Ten adults rode the same two-mile route three times, once with no motor assistance, once with moderate assistance and once with the motor turned all the way up, while a portable metabolic analyzer measured every breath. On maximum assist they used 37.4 kilocalories above resting to cover the distance. With no assistance they used 70.7, and with moderate assistance 68.6. That is about 47 percent less activity energy on full assist, and no meaningful difference between moderate assist and pedaling unaided.

The study, published in the Journal of Cycling and Micromobility Research (volume 8, article 100118, online April 29, 2026 and in the June 2026 issue), was a randomized crossover trial by researchers at HAN University of Applied Sciences in Nijmegen, the Netherlands, with a coauthor at Maastricht University Medical Center. Five of the ten participants were women. Each had their resting metabolic rate measured first under a ventilated hood, then rode 3.2 kilometers (2 miles) in each of the three conditions in randomized order, with intensity expressed in metabolic equivalents (METs) and volume in MET-minutes.

Why Full Assist Cuts the Dose

Two things changed at maximum assist. The ride got shorter, about 8 minutes against about 10 minutes in the other two conditions, and it got easier, 5.1 METs against about 7 METs. Fewer minutes at a lower intensity multiplied out to 40.6 MET-minutes on full assist compared with 70.5 on moderate assist and 73.3 with no assistance. The authors note that 5.1 METs still sits inside the range that counts as moderate-intensity activity, which starts at 3 METs, so a full-assist ride is not nothing. It is simply a smaller dose of exercise for the same trip.

Moderate assist told a different story. Riders finished in about the same time and at about the same intensity as with no motor at all, and none of the outcomes differed significantly between those two conditions. The authors' conclusion is that replacing conventional cycling with e-biking is likely to shrink the health benefit of habitual riding unless the rider compensates with longer trips.

The Limits of a Ten-Person Trial

This is a small study over a short, fixed distance, and the abstract does not report where the participants would have set the assist level if left to choose. The Crossref record for the paper lists no funder. The full text could not be opened for this article, so the limitations the authors name in their discussion are not reported here; the design itself, ten healthy adults on one bike model over two miles, is the obvious one.

Where It Sits Against the Wider Evidence

A systematic review in the November 2026 issue of the Journal of Physical Activity and Health (volume 23, issue 11, published online July 21, 2026) pulled together 31 studies with 11,890 participants on e-bikes and physical activity. Because the designs, populations and outcomes varied so much, the reviewers summarized the evidence narratively rather than pooling numbers. Their reading matches the Dutch trial on one point: compared with a conventional bicycle, an e-bike generally produces a lower physiological demand, while usually still reaching at least moderate intensity. On a second point the review widens the frame. Compared with not exercising, e-bike use was generally associated with more total daily physical activity and less sitting time, and several studies suggested e-bikes lower the barrier to riding for people who are less active or physically limited. The reviewers called for longer and more rigorous studies before the net effect on health can be stated with confidence. These are associations in observational data, not proof of cause.

The two papers together describe a trade. Against a conventional bike, an e-bike on high assist delivers less exercise per mile. Against the car it replaces, it delivers more. Which comparison applies depends on what the rider would otherwise have done.

What the Guidelines Ask For

The Physical Activity Guidelines for Americans recommend that adults accumulate 150 to 300 minutes of moderate-intensity aerobic activity each week, plus muscle-strengthening activity on at least two days. A full-assist e-bike ride counts toward the aerobic minutes at the intensity the Dutch trial measured, but as the trial shows, it covers a given commute in fewer of them. A rider who wants the ride to carry more of the weekly total has two levers from the data: turn the assist down to a moderate setting, which in this trial cost nothing in exercise compared with no motor, or ride farther.

Sources