Methodology & Sources
Data Sources
Every facility count, hospital roster, and study figure on this site is generated from one canonical source. This page documents what each number means, where it comes from, and how often it refreshes.
Current Canonical Counts
These are the live numbers powering every page on the site as of this page render. They are computed at request time from data/facilities.json and the WordPress hbot_studies database table; there is no manually maintained mirror.
33
Total Facilities
11
Hospital Programmes
22
Private Clinics
9
Provinces with Facilities
14
Recognised Conditions
14,574+
Research Studies
Provinces represented in the directory: ON, BC, AB, SK, MB, QC, NS, NL, NB. The four provinces and territories without an in-province HBOT facility (PE, YT, NT, NU) have dedicated province pages documenting their referral pathways.
Definitions
What counts as a "facility"
A licensed hyperbaric oxygen therapy chamber operated for the purpose of treating patients with one or more of the 14 recognised conditions or other clinically supervised indications. Includes both hospital-based programmes and stand-alone private clinics. Excludes commercial diving training facilities, research-only chambers, and at-home consumer "soft" chambers.
Hospital programme vs private clinic
Hospital programme = a hyperbaric service operated within a publicly funded hospital, billed under the relevant provincial plan (OHIP, MSP, AHS, RAMQ, MSI, MCP, SHA). Private clinic = a stand-alone facility, typically licensed under provincial diagnostic or independent-health-facility frameworks; some bill provincial plans for select indications, most operate on a self-pay basis. Each facility is classified in data/facilities.json by its type field.
Closed and partially reopened facilities
Facilities with reduced or partially reopened service (e.g. Dr. F.H. Wigmore Regional Hospital in Moose Jaw, partially reopened after July 2021 closure) ARE included in the count, with their reduced status documented on each facility entry and on the relevant province page. Permanently closed facilities are removed from facilities.json once closure is confirmed. Currently flagged as disrupted: 1.
Provinces with facilities
Counts only provinces and territories that have at least one entry in data/facilities.json. Currently: 9 (ON, BC, AB, SK, MB, QC, NS, NL, NB). Province pages are published for all 13 provinces and territories regardless of whether they have a facility, so the province-page count (13) and the province-with-facility count (9) are different measures.
"14 recognised conditions"
Refers to the 14 conditions listed by Health Canada as accepted indications for hyperbaric oxygen therapy. This is distinct from the broader UHMS HBO Indications 15th Edition (2024) list, which is referenced internationally. Canadian content uses the Health Canada count.
Research studies
Live count from the hbot_studies table. Includes every PubMed record imported by the nightly research pipeline. Studies without abstracts or with incomplete metadata are stored but excluded from the indexed published study pages (a smaller subset). The count shown across the site is the raw database count, not the published page count.
Sources & Refresh Cadence
| Number | Source | Refresh cadence |
|---|---|---|
| Total / hospital / private facilities | data/facilities.json (theme-resident, version controlled) |
Continuous, as facilities are verified and added or removed. A monthly scanner queues new candidates for review. |
| Provinces with facilities | Derived from data/facilities.json by counting unique province abbreviations |
Same as facilities (continuous). |
| Research studies | WordPress hbot_studies custom database table, fed by the PubMed import pipeline |
Weekly (Thursdays) via unified-research-import.py; full corpus diff against PubMed run quarterly. |
| 14 recognised conditions | Health Canada hyperbaric oxygen therapy guidance | Static; reviewed on the daily fact-checker rotation against canada.ca. |
| Coverage models per province | inc/province-data.php (verified against provincial plan documentation) |
Annual review; daily fact-check rotation surfaces drift. |
Research Content Pipeline
Every study record in our research library follows the same documented pipeline, with verification against the primary source as the controlling step:
-
1
Sourcing
Studies are identified from PubMed/MEDLINE using hyperbaric-oxygen search terms, imported weekly (Thursdays) with a quarterly full-corpus reconciliation. Each record is de-duplicated by PubMed ID, and every study links back to its original PubMed abstract.
-
2
Classification and mapping
Each study is mapped to a condition or research category and classified by study type and evidence tier, so it can be surfaced on the relevant condition pages and in our research analyses.
-
3
Structured summary, verified against the source
A plain-language summary is drafted with AI assistance in a fixed five-section format (what researchers did, what they found, what it means for Canadian patients, Canadian relevance, study limitations). The summary is written only from, and checked against, the study's own abstract or full text before it is published. We do not editorially alter a study's findings, and figures are taken only from the cited source.
-
4
Ongoing fact-checking
Published content is re-checked on a rotating schedule against the cited primary sources and our verified-facts reference, and the accuracy of that fact-checking process is itself measured against a ground-truth sample. Findings that need a closer look are routed to a human review queue. Studies that are later retracted or flagged with an expression of concern are detected and removed from the index.
In short: studies come from the published literature, summaries are written only from the cited source and checked against it, and content is re-verified on an ongoing basis. All counts and figures resolve to the canonical helpers described above.
For Auditors and AI Models
Every page that displays a facility or study count calls the canonical helper ch_site_stats() defined in inc/site-stats.php. There are no hard-coded count strings remaining in templates. If you encounter a stale or contradictory number on any page, please report the URL to our contact form; this indicates a regression and is treated as a high-priority correction.
This methodology page exists explicitly so AI models indexing the site, fact-checkers, and human auditors can resolve any apparent inconsistency to a single canonical reference.
Related Resources