Resting Heart Rate
A slow pulse at rest goes with living longer, and it costs nothing to measure — but the trials that tried to lower it on purpose mostly failed, so it is a number worth watching and a poor thing to chase.
Resting heart rate is the cheapest measurement in this entire site. It needs no lab and no appointment — though a wearable that tracks it against your own baseline over weeks tells you far more than any single reading. Across cohorts totalling well over a million people it predicts who dies, and in the one study that set the two side by side it held its own against blood pressure. It is also the clearest worked example on this site of a marker that is not a lever: three large randomised trials have now lowered it deliberately with a drug that does nothing else, and only one of them improved anything.
What the number predicts — Strong
The association with dying is one of the most consistently replicated in cardiovascular epidemiology, and it is close to linear: across the ordinary range, every step up carries a little more risk.
The largest synthesis pooled 87 prospective studies and found all-cause mortality about 17% higher for each 10 beats per minute of resting heart rate, drawing on 48 cohorts for that outcome alone[1]. That figure should be read as a direction rather than a precise quantity — the studies behind it disagree with each other enormously, which the authors quantify as a heterogeneity statistic of 94%, close to the maximum. An earlier meta-analysis — 46 studies and 1,246,203 participants, searched up to January 2015 — put it lower, at 9% per 10 beats (relative risk 1.09, 1.07–1.12), and detected both substantial heterogeneity and evidence of publication bias, meaning small studies with null results are probably missing from the literature[2]. The later analysis drew on nearly twice as many studies (87 against 46), so the gap between 9% and 17% is better read as a sign that the size of the effect is unsettled than as one figure being more cautious than the other[3].
In the earlier (46-study) analysis the risk is concentrated at the top of the range rather than spread evenly. Above 80 beats per minute, all-cause mortality ran 45% higher than in the lowest category (relative risk 1.45, 95% confidence interval 1.34–1.57). That bracketed range is where the true effect most plausibly lies; because it sits entirely above 1.0, the increase is very unlikely to be chance, whereas a range spanning 1.0 would be compatible with no effect at all. The 60–80 band carried a much smaller excess — 12% higher (relative risk 1.12, 1.07–1.17) — and cardiovascular mortality only became clearly elevated around 90 beats per minute.
Two features of the association are more informative than the headline number.
It predicts deaths that have nothing to do with the heart. In the Melbourne Collaborative Cohort Study — 41,386 adults, 9,846 deaths — each 10 beats per minute carried an 11% higher cardiovascular mortality, a 10% higher cancer mortality, and a 20% higher mortality from all remaining causes, the largest of the three[4]. Breast, colorectal, kidney and lung cancer deaths were each individually elevated. A marker that predicts kidney cancer about as well as it predicts heart disease is not describing a cardiac mechanism. A 2026 study of 615,730 cancer-free Taiwanese adults, with heart rate taken from an electrocardiogram and a median 17 years of follow-up, found almost the same figure — about 10% higher cancer mortality per 10 beats per minute, with the same upward trend across several cancer sites — though it comes from the same Taiwanese screening programme as the blood-pressure comparison below, so the two are not independent[5]. Part of what resting heart rate is picking up is almost certainly illness that has not been diagnosed yet — an inflammatory or malignant process raises resting heart rate long before it declares itself. That reverse arrow is a permanent caveat on every figure in this section, and the Melbourne analysis makes it worse by noting it did not adjust for atrial fibrillation or other irregular-rhythm disorders.
Change matters, not just level. In the same cohort, people whose resting heart rate rose by more than 15 beats per minute over roughly a decade had higher mortality than those whose rate held steady. A single reading is a snapshot; a trend is closer to a signal.
How it compares to blood pressure — Moderate (a single pooled analysis of two observational cohorts)
A 2025 analysis pooled the Taiwan MJ cohort and Norway's HUNT cohort — 692,217 adults followed for about 25 years — and set a raised resting heart rate (80–99 beats per minute) directly against hypertension (≥140/90 mmHg)[6]. Three findings are worth carrying:
- People with normal blood pressure but a high heart rate died at a higher rate than people with hypertension but a normal heart rate.
- Under age 40, hypertension was not a statistically significant predictor of death at all. A raised heart rate was significant at every age.
- The prevalence of a high resting heart rate barely moved across the lifespan — 21% in the twenties, 25% past 70 — while hypertension climbed from 5% to 57%. It is a signal that shows up early, in a group where the standard risk factor has not yet appeared.
The young-adult signal has since been replicated independently. A US analysis of adults aged 20–49, across two national survey cohorts, found that each 10 beats per minute went with about 17–26% higher long-term mortality (hazard ratios 1.17 and 1.26); non-cardiovascular deaths were raised as much as cardiovascular ones, and the link was already visible by the early-to-mid 30s[7].
The Taiwan–Norway authors also estimated life expectancy: about 10.3 years lower (8.1–12.5) for normal blood pressure with a high heart rate, against 5.5 years (3.6–7.6) for hypertension with a normal heart rate. Read that as a statement about how much predictive weight the number carries, not as a bill the reader will be charged. These are observational cohorts; nothing here shows that lowering the number buys the years back — and the next section is about exactly why it probably doesn't.
Is it just a proxy for being unfit? — Moderate
The obvious objection is that fit people have slow hearts, so resting heart rate might be measuring fitness in disguise — and cardiorespiratory fitness is the most replicated mortality predictor in the field.
Several cohorts that measured fitness with an exercise test have checked this, and they agree that resting heart rate is not just fitness in disguise. In the Cooper Clinic cohort in Dallas (50,108 adults, fitness from a maximal treadmill test), a resting rate of 80 or more still carried about 35% higher cancer mortality than one below 60 after adjusting for fitness (hazard ratio 1.35, 1.06–1.71)[8], and unfit people with a low rate died at about the same rate as fit people with a high one[9]. In about a million Swedish men examined at conscription, a high resting rate at 18 (top versus bottom fifth) still predicted slightly higher all-cause death decades later after adjusting for fitness — about 9% higher (hazard ratio 1.09, 1.05–1.12) — though not cardiovascular death[10]. Two smaller studies add useful detail.
The first is the Copenhagen Male Study. It estimated maximal oxygen uptake from a stationary-bike test in 1970–71 and measured resting heart rate by electrocardiogram fifteen years later, then followed 2,798 healthy middle-aged men for 16 years — of whom 1,082, close to four in ten, died. After adjusting for that fitness measure, leisure-time physical activity and conventional cardiovascular risk factors, mortality still rose 16% per 10 beats per minute (10–22%), and men above 90 beats per minute had roughly triple the mortality of men at or below 50[11]. That threefold figure compares two small groups at opposite extremes of the distribution, so the per-10-beat number is the more useful one. The point estimate was higher in smokers (20% per 10 beats) than non-smokers (14%), but the two overlap heavily and the difference was not statistically clear (p=0.07).
The second study is smaller still but asks a sharper question. The Belgian Physical Fitness Study put a 24-hour ambulatory heart rate and a seated clinic reading into the same model, adjusted for fitness, occupational and leisure activity, and the usual risk factors, and followed 439 men for a mean of 16.5 years[12]. Split into thirds, the 24-hour average looked stronger — about three times the risk in the highest third versus the lowest (hazard ratio 3.21, 1.22–8.44) — while the clinic reading's top third showed no clear excess once the 24-hour average was in the model (hazard ratio 1.26, 0.48–3.31). But the study recorded only 48 deaths, and when heart rate was treated as a continuous number the two readings performed almost the same: about 27% versus 22% higher risk per 10 beats, with both ranges including no effect[13]. This is one small, suggestive study. It hints that the rate your heart runs at through the day matters more than one clinic reading, but it does not show the clinic reading is useless.
Fitness and resting heart rate are related, of course. In Norway's HUNT cohort, a baseline resting heart rate above 80 beats per minute predicted a maximal oxygen uptake 4.6 mL per kilogram per minute lower in men 23 years later, compared with a baseline below 60 — though the same gap in women (1.4) was indistinguishable from no difference. Notably, being physically active largely cancelled the disadvantage of starting with a fast heart[14].
Does lowering it help? — Strong, and the answer is mostly no
This is the question that decides whether resting heart rate is worth acting on directly, and it has a real answer, because a drug exists that lowers heart rate and does almost nothing else. Ivabradine slows the sinus node — the heart's own pacemaker — without affecting blood pressure or the force of contraction. That makes it close to a clean experiment on the number itself. It has been run three times, in three populations.
In heart failure, lowering it works. SHIFT randomised 6,558 people with symptomatic heart failure, an ejection fraction — the share of blood the main pumping chamber pushes out with each beat, healthy being roughly 55–70% — of 35% or less, and a heart rate of at least 70 beats per minute. Cardiovascular death or hospital admission for worsening heart failure fell from 29% to 24% — about an 18% relative reduction, or roughly 5 fewer people affected per 100 over a median of 23 months (hazard ratio 0.82, 0.75–0.90)[15]. The benefit came almost entirely from fewer heart-failure hospitalisations and fewer heart-failure deaths.
In coronary disease without heart failure, it does nothing. SIGNIFY was much larger — 19,102 people with stable coronary artery disease and no clinical heart failure — and achieved a bigger reduction: at three months the average heart rate was 60.7 beats per minute on the drug against 70.6 on placebo[16]. Cardiovascular death or non-fatal heart attack occurred in 6.8% on the drug and 6.4% on placebo: no benefit, and if anything a nudge the wrong way (hazard ratio 1.08, 0.96–1.20). This is not a small trial that missed a real effect. Two further details matter. Among the subgroup with angina that limited their activity, the drug was associated with more events, while among those without such angina it was not (interaction p=0.02). A subgroup interaction inside a null trial is a reason to look again, not a finding in its own right — this trial did not show that lowering heart rate is harmful, it showed that lowering it did not help. And 18.0% developed a heart rate slow enough to be flagged as bradycardia, against 2.3% on placebo. The trial was funded by the drug's manufacturer.
In coronary disease with impaired pumping, it also does nothing. BEAUTIFUL lowered heart rate by 6 beats per minute in 10,917 people with coronary disease and an ejection fraction below 40%, most of them already on beta-blockers, and moved the primary endpoint not at all — a hazard ratio of exactly 1.00 (0.91–1.10)[17]. Two secondary endpoints did fall in a prespecified faster-heart-rate subgroup, which is where the SIGNIFY hypothesis came from — and SIGNIFY then tested it properly and found nothing.
In hypertension, the correlation runs backwards. Across nine randomised trials of beta-blockers as first-line treatment for high blood pressure — 34,096 people on beta-blockers — the trials that achieved the lowest heart rates were the trials with the worst outcomes: all-cause mortality, cardiovascular mortality, heart attack and heart failure all tracked in the wrong direction[18]. This is a comparison across trials rather than a randomised test of heart-rate targets, so it cannot show that the slowing caused the harm. But it is the opposite of what a simple lever model predicts.
After a heart attack with normal pumping, slowing the heart does not help either. Several large randomised trials, most reported between 2024 and 2026, asked whether heart-attack survivors whose pumping was normal still benefit from a beta-blocker, a drug that slows the heart. Pooling the individual patients from five of them — 17,801 people with an ejection fraction of 50% or more — death, a new heart attack or heart failure occurred in 8.1% with a beta-blocker and 8.3% without one, no meaningful difference (hazard ratio 0.97, 0.87–1.07)[19]. The benefit reappears where the pump is weaker: with a mildly reduced ejection fraction (40–49%), beta-blockers cut that same combined outcome by about a quarter (hazard ratio 0.75, 0.58–0.97)[20]. BETAMI-DANBLOCK, a Scandinavian trial that enrolled both groups and also contributed patients to both pooled analyses, found a modest benefit — about 15% fewer deaths or major cardiovascular events (hazard ratio 0.85, 0.75–0.98), mostly from fewer repeat heart attacks[21]. Beta-blockers do more than slow the heart and these trials were not blinded, so they are not a clean test of the number itself, but where the pump is normal they agree with the ivabradine trials: slowing the heart did not, on its own, improve outcomes.
In healthy adults, it has never been tested in a trial. No randomised trial has lowered resting heart rate in people without cardiovascular disease to see what happens. The nearest substitute is genetics: some people inherit gene variants that set their heart slightly faster for life, which works like a natural experiment. A study that identified these variants in up to 835,465 people found no sign that they lead to dying sooner, though they were linked to a higher risk of one form of heart-muscle disease (dilated cardiomyopathy) and, unexpectedly, a lower risk of atrial fibrillation and ischaemic stroke[22]. An earlier genetic study that did find a link with mortality was most likely a false positive caused by weak genetic signals. A null genetic result is not proof of no effect, but a fast heart you were simply born with does not appear to shorten life — which is what you would expect if the rate is mostly a readout.
The pattern is coherent. In heart failure, a fast heart is part of the disease — the failing ventricle compensates by beating more often, and slowing it gives it time to fill. Everywhere a pure rate-lowering drug has been tested outside heart failure, a fast heart appears to be a readout of something else: deconditioning, an over-active stress-response system, inflammation, illness. Lower the readout without changing what produced it and nothing improves. That is what a marker looks like when you push on it.
What's normal — and why the textbook range is nearly useless — Moderate
The familiar 60–100 beats per minute band describes a population, and applying it to one person throws away almost everything informative.
The best evidence on this comes from continuous wearable data rather than clinic visits. A retrospective study followed 92,457 US adults wearing wrist trackers for at least 35 weeks each, yielding roughly 33 million daily resting-heart-rate values[23]. The findings:
- The average was 65 beats per minute, but individual personal averages ran from 40 to 109 — a spread of about 70 beats between one wearer and another.
- Personal averages differed systematically by age, sex, body mass index and habitual sleep duration.
- There is a seasonal cycle: lowest in July, highest in January.
- Within a person the value was fairly stable — but one in five had at least one week where it shifted by 10 beats per minute or more.
So two people can both be entirely normal and 40 beats apart, and the same person can drift 10 beats in a week without anything being wrong. A single reading compared against a population band is close to uninformative. A reading compared against your own rolling average, built over weeks, is a real measurement. Keep the method the same, too: in about 10,900 UK adults the average was 67 beats per minute sitting, 64 lying down and 57 asleep[24], so on average a sleeping heart rate runs about 10 beats below a seated one, and an overnight wearable figure should not be compared directly with a seated clinic reading. For heart rate itself, consumer wearables are on average within about 3% of reference equipment[25].
What actually moves it
Training — Strong, but modest and slow. A systematic review of 191 studies covering 215 samples of healthy people found that every form of exercise studied lowered resting heart rate, though only endurance training and yoga reached statistical significance in both sexes[26]. That is a statement about which effects were reliably detected, not a claim that yoga and endurance training are equivalent — the review did not compare them head to head. Interval training works too: across 97 randomised trials in the general population, high-intensity interval training lowered resting heart rate by about 4 beats per minute compared with not exercising[27]. Two patterns are more useful than the pooled average: the fall is larger the higher you start, and smaller the older you are.
In sedentary adults over 60, pooled controlled trials put the effect at about 6 beats per minute (range 2–12), an 8.4% reduction — and found that training programmes shorter than about 30 weeks generally did not produce a significant fall in this age group[28]. This is a slow adaptation, not a response you will see in a fortnight. See Zone 2.
Alcohol — Strong, and the most visible thing in your own data. A controlled laboratory study in 26 healthy adults found that drinking before bed raised heart rate through the night by about 4% after one to two drinks and 14% after three to four (the lower count for women), compared with a no-alcohol night. The rise was biggest early in the night; after the lower dose it faded by morning, after the higher dose it lasted all night[29]. See Alcohol.
Omega-3, specifically DHA — Moderate. Pooling 51 randomised trials in roughly 3,000 people, supplementation lowered heart rate by 2.2 beats per minute (1.4–3.1). In the trials that gave one fatty acid on its own, DHA lowered heart rate by about 2.5 beats per minute while EPA showed no clear effect — a comparison across different trials rather than a head-to-head test, in a review partly written by staff of an omega-3 manufacturer[30]. See Omega-3.
Nicotine — Caution. Vaping briefly speeds up the heart: pooled studies found heart rate about 1.4 beats per minute higher within an hour after a session than just before it (0.8–2.0), with small rises in blood pressure and signs of temporarily stiffer arteries. The studies vary widely and smaller positive ones may be over-represented, and none of this tells you how much regular vaping shifts your resting heart rate[31]. See Smoking and nicotine.
Sauna — Weak-to-null. Pooling the eleven randomised passive-heating trials that measured it, two to fifteen weeks of regular heat produced no change in resting heart rate (0.1 beats per minute, −1.2 to 1.5)[32]. See Sauna.
Sleep timing — real but tiny. Across 255,736 nights in 557 students wearing fitness trackers, going to bed just 30 minutes later than usual was linked with a heart rate through the night that was higher by 0.18 beats per minute (0.11–0.26); going to bed an hour or more earlier than usual was also linked with a higher overnight rate[33]. Statistically unambiguous and practically negligible — a useful calibration on how small most single-night influences really are, and a reason not to over-interpret a bad morning.
Beta-blockers and other rate-slowing drugs — Strong (that they decouple the number). These decouple the number from everything it normally reflects. On a beta-blocker, resting heart rate stops being a readout of autonomic state and becomes a readout of the dose.
GLP-1 weight-loss drugs (glucagon-like peptide-1 receptor agonists) — Strong (that they push it up). Semaglutide, liraglutide, tirzepatide and similar drugs raise resting heart rate, by about 3 to 4 beats per minute on average in people without diabetes, and more at the highest doses of some newer drugs[34]. For tirzepatide, in people with type 2 diabetes, the rise grows with the dose[35]. In pig experiments, GLP-1 itself — the hormone these drugs mimic — sped up the heart's own pacemaker directly, even with the nerve signals to the heart blocked[36]. So on these drugs a rising number is expected and is not a sign of lost fitness; compare against your on-drug baseline.
When a single reading means nothing — Weak / qualitative
Any of the following makes one measurement uninterpretable: a recent meal, caffeine, alcohol the night before, standing rather than sitting, time of day, the month of the year, an infection you are incubating, dehydration, a poor night's sleep, anxiety in a clinic room, or a thyroid that is over- or under-active — see Thyroid. None of these has a well-quantified pooled effect size, and no attempt is made here to invent one — which is itself the point: the list is long, the magnitudes are unknown, and the sensible response is to stop reading single numbers.
When a low heart rate is a problem — consensus criteria; mechanism contested
For most people a slow pulse is good news, and endurance athletes routinely sit in the 40s. But the bottom of the range is not unconditionally benign.
The consensus criteria for interpreting athletes' electrocardiograms treat sinus bradycardia above 30 beats per minute in an asymptomatic trained athlete as a normal training-related finding needing no further evaluation. At or below 30 beats per minute, or with pauses of 3 seconds or longer, the response of heart rate to exercise should be assessed[37]. Such readings are uncommon but not rare. On 24-hour monitoring of 465 mostly young elite endurance athletes (median age 23), about 4 in 10 dipped to 40 beats per minute or below, 1 in 50 to 30 or below, and 3% had pauses of 3 seconds or more. Over five and a half years, slow rates and pauses were not linked to health problems, though the group with the most extreme readings was very small. Athletes whose genetic profile predicted a lower heart rate were about twice as likely to be bradycardic, so genes as well as training appear to shape how low the rate goes[38].
The discriminator is not the resting number but the response. A trained heart accelerates normally on demand and causes no symptoms. A sinus node that is failing does not — and the giveaways are dizziness, fainting, breathlessness or exhaustion on exertion out of proportion to fitness. Those symptoms with a slow pulse warrant evaluation regardless of how much you train.
It is worth knowing that why athletes' hearts slow is genuinely unsettled. The traditional explanation is high vagal tone — the parasympathetic brake described in heart rate variability. The competing account, argued in a formal exchange in The Journal of Physiology, is that endurance training structurally remodels the sinus node itself, downregulating the ion channel that drives its pacemaker current[39][40]. If the second account is even partly right, some of training-induced bradycardia is a lasting structural change rather than a fully reversible shift in tone. Registry data are consistent with this, though they cannot tell the two accounts apart. Among about 209,000 Swedish long-distance cross-country skiers, men had about 17% more pacemaker implants than non-skiers (hazard ratio 1.17). The highest rates were in men who had finished the most races and posted the fastest times, and a malfunctioning sinus node (sick sinus syndrome) was a more common reason for the pacemaker than among non-skiers. Women showed no excess, and among skiers, death rates did not differ between those who had a pacemaker and those who did not[41]. An earlier study of the same race found that men with five or more finishes were about twice as likely to be admitted to hospital with a slow heart rhythm as men who had finished once (hazard ratio 2.10)[42]. These are associations in registry data, not proof that the training caused it.
Resting heart rate versus heart rate variability — a comparison, not a finding
These are different measurements of the same beat sequence, and they are easy to conflate. Resting heart rate is how often the heart beats; heart rate variability is how much the interval between beats fluctuates. They move in opposite directions in health — a slower heart usually varies more — and that inverse relationship is strong enough that an uncorrected variability figure partly just reflects the rate.
For practical purposes, resting heart rate is the more robust of the two: it is easier to measure, and its association with mortality rests on far larger cohorts. In a 2025 overnight test against an electrocardiogram reference — 13 adults over 536 nights — wearables' resting heart rate was typically off by about 3% or less per night, against roughly 6–16% for variability, though agreement still differed between brands[43]; the details are in heart rate variability. Variability carries information that rate does not, but it is noisier and less comparable between devices and between people. Between the two, rate is the one to track if you are only going to track one.
Practical guidance
- Measure against yourself, not a chart. The only useful comparison is your own rolling average over weeks. A population "normal range" of 60–100 beats per minute is not a target — among about 92,000 wearable users, personal averages ran anywhere from about 40 to 109[44].
- Prefer continuous or overnight measurement to a spot check. An average over many hours is less exposed to the noise of a single moment, and in the one small study that compared the two, the 24-hour average looked like the stronger predictor — though with only 48 deaths the difference was not clear-cut. Wearables measure the rate more reliably than they measure variability, so this is the number your device is best at.
- If you take it manually, do it the same way each time: seated, at rest for several minutes, before caffeine, at a consistent time of day. Count for a full 60 seconds.
- Watch the trend, not the day. A rise sustained over weeks or months is the signal. A bad morning is noise, and the noise is larger than most single influences on it.
- Treat a rise as a prompt to look for a cause, not a number to correct. New medication, poor sleep, drinking more, an untreated thyroid problem, a developing illness, deconditioning. The number is the smoke.
- There is no case for taking a drug to lower it if you are otherwise well. Ivabradine was tested in 19,102 people with coronary disease and did not help, and in 17,801 heart-attack survivors with normal pumping function, randomised to a beta-blocker or none, the beta-blocker did not lower their risk of death, another heart attack or heart failure either[45]. There is no version of this experiment in healthy people, and no reason from the existing trials to expect a different answer.
- The thing that lowers it durably is aerobic training, and it works slowly — plan in months, not weeks, and expect more movement if you are starting from a fast baseline.
- A slow pulse with symptoms is not athletic. Dizziness, fainting or breathlessness on exertion alongside a low heart rate deserves evaluation.
What's overrated
- Chasing a lower number. The trials that lowered it directly failed everywhere except heart failure, and one of them produced bradycardia in nearly a fifth of participants.
- The 60–100 range as a personal verdict. Among about 92,000 wearable users, individual averages spanned 40 to 109; the band tells you almost nothing about one person.
- A single clinic reading. It captures one noisy moment. In the one small study that compared the two, a 24-hour average looked like the better predictor, though the evidence is thin.
- Reading it as a fitness score. It survives adjustment for a directly assessed fitness measure, so it is not simply fitness — and conversely, an active person with a fast heart largely closes the fitness gap anyway.
- Precision about the effect size. "17% per 10 beats" is quoted everywhere; the studies behind it disagree so violently that the honest claim is a direction and a rough magnitude, not a figure to two significant digits.
Further reading
- Aune D, et al. Resting heart rate and the risk of cardiovascular disease, total cancer, and all-cause mortality: a systematic review and dose-response meta-analysis of prospective studies. Nutr Metab Cardiovasc Dis 2017.[46]
- Zhang D, et al. Resting heart rate and all-cause and cardiovascular mortality in the general population: a meta-analysis. CMAJ 2016.[47]
- Jensen MT, et al. Elevated resting heart rate, physical fitness and all-cause mortality: a 16-year follow-up in the Copenhagen Male Study. Heart 2013.[48]
- Seviiri M, et al. Resting heart rate, temporal changes in resting heart rate, and overall and cause-specific mortality. Heart 2018.[49]
- Wen CP, et al. Resting heart rate — the forgotten risk factor? Comparison of resting heart rate and hypertension as predictors of all-cause mortality in 692,217 adults in Asia and Europe. Prog Cardiovasc Dis 2025.[50]
- Fox K, et al. Ivabradine in stable coronary artery disease without clinical heart failure (SIGNIFY). N Engl J Med 2014.[51]
- Swedberg K, et al. Ivabradine and outcomes in chronic heart failure (SHIFT): a randomised placebo-controlled study. Lancet 2010.[52]
- Bangalore S, et al. Relation of beta-blocker-induced heart rate lowering and cardioprotection in hypertension. J Am Coll Cardiol 2008.[53]
- Quer G, et al. Inter- and intraindividual variability in daily resting heart rate and its associations with age, sex, sleep, BMI, and time of year: retrospective, longitudinal cohort study of 92,457 adults. PLoS One 2020.[54]
- Reimers AK, et al. Effects of exercise on the resting heart rate: a systematic review and meta-analysis of interventional studies. J Clin Med 2018.[55]
- Korshøj M, et al. The relation of ambulatory heart rate with all-cause mortality among middle-aged men: a prospective cohort study. PLoS One 2015.[56]
- Drezner JA, et al. International criteria for electrocardiographic interpretation in athletes: consensus statement. Br J Sports Med 2017.[57]