Predictors for successful psychotherapy: Does migration status matter?

by Friederike Kobel, Yesim Erim, Eva Morawa

Background

We investigated, if migration status, and additional sociodemographic and clinical factors, are associated with somatization and depressiveness at admission and with remission after inpatient psychotherapy.

Methods

Multiple linear and binary logistic regression analyses were used to identify predictors for severity of somatoform and depressive symptoms at admission of inpatient psychotherapy (T0), and for remission after inpatient psychotherapy (T1). We tested the association between symptoms concerning somatization (PHQ-15: Patient-Health-Questionnaire Somatization Module) and depression (PHQ-9: Patient-Health-Questionnaire Depression Module) and several sociodemographic and clinical factors in 263 patients at admission. For remission after treatment, we additionally included severity of symptoms at admission, number of diagnoses and duration of treatment in the regression models. Remission after treatment was defined as response plus a post value of less than 10 points in the respective questionnaire. Clinical relevance was interpreted using effect sizes (regression coefficients, Odds Ratio (OR)) and Confidence Intervals (CI).

Findings

Significant and clinically relevant predictors for high symptom severity at T0 were lower education (β = -0.13, p = 0.04), pretreatment(s) (β = 0.205, p = 0.002) and migration status (β = 0.139, p = 0.023) for somatization, and potential clinically relevant predictors (|β|>0.1) for depression were living alone (β = -0.116, p = 0.083), pretreatment(s) (β = 0.118, p = 0.071) and migration status (β = 0.113, p = 0.069). At T1 patients with pretreatment(s) (OR = 0.284 [95% CI: 0.144, 0.560], p<0.001) and multiple diagnoses (OR = 0.678 [95% CI: 0.472, 0.973], p = 0.035) were significantly and clinically relevant less likely to show a remission of depressive symptoms. In addition, a potentially clinically meaningful effect of migration status on remission of depressive symptoms (OR = 0.562 [95% CI: 0.264, 1.198], p = 0.136) cannot be ruled out. For somatoform symptoms pretreatment(s) (OR = 0.403, [95% CI: 0.156, 1.041], p = 0.061) and education (OR = 1.603, [95% CI: 0.670, 3.839], p = 0.289) may be regarded as clinically relevant predictors for remission.

Conclusion

The results of our study suggest that migration status has a clinically relevant influence on severity of somatoform and depressive symptoms at admission. Clinical relevance of migration status can also be assumed regarding the remission of depression. Migration status and further factors affecting the effectiveness of the treatment should be analyzed in future research among larger samples with sufficient power to replicate these findings.