Wearables on a GLP-1 During Menopause: What Your Resting Heart Rate and HRV Actually Mean

Because that is what these medications do, and it has been known since long before anyone was wearing a ring to bed.

A meta-analysis of GLP-1 receptor agonist trials by Sun and colleagues found consistent small increases in heart rate across the class, averaging in the region of 2 to 3 beats per minute compared with control (Journal of the American Heart Association, 2015). The effect appears in the large obesity and diabetes trials too, and it is durable rather than transient: it typically appears during dose escalation and persists at maintenance dose.

The mechanism is not fully settled. GLP-1 receptors are present in the sinoatrial node, the heart's own pacemaker, which offers a direct route. There is also evidence of modest sympathetic nervous system activation. Both are probably contributing.

What surprises people is that the same drug lowers blood pressure at the same time. In SURMOUNT-1, systolic blood pressure fell by roughly 7 mmHg on tirzepatide (Jastreboff et al., NEJM 2022), and STEP 1 showed similar reductions with semaglutide (Wilding et al., NEJM 2021). Heart rate up, blood pressure down, in the same person on the same drug. That combination confuses anyone reading a wearable dashboard, because consumer apps tend to code a higher heart rate as unambiguously worse.

The clinical answer to whether the heart rate rise matters came from SELECT, which followed more than 17,000 people with cardiovascular disease and overweight or obesity for over three years. Semaglutide 2.4 mg reduced major adverse cardiovascular events by 20% versus placebo (Lincoff et al., NEJM 2023). The heart rate increase was present in that trial too. Whatever the small rise represents, it did not translate into harm at the outcome level, and the net cardiovascular effect was clearly favourable.

Independently, and in the same direction your wearable will flag as worse.

Heart rate variability, or HRV, measures the beat-to-beat variation in the time between heartbeats. Counterintuitively, more variation is better. High HRV reflects a well-functioning parasympathetic nervous system, the rest-and-digest branch that dominates when you are recovered and calm. Low HRV means the sympathetic branch, the fight-or-flight side, has more relative influence.

Estrogen supports parasympathetic tone through several routes, including effects on the vagus nerve and on nitric oxide-mediated vascular function. As estradiol declines through perimenopause and into postmenopause, that support weakens and autonomic balance shifts sympathetically. Studies of HRV across the menopausal transition have consistently found reduced parasympathetic indices in postmenopausal compared with premenopausal women.

This is the same underlying shift that produces several symptoms women already know well. Heart palpitations, covered in [why your heart races in menopause](/blog/menopause-heart-palpitations-why-your-heart-races-and-what-helps), are one expression. So is the 3am wake-up with a pounding heart, and so is the sense of being wired but exhausted.

There is a second, larger factor that wearables handle badly: sleep. HRV is measured overnight, and night sweats fragment sleep in ways that crush HRV readings. A hot flash produces a genuine sympathetic surge, with a measurable heart rate spike. If you are having four of those a night, your overnight HRV average is being pulled down by the vasomotor symptoms, not only by baseline hormone status. Our guide to [sleep on a GLP-1 during menopause](/blog/sleep-on-glp1-during-menopause-night-sweats-and-nausea) covers that collision in detail.

So the falling HRV number is real. What it usually reflects is disrupted sleep and shifted autonomic balance, both of which are addressable.

Usually not, and it helps to understand what those scores actually are.

A recovery or readiness score is a proprietary composite, typically blending overnight HRV, resting heart rate, respiratory rate, sleep duration and sleep stages into a single number, then comparing it against your own recent baseline. It is not a clinical measurement, it is not validated against any medical outcome, and each company computes it differently.

Here is the problem when you start a GLP-1 in menopause. Two of the score's main inputs move in the unfavourable direction for reasons that are physiologically expected. Your resting heart rate rises by a few beats because of the medication. Your HRV falls because of hormonal change and disrupted sleep. The algorithm has no way to know why, so it reports that you are under-recovered. Some people respond by training less, eating less, and worrying more, at exactly the point when resistance training and adequate protein matter most.

This is the same interpretive trap we describe in [why your VO2 max number misleads on a GLP-1](/blog/vo2-max-on-glp1-during-menopause-why-the-number-misleads), where estimated fitness scores respond to weight change and heart rate in ways that misrepresent what is actually happening to your body.

Three rules make wearables useful rather than corrosive here.

Track trends, not days. A single night means very little. A four-week direction of travel means something.

Re-baseline after a dose change. Give it three to four weeks at the new dose before drawing conclusions. The first two weeks after an escalation are dominated by transient effects.

Never compare with other people. HRV in particular varies enormously between individuals for reasons that have nothing to do with health. A 90th-percentile HRV in one person and a 20th-percentile in another can represent identical fitness. The only meaningful comparison is with your own previous self. Our guide to [tracking progress beyond the scale](/blog/tracking-progress-on-glp1-in-menopause-beyond-the-scale) applies the same logic to weight.

Some wearable data genuinely earns its place. The trick is knowing which metrics respond to things you can change.

Steps and daily movement. The least glamorous number and the most useful. It is directly actionable, well correlated with outcomes, and unaffected by algorithmic interpretation. On low-energy days early in dose escalation, watching step count fall is often the first honest signal that you need to eat more rather than push harder.

Sleep duration and wake events. Not the sleep stages, which consumer devices estimate with limited accuracy against polysomnography, but total time asleep and number of awakenings. Both are reliable enough to be useful, and both respond to interventions like treating night sweats or shifting your last meal earlier.

Resting heart rate trend over months. Over the first few weeks it will rise on a GLP-1. Over six to twelve months, as weight comes down and cardiorespiratory fitness improves, it often falls back below where it started. That longer curve tells you something real.

Blood pressure, measured properly. Not from a wrist device. A validated upper-arm cuff, seated, after five minutes of rest. This is the cardiovascular number that actually matters, it is the one improving most on a GLP-1, and no consumer wearable measures it reliably. Given that blood pressure rises in menopause anyway, as covered in [why blood pressure rises after 45](/blog/menopause-blood-pressure-why-it-rises-and-what-helps), this is worth doing properly.

Strength and function. Grip strength, how many stairs you can climb, whether you can carry the shopping in one trip. These track the lean mass question that matters most during rapid weight loss, and they are covered in [grip strength as the muscle marker that matters](/blog/grip-strength-on-glp1-the-muscle-marker-that-matters).

And one worth deliberately ignoring: calorie burn estimates. These are among the least accurate outputs consumer wearables produce, with substantial error rates in validation studies. Making eating decisions from them on a GLP-1, when under-eating is already the more likely risk, is a bad trade.

There is a real signal buried in all this noise, and it is worth being able to recognise it.

Contact your clinician if any of the following apply:

• •Resting heart rate consistently above 100 bpm, which is tachycardia and warrants assessment regardless of cause
• •A sustained rise of more than 10 bpm over your established baseline lasting more than a week, which is well beyond the expected class effect
• •Palpitations with chest pain, breathlessness at rest, or fainting, which need urgent assessment rather than a routine appointment
• •An irregular rhythm flagged by your device, particularly a possible atrial fibrillation notification, which needs an actual ECG to confirm or exclude
• •A rapid heart rate alongside ongoing vomiting or diarrhoea, which usually means volume depletion and may mean holding your dose

Two alternative explanations are worth ruling out before attributing anything to the GLP-1, because both are common in this exact population and both are treatable. Thyroid dysfunction causes tachycardia and is frequently missed in midlife because its symptoms overlap so heavily with perimenopause, as covered in [thyroid or perimenopause](/blog/thyroid-or-perimenopause-overlapping-symptoms-what-to-test). And iron deficiency anaemia causes a compensatory rise in heart rate; it is common in perimenopause because of heavy bleeding and is made more likely by reduced food intake on a GLP-1, as covered in [iron deficiency on a GLP-1 during perimenopause](/blog/iron-deficiency-on-glp1-during-perimenopause-what-to-know).

Both are settled with a blood test. If your heart rate has risen more than the expected few beats, ask for thyroid function and a full blood count with ferritin before assuming the medication is responsible.

The overall framing is worth holding onto. A device that shows you a slightly higher resting heart rate and a slightly lower recovery score is not showing you harm. It is showing you two known physiological effects with a favourable outcome profile behind them. Use the data to catch the genuine outliers, and do not let a daily score talk you out of the training and protein intake that are doing the real work.