Childhood psychiatric conditions often require that clinicians make individualized treatment decisions. In an exciting trend in child and adolescent mental health (CAMH) research and practice, there is growing interest and awareness in developing individualized decision rules (IDRs). IDRs use sociodemographic information, illness history, comorbidities, family/interpersonal measures, or biomarkers as inputs and then recommend the type, modality, and intensity or delivery of subsequent treatment(s) as outputs. IDRs provide a framework for personalized medicine: the level of personalization increases in proportion to the amount of useful information (tailoring variables) used as inputs in an IDR. The development of IDRs for the treatment of children with mental illness is an important """"""""next step"""""""" in interventions research. Large scale clinical trials have established efficacious interventions fr many of the common CAMH disorders. However, despite a strong evidence base for treatments that are effective on average, little is known about which children are more or less likely to benefit from one treatment or another. Before IDRs can fulfill their potential, much methodological work remains to be done regarding how to use currently available data to assist in their development. A key step in developing effective IDRs is identifying tailoring variables to use as inputs. Tailoring variables are special types of pre-treatment measures that pinpoint which treatment is best for whom. Existing methods do not explicitly (a) identify which measures are most useful (tailoring variable selection), or (b) combine them in an optimal fashion (tailorin variable feature construction).
The specific aims of this study are:
Aim 1 (Methods). To develop and evaluate a framework for (a) tailoring variable selection and (b) tailoring variable feature construction for individualizing CAMH treatment. The new method will be used to address scientific questions such as: (a) What are the most important measures (e.g., demographic, severity and duration of illness, comorbidity, parental psychopathology, psychosocial environment including family and peer relationships, biological) to use for tailoring treatment? (b) How do we combine them in order to guide treatment decisions (e.g., individual psychotherapy treatment, medication, or their combination)? Aim 2 (Application): To illustrate and evaluate the method by applying it to the Child/Adolescent Anxiety Multimodal Study randomized trial data to construct an IDR for choosing between SSRI, CBT, or SSRI+CBT for children with anxiety disorders. We will apply the new methodology to identify variables and features pinpointing children who are more or less likely to benefit from SSRI, CBT, or combined SSRI+CBT. Results will be used to construct a proposed IDR for pediatric anxiety disorders. We will compare how average symptom and severity outcomes differ on the IDR vs SSRI for all vs CBT for all vs SSRI+CBT for all. This study will provide researchers in CAMH with a novel tool for building IDRs. This study has significant public health impact because it aims to develop tools to personalize treatment for children and adolescents with mental health disorders. This is a high-priority area of research for the National Institute of Mental Health.
This project aims to develop a new tool for deciding how to personalize (that is, individualize or customize) treatment for children and adolescents with mental health disorders. This has important clinical and public-health relevance because, although research has established efficacious treatments for many child and adolescent mental health disorders, little is known about which children are more or less likely to benefit from one treatment or another.
|Kidwell, Kelley M; Seewald, Nicholas J; Tran, Qui et al. (2018) Design and Analysis Considerations for Comparing Dynamic Treatment Regimens with Binary Outcomes from Sequential Multiple Assignment Randomized Trials. J Appl Stat 45:1628-1651|
|Wodtke, Geoffrey T; Almirall, Daniel (2017) Estimating Moderated Causal Effects with Time-varying Treatments and Time-varying Moderators: Structural Nested Mean Models and Regression with Residuals. Sociol Methodol 47:212-245|
|Gunlicks-Stoessel, Meredith; Mufson, Laura; Westervelt, Ana et al. (2016) A Pilot SMART for Developing an Adaptive Treatment Strategy for Adolescent Depression. J Clin Child Adolesc Psychol 45:480-94|
|Almirall, Daniel; Chronis-Tuscano, Andrea (2016) Adaptive Interventions in Child and Adolescent Mental Health. J Clin Child Adolesc Psychol 45:383-95|
|Chronis-Tuscano, Andrea; Wang, Christine H; Strickland, Jennifer et al. (2016) Personalized Treatment of Mothers With ADHD and Their Young At-Risk Children: A SMART Pilot. J Clin Child Adolesc Psychol 45:510-21|
|Lu, Xi; Nahum-Shani, Inbal; Kasari, Connie et al. (2016) Comparing dynamic treatment regimes using repeated-measures outcomes: modeling considerations in SMART studies. Stat Med 35:1595-615|
|Almirall, Daniel; DiStefano, Charlotte; Chang, Ya-Chih et al. (2016) Longitudinal Effects of Adaptive Interventions With a Speech-Generating Device in Minimally Verbal Children With ASD. J Clin Child Adolesc Psychol 45:442-56|
|Klasnja, Predrag; Hekler, Eric B; Shiffman, Saul et al. (2015) Microrandomized trials: An experimental design for developing just-in-time adaptive interventions. Health Psychol 34S:1220-8|
|Peris, Tara S; Compton, Scott N; Kendall, Philip C et al. (2015) Trajectories of change in youth anxiety during cognitive-behavior therapy. J Consult Clin Psychol 83:239-52|
|Almirall, Daniel; Nahum-Shani, Inbal; Sherwood, Nancy E et al. (2014) Introduction to SMART designs for the development of adaptive interventions: with application to weight loss research. Transl Behav Med 4:260-74|
Showing the most recent 10 out of 14 publications