The comorbidity of major depression and substance use disorder (SUD), which includes both alcohol and drugs, remains very common in medical and psychiatric settings (Fig. 19-1).^1–3 Both major depression and SUD are among of the most common psychiatric disorders, each leading to significant morbidity and mortality and placing a heavy burden on the health care system.⁴ In clinical settings, however, SUDs are often underdiagnosed or missed entirely. Even if the SUD is correctly identified, treatment is often inadequate or not offered at all.^5,6 Studies have consistently noted that patients with both depression and SUD present with worse symptoms and have poorer outcomes, including a higher incidence of suicide attempts, than patients with either disorder alone (Fig. 19-2).^7,8 Research on the dually diagnosed population is still in its early stages, and significant gaps still remain.⁹ Research on the comorbidity between dual diagnosis and medical illness is even less frequently studied. Nevertheless, when approaching a depressed patient with comorbid SUD, clinicians need to be aware of the specific issues unique to this population.

The comorbidity between SUD and psychiatric disorders has been documented in all major epidemiological studies conducted to date. The Epidemiological Catchment Area (ECA) study reported a lifetime prevalence of 5.9% for major depression and 16.7% for any SUD in the general population.³ Individuals with a diagnosis of major depression or other affective disorders showed elevated rates for SUD—16.5% for alcohol use disorder and 18% for drug use disorder. Conversely, individuals with any SUD were 1.9 times and 4.7 times more likely to have a diagnosis of depression or other affective disorders, respectively.

The National Epidemiological Survey of Alcohol and Related Conditions study (NESARC)¹⁰ utilized face-to-face surveys with a broad sample of noninstitutionalized US citizens. It provided significant advantages over prior surveys by using the DSM criteria, differentiating primary mood disorders from substance-induced mood disorders, and including both substance abuse and dependence. Overall, 9.4% of the US population met DSM-IV criteria for any SUD within the past year. Among respondents meeting criteria for any SUD in the past year, 19.7% also met criteria for at least one independent mood disorder in the past year, and 17.7% met criteria for at least one anxiety disorder in the past year (Box 19-1; Fig. 19-2). For those with drug use disorder, 31.8% of respondents also had an independent mood disorder, compared to 12.3% for those with any alcohol use disorder. Similarly, 25.4% of all respondents with any drug use disorder also met criteria for an independent anxiety disorder, as compared to 17.1% for those with any alcohol use disorder. Overall, the most common comorbid diagnoses for respondents with any SUD were major depressive disorder (MDD) at 14.5% and specific phobia at 10.5%.¹⁰

Comorbidity with mood disorders was lower in those with substance disorder than in those with substance dependence. For instance, 40.0% of those with drug dependence had any mood disorder in the past year, compared to 15.8% of those with drug abuse. Of those with alcohol dependence, 20.5% also had any mood disorder in the past year, compared to 8.2% of those with alcohol abuse only.¹⁰

When the index disorder is any mood disorder, study results indicated that 20.0% also met criteria for any SUD in the past year. When the index disorder is any anxiety disorder, 15.0% also met criteria for any SUD in the past year. Respondents endorsing any mania or hypomania had the highest rates of comorbid SUD, reporting 27.9% and 26.6%, respectively. Overall, for those respondents with an alcohol use disorder, the odds ratio of having any mood disorder was 2.6. When separated by race, the odds ratio are as follows: whites 2.4, blacks 3.3, Native Americans 2.8, Asians 3.7, and Hispanics 2.6.¹⁰

The prevalence of depression is significantly higher in patients seeking treatment for SUD than in the general population. The lifetime prevalence of major depression was 42.2% in treatment-seeking alcohol-dependent patients and 43.7% in patients enrolled in substance disorder treatment programs.¹¹ In the NESARC study,⁹ among individuals with any mood disorder seeking treatment, about 20% also met criteria for any SUD. These findings demonstrate the very high rates of SUD and mood disorder co-occurring, particularly in the treatment-seeking population.

The relationship between depression and SUD has been debated for some time. A frequently cited link is that depression is a risk factor for SUD. In this model, individuals experiencing depression or anxiety attempt to “self-medicate” and control distressing symptoms by using substances (Box 19-2).¹² For example, a patient with anxiety might choose to consume alcohol for its anxiolytic effects. Up to a quarter of individuals in the general population with a mood disorder admit to using substances to relieve symptoms,¹³ and the majority of them report self-medicating using alcohol only. Research in the use of antidepressants for those with co-occurring alcohol dependence and depression may support this theory. In a trial of depressed patients with comorbid alcohol dependence, those receiving fluoxetine showed greater improvement in both depression and drinking outcomes than those on placebo.¹⁴ Nevertheless, the self-medication hypothesis has not been empirically tested, and research has failed to identify specific mood changes that are associated with specific drugs.¹⁵

Another theory proposes that chronic use of substances leads to mood or anxiety symptoms.¹⁶ Acute intoxication as well as withdrawal can both be associated with significant affective, psychotic, or anxiety symptoms.¹⁷ However, such symptoms often remit rapidly after a period of abstinence.^18,19 Indeed, in order to diagnose a substance-induced mood disorder, as opposed to an independent mood disorder, the DSM suggests that the mood or anxiety symptoms remit within approximately 4 weeks after cessation of substance use. Nevertheless, even though substance use clearly plays a role in producing depressive symptoms, the prevalence of substance-induced mood disorders is much lower than that of independent mood disorders in those with SUD.¹⁰

Another theory posits that mood disorders and SUDs are related through a shared genetic vulnerability, since both disorders have a high heritability.^20,21 Based on twin studies, shared environmental factors likely contribute only about 1% of the variance in the risk for alcohol dependence, while genetic factors contribute over 40% of the variance.²² However, the evidence remains inconsistent.. The risk of depression is three times higher in the offspring of parents with histories of major depression than it is in the offspring of parents without major depression.²³ The same study revealed a nonsignificant trend for greater incidence of substance dependence in the offspring of parents with histories of depression compared to offspring of parents without depression. In another long-term study, the offspring of depressed parents reported more symptoms of depression than the offspring on nondepressed parents,²⁴ but the incidence of alcohol or tobacco use did not differ in these two groups of offspring.²⁵

Research examining the pathophysiology of SUDs has strongly implicated the involvement of the mesolimbic reward pathway.²⁶ In response to both substances and natural rewards (i.e., food, sex, social interactions), an increase in dopamine in the nucleus accumbens appears to function as a prediction–error signal that helps the organism efficiently obtain rewards.²⁷ Although this system helps to shape survival behaviors related to procuring food and mates, substances appear to hijack this system. Even though research on the pathophysiology of depression has largely focused on the role of hippocampal and frontal cortical regions, there is now increasing recognition that the mesolimbic reward pathways may also play a significant role in the manifestation of depression.²⁸ While it remains unclear what role the reward pathway plays in depression, this line of research points to the possibility of a shared therapeutic target for both depression and SUDs.

Thus, the relationship between depression and substance use is a complex interaction between genetics and environment, and no single theory can capture this relationship sufficiently.

The current DSM 5 diagnostic criteria for SUD are listed in Table 19-1. It is important to recognize that tolerance and withdrawal are elements of “physiologic dependence.” Although physiologic dependence may be prominent in many patients with SUD, physiologic dependence alone is neither sufficient nor necessary for the diagnosis. Indeed, physiologic dependence can be observed from using medications that have no abuse liability. For example, a patient taking pain medications for chronic pain may report the development of tolerance and withdrawal, but may not manifest any behaviors of SUD. The central feature of the SUD diagnosis is the loss of control over the use of the substance, that is, inability to cut down, using more than intended, using despite knowledge of harm. Therefore, it is the inability to regulate the consumption of the substance that is the core feature of a SUD.

Comorbid SUD is found in about a fifth of those patients seeking treatment for a mood disorder, and depression is found in about half of those patients seeking treatment for SUD.¹⁰ The symptoms of major depression are likely to overlap with the symptoms of SUD. For example, a patient with alcohol dependence may present with chronic sleep problems, loss of appetite, difficulty concentrating, and depressed mood, all driven largely by the effects of alcohol. A patient with opioid dependence may present with anhedonia, weight loss, and psychomotor retardation. A cocaine-dependent patient may present with psychomotor agitation, perceptual disturbances, sleep and appetite disturbances, and poor concentration. Clinicians should routinely screen for substance use in patients seeking treatment for depression, and should assess not just current symptoms but the longitudinal history and the relationship of these symptoms to the use of substances.

In the Sequenced Treatment Alternatives to Relieve Depression (STAR*D) trial, 28% of depressed patients reported symptoms consistent with current SUD.⁸ When compared to those without a current SUD, these individuals were more likely to be male, be divorced or never married, have a younger age of onset of depression, and to have had more suicide attempts (Box 19-3). Those with comorbid SUD also reported higher levels of hypersomnia, anxious mood, and suicidal ideation.

Patients frequently present in clinical settings with depressed mood in the context of acute intoxication with a variety of substances, notably alcohol. Although the acute mood effects of alcohol include euphoria and anxiolysis, alcohol is a CNS depressant and acts to inhibit executive functions, leading to agitation, irritability, and anger. Paranoia, and suicidal ideation may be present in severe situations.²⁴ Given that a large proportion of suicides occur in the context of substance use,²⁹ patients endorsing suicidal thoughts require close observation even when acutely intoxicated. Acute intoxication with alcohol can worsen depression while intoxication from stimulants, hallucinogens, or dissociatives can cause anxiety and paranoia. Acute withdrawal from alcohol, benzodiazepines, cocaine, and opioids may present with depressed, anxious, or irritable mood. Affective symptoms are particularly common and often very prominent in acute withdrawal, and so this constellation of symptoms alone may not aid in differentiating the effects of acute withdrawal from a substance-induced mood disorder or an independent mood disorder. Nevertheless, many patients will experience a rapid improvement in mood following detoxification.¹⁹

In order to diagnose a patient with a substance-induced mood disorder, the DSM 5 requires the remission of mood symptoms within a substantial period of time (e.g., 4 weeks) following cessation of substance use. However, in clinical practice, continuous abstinence for 4 weeks is often difficult to maintain regardless of the presence of mood symptoms. While the mood symptoms are likely to lessen after cessation of substance use, relapse is a common occurrence early in treatment, and the persisting mood symptoms can hinder the patient’s ability to remain abstinent. Thus, distinguishing an independent mood disorder from a substance-induced mood disorder at the time of detoxification is difficult and requires careful review of the patient’s history (Box 19-4).³⁰

Several medical illnesses commonly associated with SUD can also present with depression. Depression may be a presenting complaint for nutritional deficiencies in alcohol-dependent patients. Pellagra, caused by niacin deficiency, is a disease characterized by dermatitis, diarrhea, and dementia. However, early in the course of illness, depression may be prominent.³¹ Wernicke–Korsakoff syndrome, a potential complication of thiamine deficiency, may be complicated by depression.³² Depression can also be a prominent complaint in chronic pain patients (see Chapter 17) who are using opioids for nonmedical reasons. Depressed chronic pain patients are less likely to experience analgesia from opioids, and are more likely to misuse their pain medications.³³ Finally, hepatitis C is an extremely common diagnosis in opioid-dependent patients, with reports indicating prevalence rates as high as 66%.³⁴ Although newer treatments have now become available, depression is a well-described complication of interferon treatment.³⁵

Finally, regardless of the history, given the high prevalence of depression in the substance using population, the index of suspicion for an independent mood disorder should remain high, especially in those seeking treatment for SUD.

The natural history of major depression with comorbid SUD tends to follow a characteristic course: depression outcomes are worse in those who also use substances even though such patients are more likely to receive more intensive treatment.³⁶ Depressed patients with comorbid SUD are also more likely to attempt suicide, and experience greater social and personal dysfunction.³⁷ In a study of fluoxetine treatment for depression, the degree of alcohol consumption at baseline was a strong predictor of worse treatment outcomes.³⁸ In a study of veterans with depression and SUD, greater frequency of substance use predicted worse depressive symptoms at every time point during the follow-up period.³⁹ In the STAR*D study, the individuals with major depression and comorbid SUD reported more severe depression, longer duration of depression, more anxiety, and more atypical and melancholic features than those without comorbid SUD.^40,41 Interestingly, the comorbidity with alcohol dependence alone, or drug dependence alone, did not significantly alter the rates of response or time to achieve response.⁴² However, those with both alcohol and drug use had significantly lower rates of remission and an increased time to remission from both depression and substance use. Furthermore, participants with a comorbid SUD experienced more psychiatric adverse events (e.g., suicidal ideation, suicide attempts) and psychiatric hospitalizations.