The article you requested is

Prevalence and Treatment Outcome in Anxious Versus Nonanxious Depression: Results From the German Algorithm Project

J Clin Psychiatry 2010;71(8):1047-1054
10.4088/JCP.09m05650blu

Objective: The objective of this study was to explore the prevalence of anxious depression in an inpatient population, to describe its clinical and sociodemographic correlates, and to compare treatment outcomes between patients with anxious and nonanxious depression. Furthermore, the efficacy of algorithm-guided treatment versus treatment as usual in patients with anxious versus nonanxious depression was evaluated.

Method: Data were collected on 429 inpatients with the diagnosis of a depressive episode (according to ICD-10) and a score of 15 on the 21-item Hamilton Depression Rating Scale (HDRS-21). The German Algorithm Project, phase 3 (GAP3), was conducted between 2000 and 2005 in 10 psychiatric departments throughout Germany. A baseline HDRS-21 anxiety/somatization factor score of 7 was considered indicative of anxious depression. Remission was defined as an HDRS-21 score 9. To evaluate the efficacy of algorithm-guided treatment, patients were randomly assigned into 3 groups: 2 different treatment algorithms or treatment as usual.

Results: The prevalence of anxious depression was 49%. Patients with anxious depression were more likely than those with nonanxious depression to be older (mean ± SD = 45.3 ± 12.8 vs 42.9 ± 12.0 years, odds ratio [OR] = 1.02 [95% CI, 1.00–1.03], P = .046), retired (70% vs 30%, OR = 3.09 [95% CI, 1.70–5.62], P = .000), without school qualification (74% vs 26%, OR = 3.11 [95% CI, 1.09–8.83], P = .035), more severely depressed (mean ± SD HDRS-21 score = 20.1 ± 5.0 vs 18.5 ± 4.4, OR = 1.08 [95% CI, 1.03–1.12], P = .001), and more likely to have a longer duration of the current episode (mean ± SD = 20.9 ± 26.2 vs 13.7 ± 14.3 weeks, OR = 1.02 [95% CI, 1.01–1.03], P = .011). Patients with anxious depression were more likely to display a variety of melancholic features. In patients with anxious depression compared to those with nonanxious depression, remission was less likely to be achieved (48.6% vs 61.5%, OR = 0.63 [95% CI, 0.42–0.92], P = .018) and took longer to occur (mean ± SD = 44 ± 3.4 vs 30 ± 2.8 days, HR = 0.65 [95% CI, 0.50–0.85], P = .001). There was no significant interaction with the treatment mode with regard to remission (Wald = 0.20, P = .890).

Conclusions: Anxious depression is common in patients diagnosed with depression. The poorer treatment outcome in patients with anxious depression demonstrates the need to address the issue of specific treatment strategies for this subgroup. However, anxious depression has no moderating effect on the efficacy of algorithm-guided treatment.

Trial Registration: http://www.germanctr.de/ Identifier: DRKS00000161

J Clin Psychiatry 2010;71(8):1047–1054

Submitted: August 28, 2009; accepted March 15, 2010.

Online ahead of print: July 13, 2010 (doi:10.4088/JCP.09m05650blu).

Corresponding author: Katja Wiethoff, PhD, Charité—Universitätsmedizin Berlin, Department of Psychiatry and Psychotherapy, Campus Charité Mitte, Charitéplatz 1, 10117 Berlin, Germany (katja.wiethoff@charite.de).

Prevalence and Treatment Outcome in Anxious Versus Nonanxious Depression: Results From the German Algorithm Project

Objective: The objective of this study was to explore the prevalence of anxious depression in an inpatient population, to describe its clinical and sociodemographic correlates, and to compare treatment outcomes between patients with anxious and nonanxious depression. Furthermore, the efficacy of algorithm-guided treatment versus treatment as usual in patients with anxious versus nonanxious depression was evaluated.

Method: Data were collected on 429 inpatients with the diagnosis of a depressive episode (according to ICD-10) and a score of 15 on the 21-item Hamilton Depression Rating Scale (HDRS-21). The German Algorithm Project, phase 3 (GAP3), was conducted between 2000 and 2005 in 10 psychiatric departments throughout Germany. A baseline HDRS-21 anxiety/somatization factor score of 7 was considered indicative of anxious depression. Remission was defined as an HDRS-21 score 9. To evaluate the efficacy of algorithm-guided treatment, patients were randomly assigned into 3 groups: 2 different treatment algorithms or treatment as usual.

Results: The prevalence of anxious depression was 49%. Patients with anxious depression were more likely than those with nonanxious depression to be older (mean ± SD = 45.3 ± 12.8 vs 42.9 ± 12.0 years, odds ratio [OR] = 1.02 [95% CI, 1.00–1.03], P = .046), retired (70% vs 30%, OR = 3.09 [95% CI, 1.70–5.62], P = .000), without school qualification (74% vs 26%, OR = 3.11 [95% CI, 1.09–8.83], P = .035), more severely depressed (mean ± SD HDRS-21 score = 20.1 ± 5.0 vs 18.5 ± 4.4, OR = 1.08 [95% CI, 1.03–1.12], P = .001), and more likely to have a longer duration of the current episode (mean ± SD = 20.9 ± 26.2 vs 13.7 ± 14.3 weeks, OR = 1.02 [95% CI, 1.01–1.03], P = .011). Patients with anxious depression were more likely to display a variety of melancholic features. In patients with anxious depression compared to those with nonanxious depression, remission was less likely to be achieved (48.6% vs 61.5%, OR = 0.63 [95% CI, 0.42–0.92], P = .018) and took longer to occur (mean ± SD = 44 ± 3.4 vs 30 ± 2.8 days, HR = 0.65 [95% CI, 0.50–0.85], P = .001). There was no significant interaction with the treatment mode with regard to remission (Wald = 0.20, P = .890).

Conclusions: Anxious depression is common in patients diagnosed with depression. The poorer treatment outcome in patients with anxious depression demonstrates the need to address the issue of specific treatment strategies for this subgroup. However, anxious depression has no moderating effect on the efficacy of algorithm-guided treatment.

Trial Registration: http://www.germanctr.de/ Identifier: DRKS00000161

J Clin Psychiatry 2010;71(8):1047–1054

Submitted: August 28, 2009; accepted March 15, 2010.

Online ahead of print: July 13, 2010 (doi:10.4088/JCP.09m05650blu).

Corresponding author: Katja Wiethoff, PhD, Charité—Universitätsmedizin Berlin, Department of Psychiatry and Psychotherapy, Campus Charité Mitte, Charitéplatz 1, 10117 Berlin, Germany (katja.wiethoff@charite.de).

Mood and anxiety disorders represent the largest group of mental disorders in Europe and the United States.1–3 They are not only highly prevalent but frequently coexist. Studies on patients with mood disorders show comorbidity rates of 40%–50%.4–7 There are 2 different approaches to address the relationship between major depressive disorder (MDD) and anxiety: anxious depression, defined as MDD with high levels of anxiety symptoms, follows a dimensional approach, whereas, when defined by comorbidity of major depression with comorbid anxiety disorder, a syndromal approach is used.8 The dimensional approach seems more applicable to clinical practice since many patients with MDD have prominent anxiety symptoms that are not clearly identifiable as a distinct entity or do not fully meet criteria for a DSM-IV or ICD-10 diagnosis.6 Although neither classification system (DSM-IV, ICD-10) specifically addresses anxious depression as a subtype of MDD, there is evidence that this may be a valid diagnostic subtype.9 The presence of anxious depression is related to greater illness severity and functional impairment,8–10 greater chronicity,10,11 delayed response to treatment,12,13 and an increased risk of suicidality.8,9,14 Recently, Fava et al8,9 described sociodemographic features associated with anxious depression: namely, patients were more likely to be older, unemployed, less educated, female, and in a relationship. Patients with anxious depression were more likely to show melancholic features of depression. Moreover, a difference in response to antidepressant treatment was found in some15,16 but not in all14,17 short-term studies. Recently, Fava et al13 have shown in a sample of 2,876 outpatients (Sequenced Treatment Alternatives to Relieve Depression [STAR*D] study) that remission was significantly less likely and took longer to occur in patients with anxious depression. Side effect frequency, intensity, and burden were also greater in the anxious subgroup. Papakostas et al18 reported that the presence/severity of psychic and somatic anxiety symptoms of MDD at baseline predicted an increased likelihood of nonresponse to fluoxetine in MDD.

While all of the previously mentioned studies were carried out in outpatients, this is the first study to our knowledge to investigate the concept of anxious depression in a sample of inpatients. The study was part of the multiphase German Algorithm Project, phase 3 (GAP3),19,20 the third and final phase of the project. The GAP3 evaluated algorithm-guided treatment of inpatients with MDD compared to treatment as usual.21 The GAP3 was part of a naturalistic study within the German Research Network on Depression, which aimed at assessing all inpatients with the primary diagnoses of a depressive episode according to ICD-1022 from admission to discharge. The objective of the presented post hoc analysis of the GAP3 database was to determine the prevalence of anxious depression in the GAP3 sample, to define its clinical correlates, to compare treatment response to that of nonanxious patients, and to test the efficacy of algorithm-guided treatment in patients with anxious versus nonanxious depression.

METHOD

Study Overview

Study subjects were drawn from GAP3, a randomized controlled multicenter trial to compare 2 different treatment algorithms (standardized stepwise drug treatment regimen and computerized decision and expert system) with treatment as usual. Within standardized stepwise drug treatment regimen, 3 different “second-step strategies” (lithium augmentation, high dose monotherapy, change of antidepressant) were compared in patients nonresponsive to a 4-week antidepressant monotherapy during inpatient treatment. For standardized stepwise drug treatment regimen–treated patients, physicians could choose from 4 different antidepressants (sertraline, a selective serotonin reuptake inhibitor [SSRI]; venlafaxine, a serotonin-norepinephrine reuptake inhibitor; reboxetine, a norepinephrine reuptake inhibitor; amitriptyline, a tricyclic antidepressant). The computerized decision and expert system linked individual patient response data to a probability matrix. Depending on the patient’s probability of responding or not responding to current treatment, the computerized decision and expert system proposes either continuing or changing the present strategy, without providing explicit recommendations. Lorazepam and nonbenzodiazepine hypnotics (zopiclone and zolpidem) were permitted for all patients to manage agitation, anxiety, or sleeping problems.

The GAP3 was performed between 2000 and 2005 in 6 academic and 4 nonacademic hospitals throughout Germany.

Participants

Adult inpatients (aged 18–70 years) with the primary diagnosis of a depressive episode (mild, moderate, or severe with or without psychotic symptoms) according to ICD-10,22 single or recurrent, were eligible for the study. An additional inclusion criterion was a score of 15 or higher on the 21-item Hamilton Depression Rating Scale (HDRS-21).23 Exclusion criteria were depression caused by another medical condition, pregnancy/breastfeeding, preexisting long-term medication treatment that could not be discontinued, and medical conditions that presented a limitation for any of the possible treatments in the study. All patients admitted to either of the participating centers were systematically assessed for eligibility and randomly assigned into 1 of the 5 study groups. The study was fully approved by the local ethics committees for each participating site. Study participants gave their written informed consent to initial treatment as well as to subsequent treatment steps in case of nonresponse.

Clinical Measurements

Diagnostic evaluation was confirmed with the Structured Clinical Interview for DSM-IV Axis I Disorders (SCID-I),24 a fully structured diagnostic interview for the assessment of mental disorders. Comorbid personality disorders were assessed with the Structured Clinical Interview for DSM-IV Axis II Personality Disorders (SCID-II).25 Baseline and follow-up evaluations included the Beck Depression Inventory (BDI),26 the HDRS-21,23 and the Utvalg for Kliniske Undersogelser (UKU) Side Effect Rating Scale27 for the assessment of medication-induced side effects. For further analysis, only those symptoms were taken into account in which causal relationship with the medication was rated as likely. At baseline, clinical and sociodemographic characteristics were assessed using a systematic basic assessment scale of clinical and sociodemographic variables in psychiatry (Basic Documentation).28

Treatment outcome and tolerability were assessed every 2 weeks (± 3 days) by research staff who, while not blinded for the treatment group, were not involved in the patient treatment. Details of algorithm-guided decision making are published elsewhere.21 The primary outcome criteria were the remission and response rates and time to remission. The primary outcome variable was the score on the HDRS-21. Remission was defined as an HDRS-21 score 9. Participants remained inpatient until remission and were rated until discharge, unless they dropped out of the study and refused further ratings. Follow-up assessments were performed at 1 and 2 years after discharge.

Definition of Anxious Depression

Anxious depression, defined as MDD with high levels of anxiety symptoms, was determined based on the Hamilton anxiety/somatization factor score.7–9 The anxiety/somatization factor, derived from a factor analysis of the HDRS-17 scale conducted by Cleary and Guy,29 includes 6 items: anxiety (psychic), anxiety (somatic), somatic symptoms (gastrointestinal), somatic symptoms (general), hypochondriasis, and insight. In accordance with previous studies,8,9,13 a Hamilton anxiety/somatization factor score 7 based on the HDRS-21 assessment at baseline was used to define anxious depression.

Statistical Analysis

Statistical Package for Social Science (SPSS) Version 18.0 (SPSS Inc, Chicago, Illinois) was used for statistical operations. Descriptive statistics included frequencies and percentages for categorical variables, mean and standard deviations or median and range for continuous variables. To allow for adjustments for baseline severity of depression, bivariate logistic regression models were used to assess the association between the independent variables of interest and the presence of anxious depression. Baseline depression severity was measured by the HDRS-21 score, not including the items used to identify anxious depression. Survival analysis was conducted to be able to use all patient data including right-censored cases due to dropout. It was assumed that treatment response did not differ between patients remaining in the study until remission and dropout. Median survival times were calculated using Kaplan-Meier statistics. Differences in probability (hazard) of remission between groups were analyzed using log-rank test and Cox regression modeling. Anxious depression, treatment mode (standardized stepwise drug treatment regimen, computerized decision and expert system, treatment as usual), depression severity (HDRS-21 baseline without anxiety-specific items), age, education, employment, and duration of current episode were included as independent variables as they were significantly different in the comparison of anxious versus nonanxious depression. The direct entry method was chosen for the multivariate analysis.

No adjustments were made for multiple testing. Therefore, all significant associations must be viewed with caution. Sample sizes may differ between specific steps of analysis depending on available data.

RESULTS

Sample Description

Of 593 patients who entered the naturalistic study between 2000 and 2005, 475 (80.1%) took part in GAP3. Of these 475 patients, 46 had to be excluded post hoc after randomization (9 patients because of incomplete data, 11 patients because of baseline HDRS-21 scores < 15, 26 because of a diagnosis other than major depressive episode), so that 429 were included in the final analysis. Table 1 summarizes the clinical and sociodemographic characteristics of the sample.

Table 1

Click figure to enlarge

Figure 1

Click figure to enlarge

The majority of the patients were female (63%), suffered from recurrent depression (53%), and were rated as having moderate to severe depression, with a mean ± SD HDRS-21 baseline score of 25.9 ± 6.0. The prevalence of anxious depression in this population was 49%. There was no significant difference in distribution among the 3 treatment modes (anxious depression: standardized stepwise drug treatment regimen (48.3%), computerized decision and expert system (50.6%), treatment as usual (61.9%); χ2 = 4.48, P = .093). The distribution of the HDRS-21 anxiety/somatization factor scores appears to be continuous, with a range of 0 to 14 and a median score of 7 (Figure 1).

Sociodemographic and Clinical Features at Baseline

Tables 2 and 3 provide the results of the nonadjusted and adjusted (for baseline severity of depression) relationships between the presence or absence of anxious depression and both sociodemographic and clinical variables.

Anxious depression was significantly more likely among older subjects (45.3 ± 12.8 vs 42.9 ± 12.0 years, OR = 1.02 [95% CI, 1.00–1.03], P = .046), among (prematurely) retired patients (70% vs 30%, OR = 3.09 [95% CI, 1.70–5.62], P = .000), and among those without school qualifications (74% vs 26%, OR = 3.11 [95% CI, 1.09–8.83], P = .035). Patients with more severe depression, as measured by the BDI (32.1 ± 10.3 vs 26.2 ± 9.7, OR = 1.05 [95% CI, 1.03–1.08], P = .000) and the HDRS-21 (not including the anxiety specific items, 20.1 ± 5.0 vs 18.5 ± 4.4, OR = 1.08 [95% CI, 1.03–1.12], P = .001), and those with a longer duration of the current episode (20.9 ± 26.2 vs 13.7 ± 14.3 weeks, OR = 1.02 [95% CI, 1.01–1.03], P = .011) were more likely to have anxious depression (see Tables 2 and 3). The differences remained significant after adjusting for baseline severity of depression.

Table 2

Click figure to enlarge

Table 3

Click figure to enlarge

There was no significant difference between patients with anxious and nonanxious depression in gender, marital status, number of children, age at illness onset, number of depressive episodes, number of psychiatric comorbidities, suicidality, or presence of a comorbid personality disorder (see Tables 2 and 3).

Patients with anxious depression were more likely, after we adjusted for baseline depression severity, to agree with the BDI items concerning sadness, negative body image, retardation, fatigability, loss of appetite, and somatic preoccupation (Table 4).

Table 4

Click figure to enlarge

Response to Treatment

Patients with anxious depression had significantly lower response (59.5% vs 69.7%, OR = 0.63 [95% CI, 0.42–0.94], P = .023) and remission rates (48.6% vs 61.5%, OR = 0.63 [95% CI, 0.42–0.92], P = .018) during the study period. These results remained significant even after controlling for baseline severity of depression (Table 5).

Table 5

Click figure to enlarge

The presence or absence of anxious depression was not the only significant predictor of remission and response. The total score of the HDRS anxiety/somatization factor also showed a significant association to remission (OR = 0.91 [95% CI, 0.85–0.98], P = .016) and response at the study endpoint (OR = 0.92 [95% CI, 0.85–0.99], P = .032), again even after controlling for baseline severity of depression.

Kaplan-Meier survival analysis showed a significant difference in median ± SD time to remission between the 2 groups (30 ± 2.8 days for nonanxious depression vs 44 ± 3.4 days for anxious depression; log-rank test, χ2 = 16.99, P = .000). The subsequent Cox regression analysis revealed a significantly lower probability of achieving remission for anxious depression as compared to nonanxious depression (HR = 0.65 [95% CI, 0.50–0.85], Wald = 10.15, P = .001), even when adding baseline severity of depression (HR = 0.96 [95% CI, 0.93–0.99], Wald = 6.16, P = .013) or treatment mode (Wald = 8.97, P = .011) to the model (Figure 2). Compared to treatment as usual, patients from standardized stepwise drug treatment regimen had a higher probability of achieving remission (HR = 1.48 [95% CI, 1.05–2.07], Wald = 4.88, P = .027), but there was no difference between computerized decision and expert system and treatment as usual (HR = 0.95 [95% CI, 0.61–1.48], Wald = 0.05, P = .822). Even when factoring in age (HR = 1.02 [95% CI, 0.99–1.03], P = .084), education (HR = 0.59 [95% CI, 0.23–1.48], P = .259), employment (Wald = 7.00, P = .072), and duration of illness (HR = 0.99 [95% CI, 0.97–1.01], P = .224), we found that anxious depression remained a significant predictor (HR = 0.66 [95% CI, 0.46–0.93], P = .017). A further Cox regression analysis revealed no significant statistical interaction between treatment mode (standardized stepwise drug treatment regimen vs computerized decision and expert system vs treatment as usual) and presence of anxious symptoms (anxious vs nonanxious depression) (Wald = 0.20, P = .890).

Figure 2

Click figure to enlarge

Anxious patients in standardized stepwise drug treatment regimen (n = 154) displayed a significantly lower probability of achieving remission after an initial antidepressant monotherapy compared to patients with nonanxious depression, even when controlling for severity of depression (OR = 0.63 [95% CI, 0.42–0.92], P = .018). Anxious and nonanxious patients did not differ in their use of tranquilizers (anxious, 61.5%, vs nonanxious, 56.6%; χ2 = 0.87, P = .350). They also did not show differences in tranquilizer dosages used per day (F = 1.22, P = .270).

Compared to patients with nonanxious depression, mean ± SD length of treatment until discharge was longer for patients with anxious depression (59.16 ± 40.51 days vs 50.25 ± 35.54 days, t = −2.416, P = .016). At discharge, patients with anxious depression had a lower probability of being remitted than patients with nonanxious depression (73% vs 82%, χ2 = 3.85, P = .050), independent of study completion or dropout.

Treatment Tolerability

A significant difference between anxious and nonanxious patients was found neither in overall dropout frequency (43.2% vs 34.6%, χ2 = 3.30, P = .069) nor in side effect–related dropout frequency (7% vs 5.8%, χ2 = 3.71, P = .716).

There was also no significant difference between anxious and nonanxious patients in reporting versus never reporting side effects (51.1% vs 60.6% reporting side effects, χ2 = 3.36, P = .067). Of those patients with side effects (n = 205), 5.9% reported side effects up to 25% of the time; 31.7%, up to 50% of the time; 18%, up to 75% of the time; and 44.4%, up to 100% of the time—with no significant difference between anxious and nonanxious patients (χ2 = 3.77, P = .287).

DISCUSSION

We showed in this multisite study of inpatients diagnosed with MDD that anxious depression is common in a German inpatient sample, with a prevalence of 49%. This finding is consistent with the frequencies of anxious depression found in other studies (44%–46% in the STAR*D outpatient sample8,9) and with the lifetime comorbidity rate for depression and anxiety disorders of 40%–50%6,30 or for bipolar disorder and anxiety of 50%.31

As in previous studies, the incidence of anxious depression was related to distinct sociodemographic and clinical correlates, even after adjustment for severity of depression. Our study showed that patients with anxious depression were older, less educated, and more often (prematurely) retired, before and after controlling for depression severity at baseline. Although patients in the STAR*D study were more likely to be unemployed than retired, the results taken together confirm the view of anxiously depressed patients as more often carrying unfavorable social determinants and negative predictors of disease course.13 Patients with anxious depression were also more likely to report greater illness severity, as measured with the HDRS-21 and the BDI self-rating instrument, which also is in line with the results of the STAR*D study,8,9 although different instruments were used to measure illness severity ( 30-item Inventory of Depressive Symptomatology–Clinician-Rated and the 16-item Quick Inventory of Depressive Symptomatology–Self-Report). Consistent with the greater illness severity at baseline is the significantly longer duration of the current episode among patients with anxious depression in our sample, both factors being possible indicators of an association between anxious depression and chronicity of depression as earlier described by Fava et al8,9 and VanValkenburg et al.11 However, in contrast to their results, we did not find a significant difference in suicidality between anxious and nonanxious patients.

As in the STAR*D sample, our study observed differences in the self-assessment of patients with anxious depression, ie, a higher frequency of items related to sadness, retardation, fatigability, and loss of appetite, although different rating scales were used. This lends further support to the assumption by Fava et al13 that anxious depression may share some core features with the endogenous, melancholic subtype of depression.

In addition to the clinical and sociodemographic differences, the present study shows that patients with anxious depression had poorer treatment outcomes than patients with nonanxious depression, independent of the predictor used (total score of HDRS anxiety/somatization factor or presence/absence of anxious depression). In our sample, about 49% of anxious patients achieved remission in contrast to 61% of nonanxious patients. In addition, anxious patients also took longer to remit than nonanxious patients, even after controlling for depression severity. The 10% difference in remission rates between anxious and nonanxious patients and the longer time to remission for anxious patients highlights the similarity of Fava and colleagues’13 findings and the results of the present study.

Dunlop et al32 and Furukawa et al33 discuss that benzodiazepines appear to improve treatment outcomes for depression characterized by anxious features. Poorer treatment outcome of anxious depressed patients in our sample cannot be attributed to restrictions in the use of benzodiazepines as they were equal for patients with and without anxious features.

In contrast to the results of Fava et al,13 our study did not find any difference between anxious and nonanxious patients with regard to the reporting of side effects, side effect frequency, or side effects related dropouts. This difference in results could partly be due to the different treatment modalities of the 2 studies. The STAR*D sample is an outpatient sample, whereas the GAP sample is an inpatient sample. Patients with anxious depression may be particularly prone to overreporting side effects, particularly under the conditions of outpatient treatment. In inpatients, closer monitoring and reassurance by the treating staff may reduce the sensitivity to side effects. In addition, side effects in our study were clinician-rated, which further minimizes the risk of side effect overreporting in anxious patients. The inpatient setting may also explain the lack of difference in the use of tranquilizer between anxious and nonanxious patients in our sample. Continuous reassurance by the staff as well as close monitoring of medication and daily controlled prescription of tranquilizers may equalize their use in the 2 subgroups. However, prescription rates for tranquilizers were rather high in both groups (61.5% and 56.6%), which further minimizes the chance of showing a difference.

A noteworthy finding was that anxious depression had no moderating effect on the efficacy of algorithm-guided treatment found in the total population. The superiority of an algorithm-guided treatment according to standardized stepwise drug treatment regimen equally applies to the anxious and nonanxious subtype of MDD. Among remitted patients, those with nonanxious depression were more likely to respond to an initial antidepressant monotherapy than patients with anxious depression.

In summary, the current study shows that patients with anxious depression are a “harder-to-treat” subgroup and that further research is needed to address the issue of specific first-line treatments for these patients to avoid and overcome treatment resistance and chronic courses of illness. On the basis of a literature review, Silverstone et al34 conclude that drugs inhibiting the reuptake of both norepinephrine and serotonin may have greater clinical utility than single-acting drugs such as the SSRIs. Further, in a pooled analysis of 10 studies, Papakostas et al35 found a modest advantage for the SSRIs compared to bupropion in the treatment of anxious depression, a finding that stresses the effectiveness of a serotonergic mechanism in these patients. So far, there are no clear recommendations for the treatment of anxious depression. Whether antidepressants that have shown their effectiveness in panic disorder or generalized anxiety disorder36 are also more effective in patients with anxious depression requires further study.

Our results highlight the need for a stronger emphasis on the identification of symptoms of anxious depression in patients with MDD. The question as to whether anxious depression should be regarded as a distinct subtype of MDD, as proposed by Fava et al,13 cannot be conclusively answered with our data. Robins and Guze37 and Kendler38 describe differentiating features, evidence of familiality, and specific treatment responsivity as indicators of a distinct psychiatric diagnostic entity. In our study, anxious depression appears to be associated with specific features and different response patterns. As we did not obtain any data on anxious depression in relatives, we cannot comment on familiality. Recently, the functional −399C/T polymorphism of the neuropeptide Y gene (rs16147) has been shown to be associated with antidepressant response in the clinical phenotype of anxious depression,39 which supports a genetic difference between anxious and nonanxious depression. The authors propose that the current state of genetic and imaging genetics research on the clinical phenotype of anxious depression might inspire a redefinition and restructuring of the present nosologic concepts in DSM-V. Clearly, further studies are needed to provide information on the specific treatment responsivity and on familiality to support the utility of anxious depression as a distinct subtype of MDD. The question remains if anxious depression should be regarded as a distinct subtype or rather as a clinical phenotype of major depression that, as with severe depression, merits a higher intensity of care.

More importantly from our point of view is the need for stronger diagnostic and therapeutic attention toward patients with anxious depression. The dimensional approach and the use of a threshold value instead of a continuous dimension could be of particular clinical utility. We used the definition of anxious depression based on the HDRS anxiety/somatization factor score 7, as had been done in previous studies8,9,13,14 to make results comparable. Undoubtedly, this cutoff point should be examined for sensitivity and specificity in future studies.

Generalizability of our results is limited to an inpatient population that is generally more severely depressed than an outpatient sample, although we have controlled for depression severity. Further, a limitation in comparability to the STAR*D sample is the lower chronic disease burden of the GAP3 sample. Our data have been collected within the German health care system, which is characterized by a higher rate of inpatient admissions that occur rather early in the course of the disease. Another limitation of our results is the inclusion of patients with psychotic symptoms, which, although limited in number (n = 28), might have confounded the analysis. Despite the different treatment modalities, Fava et al8,9,13 and our study come up with very similar results with regard to sociodemographic and clinical characteristics as well as response patterns of patients with anxious depression.

In conclusion, our results show that special attention should be paid to the assessment and diagnosis of anxious depression in clinical practice. There is growing evidence that anxious depression might be a valid clinical predictor for treatment response.

Drug names: bupropion (Aplenzin, Wellbutrin, and others), fluoxetine (Prozac and others), lithium (Lithobid and others), lorazepam (Ativan and others), sertraline (Zoloft and others), venlafaxine (Effexor and others), zolpidem (Zolpimist, Ambien, and others), zopiclone (Lunesta).

Author affiliation: Department of Psychiatry and Psychotherapy, Charité—Universitätsmedizin Berlin, Campus Charité Mitte (Drs Wiethoff, Hollinde, and Adli and Ms Kiermeir); Department of Psychiatry and Psychotherapy, St.Joseph-Krankenhaus (Dr Hauth); Department of Psychiatry and Psychotherapy, Auguste-Viktoria-Krankenhaus (Dr Zeiler), Berlin; Department of Psychiatry and Psychotherapy, Universitätsklinikum Carl Gustav Carus, Dresden (Dr Bauer); Department of Psychiatry and Psychotherapy, Ludwig-Maximilians-University of Munich, München (Drs Baghai and Möller); Bezirksklinikum Gabersee, Wasserburg am Inn (Dr Laux); Department of Psychiatry and Psychotherapy, Heinrich-Heine-Universität, Düsseldorf (Dr Cordes); Department of Psychiatry, Psychotherapy and Psychosomatic Medicine, Bezirkskrankenhaus Kempten, Kempten (Dr Brieger); Department of Psychiatry, Zentrum für Psychosoziale Medizin, Universitätsklinikum Heidelberg, Heidelberg (Dr Kronmüller), Germany; and East London NHS Foundation Trust, City and Hackney Centre for Mental Health, Donald Winnicott Centre, London, United Kingdom (Dr Fisher).

Potential conflicts of interest: Dr Bauer has been a consultant to Eli Lilly, Servier, GlaxoSmithKline, Bristol-Myers Squibb, AstraZeneca, Janssen-Cilag, Lundbeck, and Otsuka; and has received honoraria from Eli Lilly, AstraZeneca, Bristol-Myers Squibb, Servier, GlaxoSmithKline, Lundbeck, and Pfizer. Dr Baghai accepted paid speaking engagements and acted as a consultant for AstraZeneca, GlaxoSmithKline, Janssen-Cilag, Organon, Pfizer, and Servier. Dr Möller has received grants from or is a consultant to and has served on speakership bureaus of AstraZeneca, Bristol-Myers Squibb, Eisai, Eli Lilly, GlaxoSmithKline, Janssen-Cilag, Lundbeck, Merck, Novartis, Organon, Pfizer, Sanofi-Aventis, Schwabe, Sepracor, Servier, and Wyeth. Dr Laux has received grants or is a consultant to and on the speakership bureaus of AstraZeneca, Bayer, Boehringer Ingelheim, Janssen-Cilag, Eli Lilly, Lundbeck, Merz, Novartis, Organon, Pfizer, Servier, Steigerwald, Teva, Wyeth, and the German Federal Ministry of Education and Research. Dr Cordes has participated in speakership bureaus for Janssen-Cilag, Eli Lilly, Servier Deutschland GmbH, Alpine-Biomed, Eli Lilly; and has received grant/research support from Eli Lilly, Janssen-Cilag, Lundbeck GmbH, Servier Deutschland GmbH, Pfizer, and the German Federal Ministry for Education and Research. Dr Brieger has been a consultant to and has served on the speakership bureaus of Pfizer, AstraZeneca, Lundbeck, and Bristol-Myers Squibb; and has received grant support from the German Federal Ministry of Education and Research. Dr Kronmüller has received grants or is a consultant to and on the speakership bureaus of AstraZeneca, Janssen-Cilag, Lilly, Lundbeck, Pfizer, Servier, Wyeth, and the German Federal Ministry of Education and Research. Dr. Adli has received grant/research support from the German Federal Ministry of Education and Research, Pharmacia, Pfizer, Eli Lilly, Janssen-Cilag, Wyeth, esparma, and Lundbeck; has received speaker honoraria from AstraZeneca, Eli Lilly, GlaxoSmithKline, Pfizer, Boehringer Ingelheim, Sanofi-Aventis, esparma, and Wyeth; and has been a consultant to Bristol-Myers Squibb, esparma, and Pfizer. Miss Kiermeir has received grants or is a consultant to and on the speakership bureaus of Eli Lilly, Boehringer Ingelheim, and Janssen-Cilag. Drs Wiethoff, Fisher, Hollinde, Hauth, and Zeiler have no conflict of interest to declare.

Funding/support: The German Algorithm Project, phase 3, was supported by the German Federal Ministry for Education and Research through the program “German Research Network on Depression.” Additional unrestricted research grants were received from Eli Lilly, Janssen-Cilag, Pfizer, Pharmacia, and Wyeth-Ayerst.

Acknowledgment: The study was conducted in 10 psychiatric hospitals throughout Germany (6 academic and 4 nonacademic centers): Charité—Universitätsmedizin Berlin, Campus Charité Mitte (leading study site), Campus Benjamin Franklin, Ludwig-Maximilians-University Munich, University of Halle, University of Dusseldorf, University of Heidelberg, St. Hedwig Hospital Berlin, St. Joseph Hospital Berlin, Vivantes Auguste-Viktoria Hospital Berlin, and District Hospital Gabersee.

REFERENCES

1. Wittchen HU, Jacobi F. Size and burden of mental disorders in Europe—a critical review and appraisal of 27 studies. Eur Neuropsychopharmacol. 2005;15(4):357–376. PubMed doi:10.1016/j.euroneuro.2005.04.012

2. Kessler RC, Chiu WT, Demler O, et al. Prevalence, severity, and comorbidity of 12-month DSM-IV disorders in the National Comorbidity Survey Replication. Arch Gen Psychiatry. 2005;62(6):617–627. PubMed doi:10.1001/archpsyc.62.6.617

3. Kessler RC, Gruber M, Hettema JM, et al. Co-morbid major depression and generalized anxiety disorders in the National Comorbidity Survey follow-up. Psychol Med. 2008;38(3):365–374. PubMed doi:10.1017/S0033291707002012

4. Melartin TK, Rytsälä HJ, Leskelä US, et al. Current comorbidity of psychiatric disorders among DSM-IV major depressive disorder patients in psychiatric care in the Vantaa Depression Study. J Clin Psychiatry. 2002;63(2):126–134. PubMed

5. Regier DA, Rae DS, Narrow WE, et al. Prevalence of anxiety disorders and their comorbidity with mood and addictive disorders. Br J Psychiatry Supp. 1998;34:24–28. PubMed

6. Sanderson WC, Beck AT, Beck J. Syndrome comorbidity in patients with major depression or dysthymia: prevalence and temporal relationships. Am J Psychiatry. 1990;147(8):1025–1028. PubMed

7. Zimmerman M, Chelminski I, McDermut W. Major depressive disorder and axis I diagnostic comorbidity. J Clin Psychiatry. 2002;63(3):187–193. PubMed

8. Fava M, Alpert JE, Carmin CN, et al. Clinical correlates and symptom patterns of anxious depression among patients with major depressive disorder in STAR*D. Psychol Med. 2004;34(7):1299–1308. PubMed doi:10.1017/S0033291704002612

9. Fava M, Rush AJ, Alpert JE, et al. What clinical and symptom features and comorbid disorders characterize outpatients with anxious major depressive disorder: a replication and extension. Can J Psychiatry. 2006;51(13):823–835. PubMed

10. Joffe RT, Bagby RM, Levitt A. Anxious and nonanxious depression. Am J Psychiatry. 1993;150(8):1257–1258. PubMed

11. Van Valkenburg C, Akiskal HS, Puzantian V, et al. Anxious depressions: clinical, family history, and naturalistic outcome—comparisons with panic and major depressive disorders. J Affect Disord. 1984;6(1):67–82. PubMed doi:10.1016/0165-0327(84)90009-0

12. Clayton PJ, Grove WM, Coryell W, et al. Follow-up and family study of anxious depression. Am J Psychiatry. 1991;148(11):1512–1517. PubMed

13. Fava M, Rush AJ, Alpert JE, et al. Difference in treatment outcome in outpatients with anxious versus nonanxious depression: a STAR*D report. Am J Psychiatry. 2008;165(3):342–351. PubMed doi:10.1176/appi.ajp.2007.06111868

14. Tollefson GD, Holman SL, Sayler ME, et al. Fluoxetine, placebo, and tricyclic antidepressants in major depression with and without anxious features. J Clin Psychiatry. 1994;55(2):50–59. PubMed

15. Flint AJ, Rifat SL. Anxious depression in elderly patients: response to antidepressant treatment. Am J Geriatr Psychiatry. 1997;5(2):107–115. PubMed doi:10.1097/00019442-199700520-00003

16. Davidson JR, Meoni P, Haudiquet V, et al. Achieving remission with venlafaxine and fluoxetine in major depression: its relationship to anxiety symptoms. Depress Anxiety. 2002;16(1):4–13. PubMed doi:10.1002/da.10045

17. Russell JM, Koran LM, Rush J, et al. Effect of concurrent anxiety on response to sertraline and imipramine in patients with chronic depression. Depress Anxiety. 2001;13(1):18–27. PubMed doi:10.1002/1520-6394(2001)13:1<18::AID-DA3>3.0.CO;2-M

18. 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(1):116–120. PubMed doi:10.1016/j.psychres.2008.02.011

19. Adli M, Berghöfer A, Linden M, et al. Effectiveness and feasibility of a standardized stepwise drug treatment regimen algorithm for inpatients with depressive disorders: results of a 2-year observational algorithm study. J Clin Psychiatry. 2002;63(9):782–790. PubMed

20. Bauer M, Pfennig A, Linden M, et al. Efficacy of an algorithm-guided treatment compared with treatment as usual: a randomized, controlled study of inpatients with depression. J Clin Psychopharmacol. 2009;29(4):327–333. PubMed doi:10.1097/JCP.0b013e3181ac4839

21. Adli M, Bauer M, Rush AJ. Algorithms and collaborative-care systems for depression: are they effective and why? a systematic review. Biol Psychiatry. 2006;59(11):1029–1038. PubMed doi:10.1016/j.biopsych.2006.05.010

22. World Health Organization. The ICD-10 Classification of Mental and Behavioral Disorders. Clinical Descriptions and Diagnostic Guidelines. Geneva, Switzerland: World Health Organization; 1992.

23. Hamilton M. A rating scale for depression. J Neurol Neurosurg Psychiatry. 1960;23(1):56–62. PubMed doi:10.1136/jnnp.23.1.56

24. Wittchen HU, Wunderlich U, Gruschwitz MS, et al. SKID-I, Strukturiertes Klinisches Interview für DSM-IV. Göttingen: Hogrefe; 1997a.

25. Wittchen HU, Zaudig M, Fydrich T. SKID: Strukturiertes Klinisches Interview für DSM-IV, Achse II. Göttingen: Hogrefe; 1997b.

26. Beck AT, Ward CH, Mendelson M, et al. An inventory for measuring depression. Arch Gen Psychiatry. 1961;4:561–571. PubMed

27. Lingjaerde O, Ahlfors UG, Bech P, et al. The UKU Side Effect Rating Scale. A New Comprehensive Rating Scale for Psychotropic Drugs and a Cross-Sectional Study of Side Effects in Neuroleptic-Treated Patients. Acta Psychiatr Scand suppl. 1987;76(s334):1–100. doi:10.1111/j.1600-0447.1987.tb10566.x

28. Cording C, Gaebel W, Spengler A. Die neue psychiatrische basisdokumentation. Eine empfehlung der DGPPN zur qualitätssicherung im (teil-)stationären bereich (The new psychiatric basic documentation. A recommendation by the DGPPN for quality assurance in inpatient treatment). Spektrum Psychiatrie Nervenheilkunde; 1995;24:3–41.

29. Cleary P, Guy W. Factor analysis of the Hamilton Depression Scale. Drugs Exp Clin Res. 1977;1:115–120.

30. Fava M, Rankin MA, Wright EC, et al. Anxiety disorders in major depression. Compr Psychiatry. 2000;41(2):97–102. PubMed doi:10.1016/S0010-440X(00)90140-8

31. Simon NM, Otto MW, Wisniewski SR, et al. Anxiety disorder comorbidity in bipolar disorder patients: data from the first 500 participants in the Systematic Treatment Enhancement Program for Bipolar Disorder (STEP-BD). Am J Psychiatry. 2004;161(12):2222–2229. PubMed doi:10.1176/appi.ajp.161.12.2222

32. Dunlop BW, Davis PG. Combination treatment with benzodiazepines and SSRIs for comorbid anxiety and depression: a review. Prim Care Companion J Clin Psychiatry. 2008;10(3):222–228. PubMed doi:10.4088/PCC.v10n0307

33. Furukawa TA, Streiner DL, Young LT. Antidepressant and benzodiazepine for major depression. Cochrane Database Syst Rev. 2002;1(1):CD001026. PubMed

34. Silverstone PH, von Studnitz E. Defining anxious depression: going beyond comorbidity. Can J Psychiatry. 2003;48(10):675–680. PubMed

35. Papakostas GI, Stahl SM, Krishen A, et al. Efficacy of bupropion and the selective serotonin reuptake inhibitors in the treatment of major depressive disorder with high levels of anxiety (anxious depression): a pooled analysis of 10 studies. J Clin Psychiatry. 2008;69(8):1287–1292. PubMed doi:10.4088/JCP.v69n0812

36. Bandelow B, Zohar J, Hollander E, et al; WFSBP Task Force on Treatment Guidelines for Anxiety, Obsessive-Compulsive and Post-Traumatic Stress Disorders. World Federation of Societies of Biological Psychiatry (WFSBP) guidelines for the pharmacological treatment of anxiety, obsessive-compulsive and post-traumatic stress disorders—first revision. World J Biol Psychiatry. 2008;9(4):248–312. PubMed doi:10.1080/15622970802465807

37. Robins E, Guze SB. Establishment of diagnostic validity in psychiatric illness: its application to schizophrenia. Am J Psychiatry. 1970;126(7):983–987. PubMed

38. Kendler KS. The nosologic validity of paranoia (simple delusional disorder): a review. Arch Gen Psychiatry. 1980;37(6):699–706. PubMed

39. Domschke K, Dannlowski U, Hohoff C, et al. Neuropeptide Y (NPY) gene: impact on emotional processing and treatment response in anxious depression. Eur Neuropsychopharmacol. 2010;20(5):301–309. PubMed doi:10.1016/j.euroneuro.2009.09.006