Episode Transcript
[00:00:00] Welcome to this Week in Heart Rate Variability, the show where we take the newest peer reviewed heart rate variability research and turn it into something clinicians, researchers, coaches and practitioners can actually use in their work. I'm your host, Matt Bennett, and I'm glad you're here. Before we get started, a quick disclaimer Everything you hear on this podcast is for educational and informational purposes only. Nothing we discuss should be taken as medical advice, diagnosis or treatment recommendation, and none of it should replace a conversation with a qualified professional who knows your full history or or the full history of your client or patient. Heart rate variability research moves quickly and a single study, however well designed, is rarely the last word on any question. We try to be honest about that every single week, and this week is no exception. You'll also notice that we talk a lot on this show about the difference between an association and a cause. And that's not just academic hair splitting. Most of the studies we cover are observational by design, meaning researchers are watching what naturally happens in a group of people rather than deliberately assigning some people to one condition and others to another and comparing the results.
[00:00:53] That kind of observational evidence is genuinely valuable and it's often the only ethical or practical way to study a question, but it does mean we have to be careful about the language we use when we talk about it, both here on the show and in our own clinical or research work. This is a full episode with eight studies on the docket and it's a genuinely wide ranging one. We're going to look at how alcohol related cues and a laboratory stress task affect craving, subjective stress and heart rate variability in young adults who drink hazardously. We're going to look at how short term heart rate variability tracks changing suicide risk over the course of a single week in people hospitalized for depression. We're going to look at a statistical analysis of how simple changes in posture shift heart rate variability in young healthy adults. We're going to look at a multi year study following preterm infants from birth through early development, tracking how their heart rate, heart rate variability, respiration and oxygen saturation mature over their first weeks of life and and how that maturation connects to outcomes two years later. We're going to look at a small but clever pilot study trying to design a hospital nurse call sound that's easy to notice without spiking a nurse's stress response. We're going to look at a large, wearable based study connecting heart rate variability across the menstrual cycle to premenstrual disorder symptoms. We're going to look at a massive database analysis of seasonal changes in heart rate variability across different regions of Japan. And we're going to close with a study looking at how heart rate variability behaves during and after moderate exercise in people living with obstructive sleep apnea. Addiction, psychiatry, basic measurement science, neonatology, occupational health, women's health, population physiology, and exercise science all in one sitting. Let's get right into it. Our first study was published in Addictive Behaviors and is titled the Relationship between Subjective Arousal and Craving and Physiological Heart Rate Variability, Stress and Alcohol Cue Reactivity in Young Adults who Engaged in Hazardous Drinking A Counterbalance Study the authors are Insign Firdaus, Enna Kleibor, Annemarie Ekog, and Anya C. Huizink. Anyone who has worked with a client trying to change their drinking knows that craving rarely shows up in a vacuum. It's triggered by something specific a stressful day at work, the sight of a bottle on a counter, the smell of a familiar bar. Researchers have long debated which of those triggers matters more, and whether the body's stress response and its craving response are really the same underlying phenomenon wearing two different masks, or whether they're separate systems that just happen to overlap. And in the messiness of daily life, that distinction isn't just academic. If stress and craving really are the same process, then reducing stress should reliably reduce craving. But if they're separable, a person could feel perfectly calm and still experience a powerful pull toward alcohol, or feel intensely stressed with no craving at all. And a clinician who assumes the two always travel together could easily miss what's actually driving a client's risk in a given moment. This study set out to pull those two threads apart in a controlled setting and using the beat to beat rhythm of the heart as an objective window into what was happening physiologically. While participants reported in real time what they were feeling, Firdaus and colleagues recruited 56 young adults who met criteria for hazardous drinking, meaning their alcohol use placed them at meaningfully elevated risk for harm, even without necessarily meeting full criteria for an alcohol use disorder diagnosis. Each participant went through a counterbalanced laboratory protocol, meaning every person completed both the stress task and and an alcohol cue task, with the order of those two tasks varied systematically across participants so that the sequence itself couldn't quietly bias the results. The stress task was a mental arithmetic challenge, a classic and reliable way to provoke measurable psychological stress in a controlled setting. The alcohol cue task involved actual physical handling of an alcoholic beverage, engaging sight, smell, and touch simultaneously rather than relying on a photograph or a verbal prompt alone, which matters because multi sensory cues tend to be more ecologically realistic triggers than a single sense in isolation.
[00:04:34] Throughout both tasks, the researchers recorded continuous heart rate variability alongside self reported craving and subjective stress ratings, allowing them to directly compare the physiological signal against what participants said they were experiencing moment to moment rather than relying on self report alone. The findings were more nuanced than a simple story of stress equals craving. Handling the alcoholic beverage increased both craving and heart rate variability, which on its face might seem counterintuitive since we often associate heightened autonomic nervous system arousal, meaning activation of the body's rapid stress and recovery control system, with a drop in heart rate variability rather than a rise. But the mental arithmetic stress task told a genuinely different story. It reliably increased subjective stress without increasing craving at all. In other words, feeling stressed and wanting a drink were not automatically linked in the sample, at least not in the moment the researchers were measuring. Perhaps the most striking pattern emerged when the two tasks were sequenced one after another, when the stress task came first and the alcohol cue came second. Perceived stress actually went down following the alcohol cue exposure, suggesting that for these hazardous drinkers, exposure to alcohol related cues may function as a kind of short term emotional reset even before any alcohol is actually consumed. The researchers also found that heart rate variability changes during the stress task tracked more closely with how stable participants cravings stayed afterward than with how stressed participants said they felt, hinting that the physiological signal was picking up something the self report scales were missing entirely. Clinically, this has real implications for how we think about relapse prevention and craving management. If alcohol cues can function as stress relief independent of craving, then simply instructing a client to avoid triggers may be an incomplete strategy because the cue itself may be serving a soothing function that the client isn't necessarily even framing internally as craving. It also suggests that heart rate variability monitoring done thoughtfully and combined with careful clinical interviewing could eventually help identify moments where a person's physiology is shifting toward vulnerability even when their self report doesn't yet reflect it. For a clinician doing motivational interviewing or relapse prevention work, that's a genuinely useful reframe. The presence or absence of reported stress may simply not be a reliable proxy for craving risk in every client, and this study gives us physiological grounding for why that might be true. It's also worth thinking about this study alongside the broader addiction literature on cue reactivity, which has generally treated craving as the primary outcome of interest in treated stress. As one input among several that can amplify it. What this study contributes is a more precise picture of the traffic between those two lanes, showing that they can diverge sharply depending on the specific trigger involved. A therapist doing exposure based work around alcohol cues might reasonably expect a client's subjective stress to rise alongside craving during a cue exposure exercise. This study suggests that expectation may not hold and that craving can rise even as reported stress stays flatter drops, which has real implications for how we interpret a client's in session presentation during that kind of work. It's also worth noting how this finding might reshape brief motivational interventions delivered outside of formal therapy, for instance in primary care or emergency department settings where a clinician may have only a few minutes to address risky drinking. If alcohol cues can function as a stress reduction strategy independent of craving, then a brief intervention that only addresses craving directly may be leaving an important piece of the puzzle unaddressed since the client may be using alcohol cues and eventually alcohol itself to manage stress in a way that has nothing to do with wanting a drink in the moment. As with any laboratory study, there are real limits on how far we can extend these findings. 56 participants is a workable sample for a counterbalance within subject design where every participant serves as their own comparison, but it's not large enough to draw firm conclusions and about subgroups such as whether men and women or people at different levels of drinking severity respond in the same way. This was a controlled laboratory paradigm and laboratory induced arousal doesn't always map cleanly onto real world cravings that occur at a bar, at a party, or after an especially hard day where context, social pressure and prior alcohol exposure all add layers of complexity a lab can't fully replicate because every participant experience both conditions. We also can't fully rule out some carryover effect between tasks if even with counterbalancing specifically designed to minimize that risk. And this remains a cross sectional single session snapshot. Of reactivity. It can't tell us how these patterns evolve over repeated exposures or whether they predict actual drinking behavior outside the laboratory. The authors were appropriately cautious in how they framed their conclusions, and we should be too. This is an association between physiological and subjective reactivity patterns, not proof that any one pathway causes hazardous drinking to escalate. The takeaway for practitioners is craving and stress may travel together in some moments and move in entirely separate directions in others, and a single self report question may not capture what a person's autonomic nervous system is actually doing underneath the surface. If you work with clients around substance use, this is a good nudge to hold both the subjective and physiological pictures loosely to keep distinguishing are you stressed? From what does this specific cue actually mean for you right now? And to remember that reducing stress alone may not be sufficient to reduce craving in every client you sit with. Our second study comes from the Journal of Affective Disorders and is titled Short Term Heart Rate Variability Dynamics Track Suicide Risk Changes in Depressive in a Retrospective One Week Longitudinal Study. The authors are Yu Wang, Chao Hua Huang, Bao Yizhong, Yang, Liu Haoyu, Yuji Liu, Len Yanggao, Bo Xiang, Ting Ting Wang, and Kei Xiliu. Suicide risk assessment remains one of the most difficult problems in clinical psychiatry, largely because it relies so heavily on what a person is willing and able to to disclose to a clinician. A patient who has decided, for whatever reason, not to talk openly about their risk can present as clinically stable right up until the moment they aren't. And that gap between what's disclosed and what's actually happening internally is one of the most persistent challenges in inpatient psychiatric care. That's part of why there's been sustained interest in finding objective physiological markers that might supplement clinical judgment rather than replace it. And heart rate variability as a window into the balance between the sympathetic and parasympathetic branches of of the autonomic nervous system has become one of the more promising candidates. In that search. This study asked a very practical, very clinically grounded question. Does heart rate variability change in step with suicide risk as it rises and falls over the course of a single hospital admission closely enough that it might eventually serve as a meaningful adjunct signal at the bedside? Wang and colleagues conducted a retrospective analysis, meaning they looked back at existing clinical and physiological data rather than designing an entirely new prospective experiment from scratch. Following 177 inpatients diagnosed with either major depressive disorder or bipolar depression across one week of hospitalization at admission and again after that first week. Clinicians assessed suicide risk using structured clinical tools alongside standard measures of depression severity, anxiety, and sleep quality, while continuous heart rate variability metrics were recorded in parallel, including the standard deviation of the intervals between normal heartbeats, the root mean square of successive differences between heartbeats, high frequency power, and the ratio between low frequency and high frequency power, sometimes used as a rough index of sympathovagal balance, meaning the shifting balance between the body's activating and calming autonomic branches. At the start of the study, patients with moderate to high suicide risk looked physiologically distinct from lower risk patients in several consistent ways. Their heart rates ran faster, their low frequency to high frequency ratio was elevated and their heart rate variability across the standard deviation, root mean square and high frequency measures was suppressed relative to their lower risk peers. That pattern is broadly consistent with the nervous system caught in a more activated, less flexible physiological state. What's genuinely interesting is what happened over the following week. For many patients, those baseline differences narrowed considerably or disappeared entirely as inpatient treatment progressed. And critically, the degree to which a given patient's suicide risk dropped over that week correlated with the degree to which their heart rate slowed and their heart rate variability recovered. And this relationship held up even after the researchers statistically accounted for changes in depression severity, anxiety symptoms and sleep quality. That last point matters enormously because it suggests the heart rate variability signal wasn't simply riding along on general symptom improvement across the board. It appeared to be tracking something more specific to shifting suicide risk itself distinct from the broader clinical picture. For clinicians working in inpatient psychiatric settings, this raises a genuinely exciting possibility an adjunctive, objective, continuously available physiological signal that might help flag when a patient's risk is trending in the wrong direction. Potentially before that shift becomes fully obvious in conversation or in a structured risk assessment interview, it's not hard to imagine this eventually feeding into bedside risk monitoring protocols alongside and never in place of direct clinical assessment, structured safety planning and the ongoing therapeutic relationship that remains central to suicide risk management.
[00:12:52] It's worth situating this study within the longer arc of research trying to find objective biomarkers for suicide risk, assuming that has produced plenty of promising leads over the years and relatively few that have survived rigorous replication. What sets this particular finding apart is the within person design. Rather than simply comparing high risk patients to low risk patients at a single point in time, which is vulnerable to countless confounding differences between the two groups, the researchers track the same patients as their own risk level shifted over the course of a week, which is a considerably stronger form of evidence.
[00:13:23] That doesn't eliminate the need for replication, but it does mean this finding rests on firmer methodological ground than a simple cross sectional comparison would. It's also worth thinking about what a signal like this could realistically add to existing clinical workflows rather than replace them outright. Structured suicide risk assessments already exist and remain the standard of care, but they typically happen at fixed intervals, on admission, at scheduled check ins, or when a clinician has a specific reason for concern. A continuously available physiological signal wouldn't replace those structured assessments, but it could plausibly help identify which patients warrant a closer, more immediate look between those scheduled touchpoints functioning as a kind of early warning layer rather than a standalone diagnostic tool in its own right. We do need to be careful with the caveats here, and there are several worth naming explicitly. This was a retrospective observational study, so what we're describing is an association between changing heart rate variability and changing suicide risk over the course of a week, not a demonstration that one causes the other. A third factor, such as an underlying shift in neurobiological state driven by treatment itself, could plausibly be producing both the physiological change and the risk change in parallel. The sample was drawn from a single hospital system in China, which raises open questions about how well these patterns would generalize to other populations, treatment protocols, and cultural or healthcare system contexts. A one week window is also relatively short in the context of psychiatric recovery. We don't yet know whether this relationship holds over longer hospitalizations during outpatient follow up care or or during the especially high risk period immediately following discharge, which is precisely when many suicide risk models suggest vulnerability can spike. And because suicide risk itself is a notoriously difficult construct to measure reliably even with well validated structured clinical tools, some portion of the apparent tracking we're seeing here could reflect shared measurement noise rather than a genuinely clean physiological signal. Still, this is one of the more compelling associational findings in this space precisely because the effect held up independent of general symptom change across depression, anxiety and sleep. If you work in inpatient, psychiatric or behavioral health settings, this is a study worth watching closely as the underlying research matures, particularly as more prospective multi site replications begin to appear, and as researchers start testing whether the signal holds predictive value looking forward rather than only descriptive value looking backward. Our third study appears in Presidia Computer Science and is titled Postural Impacts based on Heart Rate Variability in Young A Statistical Analysis. The authors are Prashant Kumar, Ashish Kumar Das and Sumanhalder. If you've ever done heart rate variability biofeedback work with a client, you already know intuitively that posture matters, that sitting slumped in a chair feels different in the body than sitting upright, and standing feels different again from either. But intuition and rigorous data are two different things, and this study set out to quantify parameter by parameter exactly how much a simple change in posture shifts the number so many of us rely on every day in clinical and research work. Rather than leaving that relationship as an assumed but unverified detail of protocol design, the researchers recruited young adults ranging from 19 to 32 years old and recorded five minute electrocardiogram segments, meaning direct recordings of the heart's electrical activity in three different postures. From those recordings they extracted the standard battery of heart rate variability parameters. Clinicians and researchers commonly rely time domain measures capturing overall variability across the recording frequency domain measures like low frequency and high frequency power that are often used with appropriate caveats as rough proxies for sympathetic and parasympathetic activity, respectively and nonlinear measures including approximate entropy, which attempts to quantify the unpredictability or complexity of the heartbeat pattern over time rather than simply its magnitude. The three postures compared in the study represent the range most commonly encountered in everyday clinical and research settings, giving the analysis direct relevance to the kinds of protocol decisions practitioners actually have to make rather than testing some more exotic or unusual body position that would have limited real world applicability.
[00:17:02] The headline finding was that posture produced statistically significant shifts across nearly the entire panel of heart rate variability parameters tested. With a single notable exception, approximate entropy did not differ meaningfully across the three postures, suggesting that whatever complexity based signal that particular measure is capturing may be more stable or at least considerably less posture sensitive than the other metrics examined in this study. The researchers also found that short term variability measures, generally understood as more reflective of fast parasympathetically mediated changes and long term variability measures generally understood as more reflective of slower sympathetically influenced changes, were strongly correlated with one another regardless of which posture a person happened to be in at the time of recording. And a commonly used stress index derived from heart rate variability showed a strong negative correlation with heart rate variability overall, again holding steady in direction and strength across all three postures tested even as the absolute values shifted with posture itself. For anyone doing clinical heart rate variability assessment or biofeedback, the practical implication is straightforward but genuinely important. Posture is not a minor detail to standardize away as an afterthought in protocol design. If you're comparing a client's numbers across sessions or comparing a research participant's baseline reading to their post intervention reading and posture wasn't held rigorously constant between those recordings. You may be looking at a posture effect layered on top of or even masking a true change in autonomic function. This study gives us a much more granular parameter by parameter picture of exactly where those posture effects show up most strongly and where, as with approximate entropy, they may be more negligible and therefore more forgiving of minor postural inconsistency. This kind of methodological parameter level study rarely makes headlines, but it does a genuine service to the field by giving practitioners a concrete evidence based reference rather than a vague instruction to simply keep posture consistent. For anyone building a standard operating procedure for heart rate variability assessment in a clinic, gym, or research lab, having actual numbers on which parameters shift the most and which barely move at all turns a soft best practice into a defensible evidence based protocol decision that can be explained clearly to colleagues, supervisors, or an institutional review board reviewing a study design. The limitations here are worth naming clearly this is a cross sectional design conducted in a narrow age band of young, presumably healthy adults, so we cannot say with any confidence how these specific posture effects generalize to older adults, to people with cardiovascular or autonomic conditions, or to clinical populations more broadly where postural response is already known to behave quite differently, sometimes dramatically so, in conditions involving orthostatic dysfunction. Five minute recordings, while standard practice in much of this literature, are also relatively short and some nonlinear and frequency domain measures in particular are known to be sensitive to recording length, which could shape how these specific results would look with longer recordings. And because every comparison here is a snapshot in time rather than a within person comparison of the same measure changing over a longer period or across repeated days, we should describe these findings as an association between posture and heart rate variability parameters at the specific moment of measurement, not a demonstration that posture causes some lasting or cumulative shift in autonomic function over time. The clinical and research takeaway is a simple but consequential procedural one standardize posture, deliberately document it explicitly in every protocol, and treat this study as a useful granular reference table for which specific parameters are most posture sensitive and therefore most in need of careful control, and which, like approximate entropy, may be somewhat more forgiving. Our fourth study was published in Pediatric Research and is titled Maturational Physiology in Preterm Morbidity Impact and Two Year Neurodevelopmental Outcome. The authors are Julia Palladino, Julius Meyer, Marline W. Schenink, Peter Andreessen, Hendrik J. Neimarkt, and Carolla van Poel. Preterm birth remains one of the more urgent and persistent challenges in pediatric medicine, and one of the most difficult clinical questions clinicians face is figuring out as early as possible which infants are at meaningfully elevated risk for long term developmental problems long before those problems become visible in a toddler's behavior or a preschool assessment. Bedside physiological monitoring of the kind that's already happening continuously in every neonatal intensive care unit around the clock, has long been seen as an underused resource for exactly this purpose. Since the raw data is often already being collected for other clinical reasons, the study asked whether the way an infant's physiology matures over their first weeks of life. Not just any single snapshot reading, but the actual shape of that developmental trajectory over time could help identify risk earlier than current clinical practice allows. Palladino and colleagues followed 251infants born before 30 weeks of gestation, tracking heart rate, heart rate variability, respiration, and blood oxygen saturation continuously across a six week window immediately after birth. They then followed those same infants out to 2 years of age to assess motor and cognitive development using standardized developmental assessment tools. Infants were split into two groups for those who experienced no major morbidity during their hospital stay and those who did, meaning significant complications of prematurity such as serious infection, chronic lung disease, or brain injury that are already independently known to elevate developmental risk on their own. Among infants without major morbidity, the researchers identified three distinct, fairly consistent maturation phases across heart rate, heart rate variability, respiration, and oxygen saturation over those six weeks, essentially a predictable physiological growth curve that these healthier infants tended to follow. Infants who experienced major morbidity showed disrupted versions of that same underlying trajectory, with the phases less clearly defined or delayed relative to their peers. And at the two year follow up, those infants had rates of motor and cognitive impairment that were two to three times higher than infants who had not experienced major morbidity. The implication is that the shape of physiological maturation itself over those early weeks weeks, not just the presence or absence of a specific name complication may carry independent prognostic information about a child's developmental future, information that a single diagnosis code alone might not fully capture. This is a meaningful finding for neonatology and for anyone working downstream with children born extremely preterm, including pediatric therapists, developmental specialists, and early intervention providers. If continuous, already collected bedside monitoring data can help flag infants whose physiological maturation is going off track relative to their peers, that opens the door to earlier developmental surveillance, earlier referral to early intervention services, and potentially earlier, more informed family counseling and support, all well before a formal two year developmental assessment would otherwise catch a problem that has already been quietly unfolding. As always, we have to be honest about what a design like this can and cannot tell us. This is an observational longitudinal cohort study. So what we're describing is an association between disrupted physiological maturation and later developmental impairment, not proof that the physiological disruption itself is the direct cause of that later impairment. Both could easily be downstream consequences of whatever underlying complications produce the major morbidity in the first place, with the physiological signal simply serving as an early marker rather than a causal mechanism. 251 infants followed over 2 full years with the inevitable dropout and missed follow up visits that entails is a solid but not enormous sample for teasing apart the many overlapping factors that shape neurodevelopment and in extremely preterm infants, including factors like socioeconomic context and family environment that this study wasn't necessarily designed to fully capture. And the six week monitoring window, while clinically practical and aligned with typical hospital stay lengths, is only a narrow slice of a child's early development. We don't yet know whether extending that monitoring window further would sharpen the predicted picture, complicate it, or reveal additional maturation phases beyond the three identified here. It's also worth noting how this kind of trajectory based analysis is differs from the way physiological monitoring is often used at the bedside today, which tends to focus heavily on single moment alarms and thresholds, flagging a value the incident crosses a predetermined line. What this study points toward is a complementary approach looking at the overall shape of an infant's physiological development across weeks, not just individual moments where a number briefly spikes or dips. That shift in framing from threshold based alerting to trajectory based pattern recognition is likely to become more feasible as neonatal units increasingly adopt more sophisticated longitudinal data infrastructure. The takeaway for anyone in neonatal or pediatric care is that physiological trajectories, not just isolated readings taken at any single point in time, deserve more clinical attention as a potential early signal of developmental risk. This is exactly the kind of study that argues for smarter, more longitudinal use of data. Hospitals are often already collecting as a matter of routine care, rather than requiring entirely new monitoring infrastructure or additional burden on already stretched neonatal intensive care staff. Now let's take a quick break to hear from the sponsor of this show. Optimal hrv. Optimal HRV was built for exactly the kind of work we talk about on this podcast every week, helping clinicians, coaches and everyday users actually put heart rate variability science into practice in a sustainable, everyday way. The Optimal HRV app centers on a simple morning measurement protocol so you get a consistent, comparable reading first thing in the day before the noise of daily life, caffeine, movement and stress start to shift your numbers around. From there, Optimal HRV builds longitudinal tracking over time, so instead of chasing a single reading in isolation, you can start to see your own trends, your own patterns and your own baseline shifting in response to training, load, stress, sleep quality or clinical intervention. And built directly into the app are biofeedback tools designed to help you actually train your physiology in the moment, not just passively observe it after the fact. For clinicians listening today, that combination of consistent morning measurement, longitudinal tracking, and in app biofeedback tools maps directly onto several of the studies we covered this episode. If posture and measurement timing can meaningfully shift the numbers, as we heard earlier, then a standardized morning protocol is exactly the kind of consistency that makes a client's or patient's data genuinely comparable across weeks and months, rather than muddied by whatever position or time of day a reading happened to be taken.
[00:26:20] Optimal HRV is also proud to support two continuing education opportunities for professionals working in this space. The first is Dr. Enakazin's heart rate variability Biofeedback Training, a Biofeedback Certification International alliance aligned program offering 16American Psychological association continuing education credits built for clinicians who want a rigorous applied foundation in heart rate variability biofeedback practice. The second is Dr. Donald Moss's course on ethical Principles and Practice Standards in Clinical Biofeedback. Also Biofeedback Certification International Alliance Aligned Offering 3American Psychological Association Continuing Education Credits, a resource specifically designed for practitioners who want to make sure their clinical biofeedback work is grounded and sound and defensible ethical practice. Full registration details and links for both trainings are available in the show Notes for this episode. Our fifth study was published in Environmental and Occupational Health Practice and is titled Development of Easy to Notice Nurse Call with Low A Pilot Study Using Heart Rate Variability. The authors are Mako Katagiri, Masao Hira and Yoshiaki Sakurai. This next one is a smaller, more applied study, but it tackles a problem that will feel immediately familiar to anyone who has spent real time on a hospital ward. Alarm fatigue and alarm Related Occupational stress Nurse call systems have to solve two competing demands simultaneously, and that tension is harder to resolve than it might first appear. The sound has to be noticeable enough to reliably grab attention, often in a noisy, high stakes clinical environment full of competing alerts and conversations. But a sound engineered purely to maximize noticeability is often also maximally jarring, and repeated exposure to a jarring alert dozens or even hundreds of times across a single shift can itself become a meaningful cumulative occupational stressor or over the course of a career. This study set out to see whether it's actually possible to engineer a nurse call tone that threads that needle successfully, rather than simply accepting the trade off as unavoidable. Categiri and colleagues tested nine different candidate sounds with 10 nurses using heart rate variability as an objective physiological measure of each sound stress impact. Specifically drawing on a cardiac sympathetic index, a heart rate variability derived measure intended to capture activation of the sympathetic branch of the autonomic nervous system system, the branch broadly associated with alertness, arousal and the body's stress response. Each nurse's physiological reaction to each candidate's sound was measured directly and repeatedly rather than relying solely on subjective preference ratings collected afterward, which allowed the researchers to capture reactions that a nurse might not consciously report or even notice in the moment. Testing nine separate candidate sounds against a physiological outcome rather than relying on a smaller handful of obvious options also reflects a genuinely systematic approach to alarm design, one that treats sound engineering as an empirical question to be tested rather than a matter of institutional habit or simple convention carried over from decades old equipment. The traditional tremolo tone, the classic familiar nurse call sound that most wards already use as their default, came out as the most noticeable of the nine candidates tested, but it also produced the highest cardiac sympathetic index of any sound in the study, meaning it was physiologically the most stress inducing option available. A newly designed alternative built from two combined musical tones referred to in the study as the C sound, performed nearly as well on noticeability, but triggered a meaningfully smaller phys physiological stress response than the tremolo tone. The researchers then took the Csound into a follow up test on an actual hospital ward under real working conditions rather than a controlled laboratory setting, and the results there supported it as a genuinely promising, less stressful alternative to the traditional tremolo tone in real clinical practice, not just under artificial testing conditions. This has a very direct practical application for hospital administrators, biomedical engineers and nursing leadership alike. Alarm design isn't just an engineering afterthought bolted onto a device late in its development. It's a genuine occupational health variable in its own right, and heart rate variability offers a real objective tool for evaluating alert design choices before they're rolled out worldwide to hundreds of staff. Chronic cumulative exposure to high stress alert tones repeated many times over a single shift and across an entire nursing career is exactly the kind of slow burn occupational stressor that heart rate variability research is well positioned to help quantify and ideally help meaningfully reduce through smarter design choices, it's worth remembering that nursing burnout and turnover are already major concerns across the hospital sector. And while alarm sound design is obviously just one small piece of that much larger picture, it's a piece that's entirely within an institution's control. Unlike staffing ratios or patient acuity, which are much harder levers to pull, a relatively low cost intervention like swapping a default alert tone for a less physiologically taxing alternative is the kind of change that could plausibly be implemented hospital wide without requiring new hardware, new staff or new workflow, which makes this line of research more actionable than it might first appear. The limitations here are real and worth naming. Plainly, given how small this study was in scale, 10 nurses is a genuinely small sample, entirely appropriate for an early stage pilot study designed to generate a promising signal, but nowhere near large enough to generalize confidently across different hospital units, shift patterns, nursing specialties, or broader nursing populations with different baseline stress levels. This was also, by explicit design, a pilot study, meaning its purpose was to identify a promising direction worth pursuing further, not to provide definitive practice changing evidence sufficient on its own to justify a full scale hospital rollout. We also don't yet know how the C sound performs over longer exposure. Whether nurses habituate to it differently than they habituate to the tremolo tone over weeks or months of daily repeated use is an open question, and that kind of habituation effect over time could meaningfully change the overall calculus and in either direction. The takeaway is an encouraging one though. This is a nice concrete example of heart rate variability being used not just to study a health outcome after the fact, but to actively engineer a better, less stressful working environment for the people providing frontline patient care every day. It's exactly the kind of applied, occupational health oriented research this show likes to highlight when we come across it. Our sixth study appears in Archives of Women's Mental Health and is titled Wearable Measured Heart Rate Variability and Premenstrual Disorder Symptoms across Menstrual Cycle. The authors are Qing Pan, Jing Zhou Min, Chen Peijie, Zhang Xin Yishu Yifei, Lin Jinhuang Yu, Chen Li and Dong Hao Lu. Premenstrual disorders sit at an uncomfortable intersection in medicine. The symptoms are real, often genuinely disabling, and yet they've historically been notoriously difficult to track objectively, since so much of what clinicians and researchers currently rely on is retrospective self report filled in well after the fact and vulnerable to all the usual distortions of memory and recall bias. Wearable devices offer an intriguing way around that fundamental problem, since they can passively and continuously collect physiological data across an entire menstrual cycle without requiring the person wearing them to stop, remember and log anything by hand. This study followed 193 women, 68 of whom had a confirmed premenstrual disorder diagnosis across 293 menstrual cycles total, using Huawei fitness trackers to continuously capture heart rate variability metrics and including the standard deviation of normal heartbeat intervals, the root mean square of successive differences, and high frequency power throughout each individual cycle rather than at a single isolated point. Across the entire sample, heart rate variability metrics followed a remarkably consistent pattern. They dropped in the days leading up to menses and rose again in the days afterward. That basic cyclical rhythm showed up in essentially everyone studied whether or not they had a diagnosed premenstrual disorder, but the genuinely interesting divergence appeared specifically in the rebound afterward. Women without a premenstrual disorder showed a noticeably more pronounced recovery in their heart rate variability after menses than women with a diagnosed premenstrual disorder did. And within the group of women who did have a premenstrual disorder, specifically, lower heart rate variability in the week before and the week after menses was associated with worse symptom severity overall, a relationship that simply wasn't present at all in women without a premenstrual disorder. That combination of findings is genuinely useful from a clinical standpoint. It suggests that premenstrual disorders may involve not just the well established hormonal shifts and that occur across the cycle, which have long been the primary focus of research and treatment, but also a blunted autonomic recovery response specifically around menses, which could eventually help explain why some people experience such pronounced disabling symptoms while others, navigating the exact same hormonal swings, do not experience nearly the same degree of impairment. One plausible way to think about this is that the hormonal fluctuations of the menstrual cycle may act as a kind of physiological stress test each month. And while most women's autonomic nervous systems flex and recover from that test without much difficulty, some women's systems appear to struggle specifically with their recovery phase. And it's that struggling recovery, more than the initial dip itself that seems to track most closely with symptom severity. It also opens a promising door to wearable heart rate variability tracking as a genuinely useful clinical adjunct for identifying and monitoring premenstrual disorder symptom severity over time, potentially catching meaningful cycle to cycle changes that an infrequent monthly clinical check in visit would simply miss entirely. This also fits into a broader pattern we see across women's health research more generally, where conditions tied to the menstrual cycle have historically been underfunded and under researched relative to their prevalence and impact, and where objective physiological measurement tools have often lagged behind what's available in other areas of medicine. A finding like this, built on wearable devices that many people already own and use daily, represents a genuinely low barrier path toward closing some of that research and monitoring gap without requiring patients to purchase specialized new equipment or attend additional in person appointments purely for data collection. We do need to hold this as an association rather than a causal story, and that framing matters here.
[00:35:38] Specifically, this is an observational cohort study, so we cannot conclude that blunted heart rate variability recovery causes worse premenstrual symptoms or conversely, that worse symptoms themselves cause the blunted recovery. The two could easily be driven by a shared underlying hormonal or autonomic mechanism that this particular study design wasn't built to isolate or disentangle. Consumer wearable devices, while increasingly sophisticated, also don't offer the same measurement fidelity as clinical grade electrocardiogram recordings, and heart rate variability metrics derived from optical wearable sensors can be more vulnerable to motion artifact and general signal noise than gold standard clinical measurement tools. And while nearly 200 women tracked across close to 300 menstrual cycles is a genuinely strong sample size for this kind of naturalistic longitudinal tracking study. Premenstrual disorder presentations vary enormously from person to person, and this study, built on group level averages, can't tell us with confidence how well these broader patterns apply to any single individual patient sitting in front of a clinician. For clinicians working with patients who have premenstrual disorders, the practical takeaway is that consumer grade wearable heart rate variability tracking may be a genuinely useful low burden adjunct tool for monitoring symptom severity across the cycle over time, particularly around menses itself, where this study suggests the most clinically meaningful physiological signal actually lives. Our seventh study was published in Big Data and Cognitive Computing and is titled Seasonal Variation in Heart Rate Variability Associated with Physical Activity and Regional Variability Observed in the All Star Holter Electrocardiogram Database. The authors are Yutaka Yoshida and Junichiro Hayano. This next study operates at a completely different scale than most of what we've covered so far today.
[00:37:11] Rather than a few dozen or a few hundred participants, Yoshida and Hayano drew on a truly massive database of over 130,000 24 hour Holter electrocardiogram and accelerometer recordings collected across eight distinct regions of Japan. A Holter recording, for anyone unfamiliar with the term, is a small portable device that continuously records the heart's electrical activity, typically worn for a full 24 hour day, allowing researchers to capture heart rate variability across an entire day and night cycle rather than relying on a single brief snapshot recording. The core question here was whether seasonal changes in heart rate variability, something that's been observed in prior research actually track along with seasonal changes in physical activity levels and whether that underlying relationship looks the same across genuinely different regions of the country. What the researchers found was more selective and more interesting than a simple More activity in summer means more heart rate variability.
[00:38:02] The link between physical activity and seasonal heart rate variability changes was specific to certain heart rate variability and indices, rather than showing up uniformly across every single metric tested. And that relationship also differed meaningfully between the period before the COVID 19 pandemic and the period during it, suggesting that broad society wide shifts in population level behavior and daily routine can meaningfully reshape this seasonal pattern in ways that are worth taking seriously. Just as interesting was what the researchers found when they turned their attention to regional differences across the country.
[00:38:33] Rather than being explained mainly by each region's differing underlying sensitivity to physical activity, regional variation in heart rate variability was driven mostly by what the researchers described as an unexplained residual component, essentially a leftover currently unaccounted for source of regional difference that current statistical models of activity and seasons simply don't capture. Hokuriku, a region on Japan's western coast known for its distinct climate, showed the highest residual variability of the eight regions studied, while Minami Kanto, the Greater Tokyo metropolitan region region, showed the lowest. It's worth pausing on just how unusual it is to have enough data to even ask a question at this level of geographic resolution. Most heart rate variability research simply cannot support meaningful regional comparisons within a single country, since doing so requires both a very large sample and a measurement protocol standardized well enough across sites to make those regional comparisons trustworthy in the first place. For researchers and public health practitioners, this is a useful and somewhat humbling reminder that population level heart rate variability patterns are shaped by a genuinely tangled web of overlapping seasonal, behavioral, and regional factors that don't reduce neatly down to any single explanatory variable like activity level alone. However tempting that simpler story might be, it also underscores how disruptive society wide events like a pandemic can ripple through physiological patterns in ways that are easy to miss entirely unless researchers are working with a data set large enough and collected over a long enough period to actually capture that kind of shift as it happens. This is also a good example of how database studies of this scale can serve a different but equally valuable purpose than smaller, mechanistic studies. A study of 50 or 100 participants can tell us a great deal about the specific physiological pathway connecting a stimulus to a heart rate variability response, but it can tell us whether that pathway actually plays out consistently across an entire national population, through changing seasons, across regional climate differences, and through a once in a generation disruption like a global pandemic. That's precisely the kind of question this data set, and really only a data set of this scale is equipped to answer. Given the sheer scale of this data set, our usual concerns about small sample size obviously don't apply here in the way they do for many of today's other studies, but different limitations take their place. Instead, this is a retrospective observational analysis of existing database records. So what we're describing is an association between activity season, region and heart rate variability patterns at the population level, not a controlled experimental demonstration of what's actually causing any individual person's seasonal variability to shift up or down. A data set this size, for all its statistical power, also can't fully capture individual level context things like a person's underlying health conditions, medications, occupation or personal daily routines, any of which might help explain some of that unexplained residual regional variability the researchers identified but couldn't fully account for. And because this data comes specifically from Japan, with its own particular climate patterns, cultural activity norms, and distinct pandemic experience, we should be cautious about assuming these exact seasonal and regional patterns would generalize directly and cleanly to other countries with meaningfully different climates, activity norms, and pandemic trajectories. The broader takeaway is this seasonal variation in heart rate variability is genuinely real and measurable at a population level, but it's not simply a proxy for how physically active a population happens to be in any given month. There's meaningful regional texture underneath that broad seasonal pattern that current explanatory models don't yet fully capture. Which is exactly the kind of open, unresolved question that large population level data sets like this one are best positioned to keep chipping away at over time. Our eighth and final study appears in Revista Brasileira de Sine Anthropometria e desimpeno Humano and is titled Analysis of Heart Rate Variability and Submaximal Exercise Exercise in People with Obstructive Sleep Apnea. The authors are Sheila Souza Defreitas Raisa, Helena Rodriguez Machado, Leonardo Brynaramos d', Souza, and Laura Maria Tomasi Neves. We'll close today's episode with a study addressing a genuinely practical, everyday clinical question. How does the autonomic nervous system of someone living with obstructive sleep apnea, a condition already well established in the literature, to disrupt normal cardiovascular and autonomic regulation, actually respond to a bout of moderate exercise, and just as importantly, how quickly does that system recover once the exercise ends? Susan Defraitis and colleagues studied 30 adults diagnosed with obstructive sleep apnea having each participant complete a six minute step test, a widely used, low cost and accessible submaximal exercise protocol that doesn't require the specialized laboratory equipment of a full cardiopulmonary exercise test, making it far more practical for routine clinical use. The researchers measured heart rate variability continuously across three distinct points in the at rest before the test began, during the exercise itself, and again shortly afterward during the recovery period, tracking both time domain measures, including the standard deviation of normal heartbeat intervals, the root mean square of successive differences, and a triangular interpolation index alongside frequency domain measures, specifically low frequency and high frequency power. Choosing a submaximal protocol rather than pushing participants to a maximal exertion test was itself a deliberate and clinically sensible choice, since submaximal testing is both safer and more representative of the kind of exercise exercise intensity a clinician might realistically prescribe for a patient managing a chronic respiratory condition alongside everyday life. During the exercise itself, heart rate variability dropped across all of these measures, which is exactly what we'd expect physiologically reflecting a natural shift toward greater sympathetic activating autonomic influence and reduced parasympathetic calming influence as the body works to meet the increased metabolic demands of physical movement. What's clinically reassuring and arguably the most important finding in this study, is the of one what happened next? Heart rate variability values returned as something close to resting baseline levels fairly quickly after the exercise ended, suggesting that despite the autonomic disruption we already know tends to accompany obstructive sleep apnea, particularly during sleep itself. These participants showed relatively fast, largely intact autonomic recovery capacity following a single bout of daytime moderate intensity exercise. That's a genuinely useful and somewhat reassuring finding for clinicians and exercise professionals working with this specific population.
[00:44:25] Obstructive sleep apnea is already independently linked to elevated cardiovascular risk in the broader medical literature, and there's understandable clinical caution among practitioners about recommending exercise programs without a clear evidence based sense of how a given patient's autonomic system will handle and recover from that added physical stress. This study offers early encouraging evidence that submaximal moderate intensity exercise of the kind represented well by a simple accessible step test doesn't appear to meaningfully overwhelm automatically recovery capacity in this population, at least not within this relatively short window immediately following exercise, which is a reasonably encouraging signal for exercise professionals building structured conditioning programs for patients living with this condition. It's worth connecting this back to something we touched on earlier in the episode in the posture study the value of standardized low tech assessment protocols that can be deployed broadly in real clinical settings without specialized equipment. A six minute step test is exactly that kind of tool, and pairing it with heart rate variability measurement rather than relying on heart rate alone gives clinicians a more textured picture of autonomic recovery than a simple return to resting heart rate would provide on its own. For a sleep medicine clinic without ready access to a full exercise physiology laboratory, that combination is genuinely practical. The limitations here are worth stating clearly before we lean too hard on that reassurance. In clinical practice, 30 participants is a modest sample sufficient for detecting the general pattern described here, but nowhere near enough to confidently break results results down further by disease severity, age, sex, or the many other individual factors that likely matter for how well any given patient actually tolerates and recovers from structured exercise. It's also worth noting that a six minute step test is exactly the kind of low cost, easily repeatable protocol that could realistically be built into routine follow up visits for patients with obstructive sleep apnea rather than remaining confined to specialized research settings. Studies like this one are part of a broader push to make better use of simple and accessible assessment tools that clinical teams can actually deploy in everyday practice rather than requiring specialized exercise physiology equipment that most sleep medicine clinics simply don't have on hand. This was also a single bout of submaximal exercise measured over a fairly short recovery window. We don't yet know how these autonomic recovery patterns would hold up across a longer sustained training program, at more vigorous exercise intensities than a simple step test or over a longer post exercise observation period extending well beyond the immediate recovery window study.
[00:46:34] And because there was no comparison group of participants without obstructive sleep apnea running through the identical protocol alongside them, we can't say with full certainty how much of this observed recovery pattern is genuinely specific to people with this condition versus simply reflecting normal expected autonomic recovery physiology in the general population. More broadly, for exercise physiologists, sleep medicine clinicians, and rehabilitation professionals, the practical takeaway is encouraging but appropriately provisional. Given these limitations, moderate submaximal exercise appears to be well tolerated autonomically in this population in the short term, which supports continued clinical interest in structured supervised exercise programs for people living with obstructive sleep apnea. While further research with larger samples, comparison groups, and longer follow up continues to fill in the rest of the picture that brings us through all eight studies for this episode, and a few threads are worth deliberately pulling together before we close things out. Several of today's studies remind us that heart rate variability rarely tells a simple single direction story, and that's a theme worth sitting with rather than rushing past.
[00:47:30] In our alcohol cue study, physiological arousal and subjective craving moved independently of each other rather than in lockstep, defying the intuitive assumption that stress and craving are simply the same experience described two different ways. In our suicide risk study, the physiological signal tracked something more specific than general symptom improvement across depression, anxiety, and sleep combined. And in our premenstrual disorder study, the same basic cyclical dip and rise showed up in essentially everyone studied, while the clinically meaningful difference lived specifically in how well the body recovered afterward, not in the initial drop itself. Across all of these findings, recovery, adaptability, and the overall shape of a physiological trajectory over time keep showing up as more genuinely informative than any single isolated reading taken in isolation. We also saw across today's episode just how many different clinical and research contexts heart rate variability work is currently being applied to addiction and craving, psychiatric risk assessment, basic measurement, standardization, neonatal development, occupational health and alarm design, women's mental health, population level, seasonal physiology, and exercise tolerance in a chronic respiratory condition. That range is genuinely exciting to see laid out in a single episode, and it's also a good grounding reminder to stay appropriately humble about what any single study's conclusions can and cannot support. Several of the studies we covered today were observational, retrospective, or cross sectional in design, meaning what they show us is association, not causation. And that distinction matters just as much for how we talk about this research with colleagues, clients and patients, as it does for how we interpret it quietly to ourselves. It's also worth noting how differently heart rate variability was measured across today's studies, because that variation is itself part of the story of where this field currently stands. We heard about laboratory electrocardiogram recordings and a counterbalanced stress protocol, continuous inpatient monitoring tracked across a full week, brief five minute postural recordings, weeks of continuous neonatal intensive care monitoring, heart rate variability captured around a specific occupational stimulus, consumer wearable tracking across an entire menstrual cycle, 24 hour holder recordings pulled from a database of well over 100,000 people and heart rate variability captured before, during and after a brief exercise test. Each of those measurement approaches carries its own strengths and its own blind spots. And part of becoming a sophisticated consumer of this literature is learning to ask every time not just what a study found, but how the underlying signal was actually captured over what time frame and with what degree of measurement precision. There's also a quieter thread running through several of today's studies about who gets studied and in what context. We heard about young healthy adults in a laboratory setting, hospitalized psychiatric patients, extremely preterm infants, hospital nursing staff, women tracked across their reproductive years, an entire national population sampled through routine cardiology monitoring, and adults living with a specific chronic respiratory condition or heart rate variability. Research increasingly spans this full range of human contexts, from the healthiest possible participants to some of the most clinically vulnerable. And that breadth is exactly what allows a show like this one to speak meaningfully to clinicians, researchers, coaches and practitioners across such different corners of the field, all in a single week's worth of reading. If there's one closing thought to carry out of this episode, it's that heart rate variability continues to earn its place as a genuinely versatile physiological signal you useful for understanding craving and stress tracking psychiatric risk standardizing basic measurement practice monitoring the earliest weeks of a preterm infant's life engineering better hospital environments understanding women's health across the menstrual cycle mapping population level, seasonal physiology Engaging exercise tolerance in a chronic condition no single study we covered today closes the book on any of these questions, and none of them were meant to. Each one simply adds one more careful peer reviewed data point to a much larger ongoing conversation and that steady, incremental accumulation of evidence is exactly what good science is supposed to look like. That's also, in a way, the whole premise of this show. We're not here to hand you a single miracle finding each week, and real research almost never works that way. We're here to walk through the evidence carefully, name what it does and doesn't tell us, and trust that clinicians, researchers, coaches and practitioners listening are capable of holding that nuance and putting it to good use in their own work. Thank you for spending this time with us today. As always, you'll find full study details and links in the show notes for this episode, along with more information about our sponsor, Optimal HRV and the two continuing education trainings we mentioned earlier in the show. Until next week, keep measuring, keep questioning and keep learning. This has been this week in heart rate variability.
