Skip to content

PHARMAC data via eolas

Te Pātaka Whaioranga – PHARMAC is New Zealand's pharmaceutical management agency — it negotiates pricing and decides which medicines are funded under the public health system. eolas serves 4 datasets from PHARMAC: the current community + hospital medicines schedules plus monthly historical snapshots back to 2011.

If you're doing pharmaceutical policy research, market access work, generic-launch tracking, or studying NZ pharmaceutical funding decisions over time — PHARMAC is the source.

What's in the catalogue

Dataset Description
pharmac_schedule Current month's community pharmaceutical schedule (CPS) — every funded community medicine, with brand, strength, formulation, subsidy, special-authority status.
pharmac_schedule_history Monthly snapshots of the CPS from September 2010 to present (~14 years of monthly snapshots, ~840k rows).
pharmac_hospital_medicines_list Current month's hospital medicines list (HML) — medicines funded for inpatient use.
pharmac_hml_history Monthly snapshots of the HML from January 2011 to present.

Refresh schedule

Monthly. PHARMAC publishes both schedules monthly; our refresh runs weekly to catch new releases promptly. New schedule editions typically issue mid-month with funding changes effective from the 1st of the following month.

meta = client.info("pharmac_schedule")
meta["last_refreshed_at"]
meta["source_last_modified_at"]

License

All PHARMAC data is published under CC-BY 4.0. Commercial use is fine; attribution required.

Recommended attribution: "Source: Te Pātaka Whaioranga – PHARMAC, served via eolas (eolas.fyi). CC-BY 4.0."

Common patterns

What's funded for a specific medicine?

sched = client.pharmac("pharmac_schedule")
# All funded brands for paracetamol
para = sched[sched["chemical_name"].str.lower().str.contains("paracetamol")]
print(para[["brand", "strength", "form_type", "subsidy"]])

library(dplyr)
sched <- eolas_get_pharmac("pharmac_schedule")
para <- sched |> filter(grepl("paracetamol", tolower(chemical_name)))
print(para[, c("brand", "strength", "form_type", "subsidy")])

When was a medicine first funded?

hist = client.pharmac("pharmac_schedule_history")
# Find first appearance of "semaglutide" (Ozempic / Wegovy active ingredient)
hits = hist[hist["chemical_name"].str.lower() == "semaglutide"]
print(f"First listed: {hits['snapshot_date'].min()}")

Generic-launch tracking

hist = client.pharmac("pharmac_schedule_history", start="2020-01-01")
# Brands per chemical, per month — proxy for generic entry
by_month = hist.groupby(["snapshot_date", "chemical_name"])["brand"].nunique().reset_index()
# Chemicals where brand count grew
multi_brand = by_month.groupby("chemical_name")["brand"].max()
new_generics = multi_brand[multi_brand > 1].sort_values(ascending=False).head(20)
print(new_generics)

Subsidy spend over time

hist = client.pharmac("pharmac_schedule_history")
# Total subsidised cost by month (where unit price + quantity available)
hist["est_cost"] = hist["price_per_unit"] * hist["dispensings"]
by_month = hist.groupby("snapshot_date")["est_cost"].sum()
by_month.plot(title="Estimated CPS subsidy spend by month")

Source-specific notes

  • CPS vs HML: the Community Pharmaceutical Schedule covers medicines dispensed in pharmacies (subsidised + special-authority); the Hospital Medicines List covers medicines used in hospitals. Funding rules differ; both schedules update independently.
  • Snapshot history is a goldmine: ~14 years of monthly snapshots lets you see when each medicine was first funded, brand changes, price changes, special-authority restrictions added or removed. This is unusual longitudinal data — most pharmaceutical regulators don't publish full historical schedules.
  • Special Authority: many medicines have access restrictions (only funded for specific conditions, after first-line failure, etc.). The special_authority field flags these.
  • Subsidy ≠ price: PHARMAC subsidises medicines; patients sometimes still pay a co-payment ($5 standard for funded scripts). Multi-source brands may have different patient prices despite identical funding.
  • Funding decisions are precedent-setting: when PHARMAC funds a new medicine, that decision is sometimes the model overseas regulators reference. Tracking historical funding decisions has international value.
  • Section 29 medicines: prescribed but unfunded — patients pay full retail. Not in the funded schedule; check pharmacy retail databases instead.

Where to find more