Clinical and genetic predictors of delayed remission after multiple levels of antidepressant treatment: Toward early identification of depressed individuals for advanced care options

Academic Article


  • Objective: To identify clinical and genetic characteristics that can be used to recognize depressed patients who are likely to respond quickly versus those who will have a more delayed response following multiple treatment trials. Methods: The data used were obtained from the National Institute of Mental Health–sponsored Sequenced Treatment Alternatives to Relieve Depression (STAR*D) study, which was conducted between July 2001 and September 2006. Of the 4,041 treatment-naive participants in the original study, 1,953 with DNA samples were included. Major depressive disorder (DSM-IV criteria) was defined as baseline score > 14 on the 17-item Hamilton Depression Rating Scale. Time to remission was defined from the entry point to when a score ≤ 5 on the Quick Inventory of Depressive Symptomatology, Clinician Rating was achieved, irrespective of the type or number of treatments received. A Kaplan-Meier estimator was used for data description, proportional hazard regression for model building, and logistic regression for measures of predictive accuracy. Results: The overall rate of remission across all levels of treatment was 65.6%, and the overall median (interquartile range) of time to remission was 11.4 (6.0–17.9) weeks. The predictors of delayed remission included unemployment (P= .004), severe medical comorbidity (P< .0001), severe baseline depression (P< .0001), more than 4 dysthymic symptoms (P= .005), more than 9 posttraumatic stress symptoms (P= .005), and serotonin receptor 1A (P= .006) and cytochrome P450 2D6 (P= .002 for C/T and P= .0004 for T/T) genetic variants. The final model had good predictive measures of accuracy of area under the curve (70%) and sensitivity (88%). Conclusions: The results offer clinical tools for clinicians to identify depressed individuals who are likely to have delayed remission with multiple antidepressant treatments and therefore might be candidates for advanced care options.
  • Published In

    Digital Object Identifier (doi)

    Author List

  • Falola MI; Limdi N; Shelton RC
  • Start Page

  • e1291
  • End Page

  • e1298
  • Volume

  • 78
  • Issue

  • 9