Substance abuse can produce symptoms similar to other psychiatric disorders, thus confusing the diagnostic picture. This paper attempts to elucidate how misdiagnosis in bipolar disorder might be explained by the presence of substance abuse comorbidities. The overlap of symptoms, limited information about symptom onset, and inexperienced clinicians can result in the misinterpretation of symptoms of substance abuse disorders for bipolar disorder. The present study found that the presence of a substance abuse comorbidity, the polarity of last episode (depressed, manic, mixed, not otherwise specified), and the total number of comorbidities affected the reliability of a bipolar disorder diagnosis.
Clinically, the symptoms of Bipolar Disorder (BPD) during manic episodes are quite distinct and relatively easy to identify including elevated mood, rapid speech, agitation, and participation in high-risk behaviors . However, during depressive, mixed, or hypomanic episodes, or when accompanied by psychotic features, BPD shares symptoms with major depressive disorder, schizophrenia, substance abuse disorders, and several personality disorders and can therefore be difficult to distinguish.
It is this overlap in symptoms that makes the diagnostic process challenging [2–4]. In fact, misdiagnosis is common in BPD [5, 6]. For example, Zimmerman and colleagues  examined 700 psychiatric patients who reported that they had been previously diagnosed with BPD. Each person was reevaluated using the Structured Clinical Interview for DSM Disorders (SCID for DSM-IV) . They found that only 43.4% of patients who claimed they had been previously diagnosed with BPD met criteria based on the SCID. This is consistent with other studies [5, 8].
The consequences of an incorrect diagnosis are apparent. Treatment decisions are based on diagnosis and, therefore, inadequate and/or incorrect pharmacological treatments might be applied which lead to unpleasant side effects without the benefit of symptom reduction . These consequences are costly with regard to human suffering and health care service utilization .
In addition to overlapping symptoms, comorbidities such as substance abuse, which occurs in 65% of those diagnosed with BPD , can produce symptoms that muddle the diagnostic picture . Goldberg and colleagues  interviewed patients with substance abuse problems using structured diagnostic interviews during substance-free time periods. They found that only 32.9% of participants previously diagnosed with BPD met full DSM-IV criteria for bipolar I or II disorders. This suggests that prior substance use had contributed to the misdiagnosis. Likewise, Stewart and El-Mallakh  studied patients in a substance-abuse treatment program who had been previously diagnosed with BPD. They found that only 42.9% of participants met criteria for BPD.
Substance abuse disorders are prevalent comorbidities among people with BPD . These disorders may begin as primary disorders or may result from self-medication to reduce or alleviate symptoms of BPD . Commonly abused substances include alcohol, cannabis, cocaine, and stimulants. Not all clinicians are familiar with the signs and symptoms of substance abuse and dependence and could easily mistake them as evidence of a mood disorder  because of their effect on mood and behavior. Substance intoxication or withdrawal symptoms may present as symptoms of mania or depression, respectively, thereby misleading clinicians . Errors can easily occur if clinicians rely too much on global heuristics to diagnose patients rather than thoroughly evaluating all symptoms of a disorder [17, 18].
Studies have shown that utilizing structured diagnostic assessments can improve diagnostic accuracy across psychiatric disorders (e.g., [8, 19]), but less specific guidance has been provided regarding the mistakes made by diagnosticians and how they might be avoided. A better understanding of common sources of error in diagnosis might provide clinicians who do not have access to structured diagnostic methods, such as the SCID, with information that improves the accuracy of their diagnoses. For example, if prior or concurrent substance abuse or dependence is common among patients about whom clinicians disagree on a diagnosis of BPD, then comorbid mood symptoms and substance use might cue the need to invest more time and effort in gathering diagnostic information. Similarly, since structured diagnostic assessments are time intensive and costly, they cannot be provided for all patients. If it was determined that diagnostic error was more likely to occur for those suspected of having BPD along with several comorbidities, then using structured methods might be justified in these cases.
The present study reexamined diagnostic accuracy data from Basco et al.,  to determine if cases in which clinicians disagreed on a diagnosis of BPD could be explained by the presence of substance abuse or dependence, number of comorbidities, or polarity of last episode. Disagreements were cases in which a primary diagnosis of BPD was given by either a treating psychiatrist using routine clinical methods, a nurse using the SCID, or an expert diagnostician using all available data, but was not confirmed by the other sources. It was hypothesized that the presence of substance use disorders would lead to greater diagnostic disagreement because these disorders would present with mood symptoms that could be misinterpreted as a mood disorder. Additionally, the total number of comorbidities identified by the expert or gold standard diagnostician was compared for cases in which diagnostic agreement was achieved between clinicians as compared to those in which there was disagreement. It was hypothesized that a greater number of comorbidities occurring concurrently with BPD would be consistent with more diagnostic discrepancies. Finally, the polarity of the most recent episode was evaluated to determine if discrepancies were more likely to occur when the patient was in a manic, depressed, or mixed state. It was hypothesized that there would be fewer diagnostic discrepancies when patients presented with manic symptoms than with mixed or depressive symptoms as manic symptoms tended to be more striking and stereotypic of the disorder.
to read the full article
This study attempted to explain some of the factors that might interfere with diagnostic accuracy in a community mental health sample of patients with significant mood symptoms. It was found that the presence of substance abuse or dependence, symptoms of mania, and increased number of comorbidities were related to diagnostic disagreements for bipolar disorder. These findings are not surprising. Experienced diagnosticians can attest to the fact that the more complicated the symptom presentation, the more difficult it is to accurately disentangle symptoms, particularly when the same symptoms are common across several disorders.
The draw of substance use is often the alteration in mood. It is this effect that contributes to the confusion in psychiatric evaluations. While structured methods can help organize diagnostic information, clinicians must still make judgments as to the origin of symptoms (i.e., substance related or not). What our findings suggest is that when manic symptoms are present and a substance use history is endorsed, extra caution should be taken in compiling a detailed history of the onset and offset of each. If substance use predates symptom onset that is close in time, a substance-related mood disorder diagnosis is likely. If self-medication with substances of abuse occurs after the onset of mood symptoms, then a mood disorder may be more likely. Comorbidities are best sorted out by use of a life chart  or time line where the onset and offset of BPD symptoms and substance abuse symptoms can be documented. This method was used in the original study to help differential diagnoses. However, when mood and substance use symptoms occur simultaneously, it may not be possible to differentiate the two until the patient discontinues his or her use of substances long enough for its effect on symptoms to dissipate.
As an additional note, as the mental health community prepares for the introduction of the DSM-V (expected in 2013), this is an opportunity to make adjustments to highlight the importance of ruling out substance abuse disorders when diagnosing a patient with BPD, as well as to clarify the differences between both disorders.