- Email: [email protected]
Do comorbid anxiety disorders impact treatment planning for outpatients with major depressive disorder?
Psychiatry Research 169 (2009) 7–11
Contents lists available at ScienceDirect
Psychiatry Research j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / p s yc h r e s
Do comorbid anxiety disorders impact treatment planning for outpatients with major depressive disorder? Timothy Petersen a,⁎, Charissa F. Andreotti a, Iwona Chelminski b, Diane Young b, Mark Zimmerman b a b
Massachusetts General Hospital, Department of Psychiatry, Division of Postgraduate Education, 1 Bowdoin Square, 7th floor, Boston, MA 02114, United States Rhode Island Hospital Department of Psychiatry and Human Behavior, Brown University School of Medicine, Rhode Island Hospital, Providence, RI 02905, United States
a r t i c l e
i n f o
Article history: Received 1 October 2007 Received in revised form 4 August 2008 Accepted 23 October 2008 Keywords: Major depressive disorder Treatment planning Comorbid anxiety disorders
a b s t r a c t Research indicates that depressed patients with comorbid anxiety disorders have a poorer long-term course of illness, are less responsive to treatment, and may experience greater deficits in psychosocial functioning, when compared with depressed patients without comorbid anxiety disorders. The objective of this study was to examine, through use of a large, well-characterized clinical database, how clinicians may modify treatment recommendations in depressed outpatients when anxiety disorders are present. A group of 346 case records, derived from the Methods to Improve Diagnostic Assessment and Services (MIDAS) project at Rhode Island Hospital, were examined to determine what treatment recommendations were made immediately after diagnosis. Psychopharmacological and psychotherapeutic treatments were classified to capture differences in recommendations between groups. Demographic and clinical characteristics were compared for patients with (n = 248) and without (n = 98) comorbid anxiety disorders. Utilizing logistic regression models, we found patients with anxiety disorders had a greater number of psychopharmacological therapies included as part of their initial treatment plan, but no differences were found in initial psychotherapeutic interventions. Our results indicate that practitioners are making unique recommendations based on comorbid anxiety diagnoses, but outcome studies are now needed to determine the most effective treatment methods for this patient population. © 2008 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Anxiety disorders are the most common comorbid psychiatric condition in patients with major depressive disorder (Zimmerman et al., 2000; Rush et al., 2004). Analyses of epidemiological databases indicate that 50% of all depressed patients meet criteria for at least one anxiety disorder, and a large proportion of these cases meet criteria for multiple anxiety disorders (Gorman, 1996–1997; Zimmerman et al., 2000). Recent findings from the Sequenced Treatment Alternatives to Relieve Depression (STAR⁎D) investigation confirm the high prevalence of comorbid anxiety symptoms in depressed outpatients. In this effectiveness study of over 4000 depressed outpatients, treated within both primary and specialty care settings, 46% of patients were found to meet criteria for anxious depression, defined as Major Depressive Disorder (MDD) with high levels of anxiety symptoms (Fava et al., 2004). The presence of anxious depression was associated with greater severity of depressive symptoms and a higher rate of suicidal ideation (Fava et al., 2004; Rihmer, 2007). Other research suggests that the presence of comorbid anxiety disorders is associated with poorer antidepressant treatment outcomes, higher levels of health care utilization, a worse long-term course of depression, occurrence of cluster A and C personality disorders (Farabaugh et al., ⁎ Corresponding author. E-mail address: [email protected] (T. Petersen). 0165-1781/$ – see front matter © 2008 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.psychres.2008.10.036
2005), and lower psychosocial functioning (Brown et al., 1996). Despite these findings, comorbid anxiety disorders are often underrecognized in community-based care settings (Zimmerman and Chelminski, 2003). A number of studies have examined psychopharmacologic treatment approaches for depressed patients with anxiety symptoms, which include the use of antidepressants, anxiolytics, and other novel compounds (Kuzel, 1996; Nutt, 1997). Psychotherapeutic approaches have also been tested and include the use of traditional cognitivebehaviorally based approaches (Beck, 1979) as well as interpersonally oriented psychotherapies (Feske et al., 1998). While some outcome studies suggest that standard evidence-based treatments are effective in reducing both depressive and anxiety symptoms (Joffe et al., 1993; Tollefson et al., 1994), other studies recommend modifications to traditional treatment approaches in order to optimally manage depression with comorbid anxiety (Kuzel, 1996; Gorman, 1996– 1997). For example, Bakish (1999) supports the use of the novel serotonin–norepinephrine reuptake inhibitor, venlafaxine, for depression with concomitant anxiety, and work by Kuzel (1996) finds nefazodone to also aid in treating depression with comorbid anxiety features. To date, however, no psychopharmacologic or psychotherapeutic treatment has been established in randomized controlled trials as superior in the treatment of depression with comorbid anxiety. In order to better understand what treatments are most effective in managing depressed patients with comorbid anxiety disorders, it is
8
T. Petersen et al. / Psychiatry Research 169 (2009) 7–11
critical to first understand how community-based practitioners are currently approaching treatment of this patient population. The objective of this study was to examine, in a large sample of outpatients with major depressive disorder, how the presence of comorbid anxiety disorders may impact treatment choices. Based on previous research suggesting that standard treatments may need to be modified to enhance outcomes for depressed patients with comorbid anxiety disorders, we hypothesized that treatment selections would significantly differ between depressed outpatients with and without comorbid anxiety disorders; specifically that patients with comorbid anxiety disorders would have a greater total number of treatments (psychopharmacologic and/or psychotherapeutic) initiated. 2. Method
utilized in this practice consists primarily of a semi-structured interview (Structured Clinical Interview for DSM-IV (SCID), which is used to determine a comprehensive diagnostic profile for each individual. This profile consists of formal, multiaxial DSM-IV diagnoses as well as several other clinical characteristics, which include number of lifetime depressive episodes, length of the current depressive episode, number of psychiatric hospitalizations, a rating of the number of work weeks missed due to depressive illness, and a rating of current social functioning. For the purposes of data entry, a coding scheme was created to document current and recommended treatments. Psychopharmacologic treatments were grouped into categories based on drug classification including selective serotonin reuptake inhibitor (SSRI), tricyclic, monoamine oxidase inhibitor (MAOI), and serotonin–norepinephrine reuptake inhibitor (SNRI) antidepressants as well as benzodiazepines, other anxiolytics, hypnotics, mood stabilizers, anticonvulsants, and stimulants. Treatment strategies were grouped into five categories (Add, Discontinue, Increase Dose, Decrease Dose, or Continue) in order to represent recommendations made at intake. This coding scheme allowed accurate capture of all possible treatment combinations, including existing treatment regimens as well as changes made after the initial evaluation. All data entry and analyses were conducted using STATA/SE version 9.2 (STATA/SE 9.2, 2007).
2.1. Overview 2.4. Data analysis Data for this investigation were derived from the Methods to Improve Diagnostic Assessment and Services (MIDAS) project, which represents an integration of research methodology into a community-based outpatient practice affiliated with an academic medical center (Zimmerman et al., 2002). To date, 2600 psychiatric outpatients have been evaluated with a semistructured diagnostic interview in this Rhode Island Hospital Department of Psychiatry outpatient practice. This private practice group predominantly treats individuals with medical insurance (including Medicare but not Medicaid) on a fee-for-service basis, and it is distinct from the hospital's outpatient residency training clinic that predominantly serves lower income, uninsured, and medical assistance patients. The primary referral sources to the practice are primary care physicians and psychotherapists. Not all patients who present for treatment at this practice participate in the research protocols. Patients are offered the opportunity to have a more comprehensive evaluation as part of the research program, though they are not required to undergo this evaluation. Importantly, as reported previously, patients in this practice who have participated in the comprehensive research evaluation have been shown to be similar to non-participating patients (Zimmerman and Mattia, 1999). Patients were interviewed by a diagnostic rater who administered the Structured Clinical Interview for DSM-IV (SCID). To study the psychometric performance of the DSM-IV symptom criteria for major depression, it was necessary to modify the SCID and eliminate the skip-out that curtails the depression module for patients who did not report either depressed mood or anhedonia. Thus, we inquired about all of the symptoms of depression for all patients. For compound criteria that encompass more than one symptom (e.g., indecisiveness or impaired concentration; increased sleep or insomnia), we made separate ratings of each component of the diagnostic criterion. Thus, the nine DSM-IV symptom criteria were broken down into 17 separate items. The diagnostic raters are highly trained and have been monitored throughout the project to minimize rater drift. Diagnostic raters included PhD-level psychologists and research assistants with college degrees in the social or biological sciences. Research assistants received 3 to 4 months of training during which they observed at least 20 interviews, and they were observed and supervised in their administration of more than 20 evaluations. Psychologists only observed five interviews; however, they, too, were observed and supervised in their administration of 15 to 20 evaluations. During the course of training, the senior author met with each rater to review the interpretation of every item on the SCID. Also during training every interview was reviewed on an item-byitem basis by the senior rater who observed the evaluation. At the end of the training period, the raters were required to demonstrate exact, or near exact, agreement with a senior diagnostician on five consecutive evaluations. Throughout the MIDAS project, ongoing supervision of the raters consisted of weekly diagnostic case conferences involving all members of the team. In addition, the item ratings of every case were reviewed by the senior author. The Rhode Island Hospital institutional review committee approved the research protocol, and all patients provided informed, written consent. 2.2. Selection and description of participants Treatment planning data were collected from evaluation reports of depressed outpatients assessed through the MIDAS program. Patients who were 18 years old or older and met criteria for a current, principal diagnosis of major depressive disorder were eligible for inclusion in this data analysis. Patients were selected only if the principal diagnosis listed on their evaluation report was major depressive disorder. This did not include patients with principal diagnoses of other disorders, including bipolar or anxiety disorder, even if major depressive disorder was listed as a secondary diagnosis. The first 2000 evaluation reports collected as part of the MIDAS program were screened, and a total of 346 (17.3%) participants (215 females and 131 males), were included in this analysis.
Our data analytic approach consisted of first calculating descriptive statistics for the entire sample, and then by patient group (with vs. without comorbid anxiety disorders). A patient was characterized as belonging to the “with a comorbid anxiety disorder” group if he/she met DSM-IV criteria for one or more comorbid anxiety disorders. The second step of our data analytic approach was to utilize appropriate parametric (unpaired t-tests) and non-parametric techniques (chi-square analyses) to evaluate differences between groups in current treatments, treatment recommendations, illness characteristics and demographic features. Post hoc tests were used, as needed, to determine the specific nature of these differences. Logistic regression models were developed to ascertain which variables and features best account for group membership (with or without comorbid anxiety disorders). Independent variables representing treatment approaches as well as demographic and clinical characteristics on which the groups differed were included in the logistic regression analyses. Adjustments in threshold for significance were made for multiple comparisons.
3. Results 3.1. Sample description A description of demographic and clinical characteristics of the sample is shown in Table 1. Mean Severity of illness, as measured by the Clinical Global Impression of Severity Scale (CGI-S) was 2.9 ± 0.1, suggesting a moderate overall illness severity. Mean Global Assessment of Functioning was 52.4 ± 0.4, indicating a moderate to serious level of symptomatology. Although there was no significant difference in the percentages of males and females in the sample, women were more likely to receive a secondary diagnosis for one or more comorbid anxiety disorders, (n = 346, χ2 = 7.18, df = 1, P = 0.007). 3.2. Group comparison Comparison of patient groups on demographic and clinical characteristics is also shown in Table 1. Patients with comorbid anxiety disorders had a significantly higher Clinical Global Impression S Scale, lower Global Assessment of Functioning,), and poorer current social functioning. The comorbid anxiety disorder group also had a higher number of psychiatric hospitalizations over the last 5 years, missed a greater amount of work due to depressive symptoms over the last 5 years, had a greater number of lifetime depressive episodes, and a longer duration of current depressive episode. Substance use disorders were significantly more common in the patient group with comorbid anxiety disorders (52.8% group with, 34.7% group without, n = 346, χ2 = 9.25, df = 1, P = 0.002). No significant differences were found in the presence of bipolar spectrum disorders, psychotic disorders, dysthymia, and axis II disorders between the two groups. 3.3. Frequency of comorbid anxiety disorders
2.3. Study procedures The following information was gathered from the evaluation reports: current treatment regimen at intake (including medication and non-medication approaches) and the treatments recommended at the conclusion of the evaluation. The evaluation
In the sub-group of patients with comorbid anxiety disorders (n = 248), the frequency of specific diagnoses is shown in Table 2. As shown in Table 2, social phobia, generalized anxiety disorder, and
T. Petersen et al. / Psychiatry Research 169 (2009) 7–11
9
Table 1 Descriptive statistics and group comparisons for demographic and clinical characteristics. Descriptive statistic
Gender (female) Age (years) Initial medications Lifetime depressive episodes Cognitive global impression-severity Global assessment of functioning Current social functioning Psychiatric hospitalization rate in past 5 years Number of work weeks missed Length of current depressive episode
Anxiety disorder Entire sample (n = 346)
Absent (n = 98)
Present (n = 248)
Test statistic
P
62.1% 40.4 ± 0.7 2.6 ± 0.1 5.5 ± 0.8 2.9 ± 0.1 52.4 ± 0.4 3.2 ± 0.1 0.4 ± 0.1 2.2 ± 0.1 207.7 ± 21.8
51.0% 40.8 2.4 2.9 2.7 55.3 3.4 0.1 1.7 126
66.5% 40.3 2.7 6.5 2.9 51.3 2.8 0.5 2.4 293
df = 1, χ2 = 7.18 df = 344, t = 0.3 df = 344, t = − .8 df = 344, t = − 2.1 df = 344, t = − 2.5 df = 344, t = 4.6 df = 344, t = − 4.4 df = 344, t = − 3.6 df = 344, t = − 3.4 df = 344, t = − 2.4
0.007 0.74 0.4 0.04 0.014 b 0.001 b 0.001 b 0.001 b 0.001 b 0.001
post-traumatic stress disorder were most prevalent as comorbid anxiety disorders. The total number of comorbid anxiety disorders for participants in this sub-group ranged from one to five, with 38% of patients (n = 133) having one diagnosed disorder, 19% (n = 66) having two disorders, 11% (n = 38) having three disorders, 2.9% (n = 10) having four disorders, and 0.29% (n = 1) having five disorders.
3.4. Evaluation of differences between groups in current and recommended treatments The distribution of pharmacotherapies, by patient group, taken before evaluation (current treatments) is shown in Table 3. No statistically significant differences were found between the two groups in the proportion of any one class of pharmacotherapy taken before evaluation (for all comparisons, P N 0.05). Similarly, no statistically significant difference was found between the two groups in the proportion of patients currently engaged in psychotherapy prior to intake, (36.7% group with, 33.2% group without, n = 346, χ2 = 0.39, df = 1, P = 0.53). The distribution of pharmacotherapies, by patient group, added after initial evaluation is shown in Table 4. No significant differences were found to exist between the two groups in the proportion of any one class of pharmacotherapy added. Logistic regression models were utilized with independent variables representing treatment approaches as well as clinical characteristics on which the groups differed, including gender, CGI-S, GAF, current social functioning, number of psychiatric hospitalizations, hospitalization rate over a 5year period, missed weeks of work due to depressive illness, number of lifetime depressive episodes, duration of current depressive episodes, and frequency of substance use disorders. The dependent variable was group membership (with vs. without anxiety disorders). Patients with comorbid anxiety disorders had a greater total number of psychopharmacological therapies added when compared with patients without comorbid anxiety disorders, (F = 2.50, df = 5, 340, P = 0.031). No differences in treatment recommendations to discontinue, (F = 0.28, df = 9, 336, P = 0.98); increase dosage, (F = 0.75, df = 9, 336, P = 0.66); or decrease dosage (F = 1.59, df = 9, 336, Table 2 Frequency of current DSM-IV anxiety disorder diagnoses. Anxiety disorder diagnosis
Frequency
Social phobia Generalized anxiety disorder Post traumatic stress disorder Panic disorder with agoraphobia Specific phobia Panic disorder Obsessive-compulsive disorder Agoraphobia without panic disorder
102 71 65 62 36 27 21 8
Patients could receive more than one diagnosis.
(41%) (29%) (26%) (25%) (15%) (11%) (8%) (3%)
Cohen's d
Cramer's V 0.144
0.04 − 0.10 − 0.29 − 0.29 0.55 − 0.53 − 0.47 0.19 − 0.31
P = 0.12), of existing pharmacotherapies were found. In addition, although the percentage of patients was higher in the group with comorbid anxiety disorders, no significant differences were found in the frequency of recommendations to begin psychotherapy (0.52 vs. 0.69, χ2 = 13.9, df = 1, n = 346, P = 0.128). To further evaluate differences in added pharmacotherapies, Fig. 1 depicts the frequency with which various pharmacotherapies were added at the end of the initial evaluation. No statistically significant differences were found between the two groups in the proportion of pharmacotherapies added at the end of the initial evaluation for any drug class (P N 0.05 for all drug classes). Within the group of patients with comorbid anxiety disorders, we were interested in examining the relationship between the burden of anxiety comorbidity and the total number of pharmacotherapies added after the initial evaluation. Using a linear regression model, no significant relationship was found to exist between the total number of anxiety disorders and the total number of pharmacotherapies added at intake (F = 0.22, df = 1, 344, P = 0.64).
4. Discussion To our knowledge, this investigation represents the first systematic examination of differences in initial treatment choices in a carefully characterized sample of depressed outpatients (with or without comorbid anxiety disorders) seen within a community-based practice. Our primary finding, that clinicians prescribe a greater total number of Table 3 Proportional distribution of pharmacotherapies being taken before evaluation. Drug type
Anxiety disorder Entire sample Absent (n = 346) (n = 98)
SSRI
47.7% (165)
SNRI
20.5% (71)
Benzodiazepine 24.3% (84) Hypnotic
5.2% (18)
TCA
1.4% (5)
Stimulant
1.2% (4)
Mood stabilizer 4.3% (15) Anticonvulsant
1.7% (6)
Other anxiolytic 0.9% (3) MAOI
0.3% (1)
41.8% (41)
Present (n = 248)
Test statistic
50.0% (124) df = 1, χ2 = 1.91 23.5% (23) 19.4% (48) df = 1, χ2 = 0.73 19.4% (19) 26.2% (65) df = 1, χ2 = 1.78 4.1% (4) 5.6% (14) df = 1, χ2 = 0.34 1.0% (1) 1.6% (4) df = 1, χ2 = 0.17 0 1.6% (4) df = 1, χ2 = 1.60 2.0% (2) 5.2% (13) df = 1, χ2 = 1.74 1.0% (1) 2.0% (5) df = 1, χ2 = 0.41 1.0% (1) 8% (2) df = 1, χ2 = 0.04 0 .4% (1) df = 1, χ2 = 0.39
P
Cramer's V
0.17
0.07
0.39 0.05 0.18
0.07
0.55 0.03 0.68 0.02 0.21 0.07 0.19
0.07
0.52 0.1 0.85 0.01 0.53 0.03
10
T. Petersen et al. / Psychiatry Research 169 (2009) 7–11
Table 4 Frequency of pharmacological treatment recommendations. Drug type
Anxiety disorder Entire sample Absent (n = 346) (n = 98)
SSRI
43.6% (151)
SNRI
33.2% (114)
Benzodiazepine
11.0% (38)
Hypnotic
4.3% (15)
TCA
4.3% (15)
Stimulant
2.6% (8)
Mood Stabilizer
0.8% (2)
Anticonvulsant Other anxiolytic MAOI
0 0 0
Present (n = 248)
Test statistic
42.9% (42) 44.1% (109) df = 1, χ2 = 0.04 28.6% (27) 35.0% (87) df = 1, χ2 = 1.80 16.1% (16) 9.0% (22) df = 1, χ2 = 4.01 5.4% (5) 4.0% (10) df = 1, χ2 = 0.19 5.4% (5) 4.0% (10) df = 1, χ2 = 0.19 1.8% (2) 2.8% (6) df = 1, χ2 = 0.04 0 1.1% (2) df = 1, χ2 = 0.79 0 0 0 0 0 0
P
Cramer's V
0.85 0.01 0.18 0.07 0.05 0.11 0.65 0.02 0.65 0.02 0.83 0.01 0.37 0.05
pharmacotherapies to depressed outpatients with one or more comorbid anxiety disorders, when compared with depressed patients without comorbid anxiety, is consistent with recent studies suggesting that different treatment approaches may be needed for depressed patients with the additional burden of comorbid anxiety. As an example, Brown and colleagues (Brown et al., 1996) examined treatment outcomes for depressed patients with and without a lifetime anxiety disorder, and found that certain psychopharmacologic and psychotherapeutic techniques are less effective in those patients with comorbid anxiety disorders. In this study, 276 patients were included in a randomized, controlled trial comparing three treatment conditions: usual care without any specialized treatment for depression, interpersonal psychotherapy, and nortriptyline. Patients in the latter two groups demonstrated significant improvement, regardless of the presence of anxiety disorders. However, improvement was greater for patients without comorbid anxiety disorders, and treatment discontinuation rates were lower for this group as well. As compared with the current investigation, Brown and colleagues' study was focused primarily on formally comparing treatment outcomes among randomly assigned patients, whereas our investigation focused on capturing treatment selections, not treatment outcome.
In another related investigation comparing sertraline with desipramine for the treatment of patients meeting criteria for both major depressive disorder and obsessive compulsive disorder, Hoehn-Saric et al. (2000) found that sertraline was more effective than desipramine in reducing both depressive and anxiety symptoms. While this study was limited to comparing two specific psychopharmacologic treatments, it nonetheless underscores the importance of making differential treatment recommendations when both depression and anxiety disorders are present. Finally, a review of evidencebased treatments for anxiety disorders with comorbid psychiatric conditions suggests that while SSRI antidepressants should often be used as first-line treatments for patients with depression and comorbid anxiety disorders, other compounds including benzodiazepines, anticonvulsants, and atypical antipsychotics may be essential in supplementing standard treatments, in order to maximize effectiveness (Pollack, 2005). As these studies highlight, the presence of comorbid anxiety in the depressed patient should alert the clinician that treatment modifications may be necessary to maximize chances for positive outcomes. Despite having been established as the most common comorbid psychiatric condition in depressed patients, and that anxiety comorbidity has been associated with lower psychosocial functioning and worse treatment outcomes, definitive diagnostic subtyping and associated consensus treatment guidelines for depression with comorbid anxiety, equivalent to the APA practice guidelines (American Psychiatric Association, 2000), for example, have not yet been created. As the field of clinical research comes to rely more heavily on effectiveness trial designs – compared with traditional efficacy trials – more and more information will become available to elucidate what psychopharmacologic and psychosocial treatment approaches are most useful for these more complex clinical presentations. This will likely lead to more formal definition of the depression with comorbid anxiety subtype (e.g. qualifying as a DSM diagnosis) and such findings will be more readily generalize to front line practitioners. Although significant progress has been made in establishing the existing of the anxious depression subtype (Fava et al., 2004, 2006) and studying related clinical characteristics and treatment outcomes (Souery et al., 2007), such research often does not directly impact the front line practitioner as much as changes in formal diagnostic classification systems (e.g. DSM nosology). Reimbursement systems in the United States reinforce this reliance on acceptable diagnostic coding in order to insure payment. Findings from this investigation confirm that practitioners do in fact alter treatment approaches for the depressed
Fig. 1. Added pharmacotherapies for patients with and without comorbid anxiety disorders.
T. Petersen et al. / Psychiatry Research 169 (2009) 7–11
patient with comorbid anxiety, but the lack of specificity of our findings (e.g. one particular medication added more frequently than others) underscores a need for more definitive treatment guidelines for this common subtype of depression. Such guidelines should also include recommendations as to what type of psychotherapy, when added to the treatment regimen, can maximize positive outcomes for this patient population. The greatest strength of this investigation is that data were derived from a large database containing well characterized information concerning patient demographics and clinical features, as well as initial treatment recommendations. Additionally, these data represent patients seen in a general outpatient practice setting rather than those seen through traditional clinical trials. Despite this, it is important to note that all patients for whom data were analyzed were insured, and as a result, our findings may not generalize to the uninsured patient population, whose demographic and clinical presentation may differ in certain ways (Lesser et al., 2007). In addition, all study data were collected from a single practice, also limiting generalizability, and several clinical and demographic characteristics of this sample, including gender ratio, illness duration, and rate of comorbid anxiety disorders, were not representative of the general patient population. Further studies focusing on the outcomes of the treatment choices for depressed patients with comorbid anxiety are thus essential in formulating the most effective method of treatment and improving notoriously low success rates in this large, complex patient population. Depressed patients with comorbid anxiety disorders often experience lower levels of psychosocial functioning, greater illness severity and poorer treatment outcomes when compared with depressed patients without such comorbidities. Unfortunately, little empirical evidence is available to guide what treatment modifications may be necessary to address these more complex clinical presentations often encountered in the community-based practice setting. The current investigation is the first to systematically collect data concerning how the presence of comorbid anxiety, in the depressed outpatient, may impact treatments selected at the end of initial evaluation. Results indicate that practitioners tend to prescribe a greater number of psychotropic medications to depressed patients with comorbid anxiety, but do not make any other substantive changes in treatment approach. Further investigation is needed to establish which approaches – both pharmacologic and psychotherapeutic – are most efficacious for this patient population.
References American Psychiatric Association, 2000. Practice Guideline for the Treatment of Patients with Major Depressive Disorder (revision). APA, Washington, DC. Bakish, D., 1999. The patient with comorbid depression and anxiety: the unmet need. Journal of Clinical Psychiatry 60, 20–24. Beck, A.T., 1979. Cognitive Therapy of Depression. Guilford Press, New York.
11
Brown, C., Schulberg, H.C., Madonia, M.J., Shear, M.K., Houck, P.R., 1996. Treatment outcomes for primary care patients with major depression and lifetime anxiety disorders. American Journal of Psychiatry 153, 1293–1300. Farabaugh, A., Fava, M., Mischoulon, D., Sklarsky, K., Petersen, T., 2005. Relationships between major depressive disorder and comorbid anxiety and personality disorders. Comprehensive Psychiatry 46, 266–271. Fava, M., Alpert, J.E., Carmin, C.N., Wisniewski, S.R., Trivedi, M.H., Biggs, M.M., ShoresWilson, K., Morgan, D., Schwartz, T., Balasubramani, G.K., Rush, A.J., 2004. Clinical correlates and symptom patterns of anxious depression among patients with major depressive disorder in STAR⁎D. Psychological Medicine 34, 1299–1308. Fava, M., Rush, A.J., Alpert, J.E., Carmin, C.N., Balasubramani, G.K., Wisniewski, S.R., Trivedi, M.H., Biggs, M.M., Shores-Wilson, K., 2006. What clinical and symptom features and comorbid disorders characterize outpatients with anxious major depressive disorder: a replication and extension. Canadian Journal of Psychiatry 51, 823–835. Feske, U., Frank, E., Kupfer, D.J., Shear, M.K., Weaver, E., 1998. Anxiety as a predictor of response to interpersonal psychotherapy for recurrent major depression: an exploratory investigation. Depression and Anxiety 8, 135–141. Gorman, J.M., 1996–1997. Comorbid depression and anxiety spectrum disorders. Depression and Anxiety 4, 160–168. Hoehn-Saric, R., Ninan, P., Black, D.W., Stahl, S., Greist, J.H., Lydiard, B., McElroy, S., Zajecka, J., Chapman, D., Clary, C., Harrison, W., 2000. Multicenter double-blind comparison or sertraline and desipramine for concurrent obsessive–compulsive and major depressive disorders. Archives of General Psychiatry 57, 76–82. Joffe, R.T., Bagby, R.M., Levitt, A., 1993. Anxious and nonanxious depression. American Journal of Psychiatry 150, 1257–1258. Kuzel, R., 1996. Treating comorbid depression and anxiety. Journal of Family Practice 43, S45–S53. Lesser, I.M., Leuchter, A.F., Trivedi, M.H., Davis, L.L., Wisniewski, S.R., Balasubramani, G.K., Fava, M., Rush, A.J., 2007. Insured and non-insured depressed outpatients: how do they compare? Annals of Clinical Psychiatry 19, 73–82. Nutt, D., 1997. Management of patients with depression associated with anxiety symptoms. Journal of Clinical Psychiatry 58, 11–16. Pollack, M.H., 2005. Comorbid anxiety and depression. Journal of Clinical Psychiatry 66, 22–29. Rihmer, Z., 2007. Suicide risk in mood disorders. Current Opinions in Psychiatry 20, 17–22. Rush, A.J., Fava, M., Wisniewski, S.R., Lavori, P.W., Trivedi, M.H., Sackeim, H.A., Thase, M.E., Nierenberg, A.A., Quitkin, F.M., Kashner, T.M., Kupfer, D.J., Rosenbaum, J.F., Alpert, J., Stewart, J.W., McGrath, P.J., Biggs, M.M., Shores-Wilson, K., Lebowitz, B.D., Ritz, L., Niederehe, G., STAR⁎D Investigators Group, 2004. Sequenced Treatment Alternatives to Relieve Depression (STAR⁎D): rationale and design. Controlled Clinical Trials 25, 119–142. Souery, D., Oswald, P., Massat, I., Bailer, U., Bollen, J., Demyttenaere, K., Kasper, S., Lecrubier, Y., Montgomery, S., Serretti, A., Zohar, J., Mendlewicz, J., Group for the Study of Resistant Depression, 2007. Clinical factors associated with treatment resistance in major depressive disorder; results from a European multicenter study. Journal of Clinical Psychiatry 68, 1062–1070. STATA Version 9.2 SE 2007. Statacorp LP. 2007. Tollefson, G.D., Holman, S.L., Sayler, M.E., Potvin, J.H., 1994. Fluoxetine, placebo, and tricyclic antidepressants in major depression with and without anxious features. Journal of Clinical Psychiatry 55, 50–59. Zimmerman, M., Chelminski, I., 2003. Clinician recognition of anxiety disorders in depressed outpatients. Journal of Psychiatric Research 37, 325–333. Zimmerman, M., Mattia, J.I., 1999. The reliability and validity of a screening questionnaire for 14 DSM-IV axis I disorders (The Psychiatric Diagnostic Screening Questionnaire). Journal of Clinical Psychiatry 60, 677–683. Zimmerman, M., McDermut, W., Mattia, J.I., 2000. Frequency of anxiety disorders in psychiatric outpatients with major depressive disorder. American Journal of Psychiatry 157, 1337–1340. Zimmerman, M., Posternak, M.A., Chelminski, I., 2002. Symptom severity and exclusion from antidepressant efficacy trials. Journal of Clinical Psychopharmacology 22, 610–614.
Contents lists available at ScienceDirect
Psychiatry Research j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / p s yc h r e s
Do comorbid anxiety disorders impact treatment planning for outpatients with major depressive disorder? Timothy Petersen a,⁎, Charissa F. Andreotti a, Iwona Chelminski b, Diane Young b, Mark Zimmerman b a b
Massachusetts General Hospital, Department of Psychiatry, Division of Postgraduate Education, 1 Bowdoin Square, 7th floor, Boston, MA 02114, United States Rhode Island Hospital Department of Psychiatry and Human Behavior, Brown University School of Medicine, Rhode Island Hospital, Providence, RI 02905, United States
a r t i c l e
i n f o
Article history: Received 1 October 2007 Received in revised form 4 August 2008 Accepted 23 October 2008 Keywords: Major depressive disorder Treatment planning Comorbid anxiety disorders
a b s t r a c t Research indicates that depressed patients with comorbid anxiety disorders have a poorer long-term course of illness, are less responsive to treatment, and may experience greater deficits in psychosocial functioning, when compared with depressed patients without comorbid anxiety disorders. The objective of this study was to examine, through use of a large, well-characterized clinical database, how clinicians may modify treatment recommendations in depressed outpatients when anxiety disorders are present. A group of 346 case records, derived from the Methods to Improve Diagnostic Assessment and Services (MIDAS) project at Rhode Island Hospital, were examined to determine what treatment recommendations were made immediately after diagnosis. Psychopharmacological and psychotherapeutic treatments were classified to capture differences in recommendations between groups. Demographic and clinical characteristics were compared for patients with (n = 248) and without (n = 98) comorbid anxiety disorders. Utilizing logistic regression models, we found patients with anxiety disorders had a greater number of psychopharmacological therapies included as part of their initial treatment plan, but no differences were found in initial psychotherapeutic interventions. Our results indicate that practitioners are making unique recommendations based on comorbid anxiety diagnoses, but outcome studies are now needed to determine the most effective treatment methods for this patient population. © 2008 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Anxiety disorders are the most common comorbid psychiatric condition in patients with major depressive disorder (Zimmerman et al., 2000; Rush et al., 2004). Analyses of epidemiological databases indicate that 50% of all depressed patients meet criteria for at least one anxiety disorder, and a large proportion of these cases meet criteria for multiple anxiety disorders (Gorman, 1996–1997; Zimmerman et al., 2000). Recent findings from the Sequenced Treatment Alternatives to Relieve Depression (STAR⁎D) investigation confirm the high prevalence of comorbid anxiety symptoms in depressed outpatients. In this effectiveness study of over 4000 depressed outpatients, treated within both primary and specialty care settings, 46% of patients were found to meet criteria for anxious depression, defined as Major Depressive Disorder (MDD) with high levels of anxiety symptoms (Fava et al., 2004). The presence of anxious depression was associated with greater severity of depressive symptoms and a higher rate of suicidal ideation (Fava et al., 2004; Rihmer, 2007). Other research suggests that the presence of comorbid anxiety disorders is associated with poorer antidepressant treatment outcomes, higher levels of health care utilization, a worse long-term course of depression, occurrence of cluster A and C personality disorders (Farabaugh et al., ⁎ Corresponding author. E-mail address: [email protected] (T. Petersen). 0165-1781/$ – see front matter © 2008 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.psychres.2008.10.036
2005), and lower psychosocial functioning (Brown et al., 1996). Despite these findings, comorbid anxiety disorders are often underrecognized in community-based care settings (Zimmerman and Chelminski, 2003). A number of studies have examined psychopharmacologic treatment approaches for depressed patients with anxiety symptoms, which include the use of antidepressants, anxiolytics, and other novel compounds (Kuzel, 1996; Nutt, 1997). Psychotherapeutic approaches have also been tested and include the use of traditional cognitivebehaviorally based approaches (Beck, 1979) as well as interpersonally oriented psychotherapies (Feske et al., 1998). While some outcome studies suggest that standard evidence-based treatments are effective in reducing both depressive and anxiety symptoms (Joffe et al., 1993; Tollefson et al., 1994), other studies recommend modifications to traditional treatment approaches in order to optimally manage depression with comorbid anxiety (Kuzel, 1996; Gorman, 1996– 1997). For example, Bakish (1999) supports the use of the novel serotonin–norepinephrine reuptake inhibitor, venlafaxine, for depression with concomitant anxiety, and work by Kuzel (1996) finds nefazodone to also aid in treating depression with comorbid anxiety features. To date, however, no psychopharmacologic or psychotherapeutic treatment has been established in randomized controlled trials as superior in the treatment of depression with comorbid anxiety. In order to better understand what treatments are most effective in managing depressed patients with comorbid anxiety disorders, it is
8
T. Petersen et al. / Psychiatry Research 169 (2009) 7–11
critical to first understand how community-based practitioners are currently approaching treatment of this patient population. The objective of this study was to examine, in a large sample of outpatients with major depressive disorder, how the presence of comorbid anxiety disorders may impact treatment choices. Based on previous research suggesting that standard treatments may need to be modified to enhance outcomes for depressed patients with comorbid anxiety disorders, we hypothesized that treatment selections would significantly differ between depressed outpatients with and without comorbid anxiety disorders; specifically that patients with comorbid anxiety disorders would have a greater total number of treatments (psychopharmacologic and/or psychotherapeutic) initiated. 2. Method
utilized in this practice consists primarily of a semi-structured interview (Structured Clinical Interview for DSM-IV (SCID), which is used to determine a comprehensive diagnostic profile for each individual. This profile consists of formal, multiaxial DSM-IV diagnoses as well as several other clinical characteristics, which include number of lifetime depressive episodes, length of the current depressive episode, number of psychiatric hospitalizations, a rating of the number of work weeks missed due to depressive illness, and a rating of current social functioning. For the purposes of data entry, a coding scheme was created to document current and recommended treatments. Psychopharmacologic treatments were grouped into categories based on drug classification including selective serotonin reuptake inhibitor (SSRI), tricyclic, monoamine oxidase inhibitor (MAOI), and serotonin–norepinephrine reuptake inhibitor (SNRI) antidepressants as well as benzodiazepines, other anxiolytics, hypnotics, mood stabilizers, anticonvulsants, and stimulants. Treatment strategies were grouped into five categories (Add, Discontinue, Increase Dose, Decrease Dose, or Continue) in order to represent recommendations made at intake. This coding scheme allowed accurate capture of all possible treatment combinations, including existing treatment regimens as well as changes made after the initial evaluation. All data entry and analyses were conducted using STATA/SE version 9.2 (STATA/SE 9.2, 2007).
2.1. Overview 2.4. Data analysis Data for this investigation were derived from the Methods to Improve Diagnostic Assessment and Services (MIDAS) project, which represents an integration of research methodology into a community-based outpatient practice affiliated with an academic medical center (Zimmerman et al., 2002). To date, 2600 psychiatric outpatients have been evaluated with a semistructured diagnostic interview in this Rhode Island Hospital Department of Psychiatry outpatient practice. This private practice group predominantly treats individuals with medical insurance (including Medicare but not Medicaid) on a fee-for-service basis, and it is distinct from the hospital's outpatient residency training clinic that predominantly serves lower income, uninsured, and medical assistance patients. The primary referral sources to the practice are primary care physicians and psychotherapists. Not all patients who present for treatment at this practice participate in the research protocols. Patients are offered the opportunity to have a more comprehensive evaluation as part of the research program, though they are not required to undergo this evaluation. Importantly, as reported previously, patients in this practice who have participated in the comprehensive research evaluation have been shown to be similar to non-participating patients (Zimmerman and Mattia, 1999). Patients were interviewed by a diagnostic rater who administered the Structured Clinical Interview for DSM-IV (SCID). To study the psychometric performance of the DSM-IV symptom criteria for major depression, it was necessary to modify the SCID and eliminate the skip-out that curtails the depression module for patients who did not report either depressed mood or anhedonia. Thus, we inquired about all of the symptoms of depression for all patients. For compound criteria that encompass more than one symptom (e.g., indecisiveness or impaired concentration; increased sleep or insomnia), we made separate ratings of each component of the diagnostic criterion. Thus, the nine DSM-IV symptom criteria were broken down into 17 separate items. The diagnostic raters are highly trained and have been monitored throughout the project to minimize rater drift. Diagnostic raters included PhD-level psychologists and research assistants with college degrees in the social or biological sciences. Research assistants received 3 to 4 months of training during which they observed at least 20 interviews, and they were observed and supervised in their administration of more than 20 evaluations. Psychologists only observed five interviews; however, they, too, were observed and supervised in their administration of 15 to 20 evaluations. During the course of training, the senior author met with each rater to review the interpretation of every item on the SCID. Also during training every interview was reviewed on an item-byitem basis by the senior rater who observed the evaluation. At the end of the training period, the raters were required to demonstrate exact, or near exact, agreement with a senior diagnostician on five consecutive evaluations. Throughout the MIDAS project, ongoing supervision of the raters consisted of weekly diagnostic case conferences involving all members of the team. In addition, the item ratings of every case were reviewed by the senior author. The Rhode Island Hospital institutional review committee approved the research protocol, and all patients provided informed, written consent. 2.2. Selection and description of participants Treatment planning data were collected from evaluation reports of depressed outpatients assessed through the MIDAS program. Patients who were 18 years old or older and met criteria for a current, principal diagnosis of major depressive disorder were eligible for inclusion in this data analysis. Patients were selected only if the principal diagnosis listed on their evaluation report was major depressive disorder. This did not include patients with principal diagnoses of other disorders, including bipolar or anxiety disorder, even if major depressive disorder was listed as a secondary diagnosis. The first 2000 evaluation reports collected as part of the MIDAS program were screened, and a total of 346 (17.3%) participants (215 females and 131 males), were included in this analysis.
Our data analytic approach consisted of first calculating descriptive statistics for the entire sample, and then by patient group (with vs. without comorbid anxiety disorders). A patient was characterized as belonging to the “with a comorbid anxiety disorder” group if he/she met DSM-IV criteria for one or more comorbid anxiety disorders. The second step of our data analytic approach was to utilize appropriate parametric (unpaired t-tests) and non-parametric techniques (chi-square analyses) to evaluate differences between groups in current treatments, treatment recommendations, illness characteristics and demographic features. Post hoc tests were used, as needed, to determine the specific nature of these differences. Logistic regression models were developed to ascertain which variables and features best account for group membership (with or without comorbid anxiety disorders). Independent variables representing treatment approaches as well as demographic and clinical characteristics on which the groups differed were included in the logistic regression analyses. Adjustments in threshold for significance were made for multiple comparisons.
3. Results 3.1. Sample description A description of demographic and clinical characteristics of the sample is shown in Table 1. Mean Severity of illness, as measured by the Clinical Global Impression of Severity Scale (CGI-S) was 2.9 ± 0.1, suggesting a moderate overall illness severity. Mean Global Assessment of Functioning was 52.4 ± 0.4, indicating a moderate to serious level of symptomatology. Although there was no significant difference in the percentages of males and females in the sample, women were more likely to receive a secondary diagnosis for one or more comorbid anxiety disorders, (n = 346, χ2 = 7.18, df = 1, P = 0.007). 3.2. Group comparison Comparison of patient groups on demographic and clinical characteristics is also shown in Table 1. Patients with comorbid anxiety disorders had a significantly higher Clinical Global Impression S Scale, lower Global Assessment of Functioning,), and poorer current social functioning. The comorbid anxiety disorder group also had a higher number of psychiatric hospitalizations over the last 5 years, missed a greater amount of work due to depressive symptoms over the last 5 years, had a greater number of lifetime depressive episodes, and a longer duration of current depressive episode. Substance use disorders were significantly more common in the patient group with comorbid anxiety disorders (52.8% group with, 34.7% group without, n = 346, χ2 = 9.25, df = 1, P = 0.002). No significant differences were found in the presence of bipolar spectrum disorders, psychotic disorders, dysthymia, and axis II disorders between the two groups. 3.3. Frequency of comorbid anxiety disorders
2.3. Study procedures The following information was gathered from the evaluation reports: current treatment regimen at intake (including medication and non-medication approaches) and the treatments recommended at the conclusion of the evaluation. The evaluation
In the sub-group of patients with comorbid anxiety disorders (n = 248), the frequency of specific diagnoses is shown in Table 2. As shown in Table 2, social phobia, generalized anxiety disorder, and
T. Petersen et al. / Psychiatry Research 169 (2009) 7–11
9
Table 1 Descriptive statistics and group comparisons for demographic and clinical characteristics. Descriptive statistic
Gender (female) Age (years) Initial medications Lifetime depressive episodes Cognitive global impression-severity Global assessment of functioning Current social functioning Psychiatric hospitalization rate in past 5 years Number of work weeks missed Length of current depressive episode
Anxiety disorder Entire sample (n = 346)
Absent (n = 98)
Present (n = 248)
Test statistic
P
62.1% 40.4 ± 0.7 2.6 ± 0.1 5.5 ± 0.8 2.9 ± 0.1 52.4 ± 0.4 3.2 ± 0.1 0.4 ± 0.1 2.2 ± 0.1 207.7 ± 21.8
51.0% 40.8 2.4 2.9 2.7 55.3 3.4 0.1 1.7 126
66.5% 40.3 2.7 6.5 2.9 51.3 2.8 0.5 2.4 293
df = 1, χ2 = 7.18 df = 344, t = 0.3 df = 344, t = − .8 df = 344, t = − 2.1 df = 344, t = − 2.5 df = 344, t = 4.6 df = 344, t = − 4.4 df = 344, t = − 3.6 df = 344, t = − 3.4 df = 344, t = − 2.4
0.007 0.74 0.4 0.04 0.014 b 0.001 b 0.001 b 0.001 b 0.001 b 0.001
post-traumatic stress disorder were most prevalent as comorbid anxiety disorders. The total number of comorbid anxiety disorders for participants in this sub-group ranged from one to five, with 38% of patients (n = 133) having one diagnosed disorder, 19% (n = 66) having two disorders, 11% (n = 38) having three disorders, 2.9% (n = 10) having four disorders, and 0.29% (n = 1) having five disorders.
3.4. Evaluation of differences between groups in current and recommended treatments The distribution of pharmacotherapies, by patient group, taken before evaluation (current treatments) is shown in Table 3. No statistically significant differences were found between the two groups in the proportion of any one class of pharmacotherapy taken before evaluation (for all comparisons, P N 0.05). Similarly, no statistically significant difference was found between the two groups in the proportion of patients currently engaged in psychotherapy prior to intake, (36.7% group with, 33.2% group without, n = 346, χ2 = 0.39, df = 1, P = 0.53). The distribution of pharmacotherapies, by patient group, added after initial evaluation is shown in Table 4. No significant differences were found to exist between the two groups in the proportion of any one class of pharmacotherapy added. Logistic regression models were utilized with independent variables representing treatment approaches as well as clinical characteristics on which the groups differed, including gender, CGI-S, GAF, current social functioning, number of psychiatric hospitalizations, hospitalization rate over a 5year period, missed weeks of work due to depressive illness, number of lifetime depressive episodes, duration of current depressive episodes, and frequency of substance use disorders. The dependent variable was group membership (with vs. without anxiety disorders). Patients with comorbid anxiety disorders had a greater total number of psychopharmacological therapies added when compared with patients without comorbid anxiety disorders, (F = 2.50, df = 5, 340, P = 0.031). No differences in treatment recommendations to discontinue, (F = 0.28, df = 9, 336, P = 0.98); increase dosage, (F = 0.75, df = 9, 336, P = 0.66); or decrease dosage (F = 1.59, df = 9, 336, Table 2 Frequency of current DSM-IV anxiety disorder diagnoses. Anxiety disorder diagnosis
Frequency
Social phobia Generalized anxiety disorder Post traumatic stress disorder Panic disorder with agoraphobia Specific phobia Panic disorder Obsessive-compulsive disorder Agoraphobia without panic disorder
102 71 65 62 36 27 21 8
Patients could receive more than one diagnosis.
(41%) (29%) (26%) (25%) (15%) (11%) (8%) (3%)
Cohen's d
Cramer's V 0.144
0.04 − 0.10 − 0.29 − 0.29 0.55 − 0.53 − 0.47 0.19 − 0.31
P = 0.12), of existing pharmacotherapies were found. In addition, although the percentage of patients was higher in the group with comorbid anxiety disorders, no significant differences were found in the frequency of recommendations to begin psychotherapy (0.52 vs. 0.69, χ2 = 13.9, df = 1, n = 346, P = 0.128). To further evaluate differences in added pharmacotherapies, Fig. 1 depicts the frequency with which various pharmacotherapies were added at the end of the initial evaluation. No statistically significant differences were found between the two groups in the proportion of pharmacotherapies added at the end of the initial evaluation for any drug class (P N 0.05 for all drug classes). Within the group of patients with comorbid anxiety disorders, we were interested in examining the relationship between the burden of anxiety comorbidity and the total number of pharmacotherapies added after the initial evaluation. Using a linear regression model, no significant relationship was found to exist between the total number of anxiety disorders and the total number of pharmacotherapies added at intake (F = 0.22, df = 1, 344, P = 0.64).
4. Discussion To our knowledge, this investigation represents the first systematic examination of differences in initial treatment choices in a carefully characterized sample of depressed outpatients (with or without comorbid anxiety disorders) seen within a community-based practice. Our primary finding, that clinicians prescribe a greater total number of Table 3 Proportional distribution of pharmacotherapies being taken before evaluation. Drug type
Anxiety disorder Entire sample Absent (n = 346) (n = 98)
SSRI
47.7% (165)
SNRI
20.5% (71)
Benzodiazepine 24.3% (84) Hypnotic
5.2% (18)
TCA
1.4% (5)
Stimulant
1.2% (4)
Mood stabilizer 4.3% (15) Anticonvulsant
1.7% (6)
Other anxiolytic 0.9% (3) MAOI
0.3% (1)
41.8% (41)
Present (n = 248)
Test statistic
50.0% (124) df = 1, χ2 = 1.91 23.5% (23) 19.4% (48) df = 1, χ2 = 0.73 19.4% (19) 26.2% (65) df = 1, χ2 = 1.78 4.1% (4) 5.6% (14) df = 1, χ2 = 0.34 1.0% (1) 1.6% (4) df = 1, χ2 = 0.17 0 1.6% (4) df = 1, χ2 = 1.60 2.0% (2) 5.2% (13) df = 1, χ2 = 1.74 1.0% (1) 2.0% (5) df = 1, χ2 = 0.41 1.0% (1) 8% (2) df = 1, χ2 = 0.04 0 .4% (1) df = 1, χ2 = 0.39
P
Cramer's V
0.17
0.07
0.39 0.05 0.18
0.07
0.55 0.03 0.68 0.02 0.21 0.07 0.19
0.07
0.52 0.1 0.85 0.01 0.53 0.03
10
T. Petersen et al. / Psychiatry Research 169 (2009) 7–11
Table 4 Frequency of pharmacological treatment recommendations. Drug type
Anxiety disorder Entire sample Absent (n = 346) (n = 98)
SSRI
43.6% (151)
SNRI
33.2% (114)
Benzodiazepine
11.0% (38)
Hypnotic
4.3% (15)
TCA
4.3% (15)
Stimulant
2.6% (8)
Mood Stabilizer
0.8% (2)
Anticonvulsant Other anxiolytic MAOI
0 0 0
Present (n = 248)
Test statistic
42.9% (42) 44.1% (109) df = 1, χ2 = 0.04 28.6% (27) 35.0% (87) df = 1, χ2 = 1.80 16.1% (16) 9.0% (22) df = 1, χ2 = 4.01 5.4% (5) 4.0% (10) df = 1, χ2 = 0.19 5.4% (5) 4.0% (10) df = 1, χ2 = 0.19 1.8% (2) 2.8% (6) df = 1, χ2 = 0.04 0 1.1% (2) df = 1, χ2 = 0.79 0 0 0 0 0 0
P
Cramer's V
0.85 0.01 0.18 0.07 0.05 0.11 0.65 0.02 0.65 0.02 0.83 0.01 0.37 0.05
pharmacotherapies to depressed outpatients with one or more comorbid anxiety disorders, when compared with depressed patients without comorbid anxiety, is consistent with recent studies suggesting that different treatment approaches may be needed for depressed patients with the additional burden of comorbid anxiety. As an example, Brown and colleagues (Brown et al., 1996) examined treatment outcomes for depressed patients with and without a lifetime anxiety disorder, and found that certain psychopharmacologic and psychotherapeutic techniques are less effective in those patients with comorbid anxiety disorders. In this study, 276 patients were included in a randomized, controlled trial comparing three treatment conditions: usual care without any specialized treatment for depression, interpersonal psychotherapy, and nortriptyline. Patients in the latter two groups demonstrated significant improvement, regardless of the presence of anxiety disorders. However, improvement was greater for patients without comorbid anxiety disorders, and treatment discontinuation rates were lower for this group as well. As compared with the current investigation, Brown and colleagues' study was focused primarily on formally comparing treatment outcomes among randomly assigned patients, whereas our investigation focused on capturing treatment selections, not treatment outcome.
In another related investigation comparing sertraline with desipramine for the treatment of patients meeting criteria for both major depressive disorder and obsessive compulsive disorder, Hoehn-Saric et al. (2000) found that sertraline was more effective than desipramine in reducing both depressive and anxiety symptoms. While this study was limited to comparing two specific psychopharmacologic treatments, it nonetheless underscores the importance of making differential treatment recommendations when both depression and anxiety disorders are present. Finally, a review of evidencebased treatments for anxiety disorders with comorbid psychiatric conditions suggests that while SSRI antidepressants should often be used as first-line treatments for patients with depression and comorbid anxiety disorders, other compounds including benzodiazepines, anticonvulsants, and atypical antipsychotics may be essential in supplementing standard treatments, in order to maximize effectiveness (Pollack, 2005). As these studies highlight, the presence of comorbid anxiety in the depressed patient should alert the clinician that treatment modifications may be necessary to maximize chances for positive outcomes. Despite having been established as the most common comorbid psychiatric condition in depressed patients, and that anxiety comorbidity has been associated with lower psychosocial functioning and worse treatment outcomes, definitive diagnostic subtyping and associated consensus treatment guidelines for depression with comorbid anxiety, equivalent to the APA practice guidelines (American Psychiatric Association, 2000), for example, have not yet been created. As the field of clinical research comes to rely more heavily on effectiveness trial designs – compared with traditional efficacy trials – more and more information will become available to elucidate what psychopharmacologic and psychosocial treatment approaches are most useful for these more complex clinical presentations. This will likely lead to more formal definition of the depression with comorbid anxiety subtype (e.g. qualifying as a DSM diagnosis) and such findings will be more readily generalize to front line practitioners. Although significant progress has been made in establishing the existing of the anxious depression subtype (Fava et al., 2004, 2006) and studying related clinical characteristics and treatment outcomes (Souery et al., 2007), such research often does not directly impact the front line practitioner as much as changes in formal diagnostic classification systems (e.g. DSM nosology). Reimbursement systems in the United States reinforce this reliance on acceptable diagnostic coding in order to insure payment. Findings from this investigation confirm that practitioners do in fact alter treatment approaches for the depressed
Fig. 1. Added pharmacotherapies for patients with and without comorbid anxiety disorders.
T. Petersen et al. / Psychiatry Research 169 (2009) 7–11
patient with comorbid anxiety, but the lack of specificity of our findings (e.g. one particular medication added more frequently than others) underscores a need for more definitive treatment guidelines for this common subtype of depression. Such guidelines should also include recommendations as to what type of psychotherapy, when added to the treatment regimen, can maximize positive outcomes for this patient population. The greatest strength of this investigation is that data were derived from a large database containing well characterized information concerning patient demographics and clinical features, as well as initial treatment recommendations. Additionally, these data represent patients seen in a general outpatient practice setting rather than those seen through traditional clinical trials. Despite this, it is important to note that all patients for whom data were analyzed were insured, and as a result, our findings may not generalize to the uninsured patient population, whose demographic and clinical presentation may differ in certain ways (Lesser et al., 2007). In addition, all study data were collected from a single practice, also limiting generalizability, and several clinical and demographic characteristics of this sample, including gender ratio, illness duration, and rate of comorbid anxiety disorders, were not representative of the general patient population. Further studies focusing on the outcomes of the treatment choices for depressed patients with comorbid anxiety are thus essential in formulating the most effective method of treatment and improving notoriously low success rates in this large, complex patient population. Depressed patients with comorbid anxiety disorders often experience lower levels of psychosocial functioning, greater illness severity and poorer treatment outcomes when compared with depressed patients without such comorbidities. Unfortunately, little empirical evidence is available to guide what treatment modifications may be necessary to address these more complex clinical presentations often encountered in the community-based practice setting. The current investigation is the first to systematically collect data concerning how the presence of comorbid anxiety, in the depressed outpatient, may impact treatments selected at the end of initial evaluation. Results indicate that practitioners tend to prescribe a greater number of psychotropic medications to depressed patients with comorbid anxiety, but do not make any other substantive changes in treatment approach. Further investigation is needed to establish which approaches – both pharmacologic and psychotherapeutic – are most efficacious for this patient population.
References American Psychiatric Association, 2000. Practice Guideline for the Treatment of Patients with Major Depressive Disorder (revision). APA, Washington, DC. Bakish, D., 1999. The patient with comorbid depression and anxiety: the unmet need. Journal of Clinical Psychiatry 60, 20–24. Beck, A.T., 1979. Cognitive Therapy of Depression. Guilford Press, New York.
11
Brown, C., Schulberg, H.C., Madonia, M.J., Shear, M.K., Houck, P.R., 1996. Treatment outcomes for primary care patients with major depression and lifetime anxiety disorders. American Journal of Psychiatry 153, 1293–1300. Farabaugh, A., Fava, M., Mischoulon, D., Sklarsky, K., Petersen, T., 2005. Relationships between major depressive disorder and comorbid anxiety and personality disorders. Comprehensive Psychiatry 46, 266–271. Fava, M., Alpert, J.E., Carmin, C.N., Wisniewski, S.R., Trivedi, M.H., Biggs, M.M., ShoresWilson, K., Morgan, D., Schwartz, T., Balasubramani, G.K., Rush, A.J., 2004. Clinical correlates and symptom patterns of anxious depression among patients with major depressive disorder in STAR⁎D. Psychological Medicine 34, 1299–1308. Fava, M., Rush, A.J., Alpert, J.E., Carmin, C.N., Balasubramani, G.K., Wisniewski, S.R., Trivedi, M.H., Biggs, M.M., Shores-Wilson, K., 2006. What clinical and symptom features and comorbid disorders characterize outpatients with anxious major depressive disorder: a replication and extension. Canadian Journal of Psychiatry 51, 823–835. Feske, U., Frank, E., Kupfer, D.J., Shear, M.K., Weaver, E., 1998. Anxiety as a predictor of response to interpersonal psychotherapy for recurrent major depression: an exploratory investigation. Depression and Anxiety 8, 135–141. Gorman, J.M., 1996–1997. Comorbid depression and anxiety spectrum disorders. Depression and Anxiety 4, 160–168. Hoehn-Saric, R., Ninan, P., Black, D.W., Stahl, S., Greist, J.H., Lydiard, B., McElroy, S., Zajecka, J., Chapman, D., Clary, C., Harrison, W., 2000. Multicenter double-blind comparison or sertraline and desipramine for concurrent obsessive–compulsive and major depressive disorders. Archives of General Psychiatry 57, 76–82. Joffe, R.T., Bagby, R.M., Levitt, A., 1993. Anxious and nonanxious depression. American Journal of Psychiatry 150, 1257–1258. Kuzel, R., 1996. Treating comorbid depression and anxiety. Journal of Family Practice 43, S45–S53. Lesser, I.M., Leuchter, A.F., Trivedi, M.H., Davis, L.L., Wisniewski, S.R., Balasubramani, G.K., Fava, M., Rush, A.J., 2007. Insured and non-insured depressed outpatients: how do they compare? Annals of Clinical Psychiatry 19, 73–82. Nutt, D., 1997. Management of patients with depression associated with anxiety symptoms. Journal of Clinical Psychiatry 58, 11–16. Pollack, M.H., 2005. Comorbid anxiety and depression. Journal of Clinical Psychiatry 66, 22–29. Rihmer, Z., 2007. Suicide risk in mood disorders. Current Opinions in Psychiatry 20, 17–22. Rush, A.J., Fava, M., Wisniewski, S.R., Lavori, P.W., Trivedi, M.H., Sackeim, H.A., Thase, M.E., Nierenberg, A.A., Quitkin, F.M., Kashner, T.M., Kupfer, D.J., Rosenbaum, J.F., Alpert, J., Stewart, J.W., McGrath, P.J., Biggs, M.M., Shores-Wilson, K., Lebowitz, B.D., Ritz, L., Niederehe, G., STAR⁎D Investigators Group, 2004. Sequenced Treatment Alternatives to Relieve Depression (STAR⁎D): rationale and design. Controlled Clinical Trials 25, 119–142. Souery, D., Oswald, P., Massat, I., Bailer, U., Bollen, J., Demyttenaere, K., Kasper, S., Lecrubier, Y., Montgomery, S., Serretti, A., Zohar, J., Mendlewicz, J., Group for the Study of Resistant Depression, 2007. Clinical factors associated with treatment resistance in major depressive disorder; results from a European multicenter study. Journal of Clinical Psychiatry 68, 1062–1070. STATA Version 9.2 SE 2007. Statacorp LP. 2007. Tollefson, G.D., Holman, S.L., Sayler, M.E., Potvin, J.H., 1994. Fluoxetine, placebo, and tricyclic antidepressants in major depression with and without anxious features. Journal of Clinical Psychiatry 55, 50–59. Zimmerman, M., Chelminski, I., 2003. Clinician recognition of anxiety disorders in depressed outpatients. Journal of Psychiatric Research 37, 325–333. Zimmerman, M., Mattia, J.I., 1999. The reliability and validity of a screening questionnaire for 14 DSM-IV axis I disorders (The Psychiatric Diagnostic Screening Questionnaire). Journal of Clinical Psychiatry 60, 677–683. Zimmerman, M., McDermut, W., Mattia, J.I., 2000. Frequency of anxiety disorders in psychiatric outpatients with major depressive disorder. American Journal of Psychiatry 157, 1337–1340. Zimmerman, M., Posternak, M.A., Chelminski, I., 2002. Symptom severity and exclusion from antidepressant efficacy trials. Journal of Clinical Psychopharmacology 22, 610–614.