Abstract
Objective:
This is a preliminary report on the development of a scale to measure the symptoms of unmāda (psychosis) attributable to tridosa (metabolic principles) by using the concepts of Ayurvedic medicine.
Design:
The 67-item unmāda specific symptom scale was developed on the basis of translation of Sanskrit verses describing vātaja (V), pittaja (P), and kaphaja (K) unmāda (specific symptoms of psychosis due to the imbalances of metabolic components) and by taking the opinions of experts (15 Ayurveda experts, 5 psychiatrists, and 5 psychologists).
Settings:
The setting for this study was Spandana Psychiatric Nursing Home Bangalore, India.
Subjects:
The scale was administered by an unblinded assessor to 30 consecutive patients with nonaffective psychotic disorders.
Results:
The unmāda specific symptom scale was associated with excellent internal consistency. The Cronbach's α for V, P, and K scales were 0.98, 0.98, and 0.97, respectively. The split-half reliability for V, P, and K scales were 0.97, 0.97, and 0.88 respectively. Scores on vātaja, pittaja, and kaphaja scales were inversely correlated, suggesting that they are mutually exclusive. The three subgroups of psychoses—paranoid schizophrenia, schizophrenia (unspecified), and unspecified nonorganic psychosis—had significantly different loadings on the three scores, having high scores on vātaja, pittaja, and kaphaja, respectively.
Conclusions:
The tridoshas in psychotic disorders can be measured reliably by this instrument. The scores on each of these doşas help in differentiating three types of psychosis (according to Āyurveda) that have good correspondence with prevailing classification. However, this scale must also be applied to the other 28 separate subcategories of the psychoses that are identified in the International Statistical Classifications of Diseases (version 10) and to the 10 variants of psychosis as defined by the American Psychiatric Association Diagnostic and Statistical Manual-IV-R to help better understand the true utility for use here with the various subcategories of schizophrenia.
Introduction
Several literary studies 6 –9 are available in the area of concepts of mental illness in Atharva Veda and Āyurveda classics. Studies also explain about the traditional classification of mental illnesses as described in the ancient texts of Āyurveda. The possible correlation with the modern classification has also been explained.
A study, proposing independent scientific evidence for the identification of the dosha, is available. 10 Studies on quantification of tridoshas are available (e.g., a questionnaire to assess the constitution of an individual developed on the basis of classical description and opinions of the experts is available 11 ). A preliminary version of model interview to identify the doshas has been published. 12
A simple and standardized instrument to measure the types of psychoses according to Ayurvedic concepts is not available. Hence, the present investigation was carried out to develop and evaluate an unmāda-specific symptom scale (USS).
Methods
USS was developed based on 75 Sanskrit statements (items) from available, authoritative, and ancient texts describing symptoms typical of vātaja, pittaja, and kaphaja unmāda (psychosis). Translated into English, these items were presented to 5 Ayurveda experts, they were asked about correctness of each statement, and to check whether any of the items were repeated. Of these, 72 items agreed on by all 5 experts were retained. The three items dropped had similar meanings. Ten (10) independent senior Ayurveda doctors examined each item further to indicate whether (1) the symptoms of vātaja, pittaja, and kaphaja unmāda selected for the scale are correct, and (2) it is an acceptable translation of the Sanskrit statements in the original texts. On their suggestions, some items were refined.
Items were then presented to 5 psychiatrists. For each item, they were asked to indicate whether these statements can be presented to patients with psychotic disorders and informants. All 72 items were found suitable. Five (5) independent psychologists reviewed the format of this scale and recommended a two-point scoring (0 and 1), which was adopted in the final USS.
Description of final scale
The USS has 72 items: 25 items for vātaja unmāda, 24 items for pittaja unmāda, and 23 items for kaphaja unmāda subscales. A clinician (using scripted interview) administered the scale and marked the presence or absence of each feature as 0 or 1 on the basis of responses of both the patient and the informant. The score of each subscale is the sum of all positive responses (Appendix 1).
Subjects
The patients in this study were outpatients from Spandana Psychiatric Nursing Home in Bangalore. Consecutive consenting patients of both sexes (15 males) with an age range of 22–55 years (mean ± standard deviation age = 38.6 ± 10.2 years) formed the sample (n = 30). A senior psychiatrist, S. Srinivas, with 25 years of experience with clinical interviews, diagnosed these patients as paranoid schizophrenia (n = 17), schizophrenia, unspecified (n = 7), and unspecified nonorganic psychosis (n = 6) according to International Classification of Diseases, 10th Revision (ICD-10). All these patients were under treatment for 6 months or longer and were accompanied with an informant who had lived with the patient since the onset of the disease or longer.
The 72-item USS was administered and rated by an author (S.P.S.) based on the information from the patient and the informant. The Statistical Package for Social Sciences (SPSS-10) was used for data analysis.
The item difficulty level was first assessed. The data were next analyzed for reliability. The split-half and Cronbach's α tests were applied for reliability analysis. Pearson's correlation analysis was done to check the degree of association between vāta, pitta, and kapha scores. The differences in mean scores of each diagnostic group were analyzed, using the Kruskal-Wallis test.
Results
Content validity
Of the 10 Āyurveda experts, who served as judges, 6 of them agreed on all 72 items of the USS. Three (3) judges agreed on 95% and 1 on 85% of the items.
Item difficulty level
This is defined as the presence of a said symptom expressed as the percentage of patients who score positive to that item. 13 Items that had poor difficulty level (absent in >50% of patients) were V5 and V6 (absent in 70%), P24 (absent in 89.5%), and K9 and K10 (absent in 89.5%). Data analysis was carried out after deleting the five items with low difficulty level. Final analysis was done on 67 items of the scale.
Internal consistency
The Cronbach's alpha for V, P and K scales were 0.98, 0.98, and 0.97, respectively. The split-half reliability for V, P, and K scale were 0.97, 0.97, and 0.88, respectively. This shows that the three scales have good internal consistency. 14
Correlations
The subscales correlated significantly (negatively) with each other:
V vs. P = −0.55 p < 0.05, V vs. K = −0.53 p < 0.05, and P vs. K = −0.35 p < 0.05
Typological validation
To determine whether USS differentiated different types of psychoses (paranoid, schizophrenia, unspecified and unspecified nonorganic psychosis), comparisons of mean scores were done. Nearly all patients (16/17) with paranoid schizophrenia scored above the 75th percentile of the vātaja score. Likewise, all patients with schizophrenia, unspecified scored above the 75th percentile of the pittaja score and all patients with unspecified nonorganic psychosis scored above the 75th percentile of the kaphaja score (Table 1).
The percentile scores of all the subjects in three subscales are shown.
Sixteen (16) subjects who scored above 75 percentile of the vātaja score belonged to the paranoid schizophrenia category. Likewise, subjects (n = 7) who scored above 75 percentile of the pittaja score were the patients with schizophrenia not otherwise specified (NOS) scored and subjects (n = 6) who scored above the 75th percentile of the kaphaja score were the patients with psychosis NOS.
SD, standard deviation.
Discussion
This preliminary study has described the development of a 67-item (symptoms of three doshas in unmāda), interview-based, unmāda specific symptom scale (USS) as an instrument for measuring tridosha (vātaja, pittaja, and kaphaja) in psychotic disorders.
Items of USS were generated from Sanskrit statements described in available, authoritative texts of Ayurveda. Content validity from 15 Ayurveda experts, and the opinion of 5 psychiatrists and 5 psychologists were used to refine the items in the scale. The subscales were constructed mainly on the basis of theoretical and psychometric considerations. Primary empirical analysis indicated that the items selected were appropriate to constructs. After item analysis, 67 items were retained. Correlation between vātaja, pittaja, and kaphaja scale scores was negative, suggesting discriminative validity (Table 2). The reliability of subscales of USS was supported by the Cronbach's α coefficient and the split-half analysis. This provided the evidence of stability over items.
Bolding shows that three groups scored differently in different scales.
Kruskal-Wallis test.
Sig, significance.
Applying the scale to the patients with psychosis further validated these three dimensions of doshas. Nearly all patients of (F20.0) paranoid schizophrenia scored above the 75th percentile of the vāta dosha subscale, all patients of (F20.9) schizophrenia unspecified scored above the 75th percentile of the pitta dosha subscale, and all patients of (F29.) unspecified nonorganic psychosis scored above the 75th percentile of the kapha dosha subscale (Table 1).
Though the study provides evidence of categorizing patients on the basis of doshas, diagnosis of three categories was not established by using standard instruments or methods. Additional studies are required to confirm these findings in a similar set of patients and with those blinded to the conventional diagnoses. This tridosha scale has inherent limitations. As its name indicates, it can classify individuals into three subtypes, vāta, pitta, and kapha. To examine its utility in classifying patients with psychotic disorder, we had a sample of 30 patients with functional psychotic disorder attending the outpatient clinic. It is a coincidence that the diagnoses in this unselected sample were only three. Psychotic disorders have been classified into many other types. ICD-10 lists 28 more diagnostic subtypes as can be subsumed under functional psychoses. Therefore, the sample studied cannot be considered representative. Also, the diagnoses were made on ICD-10. 15 (Appendix 2 provides other subcategories of schizophrenia in ICD-10). Yet another widely used diagnostic system, DSM-IV-TR (Diagnostic and Statistical Manual for Psychiatric Diagnosis, 4th edition, text revision) 16 was not used in the study. However, paranoid schizophrenia, schizophrenia unspecified, and unspecified nonorganic psychosis categories of ICD-10 correspond to paranoid schizophrenia (295.30), schizophrenia undifferentiated type (295.9), and psychosis not otherwise specified (298.90), respectively, according to DSM-IV-TR classification. However, diagnosis corresponding to the other 10 categories of DSM-IV-TR, namely: disorganized schizophrenia (295.10), catatonic schizophrenia (295.20), residual type of schizophrenia (295.60), schizophreniform disorder (295.40), schizoaffective disorder (295.70), delusional disorder (297.1), brief psychotic disorder (298.8), shared psychotic disorder (297.3), psychotic disorder due to general medical condition with delusions (293.81), psychotic disorder due to general medical condition with hallucinations (293.82) were not included in the study.
Furthermore, it remains to be seen whether the tridosha scale can classify the corresponding diagnoses in DSM-IV-TR, which is a diagnostic system that has been developed by the American Psychiatric Association (1992). These limitations demand a study on this scale to include a larger and representative sample that has been diagnosed on both diagnostic systems using a structured instrument. Such a study can indicate which diagnostic subtypes are included in each of the tridosha types.
The raters in this study were not blinded to the original diagnosis. There is also a need to demonstrate inter-rater reliability, test–retest reliability, criterion validity, construct validity, and treatment effects on this scale.
In the current psychiatric practice, all three subtypes of psychoses would have received antipsychotic drugs. No guide is available to match each of these diagnoses with a drug. On the other hand, the tridosha profile differentiates these psychoses subtypes. Āyurveda provides different matched management plans for each dosha (viz., diet). The instrument reported here may thus have a potential use in clinical practice to help treatment matching in Ayurveda.
Conclusions
The results here suggest that the Unmāda Specific Scale merits serious further study as an instrument to measure tridoshas symptoms of psychotic disorders.
The scores of patients from three diagnostic categories on each of the subscales (Vāta, Pitta, and Kapha) help in subclassifying unmāda, which has good correspondence with prevailing classification of psychosis. However, this is a preliminary result. Larger studies employing blinded reviewers are necessary to help confirm the validity of the scale and its overall use for diagnosis and treatment of the psychoses. A tridosha measure may point to a better choice of treatment options using Ayurveda.
Footnotes
Acknowledgments
We thank Dr. Srinivas, Director, Spandana Psychiatric Nursing Home, Dr. Kishore, Ph.D. scholar, the National Institute of Mental Health and Neurosciences, and Ayurveda experts in Government Āyurveda College, Bangalore for their support and participation in the study.
Disclosure Statement
No competing financial interests exist.