Trajectories of depression and anxiety symptom change during psychological therapy

• Published in: Journal of affective disorders 2019-04, Vol.249, p.327-335
• Main Authors: Saunders, Rob, Buckman, Joshua E.J., Cape, John, Fearon, Pasco, Leibowitz, Judy, Pilling, Stephen
• Format: Article
• Language: English
• Subjects: Adult; Anxiety; Anxiety Disorders - complications; Anxiety Disorders - psychology; Anxiety Disorders - therapy; Depression; Depressive Disorder - complications; Depressive Disorder - psychology; Depressive Disorder - therapy; Female; Humans; IAPT; Latent class growth analysis; Logistic Models; Male; Middle Aged; Psychotherapy; Psychotherapy - methods; Treatment Outcome
• ISSN: 0165-0327; 1573-2517; 1573-2517
• DOI: 10.1016/j.jad.2019.02.043

•Four distinct depressive & 5 distinct anxious symptom trajectories were identified.•Most were distinguishable by session 3 but two depressive trajectories were not.•Determining “not on track” status at session 3 may lead to missing late-responders.•Baseline severity, functioning and diagnosis predicted some trajectories.•Identifying likely trajectories could inform decision-making and optimise care. Forty-percent of the variance in psychological treatment outcomes is estimated to be explained by symptom change by the third treatment session. However, change may not be uniform across patient groups and symptom domains. This study aimed to identify subgroups of patients with different trajectories of depression and anxiety symptom change during psychological therapy and identify baseline patient characteristics associated with these trajectories. 4394 patients attending two psychological treatment services completed sessional, self-report depression and anxiety measures. Trajectories of symptom change were investigated using latent class growth analysis. Multinomial logistic regression was used to explore associations between baseline patient characteristics and trajectory classes. A number of distinct trajectories were identified. Anxiety symptom trajectories could be distinguished by the third treatment session, but for depression symptoms there was a class displaying limited change until session six followed by rapid improvement in symptoms thereafter. Compared to the non-responding trajectories, depression and anxiety trajectories indicating treatment response were associated with lower baseline severity, better social functioning and lower incidence of phobic anxiety, but not with medication prescription status. Data came from two services, so wider generalisability is unknown. Predictors were limited to data routinely collected in the services; unmeasured factors may have improved the prediction of trajectories. Baseline characteristics and symptom change early in therapy can help identify different trajectories of symptom change. This knowledge could aid clinical decision making and help improve treatment outcomes. By ignoring distinct trajectories, clinicians may incorrectly consider patients as “not-on-track” and unnecessarily change or end therapy that would otherwise benefit patients.