Snellen vs. ETDRS: When to Use Each in Clinical Practice

Snellen and ETDRS charts both measure visual acuity, but they are built on different principles and answer different clinical questions. Understanding when to reach for each one — and how to convert between them — keeps your documentation defensible and your measurements comparable across visits.

How the Snellen chart works

The familiar Snellen chart, introduced in 1862, presents a large single letter at the top and progressively more letters per line as acuity improves. Letter sizes step down in irregular increments, the number of letters per line varies, and the spacing between letters is not standardized. This makes Snellen fast and intuitive for routine screening, but it also introduces variability: the 20/20 line has more letters and tighter crowding than the 20/200 line, so the test is not equally difficult at every level.

How the ETDRS chart works

The ETDRS chart was developed for the Early Treatment Diabetic Retinopathy Study to remove that variability. Its design is deliberately uniform:

• Five letters on every row — equal crowding at every acuity level
• Equal logarithmic progression — each row is 0.1 logMAR smaller than the one above it
• Equal spacing between letters and rows, proportional to letter size
• Letter-by-letter scoring — each of the 5 letters is worth 0.02 logMAR, giving finer resolution than line-by-line Snellen scoring

Because every line is equally difficult, ETDRS produces more reproducible measurements — which is exactly why it is the required standard for clinical trials and longitudinal disease monitoring.

When to use each

Use Snellen for	Use ETDRS / logMAR for
Routine acuity screening	Retinal disease monitoring (AMD, diabetic retinopathy)
Fast throughput clinics	Anti-VEGF treatment response tracking
General refraction checks	Clinical trials and research protocols
Patients with normal vision	Low-vision and small inter-visit change detection

Converting between Snellen and logMAR

logMAR (the logarithm of the minimum angle of resolution) is the score derived from ETDRS testing. The conversion is fixed:

Snellen (US)	Snellen (metric)	logMAR
20/200	6/60	1.0
20/100	6/30	0.7
20/40	6/12	0.3
20/20	6/6	0.0
20/16	6/4.8	-0.1

A lower logMAR is better, and 0.0 logMAR equals 20/20. A change of 0.1 logMAR corresponds to one ETDRS line — or five letters.

Scoring ETDRS letter by letter

The single biggest gain from ETDRS comes not from the chart but from how it is scored. Line-by-line Snellen scoring — “20/40 minus two” — discards most of the information the patient just gave you. ETDRS is scored by counting every letter read correctly across the whole chart, and because each letter is worth 0.02 logMAR, the letter count converts directly into a score.

The exact arithmetic depends on which chart version you are using and which line you start from, so follow the scoring card that came with your chart rather than a formula from memory. The principle is constant: five letters is one line is 0.1 logMAR.

Push to forced choice. The protocol asks the patient to guess at every letter, even when they say they cannot see it, and to keep going until they miss all five on a line. Patients who are allowed to stop when a line “looks blurry” routinely score one to two lines worse than their true acuity, and that error is not consistent between visits — which is precisely the variability ETDRS exists to remove.

What counts as a real change

Because ETDRS lines are equally spaced in logMAR, change means something specific:

• 5 letters = one line = 0.1 logMAR
• 10 letters = two lines = 0.2 logMAR
• 15 letters = three lines = a doubling of the minimum resolvable angle

That 15-letter threshold is the convention retinal trials use as a clinically meaningful change, and it is the reason anti-VEGF response is documented in letters rather than in Snellen fractions. A patient who goes from 20/80 to 20/40 has gained three lines; describing that as “improved to 20/40” records the endpoint but loses the magnitude.

Test–retest variability on a well-run ETDRS chart is roughly one line in eyes with good vision and wider in low vision. A single line of apparent change, on its own, is not evidence of anything.

Why the same eye scores differently on each chart

Two charts, one eye, two numbers — and the difference is systematic rather than random. Snellen’s upper lines carry one or two widely spaced letters, so there is almost no crowding to contend with; the lower lines are dense. An eye with reduced acuity is therefore tested under easier conditions on Snellen than on ETDRS, and Snellen tends to report better acuity than ETDRS in low vision, with the gap widening as vision worsens.

The practical consequence: do not compare a Snellen measurement at one visit with an ETDRS measurement at the next and treat the difference as disease progression. If a patient is being monitored, the chart type, the testing distance, and the refractive state all have to stay constant, and all three belong in the record alongside the number.

When acuity falls below the chart

If a patient cannot read the top line at the standard testing distance, the next step is to move them closer rather than to jump straight to counting fingers. Halving the distance makes every optotype subtend twice the angle, which recovers the ability to measure — and the score is then adjusted for the distance actually used, which is why the distance must be recorded.

Counting fingers, hand motion, and light perception are useful floors but they are coarse. Two patients recorded as “CF at 2 feet” can have meaningfully different vision, and neither can be tracked for the small changes that treatment decisions turn on. Exhaust the chart first.

Five habits that protect reproducibility

1. Record the chart, the distance, and the correction with every measurement. A number without those three is not comparable to anything.
2. Do not switch chart types mid-course for a patient under monitoring.
3. Keep the testing order constant — same eye first, every visit. The second eye tested benefits from familiarity with the letters.
4. Randomize or rotate optotypes where the system allows it. Patients seen frequently do memorize charts, and a memorized line reads as improvement.
5. Hold the room lighting steady. Chart luminance and ambient light both affect the result, and neither is usually written down.

The 20/20 misconception

20/20 is not perfect vision or an aspirational target; it is the acuity of a normal eye resolving a letter subtending five minutes of arc at the testing distance. Plenty of healthy young eyes read 20/15 or 20/12.5, which is why ETDRS charts extend past 0.0 logMAR into negative values. Recording “20/20” when the patient in fact read two further lines discards real clinical information — and it removes the headroom you would need later to detect an early decline.

How AcuityMaster handles both

AcuityMaster includes both Snellen (US, metric, and decimal notation) and full ETDRS/logMAR charts in one application, with optotypes calculated from your patient-to-screen distance, designed to ANSI Z80.21 and ISO 8596. You can switch between them instantly without changing hardware, so the same lane supports a fast screening and a research-grade measurement on the next patient.