TY - JOUR

T1 - Does outcome expectancy predict outcomes in online depression prevention? Secondary analysis of randomised-controlled trials

AU - Thielecke, Janika

AU - Kuper, Paula

AU - Ebert, David

AU - Cuijpers, Pim

AU - Smit, Filip

AU - Riper, Heleen

AU - Lehr, Dirk

AU - Buntrock, Claudia

N1 - Publisher Copyright: © 2023 The Authors. Health Expectations published by John Wiley & Sons Ltd.

PY - 2024/2

Y1 - 2024/2

N2 - Background: Evidence shows that online interventions could prevent depression. However, to improve the effectiveness of preventive online interventions in individuals with subthreshold depression, it is worthwhile to study factors influencing intervention outcomes. Outcome expectancy has been shown to predict treatment outcomes in psychotherapy for depression. However, little is known about whether this also applies to depression prevention. The aim of this study was to investigate the role of participants' outcome expectancy in an online depression prevention intervention. Methods: A secondary data analysis was conducted using data from two randomised-controlled trials (N = 304). Multilevel modelling was used to explore the effect of outcome expectancy on depressive symptoms and close-to-symptom-free status postintervention (6–7 weeks) and at follow-up (3–6 months). In a subsample (n = 102), Cox regression was applied to assess the effect on depression onset within 12 months. Explorative analyses included baseline characteristics as possible moderators. Outcome expectancy did not predict posttreatment outcomes or the onset of depression. Results: Small effects were observed at follow-up for depressive symptoms (β = −.39, 95% confidence interval [CI]: [−0.75, −0.03], p =.032, padjusted =.130) and close-to-symptom-free status (relative risk = 1.06, 95% CI: [1.01, 1.11], p =.013, padjusted = 0.064), but statistical significance was not maintained when controlling for multiple testing. Moderator analyses indicated that expectancy could be more influential for females and individuals with higher initial symptom severity. Conclusion: More thoroughly designed, predictive studies targeting outcome expectancy are necessary to assess the full impact of the construct for effective depression prevention. Patient or Public Contribution: This secondary analysis did not involve patients, service users, care-givers, people with lived experience or members of the public. However, the findings incorporate the expectations of participants using the preventive online intervention, and these exploratory findings may inform the future involvement of participants in the design of indicated depression prevention interventions for adults. Clinical Trial Registration: Original studies: DRKS00004709, DRKS00005973; secondary analysis: osf.io/9xj6a.

AB - Background: Evidence shows that online interventions could prevent depression. However, to improve the effectiveness of preventive online interventions in individuals with subthreshold depression, it is worthwhile to study factors influencing intervention outcomes. Outcome expectancy has been shown to predict treatment outcomes in psychotherapy for depression. However, little is known about whether this also applies to depression prevention. The aim of this study was to investigate the role of participants' outcome expectancy in an online depression prevention intervention. Methods: A secondary data analysis was conducted using data from two randomised-controlled trials (N = 304). Multilevel modelling was used to explore the effect of outcome expectancy on depressive symptoms and close-to-symptom-free status postintervention (6–7 weeks) and at follow-up (3–6 months). In a subsample (n = 102), Cox regression was applied to assess the effect on depression onset within 12 months. Explorative analyses included baseline characteristics as possible moderators. Outcome expectancy did not predict posttreatment outcomes or the onset of depression. Results: Small effects were observed at follow-up for depressive symptoms (β = −.39, 95% confidence interval [CI]: [−0.75, −0.03], p =.032, padjusted =.130) and close-to-symptom-free status (relative risk = 1.06, 95% CI: [1.01, 1.11], p =.013, padjusted = 0.064), but statistical significance was not maintained when controlling for multiple testing. Moderator analyses indicated that expectancy could be more influential for females and individuals with higher initial symptom severity. Conclusion: More thoroughly designed, predictive studies targeting outcome expectancy are necessary to assess the full impact of the construct for effective depression prevention. Patient or Public Contribution: This secondary analysis did not involve patients, service users, care-givers, people with lived experience or members of the public. However, the findings incorporate the expectations of participants using the preventive online intervention, and these exploratory findings may inform the future involvement of participants in the design of indicated depression prevention interventions for adults. Clinical Trial Registration: Original studies: DRKS00004709, DRKS00005973; secondary analysis: osf.io/9xj6a.

KW - CBT

KW - depression

KW - expectancy

KW - online intervention

KW - prediction

KW - prevention

KW - secondary analyses

KW - Psychology

UR - http://www.scopus.com/inward/record.url?scp=85181205847&partnerID=8YFLogxK

UR - https://www.mendeley.com/catalogue/9c0f331c-49bc-310b-8905-98dfaced0da1/

U2 - 10.1111/hex.13951

DO - 10.1111/hex.13951

M3 - Journal articles

AN - SCOPUS:85181205847

VL - 27

JO - Health Expectations

JF - Health Expectations

SN - 1369-6513

IS - 1

M1 - e13951

ER -