HRV and PEM: What Your Numbers Do Before, During and After a Crash

Post-exertional malaise is cruel partly because of its timing: you feel fine on the day you overdo it, and the bill arrives a day or two later. Here is what heart rate variability can and cannot tell you about a crash, what the research actually shows, and how to use your morning reading without turning it into a source of worry.

What post-exertional malaise actually is

Post-exertional malaise (PEM) is a worsening of symptoms after activity that would not trouble a healthy person. The UK’s NICE guideline describes it as a worsening that can follow minimal cognitive, physical, emotional or social activity, with symptoms that typically worsen 12 to 48 hours after the activity and can last days or even weeks.

Three features make PEM different from ordinary tiredness:

• It is delayed. The trigger and the crash are often separated by a night or two, which makes the cause easy to miss.
• It is out of proportion. A shower, a phone call or a short walk can be enough on a bad week.
• It is not only physical. Concentration, emotional strain, noise and social effort all count against the same budget.

It also helps to separate PEM from orthostatic symptoms. Feeling lightheaded, shaky or racing within minutes of standing is usually orthostatic intolerance, which many people with long COVID and ME/CFS also have. PEM is the slower wave that follows effort. Both can show up in your data, and they call for different responses.

Why HRV is worth watching

Heart rate variability is the small, constant variation in time between heartbeats. Much of the fast, beat-to-beat variation comes from the vagus nerve, the main brake of the autonomic nervous system, so a morning HRV reading gives a rough window into how much recovery capacity your nervous system has that day. If you want the full background, start with the complete guide to HRV.

Two cautions belong right at the start. First, group averages are less useful than you might expect. A 2026 meta-analysis of 11 studies found only a trend toward lower HRV in post-COVID condition, with wide differences between studies. Second, HRV varies enormously from person to person. What matters is not whether your number is low compared with a chart, but whether it is low compared with you.

Before a crash: what the research shows

The largest dataset so far

The biggest study on this question looked at 4,244 people with long COVID, ME/CFS and other energy-limiting conditions who took a 60-second morning reading with a phone camera or an armband (Aitken et al., npj Digital Medicine, 2026). The data came from the Visible app, and several authors work or have worked for the company, which is worth knowing when you read it.

The findings that matter most for you:

• Within-person changes mattered. On mornings when someone’s HRV was lower than their own usual and their heart rate was higher than their own usual, evening reports of crash, fatigue and brain fog tended to be worse.
• Instability mattered too. Bigger swings in HRV and heart rate across the previous seven days were linked to more symptoms.
• The gain was real but modest. Yesterday’s symptoms already predicted today’s crash fairly well. Adding morning biometrics improved that prediction only a little, for crash from an accuracy score (AUC) of 0.78 to 0.81.

What that means for you

A signal, not a forecast. A low morning reading raises the odds of a harder day. It does not guarantee one, and a normal reading does not guarantee a good one. The study also used uncontrolled, real-world readings, so some of the noise is measurement noise that a cleaner routine can reduce.

After effort: the slow return to baseline

A separate study followed 127 people with long COVID and 21 healthy controls who wore a chest-worn HRV monitor for several days after a supervised, submaximal exercise test (Ruijgt et al., 2025 preprint). In the patients, HRV stayed lower for 24 hours after exercise, whether that exercise was below, at or above their first ventilatory threshold. The controls showed no such dip. Nighttime HRV in patients also fell further after more intense or longer sessions.

This fits what many people notice: the hit to HRV from a big day often shows up the next morning, before the full crash arrives. It is still a preprint, though, and it did not record PEM symptoms directly, so it shows a slower autonomic recovery rather than a proven crash marker.

During a crash: why HRV can rise while you feel worse

This is the pattern that confuses people most. You are clearly crashing, yet a day or two in, your resting heart rate drifts down and your HRV climbs back up, sometimes above your usual.

One wearable company’s analysis of its own users’ self-tagged crash days found exactly this: an early phase where heart rate rises and HRV falls, then a later phase where the numbers look calmer. The data are self-reported and early, and no validated explanation exists yet. The interpretations on offer all point the same way, though: the calmer numbers may reflect a nervous system with less capacity to react, not one that has recovered.

This is not a green light. If you feel worse and your HRV looks better, believe how you feel. Treat a sudden high reading during a crash as information about the crash, not permission to catch up on everything you missed.

Patterns worth recognizing

For more on reading the two numbers together, see the piece on HRV versus resting heart rate.

How to use HRV to protect against crashes

Take a clean morning reading

Measure at the same time each day, ideally right after waking, in the same position, before coffee or your phone. Consistency matters more than the device. The article on the best time of day to measure HRV covers the routine in detail.

Compare against your own baseline

A rolling seven-day average is a practical baseline. A single reading 10 percent below it means little on its own; a reading well below it that repeats, especially with a higher resting heart rate, deserves attention.

Log the days before, not just the day of

Because PEM is delayed, the useful question is rarely “what did I do today?” It is “what did I do one, two and three days ago?” Log the obvious exertion and the easy-to-forget kind: a stressful appointment, a long conversation, a hot shower, a car journey. Over a few weeks, your own pattern starts to show.

Decide in advance what a warning sign changes

The hardest moment to make a good decision is a morning when you feel fine and the number says otherwise. Agree with yourself ahead of time what a warning morning means: perhaps shorter tasks, more lying down, postponing one thing that can wait. The pacing budget is built around exactly this kind of pre-commitment.

What HRV cannot tell you

HRV is useful, and it is also limited. Keep these in mind:

• Not every crash shows up. Cognitive and emotional exertion may not move your morning reading much.
• Other things move HRV. Poor sleep, alcohol, a late meal, infection, the menstrual cycle and some medications, including beta blockers, can all shift it. For more on overnight effects, see the piece on sleep stages and HRV.
• Measurement noise is real. Movement, a loose strap or a short reading can produce a misleading number.
• Evidence for HRV-guided pacing is still early. A 2025 feasibility trial found heart rate monitoring acceptable to people with ME/CFS and long COVID, but it was too small and too short to show clear benefit.
• Numbers can add stress. If checking HRV makes you anxious, check less often, or look only at the weekly trend.

When to involve a clinician

HRV tracking is a personal tool, not a diagnosis. Talk to a clinician, ideally one experienced with ME/CFS, long COVID or dysautonomia, if your crashes are becoming longer or more frequent, if you have new symptoms, or if you are unsure whether what you are experiencing is PEM at all. Seek urgent care for chest pain, fainting, severe shortness of breath or signs of a serious infection.

It is also worth bringing your data. A few weeks of morning readings alongside an activity and symptom log can make an appointment far more productive. Note that the NICE guideline advises against graded exercise therapy for ME/CFS, so be cautious with any plan that asks you to push through PEM.

The bottom line

HRV cannot tell you exactly when a crash is coming. It can tell you, most mornings, whether your nervous system looks more strained than usual, and that is often enough to make a better decision about the day ahead. Read it against your own baseline, look at a few mornings rather than one, connect it to what you did in the days before, and when the number and your body disagree, listen to your body.

Frequently asked questions

Can HRV predict a PEM crash? Not reliably on its own. In the largest study, a lower than usual morning HRV with a higher than usual heart rate was linked to worse symptoms that day, but it only modestly improved on simply knowing how you felt the day before.

Why does my HRV go up during a crash? Some people see HRV fall early in a crash and then rise again while they still feel unwell. The reason is not established, and it should not be read as recovery. Rest according to your symptoms.

How long after exertion does HRV drop? In one long COVID study, HRV stayed lower for at least 24 hours after exercise, and nighttime HRV fell further after harder sessions. Many people see the dip on the next morning’s reading.

How many low mornings should worry me? A single low morning is often noise. Two or more mornings below your baseline, especially with a higher resting heart rate, are worth acting on by pacing more conservatively.

Is a chest strap better than a watch or phone for this? A chest strap gives the cleanest beat-to-beat data, but consistency matters more than the device. See the comparison of HRV chest straps if you are choosing one.