Daniel Z Lieberman1, Roger Peele, Maryam Razavi. 1. Department of Psychiatry and Behavioral Sciences, George Washington University, Washington, DC 20037, USA. dlieberman@mfa.gwu.edu
Abstract
BACKGROUND: DSM-IV-TR is the standard for making psychiatric diagnoses in many countries. The diagnostic categorization of DSM-IV-TR is valued for its reliability, however it is well accepted that patients with the same diagnosis demonstrate substantial heterogeneity. In order to quantitatively characterize the degree of possible heterogeneity that can exist, combinatorial mathematics was used to calculate the number of possible ways to satisfy symptom criteria for the diagnosis of bipolar disorder as an example of this problem. SAMPLING AND METHODS: The formula n!/(r!(n - r)!), which calculates the number of different ways of choosing an unordered number of items without repetition from a larger set, was used to calculate the number of combinations which meet the criteria for the core episodes of bipolar disorder and the specifiers. RESULTS: The number of possible combinations for the core episodes ranged from 163 for a manic episode to 37,001 for a mixed episode. When the full collection of specifiers that DSM-IV-TR applies to bipolar disorder was used, the number of combinations was over 5 billion. CONCLUSIONS: The precision of medical communication about bipolar disorder is called into question by the billions of different ways that the criteria for this diagnosis can be met. As DSM-V is developed, the possible combinations for each diagnostic criterion should be calculated, and the effect this number has on clinical communication should be considered. (c) 2007 S. Karger AG, Basel
BACKGROUND: DSM-IV-TR is the standard for making psychiatric diagnoses in many countries. The diagnostic categorization of DSM-IV-TR is valued for its reliability, however it is well accepted that patients with the same diagnosis demonstrate substantial heterogeneity. In order to quantitatively characterize the degree of possible heterogeneity that can exist, combinatorial mathematics was used to calculate the number of possible ways to satisfy symptom criteria for the diagnosis of bipolar disorder as an example of this problem. SAMPLING AND METHODS: The formula n!/(r!(n - r)!), which calculates the number of different ways of choosing an unordered number of items without repetition from a larger set, was used to calculate the number of combinations which meet the criteria for the core episodes of bipolar disorder and the specifiers. RESULTS: The number of possible combinations for the core episodes ranged from 163 for a manic episode to 37,001 for a mixed episode. When the full collection of specifiers that DSM-IV-TR applies to bipolar disorder was used, the number of combinations was over 5 billion. CONCLUSIONS: The precision of medical communication about bipolar disorder is called into question by the billions of different ways that the criteria for this diagnosis can be met. As DSM-V is developed, the possible combinations for each diagnostic criterion should be calculated, and the effect this number has on clinical communication should be considered. (c) 2007 S. Karger AG, Basel