The four-flight stair test: what your time can and can’t tell you

The evidence is a 2021 scientific letter from Peteiro and Bouzas-Mosquera in Spain. They asked 165 patients who had been referred for exercise testing (mean age 66, 55% men) to climb four flights, 60 stairs, "at a fast pace without stopping but also without running". The same patients then did a treadmill test, which gave their fitness in METs.

Mean stair times by treadmill fitness:

• 10 METs or more: 46 ± 11 seconds
• 8–9.9 METs: 58 ± 21 seconds
• Under 8 METs: 82 ± 41 seconds

The figure after ± is the standard deviation, a measure of spread. Within the fittest group, times commonly ran about 11 seconds either side of 46. For scale, one MET is 3.5 mL/kg/min, so 10 METs is a VO₂max of about 35.

Stair time and treadmill METs correlated at r = 0.53. That is a moderate link: two people with the same time can have quite different fitness.

The authors tested two cut-offs. At 60 seconds, the sensitivity for reaching 10 METs was 94%: almost everyone who reached 10 METs climbed in a minute or less. The negative predictive value was 93%: of those who took longer than a minute, 93% did not reach 10 METs. A slow time, in this group, was informative.

A fast time is less so. The 8–9.9 MET group averaged 58 seconds, so many people under a minute did not reach 10 METs. At the other end, a 90-second cut-off had 96% specificity for being under 8 METs: taking longer than 90 seconds was rare among people who reached 8 METs or more.

The spread in the least fit group, 82 ± 41 seconds, shows how mixed it was. These were symptomatic patients in a clinic, not healthy adults, and the letter reports no deaths or other long-term outcomes.

A line repeated widely online says that climbing four flights in under 40–45 seconds means good fitness, often put as more than 9–10 METs. It appears to come from press coverage of this research. The published letter does not support it.

What the letter reports is a mean of 46 seconds for people who reached 10 METs. An average of 46 means many people in the fittest group took longer than 45 seconds, and the letter tests no cut-off at 40 or 45. We do not use the figure, and you should not read your time against it.

Two large cohorts asked people how many stairs they climb, then followed them. In UK Biobank, 280,423 people over 11.1 years, climbing 6–10 flights a day at home against none went with a hazard ratio of 0.91 (0.85–0.98): a 9% lower death rate, about 44–55 extra days of survival. In 8,874 Harvard alumni over 12.4 years, 35 or more floors a week against fewer than 10 went with 0.84.

Neither study found a link with cardiovascular disease. UK Biobank found a similar link with lung cancer, an outcome stairs should not affect, which points to confounding: people who climb stairs differ in other ways.

These are self-reported habits, not a timed test. They say nothing about what your four-flight time predicts.

Short timed stair climbs have been tested for reliability as measures of leg power in older adults. A 4-step stair climb power test in 50 people aged 65 and over had an intraclass correlation of 0.951 between sessions, and correlated with other leg-power measures at r = 0.85–0.96. A 12-step version in older men had ICCs of 0.91–0.97 and picked up a response to testosterone that gait speed missed.

These are power tests (body mass × height climbed ÷ time) for older, often mobility-limited people, not a four-flight fitness test. They suggest that the same stairs, timed the same way, give a repeatable number. They are not a source of norms.

Longstanding times four flights, 60 stairs, the same count Peteiro used, from the bottom step to both feet on the top landing. No population norms exist for this test, so the app’s bands are its own estimates and are labelled as such. They sit near Peteiro’s patient means but are not derived from them.

Peteiro’s patients walked fast without running. If you run, your time is not comparable with theirs. Pick one rule and keep it, on the same stairs, every quarter: your own trend is the most useful thing this test gives you. Hold the rail if you need it.

Read a single time modestly. Under a minute says you are not in the slow range of a clinical sample; it does not place you on a percentile, and it tracks treadmill fitness only moderately. For that, a field test or a lab is the better tool.

If stairs bring on chest pain or unusual breathlessness, that is a reason to see a doctor, not a score to log. This is not medical advice. The protocol is on the stair test page.