Graded exercise therapy (GET)

Start here: the honest default and the warning

No exercise program is approved as a treatment for long COVID. More than that, the specific approach of steadily increasing exertion regardless of how you feel can make people with post-exertional malaise worse, sometimes lastingly. That is why this page leads with a caution rather than a protocol.

If you crash a day or two after activity, your body is giving the clearest possible signal that the push-through model does not fit your physiology. The goal of this page is to explain why, so the warning is something you understand rather than just obey.

harm well documented in PEMnot advised when PEM is present

What graded exercise therapy is

Graded exercise therapy, or GET, is a structured program that increases physical activity on a fixed schedule, on the premise that deconditioning maintains symptoms and that reconditioning will reverse them. The defining feature is that progression is driven by the timetable, not by your symptoms.

For ordinary deconditioning, that logic is sound. The problem is that long COVID with PEM is not simple deconditioning, and a schedule that ignores symptoms collides directly with a condition defined by symptom flares after exertion.

Why it backfires in post-exertional malaise

PEM is a delayed, disproportionate worsening after activity that was once tolerable. Objective testing supports that it is physiological, not fear of activity. On a two-day repeat exercise test, people with the ME/CFS pattern fail to reproduce their first-day output on the second day, a drop healthy sedentary people do not show.

That means the system is not simply unfit; it cannot recover normally between efforts. Pushing more exertion into a body that cannot recover from the last bout is how a manageable baseline becomes a worse one. The harm is built into the mismatch.

objective: failed day-2 reproductionwell-founded harm mechanism

What the evidence and the disagreement show

Patient surveys have repeatedly found that a large share of people with PEM report deterioration after graded exercise, and on that basis the World Health Organization and long COVID rehabilitation consensus groups caution against GET when PEM is present.

In fairness, some trial meta-analyses, including work by the original GET researchers, report no excess self-rated harm in their trials. Those trials largely predate careful PEM screening and rely on different outcomes, which is why survey reports, physiological data, and guideline caution carry the day for a symptom-flaring population.

trial meta-analyses dispute excess harmguidelines caution against GET in PEM

When activity still has a place

This is not a claim that movement is forbidden. People without PEM, or whose PEM has settled, can often rebuild carefully. The distinction is symptom-guided, not schedule-driven: activity stays inside the energy envelope and never forces progression on a day the body is flaring.

Recumbent and short-interval approaches, advanced only when they do not trigger a crash, are the safer frame. The test is always the next two days. If activity reliably produces a delayed crash, the plan is wrong for now, no matter how gentle it looks on paper.

What to do instead

The evidence-based alternative for PEM is pacing, which manages activity to stay under the crash threshold rather than push past it. It is not a cure, and it does not promise reconditioning, but it prevents the boom-and-bust cycle that drives people down.

If a clinician proposes a fixed, escalating exercise schedule and you have PEM, it is reasonable to ask how the plan accounts for post-exertional malaise. A program that cannot answer that is the push-through model in new clothing.

What this means for you

If you get a delayed crash after activity, treat any push-through exercise plan with great caution and do not let a fixed schedule override your symptoms. The clearest, safest guidance the evidence supports is to pace rather than push.

Activity is not the enemy, and gentle, symptom-guided movement may have a place once PEM settles. But a program that ignores your post-exertional crashes is the wrong tool for this problem, and major guidance now agrees.

References

Each reference links to the source on PubMed, PMC, or the publisher.

1. Keller B et al. Cardiopulmonary and metabolic responses during a 2-day CPET in ME/CFS. J Transl Med 2024.
2. Vernon SD et al. Post-exertional malaise among people with long COVID compared to ME/CFS. Work 2023.
3. White PD, Etherington J. Adverse outcomes in trials of graded exercise therapy for CFS. J Psychosom Res 2021.