Accessibility

axe-core 4.13 was driven through a real browser over every route in the system: every clinical tool, the catalogue, and every page in the footer, each of them in both the light and the dark theme and at both a desktop and a phone width. Every rule set was enabled: WCAG 2.0 A and AA, WCAG 2.1 A and AA, and the tool's own best-practice rules.

The pages are allowed to settle before anything is measured. An earlier run that did not wait reported hundreds of contrast failures which turned out to be animations caught halfway through, in colours that appear in no stylesheet. A measurement that has not been checked is worth as little as no measurement.

Keyboard navigation and the tab order were then walked by hand. No automated tool can tell you whether an order makes sense.

• Text contrast meets AA everywhere it was measured, in both themes and across all three accent colours, including the unit hints beside inputs and the help text under them.
• The whole system works from the keyboard. Every control can be reached and operated by tab and by enter or space, and the control that has focus is always visibly outlined.
• The page has landmarks. A banner, a main region, the catalogue as a complementary region, the footer as navigation. A screen reader can jump between them instead of reading from the top of every page.
• Every page has one first-level heading and the headings below it run in order, which is how a screen reader user reads a long pathway.
• Every input has a label bound to it, so tapping the label focuses the field and a screen reader announces what the field is for.
• The dark theme follows your device unless you choose otherwise, and both themes were measured, not just the light one.
• The page zooms. Pinch zoom is not disabled, and text reflows rather than forcing a sideways scroll. Tables stop being tables on a narrow screen and stack into labelled cards.
• Animation is honoured as a preference. If your device asks for reduced motion, transitions and animations are suppressed.
• Nothing depends on colour alone. Where a result is banded by severity, the band is also named in words.

Stated plainly, because a statement that lists only successes is not a statement.

• No audit by a person using assistive technology. Automated rules catch roughly a third to a half of real barriers. Nobody has yet sat down with a screen reader and worked through a full clinical pathway here, and that is the single most useful thing anyone could offer this system.
• No independent or certified audit, and no formal conformance claim. What is on this page is a self-assessment with its method published so that you can repeat it.
• The printed and PDF output has not been separately assessed for contrast or structure, only the screen.
• The optional sign-in area is not covered. It has not been released and will be measured before it is.
• Content, not just code. Some pathways are long and dense by nature. Making them readable for someone with a cognitive or reading difficulty is a writing problem rather than a markup problem, and it is not solved.

The target is the Web Content Accessibility Guidelines 2.1, level AA, which is the level Indian guidance for public-facing web content points to and the one most national requirements are written around. The Rights of Persons with Disabilities Act 2016 requires that information and communication technology be accessible; WCAG is how that is measured in practice.

There is also a plainer reason. A clinician reading a dose on a ward at three in the morning, on a cracked phone screen, in bad light, with tired eyes, is not far from the reader these guidelines are written for. Almost everything done for accessibility here makes the tools better for everybody using them.

Tell us, and please be as specific as you can bear to be. Which tool. What you were using: a screen reader, the keyboard only, magnification, a particular browser. What happened. A barrier that stops you reaching a dose is treated with the same urgency as a clinical error, because in practice that is what it is.

You do not need an account and you do not need to leave your name.

Report an accessibility barrier