GlossCap

Compliance reference · Affordable Care Act § 1557 · 45 CFR Part 92 · 89 Fed. Reg. 37,522

Section 1557 captions: the ACA nondiscrimination rule, 92.207 effective communication, and the deadline cascade

Section 1557 of the Affordable Care Act (42 U.S.C. § 18116) is the federal nondiscrimination rule that reaches every healthcare-program operator that receives federal financial assistance — and the 2024 HHS final rule (89 Fed. Reg. 37,522, published 2024-05-06) is the implementation that the Office for Civil Rights enforces. For training video and patient-education video, the operative provisions are 45 CFR § 92.207 (effective communication for individuals with disabilities) and § 92.205 (auxiliary aids and services). The rule's compliance dates phase in: most provisions on 2025-05-01, the WCAG 2.1 AA web-and-mobile-content provisions on 2026-05-06 (large covered entities) and 2027-05-06 (small covered entities). Section 1557 reaches a wider operator set than HIPAA does, and through a different enforcement pathway — its substantive captioning bar is what every health-program tenant should plan against this calendar year.

TL;DR

Section 1557 of the ACA prohibits discrimination on the basis of race, color, national origin, sex, age, or disability in any health program or activity that receives federal financial assistance, is administered by an Executive agency, or is established under Title I of the ACA. The HHS final rule at 45 CFR Part 92 (89 Fed. Reg. 37,522, 2024-05-06) implements it. For captions specifically: § 92.207 (effective communication) requires covered entities to take appropriate steps to ensure communications with individuals with disabilities are as effective as communications with others, including providing auxiliary aids and services where necessary. § 92.205 (auxiliary aids and services) lists captions on video among the auxiliary aids. § 92.106 / § 92.107 (web-and-mobile content) requires conformance with WCAG 2.1 AA on web content and mobile apps. The compliance date for most provisions is 2025-05-01; the WCAG provisions phase in at 2026-05-06 (large covered entities, ≥15 employees) and 2027-05-06 (small covered entities, <15 employees). Enforcement is through OCR HHS investigations — initiated by individual complaint, by compliance review, or by referral. Section 1557 reaches the entity broadly: every covered entity, not just the federal-funded program inside it, is in scope after the 2024 rule restored the broad-application reading the 2020 rule had narrowed.

What Section 1557 reaches — the covered-entity set

The scope question is the first thing every health-system compliance lead asks: who's in? The 2024 final rule's answer is broad. Covered entities under 45 CFR § 92.4 include:

Note that Medicare Part B participation, on its own, was historically debated as a federal-financial-assistance trigger. The 2024 rule clarifies that Part B participation does count, which broadens the operator set materially — most physician practices that bill Medicare are now in scope.

How Section 1557 differs from HIPAA

This is the most common scoping confusion. Section 1557 and HIPAA reach overlapping operator sets but are different statutes addressing different concerns:

Practically: a covered entity might be in scope of both for different content. The HIPAA training-video catalogue (workforce training on PHI) needs captions because Section 504 and ADA Title II reach the workforce-training surface. The patient-education video catalogue (cardiac-rehabilitation instructions, diabetes-education, post-discharge patient-care videos) needs captions because Section 1557 § 92.207 reaches it through the effective-communication frame. The captioning workflow is the same; the regulatory rationale differs.

§ 92.207 — Effective communication (the operative caption text)

45 CFR § 92.207 is the operative effective-communication provision. The substantive text:

"A covered entity shall take appropriate steps to ensure that communications with individuals with disabilities are as effective as communications with others. The covered entity shall furnish appropriate auxiliary aids and services where necessary to afford individuals with disabilities, including applicants, participants, beneficiaries, customers, and companions, an equal opportunity to participate in, and enjoy the benefits of, the entity's health programs or activities."

"Auxiliary aids and services" is defined at § 92.205. For deaf, hard-of-hearing, deafblind, and other individuals for whom auditory communication is inaccessible or less effective, auxiliary aids include:

The effective-communication standard is functional, not merely technical. A patient-education video that has technically-present captions but mangles the procedure name and the medication dosage doesn't communicate as effectively to a deaf patient as the audio communicates to a hearing patient — and § 92.207 reaches that gap.

§ 92.106 / § 92.107 — Web-and-mobile content accessibility

The 2024 final rule added explicit web-and-mobile-content accessibility provisions at § 92.106 and § 92.107. The substantive bar is conformance with WCAG 2.1 Level AA for web content and mobile applications used to provide programs and services. The relevant Success Criteria for video specifically:

Compliance dates phase by entity size:

The phased deadline gives most operators about a year (large covered entities) or two years (small covered entities) from the time of writing to bring web and mobile content into WCAG 2.1 AA. Patient-education video on the entity's website, telehealth-platform recorded sessions, post-discharge-instruction videos, and any other web-served content is in scope.

The OCR HHS investigation pattern

The Office for Civil Rights at HHS is the enforcement body. Investigations are initiated by:

  1. Individual complaint. A patient, family member, or program participant files a complaint with OCR (web form, mailed letter, or in person at an OCR regional office). OCR triages and frequently opens the investigation within weeks.
  2. Compliance review. OCR can open a review on its own initiative, often informed by a sectoral push (post-discharge-care content sweep, Indian Health Service review, Marketplace plan review).
  3. Referral. Other federal agencies (DOJ, CMS) refer cases.

The investigation pattern, when it lands on a healthcare covered entity:

  1. Identify the relevant program or activity (specific patient-care service, the training program named in the complaint).
  2. Request the auxiliary-aids-and-services policy, the staff-training records around accessibility, and the relevant content (videos served to patients, training videos served to staff).
  3. Sample the content. The investigator opens patient-education and training videos, watches a slice with captions on, reads the caption track against the audio.
  4. Identify substantive failures. Mangled drug names, mangled procedure names, mangled instrument names, mangled regulatory citations, mangled provider names — the proper-noun mangling pattern that auto-captioning produces predictably.
  5. Open a finding letter or move to a voluntary resolution agreement. OCR's preferred remedy is a voluntary resolution with a corrective-action plan; persistent or willful non-compliance escalates to the Department of Justice.

The "watch a slice" step is what catches generic auto-captioning. A patient-education video on diabetic foot care that mangles "metformin" and "neuropathy" and "podiatrist" doesn't communicate as effectively to a deaf patient as the audio does to a hearing patient — and that observation, written into a finding letter, is hard for the entity to refute.

The covered-entity content set — what's typically in scope

For a hospital, health system, large group practice, or large nursing-facility operator, the Section-1557-relevant video catalogue typically includes:

Patient-education and discharge content is the highest-stakes layer. The communication failure mode — a deaf patient reading "neuropathy" as the auto-caption mangles it — has direct clinical consequences in a way that mangled hospital-marketing video does not.

The proper-noun failure mode in healthcare patient-education content

Healthcare operators who watch the OCR investigation pattern report consistent surfaces:

The shared pattern: the words a patient must understand to follow medical instructions or know which provider to see are exactly the words generic STT has the least training data for.

The healthcare retrofit pattern

For a covered entity sitting on a patient-education and training-video catalogue without substantively-accurate captions, the retrofit:

  1. Inventory. Cross-reference the entity's content management systems — the patient-education library (often Krames, EBSCO, Healthwise, UpToDate), the entity-authored content on the website, the LMS catalogue (commonly Healthstream or Relias), the telehealth-recording archive, the entity's YouTube channel.
  2. Triage by stake. Patient-education videos with clinical-instruction content first; discharge-instruction videos second; behavioral-health and chronic-care content third; marketing and welcome content last.
  3. Build the entity's medical glossary. Drugs in the formulary, procedures the entity performs, diagnostic codes in the EHR, provider names and titles, the entity's specific service-line language. The glossary is built once per entity and applies to every retrofit asset.
  4. Re-caption with glossary-biased output. Replace mangled or absent captions with substantively-accurate caption tracks aligned to the entity's medical glossary.
  5. Publish back to each surface. Sidecar SRT/VTT for entity-hosted video, replace caption track in the LMS, publish to the patient-portal video player, work with the patient-education-content vendor (Krames, etc.) on their captioning posture.
  6. Document for the OCR audit-evidence pack. Asset register: video URL, caption file, source, reviewer, review date, glossary version, the auxiliary-aids-and-services policy reference. This is what answers an OCR documentation request.

See pricing

Section 1557, alongside Section 504, ADA Title II, and Title III

For healthcare operators, the regulatory layering is dense. The substantive captioning bar is the same across all of them; the enforcement pathway differs:

For an in-scope operator, the cleanest posture is to caption to the WCAG 2.1 AA bar globally and let the per-statute audit-evidence pack pull from a single asset register. Glossary-biased captioning is what makes the substantive-accuracy bar achievable across the catalogue.

FAQ — Section 1557 captions

Are auto-captions ever sufficient under § 92.207?

The effective-communication standard is functional. For a low-stakes, no-clinical-content video — a thank-you message, a procedural overview without instructions — auto-captions can technically satisfy "captions exist." For patient-education with clinical-instruction content (medication administration, post-procedure care, chronic-disease management), auto-captions virtually always fall below the effective-communication bar. The defensible posture is to treat auto-captions as a draft and run a glossary-biased correction before exposure to a patient.

If we're a small physician practice (under 15 employees), does Section 1557 reach us?

If the practice receives federal financial assistance — Medicare Part A, Medicare Part B (per the 2024 rule), Medicaid, or HHS-grant participation — yes. The web-and-mobile content provisions phase to 2027-05-06 for under-15-employee covered entities; the effective-communication provisions at § 92.207 apply at the 2025-05-01 baseline.

Does Section 1557 apply to telehealth recordings?

Where the recording is preserved and re-used to provide health-program communications (post-encounter review, chronic-care management, second-opinion sharing) — yes. The recording is now prerecorded video, within the auxiliary-aids-and-services and the prerecorded-content WCAG SCs.

What's the relationship between § 92.207 (effective communication) and § 92.205 (auxiliary aids)?

§ 92.207 is the substantive standard ("communications shall be as effective"). § 92.205 lists the means of meeting it (qualified interpreters, captions, real-time captioning, etc.). § 92.205 is the "what you can use" list; § 92.207 is the "what you must accomplish" standard.

Does Section 1557 apply to insurance plans?

Health insurance issuers offering qualified health plans through Title-I-of-the-ACA marketplaces are in scope, as are issuers receiving federal financial assistance from HHS. Plan member-facing video — explanation of benefits, plan-comparison videos, member-portal walkthroughs — is in scope through § 92.207 + § 92.106/§ 92.107.

If our content vendor (Krames, Healthwise) provides the patient-education video, whose captioning obligation is it?

The covered entity's. Section 1557 reaches the entity's communications regardless of who produces the underlying video. Most patient-education-content vendors offer captioned content as standard or as a paid add-on; many do not produce captions to the substantive-accuracy bar that the OCR investigation pattern enforces. The covered-entity-side compliance posture has to verify the vendor's captions on a sampling basis, and the captions need to be re-captioned where the entity-specific glossary (formulary, provider names, service-line language) is mangled.

What's the relationship to OCR's HIPAA enforcement?

Different enforcement pathway, often the same investigator at the regional OCR office. A complaint that names both an effective-communication failure (Section 1557) and a privacy concern (HIPAA) gets opened jointly. The captioning workflow doesn't change between them — substantively-accurate captions on the patient-education video clear the § 92.207 bar; HIPAA addresses the separate question of how PHI is handled in workforce training and patient communications.

Further reading