WHO medical device policy database: country policy flags and imaging density, delivered as rows

It is one World Health Organization program with three deliverables, all maintained by the Health Product Policy and Standards team - and the search results that land here usually want all three at once.

The vocabulary. WHO publishes the working definitions the rest of the field argues by: a medical device per the Global Harmonization Task Force definition finalized in May 2012, and medical equipment as devices requiring calibration, maintenance, repair, user training and decommissioning - explicitly excluding implants and single-use devices. Companion pages extend the terms to priority medical devices (including the COVID-19 priority list), in vitro diagnostics, personal protective equipment and oxygen therapy.

The indicators. Twelve DEVICES-series codes carry the actual country-level record: eight categorical policy flags asking whether a nation maintains recommended device lists, procurement guidelines, a regulating authority, an HTA unit and the rest of the buying machinery - plus equipment-density measures recording MRI, CT, PET, mammography and radiotherapy units per million population. Verified pulls returned roughly 290-540 country-year rows per indicator, about 4,700 rows in total, reaching up to roughly 143 Member States.

The Atlas. The Global Atlas of Medical Devices folds seven survey topics into three published editions - a 2010 baseline, 2017 and 2022 - with the newest carrying structured country profiles for 177 countries.

Datadory delivers the whole stack as the WHO Medical Devices - Definitions and Policy Repository dataset: quality score 7 out of 10 against a 7.81 average across the 1,744 datasets in a catalog spanning 159 industries, and the only non-US lens among Health Care Equipment's three primary sources. Everything American in this industry stops at the border; this is the record that does not.

Three real observations captured during the August 2026 verification pass, exactly as a delivery lands:

IndicatorCode : DEVICES08     # MRI units per million population
SpatialDim    : TTO           # Trinidad and Tobago (ISO-3)
ParentLocation: Americas
TimeDim       : 2010
Value         : 2.98          NumericValue: 2.98
Comments      : Only public sector data

IndicatorCode : DEVICES08     # same series, different provenance
SpatialDim    : USA
ParentLocation: Americas
TimeDim       : 2019
Value         : 40.11         NumericValue: 40.11
Comments      : OECD 2017-19

IndicatorCode : DEVICES06     # national standards / recommended lists
SpatialDim    : VUT           # Vanuatu
TimeDim       : 2010
Value         : For different healthcare facilities and specific procedures

Read side by side, both halves of the corpus are visible in one table shape. The Trinidad and USA rows are the quantitative layer - same indicator code, same typed numeric value, twenty-nine MRI units per million apart - while the Comments column carries the provenance split: one nationally reported survey figure, one value drawn from an OECD compilation. The Vanuatu row is the policy layer: identical schema, but the value is a sentence describing which facilities the national standards list covers.

That duality is the design fact to plan around. Value is always text; NumericValue is populated only where a number exists, so filtering on it isolates the benchmarking corpus and keeping the text rows preserves the regulatory census. One delivery serves market sizing and compliance screening without a second reconciliation project between them.

Twelve DEVICES-series codes were verified live during cataloging, and they split cleanly into two jobs. Eight are categorical statements about a country's standing arrangements - the machinery that decides whether a market can buy. Four are numeric densities recording whether equipment exists there at all:

The density family extends past imaging into radiotherapy and nuclear medicine - PET, mammography, linear accelerator, telecobalt and gamma camera/nuclear medicine measures round out the twelve. Observed values span 2010 to 2021, with some equipment figures sourced from OECD compilations rather than WHO's own survey wave - the Comments field says which is which, per row.

The working questions are market-shape questions, and each maps onto named fields rather than prose guessing:

• Market-entry sizing. DEVICES08 and DEVICES09 give the installed-base denominator medtech strategy teams size against - compare forty candidate markets on scanner intensity before deciding where a launch lands.
• Regulatory-affairs mapping. The eight categorical flags turn what does this market require from a fifty-tab research project into a filtered query: which candidates run a regulating authority (DEVICES24), publish procurement guidelines (DEVICES19), maintain approved reimbursement lists (DEVICES18).
• Health-system and NGO planning. Density benchmarks against WHO-region peers give procurement programs a defensible target - how many imaging units per million a system should plan for, not just how many it has.
• Investment screens. Pairing density trends with policy-readiness flags separates markets where demand exists but purchasing machinery does not from markets ready to absorb capital.
• Citation-grade research. WHO's definitions are the ones other publishers quote; when a device-policy claim has to survive review, the organization that standardized the vocabulary is the provenance reviewers stop arguing about.

Two structural notes keep the analysis honest. Density is a ratio, so a small country can post a striking figure off a handful of machines - join population alongside before ranking. And the categorical answers are sentences, not codes, so any country-comparison matrix wants the value-mapping table built once and reused across every flag.

No single table does, which is why the question is best bought as a small stack. Within Health Care Equipment, each record ships through Datadory normalized into typed tables with its field dictionary attached, so the layers join instead of merely coexisting.

The anchor is the WHO Medical Devices - Definitions and Policy Repository: country posture and installed-base ratios across up to 143 Member States, plus the definitions layer and 177 Atlas country profiles. Around it sit the American regulator feeds:

• The openFDA Device Adverse Events (MAUDE) API holds 25,711,469 adverse-event reports - the event-level record of what went wrong with devices already on shelves.
• The openFDA Device Registration & Listing API maps 333,804 establishments worldwide, including foreign facilities marketing into the United States - the manufacturing footprint the WHO policy flags contextualize.
• The openFDA Device 510(k) Clearances API contributes 175,814 premarket notifications from 1976 forward, the clearance ledger a new market entrant's products passed through.
• The openFDA Device Recalls API adds 58,999 recall events, and the CMS Data Hub - Medicare & Medicaid Datasets supply the payer layer - 159 cataloged datasets plus 236 in the Provider Data Catalog.

Run as a sequence, the stack reads: does the target country have the regulatory and reimbursement machinery (WHO flags), is imaging capacity present to absorb the category (WHO densities), who manufactures into the space (Registration & Listing), what has gone wrong before (MAUDE, Recalls). Each layer answers a question the others cannot.

• Medtech market-entry teams screen candidate countries on the policy flags before commissioning primary research, then size the installed base with the density series.
• Regulatory-affairs analysts maintain a living matrix of which jurisdictions run regulating authorities, HTA units and reimbursement lists - kept current as a delivery, not a spreadsheet hunt.
• Journalists and academics rate this record highest in the industry at 3 of 3 relevance: citable definitions and country profiles for device-policy scholarship. See journalists academics use cases.
• Data scientists build country-year panels for global-health models at 2 of 3 relevance, a compact fully documented table that joins cleanly onto event-level FDA feeds: data scientists use cases.
• Developers and data-product builders stand up country-comparison dashboards on the same nine-field shape at 2 of 3 relevance: developers builders use cases.
• Investors and quants read policy-readiness flags beside equipment intensity when weighing which emerging markets can actually convert demand into purchases.

For teams needing a wider aperture than twelve indicators, the WHO Global Health Observatory Data Repository extends the same Member State geography to 3,000+ health indicators - the policy slice traded for breadth.

API, files, or your warehouse. Daily, weekly, or hourly.

Name the indicators, the countries and the year range when you request a sample - all twelve DEVICES-series codes across their full Member State span, or a narrow cut like MRI and CT density for a launch shortlist with the procurement-list flags beside each. The sample ships first; the ongoing arrangement follows once the rows validate.

Rows arrive as typed tables carrying the nine-field dictionary: the indicator pair (IndicatorCode, IndicatorName), geography (SpatialDim as ISO-3, ParentLocation as WHO region), year, Value as text, NumericValue where a number exists, the Comments scope footnote and the load timestamp. Three extensions are confirmed at sampling rather than guessed here: a coded value-mapping for the eight categorical policy indicators, the Atlas country-profile tables behind the 177-country edition delivered as rows keyed to the same ISO-3 codes, and per-indicator provenance notes separating nationally reported figures from OECD-derived ones.

Start with the anchor record, the WHO Medical Devices - Definitions and Policy Repository dataset page - full field dictionary, verified sample rows and coverage chips included - and request a sample cut to your indicators and countries before anything ongoing is committed.

This page is one thread of a wider map. The health care equipment data guide walks the pooled slice end to end, the health care equipment data hub indexes every dataset page in the industry, and the best health-care-equipment datasets ranking scores the leaders side by side. Request a sample from any of them - the rows prove the rest.