Reply.

The study of Sarkar and Chandra (2003) attempts to show the prevalence of alexithymia and its association with abnormal illness behaviour among a socioculturally homogeneous group of patients with somatic symptoms. A somewhat similar study (Radhika 1997) was conducted in a general hospital psychiatry unit on 71 patients ( a rather sociocultually heterogeneous but with 70% Hindu population) presenting with somatic symptoms to the medical and surgical departments and referred for psychiatric help. The prevalence of alexithymia measured by the Toronto Alexithymia Scale was rather low (7%) among this population (mean score 58.53 +_ 10.95). The results of one way analysis of variance on the score of alexithymia among the diagnostic groups (anxiety, depressive and somatoform disorder) showed that the score of alexithymia was significantly higher among those with somatoform disorder than among those with anxiety disorder (p=0.001). A higher (a trend toward significance, p=0.077) alexithymia scores were observed among the smaller number of patients (N=15) fulfilling the criteria of abnormal illness behaviour as per Illness Behaviour Assessment Schedule than the other bigger group (N=56) who did not. Again, a trend of significant positive correlation (r = 0.232, p = 0.05) was found between the alexithymia scores and the number of somatic symptoms present in an individual patient in the entire study sample as well as among the two gender groups separately. We also looked into the relation between alexithymia characteristics and the range of coping mechanisms of the patients. A non-significant negative correlation was found only in the group of male patients.


Sir,
The study of  attempts to show the prevalence of alexithymia and its association with abnormal illness behaviour among a socioculturally homogeneous group of patients with somatic symptoms. A somewhat similar study (Radhika 1997) was conducted in a general hospital psychiatry unit on 71 patients ( a rather sociocultually heterogeneous but with 70% Hindu population) presenting with somatic symptoms to the medical and surgical departments and referred for psychiatric help. The prevalence of alexithymia measured by the Toronto Alexithymia Scale was rather low (7%) among this population (mean score 58.53 +_ 10.95). The results of one way analysis of variance on the score of alexithymia among the diagnostic groups (anxiety, depressive and somatoform disorder) showed that the score of alexithymia was significantly higher among those with somatoform disorder than among those with anxiety disorder (p=0.001). A higher (a trend toward significance, p=0.077) alexithymia scores were observed among the smaller number of patients (N=15) fulfilling the criteria of abnormal illness behaviour as per Illness Behaviour Assessment Schedule than the other bigger group (N=56) who did not. Again, a trend of significant positive correlation (r = 0.232, p = 0.05) was found between the alexithymia scores and the number of somatic symptoms present in an individual patient in the entire study sample as well as among the two gender groups separately. We also looked into the relation between alexithymia characteristics and the range of coping mechanisms of the patients. A non-significant negative correlation was found only in the group of male patients.
Although such studies tend to indicate some kind of clinical association of alexithymia, certain fundamental issues remain to be addressed. Is the construct of alexithymia cross culturally applicable? Is it possible to differentiate between trait and state of alexithymia among the community population? Does the presence of alexithymia tend to have any clinical validity? Unless these issues are thoroughly addressed and resolved in future research the relevance of alexithymia to our population will continue to remain in doubt in the sceptic mind of the clinician.

REPLY
Sir, Radhika and Sengupta raise many pertinent issues about our study  in their letter to the editor. The discrepancy between the rates of alexithymia between their and our study may be on account of several factors, not least being the sociocultural heterogeneity of their sample. It has been suggested that there are significant differences between Hindu and Muslim patients who present with multiples somatic symptoms, with Muslim patients tending to have higher rates of unexplained somatic complaints (Janakiramaiah and Subbakrishna, 1980). That was the precise reason why we chose a socioculturally homogenous sample. There is much evidence to suggest that subjects from a non-English speaking background tend to have higher scores on the Toronto Alexithymia Scale (Le et al, 2002;Taylor et al, 2003) and the reasons for that have been felt to be largely related to the cultural norms, particularly to parental emotional socialisation, i.e. the manner in which emotional expression is demonstrated by parents to their children (Le et al, 2002).
The contribution of culture to the ability to experience emotional distress, accept it, label it and then express it to others cannot be underestimated especially since in eastern (including Indian) cultures, emotional distress is not accorded the status of illness and is often felt to signify a personal weakness of moral fibre in the individual expressing it, thereby leading conversely to the use of body as an idiom of distress. The construct of alexithymia has been examined and assessed in multiple cultures and the scale translated into many languages, including Hindi (Pandey et al, 1996), with adequate psychometric properties (Taylor et al, 2003) to allow the conclusion, that it is an etic (pancultural) as opposed to an emic (unicultural, culturebound) phenomenon (Taylor et al, 2003). It matters not what one calls the phenomena but clinicians in India are well aware that many of their patients complain in a somatically focussed manner when the objective view may be that the cause of their distress is psychological and the use of the term alexithymia helps to label this phenomenon and measure it.
Although there appears to be a wider recognition that alexithymia is more likely to be a trait rather than state phenomenon, the debate is still ongoing. Longitudinal studies should answer such questions by assessing for the presence of alexithymia over time, including but going beyond episodes of somatic illness. Our study was not designed to answer this question but our view is closer to that of alexithymia being a trait phenomenon, with higher baseline scores with further possible rises in scores during times of somatic illnesses of non-organic nature.
The authors ask whether alexithymia has any clinical validity. This appears to be a redundant question in the face of large amount of literature that exists about the validity of this construct. A more appropriate question may be whether it has any clinical utility. Alexithymia has been found to be associated with tonic physiological arousal, certain types of unhealthy behaviour (e.g. substance misuse, eating disorders), a biased perception and reporting of somatic sensations and symptoms, and also appears to influence excessive health care use, albeit in a complex manner (Lumley, 1996). Presence of this trait in certain patients should alert professionals to the possibility of over-reporting of physical symptoms that may not be organically based, thereby being of much use in the general medical setting. It would also help target remedial measures such as psychotherapy by helping the patient to establish links between psychological distress and somatic symptoms. By identifying such patients early, one can prevent unnecessary investigations thereby allowing better utilization of, often scant, existing resources. The advantages are therefore many and are not merely restricted to direct benefits to the patient and his doctor.