WCAG video requirements for medical device training videos

A hospital customer sends over a vendor accessibility questionnaire. A federal buyer asks for an Accessibility Conformance Report. A clinical educator mentions that one of the nurses in last month's in-service is hard of hearing. Each of these lands on the same question: do our device training videos meet accessibility requirements, and what would it take to fix them?

For video, the standard almost everyone points to is the Web Content Accessibility Guidelines (WCAG) from the W3C. This guide translates the parts of WCAG that apply to prerecorded video into decisions a device team can make during scripting, not after the final cut. It includes a worked example and a checklist you can run against your own library.

Where the requirement comes from

Accessibility requirements usually reach a device company through its customers, not through its own product regulators. Three routes come up most often in US work.

On Section 508, federal agencies ask vendors to submit an Accessibility Conformance Report (ACR), most often written on the IT Industry Council's VPAT template, and the government generally can't proceed with a purchase without one unless an exception applies. If training content or software is part of what you deliver to a federal customer, expect it to be evaluated.

On Title II, timing changed this year. The 2024 rule set compliance dates of April 24, 2026 for entities serving populations of 50,000 or more and April 26, 2027 for smaller entities and special districts. On April 20, 2026, the Department of Justice published an interim final rule pushing each date back a year, to April 26, 2027 and April 26, 2028. If a public university health system shows your training videos on its learning platform, its accessibility team may still ask you for conforming files well before then.

What WCAG video requirements actually ask for

WCAG groups video rules under Guideline 1.2, Time-based Media. For a prerecorded training or product video with sound, four success criteria do most of the work.

Two details matter more than people expect. First, WCAG defines captions as covering both speech and non-speech audio, including sound effects, music and speaker identification. A caption track that skips the occlusion alarm in an infusion pump video is missing the one sound a clinician most needs to recognize. Second, the W3C's guidance on 1.2.5 says that if the existing audio already conveys all the important visual information, no additional audio description is necessary. That one sentence is the key to doing this affordably.

Write the description into the script

Separate audio description is expensive to add after the fact, because it has to fit into pauses that the edit didn't leave. The W3C's media guidance describes the alternative: integrated description, where the visual information is carried in the main narration. It notes that this is easier and better for accessibility, and that it needs to be in the script before production.

Device training videos are well suited to this, because the narration is already instructional. The fix is usually a few words per step, telling the viewer what the hands and the screen are doing instead of saying "like this."

The second version helps more than blind or low-vision viewers. It also helps the rep watching on a phone in a parking lot and the nurse half-listening while setting up a room. It also tends to match the IFU wording more closely, which helps in medical, legal and regulatory (MLR) review. Our IFU to training video method builds this in from the first script table.

Worked example: an accessible setup video

Fictional example. "Larkfield Medical" (a fictional company) needs a four-minute setup video for its fictional "Vela" bedside pump, and a public university hospital customer has asked for WCAG 2.1 AA. Here is how we would plan it.

1. Script with integrated descriptionEach step in the script table has a narration line that names the control, its location and the visible result. On-screen text callouts are read aloud or restated in narration.
2. Mark non-speech soundsThe script lists every meaningful sound (start tone, occlusion alarm, door latch click) so captions include them, for example [high-pitched alarm repeats].
3. Check flashes in the animaticThe status LED animation blinks at 2 per second, under the three-per-second limit. Fast transitions are replaced with crossfades.
4. Deliver closed and open captionsClosed captions as SRT and WebVTT sidecar files for the hospital's learning platform, plus an open-caption version for kiosks and social clips.
5. Write the descriptive transcriptA text document with all narration, sounds in brackets and short descriptions of anything visual the narration doesn't cover. It doubles as a review aid for MLR.
6. Record what was doneA one-page note lists the WCAG criteria addressed and how, so the customer's accessibility team has what it needs for its own records.

Nothing in that plan adds a new production phase. Most of the work moves earlier, into the script and the animatic, where changes are cheap.

Caption files, formats and quality

WCAG distinguishes closed captions, which viewers can turn on and off in some players, from open captions, which can't be turned off. Either can meet 1.2.2. In practice we deliver both: closed captions for platforms that support them, and open captions for places where the player is unknown or the sound is off by default.

For closed captions, WebVTT is the W3C format designed for use with the HTML track element, and it covers captions, subtitles and text video descriptions. Many platforms also accept SRT, a simpler format. Ask the customer which their platform prefers; delivering both costs almost nothing once the timing is done.

For caption quality beyond simply being present, a useful benchmark is the FCC's 2014 caption quality standards for television: accuracy, synchronicity, completeness and placement. They were written for TV programming, not training video, but they are a fair test. Captions should match the words spoken, appear in step with the audio and stay up long enough to read, run from start to end, and not cover important visual content. That last point matters in device video, where the bottom of the frame often holds the control being demonstrated.

A checklist for your video library

• Every video with audio has captions, and the captions include meaningful non-speech sounds such as alarms and tones.
• Captions were checked by a person against the final narration, not only generated automatically.
• Device terms, drug names and units in captions match the IFU spelling exactly.
• The narration names controls, locations and visible results, so the video makes sense with the screen off.
• On-screen text that carries information (warnings, settings, values) is also spoken or included in the description.
• Nothing flashes more than three times in any one second.
• Captions don't cover the control or display being demonstrated.
• A descriptive transcript exists for each video, with the same version number as the video.
• Closed caption files (SRT, WebVTT or both) are delivered with the master, and an open-caption version exists where needed.
• Your regulatory or legal team has confirmed which WCAG version and level the customer requires.

Every MedTech Reel project includes captions plus an SRT file as standard, and integrated description is part of how we write scripts. If you're planning a refresh of an existing library, our medical device training videos page explains how we work from your approved materials.

General information as of the date of this post, not legal or regulatory advice. Accessibility obligations depend on your customers, contracts and how each video is used, so confirm requirements with your regulatory and legal teams.

Video summary

FAQ

Related services: Training videos · Video localization

Keep reading

Training video

From IFU to training video: a step-by-step method for medical device teams

A nine-step method for turning approved Instructions for Use into a training video that stays on label, moves through MLR review faster and survives the next IFU revision.

Oct 24, 2026 · 9 min read

Localization

EU MDR language requirements for medical device videos

Which device videos need translating for an EU launch? How MDR Article 10(11), the Commission language table and eIFU rules apply, plus a checklist.

Nov 3, 2026 · 8 min read