Episode Transcript
[00:00:00] Welcome back to this Week in Heart Rate Variability. I'm Matt Bennett and the show where we go through the latest peer reviewed HRV research every single week, reading the papers, unpacking the methods, interrogating the findings and asking what all of it means for the clinicians, coaches, researchers and practitioners who are actually trying to apply this science. Before we get into today's studies, the standard reminder is that everything you hear on this podcast is strictly for educational and informational purposes. Nothing we discuss constitutes medical advice, clinical guidance or a recommendation to change any treatment protocol.
[00:00:29] The studies we cover represent the author's findings and interpretations. Science moves through replication, debate and revision, and a single paper, however well designed, is never the final word. I always encourage you to go read the original papers yourself, follow the citations and form your own views. Links to everything we discuss are in the show Notes with that said, let me tell you what we're covering this week because it is a genuinely rich lineup. We open with a controlled laboratory study from the German Sport University Cologne that takes a systematic look at something that sounds straightforward but turns out to be surprisingly complex.
[00:01:01] What actually happens to your heart when you breathe fast? On purpose, the researchers tested four different breathing frequencies, including two that qualify as fast paced and simultaneously measured both HRV and cardiac contractility. The results paint a picture of frequency dependent cardiac responses that challenge some assumptions practitioners may have about how fast paced breathing works. Study two is a PRISMA guided systematic review that maps the triangular relationship between gut microbiota dysbiosis, the autonomic nervous system and immune function in the context of infectious diseases. This is a paper that sits at the intersection of several fast moving fields and it asks a question that is directly relevant to understanding why HRV drops during infection and what that drop actually represents biologically. Third, we look at a study that brings HRV into the driver's seat. Literally. A research team has developed a system using capacitive electrocardiography electrodes embedded in a car seat backrest that can continuously record HRV during driving and and they combine that with machine learning and derivative based signal analysis to detect drowsiness before behavioral signs appear. The detection lead time, they report, is meaningful enough to potentially save lives. Fourth, we step into the neurointensive care unit for a study examining what happens to hrv, baroreflex sensitivity and signal complexity in patients recovering from one of the most devastating strokes you can have malignant middle cerebral artery infarction requiring decompressive cranectomy. This paper asks not just whether autonomic measures are abnormal in this population, but which ones actually predict who recovers and who doesn't? And and we close with something different in flavor. A single case report about an 84 year old man whose post meal dizziness and near fainting episodes were treated with osteopathic manipulative therapy targeting the vagus nerve. It's a paper that cannot establish efficacy on its own, but it raises interesting questions about autonomic modulation as a therapeutic target and uses HRV to track the autonomic response, making the clinical narrative more precise. Our first study was published in Psychophysiology and is titled Specifying the Cardiorespiratory Patterns during Fast Paced Breathing. The authors are Masha Iskra, Sylvain Lebon, Tasha Papa, Katerina Salvodi, Eliza Vinen, Marcus Robb and Laura Voigt. Let's start with the core problem this paper is trying to solve because it's one of those problems that becomes more interesting the longer you stare at it Fast Paced Breathing Breathing intentionally at frequencies above 20 cycles per minute has been gaining traction as a performance preparation tool. Studies have found that it can reduce reaction time and movement time in motor tasks, increase grip strength and enhance perceived alertness, energy and focus.
[00:03:26] Athletes and coaches in high performance environments have begun using it as part of pre competition activation routines. People working in tactical and operational settings have explored it as a tool for mental readiness. The interest is real, the reported performance benefits appear genuine. And yet, as Isker and colleagues point out with admirable directness, the physiological mechanisms behind those effects remain in their framing, largely speculative. Think about what that means for a moment. We have an intervention that people are actively using to prepare their bodies and minds for demanding performance, but we do not have a clear account of what that intervention is actually doing to the cardiovascular system. That is a significant gap and it matters not just theoretically but practically, because if we do not understand the physiological machinery that fast paced breathing operates on, we cannot make principled decisions about frequency, duration, timing or population specific application. The comparison with slow paced breathing is instructive here. Six cycles per minute, the resonance frequency range extensively studied in HRV biofeedback research has a well characterized physiological profile. HRV rises substantially, respiratory sinus arrhythmia strengthens, vagal tone increases and the heart and respiratory systems enter a state of strong phase synchronization. We can describe what that state looks like, how it feels and which functions it likely supports. For fast paced breathing we have no comparable level of characterization. We know performance outcomes occur, but we do not have a clear picture of the autonomic cardiac signature that accompanies them. To understand why characterizing that signature is theoretically important, you need to know about a debate that has been running in the autonomic psychophysiology field for some time. A debate about what changes in respiratory sinus arrhythmia actually tell you when the breathing pattern is deliberately manipulated. Two distinct positions have developed. The phasic position holds that changes in the magnitude of respiratory sinus arrhythmia primarily reflect differences in the phasic moment to moment patterning of vagal bursting activity during the breath cycle. When you breathe slowly, the long expiratory phase creates a large window for vagal bursting, producing high hrv. When you breathe fast, that window shrinks and HRV falls. But according to this view, the mean level of cardiac vagal discharge has not actually changed, only its patterning has. The tonic position, by contrast, argues that voluntary breathing manipulation genuinely alters overall cardiac vagal activity through pathways including vagal afferent signaling and effects on brainstem nuclei, producing real changes in tonic autonomic state that are not merely a phasic redistribution. Both positions agree that respiratory sinus arrhythmia is predominantly vagally mediated. They disagree about what a change in it tells you about the underlying autonomic state. This matters enormously for anyone using HRV to track how a breathing intervention affects physiology, and it cannot be resolved without pharmacological blockade studies, which are beyond the scope of most lab experiments. The present study does not resolve this debate. It acknowledges that up front, but it navigates it carefully, measuring cardiac parameters that speak to both positions and interpreting the findings with appropriate nuance. The second index the team used alongside RMSSD was the pre ejection period. This measure, derived from impedance cardiography, represents the interval between the onset of ventricular electrical activation and the mechanical opening of the aortic valve, a window that reflects cardiac sympathetic beta adrenergic activity and myocardial contractility. A shorter pre ejection period indicates faster, more forceful cardiac contraction, suggesting elevated sympathetic drive. A longer pre ejection period indicates the opposite. The combination of RMSSD and pre ejection period gives you a two dimensional map of the autonomic cardiac state. You can see whether the parasympathetic system is withdrawing, whether the sympathetic system is activating, or both simultaneously. That simultaneous picture, what the literature calls reciprocal activation, is the expected cardiac signature of a genuine sympathovagal shift toward activation. Prior research on stress and exercise had established that different demands produced different patterns in this two dimensional space. Fast paced breathing had been proposed to produce reciprocal activation, but this had never been rigorously tested across a range of frequencies. The investigators recruited physically active adults between 18 and 30 years old from a local university defining physical activity as at least two hours per week. The final sample analyzed for the main hypotheses comprised 38 participants. Sample size was determined to prior using power analysis based on the lower bound of the effect size from a comparable published study with statistical power set at 0.95, a notably stringent standard. The within subject design ensured that every participant completed every condition, maximizing sensitivity to individual level effects and eliminating between person variants as a confound. The five experimental conditions were spontaneous breathing at each participant's natural resting rate, slow paced breathing at 6 cycles per minute, paced breathing at 15 cycles per minute serving as a within spontaneous range control, fast paced breathing at 35 cycles per minute and fast paced breathing at 55 cycles per minute. The selection of 35 and 55 as the two fast paced frequencies was not arbitrary. It came from a preliminary pilot study the team conducted specifically to map how RMSSD changed across a wider range of frequencies, allowing them to select two points that would capture meaningfully different levels of stimulation while remaining feasible for this participant population. Participants followed a visual pacer on screen, a ball moving upward for inspiration and downward for expiration, with each condition lasting two and a half minutes. The order of the four paced conditions was randomized and counterbalanced across participants between each condition. A five minute washout period allowed physiology to return toward baseline. Critically, participants were instructed to breathe at shallow depth throughout all conditions. This was deliberate. It distinguished fast paced breathing from hyperventilation which requires not just a higher frequency but also increased depth. Hyperventilation produces dangerous drops in carbon dioxide and a rise in blood ph, leading to dizziness, tingling and anxiety. By keeping depth shallow, the team ensured they were studying a form of fast paced breathing that could realistically be implemented in practical settings without adverse effects. The inhalation to exhalation ratio was held constant at 1, 1 across all conditions to avoid confounding effects of asymmetric breath phases. Cardiac signals were recorded using electrocardiography with a two lead setup and impedance cardiography with eight electrodes placed at the neck and thorax.
[00:09:07] RMSSD was extracted from the last two minutes of each condition's recording window. Impedance cardiography is technically demanding and the authors confronted its most significant challenge head on. The waveform morphology used to identify the pre ejection period varies substantially both between individuals and within the same individual over time, which causes standard automated algorithms to fail at an unacceptable rate. The team's solution was to develop a novel ensemble averaging technique that averaged cardiac cycles to produce a stable composite waveform, then applied adaptive detection balance to reliably identify key waveform landmarks across diverse morphologies. This methodological contribution is not peripheral to the study it is what made a valid measurement of the pre ejection period possible across the sample. The analysis also quantified the phase locking value between respiratory and cardiac signals as an exploratory measure of cardiorespiratory coupling and and examine how quickly RMSSD and pre ejection period returned to baseline during the washout periods. Subjective breathing discomfort was rated on a 10 point scale after each condition. One technical issue the team addresses transparently is cardiac aliasing, a potential signal distortion that arises when the breathing rate exceeds half the heart rate during fast paced breathing, which can create apparent changes in interbeat interval patterns that do not reflect true autonomic variation. The authors provide supplementary estimates of the potential magnitude of aliasing at both fast paced breathing frequencies and acknowledge that its impact on RMSSD and the time domain is not fully characterized in the literature. This kind of transparency about methodological limitations strengthens rather than weakens the paper's credibility. It is worth noting that 17 participants were excluded from the main pre ejection period analysis due to movement artifacts in the impedance cardiography signal, resulting in 38 valid data sets. The exploratory analyses of washout periods and cardiorespiratory coupling were conducted on subsets of 31 and 32 participants respectively. For the same reason. The authors acknowledge that the exploratory analyses are somewhat underpowered and framed them appropriately as preliminary. The central result is clean and consequential Cardiac changes during fast paced breathing are frequency dependent and the two fast paced breathing conditions produce distinctly different physiological profiles. At 35 cycles per minute, RMSSD decreased significantly compared to the paced breathing control at 15 cycles per minute, this means HRV fell. The parasympathetic modulation of heart rate was reduced. However, the pre ejection period did not change significantly at this frequency, meaning that cardiac contractility was not elevated and sympathetic beta adrenergic drive was not clearly increased. At 55 cycles per minute, the picture was different on both measures. RMSSD fell even further significantly more than at 35 cycles per minute and the pre ejection period also decreased significantly compared to baseline, indicating a genuine increase in cardiac contractility consistent with sympathetic activation. This is the reciprocally coupled response the researchers had hypothesized simultaneous parasympathetic withdrawal and sympathetic cardiac activation, mapping these conditions in the two dimensional autonomic space defined by RMSSD change and pre ejection period change. The mean vectors for all breathing conditions pointed toward the lower left quadrant representing reciprocal activation, but only the 55 cycles per minute condition achieved statistical significance for both measures simultaneously. The 35 cycles per minute condition produced a descriptively similar but statistically incomplete version of that response, with only the RMSSD reduction reaching significance. In practical terms, this means that 35 cycles per minute may primarily reduce parasympathetic modulation without reliably engaging the sympathetic patient branch of the cardiac response, whereas 55 cycles per minute does both. The threshold for engaging the full reciprocal sympathovagal activation pattern lies somewhere between these two frequencies, though the study cannot pinpoint the exact location heart period Menuna the raw interbeat interval reflecting overall heart rate did not significantly change at the mean level in either fast paced breathing condition, which might seem surprising given that a reciprocal autonomic shift would typically be expected to increase heart rate. The authors offer a mechanically plausible explanation. Because participants breathed at shallow depth, intrathoracic pressure swings were minimal, limiting the mechanical stimulation of the heart through preload and afterload changes. This may explain why the fast paced breathing in this study elicited a different heart rate response than protocols that use active or force flexilation such as the traditional yoga practice of kappa body, where forceful expiration drives more pronounced cardiovascular effects. The range of heart period values was also notably wider at 35 cycles per minute than at other conditions, suggesting diverse individual responses at that intermediate frequency that were not visible in the meantime. Perhaps the most methodologically important finding is the absence of a significant correlation between RMSSD and pre ejection period either at baseline or under any breathing condition. Despite group level patterns indicating reciprocal activation at 55 cycles per minute, the correlations for most conditions were close to zero. This independence at the individual level is a significant caution against treating HRV as a proxy for the full autonomic cardiac state. Two people can show identical RMSSD values during fast paced breathing while experiencing quite different levels of sympathetic cardiac activation.
[00:13:43] Without the pre ejection period, that difference is invisible. The breathing discomfort results confirmed expectations. Discomfort increased as frequency deviated from the spontaneous range, peaking at 55 cycles per minute, but the mean discomfort score at this highest frequency was still only moderate, around five out of 10, suggesting the protocol was feasible for this physically active population. No correlations between breathing discomfort and cardiac indices were found, indicating that participants subjective experience of discomfort did not systematically drive or or predict their cardiac responses. The washout analyses found that RMSSD returned to baseline within 2 minutes following all conditions. Pre ejection period effects from the 55 cycles per minute condition also dissipated but not instantaneously. The detailed 10 second epoch analysis showed that cardiac contractility was still elevated in the first 10 to 20 seconds after breathing stopped, with a significant drop toward baseline occurring in the 2032nd window. This gradual return is consistent with the time course of sympathetic cardiac modulation and mirrors findings from a prior study using a virtual reality stress task, suggesting a general pattern for how sympathetic cardiac activation resolves after brief stressors. The cardiorespiratory coupling analysis using phase locking value showed strong heart rate and respiratory synchronization during slow paced breathing at six cycles per minute, consistent with the resonance effect well established in the HRV biofeedback literature. With a mean above chance phase locking value of 0.83 and a phase angle difference of approximately 4 degrees, phase locking decreased monotonically with increasing breathing frequency and and under both fast paced breathing conditions, the proportion of data sets showing above chance coupling was significantly lower than a baseline. In the pace breathing control, the heart and respiratory system were essentially decoupled at fast frequencies. The authors attribute this to multiple mechanisms. Sympathetic modulation of heart rate is too slow to track the fast respiratory rhythm, the shortened respiratory cycle attenuates respiratory sinus arrhythmia and shallow breathing reduces the afferent feedback from pulmonary stretch receptors that normally contribute to heart rate, respiration and trainment. This study provides the most rigorous characterization to date of how fast paced breathing at different frequencies shapes the autonomic cardiac response. The finding that the full reciprocal sympathovagal activation pattern a simultaneously reduced HRV and increased cardiac contractility is specific to 55 cycles per minute and not reliably present at 35 cycles per minute has direct practical implications. Performance practitioners who use fast paced breathing to prime psychophysiological readiness may need to reconsider whether the frequencies they use actually produce the autonomic activation they assume. If the intended effect requires genuine sympathetic cardiac engagement, 35 cycles per minute may not reliably deliver it. The methodological contribution of the novel ensemble averaging technique to pre ejection period estimation deserves recognition beyond this single study. Reliable, non invasive measurement of sympathetic cardiac activity has long been limited by inconsistencies in automated impedance cardiography algorithms. A robust technique that works across diverse waveform morphologies would expand the range of contexts in which the pre ejection period can be validly measured, supporting more comprehensive autonomic assessment in both research and applied settings. For the broader HRV community, the independence of RMSSD and the pre ejection period at the individual level is an important finding. It adds to a body of evidence suggesting that complete characterization of the autonomic cardiac state requires measures from both branches of the autonomic nervous system. Relying on HRV alone captures vagal modulation but leaves sympathetic cardiac activity unmeasured, and this study shows that the two can diverge in ways that matter. The limitations, the authors acknowledge, are substantial and appropriate. The sample consisted exclusively of young, physically active adults, which is relevant because this is the population for which fast paced breathing protocols were explicitly designed in many prior studies. Generalizability to older adults, sedentary populations, or clinical groups is unknown. The study did not measure cognitive or emotional performance outcomes, so the question of which cardiac patterns actually mediate the performance effects of fast paced breathing remains open. The theoretical question of whether our MSSD reductions during fast paced breathing reflect genuine vagal withdrawal or only changes in phasic vagal patterning cannot be resolved without pharmacological blockade studies. Shallow breathing was used throughout to avoid hyperventilation, so the findings may not generalize to fast paced breathing protocols that use active or forceful exhalation, and the exploratory analyses, while suggestive, are acknowledged to be underpowered. Study two was published in Cureus and is titled the Gut Brain Immune Role of Microbiota, Dysbiosis, and Autonomic Nervous System in Infectious Diseases.
[00:17:53] The authors are Sridhuti Uprender Kaur, Narender Kaur, Waqasalaudin, Sayadi Khairnar, Rosibala Vipashikau Shal, and Mohit Mishra. One of the conceptual shifts that has happened quietly but decisively in cardiovascular medicine over the past decade or so is the recognition that HRV does not exist in isolation. It is not simply a cardiac metric. It is a readout of the functional state of the autonomic nervous system, which is embedded in and continuously responsive to a much larger biological ecosystem. The gut microbiome is one of the most important yet least intuitive members of that ecosystem. The gut microbiome, the community of trillions of bacteria, fungi, viruses, and other microorganisms that inhabit the gastrointestinal tract, is now understood to be in continuous bidirectional communication with the brain, the immune system, and the autonomic nervous system. The gut produces neurotransmitters and neuromodulators its microbial inhabitants ferment dietary fiber into metabolites that circulate systemically and influence gene expression in distant tissues. Its immune cells constitute the largest immune organ in the body and the vagus nerve, which is the primary neural highway between the gut and the brain, carrying both afferent signals from the gut upward and efferent autonomic commands downward threads through this entire system as both a sensor and a regulator. When the gut microbiome is disrupted, when microbial diversity collapses, beneficial species are lost and potentially pathogenic species expand.
[00:19:11] This disruption, called dysbiosis, does not stay in the gut it reverberates through the entire gut brain immune network. The question this systematic review sets out to examine is what that reverberation looks like, specifically in the context of infectious diseases. Infection is a state of acute immune challenge that tests the entire system and understanding how prior or concurrent gut dysbiosis shapes the autonomic and immune response to infection has significant clinical implications, particularly as we increasingly recognize that outcomes in infectious disease are not solely determined by the pathogen but but by the host's regulatory capacity. For the HRV practitioner, this paper matters because it provides a mechanistic account of something routinely observed but rarely fully explained. When people get sick, HRV falls. Why? The answer, if this review is correct, is not simply that inflammation suppresses vagal tone, though that is part of it. It is the gut brain immune system that operates as an integrated regulatory network, and dysbiosis can pre compromise that network before any pathogen arrives, making the autonomic response to infection both more severe and less adaptive. This was a systematic review conducted according to Preferred Reporting Items for Systematic Reviews and Meta Analyses or PRISMA guidelines. The authors searched multiple academic databases for peer reviewed studies examining the gut microbiome, autonomic nervous system function, and immune responses in the context of infectious diseases. After applying defined inclusion and exclusion criteria, 11 studies met the threshold for inclusion in the final review. The overall certainty of the evidence across the included studies was rated as low to moderate. As the authors explicitly disclose, studies included both animal models and human research and covered a range of infectious agents, disease context, and outcome measures. The heterogeneity of the included studies is substantial, an important interpretive constraint we will return to the review identified consistent evidence across the included studies that gut microbiota dysbiosis, characterized by reduced microbial diversity and depletion of beneficial bacterial species, particularly from genera that produce short chain fatty acids, increases intestinal permeability.
[00:21:01] This increased permeability, sometimes described colloquially as a leaky gut, refers to a breakdown in the tight junctions between intestinal epithelial cells that normally maintain a selective barrier between the gut lumen and the systemic circulation. When those junctions loosen, microbial products including lipopolysaccharide from the cell walls of gram negative bacteria translocate into the bloodstream. The systemic immune response to these translocated products leads to elevated levels of pro inflammatory cytokines including interleukin 6, tumor necrosis factor alpha and interferon gamma. This cytokine elevation is not a local event confined to the gut it also reaches the cardiovascular and central nervous systems. The review found that these systemic inflammatory changes interact with the autonomic nervous system in a mutually reinforcing cycle. Elevated pro inflammatory cytokines suppress vagal efferent activity which reduces the release of acetylcholine that normally activates the cholinergic anti inflammatory pathway. The the mechanism by which the vagus nerve exerts top down immune suppression. The cholinergic anti inflammatory pathway works as under healthy conditions vagal efferents release acetylcholine that binds to alpha 7 nicotinic receptors on macrophages and other immune cells. This binding suppresses the macrophages release of tumor necrosis factor alpha and other pro inflammatory mediators. This is one of the most studied neural mechanisms for immune regulation and it represents a direct pathway through which a well functioning vagal system can dampen excessive inflammatory responses. When gut dysbiosis disrupts the bidirectional signaling between the gut and brain signaling that largely runs through the vagus nerve, this anti inflammatory regulation deteriorates. The gut sends fewer and qualitatively different afferent signals to the brainstem. The brainstem's inflammatory tone shifts, efferent vagal output decreases removing the cholinergic brake on immune activation. The inflammatory cytokines that were already elevated by microbial translocation now escalate further unrestrained. The collective result is both autonomic imbalance reflected in reduced HRV and impaired immune regulation which the review associates with worse infectious disease outcomes including greater severity and higher mortality. The role of microbial metabolites in this cascade deserves particular mention. Short chain fatty acids, butyrate, propionate and acetate are produced when commensal bacteria in the large intestine ferment dietary fiber. These metabolites perform critical functions. Butyrate is the primary energy source for colonocytes, the cells lining the colon and it also signals through G protein coupled receptors to maintain gut barrier integrity and modulate immune cell behavior. Short chain fatty acid production depends entirely on the presence of the right microbial communities. When dysbiosis depletes these communities, short chain fatty acid production falls, colonocyte health deteriorates, gut barrier function weakens and the pro inflammatory cascade accelerates. The review identifies this metabolite depletion as a key mechanism linking dysbiosis to both barrier failure and immune dysregulation for the HRV community. This review provides a mechanistic framework for understanding why HRV changes during infection. The vagus nerve sits at the intersection of all three systems described here. It carries gut to brain information, it mediates the cholinergic anti inflammatory pathway and its efferent function is suppressed by the cytokines that dysbiosis and infection generate. HRV as an index of vagal modulation is therefore not merely a marker of how sick someone is it may reflect the functional integrity of the entire gut brain immune regulatory network. This also speaks to why some people respond to infectious challenges with more severe autonomic dysregulation than others. If the gut microbiome shapes baseline vagal tone and anti inflammatory capacity, then individuals with pre existing dysbiosis, which is common in populations consuming low fiber western diets, or those who have recently completed antibiotic courses, may enter infectious challenges with a compromised regulatory foundation. Their HRV is already lower, their cholinergic anti inflammatory pathway is already less robust and their inflammatory response to pathogen challenge is therefore less well regulated. This has potential implications for how we think about HRV monitoring in clinical and high risk populations. An individual with chronically low HRV in a context where gut health is also suboptimal may be carrying a compounded vulnerability to infectious disease outcomes. Conversely, interventions that improve microbiome health, dietary fiber probiotics, reduction of unnecessary antibiotic use may improve both HRV and immune resilience through overlapping mechanisms. The limitations are real and the authors address them honestly. With only 11 included studies and evidence rated as low to moderate certainty, the review is better understood as a mechanistic framework supported by preliminary evidence than as a definitive evidence base. The heterogeneity of study designs, infectious agents, animal versus human populations and outcome measures make synthesis imprecise.
[00:25:26] The associations reported are precisely associations cross sectional and observational designs dominated the included evidence and causality cannot be established. The authors explicitly call for longitudinal interventional research with standardized methods. That call is appropriate. The theoretical framework here is compelling, but it requires the kind of prospective human evidence that does not yet exist in sufficient quantity to translate it into clinical guidance. Study 3 was published in Sensors, a Journal of MDPI and is titled early drowsiness detection by a second order derivative analysis of of heart rate variability, a non contact ECG approach with machine learning. The authors are Fabrice Vossena, Abhiru Bhattacharya, Julie Payet, Alire Zasaidi, Victor Bellamine, Jordi Gabriel, Renaud Dumoulin, Sylvain G. Cloutier, and Ghyslin Gagnon. Drowsy driving is one of the most underappreciated safety hazards in modern transportation. The statistics are alarming. Drowsiness is estimated to be a contributing factor in a substantial proportion of road traffic fatalities globally, and unlike alcohol impairment or distracted driving, it is largely invisible until the moment of failure. A drunk driver may swerve visibly a drowsy driver may appear entirely normal right up until the instant their eyes close. The insidious quality of drowsiness as a hazard is that the behavioral markers that signal dangerous impairment head drooping, lane drift, slowed steering correction, micro sleep episodes typically emerge only after the driver's cognitive and physiological state has already deteriorated to a point where crash risk is elevated.
[00:26:48] Warning systems that rely on behavioral detection are therefore inherently reactive. The alert after the problem has fully materialized. The promise of HRV based monitoring in this context is precisely that the autonomic system begins to show signs of sleep pressure and a decline in arousal before behavior does. Sleep deprivation and the accumulating homeostatic sleep drive both affect the balance of sympathetic and parasympathetic activity in characteristic ways. Drowsiness is not simply a subjective state it is a physiological state with a measurable autonomic signature. HRV reflects that signature and in principle, continuous HRV monitoring could detect the physiological precursors of behavioral impairment before dangerous performance decline occurs. The practical challenge has been measurement. Traditional HRV recording requires either a chest strap wearable device or medical grade electrode attachment, all of which introduce compliance barriers and setup friction that make continuous real world implementation difficult. If you want to monitor every driver on every journey, you need an approach that requires nothing from the driver. That is the engineering problem this paper addresses. The solution is capacitive electrocardiography, recording the heart's electrical signal not through direct electrode contact with the skin, but through the capacitive coupling between the body and electrodes embedded in the car seat itself. The driver sits down and monitoring begins. No wearable, no setup, no compliance. The experimental setup used capacitive electrocardiography electrodes embedded in the car seat backrest designed to capture cardiac electrical signals through clothing without direct skin contact.
[00:28:08] Participants completed driving simulator sessions designed to induce drowsiness through monotonous highway driving scenarios, the long straight low stimulus environments most reliably associated with real world drowsiness related crashes. Behavioral markers of drowsiness and impairment were recorded throughout the simulator sessions, establishing a ground truth timeline of when participants driving performance became meaningfully impaired. The key analytical innovation was the addition of first and second derivative features of HRV signals as inputs to machine learning classification.
[00:28:35] Standard HRV analysis extracts features such as the RMSSD and frequency domain power in the low and high frequency bands. These features describe where the HRV signal is at any given moment. First and second derivative features describe how the signal is changing and how the rate of change is itself changing the trajectory of physiological evolution over time, not just the current value. This distinction is potentially important for early detection. If drowsiness related autonomic dysregulation builds gradually over time, derivative features might capture the acceleration of that deterioration before it has progressed far enough to appear in static HRV metrics. Machine learning classifiers were trained on combined feature sets. Conventional HRV metrics alone and conventional metrics augmented with derivative features and their performance in predicting pre crash drowsy states was compared. The classification target was the pre crash drowsy state. Could the system identify the physiological signature of dangerous drowsiness before it manifested in behavioral markers? The combined approach conventional HRV metrics augmented with first and second derivative features fed into machine learning classifiers was able to predict pre crash drowsy states earlier than behavioral cues alone with detection preceding observable behavioral impairment by approximately five to eight minutes. This is not a small margin at highway speeds of 100 kilometers per hour. Five minutes of warning represents roughly 8 kilometers of travel. 8 kilometers is a very long distance to be impaired without intervention. More concretely, five to eight minutes is enough time for a driver monitoring system to issue progressive alerts, reduce vehicle speed, pull toward the shoulder, and initiate a safe stop before a crash occurs. The second order derivative features contributed meaningfully to this performance advantage. The authors interpret this as reflecting the informational value of the trajectory of physiological decline, not just how dysregulated the system currently is, but how rapidly that dysregulation is accelerating. A slow drift toward lower HRV might go unnoticed by a static threshold based alert system, but the rate of change in that drift is a stronger signal that the system is heading somewhere dangerous. The machine learning classifier can weight these derivative features appropriately alongside the standard metrics. The non contact seat embedded sensor approach produced electrocardiography data of sufficient quality to support meaningful HRV analysis, though signal quality varied with driver movement, posture changes, and presumably with clothing type and thickness. The authors document this variability and acknowledge it as both a technical reality and a remaining challenge for real world deployment at population scale.
[00:30:49] Signal quality in a research setting with cooperative participants is likely to be better than in the more heterogeneous real world conditions. The significance of this study operates at two levels. At the applied level, it demonstrates a technically feasible path to continuous, passive, unobtrusive HRV monitoring in one of the environments where it could have the most direct impact on life safety. The SEED embedded sensor approach removes every barrier to implementation. There's no device to wear, no setup procedure, no compliance requirement. The monitoring simply happens and visibly as part of the normal driving experience. At the methodological level, the derivative based analytical approach may have implications beyond drowsiness detection. The principle that the trajectory of physiological change carries predictive information beyond the current state of a metric is general and potentially applicable in any context where early detection of deteriorating regulation is valuable. Intensive care monitoring, performance readiness assessment, occupational safety monitoring, and high risk environments Any domain where you want to know not just where the physiology is but where it is headed and how fast could potentially benefit from similar derivative based feature engineering. The limitations deserve careful attention. The simulator based design raises genuine questions about ecological validity. Real driving involves traffic, weather, varied road environments, emotional responses to other drivers, navigation demands, and countless other factors. Absent from a monotonous simulator scenario, drowsiness induced in a simulator laboratory may have a somewhat different physiological profile than drowsiness accumulated over a long interstate drive in the early morning hours. The machine learning models require validation and external independent data sets, ideally from naturalistic driving studies, before their performance can be considered generalizable. Signal quality from non contact sensors across the full range of real world variability in driver body type, clothing posture and seat position remains a technical challenge. Without a detailed description of the study's sample demographics and characteristics, assessing generalizability across populations is limited. This episode is brought to you by Optimal hrv. The Optimal HRV app is the tool of choice for practitioners, coaches and researchers who want professional grade HRV measurement and tracking without the complexity. Whether you're monitoring a single client or managing a large group, the platform provides the data infrastructure and clinical context to make HRV meaningful. Head to optimalhrv.com to learn more and get started. Study four comes from Neurocritical Care, a Springer journal, and is titled Intracranial Pressure Monitoring, Heart Rate Variability, Baroreflex Sensitivity, and Signal Complexity during Neurointensive Care After Decompressive Cronectomy and Malignant Middle Cerebral Artery Infarction the authors are Modar, Al Hamdan, Anders Hahnel, Timothy Howells, Odin Johnson Fartein Vela Anders Lewin, Per Emblad, and Theodor Svedung Veterwig. Malignant middle cerebral artery infarction is a clinical term that does not fully convey what it represents for the patient and family. The middle cerebral artery is the largest branch of the internal carotid artery, supplying blood to a vast territory of the hemisphere, including motor and sensory cortex, language areas, and the deep white matter connections that tie these regions together. When the proximal middle cerebral artery is occluded and collateral circulation is inadequate, the resulting infarct encompasses a substantial fraction of one entire hemisphere. As the ischemic tissue dies and swells, intracranial pressure rises. The skull is a rigid box. There's nowhere for the swelling to go. As pressure builds, it displaces viable brain tissue, compresses brain stem structures, and accelerates the secondary injury cascade that can rapidly become fatal. Decompressive craniectomy, surgically removing a large section of the skull, typically on the affected side, provides the swelling brain a route of expansion that reduces intracranial pressure and can prevent brainstem herniation. In clinical trials, it has been shown to substantially reduce mortality, but it does not reverse the initial infarct, and many survivors are left with significant disability. A central clinical challenge is identifying during the neurointensive care period following surgery which patients are at greatest risk of ongoing secondary brain injury. Ongoing elevation of intracranial pressure impaired cerebral blood flow autoregulation progressive neurological deterioration so that intensive monitoring and intervention can be concentrated where they are most needed. This study asked whether continuous bedside monitoring of autonomic and cardiovascular physiology, specifically hrv, baroreflex sensitivity, and a complexity based measure of signal dynamics, can contribute to that risk stratification. The scientific rationale is well grounded. The autonomic nervous system is regulated by neural circuits that depend on intact cortical and subcortical architecture. When a large hemisphere is destroyed or compressed, the top down regulation of autonomic function is disrupted, the disruption is measurable, and the degree of disruption may reflect the degree of ongoing secondary injury in ways that predict functional outcomes. Baroreflex sensitivity deserves a brief explanation for listeners not already familiar with it. The baroreflex is the reflex arc by which the cardiovascular system responds to changes in blood pressure. When blood pressure rises, baroreceptors in the carotid, sinus and aortic architecture send afferent signals to the brainstem, which increases vagal efferent activity to slow the heart and restore pressure. When blood pressure falls, the reverse occurs. The sensitivity of this reflex means how much heart rate changes per unit. Change in blood pressure is an index of the responsiveness of this cardiovascular regulatory system, and it reflects both vagal and sympathetic integrity. Signal complexity is the third measure in this paper, and it requires the most explanation. Multiscale entropy. The method used here quantifies the entropy, essentially the unpredictability of a physiological signal across multiple timescales simultaneously. Healthy physiological systems are neither perfectly regular nor purely random. They operate in an intermediate zone of structured variability that reflects the system's capacity to respond flexibly to changing demands. A perfectly regular signal a metronome cannot adapt a purely random signal. White noise has no structure to exploit. Physiological complexity lies between these extremes and tends to decrease with aging, disease, and acute injury. Multiscale entropy captures this across different temporal scales simultaneously, giving a richer characterization of signal dynamics than single scale entropy measures or conventional HRV metrics alone. This was a retrospective single center study conducted at a neurointensive care unit with an established continuous physiological monitoring infrastructure. 70 patients with malignant middle cerebral artery infarction who underwent decompressive craniectomy and had continuous intracranial pressure monitoring were included. Physiological signals, including intracranial pressure waveforms, arterial blood pressure, and cardiac signals, were recorded continuously over the first seven postoperative days. Hrv, baroreflex sensitivity, and signal complexity measures derived using multiscale entropy were computed from these continuous recordings. The primary clinical outcome was functional status at six months following the initial event, assessed using a standardized neurological outcome scale. The retrospective design and single center setting are the primary methodological constraints, and both are inherent to research in this clinical population.
[00:36:58] Malignant middle cerebral artery infarction is relatively uncommon, acutely life threatening, and requires highly specialized care, features that make prospective multicenter trials logistically demanding. Retrospective single center studies are the appropriate starting point for this kind of exploratory physiological investigation, but their findings must be understood as hypothesis generating rather than definitive. Autonomic nervous system function was broadly impaired across the cohort during neurointensive care. Reduced HRV and reduced baroreflex sensitivity were nearly universal findings, consistent with the expected autonomic dysregulation following catastrophic hemispheric injury. These findings were not surprising they aligned with a substantial prior literature on autonomic impairment in severe brain injury, but they established that the cohort showed the expected physiological profile. The critical and more surprising finding was about predictive value. Neither HRV nor baroreflex sensitivity independently predicted functional outcome at six months. In a population where essentially everyone shows autonomic impairment, the absolute level of that impairment did not reliably differentiate who would do better from who would do worse.
[00:37:58] What did predict the outcome was signal complexity. Lower signal complexity of mean arterial pressure, of intracranial pressure itself, and of the intracranial pressure pulse amplitude, meaning signals that were more regular, more stereotyped, less dynamically variable, were associated with significantly worse functional outcomes at 6 months. The predictive value of signal complexity was further enhanced when combined with other physiological variables.
[00:38:21] When low signal complexity co occurred with elevated intracranial pressure, low cerebral perfusion pressure, the driving pressure for cerebral blood flow, or impaired cerebrovascular pressure reactivity, the brain's ability to autoregulate its own blood flow in response to systemic pressure changes, the association with poor outcome was particularly strong. The authors framed signal complexity as a potential bedside biomarker for identifying patients at greatest risk of secondary brain injury during the neurointensive care period. The interpretation offered is physiologically coherent. When the brain's regulatory machinery for autonomic and cardiovascular control is severely damaged, the signals it generates lose their dynamic richness. The intracranial pressure waveform, which under normal conditions reflects a complex interplay of cardiac pulsatility, cerebrovascular reactivity, and respiratory influences, becomes more monotonous, more regular, more rigid. That rigidity may be a fingerprint of a system that has lost the capacity for adaptive regulation. Multiscale entropy, by capturing this across multiple timescales, may be detecting the loss of regulatory flexibility that precedes our accompanies secondary brain injury progression. The finding that signal complexity outperforms conventional HRV and baroreflex sensitivity as an outcome predictor in this population strongly supports expanding the analytical toolkit for neurointensive care monitoring. Continuous physiological monitoring is already standard in this setting. The question is not whether to monitor, but what to compute from the signals. If multiscale entropy of intracranial pressure and arterial pressure waveforms carries prognostic information that stands standard, metrics do not. Incorporating it into the clinical monitoring workflow is a reasonable next step for the HRV community. This study illustrates an important principle. The predictive value of any autonomic measure is population specific and context specific. In a cohort with near universal autonomic impairment, the absolute level of HRV offers little discrimination because the range of impairment is compressed. The dynamic behavior of the signal, its complexity, trajectory, and capacity for flexible variation may carry information precisely. In situations where static HRV metrics do not. This is not an argument against hrv, but rather an argument for methodological humility about what any single metric can accomplish in every clinical context. The limitations are clearly identified. The retrospective single center design introduces selection bias and precludes causal conclusions. The sample of 70 patients, while substantial for this high acuity population, is not large enough to support robust multivariate modeling of multiple complexity measures simultaneously.
[00:40:32] Findings may reflect institution's specific characteristics of patient management, monitoring protocols and surgical technique that do not generalize. Prospective multi center validation is required before complexity based monitoring can be incorporated into clinical practice guidelines. The authors explicitly acknowledge all of these constraints, appropriately framing the study as generating hypotheses for larger prospective trials. Our final study was published in QRIUS and is titled Osteopathic Treatment of Postprandial Cardiovascular Symptoms Suggestive of Altered Autonomic Regulation A Case Report the author is Harvey Shahada. Before we discuss the specifics of this case, it is worth pausing to say something about case reports as a form of scientific evidence because this paper sits at the far end of the evidence hierarchy spectrum from the other studies we have covered today and it deserves to be read in that light. Neither dismissed because it lacks the statistical power of a controlled trial nor over interpreted as evidence of therapeutic efficacy, case reports occupy a particular and genuinely valuable niche in scientific and clinical literature. They cannot establish causality, they cannot quantify efficacy, they cannot support treatment guidelines or clinical protocols. What they can do, and do well when carefully written, is describe clinical presentations in granular detail that builds pattern recognition across the community of readers, generate specific mechanistic hypotheses that controlled studies can subsequently test, introduce measurement approaches or therapeutic framings that other researchers and clinicians can develop further and preserve unusual cases that might otherwise be lost to the literature. This case report does several of those things reasonably well. The clinical presentation described an 84 year old man with recurrent dizziness, weakness, fatigue, near syncope and low blood pressure occurring 10 to 30 minutes after eating is a recognizable but under characterized syndrome. Postprandial hypotension, the fall in blood pressure following a meal, is most common in older adults and affects a meaningful minority of the elderly population, yet it receives relatively little attention in the autonomic cardiology literature compared to orthostatic hypotension or other forms of dysautonomia.
[00:42:20] The mechanism of postprandial hypotension involves the redistribution of blood to the splanchnic vascular bed, the mesenteric circulation that supplies the gastrointestinal tract to meet the increased metabolic demands of digestion. Under normal circumstances, this redistribution triggers a coordinated compensatory autonomic response. Heart rate increases, cardiac output rises and peripheral vascular resistance is maintained, all working together to prevent a drop in systemic blood pressure. In older adults, particularly those with autonomic dysfunction, these compensatory mechanisms may be sluggish or insufficient. The result is the clinical picture described in this case, post meal dizziness and near fainting driven by inadequate cardiovascular compensation for gastrointestinal blood pooling. The author attributes this patient's presentation specifically to vagal predominance, a state in which parasympathetic tone is disproportionately elevated relative to sympathetic activity, creating a cardiovascular regulatory imbalance that is worsened by the post meal parasympathetic surge accompanying digestion. This framing is clinically plausible, though alternative mechanisms, including delayed gastric emptying, insufficient baroreflex responsiveness and impaired sympathetic vasoconstriction can produce similar presentations and would require autonomic testing to distinguish. The therapeutic approach used is osteopathic manipulative treatment, a system of manual therapy in which trained practitioners apply hands on techniques to musculoskeletal and connective tissue structures with proposed effects on autonomic function through both mechanical and neurological mechanisms. The biological plausibility of manual therapy effects on autonomic function rests on the anatomy of the vagus nerve, which courses from the brainstem through the neck and thorax, making functional contact with cervical fascial planes and mediastinal connective tissue. Mechanical pressure stretch or mobilization along these anatomical pathways could in principle modulate the mechanical environment of the vagus nerve and influence its function. Whether it actually does so in clinically meaningful ways is an open question that requires controlled evidence to answer. This is a single case report and the methodological description reflects this. The patient received four weekly sessions of osteopathic manipulative treatment targeting anatomical structures along the course of the vagus nerve, including cervical fascia regions and thoracic connective tissue. Before and after the four session course and before and after each individual session, the following outcomes were a symptom severity score using a self report rating scale, blood pressure and heart rate measurements, and HRV frequency domain parameters including the ratio of low frequency to high frequency power and the high frequency normalized unit. The HRV frequency domain parameters deserve a brief note here. The low frequency to high frequency ratio has historically been interpreted as an index of sympathovagal balance, with higher ratios suggesting sympathetic predominance and lower ratios suggesting vagal predominance. This interpretation has been extensively debated in the literature and many HRV researchers now prefer not to use this ratio as a direct marker of autonomic balance. The high frequency normalized unit reflects the relative power of high frequency oscillations in hrv which are closely tied to respiratory sinus arrhythmia and and vagal modulation. Both measures are commonly used in clinical HRV assessment and their inclusion here allows the author to track autonomic balance across the treatment course, though as noted, interpretive caution is warranted. There was no control condition, no placebo treatment arm, no blinding, no randomization and no follow up assessment beyond the four session treatment course.
[00:45:19] The absence of these methodological features is expected in a case report and should be acknowledged without using them as a reason to discard the observations. Over the four session osteopathic manipulative treatment course, the patient's overall symptom score decreased from a mean of 2.67 to 1.33 on the scale used, representing approximately a 50% reduction in reported symptom burden. Blood pressure and heart rate showed consistent declines following each individual treatment session. The HRV frequency domain measurements changed in a direction consistent with the author's interpretation of reduced vagal predominance and improved sympath vagal balance. The low frequency to high frequency ratio decreased across the treatment course and the high frequency normalized unit increased.
[00:45:55] The author interprets these findings as consistent with osteopathic manipulative treatment modulating autonomic balance in a patient whose postprandial symptoms were driven by vagal predominance. The changes are in a clinically plausible direction. If the treatment is genuinely reducing excessive vagal activity while supporting appropriate sympathetic engagement, we would expect to see reduced parasympathetic markers in HRV and improved post meal cardiovascular compensation resulting in fewer symptoms. However, and this is important, any reader of a case report should hold that interpretation loosely. Blood pressure and heart rate declining after a relaxing hands on treatment session could reflect the general relaxation response that accompanies therapeutic physical contact of any kind. Symptom scores improving over four weeks could reflect natural fluctuation, attention and placebo effects, regression to the mean or the simple fact that the patient was being closely observed and cared for. HRV frequency domain changes over a four week period in an elderly patient could reflect many things seasonal variation, changes in activity level, medication adherence, sleep, or any number of uncontrolled variables. None of these alternative explanations can be excluded without a controlled comparison. This case report has two primary Contributions the first is clinical and descriptive. It adds a carefully documented example to the literature on postprandial autonomic dysregulation in elderly patients, a presentation that deserves more systematic investigation than it has received. The second is methodological. It demonstrates the use of HRV frequency domain measures as a concurrent quantitative assessment tool alongside symptom ratings and vital signs in evaluating a manual therapy intervention targeting the autonomic system. This measurement framework, pairing subjective symptom tracking with objective physiological markers would strengthen any future controlled study of osteopathic or other manual therapies for autonomic dysfunction. For practitioners working with older adults who present with postprandial dizziness or near syncope, this case is a reminder that the autonomic system is the central regulatory mechanism at play and that treatment for framing and monitoring should reflect this. Whether osteopathic manipulative treatment is specifically effective for this presentation is a question that only randomized controlled trials can answer. The case report is not that evidence, and it would be inappropriate to treat it as such. What it is appropriately is a hypothesis that manual therapy targeting vagal anatomy may modulate autonomic balance and postprandial dysautonomia, and that this hypothesis is worth testing rigorously. The limitations are substantial, and the author acknowledges them throughout the single patient no control, no blinding, no randomization, no follow up, and multiple confounding variables uncontrolled. These are not minor caveats. They mean that the clinical observations, however interesting, cannot be confidently attributed to the intervention. The case should be read as a signal to investigate, not as evidence for a change in practice. Five studies today, spanning four different journals, five completely different populations and clinical contexts, and methodological designs ranging from a rigorous controlled laboratory experiment to a single case report. What I want to do in closing is pull on a few threads that connect these papers in ways that I think are worth sitting with. The first thread is frequency dependence and specificity. The Fast Paced Breathing Study by Isker and colleagues is a paper about how important it is to be precise when describing a breathing intervention. 35 and 55 cycles per minute are both fast paced breathing rates. They feel different. They produce different physiological responses. At 35 cycles per minute, HRV falls, but sympathetic cardiac activation does not clearly emerge at 55 cycles per minute. Both things happen simultaneously. The performance effects of fast paced breathing have been documented across a wide range of frequencies in the literature, suggesting we have probably been grouping interventions that produce meaningfully different physiological states. Specificity matters. This lesson applies not just to breathing interventions, but to how we interpret HRV data generally. Context. The frequency, duration, depth and intent of any physiological manipulation shapes what the numbers mean. The second thread is the gut brain immune autonomic network. The systematic review by Tiwari and colleagues is a reminder that HRV does not reside solely in the cardiovascular system. It is a readout of a regulatory network that runs from the microbial communities in the gut, through the intestinal wall, through the vagus nerve, to the brain stem circuits that govern autonomic output. When we see low HRV in an infectious disease context, or in a chronically inflamed patient, or in someone with a long history of poor dietary patterns, we may be looking at a system that has been compromised at multiple levels simultaneously. The vagus nerve is the thread that runs through all of it. Understanding HRV means understanding the thread and what can we strengthen it. The third thread concerns the future of passive monitoring. The driver drowsiness study by Valsinon and colleagues represents something important HRV leaving the clinic and the laboratory and becoming part of the physical environment. The seat knows your heart not through a wearable, not through a device you chose to use, but simply by virtue of where you are sitting. The five to eight minute detection advantage over behavioral cues is meaningful, but what strikes me as more significant is the architecture monitoring that requires nothing from the monitored person. As that architecture extends to office chairs, to aircraft seats, to hospital beds, to performance environments, the question of what we do with continuous passive physiological data becomes increasingly important. The signal processing advances in this paper, including derivative based features, are one piece of the puzzle. What clinicians and practitioners do with early warnings is another. The fourth thread is about what complexity adds to the story. The conventional HRV cannot tell. The neurointensive care study by Alhamdan and colleagues found that HRV and baroreflex sensitivity did not independently predict six month outcomes after malignant middle cerebral artery infarction requiring surgery, but signal complexity did. In a population with near universal autonomic impairment, what differentiates outcomes is not simply how impaired the system is, but whether it retains any dynamic richness, any capacity for flexible variation across time scales. The multiscale entropy approach captures something about the regulatory architecture of a physiological system that static metrics cannot. This is not an argument against hrv. It is an argument for a richer toolkit. The complexity of the cardiovascular signals reflects the organizational complexity of the regulatory system that produces them. And when catastrophic injury strips that organizational complexity away, the signals become rigid and predictably poor. Outcomes follow the fifth thread. The case report invites a different kind of reflection one patient, one author, four sessions we cannot draw any definitive conclusions from this paper, and I want to be clear about that. But there is something worth noting in how the author chose to measure this case. Rather than relying solely on symptom scores and vital signs, they added HRV frequency domain measures to the evaluation, and in doing so they gave the clinical narrative a physiological dimension that points toward the mechanism, not just the outcome. That methodological instinct to reach for autonomic metrics when the clinical problem is autonomic is exactly right. If controlled trials of manual therapy for postprandial dysautonomia are ever conducted, they should look like this methodologically, but larger. The common current running through all five papers is HRV is a window into regulatory capacity, the body's ability to detect change, respond adaptively, and return to an appropriate state. That capacity lives not in a single system, but in the interactions between respiratory and cardiac, microbial and neural, cortical and autonomic, peripheral and central. The studies that teach us the most are the ones that honor that complexity rather than flattening it. That is everything for this week. Thank you for joining me on this week in heart rate variability. Links to all five studies are in the show notes, along with author information and journal details. If this episode was useful to you, please share it with someone in your field who would benefit. We will be back next week with more of the latest research. Until then, take care of yourself.
