This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in the Journal of Medical Internet Research, is properly cited. The complete bibliographic information, a link to the original publication on http://www.jmir.org/, as well as this copyright and license information must be included.
Changes in autonomic nervous system function, characterized by heart rate variability (HRV), have been associated with infection and observed prior to its clinical identification.
We performed an evaluation of HRV collected by a wearable device to identify and predict COVID-19 and its related symptoms.
Health care workers in the Mount Sinai Health System were prospectively followed in an ongoing observational study using the custom Warrior Watch Study app, which was downloaded to their smartphones. Participants wore an Apple Watch for the duration of the study, measuring HRV throughout the follow-up period. Surveys assessing infection and symptom-related questions were obtained daily.
Using a mixed-effect cosinor model, the mean amplitude of the circadian pattern of the standard deviation of the interbeat interval of normal sinus beats (SDNN), an HRV metric, differed between subjects with and without COVID-19 (
Longitudinally collected HRV metrics from a commonly worn commercial wearable device (Apple Watch) can predict the diagnosis of COVID-19 and identify COVID-19–related symptoms. Prior to the diagnosis of COVID-19 by nasal swab polymerase chain reaction testing, significant changes in HRV were observed, demonstrating the predictive ability of this metric to identify COVID-19 infection.
COVID-19 has resulted in over 41 million infections and more than 1.1 million deaths [
Digital health technology offers an opportunity to address the limitations of traditional public health strategies aimed at curbing the spread of COVID-19 [
HRV is a physiological metric that provides insight into the interplay between the parasympathetic and sympathetic nervous systems that modulate cardiac contractility and cause variability in the beat-to-beat intervals [
The primary aim of the study was to determine whether changes in HRV can differentiate participants who are infected and not infected with SARS-CoV-2. The secondary aim was to observe if changes in HRV can predict the development of SARS-CoV-2 infection prior to diagnosis by a SARS-CoV-2 nasal swab polymerase chain reaction (PCR) test. The exploratory aims were to (1) determine whether changes in HRV can identify the presence of COVID-19–related symptoms; (2) determine whether changes in HRV can predict the development of COVID-19–related symptoms; and (3) evaluate how HRV changed throughout the infection and symptom period.
Health care workers in the Mount Sinai Health System were enrolled in an ongoing prospective observational cohort study. Eligible participants were aged ≥18 years, were current employees in the Mount Sinai Health System, had an iPhone Series 6 or higher, and had or were willing to wear an Apple Watch Series 4 or higher. Participants were excluded if they had an underlying autoimmune disease or were taking medications known to interfere with ANS function. A positive COVID-19 diagnosis was defined as a positive SARS-CoV-2 nasal swab PCR test reported by the participant. Daily symptoms were collected, including fever and chills, feeling tired or weak, body aches, dry cough, sneezing, runny nose, diarrhea, sore throat, headache, shortness of breath, loss of smell or taste, itchy eyes, none, or other. This study was approved by the Institutional Review Board at The Icahn School of Medicine at Mount Sinai.
Participants downloaded the custom Warrior Watch app to complete eligibility questionnaires and sign an electronic consent form. Participants completed an app-based baseline assessment collecting demographic information, prior COVID-19 diagnosis history, occupation, and medical history and were then followed prospectively through the app. Daily survey questionnaires captured COVID-19–related symptoms, symptom severity, SARS-CoV-2 nasal swab PCR test results, serum SARS-CoV-2 antibody test results, and daily patient care–related exposure (Table S1 in
HRV was measured via the Apple Watch Series 4 or 5, which are commercially available wearable devices. Participants wore the device on the wrist and connected it via Bluetooth to their iPhone. The Apple Watch is equipped with an enhanced photoplethysmogram optical heart sensor that combines a green light-emitting diode paired with a light-sensitive photodiode generating time series peaks that correlate with the magnitude of change in the green light generated from each heartbeat [
The HRV data collected through the Apple Watch were characterized by a circadian pattern, a sparse sampling over a 24-hour period, and nonuniform timing across days and participants. These characteristics bias easily derived features, including mean, maximum, and minimum, creating the need to derive methods that model the circadian rhythm of HRV. A cosinor model was used to model the daily circadian rhythm over a 24-hour period with the following nonlinear function:
where τ is the period (τ=24 h); M is the midline statistic of rhythm (MESOR), a rhythm-adjusted mean; A is the amplitude, a measure of half of the extent of variation within a day; and Φ is the acrophase, a measure of the time at which overall high values recur on each day (Figure S1 in
where the linear coefficients
We took advantage of the longitudinal structure of the data to identify a participant-specific daily pattern, and we then measured departures from this pattern as a function of COVID-19 diagnosis or other relevant covariates. To do this, we used a mixed-effect cosinor model, where the HRV measure of participant
where M,
The model parameters and the standard errors of Equation 6 can be estimated via maximum likelihood or reweighted least squares (REWL), and hypothesis testing can be conducted for any comparison that can be written as a linear function of the
However, to test if the cosinor curve, defined by the nonlinear parameters M, A, and
The relationship between a COVID-19 diagnosis and the change in the HRV curves was evaluated. To test this association, we defined the time variant covariate Cit for participant i at time t as follows:
HRV metrics for the 14 days following the time of the first positive SARS-CoV-2 nasal swab PCR test were used to define the positive SARS-CoV-2 infection window. To evaluate the predictive ability of changes in HRV prior to a COVID-19 diagnosis and to explore its changes during the infection period, the time variant covariate was used to characterize the following 4 groups: healthy uninfected individuals (t<t0 – 7), 7 days before COVID-19 diagnosis (t≥t0 – 7, t<t0), the first 7 days post–COVID-19 diagnosis (t0≤t<t0 + 7), and 7-14 days postdiagnosis (t0 + 7≤t<t0 + 14).
To determine the association between COVID-19 symptoms and changes in HRV metrics, we defined being symptomatic as the first day of a reported symptom and compared this to all other days. To evaluate the predictive ability of HRV to identify upcoming symptom days and to explore its changes over time, the time variant covariate was used to characterize the following 4 groups: healthy asymptomatic individuals for t<t0 – 1, 1 day before COVID-19 symptoms (t≥t0 – 1, t<t0), the first day of COVID-19 symptoms (t0≤t<t0 + 1) and 1 day post–COVID-19 symptom development (t0 + 1≤t<t0 + 2).
We enrolled 297 participants between April 29 and September 29, 2020, when the data were censored for analysis (
Baseline demographics of the study participants at enrollment (N=297).
Characteristic | Value | |
|
Age (years), mean (SD) | 36.3 (9.8) |
|
BMI (kg/m2), mean (SD) | 25.6 (5.7) |
|
Female gender, n (%) | 204 (69.4) |
|
||
|
Asian | 73 (24.6) |
|
Black | 29 (9.8) |
|
White | 108 (36.4) |
|
Other | 43 (14.5) |
Hispanic ethnicity, n (%) | 44 (14.8) | |
Baseline positive SARS-CoV-2 nasal swab PCRa test, n (%) | 20 (6.7) | |
Baseline positive SARS-CoV-2 serum antibody test, n (%) | 28 (9.4) | |
|
||
|
Clinical nontrainee | 198 (68.0) |
|
Clinical trainee | 36 (12.4) |
|
Nonclinical staff | 57 (19.6) |
|
||
|
Current or past smoker | 35 (11.9) |
|
Nonsmoker or rare smoker | 259 (88.1) |
Baseline immune suppressing medication, n (%) | 4 (1.4) |
aPCR: polymerase chain reaction.
bClinical trainee is defined as a resident or fellow; clinical nontrainee is defined as a health care worker reporting at least one patient-facing day during follow-up, exclusive of residents and fellows; nonclinical staff is defined as a health care worker who did not report a patient-facing day during follow-up.
Participants classified as not having a COVID-19 diagnosis during follow-up included those with and without a diagnosis of COVID-19 prior to study enrollment. There was no significant difference in the mean MESOR, acrophase, or amplitude of the circadian SDNN pattern of participants with a positive nasal swab PCR test prior to enrollment compared to those who were never diagnosed with COVID-19. This supports the inclusion of participants with a prior COVID-19 diagnosis in our analysis (Table S2 in
The mean MESOR, acrophase, and amplitude of the circadian SDNN pattern for those without COVID-19, those during the 7 days prior to a COVID-19 diagnosis, participants during the 7 days after a COVID-19 diagnosis, and those during the 7-14 days after a COVID-19 diagnosis are described in
Of the 13 subjects diagnosed with COVID-19 during follow up, 6 reported symptoms at some point during the study period. Only 4 subjects had symptomatic COVID-19 infections, reporting symptoms between 7 days prior and 14 days after a positive SARS-CoV-2 nasal swab PCR test. Comparing participants with and without symptomatic COVID-19, no significant differences between the MESOR (28.58 milliseconds, 95% CI 18.61 to 38.56; 37.71 milliseconds, 95% CI 30.65 to 44.98,
HRV parameters in participants with and without COVID-19 diagnoses based on SARS-CoV-2 nasal swab PCR tests.
Parameter | Parameter (milliseconds), mean (95% CI) | Difference (95% CI) | ||||
|
Participants not diagnosed with COVID-19 | Participants diagnosed with COVID-19 |
|
|
||
MESORa | 43.57 (41.40 to 45.40) | 42.46 (38.90 to 45.79) | –1.12 (–4.22 to 1.73) | .46 | ||
Amplitude | 5.30 (4.97 to 5.65) | 1.23 (–1.94 to 3.11) | –4.07 (–7.29 to –2.07) |
|
||
Acrophase | –2.44 (–2.49 to –2.39) | –2.23 (–2.22 to –4.24) | 0.22 (–1.74 to 2.43) | .80 |
aMESOR: midline statistic of rhythm.
bItalic text indicates statistical significance.
Relationship between HRV circadian rhythm and COVID-19 status. Timeline (A) illustrates HRV measures from the time of COVID-19 diagnosis via nasal swab PCR test and during the following 2 weeks after subjects were deemed to be COVID-19–positive and were compared with measurements outside this window, where subjects were deemed to be COVID-19–negative. Daily HRV rhythm (B) on days with positive and negative COVID-19 diagnoses. Plots (C) showing the means and 95% confidence intervals for the parameters defining the circadian rhythm, acrophase, amplitude and MESOR, on days with positive and negative COVID-19 diagnoses. Daily HRV patterns (D, E) for days on which subjects were healthy, 7 days before a positive COVID-19 test, 7 days after a positive COVID-19 test, and 7-14 days after a positive COVID-19 test. Means and 95% confidence intervals for the acrophase, amplitude, and MESOR of the HRV measured on days when participants were healthy, 7 days before a positive COVID-19 test, 7 days after a positive COVID-19 test, and 7-14 days after a positive COVID-19 test. *
Comparison of heart rate variability parameters based on the time periods before and after diagnosis.
Parameter and first period relative to COVID-19 diagnosis | Value (milliseconds), mean (95% CI) | Second period relative to COVID-19 diagnosis | Value (milliseconds), mean (95% CI) | Difference (95% CI) | ||
|
||||||
|
7 days before | 40.56 (35.98 to 45.46) | Uninfected | 43.58 (41.88 to 45.37) | –3.03 (–6.98 to 1.02) | .13 |
|
7 days after | 40.77 (36.44 to 45.42) | Uninfected | 43.58 (41.88 to 45.37) | –2.81 (–6.73 to 1.10) | .17 |
|
7-14 days after | 43.80 (40.01 to 47.65) | Uninfected | 43.58 (41.88 to 45.37) | 0.22 (–3.39 to 3.73) | .89 |
|
7 days before | 40.56 (35.98 to 45.46) | 7-14 days after | 43.80 (40.01 to 47.65) | –3.24 (–9.63 to 3.33) | .32 |
|
7 days after | 40.77 (36.44 to 45.42) | 7-14 days after | 43.80 (40.01 to 47.65) | –3.03 (–6.98 to 1.02) | .13 |
|
7 days after | 40.77 (36.44 to 45.42) | 7 days before | 40.56 (35.98 to 45.46) | 0.217 to (–3.39 to 3.73) | .89 |
|
||||||
|
7 days before | 0.29 (–4.68 to 1.73) | Uninfected | 5.31 (4.95 to 5.67) | –5.02 (–10.14 to –3.58) |
|
|
7 days after | 1.22 (–2.60 to 3.25) | Uninfected | 5.31 (4.95 to 5.67) | –4.09 (–7.87 to –1.93) |
|
|
7-14 days after | 3.80 (–0.64 to 7.88) | Uninfected | 5.31 (4.95 to 5.67) | –1.51 (–5.79 to 2.35) | .48 |
|
7 days before | 0.29 (–4.68 to 1.73) | 7-14 days after | 3.80 (–0.64 to 7.88) | –3.51 (–10.50 to 0.22) | .20 |
|
7 days after | 1.22 (–2.60 to 3.25) | 7-14 days after | 3.80 (–0.64 to 7.88) | –2.58 (–8.44 to 2.08) | .34 |
|
7 days after | 1.22 (–2.60 to 3.25) | 7 days before | 0.29 (–4.68 to 1.73) | 0.93 (–1.92 to 5.83) | .58 |
|
||||||
|
7 days before | –1.67 (–3.78 to 1.19) | Uninfected | –2.44 (–2.49 to –2.39) | 0.78 (–1.4 to 3.62) | .45 |
|
7 days after | –0.53 (-2.39-5.89) | Uninfected | –2.44 (–2.49 to –2.39) | 1.92 (0.03 to 8.13) | .48 |
|
7-14 days after | –2.63 (–3.95 to 1.19) | Uninfected | –2.44 (–2.49 to –2.39) | –0.19 (–1.39 to 1.16) | .70 |
|
7 days before | –1.67 (–3.78 to 1.19) | 7-14 days after | –2.63 (–3.95 to 1.19) | 0.96 (–1.85 to 4.32) | .55 |
|
7 days after | –0.53 (–2.39 to 5.89) | 7-14 days after | –2.63 (–3.95 to 1.19) | 2.10 (0.10 to 8.29) | .35 |
|
7 days after | –0.53 (–2.39 to 5.89) | 7 days before | –1.67 (–3.78 to 1.19) | 1.14 (–1.34 to 7.27) | .58 |
aMESOR: midline statistic of rhythm.
bItalic text indicates statistical significance.
Of the 297 participants, 165 (55.6%) reported developing a symptom during the follow-up period, with the greatest number of participants reporting feeling tired or weak (n=87, 29.3%), followed by headaches (n=82, 27.6%) and sore throat (n=60, 20.2%) (
The mean MESOR, acrophase, and amplitude observed in the circadian SDNN patterns of participants on the first day they experienced a symptom and on all other days of follow-up are reported in
The mean MESOR, acrophase, and amplitude observed in the circadian SDNN patterns of participants on the day before symptoms developed, on the first day of the symptom, on the day following the first day of the symptom, and on all other days are reported in
Number of participants reporting each symptom (N=297).
Symptom | Participants, n (%)a |
Fever or chills | 11 (3.7) |
Fatigue or weakness | 87 (29.3) |
Body aches | 47 (15.8) |
Dry cough | 32 (10.8) |
Sneezing | 52 (17.5) |
Runny nose | 43 (14.4) |
Diarrhea | 33 (11.1) |
Sore throat | 60 (20.2) |
Headache | 82 (27.6) |
Shortness of breath | 11 (3.7) |
Loss of smell or taste | 5 (1.7) |
Itchy eyes | 53 (17.8) |
Other | 26 (8.8) |
aPercentages add to >100% because participants could report one or more symptoms.
Number of symptom days per participant when evaluating days on which participants reported symptoms.
Heart rate variability parameters on the first day of reported symptoms compared to all other symptom-free days.
Parameter | Value on the first day of symptoms (milliseconds), mean (95% CI) | Value on all other days (milliseconds), mean (95% CI) | Difference (95% CI) | |
MESORa | 46.01 (43.37 to 48.77) | 43.48 (41.77 to 45.27) | 2.53 (0.82 to 4.36) |
|
Amplitude | 2.58 (0.26 to 5.00) | 5.30 (4.95 to 5.66) | –2.73 (–5.16 to 0.31) |
|
Acrophase | –2.21 (–2.83 to –1.58) | –2.44 (–2.49 to –2.39) | 0.24 (–0.38 to 0.88) | .44 |
aMESOR: midline statistic of rhythm.
bItalic text indicates statistical significance.
Relationship between HRV circadian rhythm and symptom onset. Timeline (A) illustrates the timing of symptom onset; the HRV profiles of the day of the first symptom are compared to all other days. Daily HRV rhythm (B) on the day of the first symptom and nonsymptomatic or late-symptom days. Plots (C) showing means and 95% confidence intervals for the parameters defining the circadian rhythm, acrophase, amplitude, and MESOR, on first symptom and nonsymptomatic or late-symptomatic days. Daily HRV pattern (D) for nonsymptomatic or late-symptomatic days, the day before the first symptom, the day of the first symptom, and the day after the first symptom. Means and 95% confidence intervals for the acrophase, amplitude, and MESOR of the HRV measured on nonsymptomatic or late-symptomatic days, the day before the first symptom, the day of the first symptom, and the day after the first symptom. *
Comparison of heart rate variability parameters based on symptom state and the time periods before and after the first day of reported symptoms.
Parameter and first symptom state | Value (milliseconds), mean (95% CI) | Second symptom state | Value (milliseconds), mean (95% CI) | Difference (95% CI) | |||||||
|
|||||||||||
|
1 day after first symptom day | 44.52 (42.05 to 46.94) | Asymptomatic | 43.49 (41.74 to 45.21) | 1.03 (–0.64 to 2.67) | .21 | |||||
|
1 day before first symptom day | 43.84 (41.41 to 46.15) | Asymptomatic | 43.49 (41.74 to 45.21) | 0.34 (–1.46 to 2.23) | .73 | |||||
|
First day of symptom | 44.87 (42.42 to 47.18) | Asymptomatic | 45.49 (41.74 to 45.21) | 1.37 (–0.24 to 3.04) | .11 | |||||
|
1 day before first symptom day | 43.84 (41.41 to 46.15) | 1 day after first symptom day | 44.52 (42.05 to 46.94) | –0.69 (–3.72 to 2.47) | .66 | |||||
|
First day of symptom | 44.87 (42.42 to 47.18) | 1 day after first symptom day | 44.52 (42.05 to 46.94) | 0.34 (–1.46 to 2.23) | .73 | |||||
|
First day of symptom | 44.87 (42.42 to 47.18) | 1 day before first symptom day | 43.84 (41.41 to 46.15) | 1.03 (–0.64 to 2.67) | .21 | |||||
|
|||||||||||
|
1 day after first symptom day | 5.47 (3.16 to 7.76) | Asymptomatic | 5.32 (4.99 to 5.66) | 0.15 (–2.21 to 2.37) | .91 | |||||
|
1 day before first symptom day | 2.92 (0.50 to 5.33) | Asymptomatic | 5.32 (4.99 to 5.66) | –2.40 (–4.75 to –0.07) | .06 | |||||
|
First day of symptom | 3.07 (0.88 to 5.22) | Asymptomatic | 5.32 (4.99 to 5.66) | –2.25 (–4.38 to –0.27) |
|
|||||
|
1 day before first symptom day | 2.92 (0.50 to 5.33) | 1 day after first symptom day | 5.47 (3.16 to 7.76) | –2.55 (–6.64 to 1.65) | .25 | |||||
|
First day of symptom | 3.07 (0.88 to 5.22) | 1 day after first symptom day | 5.47 (3.16 to 7.76) | –2.40 (–4.75 to –0.06) | .06 | |||||
|
First day of symptom | 3.07 (0.88 to 5.22) | 1 day before first symptom day | 2.92 (0.50 to 5.33) | 0.15 (–2.20 to 2.37) | .91 | |||||
|
|||||||||||
|
1 day after first symptom day | –2.30 (–2.60 to –2.00) | Asymptomatic | –2.45 (–2.50 to –2.39) | 0.14 (–0.15 to 0.44) | .33 | |||||
|
1 day before first symptom day | –2.52 (–3.31 to –1.71) | Asymptomatic | –2.45 (–2.50 to –2.39) | –0.08 (–0.79 to 0.66) | .86 | |||||
|
First day of symptom | –2.26 (–2.73 to –1.79) | Asymptomatic | –2.45 (–2.50 to –2.39) | 0.19 (–0.24 to 0.63) | .36 | |||||
|
1 day before first symptom day | –2.52 (–3.31 to –1.71) | 1 day after first symptom day | –2.30 (–2.60 to –2.00) | –0.22 (–1.11 to 0.70) | .63 | |||||
|
First day of symptom | –2.26 (–2.73 to –1.79) | 1 day after first symptom day | –2.30 (–2.60 to –2.00) | 0.04 (–0.36 to 0.46) | .86 | |||||
|
First day of symptom | –2.26 (–2.73 to –1.79) | 1 day before first symptom day | –2.52 (–3.31 to –1.71) | 0.26 (–0.40 to 0.92) | .41 |
aMESOR: midline statistic of rhythm.
bItalic text indicates statistical significance.
In this prospective study, longitudinally evaluated HRV metrics were found to be associated with a positive SARS-CoV-2 diagnosis and COVID-19 symptoms. Significant changes in these metrics were observed 7 days prior to the diagnosis of COVID-19. To the best of our knowledge, this is the first study to demonstrate that physiological metrics derived from a commonly worn wearable device (Apple Watch) can identify and predict SARS-CoV-2 infection prior to diagnosis with a SARS-CoV-2 nasal swab PCR test. These preliminary results identify a novel, easily measured physiological metric that may aid in the tracking and identification of SARS-CoV-2 infections.
Current means to control COVID-19 spread rely on case isolation and contact tracing, which have played major roles in the successful containment of prior infectious disease outbreaks [
HRV has been shown to be altered during illnesses, with several small studies demonstrating changes in HRV associated with and predictive of the development of infection [
We demonstrated that significant changes in the circadian pattern of HRV, specifically the amplitude of SDNN, were associated with a positive COVID-19 diagnosis. Interestingly, when we compared these changes over the 7 days preceding the diagnosis of COVID-19, we continued to see significant alterations in amplitude when compared to individuals without COVID-19. This demonstrates the predictive ability of this metric to identify infection. Additionally, most participants diagnosed with COVID-19 in our cohort were asymptomatic. We demonstrated that there was no difference in changes in HRV metrics between participants with and without symptomatic COVID-19 infections. These findings support the utility of using wearable technology to identify COVID-19 infections, even in asymptomatic individuals. When we followed individuals 7-14 days after diagnosis with COVID-19, we found that the circadian HRV pattern began to normalize and was no longer statistically different from an uninfected pattern. As an exploratory analysis, we evaluated how HRV was impacted by symptoms associated with a COVID-19 diagnosis, as individuals may not be tested despite experiencing symptoms. We found significant changes in the amplitude of the circadian HRV pattern on the first day of symptoms, with a trend toward statistical significance on the days before and after symptoms were reported. Similarly, we found significant changes in HRV when we stratified subjects based on severe or nonsevere symptom severity. Taken together, these findings highlight the possible use of HRV collected via wearable devices to identify and predict COVID-19 infection.
There are several limitations to our study. First, the number of participants who were diagnosed with COVID-19 in our cohort was small, limiting our ability to determine how predictive HRV can be of infection. However, these preliminary findings support the further evaluation of HRV as a metric to identify and predict COVID-19 and warrant further study. An additional limitation is the sporadic collection of HRV by the Apple Watch. Although our statistical modeling was able to account for this, a denser dataset would enable expanded evaluation of the relationship between this metric and infections/symptoms. The Apple Watch also only provides HRV in one time domain (SDNN), limiting assessment of the relationship between other HRV parameters with COVID-19 outcomes. Additionally, we did not capture the times of day during which participants were awake or sleeping. Therefore, fluctuations in sleep patterns may have impacted some HRV readings and could not be controlled in the analysis. Finally, an additional limitation is that we relied on self-reported data in this study, precluding independent verification of COVID-19 diagnosis.
In summary, we demonstrated a relationship between longitudinally collected HRV acquired from a commonly used wearable device and SARS-CoV-2 infection. These preliminary results support the further evaluation of HRV as a biomarker of SARS-CoV-2 infection by remote sensing. Although further study is needed, our findings may enable the identification of SARS-CoV-2 infection during the presymptomatic period, in asymptomatic carriers, and prior to diagnosis by a SARS-CoV-2 nasal swab PCR test. These findings warrant further evaluation of this approach to track and identify COVID-19 infections and possibly other types of infection.
Supplementary material.
autonomic nervous system
heart rate variability
interbeat interval
Icahn School of Medicine at Mount Sinai
midline statistic of rhythm
polymerase chain reaction
reweighted least squares
standard deviation of the interbeat interval of normal sinus beats
Support for this study was provided by the Ehrenkranz Lab For Human Resilience, the BioMedical Engineering and Imaging Institute, The Hasso Plattner Institute for Digital Health at Mount Sinai, The Mount Sinai Clinical Intelligence Center and The Dr Henry D Janowitz Division of Gastroenterology.
RPH, MD, GN, and ZAF developed the study concept. RPH assisted with the drafting of the manuscript. RPH, MD, LT, HC, MZ, EG, SK, DH, AB, RP, AC, RM, BG, ML, IN, DR, DC, EPB, LK, MSF, GNN, ZAF, and JA critically revised the manuscript for important intellectual content. RPH, MD, LT, HC, MZ, EG, SK, DH, AB, RP, AC, RM, BG, ML, IN, DR, DC, EPB, LK, MSF, GNN, ZAF, and JA provided final approval of the version of the manuscript to be published and agree to be accountable for all aspects of the work. All authors approved the authorship list. All authors had full access to all the data in the manuscript and had final responsibility for the decision to submit for publication. RPH, ZAF, MSF, MD, and LT verified the underlying data.
RPH discloses consulting fees from HealthMode, Inc, Janssen Pharmaceuticals, and Takeda Pharmaceuticals and research support from Intralytix Inc and a Crohn’s and Colitis Foundation Career Development Award (grant number 607934). BSG has received consulting fees from Data2Discovery, Sema4, and University of California San Francisco. DC is a coinventor on patents filed by the Icahn School of Medicine at Mount Sinai (ISMMS) relating to the treatment for treatment-resistant depression, suicidal ideation, and other disorders. ISMMS has entered into a licensing agreement with Janssen Pharmaceuticals, Inc, and it has received and will receive payments from Janssen under the license agreement related to these patents for the treatment of treatment-resistant depression and suicidal ideation. Consistent with the ISMMS Faculty Handbook (the medical school policy), AC is entitled to a portion of the payments received by the ISMMS. Because SPRAVATO has received regulatory approval for treatment-resistant depression, through the ISMMS, AC will be entitled to additional payments beyond those already received under the license agreement. AC is a named coinventor on several patents filed by ISMMS for a cognitive training intervention to treat depression and related psychiatric disorders. The ISMMS has entered into a licensing agreement with Click Therapeutics, Inc, and has received and will receive payments related to the use of this cognitive training intervention for the treatment of psychiatric disorders. In accordance with the ISMMS Faculty Handbook, AC has received a portion of these payments and is entitled to a portion of any additional payments that the medical school may receive from this license with Click Therapeutics. AC is a named coinventor on a patent application filed by the ISMMS for the use of intranasally administered Neuropeptide Y for the treatment of mood and anxiety disorders. This intellectual property has not been licensed. AC is a named coinventor on a patent application in the United States and several issued patents outside the United States filed by the ISMMS related to the use of ketamine for the treatment of posttraumatic stress disorder. This intellectual property has not been licensed. EPB reports consultancy agreements with Deloitte and Roland Berger; ownership interest in Digital Medicine E. Böttinger GmbH, EBCW GmbH, and Ontomics, Inc; receiving honoraria from Bayer, Bosch Health Campus, Sanofi, and Siemens; and serving as a scientific advisor or member of Bosch Health Campus and Seer Biosciences Inc. LK declares research funding from Abbvie and Pfizer, consulting for Abbvie and Pfizer, and equity ownership/stock options in MetaMe Health and Trellus Health. MSF declares research support from Novartis and Allergenis. GNN reports employment with, consultancy agreements with, and ownership interest in Pensieve Health and Renalytix AI; receiving consulting fees from AstraZeneca, BioVie, GLG Consulting, and Reata; and serving as a scientific advisor or member of Pensieve Health and Renalytix AI. ZAF discloses consulting fees from Alexion, GlaxoSmithKline, and Trained Therapeutix Discovery and research funding from Daiichi Sankyo, Amgen, Bristol Myers Squibb, and Siemens Healthineers. ZAF receives financial compensation as a board member and advisor to Trained Therapeutix Discovery and owns equity in Trained Therapeutix Discovery as a cofounder.