A machine learning approach to understanding patterns of engagement with internet-delivered mental health interventions

Isabel Chien, Angel Enrique, Jorge Palacios, Tim Regan, Dessie Keegan, David Carter, Sebastian Tschiatschek, Aditya Nori, Anja Thieme, Derek Richards, Gavin Doherty, Danielle Belgrave

Veröffentlichungen: Beitrag in FachzeitschriftArtikelPeer Reviewed

Abstract

IMPORTANCE The mechanisms by which engagement with internet-delivered psychologicalinterventions are associated with depression and anxiety symptoms are unclear.
OBJECTIVE To identify behavior types based on how people engage with an internet-basedcognitive behavioral therapy (iCBT) intervention for symptoms of depression and anxiety.
DESIGN, SETTING, AND PARTICIPANTS Deidentified data on 54 604 adult patients assigned to theSpace From Depression and Anxiety treatment program from January 31, 2015, to March 31, 2019,were obtained for probabilistic latent variable modeling using machine learning techniques to inferdistinct patient subtypes, based on longitudinal heterogeneity of engagement patterns with iCBT.
INTERVENTIONS A clinician-supported iCBT-based program that follows clinical guidelines fortreating depression and anxiety, delivered on a web 2.0 platform.
MAIN OUTCOMES AND MEASURES Log data from user interactions with the iCBT program toinform engagement patterns over time. Clinical outcomes included symptoms of depression (PatientHealth Questionnaire-9 [PHQ-9]) and anxiety (Generalized Anxiety Disorder-7 [GAD-7]); PHQ-9 cutpoint greater than or equal to 10 and GAD-7 scores greater than or equal to 8 were used to definedepression and anxiety.RESULTSPatients spent a mean (SD) of 111.33 (118.92) minutes on the platform and completed230.60 (241.21) tools. At baseline, mean PHQ-9 score was 12.96 (5.81) and GAD-7 score was 11.85(5.14). Five subtypes of engagement were identified based on patient interaction with differentprogram sections over 14 weeks: class 1 (low engagers, 19 930 [36.5%]), class 2 (late engagers, 11 674[21.4%]), class 3 (high engagers with rapid disengagement, 13 936 [25.5%]), class 4 (high engagerswith moderate decrease, 3258 [6.0%]), and class 5 (highest engagers, 5799 [10.6%]). Estimatedmean decrease (SE) in PHQ-9 score was 6.65 (0.14) for class 3, 5.88 (0.14) for class 5, and 5.39 (0.14)for class 4; class 2 had the lowest rate of decrease at −4.41 (0.13). Compared with PHQ-9 scoredecrease in class 1, the Cohendeffect size (SE) was −0.46 (0.014) for class 2, −0.46 (0.014) for class3, −0.61 (0.021) for class 4, and −0.73 (0.018) for class 5. Similar patterns were found across groupsfor GAD-7.CONCLUSIONS AND RELEVANCEThe findings of this study may facilitate tailoring interventionsaccording to specific subtypes of engagement for individuals with depression and anxiety. Informingclinical decision needs of supporters may be a route to successful adoption of machine learninginsights, thus improving clinical outcomes overall.
OriginalspracheEnglisch
Aufsatznummere2010791
FachzeitschriftJAMA Network Open
Jahrgang3
Ausgabenummer7
DOIs
PublikationsstatusVeröffentlicht - 17 Juli 2020

ÖFOS 2012

  • 501024 Verhaltenstherapie
  • 102019 Machine Learning

Zitationsweisen