REF-1435 · Bosch et al., 2021 · International Journal of Sport Nutrition and Exercise Metabolism · Paper · PubMed

Is an elliptical or stepper as good a workout as running?

In 18 trained runners VO2max was similar on all three, and at the same share of each device's peak, heart rate and oxygen uptake did not differ. Acute; healthy. Abstract.

Main caveat: Abstract only; stage values, absolute workloads and confidence intervals not given.
limited evidenceMeasured on trained athletes

Bosch and colleagues (University of Cape Town, Stellenbosch University) noted that elliptical trainers and steppers are proposed for rehabilitating injured runners because they stress muscles and joints less than running, and compared the physiological responses to submaximal treadmill running with exercise on the two devices at the same relative workloads.

Eighteen trained runners (9 men, 9 women; 23 ± 3 years) completed randomised maximal oxygen uptake tests on all three devices, then 3-min submaximal stages at 60%, 70% and 80% of the peak workload established for each device, with breath-by-breath oxygen uptake, heart rate, fuel use and energy expenditure.

VO2max did not differ between treadmill, elliptical and stepper (49.3 ± 5.3, 48.0 ± 6.6 and 46.7 ± 6.2 mL/kg/min). Oxygen uptake, heart rate and carbohydrate and fat oxidation rose with intensity but did not differ between devices.

The authors conclude that the elliptical trainer and stepper are suitable substitutes for running when running load must be reduced, such as in rehabilitation from running injury.

Our reading: the base's only elliptical entry so far was recovery in the week after a marathon (REF-761); this one shows that, set by their own peak, both devices give a running-like cardiorespiratory stimulus with running-like heart rates. It is acute physiology in healthy runners — no training, no injured runners, no measure of joint or tendon load.

An earlier study outside the base (Turner 2010) found a lower VO2max and HRmax on an elliptical than on a treadmill, so the device and the user's skill on it may matter.

Protocol

Participants: n = 18 trained runners (9 men, 9 women); 23 ± 3 years.
Maximal tests: randomised, one on each of treadmill, elliptical trainer and stepper.
Submaximal stages: 3 min each at 60%, 70% and 80% of the peak workload individually established for each device.
Measures: breath-by-breath oxygen uptake, heart rate, carbohydrate and fat oxidation, energy expenditure.
Funding and interests: none listed in the PubMed or Crossref record; no conflict statement in the abstract record.

Key Results

MeasureTreadmillEllipticalStepper
VO2max (mL/kg/min)49.3 ± 5.348.0 ± 6.646.7 ± 6.2
VO2, heart rate, carbohydrate and fat oxidation at 60, 70 and 80% of device peakdiffered between intensitiesno difference between devicesno difference between devices
Not in the abstract: the heart-rate and VO2 values at each stage, the absolute workloads, confidence intervals for the between-device differences, perceived exertion.

Mechanisms

Measured: similar maximal oxygen uptake on all three devices and similar submaximal VO2, heart rate and fuel use at matched relative workloads.
Authors' interpretation: both devices are suitable running substitutes when running load must be reduced, because they stress muscles and joints less than running (the reduced stress is stated as background, not measured here).
Inference: both devices work large leg muscles in an upright posture, so central demand can match running once workload is scaled to the device's own peak; what they do not reproduce is impact, which is why they are chosen and why they will not prepare bone and tendon for running (PHYS-37, PHYS-38).

Practical Application

Is an elliptical or stepper as good a workout as running? In 18 trained runners VO2max was similar on all three, and at the same share of each device's peak, heart rate and oxygen uptake did not differ. Acute; healthy. Abstract.
Heart rate may carry over on group means: at the same share of each device's own peak workload, heart rate did not differ from the treadmill (values and individual agreement not reported), so running zones are a starting point to check on an elliptical or stepper, not a transfer (our inference; equal group means hid a 12-bpm individual error between bike and run in REF-1431). Intensity here was set from a maximal test on each device, not by effort.
Pick it only if it does not hurt: for Achilles pain the guideline lists a cross-trainer for pain above the heel but not at the heel bone (REF-1380, expert opinion; SYNTH-94).
What it does not show: that training on these devices keeps running fitness over weeks, any data in injured runners — the gap line in SYNTH-73 narrows to that — or the joint and tendon loads the authors assume are lower.

Limitations

Abstract only; stage values, absolute workloads and confidence intervals not given.
A null between devices in 18 people is not proof of equivalence.
Short stages (3 min); oxygen uptake may not reach a steady state at 80%.
Healthy trained runners; no injured runners, no training period.
Joint and tendon load not measured; the rehabilitation claim rests on the background assumption.
Funding and interests: not stated in the record.
How this entry was checked and changed
Why this level

(from the abstract) — a within-subject comparison of 18 (the rubric's B+ band for crossovers of 15 or more) held at B, as REF-1430 was: PubMed indexes it as an observational cross-sectional study, the between-device result the base uses (heart rate) is a null reported without values or confidence intervals, the stages lasted 3 min, and there is no training outcome