Background:Parent training interventions effectively reduce disruptive behavior in children. However, research on how participant characteristics and program factors influence the outcomes in real-world settings remains scarce.

Objective:This study aimed to identify factors predicting outcomes of the internet-based, telephone-assisted Strongest Families Parent training program.

Methods:A prospective cohort implementation study conducted within population-based screening embedded in routine health checkups targeting all children aged 4 years in Finland, to identify children with high levels of conduct problems and functional impairment. From a study population of 49,504, altogether 3911 participants completed baseline measures, 707 participants did not do so at 6 months follow-up, resulting in a sample of 3204 (1158/3186, 36.3% girl, 2028/3186, 63.7% boys). Reported duration of difficulties was 6 months in 29.57% (934/3159) of participants, 6‐12 months in 27% (853/3159) of participants, and >12 months in 43.43% (1372/3159) of participants. Most children lived with 2 biological parents (2721/3194, 85.19%). A total of 35.24% (1121/3181) of mothers and 26.18% (797/3044) of fathers had a university degree. Data was collected via parent report. Multinomial logistic regression analyses were conducted to identify which child-, family-, and program-related factors predicted changes in the Child Behavior Checklist 1.5-5 (CBCL) externalizing subscale from baseline to 6-month follow-up. The standardized change in CBCL externalizing score was created by subtracting the mean at baseline from the individual 6-month measurement, divided by the SD at baseline. The standardized change was categorized to ±0.5 SD (no change); +0.5 to +1.5 SD (moderate improvement), >+1.5 SD (large improvement), and more than −0.5 SD (deterioration). A P value of <.05 was considered significant.

Results:In 77% (2468/3204) of participants, symptoms improved at 6-month follow-up. Multinomial logistic regression analyses with α-level of <0.05 showed that >12 months duration of initial problems, callous-unemotional traits, and CBCL internalizing symptoms were linked to lower likelihood of large improvement (odds ratio [OR] 0.43, 95% CI 0.33‐0.56; P<.001; OR 0.64, 95% CI 0.57‐0.73; P<.001; OR 0.54, 95% CI 0.47‐0.63; P<.001, respectively). Definite and severe problems at baseline were linked to deterioration (OR 2.29, 95% CI 1.62‐3.24; P<.001; OR 4.38, 95% CI 2.80‐6.85; P<.001, respectively). Parental stress was linked to a lower likelihood of large improvement (OR 0.78, 95% CI 0.67‐0.91; P=.002), and anxiety to a higher likelihood of deterioration (OR 1.20, 95% CI 1.04‐1.39; P=.02).

Conclusions:Children with longer-term and more severe behavioral symptoms may require tailored intervention. Support for parents with stress may be recommended. Much of the current literature on parent training is based on randomized controlled trials, while the literature on the implementation of parenting programs and studies examining change is limited. Our study informs about predictors of treatment outcomes when interventions are implemented. These results are important clinically as they allow personalization of interventions.