Confounding Factors in Treatment-Resistant Depression (Part 2): Comorbidities and Treatment Resistance

The role of subtyping and bipolarity in TRD was discussed in Part 1 of this 2-part article. Here we review a number of the most common confounding factors of TRD but limit our scope to comorbidities that can be directly addressed and treated by psychiatrists.

In psychiatric practice, many depressed patients lack an optimal course of treatment-ie, a “simple” antidepressant medication, with or without psychotherapy-that results in complete symptom remission. The simplest definition of treatment-resistant depression (TRD) is “failure to achieve a response to a medication to any degree short of remission.”1 Addressing the complicating factors that may be responsible for TRD early may improve treatment outcomes.

The role of subtyping and bipolarity in TRD was discussed in Part 1 of this 2-part article2 (See: "Confounding Factors in TRD Part 1: The Role of Subtyping and Bipolarity). Here we review a number of the most common confounding factors of TRD but limit our scope to comorbidities that can be directly addressed and treated by psychiatrists. Subtle maladaptive personality traits or inherent temperamental factors may play a role in sustaining depression in certain individuals. Axis I diagnoses, such as eating disorders, anxiety disorders, and sleep disorders, complicate the treatment of depression unless they are directly addressed. Dissociative disorders may also complicate the treatment of depression.

The role of anxiety and anxiety disorders in TRD

Clinicians have long recognized that treatment of depression may be unsuccessful if accompanying anxiety disorders are not recognized and addressed. Even comorbid subsyndromal anxiety symptoms can complicate the treatment of depression and lead to TRD. Anxiety and depression co-occur frequently, sometimes with one or the other as the primary and preceding diagnosis, but often with a mixture of depressive and anxiety symptoms that is difficult to disentangle historically or clinically.

In their examination of generalized anxiety disorder (GAD) and its association with major depression, Moffitt and colleagues3 followed a group of approximately 1000 New Zealanders from birth to age 32. In this cohort, either MDD or GAD presented as the initial illness. High comorbidity of the two illnesses was demonstrated: 72% of patients with GAD had lifetime MDD, and 48% of MDD patients had lifetime GAD. The extent of the association between depression and anxiety disorders in general is further illustrated by the National Comorbidity Survey, which found that an anxiety disorder was present in 58% of the patients with a lifetime history of depression.4

Anxiety symptoms may make depression harder to treat. Studies have shown that comorbid anxiety in depression is associated with greater severity and longer time to recovery; poorer response to medication; multiple drug trials; more frequent relapses and recurrences; increased social and vocational disability; and perhaps most importantly, an increased rate of suicide, especially in the first year of treatment.5,6

One dissenting study found no association between GAD and TRD; however, this study had a small sample size, and it was a retrospective analysis of failed drug studies.7 Furthermore, this negative study focused narrowly on GAD, while confirmatory studies look more broadly at anxiety disorders and include subsyndromal symptoms, such as psychic anxiety, worry, phobic symptoms, and somatic anxiety.8 Subsyndromal symptoms can be debilitating, even when they do not meet DSM criteria. A comprehensive evaluation for an accompanying DSM-IV-TR anxiety disorder in patients with depression should include screening for agoraphobia, GAD, obsessive-compulsive disorder, panic disorder, PTSD, social phobia, and specific phobia (formerly simple phobia). Although any of these might accompany a specific depression diagnosis, the most commonly reported are social phobia, specific phobia, and PTSD.9

The increased risk of treatment resistance in depressed patients with anxiety symptoms has also been demonstrated in the geriatric population. A recent study of patients older than 70 years with major depression used a self-rating scale for anxiety and found that elevated scores on this scale were associated with poorer response to treatment, increased rate of recurrence in the first 2 years of illness, and increased rate of treatment drop-out.10 Anxiety was viewed in this population as creating a “brittle response” to standard antidepressant treatment. Interestingly, in this study, adjunctive lorazepam (added to paroxetine) did not improve outcomes. The authors suggest that further studies of adjunctive use of psychotherapies and atypical antipsychotic medications in the elderly are needed.

What are the implications of this strong association between anxiety and depression? The most dreaded outcome of depression is suicide, and the destabilizing effect of anxiety in patients who are struggling with suicidal impulses in the setting of major depression increases this risk. Fawcett8 exemplifies this with case reports of 3 patients who ultimately committed suicide and who serve as object lessons for those who care for patients with major depression in both inpatient and outpatient settings.

These cautionary tales illustrate the importance of monitoring for anxiety and agitation in depressed patients. Be particularly aware of escalating anxiety that can accompany the onset of treatment with antidepressants, including SSRIs and TCAs. Because patients with anxiety disorders tend to be very sensitive to the adverse effects of medications, consider beginning antidepressants at lower than usual dosages and slowly titrating the dosage in depressed patients with comorbid anxiety.

The increased suicide risk in the first year of treatment mandates that follow-up psychiatric treatment should be well-designed and reliable. The ongoing suicide risk assessment of patients with depression should give special valence to accompanying anxiety, and its presence may call for increased monitoring or hospitalization.

Pharmacological management of patients with mixed anxiety-depression might include benzodiazepines, but their risks of sedation and dependence must be considered. Although many of the SNRIs and SSRIs have received separate indications for various anxiety disorders, there may be differential responses to these medications. For instance, one study reported better outcomes with the SNRI venlafaxine than with the SSRI fluoxetine for patients with MDD and GAD.11 There may also be a role for other medications in patients with depression accompanied by anxiety, including low-dose atypical antipsychotics, anticonvulsants (divalproex, carbamazepine, gabepentin, lamotrigine, and topiramate have shown effects in smaller studies), and buspirone.6 β-Blockers are useful in blocking peripheral manifestations of anxiety, but caution is advised when treating depressed patients because of the possible depressogenic effect.

Dissociative phenomena and their relationship to TRD

In the evaluation and treatment of patients with depression, dissociation in one of its many forms is easily overlooked as a contributing factor for treatment resistance. In DSM-IV-TR, dissociative disorders are Axis I disorders in which there are significant disruptions in memory, consciousness, identity, or the perception of the environment. There are currently 5 dissociative diagnoses in DSM-IV-TR: dissociative identity disorder (formerly known as multiple personality), dissociative amnesia, dissociative fugue, depersonalization disorder, and dissociative disorder not otherwise specified (NOS).

A thorough psychiatric evaluation includes screening questions for dissociative phenomena, especially when a patient presents with a known history of trauma or abuse. However, patients may not be open about abuse histories and may also fail to spontaneously disclose dissociative experiences, especially if the focus of the evaluation is on mood symptoms. Psychiatrists who treat large numbers of patients with dissociative disorders find that in many, the condition has been undiagnosed for a number of years.

Depression and dissociative disorders both represent final common pathways for patients who have suffered abuse in childhood or have undergone traumatic incidents as adults. For this reason, when a patient has such a history, it may be important to consider using a structured interview for dissociative disorders, such as the Structured Clinical Interview for DSM-IV (SCID-D), or a screening tool, such as the Dissociative Experiences Scale.12

Using these instruments, one inpatient study of 171 consecutively admitted patients to a state hospital in Massachusetts found that 15% of these patients met criteria for a dissociative disorder. In this study, the most common dissociative diagnoses were dissociative identity disorder, dissociative amnesia, and dissociative disorder NOS. Furthermore, and relevant for considerations of TRD, there was a high rate of comorbid depressive illness among the patients with the various dissociative disorders.13

Dissociative phenomena can be mistaken for psychosis. Marcum and colleagues14 reported the case of a 29-year-old man who had failed to respond to a high-dose antidepressant and antipsychotic regimen for what was presumed to be a major depression with psychotic features. This patient had experienced childhood sexual abuse at the hands of an older brother. A sodium amytal interview gave strong evidence for dissociative identity disorder, and the patient subsequently responded to hypnotherapeutic sessions, with a marked decrease in depressive symptoms and normalization of biological markers for depression.

The role of sleep disorders in TRD

Disturbance of sleep is a common complaint among patients with MDD and other Axis I disorders. In the initial treatment approach for a patient who presents with insomnia and depressed mood, a sedative/hypnotic or low-dose trazodone at bedtime is often used to supplement the main full-dose antidepressant. Usually, sleep quality and quantity improve with depression. However, when mood symptoms fail to respond to adequate psychotropic management or psychotherapy, TRD should be suspected. Further workup for comorbid medical issues or review of the differential diagnoses is necessary.

Multiple factors related to sleep may contribute to poor response in depression treatment. Lack of exercise, insufficient time for sleep, shift work, and disruptive drugs and substances (eg, excessive or ill-timed use of alcohol and caffeine) can lead to poor sleep quality, daytime sleepiness, or fatigue.

Perhaps less identified in psychiatry offices are sleep-related breathing problems-most commonly obstructive sleep apnea (OSA). OSA is a sleep disorder characterized by repeated obstruction of the upper airways resulting in apnea (complete) or hypopnea (partial), with corresponding oxygen desaturations and arousals from sleep.15

OSA is common in both clinical and general populations. The estimated prevalence of OSA in Veterans Health Administration patients is 2.91%.13 This compares well with the 2.1% prevalence observed in a general population, cross-sectional telephone survey in Western Europe.14 However, approximately 20% of patients with MDD also have a breathing-related sleep condition.15 The converse relationship yielded a similar concurrence rate. These data confirm that depressed patients are at higher risk for sleep apnea.

While OSA and depressed mood commonly occur together, it is unclear whether depression is a primary process or the manifestation of disordered sleep. Excessive daytime sleepiness, impaired concentration, low libido, memory difficulties, and neuropsychological testing deficits are among the consequences of OSA. These symptoms overlap with MDD. Treatment of both disorders are needed to minimize their effects on the patient.

The diagnosis of OSA requires referral to a sleep specialist who can perform a specialized physical examination and who has access to a sleep laboratory for polysomnography. For financial and logistical reasons, it is not feasible to send every depressed patient for full-night polysomnography. While limited studies (nonneurophysiological recording of sleep and oxygenation patterns) have been used, polysomnography provides more definitive exclusion or confirmation of OSA. One study evaluated an optimal screening process based on patient symptoms that would initiate a more formal sleep evaluation.16 When severe snoring, observed apnea during sleep, or excessive daytime sleepiness is endorsed, referrals for polysomnography will yield sleep apnea rates up to 60%.

Eating disorders and TRD

Depression and eating disorders are known to be highly comorbid. In a critical review of the literature on the comorbidity of depressive disorders and eating disorders, Godart and colleagues17 determined that comorbidity rates for bulimia nervosa and depressive disorders range from 24% to 90%; for anorexia nervosa, comorbidity rates range from 31% to 89%. In fact, patients with bulimia not only had higher lifetime prevalence rates of depression than controls, but they also had higher rates than patients with bipolar disorder or schizophrenia.

Presnell and colleagues18 found that bulimic symptoms were predictive of future depressive symptoms. Persons with eating disorders had significantly higher scores on the Beck Depression Inventory than controls and those who have chronic pain.19 The severity of disordered eating symptoms and that of depressive symptoms is highly correlated. In addition, as the eating disorder symptoms improve and resolve, so do the mood symptoms.17 In a prospective study, Angst and colleagues20 followed patients over 20 years to characterize differences between episodic depression and long-term depression. Binge eating was more common in the group with long-term depression.

However, in anorexia nervosa, the direct effects of starvation may confuse the diagnosis of depression. Persons with anorexia often have depression-like symptoms, including fatigue, poor sleep, lack of interest in sexual activity, and flat mood.21 With chronic disordered eating, feelings of helplessness and hopelessness also often become prominent. In a study of 70 patients, Coughlin and colleagues19 showed that psychological complications related to disordered eating worsened the longer the starvation or binge-purge behavior lasted. Primary efforts toward nutritional restoration and development of healthy eating patterns should play a critical role in these patients. In 4 placebo-controlled trials, there was no evidence that antidepressants had a positive effect on weight gain, the eating disorder, or associated psychopathology.22

TRD is difficult to affirm in patients with low body weight. Correction of the eating disorder, especially in malnourished patients, may allow better attribution of depressive symptoms to an underlying mood disorder. In addition, establishing the timeline of a depressive disorder in relation to an eating disorder through detailed history and collateral reports may help justify continued antidepressant use in the midst of an eating disorder. In normal-weight patients with disordered eating, antidepressants are likely to be more effective.


Factors that contribute to TRD are not limited to those described in this article. Substance and alcohol abuse or dependence often elude an initial evaluation and undermine even the most aggressive antidepressant strategies. In addition, interactions with prescribed medications-not just drugs of abuse-can alter pharmacodynamics and/or pharmacokinetics, which necessitates closer monitoring and possible dosage adjustments.23

On the medical side, several conditions besides sleep apnea are associated with depression.24 Vigilant efforts to detect and treat anemia, hypothyroidism, heart disease, or even occult cancer remove barriers to depression treatment and improve a patient’s overall health. In addition, the issue of the relationship of chronic pain to depression is a complex one deserving of attention.

Editor’s Note: Our Category 1 CME articles are on hiatus for the summer. In the meantime, we invite you to test yourself: read the article, take the posttest on the next page, and then check the answer key on the last page of this article for the correct answers.

1. Findings from the National Comorbidity Survey indicate that ____ of patients with a lifetime history of depression have a comorbid anxiety disorder.

A. 23%
B. 27%
C. 43%
D. 58%

2. Which of the following is associated with comorbid anxiety in depression?

A. More severe episodes of psychosis
B. Higher degree of neurocognitive problems
C. Greater risk of suicide
D. All of the above

3. It is clear that depression is a manifestation of disordered sleep as a result of obstructive sleep apnea.

A. True
B. False

4. The Mini-Mental State Examination can be used to determine the co-occurrence of depression and dissociative disorder.

A. True
B. False

5. Which of the following statements is true concerning the relationship of depression to dissociative disorders?

A. Neither one can be associated with history of childhood trauma.
B. Failure to diagnose dissociative disorder may lead to unsuccessful treatment of depression.
C. In a Massachusetts inpatient study, fewer than 5% of the patients had some form of dissociative disorder.
D. Depression and dissociative disorders have low comorbidity.

6. Which of the following is used as first-line therapy for a patient who presents with co-occurring depression and insomnia?

A. An SSRI or SNRI supplemented by a sedative/hypnotic or low-dose trazodone at bedtime.
B. Several glasses of wine in the evening, along with cognitive-behavioral therapy
C. A benzodiazepine as monotherapy
D. None of the above

7. What percentage of patients with MDD also have a breathing-related sleep condition?

A. 5%
B. 20%
C. 50%
D. 75%

8. Clinicians who treat elderly patients with depression and associated anxiety should be alert for all but one of the following:

A. Poorer response to treatment
B. High rate of treatment drop-out
C. Increased rates of recurrence of illness
D. Increased rates of delusional thinking

9. Comorbidity rates for anorexia nervosa and depressive disorders range from

A. 10% to 26%
B. 24% to 90%
C. 31% to 89%
D. 30% to 93%

10. Treating the malnourishment related to an eating disorder first may allow better assessment and treatment of depressive symptoms.

A. True
B. False

Answer key to the Mine Your Mind posttest:

1, D;   2, C;   3, B;   4, B;   5, B;   6, A;   7, B;   8, D;   9, C;   10, A.



1. Barbee JG. Treatment-resistant depression: advances in assessment. Psychiatr Times. 2008;25(10):20-22.

2. Casher MI, Gih D, Bess JD, Agarwala P. Confounding factors in TRD (Part 1): the role of subtyping and bipolarity. Psychiatr Times. 2012;29(7):36-38.

3. Moffitt TE, Harrington H, Caspi A, et al. Depression and generalized anxiety disorder: cumulative and sequential comorbidity in a birth cohort followed prospectively to age 32 years. Arch Gen Psychiatry. 2007;64:651-660.

4. Rapaport MH. Prevalence, recognition, and treatment of comorbid depression and anxiety. J Clin Psychiatry. 2001;62(suppl 24):6-10.

5. Kornstein SG, Schneider RK. Clinical features of treatment-resistant depression. J Clin Psychiatry. 2001;62(suppl 16):18-25.

6. Pollack MH. Comorbid anxiety and depression. J Clin Psychiatry. 2005;66(suppl 8):22-29.

7. Petersen T, Gordon JA, Kant A, et al. Treatment resistant depression and axis I co-morbidity. Psychol Med. 2001;31:1223-1229.

8. Fawcett J. The detection and consequences of anxiety in clinical depression. J Clin Psychiatry. 1997;58(suppl 8):35-40.

9. Kessler RC, Nelson CB, McGonagle KA, et al. Comorbidity of DSM-III-R major depressive disorder in the general population: results from the US National Comorbidity Survey. Br J Psychiatry Suppl. 1996 Jun;(30):17-30.

10. Andreescu C, Lenze EJ, Dew MA, et al. Effect of comorbid anxiety on treatment response and relapse risk in late-life depression: controlled study. Br J Psychiatry. 2007;190:344-349.

11. Silverstone PH, Salinas E. Efficacy of venlafaxine extended release in patients with major depressive disorder and comorbid generalized anxiety disorder. J Clin Psychiatry. 2001;62:523-529.

12. Bernstein EM, Putnam FW. Development, reliability, and validity of a dissociation scale. J Nerv Ment Dis. 1986;174:727-735.

13. Saxe GN, van der Kolk BA, Berkowitz R, et al. Dissociative disorders in psychiatric inpatients. Am J Psychiatry. 1993;150:1037-1042.

14. Marcum JM, Wright K, Bissell WG. Chance discovery of multiple personality disorder in a depressed patient by amobarbital interview. J Nerv Ment Dis. 1986;174:489-492.

15. Hauri PJ, Sateia M, eds. The International Classification of Sleep Disorders. Diagnostic and Coding Manual. 2nd ed. Westchester, IL: American Academy of Sleep Medicine; 2005.

16. Ohayon MM. The effects of breathing-related sleep disorders on mood disturbances in the general population. J Clin Psychiatry. 2003;64:1195-1200.

17. Godart NT, Perdereau F, Rein Z, et al. Comorbidity studies of eating disorders and mood disorders: critical review of the literature. J Affect Disord. 2007;97:37-49.

18. Presnell K, Stice E, Seidel A, Madeley MC. Depression and eating pathology: prospective reciprocal relations in adolescents. Clin Psychol Psychother. 2009;16:357-365.

19. Coughlin JW, Edwards R, Buenaver L, et al. Pain, catastrophizing, and depressive symptomatology in eating disorders. Clin J Pain. 2008;24:406-414.

20. Angst J, Gamma A, Rössler W, et al. Long-term depression versus episodic major depression: results from the prospective Zurich study of a community sample. J Affect Disord. 2009;115:112-121.

21. Woodside BD, Staab R. Management of psychiatric comorbidity in anorexia nervosa and bulimia nervosa. CNS Drugs. 2006;20:655-663.

22. Claudino AM, Hay P, Lima MS, et al. Antidepressants for anorexia nervosa. Cochrane Database Syst Rev. 2006;(1):CD004365.

23. Schellander R, Donnerer J. Antidepressants: clinically relevant drug interactions to be considered. Pharmacology. 2010;86:203-215.

24. Benton T, Staab J, Evans DL. Medical co-morbidity in depressive disorders. Ann Clin Psychiatry. 2007;19:289-303.