Document Type



Doctor of Philosophy (PhD)



First Advisor's Name

Jonathan S. Comer

First Advisor's Committee Title

Committee chair

Second Advisor's Name

Jeremy W. Pettit

Second Advisor's Committee Title

Committee member

Third Advisor's Name

Daniel M. Bagner

Third Advisor's Committee Title

Committee member

Fourth Advisor's Name

Adela C. Timmons

Fourth Advisor's Committee Title

Committee member

Fifth Advisor's Name

Elizabeth D. Cramer

Fifth Advisor's Committee Title

Committee member


adaptive interventions, behavioral parenting interventions, child behavior problems, caregiver skill acquisition, clinical decision making

Date of Defense



Heterogeneity in mental health treatment outcomes and high rates of treatment nonresponse highlight the need for adaptive interventions that align with precision mental health care approaches to tailor treatments according to individual differences in progress over time. Modern clinical trial methodologies and analytic strategies can inform dynamic mental health treatment decisions, but the potential to improve patient outcomes is only as strong as the extent to which selected tailoring variables (i.e., interim response factors that dictate whether treatment should shift course) accurately detect risk for treatment nonresponse. Identifying empirically informed tailoring variables and the most appropriate timepoint(s) to assess them (i.e., critical decision points) is essential in order to design adaptive interventions.

This dissertation is comprised of three manuscripts focused on the use of early interim progress data to detect risk for mental health treatment nonresponse. First, I detail a strategy that leverages secondary data analysis to examine candidate tailoring variables at candidate critical decision points, and their relationships with treatment nonresponse. Then, I directly apply this strategy to a pooled sample of families who presented for treatment of early childhood behavior problems (N=153). This study showed that using dichotomous classifications of early interim treatment progress yielded limited utility in differentially predicting post-treatment response when examined in isolation from one another. Thus, I subsequently adopt a continuous approach to measuring early interim treatment progress and examine whether interactions between early indicators of treatment response predict symptom trajectories in a sample of families who participated in a behavioral parenting intervention (BPI) for early childhood developmental delay and behavior problems (N=70).

Findings from the third paper suggest symptom response trajectories can be predicted by examining the interaction between caregiver skills and child behavior problems displayed within the first six sessions of a BPI. Collectively, this collection of work encourages the use of routine outcome monitoring to assess multiple domains of early interim treatment progress. To improve the efficiency and effectiveness of mental health care, future work should continue to use analytic approaches that capture the dynamic interplay among multiple early interim response factors that can optimally inform clinical decision-making practices throughout treatment.



Previously Published In

Hong, N., Cornacchio, D., Pettit, J.W., & Comer, J.S. (2019). Coal-mine canaries in clinical psychology: Getting better at identifying early signals of treatment nonresponse. Clinical Psychological Science, 7(6), 1207-1221. doi: 10.1177/2167702619858111



Rights Statement

Rights Statement

In Copyright. URI:
This Item is protected by copyright and/or related rights. You are free to use this Item in any way that is permitted by the copyright and related rights legislation that applies to your use. For other uses you need to obtain permission from the rights-holder(s).