Analysis of Classes Used in the Treatment of Depression by Physician-reported Severity

| February 15, 2010 | 0 Comments

by Susan Lenderts, BA; Amir H. Kalali, MD; and Richard Weisler, MD
Ms. Lenderts is Manager, Strategic Analytics, Quintiles Commercial, Falls Church, Virginia; Dr. Kalali is Vice President, Global Therapeutic Group Leader CNS, Quintiles, Inc., and Professor of Psychiatry, University of California, San Diego, California; and Dr. Weisler is from Duke University Medical Center, Durham, North Carolina, and University of North Carolina at Chapel Hill Departments of Psychiatry, Raleigh, North Carolina.

Psychiatry (Edgemont) 2010;7(2):17–23

Funding: There was no funding for the development and writing of this article.

Financial disclosures: Dr. Weisler is or has been a consultant to, on the Speaker’s Bureaus of, and/or received research support from the following: Abbott – Speaker’s Bureau, Consultant, Received Research Support, Astra Zeneca – Speaker’s Bureau, Consultant, Received Research Support, Biovail – Speaker’s Bureau, Consultant, Received Research Support, Bristol-Myers Squibb – Speaker’s Bureau, Consultant, Received Research Support, Stockholder has held or holds stock, Burroughs Wellcome – Speaker’s Bureau, Received Research Support, Cenerx – Received Research Support, Cephalon – Speaker’s Bureau, Consultant, Received Research Support, Ciba Geigy – Speaker’s Bureau, Received Research Support, CoMentis – Received Research Support, Corcept – Consultant, Cortex – Stockholder has held or holds stock, Dainippon Sumitomo Pharma America – Received Research Support, Eisai – Received Research Support, Eli Lilly – Speaker’s Bureau, Consultant, Received Research Support, Forest Laboratories – Speaker’s Bureau, Consultant, Received Research Support, GlaxoSmithKline – Speaker’s Bureau, Consultant, Received Research Support, Janssen Speaker’s Bureau, Received Research Support, Johnson & Johnson – Speaker’s Bureau, Consultant, Received Research Support, Lundbeck – Received Research Support, McNeil Pharmaceuticals – Received Research Support, Medicinova – Received Research Support, Merck – Received Research Support, Stockholder has held or holds stock, National Institute of Mental Received Research Support, Neurochem – Received Research Support, New River Pharmaceuticals – Received Research Support, Novartis Pharmaceuticals Corporation – Speaker’s Bureau, Received Research Support, Organon – Speaker’s Bureau, Consultant, Received Research Support, Otsuka America Pharma – Consultant, Pfizer – Speaker’s Bureau, Consultant, Received Research Support, Stockholder has held or holds stock, Pharmacia – Consultant, Received Research Support, ProPhase – Consultant, Repligen – Received Research Support, Saegis – Received Research Support, Sandoz – Received Research Support, Sanofi – Speaker’s Bureau, Consultant, Received Research Support, Sanofi-Synthelabo – Speaker’s Bureau, Consultant, Received Research Support, Schwabe/Ingenix – Received Research Support, Sepracor – Received Research Support, Schering Plough – Speaker’s Bureau, Shire – Speaker’s Bureau, Consultant, Received Research Support, Solvay – Speaker’s Bureau, Consultant, Synaptic – Received Research Support, Takeda – Received Research Support, Consultant, TAP – Received Research Support, UCB Pharma – Received Research Support, Validus – Speaker’s Bureau, Consultant, Vela – Received Research Support, and Wyeth – Speaker’s Bureau, Consultant, Received Research Support.Ms. Lenderts and Dr. Kalali have no conflicts of interest relevant to the content of this article.


We analyzed use of therapeutic drug classes for the treatment of depression by the three levels of physician-reported disease severity (mild, moderate, and severe) to understand if the mix of therapeutic classes used to treat depression changes as disease severity increases. Prior to analyzing drug uses in each severity category, we established that in the 12 months ending November 2009, the majority of patients (65%) were assigned a severity rating of ‘moderate’ by their physician; 27 percent and eight percent of patients were designated as ‘mild’ and ‘severe,’ respectively. In general, we found that as disease severity changes, so too do the proportions of the various therapeutic classes used in depression. The differences in class mix by severity have persisted at least since the 12 months ending December 2008, and it appears that the mix of drug classes used in severe depression are becoming less similar to the class mix for moderate depression over time.

Key words: depression, severity, physician-reported severity, depression treatment, mild depression, moderate depression, severe depression


A previous Trend Watch explored depression treatment regimens with regard to monotherapy versus combination therapy and how various mechanisms of action are used alone or in combination,[1] but we have not yet gone in depth to understand the current trends in depression treatment by physician-reported severity. In this analysis, we look at the drug classes used for the treatment of depression and delve into the distribution of class use at three levels of physician-perceived depression severity: mild, moderate, and severe. Our objective is to understand how the major drug classes used to treat depression are represented as the reported disease state becomes increasingly severe, and to determine if there have been any major changes in therapeutic class share over time within each severity category.


We obtained physician-reported audit data on depression (ICD-9 code 311) from SDI Health’s Physician Drug & Diagnosis Audit (PDDA) for moving annual periods starting with the 12 months ending December 2008 through the 12 months ending November 2009. PDDA provides data on disease states and associated treatments from a panel of approximately 3,100 office-based physicians across 29 specialties. Use of all therapeutic drug classes associated with depression diagnoses at three physician-reported severity levels (mild, moderate, and severe) was analyzed. Trend data were analyzed on moving 12-month increments, so as to maintain a robust sample size.


In order to frame this analysis, we first investigated the distribution of patients by physician-reported depression severity. Audit data suggest that in the 12 months ending November 2009, physicians assigned a disease-severity rating of ‘moderate’ to the majority (65%) of patients diagnosed with depression (Figure 1). Comparatively, 27 percent of patients were assigned a ‘mild’ depression severity rating, and eight percent of patients were assigned a ‘severe’ depression rating. Distribution of depression severity was similar when this analysis was replicated with data on drug uses and patient visits.

Having established the distribution of patients by disease severity, we then compared the use of therapeutic drug classes by share across the three depression severity categories and found that, overall, as disease severity changes, so too does the representation of the various classes used to treat depression (Figure 2). The selective serotonin reuptake inhibitor (SSRI) drug class represents the majority of drug uses associated with mild, moderate, and severe depression; however, SSRIs as share of total drug uses decreases as severity increases. SSRIs represent 72 percent of drug uses for mild depression, 54 percent of drug uses for moderate depression, and just 41 percent of all drug uses for severe depression. This observation may be attributable to decreasing incidence of antidepressant monotherapy as physician-reported depression severity increases. We also found that as physician-reported severity increases, atypical antipsychotics and benzodiazepines become increasingly represented in the mix of class uses. For example, atypical antipsychotics represent just one percent of drug uses for mild depression, but eight percent of drug uses for severe depression. Proportionally, the use of serotonin and norepinephrine reuptake inhibitors (SNRIs) increases when severity moves from mild to moderate, but the share of SNRIs is similar in moderate and severe depression.

Trend analyses of the class mix by severity over time reveals that the difference we observe today in the class mix across the three severity categories has largely persisted over the last 12 moving annual segments. Aside from the fact that SSRIs represent a slightly increasing share of drugs used for mild depression, and atypical antipsychotics appear to have displaced some benzodiazepine use in moderate depression, the distribution of drug classes used in mild and moderate depression have remained largely unchanged from the 12 months ending December 2008 to the 12 months ending November 2009. However, there have been noticeable changes in the mix of class uses in the physician-reported severe depression category (Figure 3). In the 12 months ending December 2008, SSRIs represented 45 percent of drugs used for severe depression; comparatively, SSRIs represented just 41 percent of severe depression drug uses in the 12 months ending November 2009. Over the same time frame, atypical antipsychotics increased from four percent of severe depression drug uses (12 months ending December 2008) to eight percent of drug uses (12 months ending November 2009). Interestingly, we also see new generation antidepressants (e.g., buproprion) increase as a share of severe depression drug uses over the 12 moving annual aggregates analyzed. If anything, it appears that over the last 12 moving segments, the proportion of drug classes used in severe depression have become decreasingly similar to the proportion of drug classes used in moderate depression.

Expert Commentary—Richard Weisler, MD

Clinicians are faced with many clinical decision points as they try to optimally treat their depressed patients. They know from the Sequenced Treatment Alternatives to Relieve Depression (STAR*D) studies,[2,3] clinical trials, and their own clinical experience that remission of symptoms should be their treatment goal. Remission is also known to help decrease the high rates of relapse.[2,4,5]

Suicidal behavior in major depression is a major concern[6] with an approximate 25-fold increased risk, compared to the general population, that can be reduced for many to about an eight-fold risk with appropriate treatment. For example, in one study of 789 patients (605 with major depressive disorder [MDD], 184 with bipolar disorder),[7] suicidal status in 103 suicidal patients decreased 82 percent with antidepressant therapy in the MDD patients and with mood stabilizer therapy in the bipolar patients. Only 0.5 percent of the 656 previously nonsuicidal patients in this same study developed suicidal thoughts without attempts while undergoing antidepressant/mood stabilizing therapy.[7]

Even patients with chronic depression fall under the radar and are often undiagnosed. In the REVAMP study, 66 percent of patients with chronic depression never received an adequate trial of antidepressant therapy.[8] From STAR*D and their experience, clinicians also know that large percentages of their depressed patients will all too quickly drop out of treatment if their clinical symptoms persist and/or side-effect burden is not addressed.[2,9]

As depression increases to the highest of the three severity levels in this analysis, it seems that clinician health providers are making somewhat different pharmacotherapy choices than they do at mild- and moderate-severity levels. The greatest percentage difference in medication class usage occurred in the patients perceived by clinicians to have more severe MDD.
In this data analysis, which falls almost entirely within the year 2009, about eight percent of the patients were judged to be severely depressed by their providers. There was a corresponding increase from 4 to 8 percent for prescriptions of atypical antipsychotics for those deemed to be severely ill over the course of the last year. I suspect that part of this increase in the use of atypical antipsychotics was driven by the November 2007 United States Food and Drug Administration (FDA) approval of aripiprazole as an adjunctive agent for inadequate response to antidepressants based on pivotal trials.[10] Additionally, the scientific literature became available about quetiapine extended release (XR) being effective when used with a variety of antidepressants after an inadequate response to an antidepressant monotherapy trial of six weeks.[11] In late 2009, quetiapine XR received FDA approval for this adjunctive indication in MDD. Some of the studies on quetiapine XR for MDD monotherapy[12,13] were also published during 2009, though the monotherapy indication has not been approved by the FDA at this time nor was the monotherapy or adjunctive quetiapine therapy for MDD promoted during the time frame of this analysis. Olanzapine-fluoxetine combination therapy was also approved by the FDA for treatment-resistant depression.[14]

Based on their own clinical experience, the literature, and the above FDA approvals, I believe many clinicians think that some atypical antipsychotics could also be termed atypical antidepressants as well. Severely ill, depressed patients are more likely to be suicidal, agitated, and/or psychotic, thus may potentially benefit from cotreatment with atypical antipsychotics. In other cases, providers may be unable to determine whether a patient has unipolar or bipolar diagnosis, despite diligent diagnostic efforts. As antidepressants are generally not very effective in bipolar patients and carry a risk of switching the patient into manic or mixed states, providers could be opting to use atypical antipsychotics in such difficult cases. There appears to be less risk of mood switching, for example, if quetiapine is used in bipolar-depressed patients versus paroxetine15 should a clinician miss bipolarity despite trying to rule it out.

Additionally, one can only speculate about the impact over the last year of diminished availability of inpatient services in many states because of the combination of a depressed economy, rising unemployment, severe governmental budget cuts for mental health, and deinstitutionalization. Because of these and other factors, clinicians may have felt pressure to either quickly manage more severely ill, depressed patients with atypical antipsychotics either in short-stay inpatient settings, emergency rooms (while waiting up to days for inpatient beds), or in less restrictive outpatient settings. Clinicians know that atypical antipsychotics work in a matter of days in manic patients, and there is the suggestion, yet to be reconfirmed, that some atypicals work more rapidly to improve depressive symptoms—in MDD, statistically significant separation from placebo occurred as early as Day 4[12]—and before an SNRI.[13] Importantly, clinicians are only using atypical antipsychotics one percent of the time when their patients were judged by them to be mildly ill. Limiting use in mildly depressed patients is only appropriate given the more significant adverse event profile and greater cost of atypical antipsychotics compared to SSRIs and SNRIs.

Anxiety disorders or even just anxiety symptoms are often comorbid with MDD and this association is particularly common in moderate-to-severe MDD.[16] This anxiety association appears to fit with the prescribing data for benzodiazapines. Fawcett et al6 recently reviewed suicidality in MDD and again noted that high levels of anxiety can increase the risk of suicide and recommended reducing anxiety symptoms as a way to reduce suicide risk.6 Certainly benzodiazepines may be used in such depressed cases to help control anxiety symptoms and possibly improve sleep; poor sleep is another known risk factor for suicide. In milder depressed cases, providers use fewer benzodiazepines because of their adverse event profile, which includes memory issues and risk of abuse in all age groups.

The catch-all “other” class usage also increases significantly as depression severity rises, which is not surprising. Some of these “other” category medications include tricyclics, monamine oxidase inhibitors, lithium, and antiepileptic mood stabilizers, all of which may be of value either for their antidepressant properties and/or mood stabilizing properties. Trazodone and occasionally nefazodone are heterocyclic antidepressants that are also used adjunctively at lower doses to help improve sleep and decrease anxiety, two problems that increase with disease severity. Buspirone augmentation was shown to be helpful in some STAR*D subjects in reducing depressive and anxiety symptoms.[17] Delta 9 ligands like pregabalin and gabapentin are not effective as antidepressants, but they can improve anxiety, sleep, and help control pain when present. Lithium has also been shown to significantly reduce suicidal thoughts and acts in both unipolar[18] and bipolar depressed patients.[19] Divalproex and lamotrigine can also be of value in treating depressive symptoms in some patients with MDD. Divalproex is also used for migraine prophylaxis, a common comorbid condition in depressed patients.

Clinicians appropriately appear to be using SSRIs for initial therapy for all severity levels given their safety profile, efficacy, tolerability, and lower cost. SNRIs use clearly increases as patients are classified moderately or severely ill by their providers, and many of these patients may have already failed one or more SSRIs. The increased use of SNRIs may also have been, in part, driven by the use of duloxetine in moderately and severely depressed patients with comorbid pain conditions. Certainly, such SNRI use would be preferable if concomitant opiate use for pain control was able to be decreased or avoided outright in those depressed patients with comorbid pain.

Such a paradigm shift is particularly important at this time given recent Centers for Disease Control and Prevention (CDC) and state mortality studies showing rapidly increasing and alarmingly high death rates of coroner-determined unintentional overdoses of methadone, opiates, and other pain medications. Mixing benzodiazepines with methadone, opiates, or other narcotic pain medications also appears in the data to be associated in post-mortem toxicology screens in unintentional overdose. Tragically, the number of unintentional overdose deaths is thought to have now passed the number of deaths by suicide of all methods in some states, including North Carolina.[20] In the United States in 2006, there were 38,396 deaths from drug-induced causes according the National Center for Health Statistics (note that this number includes a number of drug-related suicides and 26,400 unintentional drug poisoning deaths, among other drug-related causes).[21] There were 33,000 deaths by suicide in the US in 2006 by all causes, with shootings being the most frequently used method by a wide margin.[22]

Some available data suggest this over a decade-long pattern of increasing unintentional drug overdose deaths has even worsened since 2006. Depressed patients with pain tend to take higher doses of narcotics for longer periods of time than their nondepressed counterparts. In a study by Dunn et al,[23] researchers observed a 116/100,000 person-years risk of serious overdoses with opioid prescriptions in patients being treated, with 20 percent of those having a history of depression and five percent a known history of substance abuse. The hazard ratio for serious overdose events rose as the opioid dose increased, and the hazard ratio was 8.39 for patients on any opioid.[21]

Clinicians are always encouraged to take advantage of either web-based or paper-based prescription drug monitoring programs if available when there is any clinical concern about doctor shopping, diversion, or unreported schedule II prescriptions. For general information on the prescription drug monitoring program, visit I recommend that you take a few minutes to complete the simple registration (if the program is available in your state) and begin to take advantage of this great clinical resource. In some of the cases of unintentional overdose death, there is evidence of doctor-shopping for opiates and other controlled drugs. In other cases, diversion is implicated emphasizing the need for every practitioner to limit prescriptions for all narcotics, especially methadone, opiates, and fentanyl to the minimal number of pills and patches necessary to manage only appropriate indications. In other cases, even well-intentioned patients, especially those who are more depressed, poor historians, or cognitively impaired because of their MDD or other reasons, will fail to mention their controlled medications to their providers or take them incorrectly. However, use by both doctor shoppers and poor historians of controlled substance is often accessible when prescription drug monitoring programs are accessed. If concomitant controlled substances are unknown to providers, there is always the potential for drug interactions that can be life threatening. For example, I suspect most providers are unaware, like I was until recently, of the fact that hydrocodone/APAP was the most frequently prescribed generic medication in the United States in 2008 at 121,266,000 generic prespcriptions. Contrast this hydrocodone prescription rate with 29,483,000 for generics of sertraline; 23,271,000 for fluoxetine; 21,846,000 for citalopram HBR; 16,701,000 for trazadone; 15,566,000 for paroxetine; 8,836,000 for ER, XL and SR buproprion; 5,452,000 for mirtazapine; and a mere 2,352,000 for lithium.[24]

We as clinicians need to keep striving to do a better job of recognizing and treating our patients’ MDD fully. This requires us to use therapeutic doses of medications while educating patients and their families about the importance of medication adherence both acutely and during maintenance treatment time frames that reflect the often high risk of recurrence off medication. With such efforts we will no doubt improve the quality of life for our depressed patients and the people they impact, as well as save lives and even money for society.

1. Lenderts S, Kalali A. Treatment of depression: an update on antidepressant monotherapy and combination therapy. Psychiatry (Edgemont) 2009;6(8):15–17.
2. Rush AJ, Trivedi MH, Wisniewski SR, et al. Acute and longer-term outcomes in depressed outpatients requiring one or several treatment steps: a STAR*D report. Am J Psychiatry. 2006;163(11):1905–1917.
3. Gaynes BN, Warden D, Trivedi MH, et al. What did STAR*D teach us? Results from a large-scale, practical, clinical trial for patients with depression. Psychiatr Serv. 2009;60(11):1439–1445.
4. Rush AJ, Kilner J, Fava M, et al. Clinically relevant findings from STAR*D. Psychiatr Ann. 2008;38(3):188–193.
5. Nierenberg AA, Husain MM, Trivedi MH, et al. Residual symptoms after remission of major depressive disorder with citalopram and risk of relapse: a STAR*D report. Psychol Med. 2010;40(1):41–50. Epub 2009 May 22.
6. Fawcett JA, Baldessarini RJ, Coryell WH, et al. Definition and management of suicidality in psychiatric patients. J Clin Psychiatry. 2009;70(10):e38.
7. Tondo L, Lepri B, Baldessarini RJ. Suicidal status during antidepressant treatment in 789 Sardinian patients with major affective disorder. Acta Psychiatr Scand. 2008;118:106–115
8. Kocsis JH, Gelenberg AJ, Rothbaum B, et al. Chronic forms of major depression are still undertreated in the 21st century: systematic assessment of 801 patients presenting for treatment. J Affect Disord. 2008;110(1–2):55–61.
9. Warden D, Rush AJ, Wisniewski SR, et al. What predicts attrition in second step medication treatments for depression?: a STAR*D Report. Int J Neuropsychopharmacol. 2009;12(4):459–473. Epub 2008 Jul 9.
10. Thase ME, Trivedi MH, Nelson JC, et al. Examining the efficacy of adjunctive aripiprazole in major depressive disorder: a pooled analysis of 2 studies. Prim Care Companion J Clin Psychiatry. 2008;10(6):440–447.
11. Bauer M, Pretorius HW, Constant EL, et al. Extended-release quetiapine as adjunct to an antidepressant in patients with major depressive disorder: results of a randomized, placebo-controlled, double-blind study. Clin Psychiatry. 2009;70(4):540–549. Epub 2009 Apr 7.
12. Weisler R, Joyce M, McGill L, et al; Moonstone Study Group. Extended release quetiapine fumarate monotherapy for major depressive disorder: results of a double-blind, randomized, placebo-controlled study. CNS Spectr. 2009;14(6):299–313.
13. Cutler AJ, Montgomery SA, Feifel D, et al. Extended release quetiapine fumarate monotherapy in major depressive disorder: a placebo- and duloxetine-controlled study. J Clin Psychiatry. 2009;70(4):526–539. Epub 2009 Apr 7.
14. Bobo WV, Shelton RC. Olanzapine and fluoxetine combination therapy for treatment-resistant depression: review of efficacy, safety, and study design issues. Neuropsychiatr Dis Treat. 2009;5(3):369–383. Epub 2009 Jul 2.
15. McElroy SL, Weisler RH, Chang W, et al; for the EMBOLDEN II (Trial D1447C00134) Investigators. A double-blind, placebo-controlled study of quetiapine and paroxetine as monotherapy in adults with bipolar depression (EMBOLDEN II). J Clin Psychiatry. 2010 Jan 26. [Epub ahead of print]
16. 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. Epub 2008 Jan 2.
17. Trivedi MH, Fava M, Wisniewski SR, et al. Medication augmentation after the failure of SSRIs for depression. N Engl J Med 2006;354(12):1243–1252.
18. Guzzetta F, Tondo L, Centorrino F, Baldessarini RJ. Lithium treatment reduces suicide risk in recurrent major depressive disorder. J Clin Psychiatry. 2007;68(3):380–383.
19. Tondo L, Baldessarini RJ. Long-term lithium treatment in the prevention of suicidal behavior in bipolar disorder patients. Epidemiol Psichiatr Soc. 2009;18(3):179–183.
20. Katherine J. Harmon, MPH. The Burden of Unintentional Poisonings in North Carolina January 2010. Injury and Violence Prevention: The Burden of Unintentional Poisonings in N.C. N.C. Division of Public Health. 2010.
21. National Vital Statistics Reports, Vol. 57, No. 14, 4/17/09, Table 21, page 93.
22. Underlying Cause-of-Death. http:// Access date: February 15, 2010.
23. Dunn KM, Saunders KW, Rutter CM, et al. Opioid prescriptions for chronic pain and overdose: a cohort study. Ann Intern Med. 2010;152(2):85–92.
24. SDI/Verispan, VONA, Full year 2008. 2008 Top 200 generic drugs by total prescriptions.

Tags: , , , , , ,

Category: Mood Disorders, Past Articles, Psychiatry, Psychopharmacology, Trend Watch

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.