Healthcare pack

Drug formulary CSV import API

Import a drug formulary from CSV. ATC-coded, GTIN-checksum-validated, Pharmacode-aware, emits FHIR Medication.

medium riskFHIR · Medicationdrug_formulary_v1

30-second curl

curl -X POST https://api.adaptivmapr.com/v1/uploads \
  -H "Authorization: Bearer $ADAPTIVMAPR_API_KEY" \
  -F "template=drug_formulary_v1" \
  -F "file=@your_data.csv"

→ 6 canonical fields · 3 validated · medium risk

Canonical columns

The whole schema, printed as it ships.

Every canonical column, the type each row carries, whether it is required, the field-level validators that fire on commit, and the multilingual header hints the cascade resolves against. This is the shipped definition, not a summary of it.

drug_formulary_v1FHIR · Medication

fields: 6
required: 2
validated: 3
hints: 16

Canonical column	Type	Required	Validators	Header hints the cascade matches
`atc_code`	string	yes	`atc_code`	atcwirkstoffcode
`gtin`	string	—	`gtin`	gtineanbarcode
`pharmacode`	string	—	`pharmacode`	pharmacodepharma-code
`name`	string	yes	—	präparatpréparationmedicamento
`dosage_form`	string	—	—	darreichungsformforme galéniqueforma
`strength`	string	—	—	stärkedosagedosis

Read the same definition as JSON at GET /v1/templates/drug_formulary_v1. A hint match resolves on layer 2 — no LLM call, no token spend, just the flat per-map fee. Hover a validator id to see what it checks.

6 canonical fields
2 required
3 validated
16 header hints, 5 languages
FHIR Medication

Why it exists

Written for the file you actually receive.

The Drug formulary template is the canonical schema for a pharmacy's medication catalog. Each row emits a FHIR Medication resource — the ATC code lands in Medication.code with the WHO ATC coding system, the GTIN goes into Medication.identifier with the GS1 system URI, and the Swiss Pharmacode (4-7 digits) carries the local registry id. The template is `medium` risk not because it contains PHI — drugs are not patients — but because dosage-form and strength errors are downstream-dangerous; a transposed digit in a strength column produces a 100x dose error in the resulting prescription system. AdaptivMapr validates ATC against the published pattern (^[A-Z]NN[A-Z]{2}NN$), GTIN against the GS1 mod-10 checksum, and Pharmacode against the 4-7 digit shape. Customers use this template when seeding an e-prescribing system, when standardising an in-house pharmacy database, or when re-importing after a vendor migration.

GTIN supports 8 / 12 / 13 / 14 digit shapes (EAN-8, UPC-A, EAN-13, GTIN-14) with mod-10 checksum verification — the same algorithm as ITF-14 barcodes. ATC five-level codes (anatomical → therapeutic → pharmacological → chemical → substance) all validate against the same regex. Pharmacode is Swiss-specific and not enforced as a checksum because the algorithm is proprietary.

Migration scenarios & the foreign headers they ship

Migration scenarios for the Drug formulary: seeding a new e-prescribing module with a clinic's in-house formulary, importing a hospital pharmacy's catalogue after a procurement-system switch, quarterly refreshes against Swissmedic / EMA / FDA bulletins, and consolidating multiple regional formularies after a healthcare-network acquisition. Foreign headers we routinely see: "ATC / Wirkstoffcode / Code ATC / GTIN / EAN / Barcode / Codice / Pharmacode / Pharma-Code / Präparat / Préparation / Medicamento / Farmaco / Darreichungsform / Forme galénique / Forma farmacéutica / Stärke / Dosage / Dosis". Trade names rarely match ATC patterns directly, so the resolver layer carries weight on this template — most imports rely on the high-confidence trade-name → ATC fallback to clear without manual intervention.

The cascade

Five layers, and the cheapest one wins.

Layers run in order and stop the moment a column resolves. That is the single biggest cost lever in the system: a column caught on layer 2 never reaches the metered layer 5.

L1Statisticsno LLM
Auto-accepts a header that past confirmations already resolved the same way, at {minN:100, minRatio:0.95} or {minN:20, minRatio:1.00}.
L2Heuristicno LLM
Normalises accents, punctuation and whitespace, then compares against the column name, the label, and every registered hint (DE / FR / IT / EN / ES).
L3Fuzzyno LLM
Token-set ratio plus Levenshtein over the normalised strings. Auto-accepts at 0.80 — it absorbs typos and reordered words.
L4Semanticcheap, cached
Embedding cosine between the header and the field’s label + hints. Catches the long tail of paraphrases.
L5LLMmetered
Everything still unresolved goes up in ONE batched, collision-aware call, constrained to this template’s column set so it cannot invent a field.

Try it

One template id, two ways in.

REST for your import pipeline, MCP for your editor. Both run the same cascade and both honour the same schema-only clamp.

REST · POST /v1/uploads

Name the template; the cascade picks up the rest. The canonical definition is read-only at GET /v1/templates/drug_formulary_v1.

bash

curl -X POST https://api.adaptivmapr.com/v1/uploads \
  -H "Authorization: Bearer $ADAPTIVMAPR_API_KEY" \
  -F "template=drug_formulary_v1" \
  -F "file=@your_data.csv"

→ upload created · mappings ready · confirm before commit

MCP · Cursor / Claude Desktop

Drop AdaptivMapr into your editor and call the same cascade as a tool. Schema-only calls leave only column names and up to three clamped sample rows.

mcp

// In Cursor or Claude Desktop with the AdaptivMapr MCP server installed:
adaptivmapr.match_headers({
  template_id: "drug_formulary_v1",
  headers: ["atc_code", "gtin", "pharmacode", "name"]
})

schema-only · headers and ≤3 rows, 80 chars each

MCP install instructions

medium-risk template

The mappings response comes back flagged. PATCH /uploads/:id/mappings returns requires_hitl: true and hitl_status: "pending_review" so you can hold the commit in your own workflow — the flag is a signal, not a queue we run. Schema-only mode (headers plus at most three sample rows, each clamped to 80 characters) is a data-minimization mode enforced at the HTTP edge. Full-data mode routes the metered layer-5 call to phi-cloud in-region under a BAA, costs 20% more on the whole map charge, and stays locked until the workspace accepts the BAA/NDA in Settings → Security & Data.

How the wallet is charged

Questions

Drug formulary CSV import — FAQ

Yes — the ATC validator falls back to a resolver that maps trade names ("Aspirin", "Dafalgan") to ATC codes when the direct validator fails, and the substitution commits when confidence ≥ 0.85.

Add a `schedule` field via a fork. The canonical template stays jurisdiction-neutral because controlled-substance lists differ per country.

Yes — Pharmacode and GTIN are both optional. The minimum is ATC + name; the rest enrich the row.

Yes. The lexicon is too large and too localised to enum-bound; teams with strict downstream systems usually post-process these fields against an internal lookup.

Ready when you are

Map drug formulary in production — without shipping raw records.

Schema-only mode leaves only headers and a handful of clamped samples. Add full-data when you need row-level AI, routed in-region under a BAA.

Get an API key More Healthcare templates →

No free tier · $10 prepaid wallet to start · every map draws a small flat fee, deterministic ones included