Hong Kong Journal of Psychiatry (1997) 7 (2) 4-8


Gabor S. Ungvari, Helen F.K. Chiu, Henry C.M. Leung, L.Y Chow, Yu Hong, Eddie M.P. So & Francis C.K. Lum

pdf Full Paper in PDF


Some aspects of the current Anglo-American diagnostic concept of schizophrenic psychoses are examined from the viewpoint of classical and modem European psychiatry. Within this context three main areas are discussed: (1) the "neo­ Kraepelinian" revival of the concept of schizophrenia in modem Anglo-Saxon psychiatry; (2) the diminishing role of the clinician-psychiatrist in the diagnostic process in current clinical practice and research, and (3) the role of clinical experience and the raecox-feeling" in the diagnosis of schizophrenia.

Keywords: schizophrenia, diagnosis, psychopathology


It was exactly 100 years ago that in the 5th edition of his textbook Emil Kraepelin first conceptualized dementia praecox, later renamed schizophrenia, as a putative disease entity (Kraepelin, 1896). Since then, the definition and diagnosis of schizophrenia has puzzled, and continues to puzzle, generations of psychiatrists, clinicians as well as researchers. Comprehensive reviews on the topic are readily available so it would be an exercise in redundancy to produce another overview of the two, three, four or five factor component model of schizophrenia (e.g. Andreasen & Carpenter, 1993; Rey et al, 1994; Arndt et al, 1995). Instead, we have compiled a few critical comments on some aspects of the currently prevailing Anglo-American diagnostic concept of schizophrenia. Our vantage point is the classic and modem continental European psychiatry which usually escapes the attention of psychiatrists trained in the British or American schools of thought.

We shall touch upon the following loosely linked issues : (1) the "neo-Kraepelinian revival" of the concept of schizophrenia; (2) the diminishing role of the clinician in the diagnostic process and (3) the role of clinical experience and the "praecox-feeling".


The past 25 years has been hailed as the eo-Kraepelinian revival" (Klerman, 1978) or "neo-Kraepelinian revolution" (Compton & Guze, 1995) in Anglo-American psychiatry claiming that recent diagnostic practices as exemplified by the subsequent editions of DSM-III, DSM-III-R and DSM-N represent a re-occurrence of descriptive psychiatry "in which careful observation of symptoms, signs and course of mental diseases become the diagnostic_ criteria themselves" (Compton & Guze, 1995). We would like to question this widely accepted theoretical position using schizophrenia as an example. In agreement with a number of authorities (e.g. Berrios, 1994; Van Praag, 1992) we submit that modem diagnostic efforts only partly correspond to what classical descriptive psychiatry has achieved and current diagnostic criteria for schizophrenia are but a diluted and simplified version of views of Kraepelin and Bleuler, the two most oft-cited founders of the concept ·of schizophrenia. Limited space permits mentioning only a few aspects relevant to our thesis.

Kraepelin, Bleuler and their contemporaries based their schizophrenia concept on observable signs and symptoms embracing a significantly larger area of descriptive psychopathology than modem clinical practice and research operate. As two eminent European psychiatrists, Parnas and Bovet (1991) put it : 'if psychiatry remains stuck with the use of structured interviews solely compatible to pre-existing operational criteria, it will cut itself off from the major variable, namely psychopathology". The following examples will illustrate this point.

By a very conservative counting, we found 95 symptoms defining schizophrenia in Bleuler's textbook (Bleuler, 1911/1950) while Kraepelin (1913/1919) listed 65 symptoms of dementia praecox. ICD-10 (WHO, 1992) and DSM-N (APA, 1994) considerably decreased the number of symptoms used as diagnostic criteria for schizophrenia to 38 and 30 respectively.

It is not only the sheer number of signs and symptoms what makes modem classification pale in comparison with classical accounts of schizophrenia but the omission of more complex psychopathological phenomena. For instance, the following 3 symptoms regarded by Bleuler (1911/1950) as fundamental in recognizing schizophrenia have all but disappeared in modem descriptive psychiatry. Autism was defined as the etachment from reality together with the relative and absolute predominance of inner life" (Bleuler, 1911/1950, p. 63). No proper word exists for Benommenheit in English. It is usually translated as clouded or doped states or mental inertia. Bleuler described Benommenheit as the "slowing up of all psychic processes ... in conjunction with an incapacily for dealing with any relatively complicated or unusual situation ...[while] ... will power seems to be relatively well, or completely, preserved [coupled with the] absence of depression and evidences of a very mild degree of confusion" (Bleuler, 1911/1950, pp. 221). Pressure of thoughts or thought-overflow, "is a pathologically increased flow of ideas" was called one of "two disturbances peculiar to schizophrenia" (Bleuler, 1911/1950, p.14) when "the patients have the feeling of being compelled to think" (Bleuler, 1911/1950, p. 32). In contrast to flight of ideas, pressure of thought is subjectively quite unpleasant, repetitive, unproductive, unstoppable, has a "made" character and not associated with any symptom seen in manic syndromes.

Kraepelin formulated the most essential features of dementia praecox in the following way:: "the weakening of judgment, of mental activily and of creative ability, the dulling of emotional interest and the loss of energy, lastly the loosening of the inner unify of psychic life would have to be reckoned among the fundamental disorders." (Kraepelin, 1913/1919, p. 248). Some of these complicated symptoms are included in the negative syndrome in contemporary diagnostic criteria but others, particularly "loosening of the unify of inner psychic life" are ignored.

Of the fundamental symptoms just mentioned, the importance of autism has never faded away in European psychiatry although its interpretation has gradually changed. Bleuler's (1911/1950) and Kraepelin's (1913/1919) definitions emphasized the patient withdrawal from social and interpersonal interactions. Minkowski (1933), Binswanger (1956) and Blankenburg (1969) put the accent on the loss of vitalily and that of common sense, that is, the inability of engagement with everyday life so evident and natural to everyone. The most recent approach to autism (Parnas & Bovet, 1991) went a step further and postulated that autism is one of the manifestations of a basic neuro-integrative defect underlying all schizophrenic psychoses. Their definition implies a subtle perceptive-cognitive defect behind autism: "a defect in the elementary perceptual / expressive anchoring of the subject in the outer world" (Parnas & Bovet, 1991).

While narrowing the clinical boundaries of schizophrenic psychoses and paying more attention to their course in recent diagnostic systems certainly constitute a shift towards the Kraepelinian model, the heavy reliance upon Schneider's first-rank symptoms in describing the schizophrenic syndrome contradicts the declared "neo-Kraepelinian" nature of modem classifications. To operationalize Schneider's views on schizophrenia was a pragmatic, expedient decision resulting in a loss of complexily covered by Bleuler and Kraepelln' s system.

"Neo-Kraepelinian revival" is a spurious claim in another respect. Although ICD-10 and DSM-N comprise 7 and 5 subtypes of schizophrenia respectively, current schizophrenia research pays only lip se1vice to the clinical heterogeneity of the disorder. A substantial proportion of research papers, biological as well as clinical, refer to their sample simply as patients meeting the criteria for schizophrenia according to, say, DSMIII-R or DSM-N. Kraepelin, while emphasizing the preliminary nature of his description and subdivision of dementia praecox, delineated 26 putative subtypes. Kraepelin underscored the uncertainly regarding the connections between his subtypes and queried if they could all be subsumed under the umbrella of dementia praecox (Kraepelin, 1913/1919, p. 89). Despite his doubts about the validi1y of his own sub-classification of dementia praecox, the attempt to describe 26 sub1ypes clearly suggests that Kraepelin was a nosologist who did not simply divide endogenous psychoses into two groups, dementia praecox and manic-depressive illness (Kraepelin, 1913/1919).

Concluding this section, it seems that modem psychiatry operates with a highly advanced methodological apparatus such as sophisticated statistical analysis, neuroimaging and the like, while the clinical characterization of schizophrenic psychoses is almost sketchy and simplified in comparison to those of the founders of the original concept. This notion has been reiterated by eminent researchers equally familiar with AngloSaxon and European psychiatry such as Van Praag who asserted that " a variety of psychopathological symptoms are ignored by the available measuring instruments, and are not properly utilized to diagnose and classify mental disorders" (Van Praag, 1992).


It is paradoxical that despite striving for objectivity, modem psychiatry has placed strong emphasis on the subjective aspects of mental state examinations i.e. those reported by patients such as the first-rank symptoms. At the same time, relatively less heed has been paid to observable behavioural signs and symptoms although lately there have been exceptions (e.g. Harvey et al. 1996). The focus on subjective symptoms in the diagnosis of schizophrenia was reinforced by Kendell's oft-cited diagnostic experiments (Kendell, 1973). Kendell asked a group of psychiatrists to diagnose a series of patients presented to them either on video-tape, or audio-tape or simply by transcripts of previously conducted interviews. To Kendell's surprise, there was no difference between diagnostic confidence and accuracy achieved by employing the three different methods. This result led Kendell to the conclusion "that psychiatrists rely almost entirely on what the patient says rather than on the way he behaves [which] can be used to justify the use of audio-tapes or transcripts for research purposes instead of live interviews and rebuts the argument that to do so involves losing vital information" (Kendell, 1975, p.53). This result was, however, entirely expected since behavioural symptoms had not featured in the making of diagnosis in the first place.

In the same series of experiments, Kendell (1973) also found that given 4 years training, further years in clinical practice failed to increase diagnostic confidence and accuracy. In other words, a basic training sufficed to reach as sound diagnostic decisions as very experienced psychiatrists did.

Having simplified the diagnostic process and diminished the importance of clinical experience, the next step was the quasielimination of the clinician psychiatrist. In the method of the socalled "best estimate life-time diagnosis" (Leckmann, 1982), a procedure used by leading American research centres (e.g. Gershon et al. 1988), different parts of the assessment, i.e. the socio-demographic background, the history, and the mental state are collected independently by different research assistants, social workers, nurse clinicians, while the final diagnosis is established by consensus conference where the participants are familiar only with certain aspects of the whole clinical picture and the decision is made on the basis of written material. This approach, characterized as "practical American technology" (Vaillant, 1984) excludes the "conversation with the patient ... [which is the]... most important method of examination" (Jaspers, 1913/1963, p. 826.)

When using structured interview schedules in establishing the diagnosis of schizophrenia, significant amount of information is lost due to the lack of longitudinal personal observation of the usually changing clinical presentation (Arndt et al. 1995). Lack of personal follow-up and the spuriously precise and simplified diagnostic criteria in clinical practice and research would be mainly responsible for the instability of the diagnosis of schizophrenia over time. In a recent large-scale 7 year-follow-up study involving 936 inpatients, 2.9 % of patients initially diagnosed with schizophrenia received another diagnosis at subsequent hospitalizations while 32.8 % of the subjects originally diagnosed with non-schizophrenic illnesses were later given the diagnosis of schizophrenia (Chen et al. 1996). Apart from diagnostic instability, the symptom-profile of schizophrenia has been found fairly unstable with the notable exception of negative symptoms (Arndt et al. 1995). In a cohort of 178 geriatric schizophrenic patients, usually regarded as clinically particularly stable, at 1-year follow up the negative, positive and mixed subtypes changed in 40 % , 50 % and 54 % respectively at 1-year follow up (Putnam et al. 1996).

To demonstrate the importance of personal longitudinal observation, a small reliability study using the Bush-Francis Catatonia Rating Scale (BFCRS), a 23-item instrument rating observational items cross-sectionally on a 4-point scale, are presented. Three equally trained clinicians jointly examined two groups of chronic schizophrenic patients and rated each subjects independently on the· BFCRS. Patients in the first group had been unknown to all raters prior to the assessment. The average scores of the three clinicians were 10.9, 11.2 and 10.0 respectively suggesting that they evaluated motor symptoms very similarly. Patients of the second group had been regularly seen by Clinician A before the actual rating exercise but unknown to Clinicians B and C. Thus Clinician A rated the subjects of the second group not only cross-sectionally but also relying on his long-term observations. In the second group, Clinician A's average score was 2 while Clinician B and C averaged only 4.9 and 5.8 respectively. Even without the results of statistical analysis, the difference in rating of the second group significantly underscores the importance of repeated personal observation of research subjects.


Kendell's above-cited diagnostic experiments (Kendell, 1973) also confirmed earlier reports (e.g. Gauron & Dickinson, 1969; Sandifer et al. 1970) that the first few minutes play a pivotal role in the diagnostic process in terms of recognizing a great number of symptoms and reaching an early diagnosis which will seldom change later. Rumke, a Dutch psychiatrist, coined the term "praecox-feeling" to denote a markedly subjective, rather instinctive feeling or intuition induced in the examiner by the schizophrenic patients right after the first seconds or minutes of their encounter (Rumke, 1941). The praecox-feeling has never been clearly defined. Rumke (1941) suggested that "impossibility of empathy", "changes of motor behaviour and speech", the "lack of exchange of affect" and an "instinctual rapport" were its main components. Simply put, praecox-feeling is a specific unease emanating in the psychiatrist reflecting the detachment, distance and alienation of the schizophrenic patient.

Praecox-feeling is regularly encountered by practicing psychiatrists even in our era of explicit diagnostic criteria and structured diagnostic instruments. According to a survey of 1196 German psychiatrists, 85.8 % of them experienced praecox-feeling, 57.9 % found it diagnostically reliable and 25% regarded it the most reliable indicator of the presence of schizophrenia (Irle, 1962). Even more surprisingly, nearly identical figures - 83. 2 %. 64.6 % and 20.7 % respectively - were found in a cohort of 257 New York-based psychiatrists (Sagi & Schwartz, 1988).

We set out to investigate the reliability of praecox-feeling; firstly, because it has never been rigorously tested particularly against modem diagnostic criteria and assessment methods and, secondly, because it has never been examined in nonWestern patients. The latter point is significant since praecoxfeeling is associated with culture-specific aspects of behaviour and emotional expressivity.

The method was simple. A psychiatrist interviewed 102 consecutively admitted consenting patients within 2 days of their admission in front of 5 passively watching psychiatrists who had never seen those patients before. This interview took place within 2 days following admission and lasted only for 2 minutes. It consisted of a few standard questions, as neutral as possible, such as name, age, address, marital status, age of parents, number and age of siblings and/or their children. The five clinicians independently rated the absence, dubious and definitive presence of praecox-feeling 30 seconds after the commencement of the interview and at its completion, that is, after 2 minutes. Then, the patient immediately left the room and the raters attempted to analyze on what grounds they had made their decision. Their choices were: general appearance, psycho-motility, overall behaviour, speech (content and its formal characteristics), and inexplicable or unidentifiable reasons.

The same or the next day following the brief initial interview, all patients were examined with the SCID by the sixth psychiatrist who had conducted the first interview. The same clinician also collated all diagnostic information on each patients and, in collaboration with the treating team, came to a final diagnostic conclusion against which the reliability of praecoxfeeling was measured. The diagnostic distribution of the cohort was as follows: 34 schizophrenia, 61 psychiatric disorders other than schizophrenia and 7 psychiatrically symptom-free subjects who were admitted for sleep assessment.

The preliminary results are shown on Table I. At 30 seconds and at 2 minutes, between 61.1% and 82.5 % of the praecoxfeelings of all clinicians was confirmed by the final research diagnosis. Not one of the psychiatrically healthy subjects were felt to suffer from schizophrenia. At the same time, false positive cases ranged between 31.1% and 50.1 % meaning that while the sensitivity of the praecox-feeling was reasonably high, its specificity was fairly low. Praecox-feeling was most frequently associated with subtle disturbances of psycho-motility, an unidentifiable sense of idiosyncrasy, appearance and overall behaviour.

We wish to draw only a tentative qualitative conclusion pertinent to our topic. Praecox-feeling seems to exist suggesting that there is a subjective element of unknown composition, magnitude and reliability when making the diagnosis of schizophrenia, thus personal observation of the patient has an important role in diagnostic decisions. Hence, the accumulated clinical experience of the psychiatrists is irreplaceable, at least in our current clinical practice and research.

We wish to emphasize that most European psychiatrists greatly appreciate, or even admire, the tremendous methodological improvement Anglo-American psychiatry has made over the past 20 years, particularly the introduction of structured and standardized assessment. No one wants to return to the old, unreliable diagnostic practice. Indeed, most European schools want to enrich and broaden the operationalized criteria by incorporating the traditions of classical descriptive psychopathology (Parnas & Bovet, 1991)


Andreasen, N.C. & Carpenter, W.T. (1993) Diagnosis and classification of schizophrenia. Schizophrenia Bulletin, 19: 199-214.

Arndt, S., Andreasen, N.C., Flaum, M. Miller, D. & Nopoulos, P. (1995) A longitudinal study of symptom dimension in schizophrenia. Archives of General Psychiatry, 52: 352- 360.

APA (1994) Diagnostic and statistical manual of mental disorders. Fourth edition. DSM-IV. Washington, DC : American Psychiatric Association.

Berrios, G.E. (1994) The language of psychiatry : a time for a change ? Journal of the Hong Kong College of Psychiatrists, 4: 5-10.

Binswanger, L. (1956) Drei Formen missgluckten Daseins Verstiengenheit, Verschrobenheit, Manieriertheit. Tubingen Niemeyer.

Blankenburg, W. (1969) Ansatze zu einer Psychopathologie des ommon sense". Confinia Psychiatrica, 12: 144-163.

Bleuler, E. (1911/1950) Dementia praecox or the group of schizophrenias. (Transl. Zinkin, J.) New York : International Universities Press.

Chen, Y.R., Swann, A.C., Burt, D.B. (1996) Stability of diagnosis in schizophrenia. American Journal of Psychiatry, 153: 682-686.

Compton, W.M. & Guze S.B. (1995) The neo-Kaepelinian revolution in psychiatric diagnosis. European Archives of Psychiatry and Clinical Neurosciences, 245: 196-201.

Gauron, E.F. & Dickinson, J.K. (1969) The influence of seeing the patient first on diagnostic decision making in psychiatry. American Journal of Psychiatry, 126: 199-205.

Gershon, E. S. DeLisi, LE. Hamovit, J. et al. (1988) A controlled family study of chronic psychoses. Archives of General Psychiatry, 45: 328-336.

Harvey, C.A., Curson, D.A., Pantelis, C., Taylor, J. Barnes T.R.E. (1996) Four behavioural syndromes of schizophrenia. British Journal of Psychiatry, 168: 562-570.

Irle, G. (1962) Das raecoxgefuhl" in der Diagnostik der Schizophrenie. Archiv fur Psychiatrie und Zeitschrift fur die gesammte Neurologie, 203: 385-406.

Jaspers, K. (1913/1963) General psychopathology. (transl. Hoenig, J., Hamilton, M.W.) Manchester : Manchester University Press.

Kendell, R. E. (1973) Psychiatric diagnoses : a study of how they are made. British Journal of Psychiatry, 122: 437-445. Kendell, R.E. (1975) The role of diagnosis in psychiatry. Oxford : Blackwell Scientific Publications.

Klerman, G.L.(1978) The evolution of scientific nosology. In : Shershow, JC. ed. Schizophrenia: science and practice. Cambridge, MA: Harvard University Press, pp. 99-121.

Kraepelin, E. (1896) Psychiatrie. 5. Auflage. Leipzig : Barth.

Kraepelin, E. (1913/1919) Dementia praecox and paraphrenia. (trans.Barclay, R.M) Edinburgh: E.S. Livingstone.

Leckman, J.F. Sholomskas, D. Thompson, D.W. et al. (1982) Best estimate lifetime psychiatric diagnosis. Archives of General Psychiatry, 39: 879-883.

Minkowski, E. (1933) La Schizophrenie. Paris : Payot.

Parnas, J. & Bovet, P. (1991) Autism in schizophrenia revisited. Comprehensive Psychiatry: 32, 7-21.

Putnam, K. M., Harvey, P.O., Parrella, M., White, L., Kincaid, M., Powchik, P. Davidson, M. (1996) Symptom stability in geriatric chronic schizophrenic inpatients : a one-year follow-up study. Biological Psychiatry 39: 92-99.

Rey, E-R. Bailer, J. Brauer, W. Handel, M. Laubenstein, D, Stein A. (1994) Stability trends and longitudinal correlations of negative and positive syndromes within a three-year follow-up of initially hospitalized schizophrenics. Acta Psychiatrica Scandinavica, 90: 405-412.


Rumke, H.C. (1941) Het Kernsyptoom der Schizophrenie en het Praecoxgevoel. In : Rumke, H.C. Studies en Voordrachten over Psychiatrie. Amsterdam : Scheltema & Holkema, 1948. pp. 53-58.

Sagi, G.A. & Schwartz, M.A. (1988) The raecox feeling" in the diagnosis of schizophrenia : a survey of Manhattan psychiatrists. Schizophrenia Research, 1: 35.

Sandifer, M.G., Hordern, A. & Green, L.M. (1970) The psychiatric interview : the impact of the first three minutes. American Journal of Psychiatry, 126: 968-973.

Vaillant, G. E. (1984) The disadvantages of DSM-JII outweigh its advantages. American Journal of Psychiatry,141: 542-545. Van Praag, H.M. (1992) Reconquest of the subjective. British

Journal of Psychiatry, 160: 266-270.

WHO (1992) The ICD-10 classification of mental and behavioural disorders. Clinical description and diagnostic guidelines. Geneva : World Health Organization.

*Gabor S. Ungvari, PhD, FRANZCP, FHKCPsych, FHl{AM(Psych), Associate Professor, Department of Psychiatry, Chinese University of Hong Kong
Helen F.K. Chiu, MRCPsych, FHKCPsych, FHKAM(PsychJ, Associate Professor and Head , Department of Psychiatry, Chinese University of Hong Kong
Henry C.M. Leung, MRCPsych, FHKCPsych, FHl{AM(Psych), Consultant Psychiatrist, Department of Psychiatry, Prince of Wales Hospital
L.Y. Chow, MRCPsych, Assistant Professor, Department of Psychiatry, Chinese University of Hong Kong
Yu Hong, M.Sc. Department of Psychiatry, Chinese University of Hong Kong
Eddie M.P. So, Medical Of ficer, Kwai Chung Hospital
Francis C.K. Lum, Medical Of ficer, Castle Peak Hospital

*Correspondence : Dr Gabor S. Ungvari, Department of Psychiatry, 11/F, Prince of Wales Hospital, Shatin, N.T. Hong Kong.

View My Stats