.55	0.69	5.97	0.70	t(21) = 2.554*	0.540
Physical punishment	1.40	0.42	1.17	0.43	t(21) = 2.188*	−0.469
CBCL internalizinga	3.96	3.77	2.0	3.52	z = −1.740	−0.648
CBCL externalizinga	3.62	3.24	1.0	4.62	z = −2.899**	−1.002
CBCL total problemsa	16.12	10.23	9.0	15.21	z = −2.397*	−0.807

Boldface indicates statistical significance (*p < .05, **p <.01).

^aBecause of non-normal distribution, Wilcoxon signed-rank tests were conducted for these variables. CBCL, Child Behavior Checklist.

Pre and Post Between-Group Hypothesis Testing Using Analysis of Covariance

Table S2 (Supplemental Digital Content 2, http://links.lww.com/JDBP/A114) shows that at Time II (Group A treated and Group B not treated), analyses indicated that Group A and B differed significantly on 3 of the 4 Parenting Stress Index subscales: total score F_(1,21) = 4.99, p = .037, parent-child dysfunctional interaction, F_(1,21) = 5.19, p = .033, and difficult child, F_(1,21) = 9.19, p = .007, and one of Parenting Practice Index subscales: physical punishment, F_(1,21) = 4.808, p = .040. As expected, at Time III after both groups were treated, Groups A and B were no longer statistically different on these variables. The remaining hypotheses were not supported (e.g., no group differences on appropriate discipline, positive verbal discipline, internalizing and externalizing behavior, and total problems). At Time II, the effect size for each outcome are as follows: total parenting stress [Partial Eta² = 0.192 (medium)], parental distress [Partial Eta² = 0.007 (small)], parent-child dysfunctional interaction [Partial Eta² = 0.198 (medium)], difficult child [Partial Eta² = 0.315 (large)], appropriate discipline [Partial Eta² = 0.007 (small)], positive verbal discipline [Partial Eta² = 0.139 (medium)], physical punishment [Partial Eta² = 0.186 (medium)], internalizing behaviors [Partial Eta² = 0.115 (medium)], externalizing behaviors [Partial Eta² = 0.088 (small)], and total problem behaviors [Partial Eta² = 0.178 (medium)].

Changes in Use of Positive Verbal Discipline, Appropriate Discipline, Parenting Stress, Use of Physical Punishment, and Child Behavior Problems

Figure 2 presents a graphical representation of the changes in parenting stress (i.e., total score, difficult child, and parent-child dysfunction) and physical punishment, for Groups A and B at Times I, II, and III. In summary, parenting stress (including total score, parent-child dysfunctional interaction, and difficult child) and physical punishment showed significant changes in the predicted direction. Follow-up assessment showed that the treatment effects noted for Group A children continued to be maintained. For the parenting stress variables, there was a parallel change in Group A when it was treated from Time I to Time II and with group B from Time II to Time III after receiving the intervention. For physical punishment, Group B seems to have improved even before they received the intervention.

Figure 2.

Changes in parenting practices and child behaviors and parental stress for Groups A and B. For all variables p , 0.05, Group A post-treatment Time II versus Group B pretreatment Time II (analysis of covariance). PSI, Parenting Stress Index; PPI, Parenting Practices Interview.

Parent Satisfaction with the Program

Parents reported very high satisfaction with the content and format of the parenting program (means ranged from 5.73 to 6.95 of 7), and 81% of parents reported interest in continuing to meet as a group after the intervention ended. Sample responses from an open-ended question on the Parent Satisfaction Questionnaire, “What do you see as the main benefit of the Incredible Years program?” include: “It teaches us how we can do things better without having to do trial and error. I think this program is a must for all parents”, “What benefited me in this program is having a good relationship not only with my son but also with my husband. This program made my family grow closer and more appreciative of each other, more expressive of our love for each other,” and “It allowed me to express my concerns and worries as a parent and it gave me the opportunity to be the best that I could by learning a new effective approach. It also made me realize that being a parent is a big rewarding opportunity.”

DISCUSSION

This preliminary study demonstrates that the Incredible Years Parenting Program was effective in decreasing self-reported use of physical punishment and decreasing parenting stress among Filipino American parents in this study sample. Intervention group parents continued to use less physical punishment and report decreased parenting stress 12 weeks after completing the program. These results are consistent with previous studies of other minority populations using the Incredible Years Parent Training Program (IYP).²² Clinically, decreases in physical punishment will lead to a reduction in coercive parenting that is necessary to reduce the coercive family dynamics that maintain and worsen problem behaviors in children. Furthermore, a parent who has less stress is more likely to be able to reflect on his/her parenting practice and be more positive, which also reduces the pattern of coercive family dynamics and improves the parent-child relationship, a protective factor for later behavioral problems.

There were no group differences in other study variables (i.e., appropriate discipline, positive verbal discipline, internalizing behaviors, externalizing behaviors, and total problems). However, because of our small sample size, strong conclusions cannot be reached concerning lack of treatment effect. To further describe the treatment effect in this study, pre-post paired t tests of the outcomes were performed and showed treatment effects in the predicted direction for the following variables: parenting stress (total), positive verbal discipline, physical punishment, externalizing behaviors, and total problem behaviors. No treatment effect was found for appropriate discipline. This may be because certain discipline strategies are novel to Filipino parents, similar to studies with Korean American mothers.²² The trend for parental report of decreasing internalizing behaviors was in the right direction but not significant (p = .082). This may also be because social and emotional coaching are new strategies to Filipino American parents.

Interestingly, both the control and experimental group parents reported less physical punishment at Time II, after the experimental group completed IYP. This may be due to 2 factors. First, there may have been a Hawthorne effect as a result of taking part in the study. Parents may have taken more interest in their children’s behavior than previously as a result of attention received from researchers during participation. Thus, they may have been more likely to seek information about positive parenting widely available in the media/internet compared with before their participation in the study. Second, this improvement may be related to a measurement issue. Given that additional questions were asked if parents answered positively on items about physical punishment during the baseline assessment, they may have been less likely to report use of physical punishment in subsequent assessments. This may be related to the fact that this study used self-report measurement as opposed to home observation data. Future studies should examine how to minimize this measurement issue.

The study results suggest that partnering with churches to offer a parenting program is a potentially useful and acceptable intervention for Filipino immigrant parents to learn positive parenting techniques. This is especially important given the high amounts of harsh discipline reported in the Philippines and the need to prevent children’s behavior problems and promote positive parent-child relationships.

Limitations

This study had several limitations. First, low sample size alone may explain the lack of significant treatment effect, as well as inflate the magnitude of the treatment effect for some outcomes. Second, this study relied on parents’ self-reports, which are less objective assessment of parenting behavior compared with behavioral observations. Third, because volunteers from our study were recruited from Catholic churches and schools, our findings may not be generalizable to the broader Filipino population in the United States. Fourth, the high refusal rate limits the generalizability of the findings to similar groups of parents with high motivation to participate in research and a parenting program. In our study, parents who participated in all data collection also attended at least 75% of the sessions. Less motivated parents may not seek a parenting program and may not have parenting or behavioral changes similar to this study once enrolled in the program. Nonetheless, the sample studied is probably representative of the increasing numbers of motivated Filipino parents in the United States (69%) who feel that being a good parent is one of the most important things in their lives.³⁴ Next, the effect size of our intervention may be over-estimated because we were only able to include participants who actually attended the program in our analyses. Finally, we were unable to calculate internal consistency for the Child Behavior Checklist (CBCL) because the software purchased from the survey’s developers does not provide raw item-level data. Although the CBCL has been shown to be highly reliable in previous studies with normed samples, future studies should examine consistency within the Filipino population.

CONCLUSION

This study contributes initial data about the efficacy, acceptability, and feasibility of the Incredible Years School Age Program among Filipino American parents in a community-based setting. Despite the promising nature of our results, our findings are preliminary. Based on the estimated effect sizes, we would need a sample of 240 parents for a future randomized controlled trial. Future research should be conducted with this larger sample size and observational data. Also, offering the program to Tagalog speakers would improve the generalizability of our findings to the broader Filipino population. Similar to previous studies involving parenting programs, more mothers participated in Incredible Years Parent Training Program than fathers. Given that both parents discipline children at home, future studies should focus on increasing participation of fathers to maximize behavioral outcomes of youth. Although the project took place in a real-world setting, several factors maximized the program’s success including a certified Incredible Years peer coach on our study team, all the preparation for the weekly group meetings were performed by research staff, and the provision of stipends to parents for participation in the program. Future studies should examine the dissemination and implementation of Incredible Years without such supports and instead with more community organizational involvement to assess long-term sustainability of the Incredible Years in community-based settings.