Subtypes of treatment-resistant depression determined by a latent class analysis in a Chinese clinical population

J Affect Disord. 2019 Apr 15:249:82-89. doi: 10.1016/j.jad.2019.02.005. Epub 2019 Feb 6.

Abstract

Background: This study aimed to explore subtypes of treatment-resistant depression (TRD).

Methods: Latent class analysis (LCA) was performed on clinical and demographic data collected from 375 patients with TRD. Clinical variables were compared across subtypes. Treatment outcomes across subtypes of TRD were compared separately for those within each subtype with anxiety (those with a HRSD-17 anxiety/somatization factor score ≥ 7) and those without anxiety. LCA subtypes were compared using Cochran's and Mantel-Haenszel χ2 test, respectively. Unordered multinomial logistic regression was used to assess clinical correlates of TRD subtypes.

Results: Three categories were detected: severe depression (66%), moderate depression with anxiety (9%) and mild depression with anxiety/somatization (25%). Gender, age, age at first onset, family monthly income, number of hospitalizations, HRSD-17 and clinical global impression-severity (CGI) scores were significantly different across the three groups. Remission rates were significantly different among anxious cases with severe (43.75%), moderate (22.73%) and mild (26.25%) depression subtypes. Compared to cases in the mild depression group, those in the severe depression group had a greater likelihood of being male, having a later age of first onset, higher numbers of hospitalization, higher HRSD-17 and CGI total scores, and lower family income. Those in the moderate depression group were more likely to be male and have lower family income than those in the mild depression group.

Limitations: Representative bias, relatively small sample size, unbalanced group size and incomplete indicator variables might have a negative effect on the validity and generalization of the findings.

Conclusions: Depression severity could be a basis for subtype classification of patients with TRD. The classification of latent class of TRD observed in our study was similar to the structure found in MDD. Longitudinal research into the stability of the latent structure of TRD across illness course is merited as is research into treatment outcomes for TRD subtypes.

Keywords: Latent class analysis; Subtype; Treatment-resistant depression.

Publication types

  • Multicenter Study
  • Randomized Controlled Trial
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adult
  • Aged
  • Anxiety / epidemiology
  • Anxiety Disorders / drug therapy
  • Asian People / ethnology*
  • China / epidemiology
  • Depressive Disorder, Treatment-Resistant / classification*
  • Depressive Disorder, Treatment-Resistant / drug therapy
  • Depressive Disorder, Treatment-Resistant / ethnology*
  • Disease Progression
  • Double-Blind Method
  • Female
  • Humans
  • Latent Class Analysis
  • Logistic Models
  • Male
  • Middle Aged
  • Treatment Outcome