Research in Developmental Disabilities 35 (2014) 2958–2975
Contents lists available at ScienceDirect
Research in Developmental Disabilities
Prevention of behavior problems in a selected population: Stepping Stones Triple P for parents of young children with disabilities Cheri J. Shapiro a,*, Janice Kilburn b, James W. Hardin a a b
University of South Carolina, United States South Carolina First Steps to School Readiness, United States
A R T I C L E I N F O
A B S T R A C T
Article history: Received 25 May 2014 Received in revised form 15 July 2014 Accepted 15 July 2014 Available online
Because young children with disabilities are at elevated risk for development of challenging behaviors, and caregivers of these children typically lack access to evidencebased parenting interventions, two randomized trials were conducted to examine the impact of an evidence-based parenting intervention, Stepping Stones Triple P (SSTP), as a selective preventive intervention. Both studies targeted parents of children under two with a variety of disabilities who were enrolled in the IDEA Part C Early Intervention (EI) system in one state. SSTP was delivered in family homes. In Study One, 49 families were randomly assigned to EI services as usual, with or without SSTP; a 52% attrition rate from treatment was seen. No significant between-group differences were seen aside from a trend toward reduced symptoms of parental depression at follow-up. Intervention group children demonstrated significant decline in behavior problems from post treatment to follow-up, and there was a trend toward improved parenting style in the intervention group during this same time frame. Study Two incorporated a separate workforce intervention for EI service coordinators; 40 families on their caseloads were then randomly assigned to receive EI services as usual with or without SSTP. Attrition from treatment was limited to 20%. No differential impact was seen on child behavior; a trend was noted post-treatment on parent symptoms of depression and on the observed parent– child relationship. At 12-month follow-up, there was a trend favoring improvement in the intervention group in parenting style; statistically significant impact was also seen on the observed quality of the parent–child relationship. SSTP shows promise as a selective preventive intervention for an early intervention population. Reasons for the differential findings between the two studies are explored and suggestions for future research are provided. ß 2014 Elsevier Ltd. All rights reserved.
Keywords: Children Disabilities Parenting Evidence-based parenting intervention Randomized design Home-based intervention
* Corresponding author at: University of South Carolina, Institute for Families in Society, 1600 Hampton Street, Fifth Floor, Columbia, SC 29208, United States. Tel.: +1 803 777 8760. E-mail address: [email protected] (C.J. Shapiro). http://dx.doi.org/10.1016/j.ridd.2014.07.036 0891-4222/ß 2014 Elsevier Ltd. All rights reserved.
C.J. Shapiro et al. / Research in Developmental Disabilities 35 (2014) 2958–2975
2959
1. Introduction Children with disabilities are at increased risk for developing behavior problems as compared to typically developing children (Handen & Gilchrist, 2006; Ozonoff, Goodlin-Jones, & Solomon, 2007; Shapiro, in press; Sofronoff, Jahnel, & Sanders, 2011), especially those with fair or poor health or those with communication difficulties (Scarborough, Hebbeler, Spikes, & Simeonsson, 2007). For example, youth with autism spectrum disorders (ASD) as well as intellectual disabilities (ID) are at higher risk for hyperactivity, conduct, and emotional problems (Totsika, Hastings, Emerson, Lancaster, & Berridge, 2011). Even minor problem behaviors that are present in typically developing children tend to occur more frequently, to a more severe degree, or for a longer length of time among children with developmental disabilities (Sanders, Markie-Dadds, & Turner, 2003). The connection between disabilities and behavior problems has been noted for children as young as age two (Baker, Blacher, Crnic, & Edelbrock, 2002; Emerson & Einfeld, 2010) and can be stable across time (Baker et al., 2002; Herring et al., 2006). Such problem behaviors can negatively impact the child, the family, and the community by increasing parental stress, disrupting parent–child relationships, contributing to family isolation, and possibly placing children at higher risk for maltreatment (Baker et al., 2002; Sanders, Markie-Dadds, et al., 2003). In addition to elevated risk for behavioral problems, children with disabilities may be at increased risk for maltreatment (Kendall-Tackett, Lyon, Taliaferro, & Little, 2005). Some studies have found increased rates of maltreatment among populations of youth with disabilities. Examples include the work of Sullivan and colleagues (Sullivan, 2009; Sullivan & Knutson, 2000) who merged archival data from schools with data from social services, foster care services, and law enforcement victimization databases. Disability status was identified from school records; higher rates of maltreatment were found for disabled students as compared to their nondisabled peers (31% vs. 9%; Sullivan & Knutson, 2000). However, studies vary widely in the level of risk that is conferred due to differing definitions of both disability and maltreatment (Jones et al., 2012; Leeb, Bitsko, Merrick, & Armour, 2012; Sullivan, 2009; Sullivan & Knutson, 2000). Because of this, KendallTackett et al. (2005) recommend that children’s disability status be included in child maltreatment studies. 1.1. Need for prevention and early intervention Given the high likelihood of behavioral problems and possibly higher risk for maltreatment among children with disabilities, prevention and early intervention approaches focusing on this population are needed. Early intervention approaches have demonstrated positive impact on a range of outcomes for children with developmental delays (Love et al., 2005; Mercy & Saul, 2009). However, behavior problems of young children with disabilities have been both under-identified and insufficiently addressed in early intervention programs (Powell, Fixsen, Dunlap, Smith, & Fox, 2007). Specific interventions for motor, language, and cognitive development are seen as more central to early intervention services; yet, despite the need to address behavior problems in young children with disabilities, this domain of functioning is often not given priority (Kaiser, 2007). 1.2. Behavioral family intervention approaches Fortunately, if behavioral functioning is targeted, research has demonstrated that behavior problems can be prevented or effectively treated in families with typically developing children as well as in families with a child with developmental delays or other disabilities (Bor, Sanders, & Markie-Dadds, 2002; Sanders, 2008, 2012; Tellegen & Sanders, 2013). Models based on behavioral family interventions have demonstrated the strongest impact (Cuijpers, 1999; McMahon & Wells, 1989; Taylor & Biglan, 1998). Eyberg, Nelson, & Boggs (2008) reviewed the literature on evidence-based psychosocial treatments for children and adolescents using the criteria developed by the task force on promotion and dissemination of psychological procedures (Chambless et al., 1996, 1998). Treatments classified as meeting the criteria for ‘‘well established’’ had two good between group experiments by different investigators that demonstrated superiority to another treatment. ‘‘Probably efficacious’’ criteria include either two good between group experiments demonstrating superiority to a wait list control group or meeting ‘‘well established’’ criteria but not by two or more different investigators (Chambless et al., 1996, 1998). Of the treatments reviewed by Eyberg et al. (2008), sixteen met the criteria for ‘‘probably efficacious’’ or ‘‘well established’’; five treatments included children four years of age and younger and derive from the behavioral family intervention perspective: Helping the Noncompliant Child (HNC), Parent–Child Interaction Therapy (PCIT), the Incredible Years (IY) Series, Parent Management Training-Oregon Model (PMT-OM), and the Triple P-Positive Parenting Program (Triple P). Of these, three (PCIT, IY, and Triple P) have shown promising effects for families with children with developmental disabilities. PCIT was developed as a parent-training program for children ages three to eight with externalizing behavior problems (Bagner & Eyberg, 2007). Based on social learning and attachment theory, the primary emphasis is on improving the quality of the parent–child relationship and changing parent–child interaction patterns. Medium to large effect sizes have been found in reducing child externalizing behavior and parental stress, and increasing positive observations of parent behavior (Thomas & Zimmer-Gembeck, 2012). Bagner and Eyberg (2007) noted improvements with PCIT for families with 3–6-year old children with mental retardation and comorbid oppositional defiant disorder. In addition, Bagner, Sheinkopf, Vohr, and Lester (2010) found that mothers of young children born prematurely who received PCIT reported significant decreases in child behavior problems and were observed to be more positive in child-led play compared with mothers of similar children in a waitlist control group.
2960
C.J. Shapiro et al. / Research in Developmental Disabilities 35 (2014) 2958–2975
The IY interventions were designed to promote emotional and social competence and to prevent, reduce and treat behavior and emotional problems in young children. IY interventions have been found to be effective in reducing children’s maladaptive behavior and increasing parents’ adaptive parenting skills (e.g., Gardner, Burton, & Klimes, 2006; Marcynyszyn, Maher, & Corwin, 2011; Webster-Stratton, 1994, 2000). Positive effects were found for a group-based parent-training program that aimed to improve the behavior of preschool children with developmental disabilities (McIntyre, 2008). The Triple P-Positive Parenting Program (Triple P; Sanders, 2008) is a multilevel parenting support system (five levels of increasing intensity of intervention) based on public health, community psychology, social learning theories and principles of behavioral family interventions. Triple P interventions have demonstrated efficacy and effectiveness in reducing child behavior problems and improving parenting outcomes in a large number of randomized controlled trials with a wide variety of populations (Sanders, 2012). Sanders, Kirby, Tellegen, and Day (2014) conducted a review and meta-analysis of more than 100 studies examining Triple P interventions; significant positive effects were obtained on child behavioral outcomes for each of the five levels of the intervention system. Positive outcomes at the parent level include improved parenting practices, parental satisfaction and efficacy, parental adjustment, and partner relationships. Stepping Stones Triple P (SSTP; Sanders, Markie-Dadds, et al., 2003) is a variant of Triple P developed specifically for parents of preadolescent children with disabilities. SSTP has been evaluated in randomized controlled trials of young children with comorbid developmental disabilities and behavior problems (Plant & Sanders, 2007; Roberts, Mazzucchelli, Studman, & Sanders, 2006), with parents of children with developmental disabilities only (i.e., not necessarily with identified behavior problems, Sofronoff et al., 2011), as well as with parents of young children of mixed disability types (Roux, Sofronoff, & Sanders, 2013). In a recent meta-analysis of 12 SSTP studies (including nine randomized controlled trials or RCTs), significant positive impact on child behavior (primary outcome of interest) as well as on a range of parent-level (secondary) outcomes were seen, including parenting style, parenting self-efficacy, and parent adjustment (Tellegen & Sanders, 2013). Effect sizes for child behavior were medium and increased with the intensity of the intervention; effect sizes for parenting style were large. Effect sizes for parent personal adjustment were small. Across these studies, child age ranged from 1 to 17; six studies focused on children age 7 or below. At least one study at each of the four intervention levels of Triple P was represented in this meta-analysis (i.e. Levels 2–5; Level 1 is a universal media and information strategy and was not represented). Effect sizes increased as the level of intervention intensity increased; effect sizes were also comparable to those found in prior meta-analyses (de Graaf, Speetjens, Smit, de Wolff, & Tavecchio, 2008a,b). In sum, evidence-based parenting interventions have demonstrated some impact on child behavior and parenting among parents of children with disabilities. However, these interventions are not typically available to parents of children with disabilities in most state early intervention systems, reflecting common research to practice gaps between what works and what is accessible to most parents. Furthermore, the evidence of impact for these interventions has been studied as a treatment for existing behavior problems; to our knowledge, examinations of these evidence-based parenting interventions as prevention strategies for parents of young children with disabilities have not been conducted. 1.3. Need for workforce development in early intervention Importantly, for young children with disabilities, effective parenting interventions alone may be insufficient to address the complex array of factors that can increase the risk for maltreatment. Identifying known risk factors for maltreatment such as parent anger, stress, and family conflict (Stith et al., 2009) requires keen observation and close contact with families; taking effective action to address these issues requires a strong, supportive relationship with parents and families in need. Within the early intervention system, all families have a service coordinator that functions as a single point of contact (www.nichcy.org/babies/overview#what). Because of their key role in supporting families within the early intervention system, this workforce is an ideal recipient of professional development approaches designed to help providers be responsive to the needs of caregivers of young children. One specific curriculum has been created to increase knowledge of protective factors that promote positive, effective working relationships between providers and parents of young children, and to increase knowledge of risk and protective factors for maltreatment. The curriculum is Preventing Child Abuse and Neglect: Parent-Provider Partnerships in Child Care (PCAN; Seibel, Britt, Gillespie, & Parlakian, 2006). This curriculum, originally developed for childcare providers, focuses on protective factors identified from the maltreatment literature, and has been broadly implemented across the United States (http://www.zerotothree.org/about-us/areas-of-expertise/trainingand-professional-development/pcan-curriculum.html). However, to date, there is no research on application or impact of this curriculum with an early intervention workforce. 1.4. Home visitation and parent training While the choice of which interventions to use to support parents is important, selecting a format for intervention delivery is equally important from an implementation perspective. Home visitation as a model of service delivery has long been recognized for its potential to support healthy child development, including competent parenting (Martin, BrooksGunn, Klebanov, Buka, & McCormick, 2008; Olds et al., 1998). Advantages of this mode of program delivery include services occurring in a familiar environment, opportunities to observe family members’ behaviors and interactions in their homes, and reducing barriers regarding transportation for services (Azzi-Lessing, 2011). Optimal outcomes occur when the full
C.J. Shapiro et al. / Research in Developmental Disabilities 35 (2014) 2958–2975
2961
dosage of the program can be administered; i.e., all home visits occur and without families dropping out (Raikes, Love, Kisker, Chazan-Cohen, & Brooks-Gunn, 2004). Flexibility is needed so that number, frequency, duration, and focus of the visits match the needs of the families (Daro, 2006; Roggman, Cook, Peterson, & Raides, 2008). It may be that home visitation is most effective, at least when prevention of child maltreatment is a measured outcome, for low-income, unmarried, first-time mothers (Karoly et al., 1998). Targets for home visitation services typically include parents of young children below school age. Multiple service systems use home visitation strategies to deliver a wide variety of interventions, including schoolreadiness, health, mental health, and early intervention for children with or at risk for disabilities. However, parenting interventions designed specifically to improve child behavior as well as parent functioning are not routinely incorporated into these interventions. There is some preliminary evidence that evidence-based parenting interventions can be effectively implemented in the home. Ware, McNeil, Masse, and Stevens (2008) examined in-home application of PCIT with five families with a child with significant behavior problems. Using an A/B design with staggered baselines, decreases in caregiver negative behavior and reported child behavior problems were observed for the three families that completed treatment. While promising, the small number of families involved limits generalizability of these findings. A second intervention that has shown promise is an adaptation of the Parenting Young Children Program (Fox & Nicholson, 2003). Holtz, Carrasco, Mattek, and Fox (2009) found that this in-home parenting intervention was effective for children ages 1–5 with significant behavior problems both with and without developmental delays; however, the lack of a control group limits interpretation of these findings. Thus, minimal evidence exists examining empirically based parenting interventions using a home visitation method of delivery; the potential impact of such interventions, particularly with parents of very young children with disabilities, is not known. This is an important area for research, as the majority of services provided in the United States for young children with disabilities (ages 0–3) who are eligible for federally mandated early intervention efforts under IDEA Part C occur in natural environments, which is typically the child’s home. 1.5. Focus of the current studies Given the relative paucity of research on the impact of evidence-based parenting interventions for parents of very young children with disabilities, and elevated risk for the development of behavior problems in this population, evaluations of preventive interventions using evidence-based parenting approaches are needed. A second identified area of need is to examine potential impact of prevention focused, evidence-based parenting interventions using a home-based (home visitation) service delivery model. To address these needs, two separate RCTs using SSTP were conducted within the early intervention service system in one state. In addition, one of the two studies also incorporated the PCAN curriculum training as a workforce enhancement strategy for the IDEA Part C service coordinators to assess feasibility and acceptability of this intervention with this professional population. 2. Study One The first study examined the following hypotheses: Will SSTP with IDEA Part C services as usual enhance parent and child functioning and improve parent–child relationships as compared to IDEA Part C services alone? 2.1. Method 2.1.1. Design This study was conducted as a randomized controlled trial with a 2 (group) 3 (time) design. 2.1.2. Participants To be eligible for the study, participants had to be the parents or legal guardians of a child below 24 months of age, with no prior history of maltreatment, who was eligible for participation in the state IDEA Part C early intervention system. Eligibility for participation in the state IDEA Part C program was defined as the child already receiving IDEA Part C services, or a child who had been evaluated for and accepted into IDEA Part C services (even if those services had not yet commenced). In addition to these criteria, participants also had to reside within the seven county region designated for this study, speak English, and have a child whose health was relatively stable and would not likely require extended hospital stays during the time frame for the project. The final study sample consisted of 49 caregivers who met these criteria. The majority of the participants (96%) were women, average age 30.94 years (SD = 8.2), and were either the biological mother, grandmother, or aunt of the child involved in the study. The majority (82%) had a high school education or beyond. Slightly less than half were in paid employment (45%) and approximately half of the sample had annual household incomes over $30,000 (49%). The majority of the caregivers (63%) were Caucasian, 27% classified themselves as African-American, and 10% ‘‘other.’’ Fortythree percent were single. No treatment-comparison group differences were found on these demographic variables. See Table 1 for demographic details for the caregivers in the intervention and comparison groups. A majority of the children (63%) were boys. The mean age of the children was 19 months (SD = 3.37). About half (51%) were Caucasian, with 24.5% African-American and 24.5% ‘‘other.’’ Most (65%) were eligible for IDEA Part C services because of global developmental delay(s), and the other children were eligible due to a diagnosis or condition that elevated future risk
2962
C.J. Shapiro et al. / Research in Developmental Disabilities 35 (2014) 2958–2975
Table 1 Study One caregiver characteristics. Intervention group (n = 25)
Age (years) Female
M
SD
28.77
7.72
Comparison group (n = 24) n
%
M
SD
33.19
8.24
p n
%
24
96
23
96
Education
Contents lists available at ScienceDirect
Research in Developmental Disabilities
Prevention of behavior problems in a selected population: Stepping Stones Triple P for parents of young children with disabilities Cheri J. Shapiro a,*, Janice Kilburn b, James W. Hardin a a b
University of South Carolina, United States South Carolina First Steps to School Readiness, United States
A R T I C L E I N F O
A B S T R A C T
Article history: Received 25 May 2014 Received in revised form 15 July 2014 Accepted 15 July 2014 Available online
Because young children with disabilities are at elevated risk for development of challenging behaviors, and caregivers of these children typically lack access to evidencebased parenting interventions, two randomized trials were conducted to examine the impact of an evidence-based parenting intervention, Stepping Stones Triple P (SSTP), as a selective preventive intervention. Both studies targeted parents of children under two with a variety of disabilities who were enrolled in the IDEA Part C Early Intervention (EI) system in one state. SSTP was delivered in family homes. In Study One, 49 families were randomly assigned to EI services as usual, with or without SSTP; a 52% attrition rate from treatment was seen. No significant between-group differences were seen aside from a trend toward reduced symptoms of parental depression at follow-up. Intervention group children demonstrated significant decline in behavior problems from post treatment to follow-up, and there was a trend toward improved parenting style in the intervention group during this same time frame. Study Two incorporated a separate workforce intervention for EI service coordinators; 40 families on their caseloads were then randomly assigned to receive EI services as usual with or without SSTP. Attrition from treatment was limited to 20%. No differential impact was seen on child behavior; a trend was noted post-treatment on parent symptoms of depression and on the observed parent– child relationship. At 12-month follow-up, there was a trend favoring improvement in the intervention group in parenting style; statistically significant impact was also seen on the observed quality of the parent–child relationship. SSTP shows promise as a selective preventive intervention for an early intervention population. Reasons for the differential findings between the two studies are explored and suggestions for future research are provided. ß 2014 Elsevier Ltd. All rights reserved.
Keywords: Children Disabilities Parenting Evidence-based parenting intervention Randomized design Home-based intervention
* Corresponding author at: University of South Carolina, Institute for Families in Society, 1600 Hampton Street, Fifth Floor, Columbia, SC 29208, United States. Tel.: +1 803 777 8760. E-mail address: [email protected] (C.J. Shapiro). http://dx.doi.org/10.1016/j.ridd.2014.07.036 0891-4222/ß 2014 Elsevier Ltd. All rights reserved.
C.J. Shapiro et al. / Research in Developmental Disabilities 35 (2014) 2958–2975
2959
1. Introduction Children with disabilities are at increased risk for developing behavior problems as compared to typically developing children (Handen & Gilchrist, 2006; Ozonoff, Goodlin-Jones, & Solomon, 2007; Shapiro, in press; Sofronoff, Jahnel, & Sanders, 2011), especially those with fair or poor health or those with communication difficulties (Scarborough, Hebbeler, Spikes, & Simeonsson, 2007). For example, youth with autism spectrum disorders (ASD) as well as intellectual disabilities (ID) are at higher risk for hyperactivity, conduct, and emotional problems (Totsika, Hastings, Emerson, Lancaster, & Berridge, 2011). Even minor problem behaviors that are present in typically developing children tend to occur more frequently, to a more severe degree, or for a longer length of time among children with developmental disabilities (Sanders, Markie-Dadds, & Turner, 2003). The connection between disabilities and behavior problems has been noted for children as young as age two (Baker, Blacher, Crnic, & Edelbrock, 2002; Emerson & Einfeld, 2010) and can be stable across time (Baker et al., 2002; Herring et al., 2006). Such problem behaviors can negatively impact the child, the family, and the community by increasing parental stress, disrupting parent–child relationships, contributing to family isolation, and possibly placing children at higher risk for maltreatment (Baker et al., 2002; Sanders, Markie-Dadds, et al., 2003). In addition to elevated risk for behavioral problems, children with disabilities may be at increased risk for maltreatment (Kendall-Tackett, Lyon, Taliaferro, & Little, 2005). Some studies have found increased rates of maltreatment among populations of youth with disabilities. Examples include the work of Sullivan and colleagues (Sullivan, 2009; Sullivan & Knutson, 2000) who merged archival data from schools with data from social services, foster care services, and law enforcement victimization databases. Disability status was identified from school records; higher rates of maltreatment were found for disabled students as compared to their nondisabled peers (31% vs. 9%; Sullivan & Knutson, 2000). However, studies vary widely in the level of risk that is conferred due to differing definitions of both disability and maltreatment (Jones et al., 2012; Leeb, Bitsko, Merrick, & Armour, 2012; Sullivan, 2009; Sullivan & Knutson, 2000). Because of this, KendallTackett et al. (2005) recommend that children’s disability status be included in child maltreatment studies. 1.1. Need for prevention and early intervention Given the high likelihood of behavioral problems and possibly higher risk for maltreatment among children with disabilities, prevention and early intervention approaches focusing on this population are needed. Early intervention approaches have demonstrated positive impact on a range of outcomes for children with developmental delays (Love et al., 2005; Mercy & Saul, 2009). However, behavior problems of young children with disabilities have been both under-identified and insufficiently addressed in early intervention programs (Powell, Fixsen, Dunlap, Smith, & Fox, 2007). Specific interventions for motor, language, and cognitive development are seen as more central to early intervention services; yet, despite the need to address behavior problems in young children with disabilities, this domain of functioning is often not given priority (Kaiser, 2007). 1.2. Behavioral family intervention approaches Fortunately, if behavioral functioning is targeted, research has demonstrated that behavior problems can be prevented or effectively treated in families with typically developing children as well as in families with a child with developmental delays or other disabilities (Bor, Sanders, & Markie-Dadds, 2002; Sanders, 2008, 2012; Tellegen & Sanders, 2013). Models based on behavioral family interventions have demonstrated the strongest impact (Cuijpers, 1999; McMahon & Wells, 1989; Taylor & Biglan, 1998). Eyberg, Nelson, & Boggs (2008) reviewed the literature on evidence-based psychosocial treatments for children and adolescents using the criteria developed by the task force on promotion and dissemination of psychological procedures (Chambless et al., 1996, 1998). Treatments classified as meeting the criteria for ‘‘well established’’ had two good between group experiments by different investigators that demonstrated superiority to another treatment. ‘‘Probably efficacious’’ criteria include either two good between group experiments demonstrating superiority to a wait list control group or meeting ‘‘well established’’ criteria but not by two or more different investigators (Chambless et al., 1996, 1998). Of the treatments reviewed by Eyberg et al. (2008), sixteen met the criteria for ‘‘probably efficacious’’ or ‘‘well established’’; five treatments included children four years of age and younger and derive from the behavioral family intervention perspective: Helping the Noncompliant Child (HNC), Parent–Child Interaction Therapy (PCIT), the Incredible Years (IY) Series, Parent Management Training-Oregon Model (PMT-OM), and the Triple P-Positive Parenting Program (Triple P). Of these, three (PCIT, IY, and Triple P) have shown promising effects for families with children with developmental disabilities. PCIT was developed as a parent-training program for children ages three to eight with externalizing behavior problems (Bagner & Eyberg, 2007). Based on social learning and attachment theory, the primary emphasis is on improving the quality of the parent–child relationship and changing parent–child interaction patterns. Medium to large effect sizes have been found in reducing child externalizing behavior and parental stress, and increasing positive observations of parent behavior (Thomas & Zimmer-Gembeck, 2012). Bagner and Eyberg (2007) noted improvements with PCIT for families with 3–6-year old children with mental retardation and comorbid oppositional defiant disorder. In addition, Bagner, Sheinkopf, Vohr, and Lester (2010) found that mothers of young children born prematurely who received PCIT reported significant decreases in child behavior problems and were observed to be more positive in child-led play compared with mothers of similar children in a waitlist control group.
2960
C.J. Shapiro et al. / Research in Developmental Disabilities 35 (2014) 2958–2975
The IY interventions were designed to promote emotional and social competence and to prevent, reduce and treat behavior and emotional problems in young children. IY interventions have been found to be effective in reducing children’s maladaptive behavior and increasing parents’ adaptive parenting skills (e.g., Gardner, Burton, & Klimes, 2006; Marcynyszyn, Maher, & Corwin, 2011; Webster-Stratton, 1994, 2000). Positive effects were found for a group-based parent-training program that aimed to improve the behavior of preschool children with developmental disabilities (McIntyre, 2008). The Triple P-Positive Parenting Program (Triple P; Sanders, 2008) is a multilevel parenting support system (five levels of increasing intensity of intervention) based on public health, community psychology, social learning theories and principles of behavioral family interventions. Triple P interventions have demonstrated efficacy and effectiveness in reducing child behavior problems and improving parenting outcomes in a large number of randomized controlled trials with a wide variety of populations (Sanders, 2012). Sanders, Kirby, Tellegen, and Day (2014) conducted a review and meta-analysis of more than 100 studies examining Triple P interventions; significant positive effects were obtained on child behavioral outcomes for each of the five levels of the intervention system. Positive outcomes at the parent level include improved parenting practices, parental satisfaction and efficacy, parental adjustment, and partner relationships. Stepping Stones Triple P (SSTP; Sanders, Markie-Dadds, et al., 2003) is a variant of Triple P developed specifically for parents of preadolescent children with disabilities. SSTP has been evaluated in randomized controlled trials of young children with comorbid developmental disabilities and behavior problems (Plant & Sanders, 2007; Roberts, Mazzucchelli, Studman, & Sanders, 2006), with parents of children with developmental disabilities only (i.e., not necessarily with identified behavior problems, Sofronoff et al., 2011), as well as with parents of young children of mixed disability types (Roux, Sofronoff, & Sanders, 2013). In a recent meta-analysis of 12 SSTP studies (including nine randomized controlled trials or RCTs), significant positive impact on child behavior (primary outcome of interest) as well as on a range of parent-level (secondary) outcomes were seen, including parenting style, parenting self-efficacy, and parent adjustment (Tellegen & Sanders, 2013). Effect sizes for child behavior were medium and increased with the intensity of the intervention; effect sizes for parenting style were large. Effect sizes for parent personal adjustment were small. Across these studies, child age ranged from 1 to 17; six studies focused on children age 7 or below. At least one study at each of the four intervention levels of Triple P was represented in this meta-analysis (i.e. Levels 2–5; Level 1 is a universal media and information strategy and was not represented). Effect sizes increased as the level of intervention intensity increased; effect sizes were also comparable to those found in prior meta-analyses (de Graaf, Speetjens, Smit, de Wolff, & Tavecchio, 2008a,b). In sum, evidence-based parenting interventions have demonstrated some impact on child behavior and parenting among parents of children with disabilities. However, these interventions are not typically available to parents of children with disabilities in most state early intervention systems, reflecting common research to practice gaps between what works and what is accessible to most parents. Furthermore, the evidence of impact for these interventions has been studied as a treatment for existing behavior problems; to our knowledge, examinations of these evidence-based parenting interventions as prevention strategies for parents of young children with disabilities have not been conducted. 1.3. Need for workforce development in early intervention Importantly, for young children with disabilities, effective parenting interventions alone may be insufficient to address the complex array of factors that can increase the risk for maltreatment. Identifying known risk factors for maltreatment such as parent anger, stress, and family conflict (Stith et al., 2009) requires keen observation and close contact with families; taking effective action to address these issues requires a strong, supportive relationship with parents and families in need. Within the early intervention system, all families have a service coordinator that functions as a single point of contact (www.nichcy.org/babies/overview#what). Because of their key role in supporting families within the early intervention system, this workforce is an ideal recipient of professional development approaches designed to help providers be responsive to the needs of caregivers of young children. One specific curriculum has been created to increase knowledge of protective factors that promote positive, effective working relationships between providers and parents of young children, and to increase knowledge of risk and protective factors for maltreatment. The curriculum is Preventing Child Abuse and Neglect: Parent-Provider Partnerships in Child Care (PCAN; Seibel, Britt, Gillespie, & Parlakian, 2006). This curriculum, originally developed for childcare providers, focuses on protective factors identified from the maltreatment literature, and has been broadly implemented across the United States (http://www.zerotothree.org/about-us/areas-of-expertise/trainingand-professional-development/pcan-curriculum.html). However, to date, there is no research on application or impact of this curriculum with an early intervention workforce. 1.4. Home visitation and parent training While the choice of which interventions to use to support parents is important, selecting a format for intervention delivery is equally important from an implementation perspective. Home visitation as a model of service delivery has long been recognized for its potential to support healthy child development, including competent parenting (Martin, BrooksGunn, Klebanov, Buka, & McCormick, 2008; Olds et al., 1998). Advantages of this mode of program delivery include services occurring in a familiar environment, opportunities to observe family members’ behaviors and interactions in their homes, and reducing barriers regarding transportation for services (Azzi-Lessing, 2011). Optimal outcomes occur when the full
C.J. Shapiro et al. / Research in Developmental Disabilities 35 (2014) 2958–2975
2961
dosage of the program can be administered; i.e., all home visits occur and without families dropping out (Raikes, Love, Kisker, Chazan-Cohen, & Brooks-Gunn, 2004). Flexibility is needed so that number, frequency, duration, and focus of the visits match the needs of the families (Daro, 2006; Roggman, Cook, Peterson, & Raides, 2008). It may be that home visitation is most effective, at least when prevention of child maltreatment is a measured outcome, for low-income, unmarried, first-time mothers (Karoly et al., 1998). Targets for home visitation services typically include parents of young children below school age. Multiple service systems use home visitation strategies to deliver a wide variety of interventions, including schoolreadiness, health, mental health, and early intervention for children with or at risk for disabilities. However, parenting interventions designed specifically to improve child behavior as well as parent functioning are not routinely incorporated into these interventions. There is some preliminary evidence that evidence-based parenting interventions can be effectively implemented in the home. Ware, McNeil, Masse, and Stevens (2008) examined in-home application of PCIT with five families with a child with significant behavior problems. Using an A/B design with staggered baselines, decreases in caregiver negative behavior and reported child behavior problems were observed for the three families that completed treatment. While promising, the small number of families involved limits generalizability of these findings. A second intervention that has shown promise is an adaptation of the Parenting Young Children Program (Fox & Nicholson, 2003). Holtz, Carrasco, Mattek, and Fox (2009) found that this in-home parenting intervention was effective for children ages 1–5 with significant behavior problems both with and without developmental delays; however, the lack of a control group limits interpretation of these findings. Thus, minimal evidence exists examining empirically based parenting interventions using a home visitation method of delivery; the potential impact of such interventions, particularly with parents of very young children with disabilities, is not known. This is an important area for research, as the majority of services provided in the United States for young children with disabilities (ages 0–3) who are eligible for federally mandated early intervention efforts under IDEA Part C occur in natural environments, which is typically the child’s home. 1.5. Focus of the current studies Given the relative paucity of research on the impact of evidence-based parenting interventions for parents of very young children with disabilities, and elevated risk for the development of behavior problems in this population, evaluations of preventive interventions using evidence-based parenting approaches are needed. A second identified area of need is to examine potential impact of prevention focused, evidence-based parenting interventions using a home-based (home visitation) service delivery model. To address these needs, two separate RCTs using SSTP were conducted within the early intervention service system in one state. In addition, one of the two studies also incorporated the PCAN curriculum training as a workforce enhancement strategy for the IDEA Part C service coordinators to assess feasibility and acceptability of this intervention with this professional population. 2. Study One The first study examined the following hypotheses: Will SSTP with IDEA Part C services as usual enhance parent and child functioning and improve parent–child relationships as compared to IDEA Part C services alone? 2.1. Method 2.1.1. Design This study was conducted as a randomized controlled trial with a 2 (group) 3 (time) design. 2.1.2. Participants To be eligible for the study, participants had to be the parents or legal guardians of a child below 24 months of age, with no prior history of maltreatment, who was eligible for participation in the state IDEA Part C early intervention system. Eligibility for participation in the state IDEA Part C program was defined as the child already receiving IDEA Part C services, or a child who had been evaluated for and accepted into IDEA Part C services (even if those services had not yet commenced). In addition to these criteria, participants also had to reside within the seven county region designated for this study, speak English, and have a child whose health was relatively stable and would not likely require extended hospital stays during the time frame for the project. The final study sample consisted of 49 caregivers who met these criteria. The majority of the participants (96%) were women, average age 30.94 years (SD = 8.2), and were either the biological mother, grandmother, or aunt of the child involved in the study. The majority (82%) had a high school education or beyond. Slightly less than half were in paid employment (45%) and approximately half of the sample had annual household incomes over $30,000 (49%). The majority of the caregivers (63%) were Caucasian, 27% classified themselves as African-American, and 10% ‘‘other.’’ Fortythree percent were single. No treatment-comparison group differences were found on these demographic variables. See Table 1 for demographic details for the caregivers in the intervention and comparison groups. A majority of the children (63%) were boys. The mean age of the children was 19 months (SD = 3.37). About half (51%) were Caucasian, with 24.5% African-American and 24.5% ‘‘other.’’ Most (65%) were eligible for IDEA Part C services because of global developmental delay(s), and the other children were eligible due to a diagnosis or condition that elevated future risk
2962
C.J. Shapiro et al. / Research in Developmental Disabilities 35 (2014) 2958–2975
Table 1 Study One caregiver characteristics. Intervention group (n = 25)
Age (years) Female
M
SD
28.77
7.72
Comparison group (n = 24) n
%
M
SD
33.19
8.24
p n
%
24
96
23
96
Education
