Mindfulness Exercises Reduce Acute Physiologic Stress Among Female Clinicians

Abstract

IMPORTANCE:

Approximately 50% of clinicians experience excessive emotional, physical, and mental stress, with repercussions across the entire medical system. Mindfulness exercises may mitigate this excessive stress. Heart rate variability (HRV) is an objective stress measure that can quantify which mindfulness exercises provide the greatest stress reduction.

OBJECTIVES:

To define the impact of specific mindfulness exercises on HRV, a surrogate for physiologic stress, and the relationship between physiologic (HRV) and subjective stress measured by the State-Trait Anxiety Inventory during a one-day mindfulness workshop.

DESIGN, SETTING, AND PARTICIPANTS:

This was a prospective observational pilot study performed at a quaternary children’s hospital with diverse subspecialists of pediatric nurses, nurse practitioners, and physicians.

MAIN OUTCOMES AND MEASURES:

Our primary outcome measure was change in HRV from baseline during three mindfulness exercises.

RESULTS:

The grounding, deep breathing, and body scan exercises all produced statistically significant changes in HRV among our 13 female participants. The body scan exercise produced statistically significant changes in all studied HRV parameters compared with baseline. We observed significant increases in Root Mean Square of Successive Differences between normal heartbeats (p = 0.026), high frequency (p ≤ 0.001), and the parasympathetic nervous system index (p ≤ 0.001) reflecting increased parasympathetic tone (e.g., relaxation), whereas sd 2/sd 1 ratio (p ≤ 0.001) and the stress index (p = 0.004) were decreased reflecting sympathetic withdrawal (e.g., decreased stress). Subjective stress decreased after 1-day mindfulness training (44.6 to 27.2) (p < 0.001). Individuals with the largest decrease in subjective stress also had the most improvement in HRV during the body scan exercise.

CONCLUSIONS:

Clinician stress levels (HRV) improved after participating in grounding, deep breathing, and body scan meditations, which may highlight their importance as stress reduction tools for clinicians. Monitoring of HRV during mindfulness exercises may provide deeper understanding of which specific exercises produce the greatest physiologic stress reduction for individual participants and the trend of these changes over time.

KEYPOINTS

Question: To determine the feasibility of measuring the effect of specific mindfulness exercises on clinician stress response by measuring physiologic markers of stress (heart rate variability [HRV]) using a noninvasive biometric shirt.

Findings: This prospective observational pilot study found all three mindfulness exercises (grounding, deep breathing, and body scan) produced statistically significant changes in HRV among our 13 female clinicians.

Meaning: HRV data can be captured noninvasively and continuously during mindfulness exercises and may provide deeper understanding as to what specific mindfulness exercises produce the greatest physiologic stress reduction for individual participants and the trend of these changes over time.

Stress and burnout are conditions that plague many members of the healthcare team (1, 2). Approximately 50% of clinicians experience excessive emotional, physical, and mental stress, with repercussions across the entire medical system, patients and their families, and clinicians (3). Stress is defined as the body’s internal reaction to any external stimulus (stressor) that is deemed harmful or dangerous due to an imbalance between demands and resources (4, 5). Excessive stress results in the overactivation of the sympathetic nervous system (SNS), causing increased heart rate (HR) and decreased HR variability (HRV) (4). According to Lazarus and Folkman’s transactional model of stress and coping, our bodies respond to this activation of the SNS by either problem-focused coping, changing the situation itself, or emotion-focused coping, changing the individual response to the situation (5). In recent years, the practice of mindfulness, “conscious, moment-to-moment awareness, cultivated by systematically paying attention on purpose in a particular way” (6), has emerged as an emotion-focused coping method that may help to combat stress, improve resiliency, and help individuals regulate physiologic responses of the autonomic nervous system (ANS) (7–9).

HRV is a measure of the difference in time between consecutive heart beats and serves as a reliable measure of the balance between the SNS and parasympathetic nervous system (PSNS) and is more accurate than HR alone (10). Higher overall HRV, reflecting parasympathetic predominance, is associated with resilience, recovery from acute stressors and improved coping and well-being (10, 11). Lower HRV is associated with depression, anxiety, work-related stressors, and lower cognitive performance (12–14).

Stress has an important role in clinician personal health and well-being, as well as patient outcomes (10, 15–17). Chronic clinician stress results in an increased risk of metabolic syndrome, immune dysfunction, cardiovascular disease, and can eventually lead to burnout (18–20). Clinically, stress impacts families by disrupting clinicians’ ability to empathize, present difficult information, and aid in family decision making (21–23). Stress has also been shown to worsen racial bias, increase patient and family distrust, and decrease patient satisfaction (24–26). Thus, isolating specific techniques that improve HRV and stress levels is important for clinicians and families alike.

Prior research has shown improvement in HRV among outpatient clinicians and subjective improvement in stress with healthy individuals participating in mindfulness training (8, 9, 27–30). Unfortunately, previous investigations of mindfulness training responsiveness have relied on subjective assessments, less reliable HRV technology, and focused on the general population and outpatient clinicians (29–31). Additionally, there is a lack of continuous, objective data defining what specific mindfulness interventions will provide improvement in clinician physiologic stress responses. Before widespread implementation of specific mindfulness interventions among clinicians, it is important to: 1) identify an objective biomarker of ANS tone, such as HRV, that can be measured noninvasively and capture individual responses to mindfulness exercises; 2) determine what specific mindfulness exercises appear most beneficial in reducing overactivation of the SNS; and 3) define the relationship between subjective and objective physiologic stress during specific mindfulness exercises. Defining this relationship between subjective and objective physiologic stress is important: 1) to ultimately raise participant self-awareness of this relationship since physiologic stress typically manifests before subjective awareness of stress and 2) given the significant health implications physiologic stress has on the body and mind as discussed above (32). In this pilot study, we aimed to: 1) determine the feasibility of measuring the effect of specific mindfulness interventions on clinician stress response by measuring physiologic markers of stress using a noninvasive biometric shirt and 2) to define the relationship between physiologic objective stress and subjective stress reports on a pilot population of clinicians.

MATERIALS AND METHODS

This was a single site, prospective pilot cohort study entitled “Mindfulness Training Program’s Effect on Physiologic Parameters” and approved by the institutional review board at Children’s National Hospital, Pro00016482 on September 28, 2021. Procedures were followed in accordance with institutional ethical standards of the responsible committee on human experimentation and with the Helsinki Declaration of 1975. Both male and female nurses, nurse practitioners, and physicians participating in an institutional mindfulness program designed to train interested clinicians in foundational mindfulness exercises were invited to participate in this embedded study via email solicitation. Participation was voluntary and the only exclusion criterion was a personal history of arrhythmia.

The mindfulness training day was an 8-hour day that included instruction on foundational mindfulness exercises, the science behind these exercises and panel discussions. Nurse participants received continuing nursing education for study participation and were not responsible for clinical duties. Before the start of the study, basic demographic information was obtained through a questionnaire. On the day of the mindfulness training, participants were fitted with an appropriately sized smart shirt to measure HRV continuously throughout the mindfulness training day and completed baseline subjective surveys of stress and mindfulness. HRV measurements were collected according to the guidelines of the Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology (33). A 30-minute baseline of HRV was obtained before the start of the training day. HRV analysis specifically focused on three foundational mindfulness exercises (grounding, deep breathing, and body scan) (34). The grounding exercise (exercise 1) focused on bringing awareness to the present moment and is often used at the beginning of mindfulness activities to focus participants. The second exercise was deep breathing and involved participants inhaling for a count of four, holding for a count of seven, and exhaling for a count of eight. The body scan exercise was the third exercise completed and focused on bringing attention to different areas of one’s body to release tension and appreciate bodily sensations. Each exercise was at least 5 minutes in duration, with at least a 1-hour washout period between each exercise, sufficient for HRV analysis per guidelines (33). During the washout period participants continued to receive additional instruction regarding the science of mindfulness and/or take nonclinical breaks. Participants also completed pre- and post-course subjective stress and baseline mindfulness surveys.

Objective Stress Measurements

HRV was used as a surrogate for objective physiologic stress and measured via the Hexoskin smart shirt (35, 36) (Appendix 1, http://links.lww.com/CCX/B425). The Hexoskin smart shirt is a noninvasive device that continuously monitors the user’s cardiorespiratory function and activity using a variety of embedded sensors, including a one-lead electrocardiogram (256 Hz), which is validated to obtain accurate HRV data (35, 37–39). A 30-minute baseline of HRV was obtained on the morning of the course and collected continuously throughout. The data were subsequently divided into 5-minute epochs for consistency in analysis via Kubios (Kuopio, Finland) Premium HRV software (40).

Kubios provides HRV information in three domains: 1) time, 2) frequency, and 3) nonlinear data that were used to determine the effects of mindfulness exercises on ANS balance (Appendix 2, http://links.lww.com/CCX/B425). More specifically, time-domain markers quantify the amount of variability in successive heartbeats. The most commonly used measure of parasympathetic tone is Root Mean Square of Successive Differences between normal heartbeats (RMSSD). RMSSD reflects beat-to-beat differences and measures vagally mediated changes, for example, parasympathetic activation.

The HRV signal may also be divided into different frequency domains with the high frequency (HF) band associated with PSNS tone (e.g., relaxation). Lastly, HRV findings can also be visually represented using nonlinear parameters (Poincarè plots). Poincarè plots are scattergrams created by plotting each heartbeat (RR interval) against the prior one. A normal configuration is defined as fan or comet-shaped and represents a balance between SNS/PSNS activity. A long, slender “torpedo” shape is representative of minimal variability within RR intervals and is consistent with SNS overdrive. Poincarè plots can be further analyzed by placing an ellipse on the plotted shape. The width of the ellipse represents parasympathetic activity (sd 1–sd 1) and the length of the ellipse represents sympathetic activity (sd 2–sd 2). The ratio sd 2/sd 1 is a nonlinear parameter that represents the sympathovagal balance, with higher values correlated with sympathetic activation (41).

Lastly, we used two Kubios-provided summary measures: 1) the stress index representative of SNS tone, for example, stress, that incorporates three parameters: mean HR, Baevsky’s stress index (a geometric measure of HRV reflecting cardiovascular system stress), and Poincarè plot index sd 2 and 2) the PSNS index representative of parasympathetic activation, for example, relaxation, which incorporates three parameters: mean RR interval, RMSSD, and Poincarè plot index sd 1 (40). These composite indices were included to determine if they could be used alone in future studies to represent changes more succinctly in SNS/PSNS during mindfulness.

We selected four commonly used HRV variables to broadly capture all three HRV domains including time (RMSDD [ms]), frequency (HF [normalized], and nonlinear [sd 2/sd 1]) to compare stress levels and PSNS tone throughout the mindfulness exercises to baseline measurements (27, 28, 37). Additionally, we compared the stress index and PSNS index provided by Kubios to characterize HRV changes more broadly during the different mindfulness exercises in hopes they could be used in isolation in future studies. Changes in HRV outcomes from baseline to each mindfulness condition were tested using mixed effects models where HRV outcome was the dependent variable, and mindfulness skill was the independent variable. The participant was included as a random term to allow for multiple HRV assessments per mindfulness skill and between conditions. Effect size and variability estimates from our previous studies allowed us to estimate the power to detect a significant change in HRV outcomes with a proposed sample size of 12 (42).

Subjective Stress, Mindfulness Measurements, and Comparison to HRV

Subjective stress was measured via the short form of the State-Trait Anxiety Inventory (STAI) at baseline and after the course. The short form of the STAI (STAI-6) is a validated survey used to measure anxiety in both patients and healthcare workers that includes six items used to gauge current anxiety (e.g., how the participant feels “right now”) (43, 44). Participants respond to the six questions using a 4-point Likert scale, which ranges from not at all to very much. The range of scores is 20–80, with higher scores indicating greater current anxiety (44). Mindfulness was measured via the Mindful Attention Awareness Scale (MAAS) at baseline. The MAAS is a 15-item scale used to measure mindfulness, or ability to pay attention to the present moment, with higher mean scores associated with greater mindfulness and decreased subjective stress (45, 46). Correlations between STAI score at baseline and HRV outcomes during each mindfulness skill were performed using a Pearson correlation.

RESULTS

Demographic characteristics of the 13 participants are outlined in Table 1. Nurses and nurse practitioners (61.5%) made up the majority of participants and physicians comprised the remaining 38.5%. The median age was 42 years, and all participants were female. The majority of participants had previous stress reduction training (76.9%). Qualitatively, participants cited an overwhelming amount of work, time constraints, and staffing shortages as the most common contributors to occupational stress. The study protocol is shown in Figure 1.

TABLE 1.

Demographic and Participant Characteristics

Characteristic	n (%)
Race
Asian	2 (15.4)
African-American	1 (7.7)
Caucasian	8 (61.5)
East Indian/Pakistani	1 (7.7)
Hispanic/Latino	1 (7.7)
Current role
Nurse	7 (53.9)
Nurse practitioner	1 (7.7)
Physician	5 (38.5)
Department
Anesthesia/pain clinic	3 (23.1)
Cardiology	3 (23.1)
Critical care	3 (23.1)
Emergency department	3 (23.1)
Simulation	1 (7.7)
Caffeine consumption
No	1 (7.7)
Yes	12 (92.3)
Average hours of sleep
< 6 hr	3 (23.1)
6–7 hr	10 (76.9)
Exercise regularly
No	6 (46.2)
Yes	7 (53.9)
Currently practice or had previous mindfulness or other stress reduction training
No	3 (23.1)
Yes	10 (76.9)
Experienced a major life event in the previous 6 moa
No	8 (61.5)
Yes	5 (38.5)

^aDefined as experienced death, divorce, marriage, and serious illness.

Figure 1.

Study protocol. HRV = heart rate variability, MAAS = Mindful Attention Awareness Scale, MBSR = mindfulness-based stress reduction; STAI = State-Trait Anxiety Inventory.

HRV changes from baseline during each of the three exercises are displayed in Table 2. As predicted, the grounding exercise produced statistically significant increases in HF and the PSNS index indicating increased parasympathetic activation and statistically significant decreases in sd 2/sd 1 ratio and stress index indicative of decreased sympathetic tone. The deep breathing exercise paradoxically showed a decrease in HF norm indicative of increased sympathetic activation. However, it was also associated with an increase in PSNS index and a decreased stress index as predicted consistent with lower levels of stress. Notably, the body scan exercise produced statistically significant changes in HRV data compared with baseline across all HRV parameters. RMSSD (p = 0.026), HF, and PSNS index (< 0.001) were increased reflecting increased parasympathetic tone, whereas sd 2/sd 1 ratio (< 0.001) and the stress index (0.004) were decreased reflecting sympathetic withdrawal. Poincarè plots provided visual representation of this decreased stressed response with change in shape from torpedo at baseline consistent with sympathetic over-activation to a fan shaped during the body scan exercise consistent with parasympathetic activation (Fig. 2).

TABLE 2.

Heart Rate Variability Change From Baseline Compared With Mindfulness Exercises^a

Heart Rate Variability Metricsb	Mindfulness Skill	Estimated Mean ± se	p Comparing Each Condition to Baseline
Root Mean Square of Successive Differences between normal heartbeats	Baseline	21.90 ± 2.56
	Grounding	31.62 ± 4.30	0.10
	Deep breathing	41.99 ± 7.43	0.09
	Body scan	35.68 ± 3.79	0.026
High frequency (normalized value)	Baseline	29.4 ± 3.4
	Grounding	40.0 ± 5.1	< 0.001
	Deep breathing	11.4 ± 3.0	< 0.001
	Body scan	48.9 ± 6.0	< 0.001
Parasympathetic nervous system index	Baseline	–1.46 ± 0.14
	Grounding	–0.82 ± 0.17	< 0.001
	Deep breathing	–0.28 ± 0.30	0.002
	Body scan	–0.43 ± 0.19	< 0.001
sd 2/sd 1 ratioc	Baseline	3.99 ± 0.29
	Grounding	3.21 ± 0.33	< 0.001
	Deep breathing	4.32 ± 0.29	0.89
	Body scan	3.00 ± 0.34	< 0.001
Stress index	Baseline	15.8 ± 12
	Grounding	13.5 ± 1.5	0.046
	Deep breathing	8.2 ± 0.8	< 0.001
	Body scan	11.9 ± 1.1	0.004

^an = 13 participants.

^bHeart rate variability metrics in gray are consistent with parasympathetic activation (relaxation).

^cContinuous long-term beat-to-beat interval variability sd 2 instantaneous beat-to-beat interval variability (sd 1).

Boldface values are statistically significant findings.

Figure 2.

Poincarè plots demonstrating parasympathetic predominance (relaxation) during body scan exercise. A, Participant baseline torpedo shaped measurement consistent with sympathetic overdrive. B, Same participant demonstrating parasympathetic predominance, relaxation, during the body scan exercise. X-axis RR_n (ms): RR interval, y-axis: RR_{n + 1} (ms): subsequent RR interval.

The mean STAI score for participants decreased from 44.6 to 27.2 post-mindfulness training, reflecting decreased subjective stress levels after training (p ≤ 0.001). Baseline mindfulness scores for participants were a 3.2, which is slightly lower than the average score of the general population of 3.85 (45).

Lastly, we evaluated the relationship between participants’ change from baseline subjective stress (STAI) and objective stress (HRV) to determine if those who subjectively had a greater improvement in stress post-mindfulness training also had a larger physiologic improvement in HRV. We found a relationship between subjective stress and objective physiologic stress during the body scan and grounding exercises. The body scan exercise revealed statistically significant changes in two HRV parameters (HF and the stress index) from baseline (r² = 0.496; p = 0.007 and r² = 0.308; p = 0.049, respectively). In other words, participants with the largest decrease in subjective stress after the mindfulness training were associated with increased parasympathetic tone (as evidenced by a larger positive change in HF) and sympathetic withdrawal (as evidenced by a decrease in the stress index) (Fig. 3).

Figure 3.

Relationship between subjective stress and heart rate variability markers from baseline during body scan exercise. Change from baseline (CFB) in high frequency (HF) normalized value (CFB HF norm; x-axis) and the stress index were associated with statistically significant CFB in State-Trait Anxiety Inventory (CFB STAI; y-axis) during body scan exercise (r² = 0.496; p = 0.007 and r² = 0.308; p = 0.049, respectively). A, The figure on the left shows that participants with the largest decrease in subjective stress from baseline (more negative) were associated with a positive change in HF indicating increased parasympathetic tone. B, The figure on the right shows that those with the largest decrease in subjective stress (more negative) were associated with a decrease in stress index indicative of sympathetic withdrawal during the body scan exercise.

There was also a relationship between subjective and objective physiologic data in the grounding exercise where a larger change in the HF norm (r² = 0.394; p = 0.029; relaxation) was noted in participants who reported a larger improvement in subjective stress post-training, meaning those who physiologically were most relaxed were also the individuals who had the largest subject stress decrease. We did not find any other statistically significant relationships between objective and subjective stress in the other HRV parameters or mindfulness exercises, indicating there was not a consistent association between self-assessment of stress and objective physiologic stress throughout.

DISCUSSION

Across our pilot population, we found statistically significant changes in objective physiologic stress (HRV) during three mindfulness exercises. The body scan exercise produced statistically significant changes in all of the HRV parameters we examined. We found that parasympathetic tone was increased in combination with sympathetic withdrawal. High levels of parasympathetic tone and vagal activation are associated with improved ability to respond to stressors and increased resiliency, which may translate to improved patient care (7–9). The body scan exercise is considered an essential starting point by many mindfulness experts and our findings here corroborated that this exercise may be the most helpful of those studied here for clinician acute stress reduction (34). The other exercises, grounding and deep breathing produced statistically significant changes in some but not all HRV domains. This highlights two possible explanations: 1) our sample size was not large enough to produce statistically significant changes in all HRV domains and 2) individual stress interpretation and response play an important role in determining what techniques will be beneficial in mitigating maladaptive stress responses for a given participant. This second point refers to the Lazarus and Folkman’s transactional model of stress and coping where individual coping responses are essential for overcoming the body’s stress response (5). This further highlights the importance of: 1) objectively measuring these responses to determine which techniques work most effectively for each individual and 2) introducing a diverse number of techniques and exercises to allow participants to select ones that are most beneficial in navigating and overcoming stress.

In the deep breathing exercise, we found a paradoxical decrease in HF, reflecting parasympathetic withdrawal. HR is influenced by respiratory patterns (e.g., respiratory sinus arrhythmia) and this decrease in HF may be related to a lack of synchrony between HR and breathing when individuals are not breathing at a resonance frequency (47, 48). Additionally, parasympathetic withdrawal during deep breathing may be explained by a potential increase in participant stress/mental concentration associated with learning to modulate breathing in a novel way (49, 50); however, the more general and multidimensional metrics of PSNS index and stress index revealed that stress was reduced during this exercise for clinicians making this second explanation less likely.

Subjectively, participant stress was decreased during this 1-day mindfulness training session. This is promising for acute reduction in stress. Longitudinal follow-up of participants will be important to determine if this acute stress reduction is sustained as well as long-term improvements in mindfulness. We found those with the largest subjective stress reduction after the mindfulness course also had the largest improvement in physiologic stress during the body scan and grounding exercises. This highlights that subjectively feeling less stress is associated with improvement in physiologic stress during this exercise. The lack of correlation between subjective and objective stress in the other mindfulness exercises may be secondary to: 1) sample size limitations, 2) acquisition of new mindfulness skills may be stressful, and 3) subjective improvement in stress levels may be challenging for many individuals to quantify further highlighting the importance of objective stress measures.

In summary, this study demonstrates that HRV data can be captured noninvasively and continuously during mindfulness exercises using a noninvasive smart shirt. We found the composite stress indices (stress index and PSNS index) produced by the Kubios software were reliable markers of sympathetic and parasympathetic activation respectively. Additionally, we found the Poincarè plots to reveal distinct changes in shape during the mindfulness exercises designed to elicit relaxation that are appreciable by the lay public. These findings are essential to future studies where we will harness HRV technology in real time to provide participants with same-day visual and numeric stress improvement findings without delays of statistical analysis.

Although we found important and statistically significant changes in this prospective study, we are limited by the sample size of 13 participants in this pilot study. Expansion to a larger group of participants of diverse backgrounds, including male participants, will be an important next step given there may be differences in stress appraisal and coping based on gender (51, 52). The body scan exercise produced the most robust and consistent decrease in stress response, which was the last mindfulness skill performed. In this cohort, we believe the skills had a distinct impact rather than cumulative effect given the return to baseline HRV between skills and the attempt to mitigate the cumulative effect with a washout period between each of the skills. In future studies with larger cohorts, it will be important to randomize the order of mindfulness skills to investigate the relationship between any cumulative effects. Additionally, this study focused on the acute reduction of stress measured via HRV and STAI, omitting any impact on chronic stress reduction, which is an important area of future research. Lastly, of note, the participants in this study self-selected to participant in mindfulness training and had some level of stress reduction training previously, which may impact the generalizability of these findings to those not invested in mindfulness training.

CONCLUSIONS

Specific mindfulness exercises, most notably the body scan exercise, produced improvement in stress response as measured via HRV among female clinicians. Clinicians may consider practicing mindfulness exercises as one method of stress mitigation that can be deployed quickly and without extensive training. Monitoring of HRV during mindfulness exercises may provide further granularity as to what specific mindfulness exercises produce the greatest physiologic stress reduction for individual participants and the trend of these changes over time. Importantly, Poincarè plots and composite stress indices may provide useful real-time feedback to participants in future studies regarding physiologic stress responses during mindfulness exercises. The relationship between subjective and objective stress during mindfulness training warrants further investigation.

Supplementary Material

cc9-6-e1171-s001.pdf ^{(291KB, pdf)}