Counts describe this library, not the strength of evidence.
How to read this evidence
Recovery is not one number. Cold-water studies separate soreness, power and strength; newer analyses make uncertainty more visible. HRV depends on breathing and measurement context, and a workload ratio is not an injury prediction.
In 962 military men (median age 37.6 years), every short-term heart rate variability (HRV) measure was higher during slow breathing at six breaths a minute than during spontaneous breathing, injured or not; in fast resting breathers RMSSD rose from 15.57 to 21.89 ms. It was one session in a fixed order, so it shows an immediate effect, not lasting benefit.
Across 22 randomised trials (375 participants, mostly men), one cold-water immersion after exercise did not change maximal isometric strength (g 0.08, 95% CI −0.11 to 0.26) and was followed by weaker jumping straight afterwards (g −0.68), gone by 24–48 hours. Lower soreness and creatine kinase looked favourable but weakened in sensitivity checks, and certainty was low to very low.
In a meta-analysis of 52 randomised trials, cold-water immersion after high-intensity exercise was followed 24 hours later by lower soreness (SMD −0.89), higher perceived recovery (0.66) and better muscular power (0.22), with no effect on strength. Participants could not be blinded, most were men, and many pooled results rest on few studies with high heterogeneity.
In 962 military men (median age 37.6 years), every short-term heart rate variability (HRV) measure was higher during slow breathing at six breaths a minute than during spontaneous breathing, injured or not; in fast resting breathers RMSSD rose from 15.57 to 21.89 ms. It was one session in a fixed order, so it shows an immediate effect, not lasting benefit.
In a pilot trial of 12 postgraduate students, daily 10-minute app-guided resonance frequency breathing for three weeks raised HRV relative to controls (SDNN at Week 3, p = 0.026, r = 0.78) but did not produce a significant difference in perceived stress (p = 0.199). With 6 people per group and no active control, the findings are preliminary.
In 56 people with chronic shoulder pain and 52 pain-free controls, viewing shoulder movement pictures went with lower parasympathetic HRV during (HF: p = 0.04) and after (HF: p = 0.03) the pictures. The study is cross-sectional with one exposure session, so it cannot show that pictures cause these changes or that autonomic differences cause pain.
Among 600 health sciences students in Türkiye, higher cyberchondria scores were associated with lower resting HRV (RMSSD; adjusted beta -0.256, p < 0.001), adding 4.9% to explained variance. Health anxiety was not measured and the study is cross-sectional, so it cannot say whether online health searching or underlying anxiety is responsible.
In 36 people who practised four yogic breathing techniques, calming techniques and accelerating techniques differed significantly in their effect on heart-rate variability (rMSSD) and breathing rate. The abstract gives no effect sizes or numbers for these differences, so how large they are is unknown.
In 36 human intervention studies, several reported less perceived muscle soreness with plant foods such as tart cherry, beetroot or berries, mostly within 24-72 h after exercise. Results were not uniform, and the evidence does not establish benefits for muscle-function recovery, later performance or biochemical markers.
Across 16 randomized trials contributing 120 effect sizes, caffeine had no clear effect on perceived or physiological recovery, a small benefit for functional performance during recovery (g = -0.208, 95% CI -0.393 to -0.023) and a worse sleep outcome (g = 0.751, 95% CI 0.244 to 1.258, from only 4 studies). The functional and sleep results were not stable in sensitivity analyses, and both prediction intervals crossed zero.
Across 266 sessions from 11 elite under-17 male soccer players, cumulative heart-rate exertion tracked total distance (r = 0.921) and PlayerLoad (r = 0.924) almost perfectly within each player. With only 11 players from one setting, the findings describe these players' training and may not hold for other teams or ages.
In 237 male soccer players aged 8-18, the best Random Forest model with short resting HRV predicted Yo-Yo-derived endurance with an RMSE of 2.82-2.89 ml/kg/min, versus 3.15 for a demographics-only model. The gain from HRV was modest and the models are exploratory, not ready to replace the Yo-Yo test or guide individual training.
In 9 elite speed climbers monitored for 30 weeks, higher training monotony and creatine kinase (CK) averages went with lower relative performance, and higher testosterone with better performance. The only GEE finding (CK variability, β = 0.529, P = 0.027) did not survive FDR correction, and with 9 athletes the results are tentative.
In 82 healthy people viewing emotional images, a first positive image went with higher HRV than a first negative one, but no reversal appeared when the second image had the opposite emotion. So HRV followed mood only when mood was stable, and the abstract reports no numerical effect sizes.
After a 24-week low-dose mind-body exercise program in older adults living at about 53°S in Chilean Patagonia, resting HRV did not change uniformly (RMSSD β = -0.14, 95% CI -1.41 to 1.06), but the SNS index in exercise-evoked responses shifted (β = -1.63, 95% CI -3.05 to -0.24). It was a small, uncontrolled sample with a chest-worn sensor, so the changes cannot be attributed to the program itself.
In 71 healthy young adults, a 10-week high-intensity interval training (HIIT) program, with or without daily beer, vodka or non-alcoholic drinks, produced no statistically significant between-group differences in heart rate variability (all p > 0.05). HRV was only a secondary outcome and the study was not powered to detect differences in it, so this is not evidence that alcohol or HIIT leaves HRV unchanged.
This scoping review of 11 maritime field studies found chest straps were the most common device (8 studies) and HRV measurement at sea is feasible but methodologically demanding. Reporting of recording duration, software, artifact handling and HRV metrics varied widely, so results cannot be pooled or compared.
Journal of occupational medicine and toxicology (London, England)
Among 550 adults with normal thyroid function, a 1-unit higher TSH index (TSHI) was associated with a 3.488 bpm lower 24-hour mean heart rate (95% CI -4.633 to -2.343; P < 0.001). Opposing FT3 and HRV components largely cancelled out, and this cross-sectional statistical decomposition does not show cause.
In 65 healthy women with a history of childhood emotional abuse (CEA) and 64 matched controls, the CEA group reported more depressive symptoms, pre-sleep arousal and sleep disturbance, but selected objective sleep measures and nocturnal HRV did not differ. Objective data came from one night on a wrist device, so subtle differences may have been missed.
In 15 soccer players, a 15 °C cold water immersion between two shuttle-run tests preserved shuttle-run performance (TD, p = 0.21 for change) and gave lower creatine kinase (p = 0.01) and lactate dehydrogenase (p = 0.003) at 48 hours than passive recovery. Foam rolling did little, and with only 15 players and many outcomes tested, the findings need confirmation.
In 30 healthy adults, mental arithmetic caused the largest overall HRV reductions, cold pain affected only MeanNN, and noise caused no global change, yet 30-second windows showed distinct patterns for each. With 30 participants and lab tasks, results show that stressors differ, not how everyday stress affects your HRV.