Maintaining Outcomes of Internet-Delivered Cognitive-Behavioral Therapy for Depression: A Network Analysis of Follow-Up Effects

TY - JOUR

T1 - Maintaining Outcomes of Internet-Delivered Cognitive-Behavioral Therapy for Depression: A Network Analysis of Follow-Up Effects

AU - Kaiser, Tim

AU - Boschloo, Lynn

AU - Berger, Thomas

AU - Meyer, Björn

AU - Späth-Nellissen, Christina

AU - Schröder, Johanna

AU - Hohagen, Fritz

AU - Moritz, Steffen

AU - Klein, Jan Philipp

N1 - Publisher Copyright: © Copyright © 2021 Kaiser, Boschloo, Berger, Meyer, Späth-Nellissen, Schröder, Hohagen, Moritz and Klein.

PY - 2021/4/20

Y1 - 2021/4/20

N2 - Background: Depression is a highly prevalent mental disorder, but only a fraction of those affected receive evidence-based treatments. Recently, Internet-based interventions were introduced as an efficacious and cost-effective approach. However, even though depression is a heterogenous construct, effects of treatments have mostly been determined using aggregated symptom scores. This carries the risk of concealing important effects and working mechanisms of those treatments. Methods: In this study, we analyze outcome and long-term follow-up data from the EVIDENT study, a large (N = 1,013) randomized-controlled trial comparing an Internet intervention for depression (Deprexis) with care as usual. We use Network Intervention Analysis to examine the symptom-specific effects of the intervention. Using data from intermediary and long-term assessments that have been conducted over 36 months, we intend to reveal how the treatment effects unfold sequentially and are maintained. Results: Item-level analysis showed that scale-level effects can be explained by small item-level effects on most depressive symptoms at all points of assessment. Higher scores on these items at baseline predicted overall symptom reduction throughout the whole assessment period. Network intervention analysis offered insights into potential working mechanisms: while deprexis directly affected certain symptoms of depression (e.g., worthlessness and fatigue) and certain aspects of the quality of life (e.g., overall impairment through emotional problems), other domains were affected indirectly (e.g., depressed mood and concentration as well as activity level). The configuration of direct and indirect effects replicates previous findings from another study examining the same intervention. Conclusions: Internet interventions for depression are not only effective in the short term, but also exert long-term effects. Their effects are likely to affect only a small subset of problems. Patients reporting these problems are likely to benefit more from the intervention. Future studies on online interventions should examine symptom-specific effects as they potentially reveal the potential of treatment tailoring. Clinical Trial Registration: ClinicalTrials.gov, Identifier: NCT02178631.

AB - Background: Depression is a highly prevalent mental disorder, but only a fraction of those affected receive evidence-based treatments. Recently, Internet-based interventions were introduced as an efficacious and cost-effective approach. However, even though depression is a heterogenous construct, effects of treatments have mostly been determined using aggregated symptom scores. This carries the risk of concealing important effects and working mechanisms of those treatments. Methods: In this study, we analyze outcome and long-term follow-up data from the EVIDENT study, a large (N = 1,013) randomized-controlled trial comparing an Internet intervention for depression (Deprexis) with care as usual. We use Network Intervention Analysis to examine the symptom-specific effects of the intervention. Using data from intermediary and long-term assessments that have been conducted over 36 months, we intend to reveal how the treatment effects unfold sequentially and are maintained. Results: Item-level analysis showed that scale-level effects can be explained by small item-level effects on most depressive symptoms at all points of assessment. Higher scores on these items at baseline predicted overall symptom reduction throughout the whole assessment period. Network intervention analysis offered insights into potential working mechanisms: while deprexis directly affected certain symptoms of depression (e.g., worthlessness and fatigue) and certain aspects of the quality of life (e.g., overall impairment through emotional problems), other domains were affected indirectly (e.g., depressed mood and concentration as well as activity level). The configuration of direct and indirect effects replicates previous findings from another study examining the same intervention. Conclusions: Internet interventions for depression are not only effective in the short term, but also exert long-term effects. Their effects are likely to affect only a small subset of problems. Patients reporting these problems are likely to benefit more from the intervention. Future studies on online interventions should examine symptom-specific effects as they potentially reveal the potential of treatment tailoring. Clinical Trial Registration: ClinicalTrials.gov, Identifier: NCT02178631.

KW - depression

KW - health-related quality of life

KW - internet interventions

KW - maintenance

KW - network analysis

UR - http://www.scopus.com/inward/record.url?scp=85105155247&partnerID=8YFLogxK

U2 - 10.3389/fpsyt.2021.598317

DO - 10.3389/fpsyt.2021.598317

M3 - SCORING: Journal article

AN - SCOPUS:85105155247

VL - 12

SP - 598317

JO - FRONT PSYCHIATRY

JF - FRONT PSYCHIATRY

SN - 1664-0640

M1 - 598317

ER -