On the way to expert systems

Comparing DSM-III Computer Diagnoses with CATEGO (ICD) Diagnoses in Depressive and Schizophrenic Patients
  • K. Maurer
  • H. Biehl
  • C. Kühner
  • W. Löffler


Operationalized diagnostics deal with the standardized assessment of psychiatric symptoms as well as diagnostic criteria. As a diagnostic system based on criteria, the DSM-III was chosen to identify operationalized diagnoses based on the Present State Examination (PSE-9) and some additional DSM-III specific items. By relating PSE symptoms to the diagnostic criteria of DSM-III, an easily applicable expert system leading to DSM-III diagnoses was developed. In two samples of 30 schizophrenic and 51 depressive patients the DSM-III computer diagnoses are contrasted with the ICD-8 diagnoses of the PSE/CATEGO system. In defining a “case”, only minimal differences between the two computer programs were found. In the sample of schizophrenics, CATEGO led to 114 (81%) diagnoses and the DSM-III program to 112 (79%) diagnoses; for the depressive patients 43% cases were identified by CATEGO and 45% by the DSM-III algorithm. Comparing the diagnosis of “acute schizophrenic disorders”, both programs arrived at similar percentages. (CATEGO: 51%; DSM-III: 57%). However, CATEGO is limited to two different subtypes (295.2 and 295.3), whereas the DSM-III program covers the total range of possible schizophrenic subtypes. Furthermore, the DSM-III program identified residual subtypes of schizophrenia in 23% of the diagnostic decisions. In the short-term course of the schizophrenic patients, CATEGO identified 27%–43% with affective diagnoses with high stability per cross-section. Using the DSM-III algorithm affective diagnoses were rather rare (maximum of 17%), marking unstable changes from acute to residual states of the psychosis. In the cohort of affective patients the correspondence between both programs was quite good, especially for the patients with a diagnosis of depression, but in total the DSM-III program requires stricter criteria for affective — especially manic — disorders, whereas CATEGO needs a somewhat higher symptom level for anxiety syndromes to reach the diagnostic threshold.

Key words

Computer-assisted diagnosis DSM-III algorithm PSE/CATEGO system 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. American Psychiatric Association (APA) (1980) Diagnostic and statistical manual of mental disorders (3rd edn, DSM-III). Washington, DCGoogle Scholar
  2. American Psychiatric Assosciation (APA) (1987) Diagnostic and statistical manual of mental disorders (3rd edn revised, DSM-IIIR). Washinton, DCGoogle Scholar
  3. Berner P, Katschnig H (1983) Principles of “multiaxial” classification in psychiatry as a basis of modern methodology. In: Helgason T (ed) Methodology in evaluation of psychiatric treatment. Proceedings of a workshop held in Vienna, 10–13 June 1981. Cambridge University Press, CambridgeGoogle Scholar
  4. Biehl H, Maurer K, Jung E (1985) Entwicklung von Behinderungen bei Schizophrenen — Erfassung, Differenzierung und Vorhersage. Ergebnisbericht des BMFT-Projekts 01 ZX 072/6Google Scholar
  5. Biehl H, Maurer K, Jung E, Krumm B, Schubart C (1986) Neuere Aspekte der Behinderungsforschung bei Schizophrenen — Praktische Anwendung, Versorgung und Klassifikation. Ergebnisbericht zum GSF/BMFT Projekt 0701608 6Google Scholar
  6. Biehl H, Maurer K, Jung E, Krüger G, Schubart C (1988) Reported symptoms in schizophrenic patients within five years of the onset of illness. In: Dencker SJ, Kulhanek F (eds) Treatment resistance of schizophrenia. Vieweg, Braunschweig/WiesbadenGoogle Scholar
  7. Cantor N, Smith EE, de Sales French R, Mezzich J (1980) Psychiatric diagnosis as prototype categorization. J Abnorm Psychol 89:183–193Google Scholar
  8. De Jong A, van den Brink W (1986) Developments in psychiatric classification; introduction of a prototype core classification. Tijdschr Psychiatr 28:426–442Google Scholar
  9. Hand DJ (1985) Artificial intelligence and psychiatry. Cambridge University Press, CambridgeGoogle Scholar
  10. Hayes-Roth F, Waterman DA, Lenat DB (1983) Building expert systems. Addison-Wesley, Reading, Mass.Google Scholar
  11. Jakubaschk J, Werner J (1974) Die Zuverlässigkeit psychiatrischer Diagnosen bei einer Wiederholungsuntersuchung. Nervenarzt 45:305–311Google Scholar
  12. Jakubaschk J, Werner J (1976) Konzepte psychiatrischer Diagnosen — eine Ergänzungsuntersuchung zur Reliabilität psychiatrischer Diagnosen. Arch Psychiatr Nervenkr 223:45–58Google Scholar
  13. Kalb R (1986) “Expertensysteme” in der Psychiatrie. Psychiatr Neurol Psychother Klin Prax 12:537–544Google Scholar
  14. Kreitman N, Sainsbury P, Morrissey J, Towers J, Scrivener J (1961) The reliability of psychiatric assessment: an analysis. J Ment Sci 107:887–908Google Scholar
  15. Krüger G, Maurer K, Biehl H, Jung E, Schubart C (1985) 5-Jahres-Verlauf eines Samples ersterkrankter Schizophrener — Symptomatologie —. In: Reimer F (ed) Prognose und Prophylaxe in der Psychiatrie, 16. Weinsberger Kolloquium. Weissenhof, WeinsbergGoogle Scholar
  16. Krüger G, Biehl H, Maurer K, Jung E, Bauer-Schubart C (1988) Longitudinale Symptomveränderungen bei schizophrenen Patienten über 5 Jahre nach Krankheitsbeginn. In: Bender W, Dencker SJ, Kulhanek F (eds) Schizophrene Erkrankungen. Therapie, Therapieresistenz — eine Standortbestimmung. Vieweg, Braunschweig/WiesbadenGoogle Scholar
  17. Olbrich R, Maurer K (1988) Zur 10. Revision der Internationalen Klassifikation der Krankheiten (ICD). Nervenheilkunde 7:139–141Google Scholar
  18. Philipp M, Maier W (1987) Diagnosesysteme endogener Depressionen. Springer, Berlin Heidelberg New YorkGoogle Scholar
  19. Sartorius N, Jablensky A, Cooper JE, Burke JD (1988) Psychiatric classification in an international perspective with special reference to chapter V (F) of the 10th Revision of the International Classification of Diseases “Mental, Behavioral and Development Disorders”. J Psychiatry [Suppl 1] 152:3–52Google Scholar
  20. Shepherd M, Brooke EM, Cooper JE, Lin T (1968) An experimental approach to psychiatric diagnosis. An international study. Acta Psychiatr Scand [Suppl] 201:7–89Google Scholar
  21. Spitzer RL, Fleiss JL (1974) A re-analysis of the reliability of psychiatric diagnosis. Br J Psychiatry 125:341–347Google Scholar
  22. Spitzer RL, Endicott J, Robins E (1975) Research diagnostic criteria for a selected group of functional psychoses, 2nd edn. Biometrics Research Division, New YorkGoogle Scholar
  23. Stede M (1984) Einführung in die künstliche Intelligenz. Luther, SprendlingenGoogle Scholar
  24. Waltz DL (1982) Künstliche Intelligenz. Spektrum der Wissenschaft, December 1982, pp 68–87Google Scholar
  25. Wing JK (1984) Defining the data base. World Psychiatric Association, Regional Symposium, Helsinki, June 18–21, 1984. Book of Abstracts, p 10Google Scholar
  26. Wing JK, Cooper JE, Sartorius N (1974) Measurement and classification of psychiatric symptoms. Cambridge University Press, CambridgeGoogle Scholar
  27. Wing JK, Babor T, Brugha T, Burke J, Cooper JE, Giel R, Jablensky A, Regier D, Sartorius N (1989) SCAN: Schedules for clinical assessment in neuropsychiatry. Arch Gen Psychiatry (in press)Google Scholar
  28. World Health Organization (1988) Provisional list of categories for chapter 5 (“F”) Mental Disorders in the tenth revision of the International Classification of Diseases. WHO, GenevaGoogle Scholar

Copyright information

© Springer-Verlag 1989

Authors and Affiliations

  • K. Maurer
    • 1
  • H. Biehl
    • 1
  • C. Kühner
    • 1
  • W. Löffler
    • 1
  1. 1.Zentralinstitut für Seelische GesundheitMannheim 1Germany

Personalised recommendations