Manuscript Detail

View Revisions

Shaw, D. S., Dishion, T. J., Supplee, L., Gardner, F., & Arnds, K. (2006). Randomized trial of a family-centered approach to the prevention of early conduct problems: 2-year effects of the family check-up in early childhood. Journal of Consulting and Clinical Psychology, 74(1), 1–9.

Model(s) Reviewed: Family Check-Up® For Children
Manuscript screening details
Screening decision Screening conclusion
Passes screens Eligible for review
Study design details
Rating Design Attrition Baseline equivalence Reassignment Confounding factors
High Randomized controlled trial Low Established on race/ethnicity, parents’ education, and baseline outcomes. None None
Study characteristics
Study participants The study included 120 mothers who participated in the Women, Infants and Children (WIC) program and had sons between ages 17 and 27 months at the time of recruitment in 2001. Families also must have demonstrated at least two of three possible risk factors: (1) socioeconomic status; (2) family risk factors (maternal depression or substance abuse); and (3) child risk factors, or conduct problems. Most study participants were African American (48%) or white (40%); the remaining 12% of participants were biracial. Half of the study participants were single and never married, 45% were married or living with their partner, and 5% were separated, divorced, or widowed. Two-thirds of participating mothers had a high school diploma or less, and the average family income was $15,374.
Setting The study recruited participants from eight sites of the WIC program in the Pittsburgh, PA, area.
Intervention services Family Check-Up typically involves three meetings (initial contact, assessment session, and feedback session). For the purposes of this study, the assessment was completed prior to random assignment, and thus the program group received the assessment (conducted by research staff), an interview session, a feedback session, and possible follow-ups. A trained parent consultant conducted the subsequent interview and feedback sessions. During the interview, the consultant explored parent concerns; in the feedback session, the consultant provided the results of the assessment and explored parents’ willingness to change in problem areas, reinforced parenting strengths, and identified services appropriate for the family. After the feedback session, families possibly also received up to six follow-up sessions focusing on parenting practices, family management issues, and contextual issues (such as child care resources and housing). Fifty-five of the 60 families assigned to the intervention participated in the interview and feedback sessions with the parent consultant.
Comparison conditions Families in the comparison condition received the same WIC services as the intervention group but did not receive visits or intervention from parent consultants. The comparison group also received an assessment session, conducted by research staff. Although this session typically is part of the Family Check-Up program, for the purposes of this study, the assessment was conducted prior to randomization.
Staff characteristics and training Two master’s-level therapists were trained to operate as parent consultants for the study. One had recently completed a master’s degree in social work and had no formal training in family or behavior therapy; the other was a professional therapist who had worked with families for five years. Both were trained for two and a half to three months using strategies developed by the study authors. These strategies included didactic instruction and role-playing, as well as ongoing videotaped supervision of intervention activity and weekly conference calls with one author to discuss problematic cases. Consultants also followed a manual and a book as guides to parenting support services following the intervention.
Funding sources National Institute of Mental Health Grant MH06291 and National Institute on Drug Abuse Grant DA016110.
Author affiliation Thomas Dishion, a study author, is a developer of this model.

Findings details

Child development and school readiness
Rating Outcome measure Effect Sample Timing of follow-up Sample size Intervention group Comparison group Group difference Effect size Statistical significance Notes
High CBCL Aggression Scale (Revised version for age 4 follow-up)
FavorableUnfavorable or ambiguousNo Effect
Pittsburgh sample Age 3 and Age 4 92 children <p>Mean at age 3 9.85 (<abbr title"standard deviation">SD</abbr> 4.04)</p><p> Mean at age 4 6.96, (<abbr title"standard deviation">SD</abbr> 4.76)</p> = 9.85 Mean = 8.93 Difference = -1.16 HomeVEE calculated = 30.22 Not statistically significant, p ≥ 0.05

footnote44

Submitted by user on Fri, 03/15/2019 - 14:29

<abbr title="Home Visiting Evidence of Effectiveness">HomVEE</abbr> calculated the difference in growth rates as the change over time (mean at age 4 minus mean at age 3) for the program group minus the change over time for the comparison group.

footnote46

Submitted by user on Fri, 03/15/2019 - 14:29

Statistical significance is based on the results of the authors’ analysis using a two-way repeated measures analysis of variance.

High CBCL Destructive Scale (Revised version for age 4 follow-up)
FavorableUnfavorable or ambiguousNo Effect
Pittsburgh sample Age 3 and Age 4 92 children <p>Mean at age 3 2.66 (<abbr title"standard deviation">SD</abbr> 1.82)</p><p>Mean at age 4 1.87 (<abbr title"standard deviation">SD</abbr> 1.87)</p> = 2.66 Mean = 3.21 Difference = 0.15 HomeVEE calculated = 3.00 Not statistically significant, p ≥ 0.05

footnote46

Submitted by user on Fri, 03/15/2019 - 14:29

Statistical significance is based on the results of the authors’ analysis using a two-way repeated measures analysis of variance.

High CBCL Physical Aggression
FavorableUnfavorable or ambiguousNo Effect
Pittsburgh sample Age 3 and Age 4 92 children <p>Mean at age 3 0.96 (<abbr title"standard deviation">SD</abbr> 1.07)</p><p>Mean at age 4 0.65, (<abbr title"standard deviation">SD</abbr> 0.95)</p> = 0.96 Mean = 0.67 Difference = -0.42 HomeVEE calculated = 30.27 Statistical significance not reported

footnote44

Submitted by user on Fri, 03/15/2019 - 14:29

<abbr title="Home Visiting Evidence of Effectiveness">HomVEE</abbr> calculated the difference in growth rates as the change over time (mean at age 4 minus mean at age 3) for the program group minus the change over time for the comparison group.

Positive parenting practices
Rating Outcome measure Effect Sample Timing of follow-up Sample size Intervention group Comparison group Group difference Effect size Statistical significance Notes
High HOME Involvement
FavorableUnfavorable or ambiguousNo Effect
Pittsburgh sample Age 3 and Age 4 92 families Mean = 2.00 Mean = 1.72 Difference = 0.82 HomeVEE calculated = 30.27 Statistically significant,
p < 0.05

footnote48

Submitted by user on Fri, 03/15/2019 - 14:29

Statistical significance is based on the authors’ analysis using a two-way repeated measures analysis of covariance. Authors report using a one-tailed test.

Outcome measure summary

Child development and school readiness
Outcome measure Description of measure Data collection method Properties of measure

CBCL:

  • Aggression, Scale
  • Physical Aggression
  • Destructive Scale
The CBCL is a questionnaire that assesses behavioral problems in young children. The two versions of the questionnaire, ages 2–3 and ages 4–18, have two broadband scales—(1) Internalizing, and (2) Externalizing—and narrow-band scales, which include destructive and aggressive behavior for the ages 2–3 version. The researchers focused on the two narrow-band scales from the 2–3 version: Destructive and Aggression. To approximate the use of the same constructs at age 4, the researchers revised the 4–18 version and only included items that were also on the ages 2–3 versions of the Destructive and Aggression Scales. Parent/caregiver report
  • aggression Scale: Cronbach’s α = 0.82 at age 2, Cronbach’s α = 0.84 at age 3, Cronbach’s α = 0.86 at age 4
  • Physical aggression: Not reported by author
  • Destructive Scale: Cronbach’s α = 0.60 at age 2, Cronbach’s α = 0.71 at age 3, Cronbach’s α = 0.73 at age 4
Positive parenting practices
Outcome measure Description of measure Data collection method Properties of measure

HOME: Involvement scale

The HOME assesses parenting practices and aspects of the home environment. Three items were drawn from the HOME, Involvement scale: (1) parent keeps child in visual range, (2) parent talks to child while doing housework, and (3) parent structures child’s play. Observation by a trained examiner during a home visit

Cronbach’s α = 0.53 at age 2, Cronbach’s α = 0.56 at age 3 , Cronbach’s α = 0.68 at age 4