Suicide is the 2nd leading cause of death in adolescence, and acute pediatric mental health emergency department (ED) visits have doubled in the past decade. The objective of this study was to evaluate physiologic parameters relationship to suicide severity.
This was a prospective, observational study from April 2018 thru November 2019 in a tertiary care pediatric emergency department (ED) and inpatient pediatric psychiatric unit enrolling acutely suicidal adolescent patients. Patients wore a wrist device that used photoplethysmography for 7 days during their acute hospitalization to measure heart rate variability (HRV). During that time, Columbia Suicide Severity Scores (CSSRS) were assessed at 3 time points.
There was complete device data and follow-up for 51 patients. There was an increase in the high frequency (HF) component of HRV in patients that had a 25% or greater decrease in their CSSRS (mean difference 11.89 ms/
We found an inverse correlation between parasympathetic activity measured through the HF component and suicidality in an acutely suicidal population of adolescents. Wearable technology may have the ability to improve outpatient monitoring for earlier detection and intervention.
Suicide is the 2nd leading cause of death among adolescents, exponentially rising over the past decade and accounting for more than all other non-mental health diagnoses combined, excluding trauma [
Given the significant public health burden, there is a critical need for improved outpatient monitoring that allows for early detection and intervention before an acute mental-health crisis results in a suicide attempt, associated ED visit, or hospitalization. Previous studies have identified physiologic markers that may help identify suicidal patients, including changes in heart rate or breathing related to stress [
With that in mind, the objective of this study was to identify correlations between the HF component of HRV and severity of suicidality in a cohort of acutely suicidal, inpatient adolescents. Based on evidence from prior literature [
This was a prospective, observational study from September 2017 thru November 2019 in a tertiary care pediatric emergency department and an inpatient pediatric psychiatric unit. These 2 locations are within different hospital systems, but located in close geographic proximity. The tertiary care children’s hospital where the emergency department is housed does not have an inpatient psychiatric unit, so many patients requiring admission are transferred to the enrolling inpatient pediatric psychiatric unit. Patients were eligible if they were between the ages of 13–19 years and presented to the ED or were admitted to the inpatient psychiatric unit for acute suicidality. Admission to the psychiatric facility was determined by the treating psychiatry team unrelated to the study. The indication for admission in this population is generally risk of self-harm and need for acute psychiatric interventions. Patients were excluded if they were in the Oregon Department of Human Services custody, acutely agitated as deemed by the primary medical team, were pregnant, had an initial CSSRS <4 (not suicidal), or had insufficient data quality for analysis. Acutely agitated patients were excluded due to the concern for self-harm with a study device and danger to the study team. The study was approved by the Institutional Review Board at both sites and registered with clinicaltrials.gov (NCT03030924). Informed consent was obtained from the patient or legal guardian and assent was obtained from patients <18 years.
HRV data was collected from a wrist-worn device specifically developed for the study that the patient wore for 7 days while it continuously collected photoplethysmography data. HRV data requires beat detection at a consistent point in the cardiac cycle creating a situation where motion artifact is problematic. While prior literature using ECG has used longer durations of time [
At the time of enrollment, demographic data, including past history of an ED visit or inpatient admission for suicidality, formal mental health diagnoses of depression, suicidality or anxiety, daily psychotropic medications taken, gender, age, past non-mental health medical history were collected. Suicide severity was measured by the Columbia Suicide Severity Scale (CSSRS), which has been validated in suicidal adolescents [
Descriptive statistics were calculated for demographic variables, including age, gender, previous admission to a psychiatric facility, previous ED visit for suicidality, days between enrollment in the study and ED visit, attempted self-harm for the current visit, location of enrollment and final assessment (inpatient or home). All data were examined for normality and transformed, as needed. HRV measurements were examined for outliers and across all raw HRV values we removed the highest 1% of data.
To assess for adjusted differences in sqrt (HF) we conducted multiple regression. The primary predictor was the power of sqrt (HF) component on the primary outcome of CSSRS. Analyses were performed at a single point in time (first or second HRV measurement), during Day 1 of the study, or across all days with available data. A CSSRS of 15 or greater was used as a dichotomous variable, because it is approximately 50% of the total composite score and felt by the study team to most likely indicate a patient with significant suicidality. The second way was to identify the correlation between changes in the HF power and changes in CSSRS over time, which allows future work around HRV to be used as a diagnostic tool to monitor for improvement in suicidal behavior. Because patients were actively undergoing treatment outside the study, we aimed to identify HF correlations as the CSSRS improved. To assess suicidal severity changes over time, in particular improved symptoms, a decrease in the Columbia Suicide Score of 25% between 2 times points was used. Covariates included patient gender, age, prior suicide attempts, days between the ED visit and study enrollment, location of enrollment, hour of the day and circadian rhythm. We controlled for circadian rhythm, as previously described, in which the hour of the day was converted to angular format (radians) and then converted to trigonometric format via a Fourier transform pair (sine and cosine) [
The pragmatic use of this data is to identify high risk patients at time of measurement of HRV, but also to monitor for changes over time. A higher CSSRS correlated to increased suicide severity, while a decrease over time suggests reduced suicidality during the study time period. To control for the clustering of repeated measurements within a patient, for all models with more than one measurement per patient, we used mixed effects models with a random effect for patient. To assess model fit, we examined QQ plots, plots of residuals, and influence statistics. We performed a sensitivity analysis using only nighttime data (10 pm–6 am) as the data quality was expected to be much higher with decrease movement while people slept. The overall analysis goal was to evaluate correlations in HF power with CSSRS in 2 ways: The first was to use HF power to distinguish patients with a higher CSSRS from those with lower; this would allow future work around HRV as a diagnostic tool when patients first present to a clinician. The second was to evaluate correlations in changes of HF power with changes in CSSRS over time; this would allow future research around the utility of HRV as a monitoring tool for patients outside the hospital setting.
There were a total of 104 acutely suicidal adolescents enrolled in this study. As detailed in
There were a total 1,215 HRV segments across 51 patients available for analysis across all days, and 697 HRV segments available across 51 patients on day 1. Model fit was acceptable for all regression models. Due to heteroskedastisicity, we used a sandwich estimator for variance in all models. No outliers or leverage values were identified. To accomplish the study objective, HRV measurements were focused on the transformed sqrt (HF).
This novel study was consistent with our hypothesis and found a significant negative association between the HF power, which measures the parasympathetic branch of the autonomous nervous system, and the CSSRS in a population of acutely suicidal adolescents using a wearable technology. This finding indicates that increasing suicidality is associated with decreased parasympathetic tone. The lifetime prevalence of a suicide attempt in adolescence is currently over 20% [
The rationale for this work stems from previous cross-sectional studies evaluating HRV in patients with mental health problems as a biomarker of the autonomic nervous system; in particular parasympathetic activity through the HF power of HRV. Across multiple study designs and research teams, the findings have consistently found an association between the HF and suicidality [
One significant clinical impact wearable technology that monitors physiologic parameters, such as HRV, can have is to identify dynamic changes in suicidality beyond single timepoint measurements. Previous studies have investigated more dynamic changes in HRV by using stress-inducing tasks in the laboratory setting while monitoring HRV. A large systematic review in adults included studies measuring HRV during stressful situations in patients with clinical depression [
While several studies have shown a significant inverse correlation between the HF and depression/suicidality in laboratory settings, this study is significant in that it found similar results in acutely suicidal adolescents admitted to the hospital or ED. The next phase to this area of research would be to intermittently measure HRV longitudinally and as part of a system to alert patients and their loved ones to autonomic dysregulation and higher risk of increasing suicidality. There are several potential interventions patients could implement upon receiving an alert. One area of research has focused on the impact of biofeedback, such as breathing, targeted to change HRV in patients with mental health disorders [
This pilot study has similar findings to past literature, but does have a number of limitations. As detailed in
This pilot study in acutely suicidal adolescents showed a statistically significant inverse correlation between the HF HRV and CSSRS measured by photoplethysmography using a wrist-worn device. This association was found both in comparisons using a single point in time, as well as when analyzed as a change over time. Given that many adolescent patients have a repeat visit to the ED or suicide attempt after an initial visit, the ability to detect worsening suicidality in this high-risk population may have positive impacts on patient outcomes and the healthcare system. Further research will need to confirm this relationship in larger cohorts of acutely suicidal adolescents.
The datasets generated or analyzed during the study are available from the corresponding author on reasonable request.
The authors have no potential conflicts of interest to disclose.
Conceptualization: David C. Sheridan, Steven Baker, Matthew Hansen. Data curation: David C. Sheridan, Steven Baker, Ryan Dehart, Nancy Le. Formal analysis: Amber Lin. Funding acquisiton: Craig D. Newgard, David C. Sheridan. Investigation: David C. Sheridan, Nancy Le. Methodology: David C. Sheridan, Matthew Hansen, Steven Baker, Larisa G. Tereshchenko, Bonnie Nagel. Project administration: Nancy Le. Resources: David C. Sheridan. Software: Ryan Dehart, Steven Baker. Supervision: Craig D. Newgard, Bonnie Nagel. Validation: David C. Sheridan, Amber Lin. Writing—original draft: David C. Sheridan. Writing—Reviewing and editing: all authors.
This was funded by the National Institute of Mental Health K12 award; Oregon Care Emergency Care Research MultiDisciplinary Training Program Award#5K12HL133115.
Beat to beat detection.
Final cohort enrollment.
Columbia suicide score change over time. CSSRS, Columbia Suicide Severity Score (negative is a decrease in score, ie. Improved suicide severity); sqrt(HF), square root of the high frequency component.
Patient and enrollment characteristics (N=51)
Age (yr), mean (SD) | 16.7 (1.5) |
Age (yr) | |
14 | 2 (3.9) |
15 | 11 (21.6) |
16 | 10 (19.6) |
17 | 13 (25.5) |
18 | 11 (21.6) |
19 | 2 (3.9) |
20 | 1 (2) |
21 | 1 (2) |
Sex | |
Female | 37 (72.6) |
Male | 12 (23.5) |
Unknown | 2 (3.9) |
Home medications | |
Antidepressant | 35 (68.6) |
Antipsychotic | 15 (29.4) |
ADHD medication | 8 (15.7) |
Beta blocker | 0 (0) |
Clonidine | 1 (2) |
Patient previously admitted to psychiatric facility or had an ED visit for suicidality | 24 (47.1) |
Patient attempted self-harm | 31 (60.8) |
Location of enrollment | |
Inpatient psychiatric unit | 28 (54.9) |
ED | 23 (45.1) |
Location of final 7 day data | |
Inpatient | 25 (49) |
Home | 26 (51) |
ED, emergency department; SD, standard deviation; ADHD, attention deficit/hyperactivity disorder
Adjusted association of heart rate variability with suicidality
HRV timepoints | Primary predictor | Primary cohort |
Sensitivity cohort: nighttime only (10 pm–6 am) |
||
---|---|---|---|---|---|
Sqrt(HF) change (95% CI), ms/ |
p-value | sqrt(HF) change (95% CI), ms/ |
p-value | ||
Day1 HRV measurements | CSSRS intensity≥15 | -1.69 (-11.19 to 7.82) | 0.730 | -8.34 (-17.11 to 0.43) | 0.071 |
All days with CSSRS (require >1 day with CSSRS) | Dichotomous 25% decrease since baseline CSSRS | 2.16 (-5.18 to 9.51) | 0.564 | 11.899* (3.70 to 20.10) | 0.005 |
Models adjusted for hour of HRV measurement, circadian rhythm, patient sex, age, prior suicidal ideation, days between admission and enrollment, and patient’s final location. *statistically significant. HRV, heart rate variability; CSSRS, Columbia Suicide-Severity Rating Scale; SI, suicidality; sqrt(HF), square root of high frequency component; CI, confidence interval
Unadjusted sqrt(HF) measurements
Primary cohort |
Sensitivity cohort: nighttime only (10 pm–6 am) |
|||||
---|---|---|---|---|---|---|
Overall, mean (SD) | CSSRS intensity <15, mean (SD) | CSSRS intensity >15, mean (SD) | Overall, mean (SD) | CSSRS intensity <15, mean (SD) | CSSRS intensity >15, mean (SD) | |
sqrt(HF) | sqrt(HF) | sqrt(HF) | sqrt(HF) | sqrt(HF) | sqrt(HF) | |
Day1 HRV measurements | 30.97 (17.14) | 34.74 (18.65) | 29.54 (16.3) | 27.77 (15.29) | 34.37 (17.07) | 25.10 (13.69) |
All days with CSSRS (require >1 day with CSSRS) | 31.10 (19.93) | 30.93 (20.16) | 32.03 (18.76) | 31.45 (20.56) | 30.40 (20.39) | 36.02 (21.00) |
Sqrt(HF), square root of high frequency; CSSRS, Columbia suicide-severity rating scale; SD, standard deviation; HRV, heart rate variability