Explaining functional outcomes in depression treatment: A multilevel modelling approach

Hesham F. Gadelrab, Maria Cabello, Eduard Vieta, Jesus Valle, Matilde Leonardi, Somnath Chatterji, José Luís Ayuso-Mateos

Research output: Contribution to journalArticlepeer-review


Background. The main aim of this study is to investigate the role that social, environmental and personal factors play in explaining change in function in patients with major depressive disorder (MDD). Method. The sample of the current study comes from a naturalistic and prospective cohort study of patients starting antidepressant treatment in one distinct health area of Madrid. A total of 97 adults (18 years or older) with MDD have been followed up three points in time and evaluated with the WHO-DAS II and instruments that assessed socio-demographic characteristics, clinical and contextual factors. Data were analysed using multilevel modelling statistical approach to repeated measures data. Results. Change of severity of depression symptoms was the main explanatory factor for change in patient's functioning, accounting for 32% of the variance of functioning over time. Contextual factors (environmental and personal variables), including gender, work status, societal attitudes, health satisfaction and quality of life accounted for 20% of the functioning variance after controlling for the effect of health condition variables. Conclusions. Social, environmental and personal factors seem to play a significant role in explaining depressed patients' functioning after controlling for health condition/medical factors.

Original languageEnglish
JournalDisability and Rehabilitation
Issue numberSUPPL. 1
Publication statusPublished - 2010


  • Depressive disorder
  • functional outcomes
  • ICF
  • longitudinal data
  • multilevel analysis

ASJC Scopus subject areas

  • Rehabilitation


Dive into the research topics of 'Explaining functional outcomes in depression treatment: A multilevel modelling approach'. Together they form a unique fingerprint.

Cite this