Underdiagnosing and Overdiagnosing Psychiatric Comorbidities

Diagnostic assessment of psychiatric disorders and their comorbidities is a challenge for many clinicians. In emergency settings, there is no time to conduct lengthy interviews, and collateral information is often unavailable. In acute treatment settings, patients may be too symptomatic to provide accurate information or to recall important details that would aid clinicians in deriving a valid diagnosis. In busy clinical practices, clinicians must see a high volume of patients and may not have the luxury of a psychologist on-site to conduct psychological testing or of a psychiatric nurse to administer a structured assessment. Moreover, there are no practical laboratory tests to facilitate the psychiatric diagnostic process.

In addition, acutely ill patients who present for psychiatric care often lack the cognitive ability to explain their symptoms or to describe their symptom progression, and they may not know what historical information might be critical to their doctors’ diagnostic assessments. Therefore, to meet the immediate needs of patients, psychiatrists must use their clinical judgment to make quick assessments, and they must implement treatment with the information available.

Unfortunately, accuracy and thoroughness are sometimes sacrificed for speed. Studies of diagnostic accuracy have demonstrated weaknesses when general clinical interviews are used instead of structured diagnostic methods, which are more common in research clinics.^1-3

There are 3 common problems in diagnosing comorbidities:

• Missed diagnosis. Clinicians may miss coexisting problems that require additional treatment or that interfere with treatment of the primary problem. This usually occurs when clinicians are unable or do not attempt to access sufficient information about the history of illness or do not thoroughly assess concurrent symptoms (eg, not inquiring about symptoms of alcohol or substance abuse disorder in a patient who presents with a mood disorder).

• Overdiagnosis. Clinicians may jump to conclusions when a patient presents with a few symptoms that might be indicative of other disorders (eg, assuming the presence of borderline personality disorder in a patient who engages in self-injury or who has outbursts of anger).

• Misinterpretation. Clinicians may misinterpret a symptom and establish an incorrect comorbid diagnosis (eg, observing the theatrical and dramatic behavior of a patient with histrionic personality disorder and mistaking it for hypomania).

Common diagnostic oversightsSubstance-related disorders. Clinicians may not ask about alcohol or substance use and abuse when a patient does not fit the stereotype of an addict (eg, those who seem too attractive, intelligent, accomplished, or “well put together” to have a problem with alcohol or drugs).

Age can also hinder the diagnosis of substance-related disorders. Substance use disorder is less likely to be diagnosed in the elderly than in younger adults during an office visit.⁴ This is, in part, because of the overlap between symptoms of substance abuse and symptoms of other physical and mental conditions (eg, stroke, malnutrition, reaction to prescription drugs, depression, dementia).^4,5Bipolar II disorder. The failure to inquire about a history of hypomania in a patient with major depression can lead to diagnostic and treatment errors.^6-8 Kim and associates⁸ evaluated patients who had received a diagnosis of major depression but who had not received a prior diagnosis of bipolar disorder. Using the Mood Disorder Questionnaire, 53% of this sample was found to have bipolar disorder; 30% had bipolar II disorder.

Schizo-affective disorder or psychotic depression. Clinicians may not assess whether auditory or visual hallucinations are present in patients with major depression who are quiet or pensive, have long response latencies, or are well-groomed and appropriately dressed.

Major depressive disorder (MDD). MDD is commonly overlooked when the clinician assumes that vegetative symptoms of major depression are manifestations of a primary physical illness, such as chronic pain, diabetes, or cancer⁹; dismisses MDD in a patient who has recently suffered a significant loss and does not inquire about symptoms that differentiate MDD from bereavement (ie, excessive guilt, passive suicidal ideation, worthlessness, psychomotor retardation, prolonged functional impairment, or psychosis)¹⁰; or misses concurrent depression in a patient with a primary psychotic disorder.

Posttraumatic stress disorder (PTSD). PTSD is underrecognized in patients with panic disorder, certain phobias, schizophrenia, major depression, alcohol abuse, and antisocial and borderline personality disorders.^11-13

Personality disorders. Because personality disorders share some symptoms with Axis I disorders, they can be easily overlooked (Table 1).¹⁴ Problematic personality traits can interfere with treatment adherence, stress management, and psychosocial functioning.

Cognitive dysfunction. Cognitive decline can interfere with a patient’s ability to follow treatment guidelines (eg, correct medication dosing and consistent appointment attendance) or to manage his or her life and engage in self-care. It may be difficult to identify cognitive dysfunction when marked symptoms of psychosis, mood disorders, or anxiety disorders are present. The repetitive nature of obsessive-compulsive disorder, for example, may mask memory deficits.¹⁵ Forgetfulness in taking daily medication may be mislabeled as resistance. Disorganization and poor planning may be mistaken for impulsivity or an attention deficit consistent with attention-deficit/hyperactivity disorder or bipolar disorder.

Low intelligence is another commonly overlooked cognitive deficit. The patient who appears to comprehend treatment directives but who declines to ask questions when provided the opportunity and who ultimately fails to follow instructions may have impaired intellectual functioning or a verbal comprehension level that is lower than the instructional level of care providers.

Table 2 summarizes additional diagnostic oversights in psychiatric comorbidities among children and adults.

Solutions

The key to reducing diagnostic errors of commission or omission is to gather as much information as possible before finalizing the diagnosis. Crucial elements include:

• A complete, longitudinal history of presenting problems.

• Use of a structured diagnostic method to facilitate thoroughness of evaluation.

• Collateral information from other health care providers, family members, and friends.

• A review of diagnostic criteria rather than reliance on one’s memory of DSM-IV criteria.

• Assessment of factors that affect diagnostic accuracy.

Kashner and colleagues¹⁶ found the Structured Clinical Interview for DSM-IV (SCID) to be more effective than a previous, less-structured protocol in accurately diagnosing conditions in community mental health patients. This tool resulted in a change in diagnosis for 20% of patients in whom it was used; in contrast, a change in diagnosis was made for only 2% of patients in a control group. The SCID also captured more comorbid diagnoses (51% for the SCID group vs 5% for the control group), including substance use disorder, anxiety, and eating disorders.

A summary of commonly used diagnostic tools that might facilitate accurate identification of primary psychiatric disorders and comorbidities follows. While these tools facilitate the diagnostic process, they are used primarily to provide additional information to support diagnoses, to aid evaluations by focusing the attention of providers, and to help the interviewer be more thorough in gathering diagnostic information.

Axis I

Many professionals have begun to standardize their assessment practices by replacing unstructured, open-ended interviews with more systematic methods of eliciting psychopathological criteria. As a result, they have significantly increased diagnostic accuracy.^2,17,18 For example, Miller and colleagues² compared the efficacy of structured (eg, SCID–Clinical Version, Computer Assisted Diagnostic Interview) and unstructured, traditional diagnostic assessments and found that structured methods were significantly superior.

Others have also demonstrated similar results.^1,17,19 Basco and colleagues¹ demonstrated the superiority of structured interviews in identifying important psychiatric comorbidities. Compared with unstructured methods used by clinicians in a community mental health setting, the SCID identified more than twice the number of active comorbid Axis I disorders. Unfortunately, structured interviews are time consuming and require some training to administer.

An alternative to structured interviews is Zimmerman’s Psychiatric Diagnostic Screening Questionnaire (PDSQ).²⁰ This 125-item self-report measure screens for 13 adult psychiatric disorders. It was designed to facilitate diagnostic assessment in outpatient mental health and primary care settings by providing clinicians with information on disorders that are most likely to be present. Because it focuses the interviewer’s diagnostic questions on commonly diagnosed disorders, it increases efficiency. The internal consistency of the 13 scales is high, with a mean a coefficient of 0.85.²¹ Except for the psychosis scale, sensitivity levels range from 85% to 92%. Specificity values range from a low of 50% for generalized anxiety disorder to a high of 89% for bulimia nervosa.²² There are no subscales for mania, hypomania, or dysthymia in the PDSQ.

Axis II

The reliability of Axis II diagnoses based on a single interview has been problematic. Symptoms of personality disorders can be subtle, and it may take many different interactions with a patient before a clinician can confidently make a diagnosis. Structured interviews are available to help improve the accuracy of Axis II diagnoses by guiding the interviewer in gathering sufficient information; however, they are lengthy and impractical in many clinical settings.

The Personality Disorder Interview-IV is a 2-hour semistructured interview in which an overall dimensional profile is created by rank-ordering possible diagnoses.²³ Interview questions are clustered by personality disorder and follow-up questions are provided. Two other examples are the Personality Disorder Examination²⁴ (PDE) and the International PDE²⁵; the latter was developed from the PDE to diagnose disorders in the DSM-IV and the International Classification of Diseases-10.²⁶ Another example is the SCID for Personality Disorders-II, a 120-item self-report questionnaire that is followed by a semistructured interview to examine more closely any criteria that were endorsed.²⁷

The Coolidge Axis II Inventory²⁸ is a Likert-type self-report scale that is also used to assess personality disorders; however, caution should be taken in the interpretation of these scales. Personality traits that are problematic for the patient are not necessarily evidence of a DSM-IV disorder. Patient traits that present a challenge for the clinician may reflect a problem with the therapeutic alliance.

Substance abuse

Alcohol and illicit drug use can be problematic, even at subclinical levels. While substance abuse and dependence can be assessed with the Axis I diagnostic tools described previously, a more focused measure is the Alcohol Dependence Scale.²⁹ The scale contains 25 items that measure withdrawal, control, awareness of compulsion to drink, tolerance, and addictive behavior. It can be administered as an interview or can be self-administered by computer, and it takes 5 to 10 minutes to complete. It has shown high internal consistency, with Cronbach’s a values ranging from 0.85 to 0.99. Cutoff scores are determined according to the purpose of treatment and typically range from 3 to 8.²⁹ A lower cutoff score might suggest levels of substance use that could potentially interfere with the primary Axis I problem under treatment.

Summary

There is no substitute for being thorough in conducting a diagnostic evaluation. This includes taking the time to gather information from the patient and significant others, going over prior medical records, and/or observing the patient over time and updating the diagnosis if appropriate. Keep in mind the importance of probing for comorbidities, of not jumping to conclusions about their presence when a patient presents with a few striking symptoms, and of the need to interpret symptoms correctly. Clinicians may be forced to draw quick diagnostic conclusions with limited information in busy practice settings or when patients are acutely ill. In these cases, it is important to follow up after patients have been stabilized, to reevaluate diagnoses when patients seem to be nonresponsive to treatment, and to consider the possibility that comorbid problems may be interfering with treatment outcome.

Even when the clinician is familiar with DSM-IV criteria, it is easy to forget the details over time. Most diagnostic tools provide a reminder of symptoms that are common to each category of diagnosis. If it is impractical to use such tools, it is prudent to review DSM-IV criteria when new patients present for treatment. This allows clinicians to recalibrate their assessments, refresh their memories, and reduce drift over time of how diagnoses are derived. By taking time to be thorough in diagnosing primary and comorbid psychiatric disorders, the clinician can facilitate treatment selection and treatment outcome.

Disclosures:

The authors report no conflicts of interest regardingthe subject matter of this article.

References:

References

1. Ramirez Basco M, Bostic JQ, Davies D, et al. Methods to improve diagnostic accuracy in a community mental health setting. Am J Psychiatry. 2000;157: 1599-1605.
2. Miller PR, Dasher R, Collins R, et al. Inpatient diagnostic assessments: 1. Accuracy of structured vs. unstructured interviews. Psychiatry Res. 2001;105:255-264.
3. Pfeiffer AM, Whelan JP, Martin JM. Decision-making bias in psychotherapy: Effects of hypothesis source and accountability. J Couns Psychol. 2000;47: 429-436.
4. Banta JE, Montgomery S. Substance abuse and dependence treatment in outpatient physician offices, 1997-2004. Am J Drug Alcohol Abuse. 2007;33:583-593.
5. Stelle CD, Scott JP. Alcohol abuse by older family members: a family systems analysis of assessment and intervention. Alcohol Treat Q. 2007;25:43-63.
6. Ghaemi SN, Boiman EE, Goodwin FK. Diagnosing bipolar disorder and the effect of antidepressants: a naturalistic study. J Clin Psychiatry. 2000;61:804-808.
7. Ghaemi SN, Sachs GS, Chiou AM, et al. Is bipolar disorder still underdiagnosed? Are antidepressants overutilized. J Affect Disord. 1999;52:135-144.
8. Kim B, Wang HR, Son JI, et al. Bipolarity in depressive patients without histories of diagnosis of bipolar disorder and the use of the Mood Disorder Questionnaire for detecting bipolarity. Compr Psychiatry. 2008; 49:469-475.
9. Kahn LS, Fox CH, McIntyre RS, et al. Assessing the prevalence of depression among individuals with diabetes in a Medicaid managed-care program. Int J Psychiatry Med. 2008;38:13-29.
10. Zisook S, Shear K, Kendler KS. Validity of the bereavement exclusion criterion for the diagnosis of major depressive episode. World Psychiatry. 2007;6: 102-107.
11. Brunet A, Akerib V, Birmes P. Don’t throw out the baby with the bathwater (PTSD is not overdiagnosed). Can J Psychiatry. 2007;52:501-502.
12. Kessler RC, Demler O, Frank RG, et al. Prevalence and treatment of mental disorders, 1990 to 2003. N Engl J Med. 2005;352:2515-2523.
13. Zanarini MC, Frankenburg FR, Dubo ED, et al. Axis I comorbidity of borderline personality disorder. Am J Psychiatry. 1998;155:1733-1739.
14. Shea MT, Zlotnick C, Weisberg RB. Commonality and specificity of personality disorder profiles in subjects with trauma histories. J Personal Disord. 1999; 13:199-210.
15. Rao NP, Reddy YC, Kumar KJ, et al. Are neuropsychological deficits trait markers in OCD? Prog Neuropsychopharmacol Biol Psychiatry. 2008;32:1574-1579.
16. Kashner MT, Rush JA, Suris A, et al. Impact of structured clinical interviews on physicians’ practices in community mental health settings. Psychiatr Serv. 2003;54:712-718.
17. Shear MK, Greeno C, Kang J, et al. Diagnosis of nonpsychotic patients in community clinics. Am J Psychiatry. 2000;157:581-587.
18. Ventura J, Liberman RP, Green MF, et al. Training and quality assurance with the Structured Clinical Interview for DAM-IV (SCID-I/P). Psychiatry Res. 1998; 79:163-173.
19. Corty E, Lehman AF, Myers CP. Influence of psychoactive substance use on the reliability of psychiatric diagnoses. J Consult Clin Psychol. 1993;61:165-170.
20. Zimmerman M. Psychiatric Diagnostic Screening Questionnaire (PDSQ). Los Angeles: Western Psychological Services; 2000.
21. Zimmerman M, Mattia JI. The Psychiatric Diagnostic Screening Questionnaire: development, reliability and validity. Compr Psychiatry. 2001;42:175-189.
22. Zimmerman M, Mattia JI. A self-report scale to help make psychiatric diagnoses: the Psychiatric Diagnostic Screening Questionnaire. Arch Gen Psychiatry. 2001;58:787-794.
23. Hyler SE. Personality Diagnostic Questionnaire-4. New York: Psychiatric Institute; 1994.
24. Loranger AW. Personality Disorder Examination (PDE) Manual. Yonkers, NY: DV Communications; 1988.
25. Loranger AW, Sartorius N, Andreoli A, et al. The International Personality Disorder Examination: The World Health Organization/Alcohol, Drug Abuse, and Mental Health Administration pilot study of personality disorder. Arch Gen Psychiatry. 1994;51:215-224.
26. World Health Organization. International Classification of Disease–10th revision. Geneva: WHO; 2007.
27. Spitzer RL, Williams JBW, Gibbon M, First MB. Structured Clinical Interview for DSM-III-R Personality Disorders (SCID-II). New York: New York State Psychiatric Institute; 1989.
28. Coolidge FL, Merwin MM. Reliability and validity of the Coolidge Axis II Inventory: a new inventory for the assessment of personality disorders. J Pers Assess. 1992;59:223-238.
29. Chantarujikapong S, Smith E, Fox LW. Comparison of the Alcohol Dependence Scale and diagnostic interview schedule in homeless women. Alcohol Clin Exp Res. 1997;21:586-595.
30. Abela JR, Payne AV, Moussaly N. Cognitive vulnerability to depression in individuals with borderline personality disorder. J Personal Disord. 2003;17:319-329.
31. Garno JL, Gunawardane N, Goldberg JF. Predictors of trait aggression in bipolar disorder. Bipolar Disord. 2008;10:285-292.
32. Zimmerman M, Mattia JI. Differences between clinical and research practices in diagnosing borderline personality disorder. Am J Psychiatry. 1999;156: 1570-1574.
33. Craske MG, Roy-Byrne P, Stein MB, et al. CBT intensity and outcome for panic disorder in a primary care setting. Behav Ther. 2006;37:112-119.
34. Leskin GA, Sheikh JI. Lifetime trauma history and panic disorder: findings from the National Comorbidity Survey. J Anxiety Disord. 2002;16: 599-603.
35. Vujanovic AA, Zvolensky MJ, Bernstein A. Posttraumatic stress and panic psychopathology: clinical import, research advances, and future directions–introduction to the special issue. Cogn Behav Ther. 2008;37:63-65.
36. Markowitz JC, Moran ME, Kocsis JH, Frances AJ. Prevalence and comorbidity of dysthymic disorder among psychiatric outpatients. J Affect Disord. 1992; 24:63-71.
37. Singh MK, DelBello MP, Kowatch RA, Strakowski SM. Co-occurrence of bipolar and attention-deficit hyperactivity disorders in children. Bipolar Disord. 2006;8:710-720.
38. Elia J, Ambrosini P, Berrettini W. ADHD characteristics: I. Concurrent co-morbidity patterns in children & adolescents. Child Adolesc Psychiatry Ment Health. 2008;2:15.
39. Kjelsås E, Bjørnstrøm C, Götestam KG. Prevalence of eating disorders in female and male adolescents (14-15 years). Eat Behav. 2004;5:13-25.

Evidence-Based References

Ramirez Basco M, Bostic JQ, Davies D, et al. Methods to improve diagnostic accuracy in a community mental health center. Am J Psychiatry. 2000;157:1599-1605.
Kashner MT, Rush JA, Suris A, et al. Impact of structured clinical interviews on physicians’ practices in community mental health settings. Psychiatr Serv. 2003;54:712-718.