Spotting and avoiding PEM with wearable data: heart rate, HRV, and the anaerobic threshold

In PEM, “stay below your exertion limit” is central advice – but where exactly that limit lies is hard to feel in the moment. Because the worsening often sets in only 12 to 72 hours later, the immediate feedback that warns healthy people about overload is missing.

Objective measurements can partly close that gap. Heart rate and HRV respond to exertion before a crash shows up as a symptom – making visible a limit that subjectively only registers too late.

With heart-rate pacing, you set an individual heart-rate ceiling and try to stay below it in daily life. The idea: above a certain threshold the body shifts more toward anaerobic metabolism, which in ME/CFS can disproportionately contribute to PEM.

A common starting point in the ME/CFS community is a rough estimate of about 60 percent of maximum heart rate (roughly (220 − age) × 0.6). That is only a starting point – the actual threshold is individual and is ideally refined with a therapist or through an exercise test.

The anaerobic or ventilatory threshold (VT1) marks the point where energy metabolism becomes increasingly anaerobic. Work by the Workwell Foundation using an exercise test on two consecutive days (2-day CPET) shows that this threshold drops on the second day in ME/CFS – an objective sign of impaired recovery after exertion.

In practice this means the heart rate at VT1 can serve as a personal ceiling. Staying below it engages the anaerobic range less often – a key building block for triggering crashes less frequently.

Heart-rate variability (HRV) describes the variation in the intervals between heartbeats and is seen as a clue to the state of the autonomic nervous system. A low morning HRV relative to your own baseline, or an elevated resting heart rate, can mean the body has not yet fully recovered.

In ME/CFS and long COVID, early studies suggest such autonomic markers relate to exertion intolerance and symptom severity. As a single value HRV is noisy – it becomes useful only as a trend over several days and when compared with the symptom course.

The real value emerges when heart rate, HRV, resting heart rate, sleep, and activity are viewed together over time. From that pattern, early hints of an approaching worsening can be derived before the crash is felt – this is exactly the approach behind Elara’s PEM early-warning.

The honest framing matters: this is about probabilities and early warnings, not a guarantee. No algorithm can reliably predict a crash. The data prompt you to plan more defensively – the decision still rests with the person, ideally together with the symptom diary.

Consumer wearables are not validated medical devices, and their accuracy varies by device and situation. Numbers should relieve pressure, not add it – constantly chasing “better” values can itself become a burden.

For severely affected people, even wearing or reviewing a device can be too much. Wearable data are a pacing aid, not a replacement: most valuable as an additional, individually calibrated clue alongside your own perception and medical guidance.