Homeopathic Repertorization and Rubric Selection
Educational and safety note: This guide describes a professional homeopathic documentation and analysis method. It does not establish clinical efficacy, provide medical advice or replace diagnosis, emergency assessment, referral, informed consent, applicable law or the practitioner’s professional judgment.
Quick answer: how should a homeopath repertorize a case?
A defensible homeopathic repertorization follows a traceable chain: preserve the source statement, clarify its meaning and context, decide whether it is characteristic, translate it into one or more candidate rubrics, verify those rubrics in the repertory, compare the result under more than one reasonable weighting, and then read the leading remedies in materia medica before deciding what the analysis does—or does not—support.
Do not begin by searching every sentence. Do not end with the highest score. A repertory is an index to remedy information. The chart narrows and tests a differential; it does not prescribe automatically.
Key takeaways
- Keep source language, clarification, interpretation and rubric in separate fields or lines.
- A small, justified set is usually more reviewable than a large list of overlapping rubrics.
- “Characteristic” means distinctive in this case, not merely intense, dramatic or rare in isolation.
- Read the full rubric path and nearby rubrics; a keyword match can conceal a different meaning.
- Record repertory edition or database, rubric path, weighting and exclusions.
- Verify the emerging differential in materia medica and against contradictory case information.
- Treat AI suggestions as candidates to inspect, edit or reject.
What repertorization can and cannot do
Repertories organize indexed symptoms and remedy entries so a practitioner can compare coverage across a selected totality. Different repertories embody different structures, terminology, editions and editorial decisions. A search result therefore depends on both the case representation and the source being searched.
Repertorization can help you:
- locate rubric language that corresponds to a clarified symptom;
- compare which remedies occur across several selected rubrics;
- make weighting and exclusions visible;
- notice remedies or alternatives that deserve materia medica review;
- reproduce an earlier analysis at follow-up or peer review.
It cannot determine whether the interview was accurate, whether a symptom is clinically safe to analyze rather than refer, whether a rubric faithfully represents the person, or whether a remedy is appropriate. Those remain practitioner responsibilities.
Our guide to homeopathic case taking and follow-up documentation covers the record that should exist before repertorization. The history of homeopathic repertorization explains why repertories are retrieval tools rather than self-sufficient decision engines.
The SOURCE workflow: seven steps from narrative to verification
Use SOURCE as a repeatable sequence:
- S — Save the source statement
- O — Obtain clarification
- U — Understand and rank the symptom
- R — Retrieve candidate rubrics
- C — Check wording, hierarchy and source
- E — Evaluate the chart and alternatives
- Verify — Return to materia medica and the whole case
1. Save the source statement
Copy the shortest useful verbatim phrase from the interview and retain where it came from: patient, parent, partner, practitioner observation or prior record. Add the date and relevant episode. A rubric should never become the only surviving version of the symptom.
For example:
Patient statement: “Before the meeting I keep checking the clock, and my hands turn cold.”
That sentence contains sequence, circumstance, behaviour and a physical concomitant. Searching only “anxiety” would discard most of it.
2. Obtain clarification without supplying the answer
Clarify what the person means before opening the repertory. Useful questions establish:
- onset and sequence;
- exact trigger or circumstance;
- sensation and location;
- frequency, intensity and functional effect;
- aggravating or relieving factors;
- accompanying features;
- whether the pattern is usual, new or exceptional for that person.
Avoid offering repertory words for the patient to endorse. If a person says “I feel trapped,” that may describe a physical sensation, a relationship, a work situation or a metaphor. The same word can lead to unrelated rubric families.
3. Understand and rank the symptom
Hahnemann’s Organon §153 gives particular attention to symptoms described as striking, singular, uncommon and peculiar, while warning that vague, general symptoms provide less discrimination. This is a historical methodological rule within homeopathy, not evidence that every unusual detail is reliable or therapeutically decisive.
Ask four questions before including a symptom:
| Test | Question | If the answer is weak |
|---|---|---|
| Reliability | Was it reported consistently, observed, or clearly clarified? | Mark uncertain or omit |
| Relevance | Does it belong to the current case and analysis? | Keep in notes, not totality |
| Individuality | Does it distinguish this person’s pattern from a generic complaint? | Use low weight if needed |
| Independence | Does it add information, or duplicate another selected rubric? | Combine or choose one |
Intensity alone is not individuality. A severe but common defining feature may matter medically and for monitoring while contributing little discrimination in repertorization. Conversely, novelty alone is not enough: a curious statement that was led, mistranslated or mentioned once without clarification is weak evidence.
4. Retrieve more than one candidate rubric
Translate meaning, not just vocabulary. Search synonyms and likely sections, then build a short candidate list. Preserve uncertainty rather than forcing an exact-looking match.
Worked documentation example (fictional and non-prescriptive)
| Layer | Record |
|---|---|
| Source phrase | “Before the meeting I keep checking the clock, and my hands turn cold.” |
| Clarification | Starts about one hour before presentations; repeated clock-checking; cold palms; not present before social visits |
| Interpretation | Anticipatory context appears stronger than undifferentiated anxiety; cold hands may be a concomitant |
| Candidate search | Anticipation; anxiety before an engagement; checking/repetition if independently supported; coldness of hands |
| Uncertainty | “Checking” may simply measure time, not compulsive behaviour |
| Decision | Use the well-supported anticipatory rubric; retain cold hands only if repertory wording and case relevance check out; exclude an unsupported compulsion rubric |
The example demonstrates representation, not a remedy choice. It deliberately stops before naming remedies.
5. Check the rubric, not only the search result
Before adding a rubric to the working totality, perform a five-point rubric check:
- Exact path: read the chapter, parent rubric, subrubric and qualifiers.
- Meaning: confirm that the repertory term expresses the clarified case, not merely a shared word.
- Neighbourhood: inspect adjacent rubrics and cross-references for a better fit.
- Source: identify repertory, edition or database and, where available, the provenance of additions.
- Content: review remedy entries and grades in context; notice unexpectedly narrow or very broad coverage.
This matters because repertory data are edited over time. The 2014 National Guidelines for Translation of Proving Raw Data into Repertory Rubrics were created specifically to improve the quality and reliability of converting raw proving information into rubric form. Their subject is repertory construction rather than day-to-day case taking, but the underlying lesson applies: translation into a rubric is an editorial act, not a neutral copy operation.
If two repertories use similar labels differently, document which one you used. Do not silently combine grades from unlike editions as though they were one homogeneous source.
Build the totality without double-counting
A common software-era error is selecting several rubrics generated from one sentence and treating them as independent confirmations. This can amplify one idea simply because it has many searchable formulations.
Use a one-fact, one-vote check:
- highlight rubrics derived from the same source phrase;
- identify parent/child duplication;
- identify a general rubric plus its more precise subrubric;
- keep the version that best preserves the clarified meaning;
- retain alternatives in notes rather than scoring all of them.
Weighting decision model
| Tier | Use | Typical treatment in analysis |
|---|---|---|
| A — Anchoring | Reliable, relevant, individualizing and independent | Include; consider higher emphasis |
| B — Supporting | Reliable and relevant but less distinctive | Include with ordinary/lower emphasis |
| C — Context only | Common, uncertain, derivative or duplicated | Keep visible; normally do not score |
| Safety/diagnostic | Red flag, diagnosis, test or referral information | Handle clinically; never reduce to repertory weighting |
Weighting should expose a judgment, not disguise it as mathematics. Record why an item is emphasized. Different repertory traditions order symptoms differently, so state your method rather than presenting one hierarchy as universally mandatory.
Evaluate the chart with sensitivity analysis
A result is fragile if one debatable rubric completely determines it. Run at least three views when the case is complex:
- Primary set: the justified A and B rubrics with your planned weights.
- Unweighted set: the same rubrics without weighting.
- Conservative set: remove the least certain or most interpretive rubric.
Then compare:
- Which remedies remain visible across all views?
- Which appear only because of one heavily weighted rubric?
- Does a very small rubric exclude plausible candidates too early?
- Does the result contradict important generals, modalities or the chronology?
- Is there a repertory or edition effect worth checking?
This is not a statistical confidence analysis. It is a practical robustness check that makes hidden dependence easier to see.
Return to materia medica before documenting a conclusion
Kent’s repertory and other classical repertories were designed to support comparison, but repertory entries compress richer source descriptions. After producing a manageable differential:
- read the leading remedies in an appropriate materia medica;
- verify the exact characteristic pattern and context;
- look actively for contradictions, not only confirmations;
- compare close alternatives;
- return to the original narrative and chronology;
- document uncertainty and why a candidate was retained or rejected.
The remedy with the largest numerical total is not automatically the best fit. Conversely, a favoured remedy should not be rescued by changing rubrics after seeing the chart. If the result seems wrong, return to source notes and rubric meaning rather than manipulating weights toward a preferred answer.
Repertorization audit trail: the minimum viable record
A reviewable analysis can be compact. For each selected rubric, retain:
| Field | What to record |
|---|---|
| Source note | Exact phrase or observation plus date |
| Clarification | Context that supports the interpretation |
| Rubric | Full path as displayed |
| Reference | Repertory/database and edition or version |
| Role | Anchoring, supporting or excluded |
| Weight | Value and short rationale, if used |
| Alternatives | Nearby rubric considered and why rejected |
| Verification | Materia medica source and relevant confirmation/contradiction |
Also save the final rubric set, analysis date and a brief narrative conclusion. Version later changes rather than overwriting the original analysis. HOM-CASE reporting guidance asks published case reports to make the homeopathic analysis, repertorization and rationale transparent. A private record is not a journal report, but traceability remains useful for supervision, follow-up and self-audit.
Using AI-assisted rubric suggestions responsibly
Semantic search can help locate language that a literal repertory search misses. It can also turn a plausible paraphrase into false certainty. Apply the same SOURCE workflow to every AI suggestion.
Before accepting one, ask:
- Can I see the source phrase that triggered it?
- Is the exact rubric available in an identified repertory?
- Can I inspect its full path and remedy list?
- Can I edit, reject and annotate it?
- Does the tool keep suggestions separate from verified record facts?
- Can I export the rubric set and reasoning?
- Are sensitive case notes handled under an acceptable privacy policy?
The AI repertorization software checklist provides a fuller vendor-evaluation framework. AI may accelerate retrieval; it should not diagnose, prescribe automatically or replace source reading.
Common repertorization failure modes
| Failure mode | Why it distorts the analysis | Better practice |
|---|---|---|
| Searching while the patient is still explaining | Early labels anchor later questions | Clarify and preserve the narrative first |
| Rubric by keyword | Same word may carry a different meaning | Check context and full path |
| Too many rubrics | Common and duplicate items dilute the characteristic pattern | Use a justified, compact totality |
| Parent and child both scored | One fact receives extra influence | Choose the most faithful level |
| Rare rubric treated as automatically important | Narrowness can reflect wording or source limits | Check reliability, provenance and fit |
| Highest score treated as prescription | Arithmetic replaces differentiation | Verify in materia medica and whole case |
| Preferred remedy drives rubric changes | Produces confirmation bias | Freeze a documented primary set before viewing results |
| AI output saved as fact | Interpretation loses provenance | Require practitioner review and versioning |
Reusable repertorization checklist
Before search
- Safety, diagnosis/referral responsibilities and concurrent care addressed separately
- Patient’s source language preserved
- Meaning, chronology, modalities and concomitants clarified
- Reliability and relevance marked
- Characteristic features distinguished from common defining features
During rubric selection
- More than one candidate considered where wording is ambiguous
- Full rubric path and neighbouring entries read
- Repertory and edition/version recorded
- Parent-child and same-fact duplication removed
- Exclusions and uncertainty retained in notes
- Weighting rationale stated
After the chart
- Unweighted and conservative views compared
- Fragile dependence on one rubric identified
- Leading and close alternative remedies checked in materia medica
- Contradictions actively sought
- Conclusion linked back to original source notes
- Analysis saved as a dated, versioned record
Frequently asked questions
How many rubrics should be used in repertorization?
There is no universally correct number. Use enough independent, reliable rubrics to represent the characteristic totality without padding the chart. The audit question is stronger than a numerical rule: can you justify every inclusion and show that it adds distinct information?
Is a rare rubric always more valuable?
No. A rare rubric may be precise, but it may also have limited provenance, uncertain wording or a very small remedy set. Distinctiveness must be combined with reliability and faithful fit to the case.
Should general symptoms be repertorized?
They may be important when they are individualized and well supported. Vague or common symptoms usually discriminate less. Medical significance and repertorial discrimination are different questions; an important clinical finding must still be assessed and documented even if it is not useful as a rubric.
What is the difference between a repertory and materia medica?
A repertory starts from indexed symptom language and points toward remedies for comparison. Materia medica presents remedy information in fuller context. In a disciplined workflow, the repertory helps build a differential and materia medica helps verify and differentiate it.
Can software choose and weight rubrics automatically?
Software can suggest rubrics or an initial weighting, but the practitioner should inspect the source phrase, meaning, repertory path, duplication and consequences before accepting them. Automated output should remain editable and visibly provisional.
Should repertorization be changed at follow-up?
Preserve the original analysis. If new information justifies a revised repertorization, save it as a dated version and record what changed. Do not rewrite the earlier chart as though the new information was known at the first consultation.
From a chart to a reviewable decision
Good repertorization is not the production of a convincing graph. It is a controlled translation process in which every important transformation can be inspected: narrative to characteristic symptom, symptom to rubric, rubric set to differential, and differential back to materia medica and the whole case. The case-analysis documentation guide covers the next written artifact: the dated differential, named materia medica checks and a working decision.
Sources and further reading
- Samuel Hahnemann, Organon of Medicine, §§101–104 (written case picture and follow-up comparison)
- American Institute of Homeopathy, The Organon, Section 153 (historical text and discussion of characteristic versus general symptoms)
- Eastman C. et al. (2014), National Guidelines for Translation of Proving Raw Data into Repertory Rubrics, Homœopathic Links, 27(4), 240–244, DOI: 10.1055/s-0034-1383237
- Kent J. T., Repertory of the Homoeopathic Materia Medica (1897 digitized edition), Internet Archive / Wellcome Library
- Teut M. et al. (2022), Case Reporting in Homeopathy—An Overview of Guidelines and Scientific Tools
This HomeoStudio knowledge-base article supports professional education, transparent documentation and software-assisted workflow. It does not claim that homeopathy is effective for any condition and does not encourage self-prescribing.