Six-Minute Walk Test

Six-Minute Walk Test in Cardiopulmonary Trials and Economic Evaluation

The six-minute walk test, usually shortened to 6MWT, asks a patient to cover as much ground as possible on foot in six minutes, and the result is reported as the six-minute walk distance (6MWD) in metres. For a health economist the test matters because it has carried the primary endpoint of many trials in pulmonary arterial hypertension and other cardiopulmonary conditions, so the clinical evidence behind a reimbursement decision is often expressed in metres walked and not in quality-adjusted life years. The distance has no preference weight, which means an analyst has to decide how a gain of, say, 30 metres turns into costs, utilities and survival in a model. Published minimal important differences of about 25 to 33 metres in chronic respiratory disease, and about 33 metres in pulmonary arterial hypertension, give the usual yardstick for judging whether a trial difference is large enough to matter to patients.

How the test is set up and scored

The American Thoracic Society (ATS) published its consensus guidelines for the test in 2002, approved by the ATS Board of Directors in March of that year. The statement covers indications, factors that affect results, a step-by-step protocol, safety measures, patient preparation and the clinical interpretation of results, and it drew on a Medline search of the literature from 1970 to 2001. It describes the 6MWT as a practical, simple test that requires a 100-foot hallway but no exercise equipment or advanced training for technicians.

A joint European Respiratory Society and ATS technical standard, led by Holland and published in 2014, updated that guidance for chronic respiratory disease. It describes the test as a self-paced test of walking capacity in which the patient is asked to walk as far as possible in six minutes along a flat corridor. Because the patient sets the pace, the 6MWD describes walking capacity under the patient's own effort and not peak exercise capacity. The score is the distance covered, and trials and appraisals report either the 6MWD itself or its change from baseline.

Reliability, the learning effect and methodological sensitivity

The 2014 technical standard concluded that field walking tests are valid, reliable and responsive to change with some interventions, while noting that the results are sensitive to small changes in method. The number of practice tests is the best documented of these influences.

The companion ERS and ATS systematic review by Singh and colleagues pooled 13 studies of people with chronic obstructive pulmonary disease (COPD) and found a mean improvement of 26.3 metres on the second 6MWT compared with the first. The technical standard reports good test-retest reliability for the 6MWD but strong evidence of a learning effect, and recommends that two tests be performed when the 6MWT is used to measure change over time.

For a trial-based economic evaluation the practical point is that an unrecognised learning effect inflates within-arm change. A randomised comparison removes most of that bias from the between-arm difference, but uncontrolled before-and-after studies, registry data and single-arm extensions, which often supply inputs for long-term models, do not.

The 6MWT as a trial endpoint in pulmonary arterial hypertension

Pulmonary arterial hypertension (PAH) trials have used the 6MWD as a primary endpoint, and the test asks for little more than a corridor and a stopwatch. A recent example shows how the endpoint reaches an HTA committee.

The 2026 United States prescribing information for sotatercept (Winrevair) gives the change from baseline in 6MWD at week 24 as the primary endpoint of the STELLAR trial. The committee papers for the NICE appraisal of sotatercept (TA1161) summarise the STELLAR results as a Hodges-Lehmann location shift of 40.4 metres (95% confidence interval 27.28 to 53.53) in favour of sotatercept, with 54% of the sotatercept arm and 22% of the placebo arm improving their 6MWD by at least 30 metres.

Both ways of reporting matter for a model. The Hodges-Lehmann estimate is the median of all paired differences between the sotatercept and placebo arms, so it describes a shift between arms and not the gain of a typical patient, while the proportion who gain at least 30 metres is a responder analysis that can feed directly into a model whose health states are defined by response or by risk status.

Minimal important differences from the primary studies

A minimal clinically important difference translates a change in metres into a judgement about benefit. For the 6MWT the published values come from both anchor-based and distribution-based methods, and they differ by disease.

Mathai and colleagues estimated the minimal important difference in PAH from 405 participants in the 16-week randomised trial of tadalafil, whose patients were either treatment naive or taking an endothelin receptor antagonist as background therapy. The physical component score of the SF-36 served as the anchor because its change correlated most strongly with change in 6MWD, and combining distributional and anchor-based methods gave a consensus estimate of about 33 metres.

In chronic respiratory disease the ERS and ATS systematic review by Singh and colleagues placed the minimal important difference for adults between 25 and 33 metres, and the technical standard reported that the available evidence suggests a value of 30 metres.

These figures are thresholds for change within a patient or a group, estimated in particular populations. Applying a PAH value in heart failure or neuromuscular disease, or treating a mean between-arm difference as though it were the share of patients who improved, are common misreadings.

How regulators and NICE have weighed walk distance

Regulators and HTA bodies have accepted the 6MWD as evidence while questioning how much a given difference means. The US Food and Drug Administration (FDA) advisory committee meeting of 13 September 2023 on patisiran in transthyretin amyloid cardiomyopathy considered the APOLLO-B trial, whose primary endpoint was the change in 6MWT distance at 12 months. The FDA clinical reviewer's slides reported a median difference of 14.7 metres (95% confidence interval 0.7 to 28.7) against a median baseline of 364 metres and described the effect as statistically significant but small, and the committee was asked whether patisiran has clinically meaningful benefits.

In NICE appraisals for Pompe disease the test has appeared as a secondary outcome. In the appraisal of avalglucosidase alfa for Pompe disease (TA821) the primary outcome of the COMET trial was percentage change in FVC, a spirometry measure of lung volume, and the 6MWT was among the main secondary outcomes alongside safety and health-related quality of life. The committee papers record that COMET data informed the magnitude of treatment effect in the company's model, with the 6MWT as a key secondary outcome of that trial. In the later appraisal of cipaglucosidase alfa with miglustat (TA912), the committee discussion names the 6MWT first among the tests used to judge muscle and respiratory response, which include, but are not limited to, FVC per cent predicted, time taken to stand from a chair and daily life activities.

Using walk distance in a cost-utility analysis

A cost-utility analysis needs utilities and survival, and the 6MWD supplies neither directly. Three routes are common. The first defines model health states by a responder threshold or by a risk category that includes walk distance, attaching a utility to each state from EQ-5D data collected in the trial. The second estimates a mapping function that predicts utility using 6MWD and other clinical variables. The third treats walk distance as a surrogate endpoint for hospitalisation and death, which requires evidence that a change in distance predicts a change in those outcomes.

Each route rests on assumptions that move the result. A responder threshold discards the size of gains above and below the cut-off, a mapping function may predict poorly in a population different from the one used to estimate it, and a surrogate relationship estimated across patients does not guarantee that a treatment-induced gain carries the same prognosis. The analysis should show how far the incremental cost per QALY depends on the chosen link.

Illustrative worked example of a responder-based model

All numbers in this example are illustrative and do not describe any real product. A pulmonary rehabilitation programme adds GBP 2,000 per patient to NHS costs over one year. In a trial, 45% of patients in the programme arm and 20% in the usual care arm improve their 6MWD by at least 30 metres. EQ-5D data collected in the trial show that responders gain 0.08 QALYs over the year relative to non-responders, and the model assumes no survival difference.

1. Additional responders. The difference in response is 0.45 − 0.20 = 0.25, so one extra patient responds for every four treated.

$$\text{NNT} = \frac{1}{p_T-p_C} = \frac{1}{0.45-0.20} = 4$$

where NNT is the number needed to treat for one additional responder, $p_T$ is the proportion responding with the programme and $p_C$ is the proportion responding with usual care.

2. Cost per additional responder. Treating four patients costs 4 × 2,000 = 8,000, so each additional responder costs GBP 8,000.

3. Cost per QALY. The expected QALY gain per patient treated is 0.25 × 0.08 = 0.02.

$$\text{ICER} = \frac{\Delta C}{\Delta E} = \frac{2000}{0.02} = 100000$$

where $\Delta C$ is the incremental cost per patient in GBP and $\Delta E$ is the incremental QALYs per patient.

A cost of GBP 8,000 per additional responder (at least 30 metres) sounds modest, yet the incremental cost per QALY is GBP 100,000 because the utility attached to response is small and lasts one year in this model. The result is driven almost entirely by the utility gap between responders and non-responders and by how long it is assumed to last. A report of cost per responder alone would hide that dependence, and NICE appraisals compare the cost per QALY, not the cost per responder, with a threshold.

Boundaries with other walking and function measures

The 6MWT is a clinical outcome assessment of the performance type, in which the patient carries out a standardised task. It differs from a patient-reported outcome such as the KCCQ, which records how patients say they feel and function, and from a clinician-reported outcome such as the NYHA functional classification, in which a clinician grades limitation.

Among performance measures, the Timed Up and Go Test times a short sequence of rising, walking three metres, turning and sitting, and is used mainly in older people for mobility and falls, while the Short Physical Performance Battery combines several short tasks of lower-limb function. The 2014 technical standard also covers the incremental shuttle walk test, a different field walking test whose distances are not interchangeable with the 6MWD even though both are reported in metres.

Limitations and common misreadings

The 6MWD is easy to collect and easy to misread. Several problems recur in trials and economic models.

• Learning and protocol effects. A single baseline test understates true capacity, and the pooled second-test gain of 26.3 metres in COPD is close to the minimal important difference itself.
• Small differences treated as meaningful. In APOLLO-B the FDA clinical reviewer described a median difference of 14.7 metres as statistically significant but small, and the advisory committee was asked whether the benefit was clinically meaningful.
• Thresholds borrowed across diseases. Minimal important differences were estimated in particular populations and do not transfer automatically.
• Distance treated as utility. Walk distance carries no preference weight, so any QALY figure built on it depends on a mapping, a responder utility or a surrogate relationship that needs its own evidence.

Sources

• ATS Committee on Proficiency Standards for Clinical Pulmonary Function Laboratories. ATS statement: guidelines for the six-minute walk test. American Journal of Respiratory and Critical Care Medicine. 2002;166(1):111-117. doi:10.1164/ajrccm.166.1.at1102
• Holland AE, Spruit MA, Troosters T, et al. An official European Respiratory Society/American Thoracic Society technical standard: field walking tests in chronic respiratory disease. European Respiratory Journal. 2014;44(6):1428-1446. https://erj.ersjournals.com/content/44/6/1428
• Mathai SC, Puhan MA, Lam D, Wise RA. The minimal important difference in the 6-minute walk test for patients with pulmonary arterial hypertension. American Journal of Respiratory and Critical Care Medicine. 2012;186(5):428-433. doi:10.1164/rccm.201203-0480OC
• National Institute for Health and Care Excellence. Avalglucosidase alfa for treating Pompe disease (TA821). Committee papers and final appraisal document. https://www.nice.org.uk/guidance/ta821
• National Institute for Health and Care Excellence. Cipaglucosidase alfa with miglustat for treating late-onset Pompe disease (TA912), section 3 Committee discussion. https://www.nice.org.uk/guidance/ta912/chapter/3-Committee-discussion
• National Institute for Health and Care Excellence. Sotatercept for treating pulmonary arterial hypertension (TA1161) [ID6163], committee papers. https://www.nice.org.uk/guidance/ta1161/documents/1-2
• Singh SJ, Puhan MA, Andrianopoulos V, et al. An official systematic review of the European Respiratory Society/American Thoracic Society: measurement properties of field walking tests in chronic respiratory disease. European Respiratory Journal. 2014;44(6):1447-1478. https://erj.ersjournals.com/content/44/6/1447
• US Food and Drug Administration. FDA clinical reviewer presentation, sNDA 210922/015, patisiran. Cardiovascular and Renal Drugs Advisory Committee, 13 September 2023. https://www.fda.gov/media/172042/download
• US Food and Drug Administration. Winrevair (sotatercept-csrk) prescribing information, 2026. https://www.accessdata.fda.gov/drugsatfda_docs/label/2026/761363s012s016s017lbl.pdf