EDS Research Index

Methodology

Rare diseases are studied differently from common ones. Trials are few, patient communities sometimes notice problems years before the literature does, and the diagnostic criteria for EDS have changed three times in forty years. This page sets out the rules the index follows. The site checks some of them automatically when it builds; the rest depend on editorial review.

1. Five kinds of evidence

Every source is filed under exactly one of five kinds of evidence, and every record lists its evidence by kind. The kinds are never merged into a single score.

  • Clinical: peer-reviewed articles, trials, guidelines, consensus statements, and laboratory studies.
  • Community: patterns reported across forums, support groups, and patient organizations. The index names the venue only; no individual poster, username, or verbatim post is recorded.
  • Historical: case reports from before EDS was formally classified, accounts written at the time, archival records, and documented folk practice.
  • Registry: patient registries, trial registrations, and rare-disease reference databases (Orphanet, GARD, ClinicalTrials.gov).
  • Gray literature: preprints, theses, conference abstracts, and working papers that have not been peer reviewed.

2. Evidence levels within each kind

Each kind of evidence has its own scale, so a randomized trial and a recurring forum pattern are never ranked against each other. The clinical scale runs from systematic review to case report, the community scale from structured patient survey to individual account, and the historical scale from primary historical document to folk tradition. The build rejects a record that gives a source a level from the wrong scale, or a level different from the one in the source catalog.

3. When kinds of evidence agree or disagree

A record whose evidence spans more than one kind states how that evidence relates: convergent when independent studies or reports agree, one underlying source when the different kinds restate a single study or report (a paper, its preprint, and a press release are one study), contested when they disagree, or refuted. Records with one kind of evidence make no such claim. Local anesthetic resistance is the clearest example of agreement: patient reports came first, then surveys, then a randomized trial.

4. Diagnostic criteria

What counted as “EDS” changed with each classification. The 1997 Villefranche nosology consolidated the types to six, and the 2017 International Classification defines thirteen and created hypermobility spectrum disorder (HSD) for people with symptomatic hypermobility who do not meet the hEDS criteria. Every record names the criteria in force when its sources were written, so a study from before 2017 is not read as a study of hEDS as defined today.

5. EDS types

Every record names the EDS types it applies to. A finding about hEDS is not applied to vEDS or other types with a known gene: the conditions share a name but differ in cause, risk, and management. Records that apply to all types say so.

6. How settled each claim is

  • Established: at least one clinical source at one of these levels: systematic review, meta-analysis, randomized trial, cohort study, case-control study, case series, clinical guideline, consensus statement. A research program record may instead rest on a registry source.
  • Probable: supported by other clinical or registry evidence.
  • Emerging: new and not yet replicated. Always has a reassessment date.
  • Contested: sources or studies disagree. Always has a reassessment date.
  • Community signal: a pattern patients report consistently that clinical research has not tested. Always has a reassessment date.
  • Historical record: something documented in the past. Listing it records that it happened, not that it works.
  • Refuted: contradicted by stronger evidence. Kept on the site and labeled, not deleted.

7. Review dates

Every record carries the date it was last reviewed. Emerging, contested, and community-signal records cannot be published without a reassessment date. The publication policy sets target intervals of 60 days for emerging records, 90 for contested, 90 for community signals, and 365 for established records. The build does not enforce these intervals.

8. Publication policy

These rules are quoted from the published policy file:

  • The index describes evidence and provenance; it never recommends, prescribes, or discourages a specific course of care for an individual.
  • Community evidence is recorded by venue. No individual poster, username, or verbatim post is recorded.
  • Folk and historical practices are listed as documented history. Listing a practice records that it happened, not that it works.
  • Every record declares its subtype scope. hEDS findings are never silently generalized to vEDS or other monogenic types.
  • Records carry the diagnostic-criteria era of their sources so pre-2017 claims are not read as modern nosology.
  • Records with more than one kind of evidence must state whether that evidence is convergent, from one underlying source, contested, or refuted. Disagreement is published.
  • Practices with documented harm potential carry a visible risk note rather than silent omission.
  • The log of research runs and decisions is append-only. A correction adds a new entry and marks a wrong record refuted rather than silently rewriting it.
  • Emerging, contested, and community-signal records must carry a reassessment date; undated emerging claims are not published.

Two kinds of entry can go live without review, and only as bibliographic facts: trial registrations and journal articles (title, venue, date, and identifier). Everything else needs review: every clinical claim, practice assessment, and risk note; any reading of a registry beyond its own fields; and all community, historical, and gray literature material. Preprints enter provisionally and are marked emerging.

9. Searches and the change log

The index lists the searches it uses to find new evidence (PubMed, ClinicalTrials.gov, society news, preprint servers, guideline bodies, community venues, and historical archives) and how often each should run. Each update to the data is logged with its date, the searches used, and what changed. Corrections add a new log entry; earlier entries are not edited.

10. What this index does not do

It does not diagnose. It does not recommend or discourage any course of care for any individual. It does not treat forum reports as clinical evidence, and it does not treat documented folk practice as treatment. When evidence is weak or contradictory, the record says so and says why.