Introduction: Decreasing the frequency of child abuse is a moral imperative with clear benefits to children. Of the many programs designed to reduce child abuse, we examined parent training (PT). A previous meta-analysis of many programs aimed to decrease child maltreatment identified only five PT programs and reported results in the low to moderate range (i.e., d = .36; MacLeod & Nelson, 2000). Our meta-analysis investigated 23 PT studies.

Method: Empirical studies were identified by searching three electronic databases (i.e., ERIC, PsychINFO, Social Work Abstracts) using key words associated with “parent training” and “child abuse.” To be included, studies needed to test PT as an intervention targeting child abuse, include at least five participants, and provide sufficient statistics to calculate Cohen's d. Studies targeting sexual abuse were excluded. Studies were independently coded by the authors with high inter-rater reliability. Dependent variables for parents were organized into four groups: emotional adjustment, attitudes toward childrearing, childrearing skills, and documented abuse. Separate analyses were conducted for results directly following treatment and follow-up periods.

Of the 23 studies, 17 used pre-post designs and 6 included a treatment group. Most studies targeted at-risk families (52%) compared to identified abusers (32%). Over half of PT programs involved a home delivery component (60%). Group PT was the preferred option (44%) followed by individual (24%) or a combined (32%) delivery.

Findings: General findings. Immediately following treatment, parents evidenced moderate decreases in unhealthy childrearing attitudes (d = .60, n = 11) and increases in emotional adjustment (d = .53, n = 11). Similar improvements in childrearing skills (d = .51, n = 13) and actual abuse (d = .45, n = 3) were found. Data from studies that conducted follow-up assessments suggest desirable changes in childrearing attitudes were durable, but that emotional adjustment and childrearing skills declined (ds = .19 and .32). The above values are slightly inflated in some cases, though not all, because single-subject designs produced stronger ES than those using a control group.

Moderator analyses. Moderator analyses were conducted if significant heterogeneity existed. General themes are presented from such analyses. First, families identified for actual abuse did just as well as families at-risk for abuse. Second, programs involving families in both a therapeutic setting and a home-based service had stronger outcomes than those offering only traditional PT. For example, gains in childrearing skills for those in a therapeutic setting only (d = .41, n = 5) were significantly less than those involved in a mixed program (d = .87, n = 7), Qb = 8.85, p < .01. Results did not clearly identify an advantage from group- or individually-delivered parent training; however, childrearing skills tended to increase more with individual PT whereas emotional adjustment improved more with group delivery.

Reference MacLeod, J., & Nelson, G. (2000). Programs for the promotion of family wellness and the prevention of child maltreatment: A meta-analytic review. Child Abuse & Neglect, 24, 1127-1149.