How to repertorize a case, step by step

Repertorization is the bridge between a taken case and a short list of remedies: select the symptoms that characterise the patient, map each to a rubric, combine the rubrics into a ranked analysis. The procedure is mechanical; the judgement inside it is not. What follows is the full sequence — by hand and in software — with the places it commonly fails.

Step 1 — take the case before you open the book

Hahnemann's instructions on case-taking (Organon §§83–104) precede his instructions on remedy selection for a reason: the repertory can only index what the consultation captured. Record the patient's own language, the modalities (what makes each complaint better or worse), the times, the laterality, and the concomitants — symptoms that accompany the chief complaint without obvious causal connection. A case taken as a disease label ("migraine, 10 years") cannot be repertorized; a case taken as a set of observed particulars can.

Step 2 — select the symptoms worth indexing

Not every recorded symptom enters the analysis. The selection filter is §153: the more striking, exceptional, unusual, and peculiar signs are the ones chiefly to be kept in view. Kent operationalised this into a hierarchy that most classical teaching still follows:

• Mental and emotional generals, when genuinely marked.
• Physical generals — reactions of the whole person to heat, cold, weather, time, food, sleep.
• Characteristic particulars — local symptoms with strange modalities or concomitants.
• Common symptoms — expected features of the pathology, used last or not at all.

A workable analysis usually rests on 4 to 8 symptoms drawn from the top of this hierarchy. Twenty rubrics do not make a stronger analysis; they make a polychrest lottery, because only remedies present in the largest rubrics survive the sum.

Step 3 — map each symptom to a rubric

Each selected symptom must now find its rubric — the right chapter, the right phrasing, the right level of the hierarchy. Three checks per rubric keep the mapping honest. First, meaning: read the rubric's sub-rubrics and cross-references to confirm the entry means what the patient said, not what the wording suggests at a glance. Second, size: prefer the most specific rubric that still genuinely contains the symptom; a 12-remedy sub-rubric that exactly matches beats a 280-remedy parent, but a forced sub-rubric is worse than the honest parent. Third, existence: if no rubric carries the symptom, leave it out rather than bend a neighbour to fit.

Search tooling changes the speed of this step, not its logic. Keyword search demands you guess the index's vocabulary; semantic search retrieves by meaning — "headache better lying down" in natural language surfaces matching rubrics directly. The repertory itself, paper or electronic, is where the discipline lives; tools like the free Similia repertory just shorten the lookup.

Step 4 — choose an analysis strategy and run it

With rubrics on the sheet, the combination method decides the ranking. The three classical strategies:

Hand repertorization rules a grid — rubrics as rows, remedies as columns — and counts; the Boenninghausen tradition did this on paper for a century, and the Therapeutic Pocket Book's structure was designed for exactly such combination work. Software computes all three strategies live: add rubrics from the search panel, optionally weight each rubric 0–4, read the remedy table as it updates, and switch on elimination to restrict the grid to remedies covering checked rubrics. Kingdom or family filters narrow the field further.

Weighting deserves one caution. Grading a rubric 4 because you are confident in the symptom is legitimate; grading it 4 to push a favoured remedy up the table is the analysis equivalent of leading the witness.

Step 5 — read the result as a short list, not a verdict

The output of repertorization is 3 to 5 candidates, not a prescription. Confirmation happens in the materia medica: read each candidate whole — Hering's Guiding Symptoms, Kent's lectures, Clarke, or a modern source — against the case as lived, including everything that never made it into a rubric. The materia medica read regularly overturns the table's first place, and that is the system working: the repertory fragments, the materia medica reassembles.

Record the analysis either way. A documented repertorization — rubrics, weights, strategy, candidates, and the reason for the final choice — is what makes the follow-up reviewable. On paper, the ruled sheet goes in the file; in software, the analysis attaches to the patient record automatically.

Worked example

Chief complaint: recurrent sore throats. Characteristic features from the case: pain consistently left-sided, worse from warm drinks, better from cold drinks; markedly worse after sleep; loquacity during fever noted by the spouse. A defensible sheet: "Throat; pain; left", "Throat; pain; warm drinks aggravate", "Throat; pain; cold drinks ameliorate", "Generals; sleep; aggravates after", "Mind; loquacity; fever, during". Five rubrics, all characteristic, spanning local, general, and mental registers. Any classical practitioner will recognise where this sheet points; the materia medica read — not the sheet — settles whether the portrait truly fits.

Verdict

Repertorize less and select better: a 5-rubric sheet built from §153 symptoms outperforms a 20-rubric sheet built from everything the patient said. The five steps are the same on paper, in Similia, in RadarOpus, or in Complete Dynamics — what changes is the speed of lookup and the live update of the table, not the discipline.

References

Hahnemann, S. (1842) Organon of the Medical Art, sixth edition, §§83–104 (case-taking) and §153 (characteristic symptoms); translation by W. B. O'Reilly (1996), Redmond: Birdcage Books.

Kent, J. T. (1900) Lectures on Homoeopathic Philosophy, Chicago: Ehrhart & Karl; on symptom hierarchy and the use of the repertory.

von Boenninghausen, C. (1846) Therapeutic Pocket Book for Homoeopathic Physicians, English translation by C. J. Hempel, New York: Radde.

Hering, C. (1879–1891) The Guiding Symptoms of Our Materia Medica, 10 volumes, Philadelphia: American Homoeopathic Publishing Society.

Clarke, J. H. (1900) A Dictionary of Practical Materia Medica, 3 volumes, London: Homoeopathic Publishing Company.

Similia (2026) "Searching for Rubrics (Semantic Search)" and "Repertorization and Analysis (Selecting Rubrics)", Similia Help Centre, https://similia.crisp.help/en/, fetched 2026-05-17.