rehabCare discovery research
Research methods and limitations
A useful finding needs a clear denominator, an inspectable source, and an honest boundary.
Two kinds of research
The language-access feature synthesizes federal guidance and selected provider disclosures. Country reports are original descriptive analyses of contact fields in Rehab.ai’s delivered catalog. They do not measure all information visible on every rendered page. Neither is a clinical study.
Population and unit
Published country reports include only records present in the delivered catalog for an enabled country, deduplicated by canonical listing URL. A contact indicator is present if any catalog row sharing that canonical URL records it. A record can identify a location or a service unit. Multiple records may belong to one organization. Incoming and held records do not contribute to published counts.
Field definitions
| Measure | Definition |
|---|---|
| Phone recorded | A recognized phone or intake field contains digits and is not an unknown-value placeholder. No call was placed. |
| Website recorded | A recognized provider website field contains an HTTP or HTTPS URL. Source-reference URLs do not count unless separately stored as a website. |
| Either / both / neither | Union, intersection, and complement of these two indicators, calculated per record. |
| Percentage | Indicator count divided by included records, multiplied by 100 and rounded to one decimal. Zero denominators are not assigned percentages. |
| Language support | Not estimated in this edition. Page language, program language, and interpreter arrangements must be evidenced separately. |
Reproducibility
Country downloads include the included listing paths and their two Boolean indicators. Summing the indicators reproduces the table; use logical OR for either, AND for both, and neither true for neither. Each country report includes a hash of its analyzed input. The local CSV groups records by stored city and region labels, including blank labels. Legacy listing paths are resolved through the site’s canonical route map before deduplication. It does not geocode records or estimate travel access.
Dates
The analysis date is the date of this edition. Source publication year, source retrieval time, and provider confirmation are separate. Regenerating a report does not make underlying service information newly verified. Revisions to data require a new recorded input hash and an updated edition history.
What the findings cannot establish
- Whether a phone answers or a website belongs to the specific service unit.
- Whether an organization is licensed, accepting patients, affordable, or clinically appropriate.
- How many people can obtain treatment or how many treatment places exist.
- National prevalence, market share, or a ranking of countries’ care quality.
- That an unrecorded field means a service does not exist.
Sources and commercial interest
Rehab.ai operates the directory being studied. Research is labeled by scope so readers can evaluate it independently. External source links document evidence, not endorsement. The current work was prepared with AI assistance and automated calculations; it has not received independent clinical or statistical review. No such credentials or review are implied.
Corrections and editions
Version 1.0, September 19, 2026: first contact-field analysis and exploratory language-access feature. Submit corrections with supporting source material. Preserve the cited edition and its download when using figures; later directory changes can alter counts.
Country release boundaries
Prepared research for incoming countries remains outside public assets. Research does not change listing eligibility or country release status. When a country is released through its existing checks, this builder derives its public contact report from the delivered records.