Identifying patients with therapy-resistant depression by using factor analysis

K Andreasson, V Liest, M Lunde, K Martiny, M Unden, S Dissing, P Bech

7 Citations (Scopus)

Abstract

Attempts to identify the factor structure in patients with treatment-resistant depression have been very limited. Methods: Principal component analysis was performed using the baseline datasets from 3 add-on studies [2 with repetitive transcranial magnetic stimulation and one with transcranial pulsed electromagnetic fields (T-PEMF)], in which the relative effect as percentage of improvement during the treatment period was analysed. Results: We identified 2 major factors, the first of which was a general factor. The second was a dual factor consisting of a depression subscale comprising the negatively loaded items (covering the pure depression items) and a treatment resistant subscale comprising the positively loaded items (covering lassitude, concentration difficulties and sleep problems). These 2 dual subscales were used as outcome measures. Improvement on the treatment resistant subscale was 40% in the active treatment group compared to 1730% improvement in the sham treatments. Discussion: It is possible to describe patients with therapy-resistant depression by a factor structure. Both rTMS and T-PEMF had a clinical effect on the factor-derived scales when compared to sham treatment.

Original languageEnglish
JournalPharmacopsychiatry
Volume43
Issue number7
Pages (from-to)252-6
Number of pages5
ISSN0176-3679
DOIs
Publication statusPublished - 1 Nov 2010

Keywords

  • Antidepressive Agents
  • Clinical Trials as Topic
  • Depressive Disorder, Major
  • Drug Resistance
  • Factor Analysis, Statistical
  • Female
  • Humans
  • Male
  • Placebos
  • Principal Component Analysis
  • Psychiatric Status Rating Scales
  • Transcranial Magnetic Stimulation
  • Treatment Failure
  • Treatment Outcome

Fingerprint

Dive into the research topics of 'Identifying patients with therapy-resistant depression by using factor analysis'. Together they form a unique fingerprint.

Cite this