Computerized detection and internet-based treatment of common mental disorders among college students in two Latin American LMICs

PROJECT SUMMARY/ABSTRACT: Objectives: This proposal is submitted in response to PAR-18-835 to
detect, engage, and experimentally evaluate the effects of internet-based cognitive behavior therapy (e-CBT)
to treat college students in two Latin American LMICs with major depressive disorder and/or generalized
anxiety disorder w/ or w/o comorbid common mental disorders (CMDs). Detection and engagement will be
based on inexpensive internet-based self-report surveys and the intervention on inexpensive e-CBT to address
the low resources in LMICs. Knowledge transfer to the primary LMIC collaborating institution (Mexican National
Institute of Psychiatry Ramon de la Fuente Muniz [INPRFM]) will be a goal designed to facilitate dissemination
to LMIC colleges/universities throughout Latin America. The study will build on prior research by collaborators
in the WHO World Mental Health (WMH) survey consortium and WMH International College Study (WMH-ICS)
initiative documenting high prevalence, impairment, and unmet need for treatment of CMDs among college
students around the world and significant effects of e-CBT in treating these disorders in high income countries.
Specific aims/Design/Methods: First, we will carry out a pragmatic clinical trial with students seeking
treatment for CMDs at student clinics in four universities in Colombia and Mexico. Students on waitlists for
student clinic services will be offered a possibility of receiving guided e-CBT immediately while staying on the
list. 50% of the students with CMD who express interest will be randomized to guided e-CBT and the other
50% to treatment as usual (TAU). Short-term aggregate intervention effects will be assessed 60 days after
randomization and longer-term effects 12 months after randomization. We will then use ensemble machine
learning methods to predict heterogeneity of treatment effects of guided e-CBT versus TAU and develop a
precision treatment rule (PTR) to predict which students will respond best to which intervention. We will then
implement a SMART design in which we randomize 50% to the treatment arm predicted to be optimal by the
PTR and the other 50% to randomization across arms. Additionally, we will carry out annual WMH-ICS
internet-based mental health needs assessment surveys with a probability sample of students to estimate
unmet need for treatment and barriers to treatment. Students screening positive for CMD who are not in
treatment and report resistance to in-person treatment will be randomized to guided e-CBT to determine if
access to guided e-CBT reduces the unmet need for treatment among students with CMDs who fail to seek
treatment at student health clinics. Finally, the technology for screening, delivering e-CBT, and updating the
precision treatment model, implemented initially by Harvard Medical School, will be transferred to INPRFM for
dissemination to universities in LMICs throughout Latin America. That many senior regional mental health
policy makers are trained at INPRFM and return for periodic policy summits, coupled with the close working
relationship of INPRFM with the Pan American Health Organization, bode well for such dissemination efforts.

Grant Number: 5R01MH120648-05
NIH Institute/Center: NIH
Principal Investigator: Corina Benjet