Black Lightning: Season ... | Filmy 3D / SBS | Back to login

Different levels of lack of improvement at 4 weeks of escitalopram treatment as predictors of poor 8-week outcome in MDD

Different levels of lack of improvement at 4 weeks of escitalopram treatment as predictors of poor 8-week outcome in MDD

Journal of Affective Disorders 146 (2013) 433–437 Contents lists available at SciVerse ScienceDirect Journal of Affective Disorders journal homepage...

218KB Sizes 0 Downloads 0 Views

Journal of Affective Disorders 146 (2013) 433–437

Contents lists available at SciVerse ScienceDirect

Journal of Affective Disorders journal homepage: www.elsevier.com/locate/jad

Brief report

Different levels of lack of improvement at 4 weeks of escitalopram treatment as predictors of poor 8-week outcome in MDD I. Gilaberte a,n, I. Romera a,b, V. Perez-Sola c, J.M. Menchon d, A. Schacht e a

Eli Lilly and Company, Madrid, Spain ´noma de Barcelona, Barcelona, Spain Universidad Auto c Hospital de Sant Pau i de la Santa Creu, UAB, CIBERSAM, Barcelona, Spain d Hospital Universitari de Bellvitge, CIBERSAM, Hospitalet de Llobregat, Barcelona, Spain e Eli Lilly and Company, Bad Homburg, Germany b

a r t i c l e i n f o

abstract

Article history: Received 15 June 2012 Received in revised form 26 July 2012 Accepted 26 July 2012 Available online 23 August 2012

Background: Several post-hoc studies have shown that lack of early improvement reduces the chance of later response or remission. This post-hoc analysis evaluates different cut-off points of non-improvement at 4 weeks of escitalopram treatment to predict 8-week non-response and non-remission. Method: This study consisted of MDD patients with an absence of improvement ( o30% reduction in baseline score of the HAMD-17) at Week 4 of escitalopram treatment (10 mg/day) that continued escitalopram treatment (10–20 mg/day) for a further 4-week period (n ¼ 251). Predictive, sensitivity and specificity values for the several definitions of non-improvement (r 25%, r 20% and r 15% reduction in the HAMD-17 baseline total score) at 4 weeks were calculated. Results: Overall, 70.1% (176/251) of patients did not achieve response at Week 8 and 84.5% (212/251) did not achieve remission. The predictive value for non-response was high (71.4–74.3%) for all cut-off points of non-improvement tested. The respective values for non-remission were placed between 85.0% and 87.2%. Limitations: This was a post-hoc subgroup analysis. The only drug assessed was escitalopram. Conclusions: Our data indicate that an absence of improvement, o 30% reduction in the HAMD-17, after 4 weeks of escitalopram treatment should prompt clinicians to consider a change in treatment strategy. Similar findings were previously reported for other antidepressants. & 2012 Elsevier B.V. All rights reserved.

Keywords: Depression Non-improvement Non-response Non-remission Treatment strategy

1. Introduction Several post-hoc studies have shown that lack of early improvement reduces the chance of later response or remission (Szegedi et al., 2003,2009; Nierenberg et al., 1995; Baldwin et al., 2009). Nierenberg et al. reported that lack of improvement ( o 30% decrease in the Hamilton Rating Scale for Depression [HAMD]) after 4 weeks of fluoxetine treatment was associated with an 80% chance of non-response after 8 weeks (Nierenberg et al., 2000). Similarly, Szegedi et al. conducted a study with mirtazapine and paroxetine and showed that the probabilities of not achieving a later stable response were 97% and 92% for each treatment respectively, when patients did not improve ( o 30% reduction on the HAMD) after 3 weeks of treatment (Szegedi et al., 2003). More recently, Baldwin et al. showed that the probability of non-response at 8 weeks, in patients that

n Correspondence to: Clinical Research Department, Eli Lilly and Company, Madrid, Spain. Tel.: þ34 916233340; fax: þ 34 91 6635231. E-mail addresses: [email protected], [email protected] (I. Gilaberte).

0165-0327/$ - see front matter & 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jad.2012.07.035

failed to demonstrate an onset of escitalopram treatment ˚ effect (decrease in Montgomery-Asberg Depression Rating Scale [MADRS] total score o 20%) at 4 weeks, was about 80% (Baldwin et al., 2009). These studies indicate that if improvement is not observed, clinicians should make a decision about the treatment strategy as early as after 4 weeks of treatment (NICE, 2010). However, what degree of non-improvement should trigger that therapeutic decision? Previous studies, were focused on the predictive value of early improvement for later response and remission (Baldwin et al., 2009; Nierenberg et al., 2000; Szegedi et al., 2003,2009; Lin et al., 2011; Henkel et al., 2009), evaluating several definitions of early improvement ( Z20%, Z25%, Z30%, Z50% reduction in the HAMD or MADRS baseline total score) at different time-points (at Weeks 1, 2, 3, 4). However, despite all these studies, the degree of non-improvement necessary to predict a poor outcome is not clear. Knowing the level of non-improvement that would indicate a low benefit of prolongation of the current treatment is of clinical value. Such an indicator could guide clinicians towards recognising patients that would benefit most from an early change in the treatment approach.

434

I. Gilaberte et al. / Journal of Affective Disorders 146 (2013) 433–437

This post-hoc analysis evaluates different cut-off points of non-improvement ( r25%, r20% and r15% reduction in the HAMD-17 baseline total score) at Week 4 of escitalopram treatment, to confirm its predictive value for 8-week non-response and non-remission. Predictive, sensitivity and specificity values for the different definitions of non-improvement were calculated.

2. Method This post-hoc analysis was conducted in a sub-sample of patients from a larger, randomised, double-blind study. The study evaluated two antidepressant switching strategies (early switch [n¼282]; at 4 weeks of treatment vs. conventional switch [n¼284]; at 8 weeks of treatment) in patients that did not experience improvement (o30% reduction in the HAMD-17 baseline total score) after one month of escitalopram (10 mg/day) treatment (n¼566/840) (Romera et al., 2012). The study was approved by the ethical review boards at each centre and was conducted in accordance to the Declaration of Helsinki.

that patients with non-improvement will probably remain nonresponders/non-remitters at Week 8. The NPV is the percentage of patients that have response/remission at Week 8 within the patients that have early improvement at Week 4. A high NPV signifies that patients improving at Week 4 will probably remain responders/remitters at Week 8. The sensitivity is the percentage of patients with non-response/non-remission at Week 8 that had non-improvement at Week 4. The specificity is the percentage of patients with response/remission at Week 8 that had predefined early improvement at Week 4. In addition, two ROC (receiver operating characteristic) curves for non-response and non-remission were plotted using the HAMD-17 levels of non-improvement at week 4. Statistical analysis was performed by SAS.

3. Results 3.1. Demographic characteristics

2.1. Patient population The study population consisted of male or female outpatients aged Z18 years who met the diagnostic criteria for single or recurrent episodes of major depressive disorder (MDD) according to the Diagnostic and Statistical Manual of Mental Disorders, 4th edition (DSM-IVs-TR) (American Psychiatric Association, 2000). Written, informed consent was obtained from all patients before participation in the original trial. For a detailed list of exclusion criteria see Romera et al. (2012). Starting at baseline, all patients were treated with 10 mg/day escitalopram for 4 weeks (n ¼840). Those patients with an absence of improvement at Week 4 ( o30% reduction in baseline score of the HAMD-17) continued the study (n ¼566). The present post-hoc analysis includes only those patients that were randomised at 4 weeks to continue with escitalopram treatment (10–20 mg/day) for a further period of 4 weeks (conventional switch; n ¼284) and those that completed the study (n ¼251). 2.2. Outcome measures To predict 8-week non-response and non-remission, according to previous studies (Henkel et al., 2009; Szegedi et al., 2003) 3 different cut-off points of non-improvement (r25%, r20%, and r15% reduction in the HAMD-17 baseline total score) at Week 4 of escitalopram treatment were evaluated. Non-response was defined as o50% reduction in baseline total HAMD-17 score at Week 8. Nonremission was defined as an HAMD-17 total score of 47 at Week 8.

Patients (n ¼284) randomized to the conventional switch strategy (therefore, that continued with escitalopram treatment (10–20 mg/day) for a further period of 4 weeks), had a mean age (SD) of 47.4 (12.7), 197 (69.4%) were female and presented with a mean number (SD) of previous episodes of depression of 2.4 (2.5). Mean total score (SD) of the HAMD-17 was 24.4 (3.8) indicating moderate to severe depression. The dose of 10 mg of escitalopram was maintained in 30.6% of the patients, whereas the dose was increased to 20 mg in 62.3% of the patients. Of these 284 patients, 251 stayed on escitalopram treatment at Week 8 and were included in the present analysis.

3.2. Treatment outcomes Despite that 62.3% of patients had a dosage increase to 20 mg/day, 70.1% (176/251) of patients in this sample of non-improvers at Week 4, did not achieve response at Week 8, and 84.5% (212/251) did not achieve remission. Rates of non-response/non-remission at Week 8, given various levels of improvement at Week 4, are shown in Fig. 1. Regardless of the cut-off point for the definition of nonimprovement (r25%, r20% and r15%), a high percentage of patients ended-up as non-responders and non-remitters, with small differences between groups.

The primary outcome of the present analysis was the assessment of the prognostic value of different levels of early nonimprovement at Week 4 for achieving non-response and nonremission at Week 8 of treatment. A cross-tabulation of the final condition (i.e. the non-response/non-remission criteria at Week 8) with the test outcome (i.e. the lack of improvement criteria at Week 4) was performed. Based on this cross-tabulation, the positive predictive value (PPV), the negative predictive value (NPV), sensitivity, and specificity were calculated. Analyses were performed on the completers population (n ¼251) In this study, the PPV is the percentage of patients that have non-response/non-remission at Week 8 within the patients that have non-improvement at Week 4 (i.e. that non-improvement indeed predicts non-response/nonremission). A high PPV signifies

Percentage of patients

2.3. Statistical analysis

100 90 80 70 60 50 40 30 20 10 0

8-Week Non-response

≤15%

≤ 20%

8-Week Non-remission

≤ 25%

Overall (<30%)

Levels of non-improvement at 4 weeks Fig. 1. Rates of non-response/non-remission at 8 weeks given various levels of non-improvement at 4 weeks (n¼ 252).

I. Gilaberte et al. / Journal of Affective Disorders 146 (2013) 433–437

435

Table 1 Prediction of 8-week non-response/non-remission using different levels of improvement at Week 4 (n¼ 251).

Predicted outcome:8-week non-response r15% 4-week improvement (n¼ 148) r20% 4-week improvement (n¼197) r25% 4-week improvement (n¼ 234) Predicted outcome: 8-week non-remission r15% 4-week improvement (n¼ 148) r20% 4-week improvement (n¼197) r25% 4-week improvement (n¼ 234)

Sensitivity % (95% CI)

Specificity % (95% CI)

PPV % (95% CI)

NPV % (95% CI)

62.5 (55.5; 69.7) 80.1 (74.2; 86.0) 94.9 (91.6; 98.1)

49.3 (38.0; 60.6) 25.3 (15.5; 35.2) 10.7 (3.7; 17.7)

74.3 (67.3; 81.4) 71.6 (65.3; 77.9) 71.4 (65.6; 77.2)

35.9 (26.7; 45.2) 35.2 (22.4; 47.9) 47.1 (23.3; 70.8)

60.8 (54.3; 67.4) 79.2 (73.8; 84.7) 93.9 (90.6; 97.1)

51.3 (35.6; 67.0) 25.6 (11.9; 39.3) 10.3 (0.7; 19.8)

87.2 (81.8; 92.6) 85.3 (80.3; 90.2) 85.0 (80.5; 89.6)

19.4 (11.8; 27.1) 18.5 (8.2; 26.6) 23.5 (3.4; 43.7)

NPV: negative predictive value; PPV: positive predictive value.

3.3. Predictive value of non-improvement

Sensitivity

1.0

In this post-hoc analysis, our objective was to investigate if any of 3 non-improvement cut-off points at Week 4 of treatment ( r25%, r20% and r15% reduction in the HAMD-17 baseline total score) could serve as a useful predictor of non-response and non-remission at Week 8. To this end, we determined the sensitivity, specificity and predictive values of each cut-off point of non-improvement at Week 4 of escitalopram treatment, to predict 8-week non-response and non-remission. Predictive values for 8-week non-response and non-remission were 470% and 480% respectively, for any studied level of nonimprovement ( r25%, r20% and r15% reduction in the HAMD17 baseline total score). The values were also similar between the 3 studied levels of 4-week non-improvement. This indicates a high probability of poor 8-week outcome in case of non-improvement at Week 4, regardless of the cut-off point for non-improvement. Knowing how to reliably assess non-improvement and to predict the proportion of non-improvers that will not achieve response/remission is very important in allowing the physician to

25%

x

20%

x 15%

0.0

0.1

0.2

0.3

0.4 0.5 0.6 1-Specificity

0.7

0.8

0.9

1.0

ROC Curve for Non-Remission at Week 8 Area under the curve = 0.5910

1.0

x

0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0

4. Discussion

x

0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0

Sensitivity

Sensitivity, specificity, positive predictor values (PPV) and negative predictor values (NPV) were calculated for the 3 different cut-off points of non-improvement at Week 4 (Table 1). The PPV for not achieving response and remission was high and almost stable in the 3 studied levels of improvement ( r15%, r20%, r25%). PPV for non-response were 470% and PPV for nonremission were approximately 85%. This indicates a high probability of poor outcome in the case of non-improvement at Week 4 of treatment. Non-improvement of any of the 3 cut-off points ( r15%, r20%, r25%) was also a sensitive predictor of later nonresponse or non-remission. Sensitivity was greater for higher cutoff levels, indicating a higher probability of correctly identifying non-response or non-remission. Therefore, for a r20% cut-off point of improvement, about 80% of patients with non-response/ non-remission at Week 8 had non-improvement at Week 4. This percentage increased above 90% for a cut-off-point of r25%. NPV for non-response and non-remission (i.e. the chance of having a response/remission in case of improvement at Week 4) were low at all studied cut-off points, since the study population only included patients with lack of improvement at Week 4 (r30% reduction in the HAMD-17 baseline score). Similarly, low specificity (i.e. correctly identifying remission) values were obtained. The ROC analyses (Fig. 2) show the sensitivity and specificity change between the cut-off points of 15%, 20% and 25%. The area under the curve (AUC) for both non-response and non-remission was 0.591 and the p-values based on the logistic regression (Wald test) were p¼0.0150 for non-response and p¼0.0726 for non-remission.

ROC Curve for Non-Response at Week 8 Area under the curve = 0.5908

25%

x 20%

x

0.0

0.1

0.2

0.3

15%

0.4 0.5 0.6 1-Specificity

0.7

0.8

0.9

1.0

Fig. 2. (A) ROC curve for non-response at week 8. (B) ROC curve for non-remission at week 8.

make earlier decisions with regard to treatment intervention. The reasonable predictive value for non-response/non-remission that should trigger a treatment decision is between 70% and 80% (NICE, 2010). Therefore, our results support recent guidelines where an action in the treatment regimen or strategy in the first 3 to 4 weeks of treatment could be considered (NICE, 2010; Bauer et al., 2007). Our study results are also in agreement with previous studies that used different approaches and medications (other selective serotonin re-uptake inhibitors [SSRIs], mirtazapine) and showed the value of early non-improvement to predict later non-response and non-remission (Szegedi et al., 2003,2009; Nierenberg et al., 1995; Baldwin et al., 2009). However, most of the previous studies are mainly focused on the predictive value of early

436

I. Gilaberte et al. / Journal of Affective Disorders 146 (2013) 433–437

improvement (defined as an improvement of 20% in the first 2 weeks) for later response and remission (Henkel et al., 2009; Baldwin et al., 2009; van Calker et al., 2009; Tadic´ et al., 2010; Kim et al., 2011). From a clinical point of view, it is more important to identify as soon as possible in the treatment period, patients who will not respond, in order to take action earlier and increase the likelihood of improvement. It is worth mentioning that most of the evidence regarding early onset theory and its clinical implications comes from posthoc analyses. Only two prospective studies have assessed the clinical benefit of taking early treatment decisions in case of early non-improvement (Nakajima et al., 2011; Romera et al., 2012), and both indicate certain degree of benefit of conducting an early switch in early non-improvement. Other treatment plans after early nonimprovement, combination or augmentation, could be considered, however current evidence in that regard is scarce (Nakajima et al., 2009). In our study population of non-improvers, the chances of poor outcome were high, despite the fact that 62.3% of the patients were titrated to 20 mg of escitalopram from Week 4 to Week 8. This is consistent with several trials that have shown no benefit of dose escalation after a failure to respond to initial SSRI treatment (Corruble and Guelfi, 2000; Schweizer et al., 2001). Recent post-hoc analyses have found that lack of improvement ( r20% reduction on the HAMD-17) after 2 weeks of treatment is a good indicator of high chances of non-response and nonremission (Szegedi et al., 2009). In addition, improvement of at least 20% after 2 weeks of treatment predicted with high sensitivity later response and remission (Henkel et al., 2009; van Calker et al., 2009). Although these findings are placed after only two weeks of treatment, they are in agreement with our study results and support the fact that treatment decisions, in case of lack of improvement, can be done earlier than conventionally thought. We consider several limitations in the interpretation of our results. In our study, we were addressing a selected patient population in which patients, with a clear early improvement of Z30% in the HAMD-17 per protocol population, discontinued from the study. This explains the asymmetrical distribution of PPV vs. NPV and sensitivity vs. specificity. Therefore NPV and specificity results should be interpreted accordingly. In addition, all patients were treated with escitalopram, so results may not be generalised to other antidepressants. It should be mentioned however, that there are no similar studies done with escitalopram. Escitalopram doses were flexible, so results should be interpreted accordingly. In addition, we assessed the cut-off point only at Week 4 and not at earlier weeks (Henkel et al., 2009; van Calker et al., 2009). However, 4 weeks is a usual timeframe considered by clinicians for treatment evaluation. Finally, the results of our analysis should be interpreted taking into account that patients had moderate to severe depression.

5. Conclusions On the basis of this analysis, it can be concluded that patients displaying a lack of improvement at Week 4 of escitalopram treatment (o30% reduction in the HAMD-17) have a low probability of experiencing a response at Week 8, in spite of escitalopram dose optimisation. On the basis of our data and previous reports, a change in treatment strategy as early as 4 weeks after start of treatment could be justified, if non-improvement is observed, to increase the likelihood of response and remission of patients with MDD.

Role of funding source The present study has been fully supported by funding from Eli Lilly and Company.

Conflict of interest Dr. Irene Romera, Dr. Alexander Schacht, and Dr. Inmaculada Gilaberte are full time employees of Eli Lilly. Dr. Irene Romera is also an affiliate with the Universidad Autonoma de Barcelona, Departamento de Psiquiatrı´a. Dr. Victor Perez-Sola has received grant support from Eli Lilly, Lundbeck, Boehringer, Pfizer, Astra Zeneca, and GSK; has received honoraria from servier, Eli Lilly, BMS-Otsuka, GSK, Astra Zeneca, and Boehringer; has served as a consultant for and/or on advisory boards for Eli Lilly, BMS, and Astrazeneca. Dr. Jose M Mencho´n declares that he has received funding from the University of Barcelona, the Bellvitge University Hospital-IDIBELL, and from the Spanish Ministry of Science and Innovation, Instituto de Salud Carlos III, Centro de Investigacion en Red de Salud Mental (CIBERSAM). He has also received honoraria as a consultant from Almirall, Astrazeneca, Boehringer Ingelheim, Eli Lilly, Janssen, Lundbeck, and Wyeth.

Acknowledgements We thank Maria Isabel Vazquez, Belen Yruretagoyena, Pepa Polavieja, and Deborah Quail from Lilly for their help with the protocol development and study implementation; Dr James Russell, Dr David Perahia, Dr Jesu´s Hernandez from Eli Lilly and Dr Jose´ Antonio Sacristan from Lilly Spain who provided medical advice on the study design (Jesu´s Hernandez is currently working for the Centro de Investigacio´n Me´dica Aplicada de la Universidad de Navarra, Spain); Dr. Thomas Wagner of Trilogy Writing & Consulting GmbH (Frankfurt, Germany) who provided medical writing support; and the principal investigators, and the patients for their voluntary participation.

References American Psychiatric Association, 2000. Diagnostic and Statistical Manual of Mental Disorders, Text Revision, 4th ed. Author, Washington, DC. Baldwin, D.S., Stein, D.J., Dolberg, O.T., Bandelow, B., 2009. How long should a trial of escitalopram treatment be in patients with major depressive disorder, generalised anxiety disorder or social anxiety disorder? An exploration of the randomized controlled trial database. Human Psychopharmacology-Clinical and Experimental 24, 269–275. ¨ Bauer, M., Bschor, T., Pfennig, A., Whybrow, P.C., Angst, J., Versiani, M., Moller, H.J., 2007. WFSBP Task Force on Unipolar Depressive Disorders, 2007. World Federation of Societies of Biological Psychiatry (WFSBP) guidelines for biological treatment of unipolar depressive disorders in primary care. World Journal of Biological Psychiatry 8, 67–104. Corruble, E., Guelfi, J.D., 2000. Does increasing dose improve efficacy in patients with poor antidepressant response: a review. Acta Psychiatrica Scandinavica 101, 343–348. ¨ Henkel, V., Seemuller, F., Obermeier, M., Adli, M., Bauer, M., Mundt, C., Brieger, P., ¨ Laux, G., Bender, W., Heuser, I., Zeiler, J., Gaebel, W., Mayr, A., Moller, H.J., Riedel, M., 2009. Does early improvement triggered by antidepressants predict response/remission? Analysis of data from a naturalistic study on a large sample of inpatients with major depression. Journal of Affective Disorders 115, 439–449. Kim, J.M., Kim, S.Y., Stewart, R., Yoo, J.A., Bae, K.Y., Jung, S.W., Lee, M.S., Yim, H.W., Jun, T.Y., 2011. Improvement within 2 weeks and later treatment outcomes in patients with depressive disorders: the CRESCEND study. Journal of Affective Disorders 129, 183–190. Lin, C.-H., Lane, H.-Y., Chen, C.-C., Han Juo, S.-H., Yen, C.-F., 2011. Early prediction of fluoxetine response for Han Chinese inpatients with major depressive disorder. Journal of Clinical Psychopharmacology 31, 187–193. Nakajima, S., Suzuki, T., Watanabe, K., Kashima, H., Uchida, H., 2009. Accelerating response to antidepressant treatment in depression: a review and clinical suggestions. Progress in Neuro-Psychopharmacology and Biological Psychiatry 34, 259–264. Nakajima, S., Uchida, H., Suzuki, T., Watanabe., K., Hirano., J., Yagihashi, T., Takeuchi, H., Abe, T., Kashima, H., Mimura, M., 2011. Is switching antidepressants following early nonresponse more beneficial in acute-phase treatment of depression?:a randomized open-label trial. Progress in Neuro-Psychopharmacology and Biological Psychiatry 35, 1983–1989. NICE, 2010. Depression: the Treatment and Management of Depression in Adults. NICE Clinical Guideline 91. Nierenberg, A.A., McLean, N.E., Alpert, J.E., Worthington, J.J., Rosenbaum, J.F., Fava, M., 1995. Early nonresponse to fluoxetine as a predictor of poor 8-week outcome. American Journal of Psychiatry 152, 1500–1503. Nierenberg, A.A., Farabaugh, A.H., Alpert, J.E., Gordon, J., Worthington, J.J., Rosenbaum, J.F., Fava, M., 2000. Timing of onset of antidepressant response with fluoxetine treatment. American Journal of Psychiatry 157, 1423–1428. Romera, I., Perez, V., Menchon, J.M., Schacht, A., Papen, R., Neuhauser, D., Abbar, M., Svanborg, P., Gilaberte, I., 2012. Early switch strategy in patients with

I. Gilaberte et al. / Journal of Affective Disorders 146 (2013) 433–437

major depressive disorder: a double-blind randomized study. Journal of Clinical Psychopharmacology 32, 479–486. ˜ a, F., Rickels, K., Schweizer, E., Rynn, M., Mandos, L.A., Demartinis, N., Garcia-Espan 2001. The antidepressant effect of sertraline is not enhanced by dose titration: results from an outpatient clinical trial. International Clinical Psychopharmacology 16, 137–143. ¨ Szegedi, A., Muller, M.J., Anghelescu, I., Klawe, C., Kohnen, R., Benkert, O., 2003. Early improvement under mirtazapine and paroxetine predicts later stable response and remission with high sensitivity in patients with major depression. Journal of Clinical Psychiatry 64, 413–420. Szegedi, A., Jansen, W.T., van Willigenburg, A.P.P., van der Meulen, E., Stassen, H.H., Thase, M.E., 2009. Early improvement in the first 2 weeks as a predictor of

437

treatment outcome in patients with major depressive disorder: a metaanalysis including 6562 patients. Journal of Clinical Psychiatry 70, 344–353. Tadic´, A., Helmreich, I., Mergl, R., Hautzinger, M., Kohnen, R., Henkel, V., Hegerl, U., 2010. Early improvement is a predictor of treatment outcome in patients with mild major, minor or subsyndromal depression. Journal of Affective Disorders 120, 86–93. van Calker, D., Zobel, I., Dykierek, P., Deimel, C.M., Kech, S., Lieb, K., Berger, M., Schramm, E., 2009. Time course of response to antidepressants: predictive value of early improvement and effect of additional psychotherapy. Journal of Affective Disorders 114, 243–253.