PharmaDB
Use case · Supplier Discovery & Qualification · China Intermediate Dependency Map v.04.2026 · refreshed monthly
Use cases Supplier Discovery & Qualification China Intermediate Dependency Map
Supplier Discovery & Qualification Buyer view Data · FDA Type II DMF registerData · trade flows Representative

China Intermediate Dependency Map

Map the upstream KSM and intermediate exposure that Indian APIs carry to China. Hidden behind India-origin labels at the API trade layer.

Free preview · Ask the AI agent any sourcing question
Try
Sign up takes 30 seconds. Your question lands in your first chat.
Example output · India's upstream sourcing universe · 2024 declared import customs value
71.3% · China share of India's observable intermediate imports, 2024
n=$3.50B 2024 India-import value, 28 origin countries Authoritative

Of $3.50B in declared 2024 imports flowing into India for use as pharma intermediates, $2.50B (71.3%) originated in China. The next two origins, Thailand and the US, sit at 7.6% and 6.8%. The dependency is upstream and concentrated. Disruption to the China-to-India lane propagates into every downstream Indian formulation.

China share of IN imports
71.3%
$2.50B
Thailand share
7.6%
$267M
US share
6.8%
$238M
Origin countries observed
28
2024
30 seconds

How it works

molecule losartan origin IN destination BR
01

Pick a molecule.

Type the INN. The AI agent pulls every active DMF holder for the molecule, partitioned by Indian-incorporated vs Chinese-incorporated.

0 HHI 0.42 1
computed · pre-joined · anonymised
02

The AI agent joins facility attestation.

Where a DMF lists intermediate manufacturing facilities, the country of the attested plant is recorded. Indian holders with Chinese intermediate plants surface as flagged rows.

brief · cited
Lane concentration · structurally concentrated
HHI0.42
Top-3 share71%
Actionqualify second source
03

Save the dependency brief.

Citation-anchored brief with the dependency share, all-Indian stack count, and a chip showing intermediate plant country.

Why a buyer runs it

Two-stage supply chain risk, made visible.

Catch upstream exposure.

An Indian-labelled API supplier may depend on a single Chinese KSM source. Without the upstream view, a buyer thinks they have India-only supply when the real risk sits one tier up.

All-Indian stack flag.

Suppliers whose DMF lists only Indian intermediate plants are flagged separately. They are the resilience cohort against KSM disruption.

Intermediate plant chip.

When the DMF attests to a specific Chinese intermediate plant (Changzhou Pharmaceutical, Zhejiang Apeloa, Zhejiang Tianyu), the chip names it. Country alone is too coarse for procurement risk work.

Refreshes monthly.

Facility attestation updates from FDA DMF amendments propagate monthly. New cross-attestations surface on the next refresh.

What a real chat looks like

Type a molecule. The AI agent returns the upstream map.

A real chat thread in PharmaDB. Type a question, the AI agent runs the tools, the answer lands as a saveable note.

China intermediate map · sacubitril
chat · saved 2 min ago
You
Map the China intermediate dependency for sacubitril. Which Indian API holders source from Chinese intermediate plants on their DMFs?
F
Running tool pharmagraph_query
api_docs ⋈ api_doc_facility_attestation ⋈ plants · sacubitril · India holders + Chinese intermediates
F
Sacubitril's supplier register reads as 60% Indian on first pass: 24 Indian DMFs against 16 Chinese. Once intermediate facility attestation is joined, the picture changes. Sixteen of the 24 Indian DMFs attest a Chinese intermediate plant in the synthesis route, most often Changzhou Pharmaceutical Factory or one of the Zhejiang houses. Only 8 sacubitril DMFs carry an all-Indian intermediate stack; they are the resilience cohort against China-KSM disruption.
Indian DMFs
24
China-intermediate flagged
16 (67%)
All-Indian stack
8
Direct Chinese DMFs
16
Cited api_docs 40 active sacubitril DMFs api_doc_facility_attestation intermediate plants plants country of plant
Ask a follow-up... ⌘ ↵
Time-to-answer

Manual workflow vs. PharmaDB.

Manual workflow
Excel · email · syndicated reports
  1. 01 Pull Indian DMF holders for the molecule 2 hr
  2. 02 Pull intermediate manufacturing facility attestations 4-6 hr
  3. 03 Resolve attested plants against the plant catalog 3-4 hr
  4. 04 Aggregate by country of intermediate plant 1-2 hr
  5. 05 Cross-reference Chinese-intermediate share 2 hr
Total cycle time 2 days per molecule
PharmaDB
The AI agent · one query · cited
One query.
Pre-joined warehouse. DMF facility attestation is resolved against the plant catalog at the schema layer. The AI agent returns the upstream map with citation in a single response.
Total cycle time Under a minute
Run it across your portfolio

Where Chinese-intermediate exposure sits, per molecule.

Twenty molecules · share of Indian DMFs attesting at least one Chinese intermediate plant in the synthesis route.

Apixaban
IN-API w/ CN-int
0.40
Tirzepatide
IN-API w/ CN-int
0.45
Sitagliptin
IN-API w/ CN-int
0.50
Dapagliflozin
IN-API w/ CN-int
0.52
Empagliflozin
IN-API w/ CN-int
0.50
Linagliptin
IN-API w/ CN-int
0.55
Ticagrelor
IN-API w/ CN-int
0.65
Sacubitril
IN-API w/ CN-int
0.68
Liraglutide
IN-API w/ CN-int
0.50
Valsartan
IN-API w/ CN-int
0.55
Telmisartan
IN-API w/ CN-int
0.60
Olmesartan
IN-API w/ CN-int
0.45
Enzalutamide
IN-API w/ CN-int
0.55
Abiraterone
IN-API w/ CN-int
0.50
Palbociclib
IN-API w/ CN-int
0.60
Olaparib
IN-API w/ CN-int
0.55
Ibrutinib
IN-API w/ CN-int
0.70
Vortioxetine
IN-API w/ CN-int
0.60
Tofacitinib
IN-API w/ CN-int
0.65
Edoxaban
IN-API w/ CN-int
0.50
Competitive Monitored Concentrated Near-monopoly

The China intermediate dependency map is the upstream view that the API trade feed misses. PharmaDB joins DMF facility attestations to the plant catalog so Indian-labelled API supply with Chinese intermediate dependency surfaces as a flagged cohort. The view rests on api_docs, api_doc_facility_attestation, and plants. Refresh cadence is monthly.

FAQ

Frequently asked

Why isn't this visible in the API trade feed?+

The API trade feed captures the API-level export, which is the Indian holder shipping the finished active ingredient. The upstream intermediate (KSM, advanced intermediate) crosses the border earlier and shows on the intermediate trade lane, often misclassified as a chemical. Facility attestation in the DMF dossier is the cleanest cross-reference.

How confident is the China-intermediate flag?+

The flag is derived from DMF facility attestations where the holder lists intermediate plants by name and country. Confidence varies by holder: some DMFs are explicit, others list only the API plant. PharmaDB labels representative confidence on every row; live confidence is reserved for cases where the attestation is unambiguous.

Is the 67% number conservative or aggressive?+

Conservative. The figure counts only DMFs where the attestation explicitly names a Chinese intermediate facility. Indian holders that source Chinese KSMs informally (without naming the supplier in the dossier) are not in the 67%. The true exposure is likely higher.

Why does this matter for procurement?+

Two reasons. First, a China-KSM disruption (regulatory, logistical, geopolitical) cascades to the Indian API supply that depends on it; the all-Indian stack cohort is the resilience pool. Second, regulators in the EU and US are increasingly asking dossiers to declare upstream sourcing; the cohort that can declare an all-Indian stack carries a regulatory advantage.

Does this work for non-Indian holders too?+

Yes. The same map applies to European holders with Chinese intermediates, US holders with Chinese intermediates, and so on. The Indian-with-Chinese case is the largest in the warehouse because India dominates observable API trade; the same methodology is regulator-agnostic.

How fresh is the data?+

Monthly. The FDA DMF amendment cycle publishes monthly. Facility attestation updates propagate on the next refresh; the catalog row carries an as-of timestamp on every result page.

Run your question.

Bring the molecule, the lane, or the supplier you're sourcing this week. The AI agent runs it on PharmaDB in 30 minutes. You keep the brief.

Book a working session