Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2015 May 7.
Published in final edited form as: J Affect Disord. 2013 Oct 16;0:538–542. doi: 10.1016/j.jad.2013.10.015

Changes in well-being and quality of life in a randomized trial comparing dynamic psychotherapy and pharmacotherapy for major depressive disorder

Sigal Zilcha-Mano 1, Ulrike Dinger 1,4, Kevin S McCarthy 2, Marna S Barrett 3, Jacques P Barber 1
PMCID: PMC4423535  NIHMSID: NIHMS532522  PMID: 24176534

Abstract

Background:

Major depressive disorder (MDD) is associated with a decrease in quality of life (QOL) and well-being. Therefore, researchers are increasingly complementing traditional symptom measurements with QOL and well-being assessments in order to broaden the evaluation of treatment outcomes. The current prospective study investigated the effectiveness of supportive-expressive therapy (SET), antidepressant medication (MED) and placebo (PBO) in improving QOL and well-being in patients with MDD.

Methods:

Data from a randomized controlled trial (trial registration: NCT00043550) comparing SET, MED and PBO for the treatment of depression (N = 156) were analyzed. Outcome measures addressed patients’ QOL and physical and mental well-being. Changes in outcomes were assessed across and between treatments using linear mixed models.

Results:

Across treatments, patients showed significant improvement in QOL and mental and physical health measures, as well as a reduction in interpersonal distress and depressive and anxiety symptoms (p ≤.002 for all measures). Those changes were not only the products of a decrease in depressive symptoms, but also predicted subsequent reduction in symptoms. No significant differences were found between the three treatment conditions.

Limitations:

The limitation is the study's moderate sample size.

Conclusions:

Current treatments for depression significantly improve patients' QOL and well-being. No significant differences were found between the three conditions examined in this study. The current study highlights the role of well-being in predicting subsequent symptomatic change.

Keywords: Depression, Secondary outcomes, Quality of life, Well-being, Psychodynamic Psychotherapy

Introduction

Major depressive disorder (MDD) has a lifetime prevalence of approximately 10%, and is currently the 4th leading cause of disability worldwide (Kessler, 2012). Depression is associated with a decrement in health that is significantly greater than that associated with other chronic diseases (Moussavi et al., 2007). More than 60% of patients with MDD have a clinically significant impairment in their quality of life (QOL) (Rapaport et al., 2005, Zeng et al., 2013). Medication (Fournier et al., 2010), psychodynamic psychotherapy (Barber et al., 2012), and even placebo (Walsh et al., 2002) have been shown to effectively reduce depressive symptoms. However, not to be overlooked is the ability of such treatments to increase QOL and the ability to actively participate in society.

In a recent randomized controlled trial (RCT) (Barber et al., 2012), supportive-expressive therapy (SET), antidepressant medication (MED) and placebo (PBO) were found to be equally effective in reducing depressive symptoms when treating patients with MDD. Taking into account the substantial effect of MDD on quality of life (Rapaport et al., 2005, Zeng et al., 2013), and the growing interest in complementing traditional symptom measures with additional QOL measures when evaluating treatment effectiveness (Ishak et al., 2011), we examined whether SET, MED and PBO in this setting had a significant effect on QOL and mental and physical well-being throughout treatment. In addition, we aimed to evaluate whether outcomes differed among the three treatment conditions. We further examined whether these measures are the products or predictors of changes in depressive symptoms.

Methods

Participants

Patients diagnosed with MDD were randomly assigned to one of three treatment conditions: SET, MED, or PBO (N = 156). Details regarding inclusion criteria and study procedures have been previously reported (Barber et al., 2012). The mean age was 37.5 (SD = 12.2) and 92 participants (59%) were female. Approximately half (48%) of the patients were Caucasian, 45% were African Americans, and the rest were Latino (5%) or Asian (2%). At intake 84.5% of patients had at least one comorbid Axis I disorder. Comorbidities included anxiety disorders (44.9%) and current substance abuse or past dependence disorder (35.3%). In addition, 46.2% had a comorbid Axis II personality disorder. The study (clinicaltrials.gov Identifier: NCT00043550) was approved by the Institutional Review Board, and all patients gave their informed consent in writing prior to screening.

Treatments

Treatments were provided for a total of 16 weeks. Patients treated with SET (N = 51) received 20 sessions of manualized psychodynamic therapy for depression (Luborsky et al., 1995) twice weekly during the first four weeks of treatment and weekly for the remainder of treatment. Treating psychotherapists had a minimum of 10 years of psychotherapy experience in SET. In the medication (MED; N = 55) and placebo (PBO; N = 50) conditions, patients received either Sertraline or placebo delivered by experienced psychopharmacologists using a manualized clinical management model (Fawcett et al., 1987); non-responders were switched to Venlafaxine (MED) or to a second placebo (PBO) after eight weeks. Patients in the MED and PBO were seen weekly for the first six weeks, after which they were seen every two weeks at the discretion of the treating psychopharmacotherapist.

Measures

Secondary outcomes included anxiety and depressive symptoms as well as well-being status, quality of life and interpersonal distress. Anxiety severity was measured with the structured interview guide for the Hamilton Anxiety Rating Scale (HARS; Hamilton, 1959) and with the Beck Anxiety Inventory 21-item self-report measure (BAI; Beck et al., 1988a). Severity of depressive symptoms was measured with the Beck Depression Inventory, a 21-item self-report measure (BDI; Beck et al., 1988b). Mental and physical health was measured with the Medical Outcomes Study 36-item Health Survey (Ware et al., 1994), which yields two summary scores developed from the original measure: the general mental health component score (MCS) and the physical health component score (PCS). Interpersonal distress was measured with the 64-item version of the Inventory of Interpersonal Problems (Horowitz et al., 2000), with the total mean of all items indicating the general level of interpersonal distress. Quality of life was measured with the 93-item Quality of Life Enjoyment and Satisfaction Questionnaire (Q-LESQ; Endicott, Nee, Harrison, & Blumenthal, 1993), which yields indices for various areas of functioning (i.e., physical health, mood, interpersonal relationships, household activities, and ability to complete work/ hobbies). In addition, the 17-item version of the clinician-administered Hamilton Rating Scale for Depression (HRSD; Hamilton, 1987) was used to examine the temporal relationship between the HRSD that serves as the primary outcome measure in treatments for depression and our secondary outcomes. All outcome measures were evaluated at six time points (week 0, 4, 6, 8, 12 and 16), excluding interpersonal distress, which was examined at three time points (week 0, 8 and 16).

Statistical Analyses

All analyses were carried out on an intention-to-treat basis, which included all participants randomized for the trial. Baseline differences in demographic and clinical characteristics between conditions were investigated using a one-way analysis of variance (ANOVA) for continuous variables and χ2 tests of independence/Fisher exact test for categorical variables.

Outcomes were analyzed using linear mixed models (SAS PROC MIXED), (Littell, Milliken, Stroup, Wolfinger, & Schabenberger, 2006), which take into account the unbalanced effect caused by missing observations. The analysis of each outcome consisted of two longitudinal models. The first model examined linear trends in the outcome variables over time for the entire sample (SET, MED, and PBO). The second model examined differences in this trend over time between type of treatments in predicting outcomes and included a variable for time, type of treatment (a three-category variable of condition), and the treatment × time interaction. Significant interactions were further investigated for specificity. We assumed random effects for both subject intercept and time effect. Because gender and minority status were found to be related to Hamilton Rating Scale for Depression symptom scores in the main outcome study, we repeated the data analyses controlling for gender and minority status. The sample enabled the detection of a medium effect size (0.48) with a power >80% when comparing MED or SET to PBO over the longitudinal period (Diggle, Liang, & Zeger, 1994). Within- and between-group effect sizes were computed as d, the standardized mean difference, and defined as small (d = 0.20), medium (d = 0.50), and large (d = 0.80; Cohen, 1988). Pattern-mixture models (Hedeker & Gibbons, 1997) were implemented to assess whether estimates per the linear mixed models were dependent on missing data patterns (for more details, see Gallop & Tasca, 2009).

Finally,to examine the bidirectional association between HRSD and the secondary outcome measures over time,we employed autoregressive cross-lagged modeling (ARCL;e.g., Collins and Sayer, 2001; Zilcha-Mano et al.,2014). In this ARCL model, HRSD and each one of the secondary outcomes were introduced in Level 1 of the models (within-subject) along with the dependent variable at the subsequent time (either HRSD or the secondary outcome, depending on the model).

Results

A total of 156 patients were included in the outcomes analyses. No differences between conditions were found for baseline demographics or clinical characteristics. Table 1 describes the multilevel models for changes in outcomes throughout time, showing improvement in all areas examined. Specifically, patients showed improvement in quality of life (across all the 7 subscales, p<.0001), mental health (p<.0001) and physical health (p<.002), as well as reductions in interpersonal distress (p<.0001), depressive symptoms (p<.0001) and anxiety symptoms (the latter were examined using both self-report and clinician-rated measures, p<.0001 for both measures).

Table 1.

Changes in Outcome Measures across Treatments

Outcome Mean
Pre-treatment (SD)		 Mean
Post-treatment (SD)		 F Df P Estimate
(S.E.)		 Pre-post effect size
MED SET PBO MED SET PBO MED SET PBO
MOS Physical
component
score		 70.52
(18.70)		 56.28
(21.25)		 62.64
(21.80)		 66.59
(18.70)		 71.03
(21.25)		 70.70
(21.80)		 9.50 1,130 P=.002 0.40
(0.13)		 −.21 .69 .36
Mental
component
score		 23.91
(13.04)		 26.37
(15.15)		 26.77
(13.58)		 50.54
(13.04)		 62.05
(15.15)		 53.80
(13.58)		 80.1 1,130 P<.0001 1.72
(0.19)		 2.04 2.35 1.99
IIP 92.15
(30.78)		 107.98
(35.99		 82.16
(36.65)		 87.30
(30.78)		 73.72
(35.99)		 70.44
(36.65)		 25.91 1,95 P<.0001 −0.97
(0.19)		 −.15 −.95 −.31
Q-les-Q Physical health .43
(.13)		 .35
(.17)		 .38
(.16)		 .53
(0.13)		 .58
(0.17)		 .50
(0.16)		 43.94 1,89 P<.0001 0.009
(0.001)		 .69 1.33 .75
Feeling .42
(.10)		 .38
(.14)		 .39
(.16)		 .56
(0.10)		 .62
(0.14)		 .55
(0.16)		 70.84 1,89 P<.0001 0.01
(0.001)		 1.25 1.65 .98
Work .49
(.19)		 .44
(.22)		 .47
(.16)		 .71
(0.19)		 .71
(0.22)		 .58
(0.16)		 39.42 1,68 P<.0001 0.01
(0.001)		 1.12 1.18 .66
Household
duties		 .47
(.17)		 .45
(.25)		 .41
(.19)		 .56
(0.17)		 .63
(0.25)		 .59
(0.19)		 55.40 1,74 P<.0001 0.01
(0.001)		 .55 .70 .96
Leisure .43
(.18)		 .37
(.22)		 .37
(.22)		 .59
(0.18)		 .62
(0.22)		 .52
(0.22)		 35.35 1,89 P<.0001 0.01
(0.001)		 .84 1.11 .62
Social
relationships		 .44
(.13)		 .40
(.16)		 .44
(.18)		 .53
(0.13)		 .62
(0.16)		 .54
(0.18)		 36.76 1,89 P<.0001 0.008
(0.001)		 .67 1.40 .54
General
activities		 .39
(.11)		 .34
(.14)		 .36
(.14)		 .53
(0.11)		 .59
(0.14)		 .52
(0.14)		 84.11 1,88 P<.0001 0.01
(0.001)		 1.25 1.71 1.09
HRSA 15.88
(4.74)		 17.66
(5.36)		 16.96
(5.06)		 10.06
(4.74)		 10.10
(5.36)		 9.52
(5.06)		 73.16 1,111 P<.0001 −0.39
(0.04)		 −1.22 −1.40 −1.47
BAI 9.42
(6.25)		 9.9
(6.66)		 9.08
(5.34)		 4.22
(6.25)		 2.87
(6.66)		 2.75
(5.34)		 91.14 1,131 P<.0001 −0.33
(0.03)		 −.83 −1.06 −1.18
BDI 32.09
(9.38)		 32.19
(9.86)		 29.11
(11.00)		 14.91
(9.38)		 9.68
(9.86)		 12.64
(11.00)		 165.29 1,139 P<.0001 −1.00
(0.07)		 −1.82 −2.27 −1.49
Open in a new tab

Note. Df = Degrees of Freedom: S.E. = Standard Error. Analyses were performed using PROC MIXED. For MOS and Q-les-Q, higher scores represent improvement in mental and physical health and in QOL, respectively. For IIP, HRSA, BAI and BDI lower negative scores represents a greater reduction in interpersonal distress and symptoms, respectively. Means at post-treatment as well as pre-post effect sizes were calculated based on the intent-to-treat data with post-treatment scores from the linear mixed model analyses.

Similar to the main outcome study, significant treatment differences in the time trends failed to be found for interpersonal distress, depressive and anxiety symptoms, and mental and physical health (all ps≥.09, see table 2). For quality of life, significant differences were only found for the Work subscale, in which patients receiving SET showed greater improvement than those in the PBO condition (p=0.01). This finding (d=.52 while comparing SET to PBO) was no longer significant after applying the Bonferroni correction. Results were unchanged when controlling for gender and minority status.

Table 2.

Changes in Outcome Measures Comparing MED and SET to PBO

Outcome F df P Estimate S.E Between-conditions effect size
MED SET MED SET MED SET
MOS Physical
component
score		 0.75 2,371 P-.47 −0.24 0.14 0.31 0.31 −.57 .33
Mental
component
score		 1.89 2,371 P=.15 0.70 −0.16 0.47 0.46 .05 .36
IIP 0.70 2,69 P=.49 −0.48 0.46 −0.47 0.46 .16 −.64
Q-les-Q Physical
health		 0.48 2,76 P=.61 0.001 0.003 0.003 0.003 −.06 .58
Feeling 0.19 2,75 P=.82 0.001 0.003 0.001 0.003 .27 .67
Work 3.32 2,53 P=.04 0.006 0.004* 0.01 0.004 .46 .52
Household
duties		 1.11 2,56 P=.33 −0.001 −0.004 0.003 0.003 −.41 −.26
Leisure 0.60 2,75 P=.55 0.002 0.004 0.004 0.004 .22 .49
Social
relationships		 1.89 2,75 P=.15 0.005 0.001 0.003 0.003 .13 .86
General
activities		 0.65 75,2 P=.52 0.003 0.001 0.002 0.002 .16 .62
HRSA 0.62 90,2 P=.53 0.10 −0.001 0.10 0.11 .25 .07
BAI 0.23 2,529 P=.79 −0.03 0.02 0.08 0.08 .35 .12
BDI 2.35 2,1639 P=.09 −0.35 −0.008 0.19 0.18 −.33 −.78
Open in a new tab

Note. F = represent the omnibus tests for the interactions with treatment condition. Df = Degrees of Freedom: S.E. = Standard Error. Analyses were performed using PROC MIXED. All effects of MED and SET are in comparison to PBO. For MOS and Q-les-Q, higher scores represent improvement in mental and physical health and in QOL, respectively. For IIP, HRSA, BAI and BDI lower negative scores represents a greater reduction in interpersonal distress and symptoms, respectively. Pre-post effect sizes were calculated based on the intent-to-treat data with post-treatment scores from the linear mixed model analyses.

In order to determine whether the lack of significant differences between treatments was driven by missing data, we classified two patterns of patients' available data (i.e., patients with and without data at week 16) and assessed the interaction of pattern, time, and treatment conditions using a linear mixed models analysis for each of the outcome measures. The pattern-mixture results were non-significant (ps>.14).

Finally, we employed autoregressive cross-lagged modeling (ARCL) on those measures that had sufficient validated data (MOS: PCS and MCS, BDI, BAI), in order to determine the temporal relationship between HRSD and secondary outcome measures. Findings show that for all the measures (MOS: PCS and MCS, BDI, BAI), higher well-being at a given time point predicted a lower HRSD level at the subsequent time point, while controlling for previous levels of HRSD (all psr.03). Additionally, for half of the measures (BAI, MCS), lower HRSD at a given time point predicted higher well-being at the subsequent time point, while controlling for previous levels of well-being.

Discussion

Depression is associated with a substantial decrease in QOL, and is a major contributor to decrement in general health (Rapaport et al., 2005, Moussavi et al., 2007). Medication, psychodynamic psychotherapy, and even placebo have been shown to effectively reduce observer-rated depressive symptoms of patients with MDD when compared in a recent RCT (Barber et al., 2012). However, whether these treatments lead to improvement in other areas of functioning in patients with MDD throughout and between treatments has not been previously evaluated. The results of this study showed that patients experienced not only a significant reduction in self-reported depressive symptoms and observer-rated and self-reported anxiety symptoms, but also a significant improvement in many other aspects of life, including work productivity, social relationships and vitality.

Although the three treatments presumably have different underlying mechanisms and distinct procedures, they showed equivalent efficacy not only in reducing symptoms but also in other more broad parameters of clinical progress. Importantly, the changes through time in well-being found in this study cannot be interpreted just as the consequences of symptomatic change (reduction in depressive symptoms predicting changes in other life domains), since well-being was found to predict symptomatic change and was not merely the products of such a change. Therefore, other explanations should be raised, such as the common factors explanation which suggests common mechanisms of change between different treatments (e.g., Kazdin, 2007). Specifically, all patients met a professional (either a psychopharmacologist in clinical management or psychotherapist in SET) who expressed interest in their well-being. The treatment provider may have also encouraged optimism about the possibility of positive change as well as a plausible rationale for the patient's symptoms (either related to medical or interpersonal causes).

Although the analyses suggest that missing data did not contribute to our findings, methodological shortcomings such as sample size may have contributed to our failure to find differences between treatments. For instance, although our sample was larger than many other studies and enabled us to detect large to medium effects between the conditions, it did not allow for the detection of small effects.

In sum, the possibility that different but equally effective treatments in reducing symptoms will yield different outcomes when broadening the scoop to well-being and QOL was not supported. Rather, consistent with other reports in the literature (e.g., Cuijpers et al., 2013), it provided evidence that distinct treatments for depression works equally well regardless of the specific outcome being measured. The current study also shows rare evidence that the changes in well-being were not just the "byproduct" of reduction in depressive symptoms, but were also the predictors of symptomatic change— and therefore are important in the assessment of the effects of treatments for depression.

Acknowledgments

We are grateful to the research team at the University of Pennsylvania and to those colleagues who were involved in and supported the primary trial as well as to the generous support of the Fulbright Program.

Role of funding source

The original study was conducted with support from NIMH grant R01 MH 061410 to J.P. Barber (P.I.). The sertraline and pill placebos were provided by a grant from Pfizer Corp. The current study was written with support from the Fulbright Program to S. Mano-Zilcha.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Conflict of interest

Dr. Zilcha-Mano has received funding from the Fulbright Program. Dr. Barber has received funding from the National Institute of Mental Health (NIMH), and the National Institute on Drug Abuse (NIDA); authors’ fees from Guilford Press, Basic Books, and Cambridge University Press; and honoraria from Lundbeck. The other authors declare no conflict of interest.

Contributors

Sigal Zilcha-Mano developed the research idea for the current study, conducted statistical analyses and wrote the first version of the manuscript. Kevin S. McCarthy contributed to statististical analyses and supported data management. Marna Barrett was responsible for patient recruitment and data management. Ulrike Dinger contributed to the writing process. Jacques P. Barber was P.I. of the original trial and supervised the present analyses. All authors read previous versions of the manuscript, provided feedback and approved the final version of the manuscript.

References

  1. Barber JP, Barrett MS, Gallop R, Rynn MA, Rickels K. Short-term dynamic psychotherapy versus pharmacotherapy for major depressive disorder: a randomized, placebo-controlled trial. J Clin Psychiatry. 2012;73:66–73. doi: 10.4088/JCP.11m06831. [DOI] [PubMed] [Google Scholar]
  2. Beck AT, Epstein N, Brown G, Steer RA. An inventory for measuring clinical anxiety: psychometric properties. J Consult Clin Psychol, 1988a;56:893–7. doi: 10.1037//0022-006x.56.6.893. [DOI] [PubMed] [Google Scholar]
  3. Beck AT, Steer RA, Garbin MG. Psychometric properties of the Beck Depression Inventory: Twenty-five years later. Clinical Psychology Review, 1988b;8:77–100. [Google Scholar]
  4. Cohen J. Statistical Power Analysis for the Behavioral Sciences. 2nd Erlbaum; Hillsdale, NJ: 1988. [Google Scholar]
  5. Collins LM, Sayer AG. New methods for the analysis of change. American Psychological Association; Washington, DC: 2001. [Google Scholar]
  6. Cuijpers P, Sijbrandij M, Koole SL, Andersson G, Beekman AT, Reynolds CF. The efficacy of psychotherapy and pharmacotherapy in treating depressive and anxiety disorders: a meta-analysis of direct comparisons. World Psychiatry. 2013;12:137–48. doi: 10.1002/wps.20038. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Diggle PJ, Liang K-Y, Zeger SL. Analysis of Longitudinal Data. Oxford: Oxford Science; 1994. [Google Scholar]
  8. Endicott J, Nee J, Harrison W, Blumenthal R. Quality of Life Enjoyment and Satisfaction Questionnaire: A new measure. Psychopharmacology Bulletin. 1993;29:321–326. [PubMed] [Google Scholar]
  9. Fawcett J, Epstein R, Fiester SJ, Elkin I, Autry JH. Clinical management--imipramine/placebo administration manual. Nimh treatment of depression collaborative research program. Psychopharmacol Bull. 1987;23:309–24. [PubMed] [Google Scholar]
  10. Fournier JC, Derubeis RJ, Hollon SD, Dimidjian S, Amsterdam JD, Shelton RC, Fawcett j. Antidepressant drug effects and depression severity: a patient-level meta-analysis. JAMA. 2010;303:47–53. doi: 10.1001/jama.2009.1943. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Gallop R, Tasca GA. Multilevel modeling of longitudinal data for psychotherapy researchers: II. The complexities. Psychotherapy Research. 2009;19:438–452. doi: 10.1080/10503300902849475. [DOI] [PubMed] [Google Scholar]
  12. Hedeker D, Gibbons RD. Application of random effects pattern-mixture models for missing data in longitudinal studies. Psychological Methods. 1997;2:64–78. [Google Scholar]
  13. Hamilton MA. The assessment of anxiety status by rating. British Journal of Medical Psychology. 1959;32:50–55. doi: 10.1111/j.2044-8341.1959.tb00467.x. [DOI] [PubMed] [Google Scholar]
  14. Hamilton M. Development of a rating scale for primary depressive illness. British Journal of Social and Clinical Psychology. 1967;6:278–296. doi: 10.1111/j.2044-8260.1967.tb00530.x. [DOI] [PubMed] [Google Scholar]
  15. Horowitz LM, Alden LE, Wiggins JS, Pincus AL. Inventory of Interpersonal Problems. The Psychological Corporation; Odessa, FL: 2000. [Google Scholar]
  16. Ishak WW, Ha K, Kapitanski N, Bagot K, Fathy H, Swanson B, Vilhauer J, Balayan K, Bolotaulo NI, Rapaport MH. The impact of psychotherapy, pharmacotherapy, and their combination on quality of life in depression. Harv Rev Psychiatry. 2011;19:277–89. doi: 10.3109/10673229.2011.630828. [DOI] [PubMed] [Google Scholar]
  17. Kazdin AE. Mediators and mechanisms of change in psychotherapy research. Annu Rev Clin Psychol. 2007;3:1–27. doi: 10.1146/annurev.clinpsy.3.022806.091432. [DOI] [PubMed] [Google Scholar]
  18. Kessler RC. The costs of depression. Psychiatr Clin North Am. 2012;35:1–14. doi: 10.1016/j.psc.2011.11.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Luborsky L, Mark D, Hole AV, Popp C, Goldsmith B, Cacciola J. Supportive-Expressive Dynamic Psychotherapy of Depression: A time-limited version. In: Barber JP, Crits-Christoph P, editors. Psychodynamic Psychotherapies for psychiatric disorders(Axis I) Basic Books; New York: 1995. pp. 13–42. [Google Scholar]
  20. Moussavi S, Chatterji S, Verdes E, Tandon A, Patel V, Ustun B. Depression, chronic diseases, and decrements in health: results from the world health surveys. Lancet. 2007;370:851–8. doi: 10.1016/S0140-6736(07)61415-9. [DOI] [PubMed] [Google Scholar]
  21. Rapaport MH, Clary C, Fayyad R, Endicott J. Quality-of-life impairment in depressive and anxiety disorders. Am J Psychiatry. 2005;162:1171–8. doi: 10.1176/appi.ajp.162.6.1171. [DOI] [PubMed] [Google Scholar]
  22. Walsh BT, Seidman SN, Sysko R, Gould M. Placebo response in studies of major depression :variable, substantial, and growing. JAMA. 2002;287:1840–7. doi: 10.1001/jama.287.14.1840. [DOI] [PubMed] [Google Scholar]
  23. Ware JE, Sherbourne CD. The MOS 36-item Short-Form Health Survey (SF-36): I. Conceptual framework and item selection. Medical Care. 1992;30:473–83. [PubMed] [Google Scholar]
  24. Zeng Q, Xu Y, Chun Wang W. Quality of life in outpatients with depression in china. J Affect Disord. 2013;150:513–21. doi: 10.1016/j.jad.2013.04.052. [DOI] [PubMed] [Google Scholar]
  25. Zilcha-Mano S, Dinger U, McCarthy KS, Barber JP. Does alliance predict symptomatic change, or is it the other way around? Journal of Counseling and Clinical Psychology. 2014;82:931–935. doi: 10.1037/a0035141. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES