The state of hospital website accessibility, 2026

Of the 99 health-system homepages we tested, the homepage alone showed:

• 66% had at least one serious accessibility failure (a critical or serious WCAG issue), about two in three.
• 35% had at least one critical issue, more than a third, the kind that can hard-block a screen-reader user.
• Only 12%, about one in eight, passed every automated check.
• The median site had 8 detectable issues on its homepage; the worst had 75.

These are detectable failures from an automated scan. They are the floor, not the ceiling (more on that below). And to be clear: this is 66% of the health systems we tested, a curated sample of large and well-known providers, not a census of every hospital.

We ran an identical scan over 90 online stores earlier this year, which lets us compare two sectors on the same yardstick. Healthcare came out ahead on every headline measure:

• At least one serious failure: 66% of hospitals vs 81% of stores.
• At least one critical issue: 35% vs 49%.
• Passed every automated check: 12% vs 8%.
• Median issues per homepage: 8 vs 17, roughly half.

The likely reason is not that hospitals write more careful CSS. It is that health systems tend to run on institutional content platforms with in-house web teams and, unlike most retailers, a sector-specific reason to care about access. Even so, a third of these homepages had a critical barrier on the first page, which is a low bar to be tripping over.

Ranked by the share of tested sites affected, the recurring WCAG problems were:

• Colour contrast (28%) — text that does not meet the 4.5:1 ratio, WCAG 1.4.3. As with the stores, this was the single most common WCAG failure, and our figure includes text on CSS gradients that most scanners skip.
• Touch target size (12%) — controls smaller than the 24×24px minimum, WCAG 2.5.8.
• Buttons with no accessible name (12%) — usually icon-only buttons, WCAG 4.1.2.
• ARIA elements missing required children (11%) — custom menus, tab sets, and listboxes that are wired up incorrectly, WCAG 1.3.1. This one showed up more on hospital sites than on stores, and it tends to sit on exactly the interactive widgets that matter here: appointment finders, provider search, portal menus.
• Images missing alt text (9%) — WCAG 1.1.1.
• Links with no discernible text (9%) — usually icon-only links, WCAG 2.4.4 / 4.1.2.

Structural best-practice issues were more widespread than any single WCAG failure: content not contained in landmarks showed up on 52% of sites, and out-of-order headings on 29%. Those are advisory rather than WCAG success-criterion failures, so we have kept them separate from the figures above, but they are the kind of thing that makes a page harder to navigate by screen reader.

A checkout you cannot use costs you a sale. A hospital site you cannot use can stand between someone and their care, and the people most likely to depend on assistive technology are, disproportionately, the people who need healthcare.

Healthcare also carries obligations most sectors do not. In 2024 the US Department of Health and Human Services adopted WCAG 2.1 Level AA as the accessibility standard for the websites and mobile apps of providers that receive federal funding, under Section 504, with compliance phasing in over the following years, and Section 1557 of the Affordable Care Act bars disability discrimination by those same programmes. We are not offering a legal reading of any particular site here. The point is only that for hospitals, the machine-checkable issues above sit on top of a standard that is already named in a rule.

Here is the number that matters most for honesty: the median site also had 15 items flagged as "needs human review", things an automated test cannot decide on its own, like whether alt text is accurate or whether a custom appointment widget is actually operable by keyboard.

Automated testing only covers the machine-checkable part of WCAG, which independent research puts at roughly a third of the success criteria. So the failure rates above are almost certainly an undercount of the real barriers on these sites, not an exaggeration. Hospitals doing better than stores on the automated checks does not mean their sites are actually easier to use, only that they trip fewer of the traps a machine can see.

The takeaway is not that these health systems are breaking the law, automated failures are not a legal compliance verdict, and we are naming no one. The takeaway is that even in a sector with a specific reason to get this right, a serious accessibility failure on the homepage is still the common case, and that the fixes are mostly ordinary front-end work: contrast, alt text, button and link names, correctly built widgets.

These are fixes that belong in the code itself, not in a layer added on top of it, which is why the durable path is to find the real issues and correct them at the source. You can see where your own site stands with the free scan, or wire it into CI with the CLI and GitHub Action so it does not regress after the next release.

In the interest of not overclaiming:

• Sample: homepages of 99 US health systems, scanned in June 2026. We attempted 101; 2 were excluded because they returned bot-protection or error pages instead of the real homepage. This is a curated sample of large, recognisable providers, not a random or traffic-ranked one, so read it as a snapshot of major health systems rather than a statistical census of all hospitals.
• Engine: the axe-core rules for WCAG 2.x Level A and AA, plus our own resolution of colour contrast over CSS gradients. The same engine and settings we used for the stores report, so the two are directly comparable.
• Scope: homepages only. Patient portals, provider search, and appointment booking flows often fare worse, and are exactly where the interactive-widget failures above tend to concentrate.
• As loaded: pages were scanned as they rendered, including any cookie or consent overlays, which can both add and mask issues.
• Not a compliance audit: detectable WCAG failures are not a legal determination, and automated testing does not cover the criteria that need a human. We have reported aggregate figures only and named no individual provider.