Early Steps

Early Family-Centered Prevention of Drug Use Risk
Stage 1 funding period: September 30, 2002–August 31, 2007
Current funding period: March 15, 2008–June 1, 2013
Principal Investigator, Child and Family Center: Dr. Thomas Dishion
Principal Investigator, University of Pittsburgh: Dr. Daniel Shaw; co-PIs include Dr. Anne Gill.
Principal Investigator, University of Virginia: Dr. Melvin Wilson; co-PI Dr. Frances Gardner, Oxford University
Funded by: National Institute on Drug Abuse, National Institutes of Health

The Early Steps Project is a multisite prevention study that evaluates the EcoFIT model in early childhood to reduce risk for a developmental trajectory leading to early-onset substance use/abuse and associated antisocial behaviors.

Early Steps, Stage 1

The first stage of the Early Steps Project began in 2002 and was funded for five years. A total of 731 families with a 2-year-old child were recruited at WIC offices in three states. Each family was visited in their home and asked to complete a set of activities and questionnaires when their child was 2, 3, 4, and 5 years old. The purpose of the study is to understand the contribution of the family to early social and emotional development and to evaluate the benefits of our family-centered intervention (EcoFIT) embedded in WIC service settings.

One half of the families were selected randomly to receive a Family Check-Up, which is adapted according to the home visit assessment findings. Additional services based on the results of the feedback session are designed to be collaborative, ecologically based, and brief. Some parents received periodic check-up calls following their home visit assessment date to report changes in the behavior of their young children. This study adapts the EcoFIT model (Dishion & Stormshak, 2007) to the mental health needs of parents with young children.


Summary of Findings: A sizable cadre of findings from follow-up data indicate that the Family Check-Up intervention improves child problem behavior and parents’ positive behavior support. Of particular importance, the Family Check-Up also had unanticipated positive effects, including improved child language skills, diminished child emotional distress and comorbid problems such as early childhood depressive symptoms, and reduced maternal depression. Evidence shows that the intervention is effective for families facing higher versus lower levels of social and family adversity. Fifteen publications describing Early Steps research findings are in print, in press, or under review.

Early Steps, Stage 2

The second stage of the Early Steps Project began in 2008 and is funded for an additional five years. The same 731 families from the first stage of the project are being asked to continue to participate in home visit assessments when their child is 7, 8, 9, and 10 years old. Families who previously received feedback sessions are being offered this service again, and parents will continue to receive check-up calls at 3, 6, and 9 months after their home visit assessment date, but for this stage they or their child report on the child’s activities during the preceding 24 hours.

Monitoring Project

The second stage of the Early Steps Project occurs concurrently with another grant-funded project, Parental Involvement, Extra-Familial Contexts, and Prevention of Drug Use Risk (aka the Monitoring Project), to learn about the influence of settings outside the home such as school, after-school care programs, and neighborhoods, on children and their families. This grant began June 1, 2007, and continues until April 30, 2012. The same 731 families are asked to grant permission for their child to be observed at school and in an after-school care setting each year following their Early Steps home visit assessment. They are also asked to permit their child’s teacher and after-school care provider to be contacted each year about completing questionnaires about the child. Each family’s neighborhood is observed twice per year. Current research is examining how family neighborhoods, poverty, and ethnic identity influence young children’s development. Information gathered may be shared with the parents receiving a feedback session through the Early Steps Project.


During the second five years of the Early Steps Multisite Project, the family intervention model is being adapted to provide support for children’s transition into the public school system, and developmental analyses are being conducted to better understand early emergence of school competence, early deviant peer involvement, problem behavior, and emotional distress. Assessment, including yearly direct observations and measurement of child and family characteristics, has yielded a rich opportunity to link caregiver engagement in the intervention with child and family outcomes.

2011-2012: With completion of the age 7.5 and age 8.5 assessments and data collection nearing completion for age 9.5 assessments, the team has focused on increased mastery and scheduling of the age 10.5 assessments, which include psychiatric assessments of children. In response to the changing developmental needs of the children and their adjustment to more rigorous school expectations, the parent consultants have focused their training on new and creative approaches to support the parents.

A core issue in the study of intervention effects is the formulation and monitoring of intervention fidelity. During the past year, we developed and refined a fidelity implementation coding system for videotaped parent consultant sessions. Analyses revealed that high levels of fidelity to the FCU model were prognostic of client engagement in the feedback session, which in turn predicted improvements in observed parental use of positive behavior support and resulting improvements in children’s problem behavior. These findings are a conservative estimate of intervention fidelity on outcomes, because all therapists observed were found to implement the FCU model with acceptable levels of fidelity.

2010: Training efforts were focused on examiner training for the age 9.5 assessments. Age 7.5 assessments were completed and age 8.5 assessments were to be completed by December 2011. Intervention effects have been revealed on age 7.5 teacher reports on the Achenbach TRF Oppositional Defiant Disorder factor; children in the intervention group demonstrated significantly less oppositional behavior in the classroom than did control children. A similar effect was found for the teacher-child conflict factor. A significant difference was found in the slope of child inhibitory control between intervention and control children from ages 2 to 7.5, an effect that was more pronounced when intervention families engaged in the Family Check-Up 1 to 2 years out of 4 years from age 2 to 5. An indirect intervention effect has been found on children's school achievement at age 7.5 as the result of changes associated with the FCU on child aggression between ages 2-3 and 4-5. A similar indirect intervention effect on school achievement was also found for changes in positive behavior support among primary caregivers between ages 2 and 3.

2009: The bulk of age 7.5 assessments were completed and age 8.5 assessments were begun. Five papers were published or prepared; one described the effectiveness of the FCU for reducing multiple types and combinations of problem behaviors and improving positive behavior support (Shaw et al., 2009). Gardner et al. (2009) found that few indicators of initial family risk (e.g., parental drug use, depression) moderated the probability of intervention efectiveness, suggesting that the FCU worked comparably well with extremely high-risk versus less high-risk families in our sample. Wilson et al. (2009) found that African American children in the sample experienced greater frequency of family- and of community-level stressors, but that European American children appeared to be more vulnerable to the effects of such stressors on child disruptive behavior at age 2. Families who engaged in the intervention were compared with families with similar profiles of risk in the control group. Dishion et al. (in progress) found intervention effects on growth in child disruptive behavior from ages 2 to 5. Shaw et al. (in progress) also found intervention effects on child problem behavior at age 5 for those families experiencing parental separation/divorce between child age 4 and 5.

2008: By the end of 2008, 16% of families due for assessments had participated in age 7 assessments and 26% of families in the intervention group had contact with parent consultants. Consistent patterns of intervention effects were found relevant to reductions in child conduct problems from ages 2 to 4 and to mediation of these improvements by increased positive parenting. In addition to hypothesized group differences in child conduct problems and in parenting, we also found the intervention to be associated with a number of collateral effects, including reductions in child internalizing and co-occurring internalizing and externalizing problems, maternal depression, and improvements in child language development and inhibitory control, the latter again mediated by improvements in positive parenting. We also explored potential moderators of intervention effects and specifically whether families at higher or lower risk on sociodemographic and family process factors would be more likely to benefit from the intervention. With a couple of exceptions (i.e., parents with higher levels of marital distress and teen parents benefit more), the intervention appears to be equally as effective for families that are more distressed and disadvantaged as for other families, as indexed by parents with maternal depression, low income, or a history of drug abuse.