• DOI:10.1159/000288867
• Corpus ID: 3376436

Somatization: syndromes or processes?

@article{Sensky1994SomatizationSO,
  title={Somatization: syndromes or processes?},
  author={Tom Sensky},
  journal={Psychotherapy and psychosomatics},
  year={1994},
  volume={61 1-2},
  pages={
          1-3
        }
}

• T. Sensky
• Published 1994
• Psychology, Medicine
• Psychotherapy and psychosomatics

Tom Sensky, PhD, FRCPsych, Department of Psychiatry, Charing Cross and Westminster Medical School, West Middlesex University Hospital, Isleworth, Middlesex TW7 6AF (UK) The accompanying paper [1] is Robert Kellner’s last, published posthumously. In it, he provides a scholarly review of a number of psychosomatic syndromes, discussed in further detail in his recent book [2]. As Weiner [3] has commented, such ‘reviews of the state of knowledge in a field illuminate the gaps in its fabric: they highlight what still needs to be accomplished’. Kellner’s review elegantly demonstrates the problems with existing diagnostic classifications in psychosomatic medicine. While evidence is presented supporting the validity of the disorders reviewed, there are clearly also problems with the existing nosology. Although current research shows considerable differences between syndromes (one prerequisite of any useful classification), Kellner reminds us that such differences are deceptive, for two main reasons. First, there is considerable heterogeneity within syndromes. Secondly, within individuals, presenting symptoms might have different causes over time. Also, it is difficult to distinguish somatizing patients from others on the basis of their presenting symptoms alone [4]. One response to this diagnostic conundrum is to reject the need to reach a diagnosis, as a putative means of fostering an holistic approach to the individual’s problems. Such a stance has been rightly criticized [5] because it severely limits attempts to reach a more complete understanding of the aetiology and prognosis of psychosomatic syndromes. It is worth remembering that similar arguments were applied only a few years ago to the diagnosis of schizophrenia [6]. The opposite response to such difficulties in achieving accurate diagnosis is illustrated by the development of the diagnosis of Briquet’s syndrome. Detailed and restrictive criteria were applied [7] and validated by follow-up and family studies [8]. Such an approach may well be valuable in research, but might be less useful in everyday clinical practice [9], where the prevalence of such ‘pure’ syndromes is low, and most patients present in ways which fall outside the definition of the syndrome. The DSM and ICD 10 classifications illustrate another approach, namely the classification of syndromes by operational definitions or diagnostic guidelines. However, it may be extremely difficult (as Kellner states) to classify a given symptom within the somatoform disorders of DSM-IIIR. Perhaps this is not as surprising as it might seem. The DSM classification is descriptive and atheoretical [10]. Such a classification works most effectively when the operational criteria (or, more accurately, semantic criteria [6]) happen to define syndromes which can be distinguished also in their aetiology and prognosis. As Kellner’s review illus-