What your stress score is measuring

In exam weeks, Division I college football players had 1.78 times the odds of an injury restriction compared with ordinary in-season weeks. Mann and colleagues (2016, Journal of Strength and Conditioning Research) tracked 101 players across a season and sorted every week into preseason camp, weeks with scheduled exams, or regular weeks without them. Camp was the riskiest overall. Among the players who regularly played in games, the gap between camp and exam weeks disappeared, with an odds ratio of 1.13 and a p value of 0.75. For the athletes who mattered most on game day, a week of midterms carried about the same injury risk as training camp.

The coaches could see camp coming in their practice schedule. The academic load showed up nowhere in their training data, and it hit the same bodies. A wearable stress score is an attempt to put that second kind of load on a screen, by reading the one system both kinds of stress pass through, the autonomic nervous system.

This article covers what that score is built from and why its numbers disagree with each other and with how you feel. It then separates one hard day from an accumulating burden and sets out what to change in the 24 hours after a high reading. The Titan math comes from the Stress score, Battery and notifications help articles. For the HRV basics behind all of it, start with the glossary entry on HRV.

01What HRV registers during stress

Kim and colleagues (2018, Psychiatry Investigation) pooled 37 publications from 2007 to 2017 that measured heart rate variability under psychological stress. The pattern was consistent. Stress lowered high-frequency power and raised low-frequency power, pushed the LF/HF ratio up, and reduced the time-domain measures RMSSD and pNN50. The authors read this as a drop in parasympathetic activity, the vagal brake that slows the heart between beats, and concluded that HRV supports the objective assessment of stress. They also noted that the reading should be interpreted alongside the person's medical history.

The effect holds outside the lab. Järvelin-Pasanen and colleagues (2018, Industrial Health) reviewed 10 studies that recorded HRV during real work, most with 24-hour or longer Holter ECG or heart rate monitors, and sample sizes from 19 to 653. Higher occupational stress went with lower RMSSD and HF power and a higher LF/HF ratio. That is the same signature Kim found in laboratory stress tasks.

Thayer and colleagues (2012, Neuroscience & Biobehavioral Reviews) explain why a heart signal should track a mental state. Their meta-analysis of neuroimaging studies found regions whose blood flow moved with HRV across studies, including the amygdala and the ventromedial prefrontal cortex. Their model treats HRV as a readout of how strongly the prefrontal cortex, judging a situation safe, restrains the brainstem's default threat response. When prefrontal control holds, the vagal brake stays on and HRV stays up. When it slips, the brake comes off and the heart beats more evenly and faster. Your watch never sees your thoughts. It sees the brake.

02The two branches of a stress response

The stress response runs on two timescales, and a wearable sees only one of them directly.

The fast branch is sympathetic. Harvard Health's summary of the physiology describes the amygdala signaling the hypothalamus, which fires sympathetic nerves and tells the adrenal glands to release epinephrine. Heart rate, blood pressure and breathing rise within seconds, often before the visual system has finished working out what the threat is. At the same moment, the parasympathetic brake eases off. These two moves, sympathetic push and vagal withdrawal, are what heart rate and HRV record. They happen fast enough for a heart rate sensor to catch them.

The slow branch is hormonal. As the epinephrine surge fades, the hypothalamic-pituitary-adrenal axis takes over. The hypothalamus releases corticotropin-releasing hormone, the pituitary answers with adrenocorticotropic hormone, and the adrenal cortex releases cortisol. Cortisol keeps the sympathetic system engaged while a threat persists. It rises and falls over hours. No consumer wearable measures it. When cortisol stays high, it shows up in the autonomic signal only indirectly, through a heart that stays faster and less variable than it should for the setting.

The time lag matters for reading the score. Leproult and colleagues (1997, Sleep) measured cortisol across 32 hours in healthy young men after normal sleep, partial sleep deprivation and total sleep deprivation. The morning after the short night looked normal. The change arrived the following evening, when cortisol ran 37% higher after partial deprivation and 45% higher after total deprivation, and the nightly drop to low cortisol started at least an hour late. The hormonal cost of a bad night lands the following evening, right when you are trying to fall asleep again.

What allostatic load adds to the picture

McEwen (2006, Dialogues in Clinical Neuroscience) gives the vocabulary for separating a hard day from a hard season. Allostasis is the body staying stable by changing, releasing epinephrine and cortisol to meet a demand and then switching them off. Allostatic load is "the wear and tear that results from either too much stress or from inefficient management of allostasis," for example not turning the response off when it is no longer needed.

McEwen names four ways the load accumulates. Repeated hits are many separate stressors in a row. Lack of adaptation is a failure to habituate, so the same weekly meeting provokes the same surge for months. A prolonged response is a surge that does not shut off after the stressor ends. An inadequate response is one system under-responding, which forces others to compensate. A single elevated Stress reading on a deadline afternoon fits none of these. It is allostasis doing its job. The patterns that describe load show up across days, as a response that is still elevated at bedtime or a reading that never settles on weekends.

Juster, McEwen and Lupien (2010, Neuroscience & Biobehavioral Reviews) show how researchers put a number on that burden. An allostatic load index combines markers from the neuroendocrine, immune, metabolic and cardiovascular systems, and across many studies the composite predicted illness and death beyond standard clinical markers. A wrist-worn Stress score measures one of those systems, autonomic cardiac control. Two people can post the same Stress score today while carrying very different loads, because the score cannot see their blood pressure, their inflammation or the last six months.

The practical translation is to treat a single day's number as a measure of allostasis and a run of days as an estimate of load. Titan's Recovery guidance applies the same logic to HRV. One low morning tells you less than a week of them (managing recovery and stress).

03What a wearable stress score is built from

Every consumer stress score shares one design decision. It compares your heart rate and HRV with your own history, not with a population table. The reason is the spread between people. Shaffer and Ginsberg (2017, Frontiers in Public Health), the standard reference on HRV norms, show that HRV falls with age, with the steepest decline between the second and third decades, and shifts with breathing rate, posture and time of day. A 35 ms reading is normal for one adult and a bad day for another. A fixed threshold would label some people stressed every day and others never.

Where the scores diverge is in what counts as a valid moment to measure and how the deviation becomes a number.

Oura's Daytime Stress uses daytime HRV, heart rate, temperature trends and accelerometer data. It sorts moments into four states, Stressed, Engaged, Relaxed and Restored. Oura states that stress and recovery "can only be measured during no or low movement periods," so it leaves active periods unscored. Garmin's stress feature uses algorithms built with Firstbeat Analytics. According to Lifestack's explainer of the method, it samples optical HRV continuously and tries to recognize when a raised heart rate comes from exercise so that it does not count as stress.

Titan takes a more transparent route. The formula is two ratios and an average, and the help article publishes it (Stress score).

The ratio math behind Titan's Stress score

The first ratio is your baseline HRV divided by today's HRV. The second is today's heart rate divided by your baseline resting heart rate. Titan averages the two, subtracts 1, multiplies by 100 and rounds. Anything below 0 becomes 0 and anything above 100 becomes 100. If HRV is at or above baseline and heart rate is at or below it, the score is 0. The bands are Low 0 to 25, Moderate 26 to 50, Elevated 51 to 75 and High 76 to 100.

The headline value on Today uses the same single HRV value that drives your Recovery score, a morning Mindfulness session or your average sleep HRV by default, along with today's resting heart rate. It compares them with the median of your Recovery baseline window, 60 days unless you picked 7 or 30 during onboarding (baselines). A baseline needs at least 7 days of readings covering at least 60% of the window. Until Titan has that, the headline shows 100 minus your Recovery score.

Running the formula on a range of mornings shows how the scale behaves. Every row below assumes a baseline HRV of 50 ms and a baseline resting heart rate of 50 bpm. The mornings are illustrative.

Three properties of the scale come straight out of that table.

The HRV side is inverted, so it accelerates. Dropping from 50 to 40 ms adds 0.25 to the HRV ratio. Dropping another 10 ms, from 30 to 20 ms, adds 0.83. Each millisecond lost counts for more as HRV falls, which is the right shape, since a reading far below your normal is disproportionately informative. The heart rate side is linear. Each beat above a 50 bpm baseline adds 0.02 to its ratio, whether it is the first beat or the tenth.

The Low band is wide. A morning with HRV 20% under baseline and resting heart rate 10% over it still scores 18. The headline reaches Elevated only when HRV is down by roughly half and resting heart rate has climbed several beats. That follows from the ratio design, and it means a headline in the Elevated or High band is a large deviation from your own normal. It deserves a look at what else changed in the last day or two.

The ratios ignore your normal spread. Two people with a 50 ms baseline get the same Stress score from a drop to 40 ms, even if one swings 10 ms every other day and the other rarely moves 3 ms.

Why Stress and Recovery can disagree on the same morning

The Recovery score uses the same HRV and resting heart rate as the Stress headline and handles them differently. Recovery measures how far today's HRV and resting heart rate sit from the baseline median in units of your median absolute deviation, your usual day-to-day spread. It adds shorter-term signals, passes the total through a logistic curve, then ranks the result against your last 60 Recovery calculations (Recovery score).

Take the second person above, whose HRV rarely moves 3 ms, waking to 42 ms and 53 bpm. Stress reads 13, Low. Recovery sees an HRV nearly three spreads below normal, rare for this person, and can put the morning in its Low band. Stress answers how far you are from baseline in proportion. Recovery answers how unusual today is for you. When they disagree, Recovery is the better guide to training, because it knows how much your numbers normally move. The blog's rule of thumb for acting on HRV outside the app works the same way. A morning HRV more than one median absolute deviation below your baseline median for two or more consecutive days, with resting heart rate above baseline, is a signal to pull intensity. The HRV and training readiness guide covers that rule in depth.

04Why a morning reading and a daytime reading are different measurements

The most common complaint about stress scores is that the numbers contradict each other. Sleep stress looks fine, the afternoon looks alarming, and the headline says something else. Most of this comes from comparing measurements taken under different conditions and scored against the same baseline.

Shaffer and Ginsberg (2017) put the principle bluntly. Twenty-four-hour, short-term and ultra-short-term HRV values "are not interchangeable." The formulas are the same, and "their physiological meaning can profoundly differ." A reading taken asleep or lying still after waking captures the heart at rest, with the vagal brake fully applied, breathing slow and regular, posture fixed. A reading taken at 2 p.m. captures a heart that is also responding to standing, walking, digesting, talking and caffeine. HF power "may increase at night and decrease during the day," and heart rate itself shapes HRV. Faster heart rates shorten the time between beats and leave less room for the intervals to vary. A daytime reading starts lower than a sleep reading for reasons unrelated to psychological stress.

How Titan scores five different periods

Titan scores five periods, each against its own window of data (Stress score).

Heart rate is the largest source of disagreement between them. For any window, Titan first looks for resting heart rate samples from Apple Health. Sleep and waking windows often have none, so Titan uses your average heart rate over that window. Average heart rate while you move around sits above your resting baseline, so the heart rate ratio rises.

Run the numbers on an ordinary active day, again with baselines of 50 ms and 50 bpm. Sleep averages 48 ms of HRV and 52 bpm, so sleep stress reads 4. The morning headline uses that 48 ms sleep HRV and an Apple Health resting heart rate of 50 bpm, which gives 2. The waking window includes a commute, a walk at lunch and a day on your feet. It averages 78 bpm and 32 ms of HRV. The heart rate ratio is 1.56, the HRV ratio is 1.56, and waking stress reads 56, Elevated.

Nothing about that day was stressful in the psychological sense. The waking value measured a heart doing ordinary daytime work and compared it with a baseline built from rest.

The HRV inputs differ too. The headline uses the single recovery HRV value. The timeline uses every HRV reading recorded during each interval, and Apple Watch records HRV only a few times a day (Battery). Titan picks the finest timeline interval it can fill, from 30 up to 360 minutes, and leaves an interval blank when it has no observation.

Oura handles the same problem by refusing to score movement. Titan scores the whole window and documents the heart rate fallback. Each approach has a cost. Oura leaves gaps across the most active parts of the day. Titan fills them with a number that carries movement in it.

How to read the three together

Use the headline for the state you woke in. It is the cleanest measurement, taken under the same conditions each day, and it is the value Stress Alerts check.

Use sleep stress for whether the response shut off overnight. Sleep is the closest thing to a controlled condition that a day contains. A sleep stress value that rises across several nights while daytime numbers look normal matches McEwen's prolonged response, and Titan's own guidance flags that pattern as worth attention (managing recovery and stress).

Use waking stress and the timeline as a record of what happened, with care. Compare waking stress with your own waking stress on similar days, since it will almost always sit above sleep stress. A desk day that scores higher than a hiking day is informative. An absolute waking value in the 50s on a day you walked 15,000 steps mostly measures the steps.

Battery and Stress are different instruments

Battery is Titan's other HRV-based number, and people expect it to mirror Stress. It uses different math. Battery averages every HRV reading in a smoothing window, 24 hours by default, divides by the median of your daily HRV over a fixed 60 days, multiplies by 100 and by a sensitivity setting, and clamps to 0 to 100 (Battery). It uses no heart rate, and its HRV ratio runs the other way up, today over baseline.

On the active day above, the 24-hour HRV average might come to 40 ms against a 50 ms baseline, which gives a Battery of 80, Partially Charged. The same day produces a Stress headline of 2 and a waking stress of 56. All three numbers are correct. Battery reports a smoothed day of HRV relative to normal, with yesterday evening and last night still inside the window. The Stress headline reports this morning. Waking stress reports a movement-heavy window scored against rest. Battery's labels are Charged 85 to 100, Partially Charged 60 to 84, Low 30 to 59 and Depleted 0 to 29.

When the number and your mood disagree

The other common complaint is that the score says stressed on a day that felt calm, or calm on a day that felt awful. The research expects this. Föhr and colleagues (2015, Journal of Occupational Medicine and Toxicology) compared perceived stress questionnaires with HRV-based stress measured over one to three workdays in 221 adults. Subjective stress was associated with objective stress, with a p value of 0.047, and inversely with objective recovery, at 0.046. The associations were real and weak, and different factors drove each. Older age went with higher physiological stress, and after further adjustment with lower reported stress.

A felt state and an autonomic state overlap without being the same thing. A demanding day you enjoyed still costs sympathetic activation. A day of low-grade dread spent on the couch may barely move heart rate. Treat the gap as information about which of the two systems is carrying the load. Use neither number to overrule the other.

05How training stress and life stress add up in the same body

The autonomic nervous system keeps one account for a deadline and an interval session. Both raise sympathetic drive and suppress the vagal brake, and both draw on the same recovery capacity afterward.

Stults-Kolehmainen and Sinha (2014, Sports Medicine) reviewed 168 studies on stress and exercise and examined 55 prospective ones in detail. In 76.4% of the prospective studies, psychological stress predicted less physical activity or more sedentary time, and the share rose to 85.7% among the higher-quality studies. The review also collected evidence that chronic stress delays recovery from exercise and dampens muscular and neural adaptation.

Two experiments show that effect on the muscle itself. Stults-Kolehmainen, Bartholomew and Sinha (2014, Journal of Strength and Conditioning Research) had 31 students perform a heavy leg press session and tracked maximal isometric force, energy, fatigue and soreness for 96 hours. Life event stress moderated the recovery of force, and higher stress went with worse recovery on every outcome. An earlier study with the same design found stress slowed force recovery within the first hour (Stults-Kolehmainen and Bartholomew, 2012, Medicine & Science in Sports & Exercise). Over a longer horizon, Bartholomew and colleagues (2008, Journal of Strength and Conditioning Research) followed 135 students through 12 weeks of twice-weekly weight training. Students reporting low life stress gained significantly more on the bench press and squat than high-stress students on the same program.

Exercise does not simply cancel the damage. Stults-Kolehmainen, Tuit and Sinha (2014, Stress) interviewed 395 adults about cumulative adversity. More adverse life events went with more health problems, with a correlation of 0.431. At low levels of major life event stress, more exercise went with fewer health problems. At high levels, the benefit of exercise was absent in the sample as a whole. It persisted at all stress levels for men and only at low stress for women.

HRV shows the training side of the ledger less cleanly than people assume. Bellenger and colleagues (2016, Sports Medicine) pooled 27 studies of endurance training and found that training which improved performance produced small increases in resting RMSSD, with a standardized mean difference of 0.58. Overreaching produced something less intuitive. Post-exercise RMSSD rose as well, and the candidate marker of fatigue was a slower heart rate acceleration at the start of exercise. A morning HRV number alone cannot always separate productive fatigue from overreaching. The authors call for additional measures, which is one reason Titan tracks training load separately from Stress and Recovery. For how overreaching develops across a block, see the glossary entry on overreaching.

The evidence linking life stress to injury and illness

Mann and colleagues (2016) supply the clearest applied case. Across the season, camp carried 2.05 times the odds of an injury restriction of exam weeks and 3.65 times those of regular weeks, and exam weeks carried 1.78 times the odds of regular weeks. The authors suggest academic stress may affect the players who play even more than physical load does, and advise coaches to choose training methods carefully in exam weeks.

One team's injury log proves no mechanism. The pattern fits the lab data above. Stress slows the recovery of force and dampens adaptation, and a player who has not recovered from Tuesday's practice goes into Thursday's with less margin. Nothing in a training plan records a stressful week at work. Your morning HRV and resting heart rate do.

06Stress and sleep run in both directions

Stress carries into sleep physiology. Hall and colleagues (2004, Psychosomatic Medicine) gave 59 healthy adults either a stressful speech task or a control task right before bed and recorded ECG through the night. The stressed group showed less parasympathetic modulation in both NREM and REM sleep and a higher sympathovagal balance in NREM. The usual pattern, in which vagal activity climbs with each successive sleep cycle, was blunted. The authors proposed this as one pathway to disturbed sleep. For a wearable user this is the most direct evidence that an evening stressor shows up in sleep HRV, and therefore in the next morning's Stress headline and Recovery score.

Short sleep raises the stress systems in turn. Meerlo, Sgoifo and Suchecki (2008, Sleep Medicine Reviews) reviewed human and rodent studies and found that sleep deprivation and restriction produce mild, temporary increases in sympatho-adrenal and HPA-axis activity. Over time they can also change how strongly those systems react to other stressors. McEwen (2006) cites work in which restricting sleep to four hours a night raised blood pressure, lowered parasympathetic tone, and raised evening cortisol and insulin. A stressful day costs sleep quality, the short night raises the next evening's cortisol, and the next night starts from a worse position.

Training timing enters the same loop. In 16,275 Finnish employees measured with HRV over mainly three days, Föhr and colleagues (2016, BMC Public Health) found that more physical activity went with less physiological stress on workdays and better recovery during sleep over the long term. They also noted that high activity may disturb recovery during the following night. A late hard session is a stressor that lands on your sleep stress. So is a late work email.

In Titan, sleep stress is one of six inputs to the Sleep Quality score, with a weight of 0.10, and a higher value lowers the result. It does not change the headline Sleep score (Sleep score). Its more useful role is as an early reading. An elevated sleep stress value tonight predicts a worse headline tomorrow, because the same overnight HRV feeds both.

07What to change on a high-stress day

Most advice on this question stops at "breathe, sleep, hydrate." Work through the reading in order.

Find out which number is high. A high waking value on an active day is mostly the heart rate fallback described above, and it asks for nothing. A high headline is a morning deviation large enough to matter, given how wide the Low band is. Open the Recovery screen and check the HRV and resting heart rate rows against their baselines. They show which input moved and by how much (managing recovery and stress).

Let Recovery set the session. Check where Recovery sits on the same morning. In its Low band, Titan's guidance is to reduce intensity, prioritize sleep and focus on easy aerobic work. The strength-recovery studies give a specific reason to cut heavy eccentric and high-volume lifting first. Higher life stress went with slower recovery of force across the 96 hours after a single heavy leg press session (Stults-Kolehmainen, 2014). Move the hard session, and do not stack it on the next day if tomorrow's reading is also high.

Protect tonight's sleep. Hall (2004) and Leproult (1997) point to the same target, the hours before bed. Stop work early enough that the last stressor of the day is not the last thing before sleep. Do not add a late hard workout. Skip alcohol. The HRV and training readiness guide covers how sharply it suppresses overnight HRV. Give yourself a longer sleep window than usual, since the cortisol cost of a short night arrives the following evening.

Use paced breathing as a fast lever. Slow breathing works through the same vagal brake the score measures. Lehrer and colleagues (2020, Applied Psychophysiology and Biofeedback) pooled 58 randomized trials of HRV biofeedback and found a small to moderate effect across outcomes, with the largest effects on anxiety, depression, anger and performance. The protocols, doses and the way a morning Mindfulness session changes Titan's HRV input are covered in the mindfulness and breathwork guide.

Check the next two mornings before changing the plan. One elevated day is allostasis. The rule for pulling intensity is two or more consecutive mornings with HRV more than one median absolute deviation below your baseline median and resting heart rate above baseline. If the second morning is back in range, resume the plan. If it is not, treat the week as reduced, and look at sleep first. A deliberate deload week is the standard response when the pattern runs longer.

Tag the cause. Titan ships with 15 system Journal tags, including Job Stress and Travel, plus any custom tags you add (Journal tags). Journal Insights ranks each tag by its measured impact on Stress or Recovery over 60 days and moves tags with fewer than 3 occurrences to Low Confidence (Journal Insights). Titan pairs each tag with the score from the same calendar date, and that score comes from the night before. A stressful Tuesday shows up in Wednesday's headline, so a Job Stress tag on Tuesday is compared with the morning before the stress. Read that tag's ranking with the one-day lag in mind.

For a broader view of managing total load, the glossary entry on stress management covers the terms, and the recovery plan covers the recovery side in full.

08Setting the Stress Alert Level

The setting that turns the Stress score into a trigger is Stress Alert Level, under You > Settings > Notifications. Tap your avatar on Today to reach You. Turn on Allow Notifications, turn on Stress Alert, which is off by default, and set the level with the minus and plus buttons, one point per tap from 0 to 100 (notifications).

The alert fires when the headline Stress is at or above the level. It skips the 30 minutes after a workout ends and sends at most one alert per hour. Titan checks only when the Today screen refreshes, so an alert can arrive late if the app has not refreshed. The default level is 75, the top of Elevated. Set it to 76 to hear only about High stress, as managing recovery and stress recommends.

Choose the level knowing what the math requires. Reaching 76 needs the two ratios to average about 1.755. With baselines of 50 ms and 50 bpm that is a morning like 22 ms and 62 bpm, a large deviation from normal. An alert at 76 will rarely fire, and when it does the next 24 hours should change as described above. A level of 51 fires at the start of Elevated, around 27 ms and 58 bpm on the same baselines, and gives earlier warning for people who want it. With Prioritize Sleep or Mindfulness HRV on, the default, the headline follows your morning HRV, so the alert tells you about the state you woke in. Turn that setting off under Settings > Recovery & Sleep > Recovery Preferences and the headline follows new waking HRV through the day, with the daytime caveats above (Recovery tools).

09References

• Bartholomew JB et al. (2008). Strength gains after resistance training: the effect of stressful, negative life events. Journal of Strength and Conditioning Research 22(4):1215-1221. doi:10.1519/JSC.0b013e318173d0bf
• Bellenger CR et al. (2016). Monitoring athletic training status through autonomic heart rate regulation: a systematic review and meta-analysis. Sports Medicine 46(10):1461-1486. doi:10.1007/s40279-016-0484-2
• Föhr T et al. (2015). Subjective stress, objective heart rate variability-based stress, and recovery on workdays among overweight and psychologically distressed individuals: a cross-sectional study. Journal of Occupational Medicine and Toxicology 10:39. doi:10.1186/s12995-015-0081-6
• Föhr T et al. (2016). Physical activity, body mass index and heart rate variability-based stress and recovery in 16 275 Finnish employees: a cross-sectional study. BMC Public Health 16:701. doi:10.1186/s12889-016-3391-4
• Hall M et al. (2004). Acute stress affects heart rate variability during sleep. Psychosomatic Medicine 66(1):56-62. doi:10.1097/01.psy.0000106884.58744.09
• Harvard Health Publishing (2024). Understanding the stress response. health.harvard.edu
• Järvelin-Pasanen S et al. (2018). Heart rate variability and occupational stress, systematic review. Industrial Health 56(6):500-511. doi:10.2486/indhealth.2017-0190
• Juster RP et al. (2010). Allostatic load biomarkers of chronic stress and impact on health and cognition. Neuroscience & Biobehavioral Reviews 35(1):2-16. doi:10.1016/j.neubiorev.2009.10.002
• Kim HG et al. (2018). Stress and heart rate variability: a meta-analysis and review of the literature. Psychiatry Investigation 15(3):235-245. doi:10.30773/pi.2017.08.17
• Lehrer P et al. (2020). Heart rate variability biofeedback improves emotional and physical health and performance: a systematic review and meta analysis. Applied Psychophysiology and Biofeedback 45(3):109-129. doi:10.1007/s10484-020-09466-z
• Leproult R et al. (1997). Sleep loss results in an elevation of cortisol levels the next evening. Sleep 20(10):865-870. doi:10.1093/sleep/20.10.865
• Lifestack (n.d.). How does Garmin measure stress. lifestack.ai
• Mann JB et al. (2016). Effect of physical and academic stress on illness and injury in Division 1 college football players. Journal of Strength and Conditioning Research 30(1):20-25. doi:10.1519/JSC.0000000000001055
• McEwen BS (2006). Protective and damaging effects of stress mediators: central role of the brain. Dialogues in Clinical Neuroscience 8(4):367-381. PMC3181832
• Meerlo P et al. (2008). Restricted and disrupted sleep: effects on autonomic function, neuroendocrine stress systems and stress responsivity. Sleep Medicine Reviews 12(3):197-210. doi:10.1016/j.smrv.2007.07.007
• Oura (n.d.). Discover Oura's Daytime Stress feature. ouraring.com
• Shaffer F, Ginsberg JP (2017). An overview of heart rate variability metrics and norms. Frontiers in Public Health 5:258. doi:10.3389/fpubh.2017.00258
• Stults-Kolehmainen MA, Bartholomew JB (2012). Psychological stress impairs short-term muscular recovery from resistance exercise. Medicine & Science in Sports & Exercise 44(11):2220-2227. doi:10.1249/MSS.0b013e31825f67a0
• Stults-Kolehmainen MA et al. (2014). Chronic psychological stress impairs recovery of muscular function and somatic sensations over a 96-hour period. Journal of Strength and Conditioning Research 28(7):2007-2017. doi:10.1519/JSC.0000000000000335
• Stults-Kolehmainen MA, Sinha R (2014). The effects of stress on physical activity and exercise. Sports Medicine 44(1):81-121. PMC3894304
• Stults-Kolehmainen MA et al. (2014). Lower cumulative stress is associated with better health for physically active adults in the community. Stress 17(2):157-168. doi:10.3109/10253890.2013.878329
• Thayer JF et al. (2012). A meta-analysis of heart rate variability and neuroimaging studies: implications for heart rate variability as a marker of stress and health. Neuroscience & Biobehavioral Reviews 36(2):747-756. doi:10.1016/j.neubiorev.2011.11.009