August 2010  << Back  

  Can't open the PDF? Click here for help.

 

 RESEARCH ARTICLE

Early improvements in anxiety, depression, and anger/hostility symptoms and response to antidepressant treatment

Amy Farabaugh, PhD

Depression Clinical and Research Program, Department of Psychiatry, Massachusetts General Hospital, Boston, MA, USA

Shamsah Sonawalla, MD

Psychiatry Research, Jaslok Hospital and Research Centre, Mumbai, India
Department of Psychiatry and Behavioral Sciences, Duke University Medical Center, Durham, NC, USA

Daniel P. Johnson, BA

Depression Clinical and Research Program, Department of Psychiatry, Massachusetts General Hospital, Boston, MA, USA

Janet Witte, MD, MPH

Depression Clinical and Research Program, Department of Psychiatry, Massachusetts General Hospital, Boston, MA, USA

George I. Papakostas, MD

Depression Clinical and Research Program, Department of Psychiatry, Massachusetts General Hospital, Boston, MA, USA

Tracie Goodness, BA

Depression Clinical and Research Program, Department of Psychiatry, Massachusetts General Hospital, Boston, MA, USA

Alisabet Clain, MS

Depression Clinical and Research Program, Department of Psychiatry, Massachusetts General Hospital, Boston, MA, USA

Lee Baer, PhD

Depression Clinical and Research Program, Department of Psychiatry, Massachusetts General Hospital, Boston, MA, USA

David Mischoulon, PhD, MD

Depression Clinical and Research Program, Department of Psychiatry, Massachusetts General Hospital, Boston, MA, USA

Maurizio Fava, MD

Depression Clinical and Research Program, Department of Psychiatry, Massachusetts General Hospital, Boston, MA, USA

Rebecca Harley, PhD

Department of Psychiatry, Massachusetts General Hospital, Boston, MA, USA

BACKGROUND: The purpose of this study was to examine whether treatment response to fluoxetine by depressed outpatients was predicted by early improvement on any of 3 subscales (Anxiety, Depression, and Anger/ Hostility) of the Symptom Questionnaire (SQ).

METHODS: We evaluated 169 depressed outpatients (52.6% female) between ages 18 and 65 (mean age, 40.3 ± 10.6 years) meeting DSM-III-R criteria for major depressive disorder (MDD). All patients completed the SQ at baseline (week 0) and at weeks 2, 4, and 8 of treatment with fluoxetine 20 mg/d. We defined treatment response as a ≥50% reduction in score on the 17-item Hamilton Rating Scale for Depression, and early improvement on 3 SQ subscales (Anxiety, Depression, and Anger/Hostility) as a >30% reduction in score by week 2.

RESULTS: The percentage of patients with significant early improvement in anger was significantly greater than the percentage of those with early improvements in anxiety or depression. When early improvement on the Anxiety, Depression, and Anger/Hostility subscales of the SQ were assessed independently by logistic regression, all 3 subscales were predictors of response to treatment.

CONCLUSIONS: Early improvement in anger, anxiety, and depressive symptoms may predict response to antidepressant treatment among outpatients with MDD.

KEYWORDS: major depressive disorder, anxiety, hostility, depression, Symptom Questionnaire

ANNALS OF CLINICAL PSYCHIATRY 2010;22(3):166–171

  INTRODUCTION

A growing body of research in psychiatry is exploring both biological and psychological markers of response to antidepressant treatment.1,2 Establishing such predictors has the potential to significantly improve treatment for depression, by assisting both clinicians and patients in producing an effective treatment plan. An understanding of response predictors could allow clinicians to decide within a short period (1 to 2 weeks) of initiation of treatment whether to continue a patient on the current medication and dose, increase the dose, or try another medication or intervention.3-5

Previous research has explored the link between symptom reduction and treatment response. In a study of treatment of major depressive disorder (MDD) with antidepressant medication, Papakostas and colleagues6 found that responders had a significant correlation between early improvement in depressive symptoms and lower scores at endpoint on the Anxiety subscale of the Symptom Questionnaire (SQ). We also found that early symptom changes, as assessed by clinician-rated and self-report items, have predicted outcome of treatment by using several symptom questions on the 17-item Hamilton Rating Scale for Depression (HAMD-17).7 These questions were used to assess the relationship between early changes on anxiety and somatization items and treatment outcome at endpoint in depressed patients.8 Results showed that more pronounced early improvement in general somatic symptoms was significantly correlated with remission of depression after 12 weeks of treatment with fluoxetine.8

This work expands on the previous studies and explores the potential predictive value of early improvement on the SQ subscale scores for anxiety, depression, and anger/hostility as predictors of treatment response.

  METHODS

Our sample consisted of 169 depressed outpatients (52.7% female) between ages 18 and 65 (mean age, 40.3 ± 10.6 years) who were enrolled in an 8-week open treatment trial of fluoxetine 20 mg/d at the Depression Clinical and Research Program of the Massachusetts General Hospital. Subjects were required to meet criteria for MDD diagnosed with the Structured Clinical Interview for DSM-III-R, Patient Edition (SCID-P)9 and to have a HAMD-17 score >16 at baseline. All patients were administered the SQ10 at week 0 (baseline), week 2, week 4, and week 8 (endpoint). The SQ is a 92-item, self-rated instrument with symptom subscales, including Anxiety, Depression, and Anger/Hostility. The subscales have shown good sensitivity to change following treatment.10

Prior to enrollment, patients gave written consent to participate in the study. The consent form was approved by the Institutional Review Board of the Massachusetts General Hospital. The SCID-P and HAMD-17 interviews were conducted by psychiatrists and clinicians affiliated with the Depression Clinical and Research Program who were fully trained in the use of these instruments. During the 8 weeks of treatment with fluoxetine 20 mg/d, patients were seen every 2 weeks for safety and efficacy assessments.

The following criteria were exclusionary: pregnancy, breast-feeding, use of birth control pills, serious suicide risk, history of neurologic illness including seizure disorder, serious/unstable medical illness, organic mental disorders, substance use disorders active within the last year, schizophrenia, delusional disorder, bipolar disorder, antisocial personality disorder, mood-congruent or mood-incongruent psychotic features, history of multiple adverse drug reactions or allergy to the study drugs, current use of other psychotropic drugs, and clinical or laboratory evidence of hypothyroidism. Patients were also excluded if they had a previous nonresponse to or intolerance of fluoxetine (60 to 80 mg/d), or to the combination of fluoxetine and desipramine or fluoxetine and lithium, or if they had failed to respond during the course of their current major depressive episode to at least one adequate antidepressant trial (>6 weeks of treatment with either >150 mg of imipramine or its tricyclic equivalent, >60 mg of fluoxetine or its selective serotonin reuptake inhibitor equivalent, or >60 mg of phenelzine or its monoamine oxidase inhibitor equivalent).

Statistical analyses

We defined treatment response as a ≥50% reduction in HAMD-17 scores at study endpoint and early improvement on 3 SQ subscales (Anxiety, Depression, and Anger/ Hostility) as a >30% reduction in scores by week 2, similar to the 2004 study by Papakostas et al.11 For secondary analysis, we defined remission as having a HAMD-17 score of <8 at endpoint.

The logistic regression method was used to assess whether early improvement on each subscale independently predicted response to treatment. As a secondary analysis, we examined the relationship between early changes on the 3 SQ subscales and remission at endpoint, by regression analysis.

  RESULTS

Patient demographics are summarized in TABLE 1. Out of 169 patients, 102 met criteria for response (61.4%), and 94 individuals met criteria for remission (56.6%). The percentages of patients who obtained a 30% reduction in SQ scores at different times in the study are summarized in the FIGURE. For all 3 SQ subscales, the percentage of patients achieving a 30% reduction in scores increased from week 2 through endpoint (all tested by McNemar test, P < .05).

At each visit postbaseline, the difference in improvement rates between the Anger/ Hostility subscale and both the Anxiety and Depression subscales was significant at P < .05 (McNemar test, exact probability), and the difference in improvement rates between the Anxiety and Depression subscales was not significantly different (all P values > .05 by McNemar test, exact probability).

We assessed the relationship between treatment response and age, gender, and severity of depression at baseline. These variables were not significantly associated with treatment response (depression severity, P = .077; age, P = .976; gender, P = .329), nor were they significantly associated with remission (depression severity, P = .062; age, P = .922; gender, P = .251) and thus were not controlled for in the regressions. When the SQ subscales were entered individually into a logistic regression, early improvement, defined as >30% improvement from baseline, was predictive of response to treatment for all subscales (Depression, P = .024; Anxiety, P = .021; Anger/Hostility, P = .045; TABLE 2).

As a secondary analysis, we assessed whether early improvement in any of the subscales predicted remission at endpoint. When entered individually into a logistic regression, the Depression and Anxiety/Hostility subscales were significantly predictive of remission (Depression, P = .049; Anxiety/Hostility, P = .033; TABLE 3), but the Anxiety subscale was not.

As a post hoc analysis, we ran a survival analysis to look at whether trajectories of time to 30% improvement on any of the 3 SQ subscales differed between responders and nonresponders on the HAMD-17 and found no significant differences (there were also no significant differences when we compared remitters and nonremitters). Additionally, we examined the sensitivity and specificity of 3 different cutpoints of early improvement on the SQ hostility subscale (eg, 20%, 30%, and 40% improvements at week 2) for predicting ultimate treatment response; we found sensitivities of .66, .56, and .46, respectively, for the 3 cutpoints, with corresponding specificities of .52, .61, and, .69.


Table 1

Patient demographics and clinical variables (N = 169)

Characteristics n Percentage
Female 89 52.7
Marital status:
  Never married
28 16.6
Employed 101 59.8
Education:
  Some high school
  High school/GED
  ≥4 y college
2
14
81
1.2
8.3
48.0
Clinical variables Mean SD
Age, y 40.3 10.6
HAMD-17 baseline 19.88 4.526
HAMD-17 endpoint 8.50 4.642
HAMD-17: 17-item Hamilton Rating Scale for Depression.

FIGURE Percentage of patients with 30% reduction in SQ subscale scores at weeks 2, 4, and 8*

SQ: Symptom Questionnaire.
*All tested by McNemar test, P < .05.


Table 2

Early improvement* in the SQ subscales of Anxiety, Depression, and Anger/Hostility in relationship to response to treatment when entered independently into a logistic regression

SQ subscales Wald χ2 df P valuea Exp (B) Odds Ratio
Depression 5.084 1 .024 2.546
Anxiety 5.304 1 .021 2.468
Anger/Hostility 4.033 1 .045 2.039
SQ: Symptom Questionnaire.
*≥30%
aBy McNemar test.

Table 3

Early improvement* in the SQ subscales of Anxiety, Depression, and Anger/Hostility in relationship to remission when entered independently into a logistic regression

SQ subscales Wald χ2 df P valuea Exp (b) odds ratio
Depression 3.877 1 .049 2.154
Anxiety 3.207 1 .073 1.937
Anger/Hostility 4.554 1 .033 2.102
SQ: Symptom Questionnaire.
*≥30%
aBy McNemar test.

  DISCUSSION

Our results suggest that psychological symptoms of depression such as anger and hostility, as measured by the SQ, may respond to antidepressant treatment with fluoxetine earlier and perhaps more robustly as compared with other symptoms. Fava12 has hypothesized that antidepressants that affect serotonergic neurotransmission, which is involved in the modulation of aggressive behavior, may be particularly effective in depression with irritability and anger attacks, which may in part explain our findings.

Our results also show that early changes (>30% reduction by week 2) on the SQ Depression and Anger/Hostility subscales predicted response and remission, and that early changes on the Anxiety scale predicted response only. Replication with larger samples will be necessary to determine whether early changes in symptoms can predict outcome and influence treatment decisions.

Approximately 70% of patients with early improvement on the SQ subscales were eventual responders, compared with about 53% of those who did not experience early improvement on the SQ. Using a more liberal criterion of 20% early improvement on the SQ subscales will result in increased sensitivity. In terms of clinical relevance, early response would appear to suggest a greater probability of overall response/remission. However, it is also important to determine whether early nonresponse suggests eventual nonresponse as well, since this could also impact clinical decisions regarding the timing of treatment changes.

In addition, our findings suggest that early improvement in symptoms of anger and hostility can predict response and remission despite the fact that they are not core symptoms of MDD. Early improvements in anger and hostility may occur more rapidly than improvements in symptoms of anxiety and depression, and monitoring anger and hostility symptoms early in treatment could guide treatment decisions.

A potential limitation of the study is that we did not have a placebo control arm. Further studies, with placebo controls and a variety of drug treatments, are needed to confirm our preliminary findings.

  Conclusions

In summary, we have found that early improvement in the SQ subscales may predict response and remission to antidepressant treatment with fluoxetine among outpatients with MDD. If replicated in a more rigorous manner, this finding may help establish that early changes in several psychological symptoms of depression may be predictive of treatment response and remission and may allow clinicians to gauge the efficacy of antidepressant treatment earlier than the usual 6 to 8 weeks.

ACKNOWLEDGEMENTS: This research was supported in part by a grant to M. Fava from the National Institute of Mental Health (5R10MH056057-05).

DISCLOSURES: Dr. Farabaugh has received research support from Forest Pharmaceuticals.

Dr. Witte has received research support from Forest Pharmaceuticals.

Dr. Papakostas has served as a consultant for AstraZeneca, Bristol-Myers Squibb, Eli Lilly and Company, Evotec AG, GlaxoSmithKline, Inflabloc Pharmaceuticals, Jazz Pharmaceuticals, Otsuka Pharmaceuticals, Pamlab L.L.C., Pfizer Inc, Pierre Fabre Laboratories, Shire Pharmaceuticals, and Wyeth Pharmaceuticals. He has received honoraria from AstraZeneca, Bristol-Myers Squibb, Eli Lilly and Company, Evotec AG, GlaxoSmithKline, Inflabloc Pharmaceuticals, Jazz Pharmaceuticals, Lundbeck, Pamlab L.L.C., Pfizer Inc, Pierre Fabre Laboratories, Shire Pharmaceuticals, Titan Pharmaceuticals, and Wyeth Pharmaceuticals. He has received research support from Bristol-Myers Squibb, Forest Pharmaceuticals, the National Institute of Mental Health, Pamlab L.L.C., Pfizer Inc, and Ridge Diagnostic (formerly known as Precision Human Biolaboratories). He has served on the speakers bureau of Bristol-Myers Squibb and Pfizer Inc.

Dr. Fava has received research support from Abbott Laboratories, Alkermes, Aspect Medical Systems, AstraZeneca, Bristol-Myers Squibb, Cephalon, Clinical Trial Solutions L.L.C., Eli Lilly and Company, Forest Pharmaceuticals, Ganeden, GlaxoSmithKline, J & J Pharmaceuticals, Lichtwer Pharma GmbH, Lorex Pharmaceuticals, NARSAD, NCCAM, NIDA, NIMH, Novartis, Organon Inc., Pamlab L.L.C., Pfizer Inc, Pharmavite, Roche, Sanofi/Synthelabo, Shire Pharmaceuticals, Solvay Pharmaceuticals, and Wyeth Pharmaceuticals. He has served as a consultant/advisor for Abbott Laboratories, Affectis Pharmaceuticals, Amarinn, Aspect Medical Systems, AstraZeneca, Auspex Pharmaceuticals, Bayer AG, Best Practice Project Management, Inc., BioMarin Pharmaceuticals, Biovail Pharmaceuticals, Inc., BrainCells, Inc., Bristol-Myers Squibb, Cephalon, Clinical Trials Solutions L.L.C, CNS Response, Compellis, Cypress Pharmaceuticals, Dov Pharmaceuticals, Eisai, Eli Lilly and Company, EPIX Pharmaceuticals, Euthymics Bioscience, Inc., Fabre-Kramer Pharmaceuticals, Inc., Forest Pharmaceuticals, GlaxoSmithKline, Grunenthal GmBH, Janssen L.P., Jazz Pharmaceuticals, J & J Pharmaceuticals, Knoll Pharmaceutical Company, Labopharm, Lorex Pharmaceuticals, Lundbeck, MedAvante, Inc., Merck, Methylation Sciences, Neuronetics, Novartis, Nutrition 21, Organon, Pamlab, L.L.C., Pfizer Inc, PharmaStar, Pharmavite, Precision Human Biolaboratory, PsychoGenics, Psylin Neurosciences, Inc., Ridge Diagnostics, Inc., Roche, Sanofi/Synthelabo, Sepracor Inc., Schering-Plough, Solvay Pharmaceuticals, Somaxon, Somerset Pharmaceuticals, Takeda, Tetragenex, TransForm Pharmaceuticals, Inc., Transcept Pharmaceuticals, Vanda Pharmaceuticals, Inc., and Wyeth Pharmaceuticals. He has served on the speakers bureau of AstraZeneca, Boehringer Ingelheim, Bristol-Myers Squibb, Cephalon, Eli Lilly and Company, Forest Pharmaceuticals, GlaxoSmithKline, Novartis, Organon, Pfizer Inc, PharmaStar, and Wyeth Pharmaceuticals. He has equity holdings with Compellis.

Dr. Mischoulon has received research support from Ganeden, Laxdale (Amarin), and Nordic Naturals. He has served as a consultant/advisor for Bristol-Myers Squibb. He is a speaker for and has received writing support from Pamlab, L.L.C. He receives royalty/patent income from PMS Escape (patent co-holder).

Dr. Sonawalla, Mr. Johnson, Ms. Goodness, Ms. Clain, Dr. Baer, and Dr. Harley report no financial relationships with any company whose products are mentioned in this article or with manufacturers of competing products.

    REFERENCES

  1. Papakostas GI, McGrath P, Stewart J, et al. Psychic and somatic anxiety symptoms as predictors of response to fluoxetine in major depressive disorder. Psychiatry Res. 2008;161:116–120. [Epub ahead of print]
  2. Fraguas Jr, Marci C, Fava M, et al. Autonomic reactivity to induced emotion as potential predictor of response to antidepressant treatment. Psychiatry Res. 2007;151:169–172.
  3. Fava GA, Tomba E, Grandi S. The road to recovery from depression—don’t drive today with yesterday’s map. Psychother Psychosom. 2007;76:260–265.
  4. Fava M, Rush AJ. Current status of augmentation and combination treatments for major depressive disorder: a literature review and a proposal for a novel approach to improve practice. Psychother Psychosom. 2006;75:139–153.
  5. Katz MM, Tekell JL, Bowden CL, et al. Onset and early behavioral effects of pharmacologically different antidepressants and placebo in depression. Neuropsychopharmacology. 2004;29:566–579.
  6. Papakostas GI, Petersen T, Sklarsky KG. Timing of clinical improvement and symptom resolution in the treatment of major depressive disorder. Psychiatry Res. 2007;149:1–3.
  7. Hamilton M. A rating scale for depression. J Neurol Neurosurg Psychiatry. 1960;23:56–62.
  8. Farabaugh A, Mischoulon D, Fava M, et al. The relationship between early changes on the HAMD-17 anxiety/somatization factor items and treatment outcome among depressed patients. Int Clin Psychopharmacol. 2005;20:87–91.
  9. Spitzer RL, Williams JBW, Gibbon M. Instruction Manual for the Structured Clinical Interview for DSMIII-R (SCID), 4/1/87 Revision. New York, NY: Biometrics Research Department, New York State Psychiatric Institute; 1987.
  10. Kellner R. A symptom questionnaire. J Clin Psychiatry. 1987;48:268–274.
  11. Papakostas GI, Petersen T, Iosifescu D, et al. Somatic symptoms as predictors of time to onset of response to fluoxetine in major depressive disorder. J Clin Psychiatry. 2004;65:543–546.
  12. Fava M. Depression with anger attacks. J Clin Psychiatry. 1998;59:18–22.

CORRESPONDENCE: Amy Farabaugh, PhD, Director, Psychotherapy Research, Depression Clinical and Research Program, Massachusetts General Hospital, 50 Staniford Street, Suite 401 Boston, MA 02114 USA, E-MAIL:afarabaugh@partners.org