The Future of Mental Health Diagnostic Screening

The CAD-MDD (Computerized Adaptive Diagnostic Test for Major Depressive Disorder) is a diagnostic screening tool that provides sensitivity of 0.95 and specificity of 0.87 when compared with a clinician-based 1- to 2-hour DSM diagnostic interview for depression, and it does so using an average of only 4 items and typically requires less than a minute to complete.¹ The speed and accuracy of this screening tool surpass those of any existing depression diagnostic screener.

The CAD-MDD complements our previous work² on the development of the Computerized Adaptive Test–Depression Inventory (CAT-DI) to assess level of depression severity. The CAT-DI is based on multidimensional item response theory^3,4 and, like the CAD-MDD, provides highly accurate information in a very short time. We have also developed the CAT-ANX for anxiety.⁵ Both of these dimensional tests require an average of only 12 items per person to produce correlations of 0.95 with total scores on the respective item bank (consisting of 389 items for the CAT-DI and 431 items for the CAT-ANX). Thus, these tests allow us to quickly extract the information that would have been provided by answering about 400 items using conventional testing methods.

When used together, the CAD-MDD and the CAT-DI enable a clinician to rapidly and accurately determine whether an individual has major depression and then evaluate the effectiveness of treatment over time.

An integral part of the CAT-DI is a set of 14 suicidal ideation and behavior screening questions, incorporating the now-validated Columbia–Suicide Severity Rating Scale (C-SSRS).⁶ If positive, a suicide alert can automatically be generated and sent to a clinical monitor(s) immediately via e-mail or text.

The CAD-MDD and related CAT-DI and CAT-ANX share goals similar to those of the new version of DSM: they seek to improve screening and assessment of mental health disorders in a number of ways for patients, clinicians, and caregivers, including decreasing clinician and patient burden. The third section of DSM-5 is aimed at providing tools for cross-cutting and dimensional assessment, often involving patient-reported outcomes. The electronic version of the DSM-5 will allow for the development and application of many more scales and certainly better possibilities for tracking change and the effectiveness of treatment. Another objective of DSM-5 is to improve the interface with the rest of medicine, especially primary care.⁷

The criteria for major depression did not change between DSM-IV and DSM-5, but the new manual provides more opportunities to identify subgroups and to deal with co-morbidity. The CAD and CAT methods described here and in related articles go well beyond existing approaches for dimensional assessment and screening for diagnoses by dramatically reducing patient burden, providing the possibility of screening and monitoring entire populations for mental health disorders and suicide risk (eg, primary care, active military, veterans), and serving as a model for the assessment of other forms of psychopathology that are of particular relevance for the new DSM.

Financial disclosure:Drs Gibbons, Frank, and Kupfer are founders of Psychiatric Assessments, Inc, which will ultimately distribute the CAD-MDD. Dr Gibbons has no additional conflicts of interest. Dr Frank has received honoraria from Lundbeck; is a member of the speakers/advisory board for Servier; and has received royalties from Guilford Press and American Psychological Association Press. Dr Kupfer is a consultant for the American Psychiatric Association.

References

1. Gibbons RD, Hooker G, Finkelman MD, et al. The Computerized Adaptive Diagnostic Test for Major Depressive Disorder (CAD-MDD): a screening tool for depression. J Clin Psychiatry. 2013;74(7):669–674. Abstract

2. Gibbons RD, Weiss DJ, Pilkonis PA, et al. Development of a computerized adaptive test for depression. Arch Gen Psychiatry. 2012;69(11):1104–1112. Erratum in Arch Gen Psychiatry. 2013;70(1):30. PubMed

3. Gibbons RD, Hedeker DR. Full-information item bi-factor analysis. Psychometrika. 1992;57(3):423–436. http://link.springer.com/article/10.1007/BF02295430. Accessed August 26, 2013.

4. Gibbons RD, Bock RD, Hedeker D, et al. Full-Information item bi-factor analysis of graded response data. Appl Psychol Meas. 2007;31(1):4–19. http://apm.sagepub.com/content/31/1/4.short. Accessed August 26, 2013.

5. Gibbons RD, Weiss DJ, Pilkonis PA, et al. Development of the CAT-ANX: a computerized adaptive test for anxiety [published online ahead of print August 9, 2013]. Am J Psychiatry. doi:10.1176/appi.ajp.2013.13020178. PubMed

6. Posner K, Brown GK, Stanley B, et al. The Columbia–Suicide Severity Rating Scale: initial validity and internal consistency findings from three multisite studies with adolescents and adults. Am J Psychiatry. 2011;168(12):1266–1277. PubMed

7. Kupfer DJ, Kuhl EA, Regier DA. DSM-5: the future arrived. JAMA. 2013;309(16):1691–1692. http://jama.jamanetwork.com/article.aspx?articleid=1656312. Accessed August 23, 2013.