# Disease classification in Ayurveda (vyadhi-vibhaga) > Canonical version of this page: https://nighantu.ageayurveda.com/concept/disease-classification/ ## How much weight this concept bears This is the one concept in this cluster where the situation is not simply that evidence is absent, because part of the relevant evidence has actually been collected and it is unflattering. The studies that make that sentence true are named here, with their figures, so that the claim can be checked rather than taken on trust. PMID 30653338 (Kessler et al. 2019, J Altern Complement Med) nested a reliability study of 30 knee osteoarthritis patients and four Ayurvedic experts inside a randomised controlled trial. Raw agreement was high, 95 percent consensus agreement on the main diagnosis, and the chance-corrected figures were not: the abstract reports kappas that "largely yielded fair-to-poor inter-rater agreement kappas for central diagnostic aspects such as prakriti and agni (kappa values between 0 and 0.4)", and adds that "agreement on disease-related entities was better than that on constitutional entities". PMID 24191170 (Kurande et al. 2013, Evid Based Complement Alternat Med) had fifteen registered Ayurvedic doctors with 3 to 15 years of experience independently examine twenty healthy subjects, and computed weighted kappa for all 105 doctor pairs: average pairwise kappa was 0.07 for pulse, 0.17 for tongue and 0.28 for prakriti, with the null hypothesis of random rating rejected for only twelve of the 105 pairs on prakriti, one pair on pulse and no pair on tongue. That study's own qualification has to travel with those numbers: it also found "significant evidence against random rating by software and questionnaire use" and concluded that "Prakriti assessment appears reliable when questionnaire and software assessment are used, while other diagnostic methods have room for improvement". The unflattering result is therefore about unaided expert judgement, not about every instrument, and this record does not claim more than that. PMID 23930037, the same group's methodological overview, gives the field's assessment of itself in 2013: "reliability studies in Ayurveda are in the preliminary stage". What none of the three measures is the classificatory axes this record is about. Between them they cover pulse, tongue, prakriti, agni and a final named diagnosis in one condition. No located study reports an agreement figure for assigning a disease to nija or agantu, to sharira or manasa, to a doshaja subtype, to one of the 80, 40 and 20 nanatmaja lists, or to one of Sushruta's seven bala-pravritta types. The classificatory schemes themselves are well documented in the primary texts and can be reported with confidence as doctrine; what remains unmeasured is whether two physicians apply them the same way. On prediction the record states only what it has looked for. A PubMed search on 2026-10-06 for ayurved*[tiab] AND (classification[tiab] OR nosolog*[tiab] OR vyadhi[tiab]) AND (prognos*[tiab] OR predict*[tiab] OR "treatment response"[tiab]) returned 24 records whose titles were screened, and none of them tests whether assigning a disease to an Ayurvedic class predicts treatment response or prognosis better than the symptom description the assignment is derived from. That is a keyword search on title and abstract and would miss a study that does not use those words, so it supports "none located" and not "none exists". Institutionally the picture has moved faster than the evidence, and this project now holds the artefact. The sibling repository file data/enrichment/icd-tm2.json in ayurveda-atlas is a snapshot of the ICD-11 TM2 content at release 2026-01, fetched 2026-09-17, containing 709 entities, 61 blocks and 648 categories. Five blocks bear directly on Ayurvedic disease classification: "Functional elements and humoral derangement patterns (TM2)", "Physical constituent derangement patterns (TM2)", "Power derangement and chronic accumulation patterns (TM2)", "Body constitution and temperament patterns (TM2)" and "External factors disorders (TM2)". Two facts about that chapter matter more than its size. The first is that its statistical standing is narrower than the word classification suggests: PMID 41180267 describes Chapter 26 as "a supplementary, morbidity-only chapter, enabling dual coding of traditional medicine concepts alongside conventional diagnoses", with dual coding "optional", Module I in force from 1 January 2022 and "Module II (Ayurveda, Siddha, Unani) was added in 2025". Morbidity-only means the codes are excluded from mortality statistics; the same source says Chapter 26 carries "the potential to support morbidity statistics", so it is not correct to say TM2 is excluded from morbidity statistics too. The second is that the pattern-level categories were adopted ahead of their own validation: the same source states that "The validity and reliability of Chapter 26 require dedicated study", that "Within-system reliability should be tested through interrater and test–retest assessments among traditional medicine clinicians", and that "validity is not guaranteed: even within the same system, different practitioners may not agree when assigning traditional medicine patterns". A bounded PubMed search on 2026-10-06 for "ICD-11"[tiab] AND ("traditional medicine"[tiab] OR ayurved*[tiab]) returned 19 records, titles screened, none of which reports field-testing or coding-reliability results for the ASU module. NAMASTE (PMID 37541078) stands on the same footing: the initiative report establishes that standardised ASU morbidity codes exist and who built them, not that the categories behind them have been validated or that coders apply them reliably. The constitution-genomics literature does not close the gap, and this record says so explicitly rather than omitting it while neighbouring records cite it. PMID 18782426 (Prasher et al. 2008) and PMID 26511157 (Govindaraj et al. 2015) are the two most-cited molecular results in the field. Both classify persons, not illnesses: Prasher profiled healthy individuals of extreme vata, pitta and kapha type and framed the conclusion as "differential disease predisposition", a hypothesis about susceptibility in the well, while Govindaraj correlated 52 SNPs with prakriti group membership in 262 screened males. Neither assigns a disease to a class, neither tests whether such an assignment is reproducible, and neither tests whether it predicts anything. Transfer to disease classification would require the disease axes to be derived from or validated against constitution, and the texts do not derive them that way: nija and agantu, sharira and manasa, and the doshaja triad classify the disturbance, not the person, and prakriti is the baseline a vikriti is read against rather than the vikriti itself. The one empirical hint about how the two levels behave points away from transfer, since PMID 30653338 found agreement on disease-related entities better than on constitutional entities. A replicated genomic correlate of prakriti would therefore still license no claim about vyadhi-vibhaga. There is also a textual problem that no amount of modern work addresses: Charaka and Sushruta each give more than one classification, the schemes are structurally incompatible, both authors then subordinate everything to the doshas on grounds that are stated in unfalsifiable terms, and the identification of the enumerated Sanskrit disease names with modern entities was flagged as unreliable by the public-domain translators themselves. Ayurveda does not use one disease taxonomy but several overlapping axes applied to the same illness at once. Charaka sorts diseases by origin into nija (constitutional, arising from vitiated dosha) and agantu (accidental, arising from trauma, poison, spirits, weapons or curses); by seat into sharira (bodily) and manasa (mental); by dosha into vataja, pittaja and kaphaja, with stated counts of 80, 40 and 20 nanatmaja disorders; and by the numeric enumerations of the Ashtodariya chapter, which tallies named diseases by the number of their varieties. Sushruta gives a four-way origin classification (agantuka, sharira, manasa, svabhavika) in the opening chapter and, in the chapter on the classification of diseases, a seven-way scheme running from parental-seed defect through congenital, doshic, traumatic, seasonal, providential and spontaneous causes, which he maps onto the older three-fold adhyatmika, adhibhautika and adhidaivika division. Across all of these Charaka asserts a unifying claim: however classified, constitutional disease never escapes vata, pitta and kapha. ## At a glance | Field | Value | | --- | --- | | Subject area | How disease is held to arise | | Domain, as recorded | etiology | | Contested questions | 6 | | Stated as not known | 13 | | Classical citations | 12, 10 quoting the text | | Research papers assessed | 10 | | Citations checked | 2026-10-08 | ## What is contested, and by whom These are the questions on which the sources genuinely disagree. Each position is given as its source states it. No position is presented as the settled answer, because none of them is. ### How many classifications are there, and which is primary? This is contested. - Charaka Sutra 11.44 gives three (physical, accidental, mental). - Charaka Sutra 20.2 gives four (accidental plus the three doshaja) and then immediately gives two by cause and two by substratum. - Sushruta Sutra 1 gives four (agantuka, sharira, manasa, svabhavika). - Sushruta Sutra 24 gives seven bala-pravritta types mapped onto three older ones. - The texts present these as complementary; they are not structurally reconcilable, and which is treated as primary varies by author and by clinical context. ### Is the nija/agantu distinction stable? This is contested. - It is presented as the fundamental causal division. - Charaka himself states in Ashtodariya that each converts into the other on aggravation, and in Maharoga that all four classes follow one another when aggravated, which means the distinction holds only early in the course. - Since both classes are said to share the same three inducing causes (Sutra 20.9), the distinction rests on the proximate agent alone. ### Are the nanatmaja counts of 80, 40 and 20 a taxonomy or a mnemonic? This is contested. - Traditional exposition treats the lists as exhaustive enumerations of dosha-specific disorders. - Kaviratna noted that the lists differ between printed editions and that he compiled his from two texts, and that identification of the names with modern entities may not be accurate in every instance. - Some scholars read the numeric enumerations of Ashtodariya and Maharoga as pedagogical devices for memorisation rather than as claims about the number of diseases. ### Should Ayurvedic categories be mapped into ICD, or kept separate? This is contested. - The NAMASTE and ICD-11 traditional medicine module position is that standardised coding enables morbidity statistics, insurance and research on traditional medicine. - The ethnopharmacology methodological position (Staub and colleagues 2015) is that forcing indigenous categories into a clinical-examination-based biomedical taxonomy destroys exactly the categories that carry the local explanatory content, and that patient-perception-based systems such as ICPC map better. - The practitioner-side critique visible in the ICD-11 Chapter 26 revision literature is that even a harmonised traditional classification satisfies none of the traditions it merges. ### Does a classification that always terminates in the doshas classify anything? This is contested. - The classical position, stated in Charaka's bird-and-shadow verse and in Sushruta's fatherhood-of-all-maladies verse, is that doshic reduction is the deepest truth about disease. - The criterion Sushruta actually offers, that doshic symptoms can be detected in a disease of whatsoever type, is not falsifiable as stated, so the reduction cannot be tested by the means the text provides. - Charaka's own reservation, that agantu disease is initially distinguished therefrom, preserves a class that is not doshic in origin. ### Does the constitution-genomics line bear on disease classification at all? This is contested. - The ayurgenomics position, as stated in Prasher et al. 2008 (PMID 18782426), is that Ayurvedic phenotypic classification "allows us to uncover genes that may contribute to system level differences in normal individuals which could lead to differential disease predisposition", and in Govindaraj et al. 2015 (PMID 26511157) that "the phenotypic classification of India's traditional medicine has a genetic basis". Both claims are about prakriti. - This record's position is that the claims do not transfer to vyadhi-vibhaga, because the classified object differs: prakriti classifies the person and is assessed in the healthy, while the axes in this record classify the illness. Neither study assigns a disease to a class or tests such an assignment for reproducibility or prediction. - The one relevant empirical observation runs against transfer in the direction that matters: Kessler et al. 2019 (PMID 30653338) reported that "agreement on disease-related entities was better than that on constitutional entities", so the two levels are not interchangeable even in reliability behaviour. - A position this record does not hold but cannot rule out: if a doshaja disease assignment were shown to be derived from or constrained by prakriti, the genomics work would become relevant. No located study establishes that derivation. ## What is not known Stated rather than left out. Where a negative has actually been searched for rather than assumed, the entry says so. - Whether two independent Ayurvedic physicians agree on nija versus agantu, on doshaja subtype, or on Sushruta's seven bala-pravritta types. The located reliability studies test other things: PMID 30653338 tested prakriti and agni among other central diagnostic aspects and returned kappa between 0 and 0.4, and PMID 24191170 tested pulse, tongue and prakriti and returned average pairwise kappa of 0.07, 0.17 and 0.28. No agreement figure for any of the classificatory axes in this record has been located. - Whether any Ayurvedic disease classification predicts treatment response or prognosis better than the symptom picture alone. No study establishing this has been located. The search behind that statement, run on 2026-10-06, was PubMed for ayurved*[tiab] AND (classification[tiab] OR nosolog*[tiab] OR vyadhi[tiab]) AND (prognos*[tiab] OR predict*[tiab] OR "treatment response"[tiab]), 24 records returned and titles screened. It is a keyword search in one database and would miss a study phrased differently, so this is a located-nothing statement and not a non-existence claim. - Whether the ICD-11 traditional medicine module covering Ayurveda, Siddha and Unani has been field-tested for coding reliability. No such results have been located, and WHO's own Bulletin (PMID 41180267) says the work is still to be done rather than reporting it: "The validity and reliability of Chapter 26 require dedicated study", with interrater and test-retest assessment named as what is needed. The searches behind the negative, both run on 2026-10-06, were PubMed for "ICD-11"[tiab] AND ("traditional medicine"[tiab] OR ayurved*[tiab]), 19 records with titles screened, and ("TM2" OR "Module 2" OR "Module II") AND "ICD-11", which returned nothing. - How the TM2 pattern blocks held in this project's snapshot map onto the classical axes in this record. The snapshot carries no definition text for any of its 709 entities and no entity in it mentions Ayurveda, so the correspondence between, for example, "Functional elements and humoral derangement patterns (TM2)" and Charaka's doshaja triad cannot be read off the data. No crosswalk is asserted here, and the BY-ND licence on TM2 would in any case forbid an adapted one. - How many TM2 categories there are. This project's source register and Indian government press material both give 529 disorder categories, while the 2026-01 snapshot held in the sibling repository contains 648 categories and 61 blocks. The figures have not been reconciled; the difference may be release drift, a difference between disorder and pattern counts, or a counting convention, and none of those has been established. - How many of the 80 vataja, 40 pittaja and 20 kaphaja named disorders can be identified with any modern entity with agreement between scholars. Kaviratna flagged his own identifications as possibly inaccurate. - Whether the readings of the nanatmaja lists differ enough between manuscript traditions to change the counts, given that Kaviratna compiled his list from two printed texts that differed. - How Kaviratna's verse numbering in the passages quoted here corresponds to standard Trikamji numbering. Every verse number in the classical array is labelled "(Kaviratna)" for that reason and no Trikamji equivalence is asserted. The correspondence has not been checked against a Sanskrit edition for Sutrasthana 11, 19 and 20 or Vimanasthana 1, and Kaviratna is known in this project to run one behind Trikamji elsewhere in the text. - What clinical consequence, if any, follows from assigning a disease to Sushruta's adi-bala (parental seed) versus janma-bala (congenital) class, which is the closest classical approach to a hereditary versus developmental distinction. - Whether the manasa class as defined, arising from non-attainment of the desired and accession of the disliked, picks out anything distinguishable from the psychiatric categories now used in the same settings. - Whether practitioners in different Indian states and schools use the same classification in the same way. No multi-centre concordance study has been located; PMID 23930037, the field's own methodological overview, described Ayurvedic reliability work as being "in the preliminary stage" in 2013 and set out the designs that would be needed. - How svabhavika conditions such as hunger, thirst and sleep are meant to function as a disease class rather than as normal physiology, and what makes their untimely occurrence pathological in operational terms. - Whether anything in the constitution-genomics literature bears on disease classification. The record's position is that it does not, for the reasons given in evidenceState and in the contested entry on the question, but that position is an argument from what the two studies did, not a result. No study testing whether prakriti constrains or predicts a disease-class assignment has been located. ## The terms this concept is built from | Term | What it names | | --- | --- | | nija and agantu | Charaka's primary origin axis: constitutional disease arising from one's own vitiated doshas, versus accidental disease from an external agent. The two are said to convert into each other when aggravated | | sharira and manasa | Seat axis: body and mind as the two substrata of disease. Charaka gives mental disease its own etiology, namely non-attainment of what is desired and accession of what is disliked | | doshaja triad | Vataja, pittaja and kaphaja disease, with enumerated counts of 80, 40 and 20 nanatmaja (dosha-specific) disorders in Charaka Sutra 20 | | nanatmaja and samanyaja | Disorders arising from a single dosha alone versus those arising from any dosha or combination. Charaka's Ashtodariya chapter enumerates the general ones and Maharoga the dosha-specific ones | | Sushruta's four types | Agantuka (traumatic or extraneous), sharira (bodily, from diet, blood or humours), manasa (mental, from anger, grief, fear, joy, envy, pride, greed, lust), svabhavika (natural, including hunger, thirst, decrepitude, sleep, death) | | Sushruta's seven bala-pravritta types | Adi-bala (parental seed defect), janma-bala (congenital, from maternal conduct in gestation), dosha-bala (humoral), sanghata-bala (traumatic), kala-bala (seasonal), daiva-bala (providential, including epidemic and sporadic subtypes), svabhava-bala (spontaneous, timely or untimely) | | adhyatmika, adhibhautika, adhidaivika | The older three-fold division of affliction by locus of cause, onto which Sushruta maps his seven types | | sadhya and asadhya | Prognostic axis of curable and incurable, cited at Charaka Vimana 6/3 alongside the somatic and psychological division | | [nidana-panchaka](https://nighantu.ageayurveda.com/lexicon/nidana-panchaka/) | The five diagnostic limbs (nidana, purvarupa, rupa, upashaya, samprapti) which classify the evidence for a disease rather than the disease itself; systematised in Madhava Nidana chapter 1 | ## The classical passages Each citation gives its address so a reader can check it in a printed edition, our account of what it establishes, and the result of reading it against its source. ### Charaka Samhita, Sutra Sthana, Lesson XI (Tistraishaniya), verse 44 in Kaviratna's numbering Sutra Sthana 11:44 (Kaviratna), checked against the source > Diseases are of three kinds, viz., Physical, Accidental, and Mental. Physical diseases are those that arise from some abnormal condition of the body. Accidental are those that arise from the action of spirits, of poison, of wind, of fire, of acts of violence done to the body, such as beating, etc. Mental diseases are those that arise from non-attainment of objects desired or coveted, or from accession of those that are disliked. tr. Avinash Chandra Kaviratna (1896-1913), public domain A three-way classification with each class given its own etiology, and a definition of mental disease in terms of desire and aversion rather than of dosha. Source: https://archive.org/stream/BIUSante_47357/BIUSante_47357_djvu.txt ### Charaka Samhita, Sutra Sthana, Lesson XIX (Ashtodariya), verse 47 in Kaviratna's numbering Sutra Sthana 19:47 (Kaviratna), checked against the source > All constitutional diseases that affect oneself never spring from any other source than one's wind, bile, and phlegm. Even as a bird, roving wheresoever it likes, can never transgress its own shadow, after the same manner all the diseases that afflict a person, born of the disorders of his own constituent elements, can never transgress his own wind, bile, and phlegm. Intelligent physicians observing the especial seat or place, the indications, and the exciting causes, of wind, bile, and phlegm, refer all diseases as caused by disorders of wind, or bile, or phlegm. tr. Avinash Chandra Kaviratna (1896-1913), public domain The unifying doctrine underneath all the classificatory axes: every nija disease reduces to vata, pitta and kapha. The bird-and-shadow simile is the standard proof-text for this. Source: https://archive.org/stream/BIUSante_47357/BIUSante_47357_djvu.txt ### Charaka Samhita, Sutra Sthana, Lesson XIX (Ashtodariya), closing verses Sutra Sthana 19:closing verses 1-2 of the chapter (Kaviratna), checked against the source > All the numerous diseases that appear in the body, arising from the disorders of one's constituent elements, are not separable from (disorders of) bile, phlegm and wind. All diseases, however, that are accidental, are distinguished therefrom. The Accidental becoming aggravated, follows the Constitutional disease; similarly the Constitutional, becoming aggravated, follows the Accidental. Hence, ascertaining properly the adjunct and the cause, the work of treatment should be commenced. tr. Avinash Chandra Kaviratna (1896-1913), public domain That the nija/agantu boundary is permeable in both directions once disease is aggravated, which qualifies the classification substantially. Source: https://archive.org/stream/BIUSante_47357/BIUSante_47357_djvu.txt ### Charaka Samhita, Sutra Sthana, Lesson XX (Maharoga), verses 2-6 in Kaviratna's numbering Sutra Sthana 20:2-6 (Kaviratna), checked against the source > The diseases that occur are of four kinds: They are 1. Accidental, 2. those born of wind, 3. those born of bile, and 4. those born of phlegm. Though distributed into four kinds, their character as disease is of one kind, in consequence of the common element being present in all of them of producing affliction or pain. In consequence, again, of diseases being distributed into two kinds, viz., Accidental and Constitutional, their causes are of two kinds. In consequence, again, of the distinction between mind and body, the substrata of disease are of two kinds. The modifications, again, of those four are innumerable, in consequence of the differences in respect of causes, seats, symptoms, areas, and varieties, being innumerable. tr. Avinash Chandra Kaviratna (1896-1913), public domain The explicit statement that several classifications (four-way, two-way by cause, two-way by substratum) are applied simultaneously to the same set of diseases, and that the number of resulting varieties is innumerable. Source: https://archive.org/stream/BIUSante_47357/BIUSante_47357_djvu.txt ### Charaka Samhita, Sutra Sthana, Lesson XX (Maharoga), verse 9 in Kaviratna's numbering Sutra Sthana 20:9 (Kaviratna), checked against the source > The inducing causes of both Accidental and Constitutional diseases are injudicious correlation of objects with the senses, want of wisdom or foresight, and time. tr. Avinash Chandra Kaviratna (1896-1913), public domain That the three general causes cut across the classification rather than aligning with it: all classes of disease share the same three inducing causes. Source: https://archive.org/stream/BIUSante_47357/BIUSante_47357_djvu.txt ### Charaka Samhita, Sutra Sthana, Lesson XX (Maharoga), verse 18 in Kaviratna's numbering Sutra Sthana 20:18 (Kaviratna), checked against the source > The diseases of Wind are of Eighty kinds; those of Bile are of forty kinds; those of Phlegm are of twenty kinds. tr. Avinash Chandra Kaviratna (1896-1913), public domain The canonical nanatmaja counts of 80, 40 and 20. The numbered lists that follow are of Sanskrit disease names whose identification with modern entities Kaviratna himself flagged as uncertain. Source: https://archive.org/stream/BIUSante_47357/BIUSante_47357_djvu.txt ### Charaka Samhita, Vimana Sthana, Lesson I, verse 2 in Kaviratna's numbering Vimana Sthana 1:2 (Kaviratna), checked against the source > Without doubt, having carefully noticed the particulars relating to inducing causes, premonitory indications, symptoms, applicability of medicine, diet and course of conduct, the varieties of diseases, the predominance of wind or bile or phlegm in diseases, the class under which the disease falls, the precise measures in which the faults have been excited, strength, and age or season, the physician, with concentrated mind, should specially attend to the knowledge (analysis) of the tastes, articles, faults, derangements, medicine, country, season, strength, body, diet, predominance of constituent elements (of the body, such as skin, blood, etc.), assimilability, mind, constitution, and age. tr. Avinash Chandra Kaviratna (1896-1913), public domain The nidana-panchaka elements in Charaka: nimitta (inducing cause), purvarupa (premonitory indication), rupa (symptom) and upashaya (applicability of medicine, diet and conduct), glossed individually in Kaviratna's note to this verse. Source: https://archive.org/stream/BIUSante_47357/BIUSante_47357_djvu.txt ### Sushruta Samhita, Sutra Sthana, Chapter I Sutra Sthana 1, checked against the source > The Purusha (man) is the receptacle of any particular disease, and that which proves a source of torment or pain to him, is denominated as a disease. There are four different types of disease such as, Traumatic or of extraneous origin (Agantuka), Bodily (Sharira), Mental (Manasa) and Natural (Svabhavika). A disease due to an extraneous blow or hurt is called Agantuka. Diseases due to irregularities in food or drink, or incidental to a deranged state of the blood, or of the bodily humours acting either singly or in concert, are called Sharira. Excessive anger, grief, fear, joy, despondency, envy, misery, pride, greed, lust, desire, malice, etc. are included within the category of mental (Manasa) distempers; whereas hunger, thirst, decrepitude, imbecility, death, sleep, etc. are called the natural (Svabhavika) derangements of the body. tr. Kunja Lal Bhishagratna (1907-1916), public domain Sushruta's four-way origin classification and his definition of disease, which differs from Charaka's by adding svabhavika as a fourth class and by counting hunger, thirst and sleep as derangements. Source: https://archive.org/stream/englishtranslati01susruoft/englishtranslati01susruoft_djvu.txt ### Sushruta Samhita, Sutra Sthana, Chapter XXIV (Vyadhi-samuddeshiya) Sutra Sthana 24, checked against the source > It has been stated before that anything that afflicts either the body or the living personality-self, or both, is called disease. This pain or affliction may be either physical in its character (Adhyatmikam), or due to any disturbance in the physical environments of a man (Adhibhautikam), or to the acts of God (Adhidaivikam) etc. This three-fold pain may be ultimately transformed into any of the seven kinds of diseases such as, the Adi-vala-pravritta, Janma-vala-pravritta, Dosha-vala-pravritta, Sanghata-vala-pravritta, Kala-vala-pravritta, Daiva-vala-pravritta and Svabhava-vala-pravritta. tr. Kunja Lal Bhishagratna (1907-1916), public domain Sushruta's seven-way causal classification and its mapping onto the older three-fold division of affliction. Note that this is a third scheme, incompatible in structure with both his own Chapter I four-way scheme and Charaka's. Source: https://archive.org/stream/englishtranslati01susruoft/englishtranslati01susruoft_djvu.txt ### Sushruta Samhita, Sutra Sthana, Chapter XXIV, closing on the primacy of the doshas Sutra Sthana 24, checked against the source > The deranged bodily humours such as, Vayu, Pittam and Kapham should be looked upon as the primary sources of all diseases, inasmuch as symptoms characteristic of each of them may be detected in the case of a disease of whatsoever type, (which usually abates with their corresponding subsidence), and also because the Shastras have ascribed to them the fatherhood of all maladies that assail the human frame. tr. Kunja Lal Bhishagratna (1907-1916), public domain That Sushruta, having given a seven-way classification, then subordinates all of it to the doshas, matching Charaka's bird-and-shadow claim. The argument offered is that doshic symptoms can be detected in any disease type, which is an unfalsifiable criterion as stated. Source: https://archive.org/stream/englishtranslati01susruoft/englishtranslati01susruoft_djvu.txt ### Charak Samhita New Edition (carakasamhitaonline.com), entry 'Roga' Cites Charaka Vimana Sthana 6/3 for both the somatic/psychological division and the curable/incurable division. Located, not quoted. Source: https://www.carakasamhitaonline.com/index.php/Roga ### Madhava Nidana, chapter 1, systematises the nidana-panchaka as the five limbs of diagnosis and is the standard reference work for Ayurvedic disease description The classical framework for diagnostic evidence. No public-domain English translation exists, so nothing is quoted and no verse is asserted beyond the chapter. The Bulletin of the Indian Institute of History of Medicine carried two short papers specifically on the definition of nidana-panchaka in this chapter (PMID 12585278 and PMID 12572581). ## What the research actually shows Every paper located for this concept, with what it measured and what it does and does not establish. The assessment is ours. ### Reliability of Ayurvedic Diagnosis for Knee Osteoarthritis Patients: A Nested Diagnostic Study Within a Randomized Controlled Trial. J Altern Complement Med · 2019 · PMID 30653338 (https://pubmed.ncbi.nlm.nih.gov/30653338/) Four Ayurvedic experts independently diagnosing 30 knee osteoarthritis patients, nested in an RCT, with a nominal group technique as the consensus procedure. High raw agreement on main diagnostic entities (95 percent consensus agreement on the main diagnosis) was not reflected in the kappa values, which "largely yielded fair-to-poor inter-rater agreement kappas for central diagnostic aspects such as prakriti and agni (kappa values between 0 and 0.4)". The authors also note that "agreement on disease-related entities was better than that on constitutional entities". **How much it shows.** The single most useful empirical study for this concept. Properly designed reliability study nested in an RCT, international collaboration, appropriate statistics. Limits: 30 patients, one disease, four raters, a setting where a consensus procedure was available. The gap between raw agreement and kappa is the key finding and is often misreported. ### Interrater reliability of diagnostic methods in traditional Indian ayurvedic medicine. Evid Based Complement Alternat Med · 2013 · PMID 24191170 (https://pubmed.ncbi.nlm.nih.gov/24191170/) Fifteen registered Ayurvedic doctors with 3-15 years of experience independently examined twenty healthy subjects; weighted kappa was computed for all 105 doctor pairs. Average pairwise kappa was 0.07 for pulse (nadi), 0.17 for tongue (jivha) and 0.28 for body constitution (prakriti). The null hypothesis of random rating was rejected for only twelve of the 105 pairs on prakriti, one pair on pulse and no pair on tongue. The abstract's own qualification travels with those figures: there was "significant evidence against random rating by software and questionnaire use", and the authors conclude that "Prakriti assessment appears reliable when questionnaire and software assessment are used, while other diagnostic methods have room for improvement". So the finding is against unaided expert judgement, not against every instrument. **How much it shows.** The most direct chance-corrected reliability figures in the Ayurvedic literature and correctly analysed. Small (20 subjects), healthy volunteers only, single site, and confined to pulse, tongue and constitution. It measures none of the classificatory axes this record is about: it does not test nija versus agantu, sharira versus manasa, doshaja subtype, the nanatmaja lists, or Sushruta's seven bala-pravritta types. Cited here for what reliability has been measured and for the boundary of that measurement, not as evidence about disease classification. ### Reliability studies of diagnostic methods in Indian traditional Ayurveda medicine: An overview. J Ayurveda Integr Med · 2013 · PMID 23930037 (https://pubmed.ncbi.nlm.nih.gov/23930037/) The same group's methodological overview of reliability study design for Ayurvedic diagnosis. It states the field's own position on itself: "The investigation of the reliability of traditional Chinese, Japanese and Sasang medicine diagnoses is in the formative stage. However, reliability studies in Ayurveda are in the preliminary stage." It sets out the designs and statistics that future studies would need, and names classification (explicitly prakriti classification) as one of the places where reliability is crucial. **How much it shows.** A methodological overview, not primary evidence, and written by the group that then produced the reliability data, so it is the field's self-assessment rather than an independent audit. Cited for that self-assessment and for the fact that no reliability programme covering disease classification had been laid out as of 2013. ### International Statistical Classification of Diseases and data gaps in traditional medicine. Bull World Health Organ · 2025 · PMID 41180267 (https://pubmed.ncbi.nlm.nih.gov/41180267/) Bulletin of the WHO perspective on the institutional standing of ICD-11 Chapter 26 and on what traditional medicine data the ICD does and does not capture. Four statements are quoted in this record: Chapter 26 "is a supplementary, morbidity-only chapter, enabling dual coding of traditional medicine concepts alongside conventional diagnoses"; the chapter "enables optional dual coding of traditional medicine disorders and patterns alongside conventional diagnoses" with "the potential to support morbidity statistics"; "ICD-11 took effect on 1 January 2022 with Module I (East-Asian systems)" and "Module II (Ayurveda, Siddha, Unani) was added in 2025"; and on validation, "The validity and reliability of Chapter 26 require dedicated study", "Within-system reliability should be tested through interrater and test–retest assessments among traditional medicine clinicians", and "validity is not guaranteed: even within the same system, different practitioners may not agree when assigning traditional medicine patterns; and across systems, the underlying concepts may not align. Validity must be demonstrated before multinational comparisons are relied upon." **How much it shows.** Policy perspective piece, not research. No abstract, no data, and its authors are writing from inside the institution whose product they are describing. It is nonetheless the best available peer-reviewed address for the statistical standing of Chapter 26 and for the statement that its validity and reliability have yet to be studied, which is why it now carries those facts instead of government press material. Not cited as evidence about Ayurvedic nosology itself. ### Initiatives: National Ayush Morbidity and Standardized Terminologies Electronic (NAMASTE) Portal. J Ayurveda Integr Med · 2023 · PMID 37541078 (https://pubmed.ncbi.nlm.nih.gov/37541078/) Describes the National Ayush Morbidity and Standardized Terminologies Electronic (NAMASTE) portal, India's standardised terminology and morbidity coding system for Ayurveda, Siddha and Unani. **How much it shows.** Short initiative report, no abstract, no evaluation data. Establishes that the coding system exists and who built it. It does not establish that codes are applied reliably or that the classification has been validated. NAMASTE is treated here as a comparator, not as a source of definitions. ### A Proposed Revision of the International Classification of Diseases, 11th Revision, Chapter 26. Integr Cancer Ther · 2020 · PMID 32070150 (https://pubmed.ncbi.nlm.nih.gov/32070150/) Argues that ICD-11 Chapter 26, the traditional medicine chapter, is a union set of harmonised Chinese, Japanese and Korean pattern classifications and proposes a structural revision. **How much it shows.** Single-author proposal paper, not empirical. Included because it documents that even the better-established traditional medicine module in ICD-11 is contested in structure and terminology by practitioners of the traditions it encodes. Note that Chapter 26 Module 1 covers the East Asian traditions; the Ayurveda, Siddha and Unani content sits in Module 2, released later. ### Classifying diseases and remedies in ethnomedicine and ethnopharmacology. J Ethnopharmacol · 2015 · PMID 26342522 (https://pubmed.ncbi.nlm.nih.gov/26342522/) Systematic comparison of ICD, the Economic Botany Data Collection Standard and ICPC for classifying disease in ethnomedical systems, concluding that ICD's reliance on clinical-examination-based categories makes it a poor fit and that ICPC, being built on patient-reported categories, maps better onto ethnomedical reality. **How much it shows.** Methodological analysis rather than an empirical study, but a careful and directly relevant one. It makes explicit the general problem that the Ayurvedic case instantiates: translating an indigenous nosology into a biomedical one loses the categories that did the original explanatory work. ### Definition of Nidana Pancaka (first chapter of Madhavanidana). Bull Indian Inst Hist Med Hyderabad · 1999 · PMID 12585278 (https://pubmed.ncbi.nlm.nih.gov/12585278/) Short historical paper on the definition of nidana-panchaka in the first chapter of Madhava Nidana. **How much it shows.** Historical and philological note in a low-circulation journal, no abstract available through PubMed, no DOI. Cited for textual provenance only. ### Whole genome expression and biochemical correlates of extreme constitutional types defined in Ayurveda. J Transl Med · 2008 · PMID 18782426 (https://pubmed.ncbi.nlm.nih.gov/18782426/) Genome-wide peripheral blood expression profiling plus biochemical and haematological measurement in healthy individuals selected as extreme vata, pitta and kapha types in an Indian population of Indo-European origin. Reports differences between the groups and enrichment of differentially expressed genes in transport, regulation of cyclin-dependent protein kinase activity, immune response and regulation of blood coagulation. The conclusion is framed as susceptibility, not diagnosis: the phenotyping "allows us to uncover genes that may contribute to system level differences in normal individuals which could lead to differential disease predisposition". **How much it shows.** A genuine molecular study and the foundation of the ayurgenomics line, but small, single-population, with prakriti assignment by the hypothesis-forming team, a multiple-comparison burden inherent in microarray work, and no independent replication of the expression signature. Carried here only to support the record's explicit statement that the constitution-genomics line does not transfer to disease classification: the subjects are healthy, the classified object is the person and not the illness, and no disease is assigned to a class anywhere in the study. It is a comparator, not evidence about vyadhi-vibhaga. ### Genome-wide analysis correlates Ayurveda Prakriti. Sci Rep · 2015 · PMID 26511157 (https://pubmed.ncbi.nlm.nih.gov/26511157/) Genome-wide SNP analysis (Affymetrix 6.0) of 262 well-classified male individuals screened from 3,416 subjects across the three prakritis. 52 SNPs differed at p <= 1x10^-5 after 10^6 permutations and control for stratification; principal component analysis on those SNPs separated the 262 individuals into their prakriti groups irrespective of ancestry; the finding was then checked against 297 Indian population samples of known ancestry; PGM1 was reported as correlating with the pitta phenotype. **How much it shows.** Methodologically the strongest study in the area: large screening, permutation testing and explicit stratification control. Caveats that bear on how far it can be pushed: males only, extreme-phenotype selection (262 of 3,416), the PCA separation is computed on SNPs selected in the same sample, which inflates apparent classification power, the single-gene narrative link to a Charaka phenotype description is interpretation rather than result, and the 52-SNP set has not been independently replicated in a separate prakriti cohort. Like Prasher 2008, it classifies persons and not diseases, and is carried here as the comparator behind the record's non-transfer statement. ## Citation Age Ayurveda Nighantu, "Disease classification in Ayurveda (vyadhi-vibhaga)", https://nighantu.ageayurveda.com/concept/disease-classification/, CC BY 4.0. Whole concept set: https://nighantu.ageayurveda.com/concepts.json and https://nighantu.ageayurveda.com/concepts.csv. Educational reference only. This page describes what a body of theory claims and how well it is evidenced. Nothing here is advice, and nothing here describes how any condition should be treated.