Outcomes of evidence-based psychotherapy for healthcare workers: a single-arm prospective cohort study

Abstract

Background

Healthcare workers (HCWs) experience high rates of depression and anxiety, yet face substantial barriers to accessing mental health care, including limited availability, scheduling constraints, cost, and concerns about confidentiality and stigma. Despite widespread efforts to support HCWs well-being, there is limited research on improving access to high-quality mental health care for this population.

Objective

To evaluate engagement and clinical outcomes associated with a scalable, employer-sponsored digital psychotherapy program implemented within a large health system to improve access to high-quality mental health care for HCWs and their dependents.

Methods

This prospective cohort study was conducted from March 2022 through July 2024 within an employer-sponsored digital mental health program at a large integrated healthcare system. Participants were 496 adults (employees or adult dependents) who enrolled in online psychotherapy and scored ≥ 5 on the Patient Health Questionnaire-9 (PHQ-9) or Generalized Anxiety Disorder-7 (GAD-7) at baseline. Of these, 348 (70.2%) completed at least one follow-up assessment. Participants received weekly evidence-based psychotherapy delivered via a digital platform. Primary outcomes were change in PHQ-9 and GAD-7 scores from baseline to last follow-up; secondary outcomes included remission (score < 5 among those ≥ 10 at baseline) and treatment engagement (sessions attended, treatment completion).

Results

Of the 496 participants who initiated psychotherapy, 348 (70.2%) completed at least one follow-up assessment. Among followed participants (mean age 35 years; 80% female), mean symptom reductions were 6.4 points on the PHQ-9 (95% confidence interval [CI], 5.7, 7.1) and 6.3 on the GAD-7 (95% CI, 5.8, 6.9; both ps < 0.001; d ≈ 1.3). In the clinical subsample, 36.5% achieved depression remission and 48.8% achieved anxiety remission. Engagement was high, with participants attending a mean of 15 sessions and 81.9% completing treatment. Treatment completion was significantly associated with greater symptom improvement.

Conclusions

A scalable, health system–based digital psychotherapy model was feasible to implement and achieved high engagement and clinically meaningful improvements among HCWs receiving care. These findings suggest that health systems can expand access to high-quality mental health care for their workforce through digitally delivered, evidence-based care models.

Clinical trial number

Not applicable

Introduction

Healthcare workers (HCWs) experience high rates of burnout, depression, anxiety, and related mental health problems even before the COVID-19 pandemic [1–4]. Prior to the pandemic, estimates of burnout ranged from 24% to 70%, depending on specialty and setting [5, 6]. The pandemic precipitated sharp increases in clinically significant anxiety, depression, and trauma-related symptoms among HCWs [7, 8]. Left unaddressed, these conditions contribute to diminished quality of care, adverse patient outcomes [9, 10], workforce attrition and substantial organizational costs [11, 12].

There is now a large literature on interventions for HCW burnout and mental health focused primarily on non-clinical supports such as psychoeducation [13], peer support [14], cognitive-behavioral stress management [15, 16]. In addition, a growing body of work has examined system-level interventions aimed at strengthening institutional mental health infrastructure, including screening, triage, peer support, and care navigation for healthcare workers [17, 18]. However, little research has examined how to improve healthcare workers’ access to high-quality mental health treatment.

Healthcare workers face the same significant barriers to accessing mental health services as other employed populations, including limited provider availability, scheduling constraints, high out-of-pocket costs, stigma, and limited understanding of treatment options and how to access care [19, 20]. They also encounter profession-specific concerns related to confidentiality, stigma within healthcare culture, and potential licensing consequences associated with seeking care [21, 22]. These barriers highlight the need for scalable, health system–based approaches that can improve access to high-quality mental health care for healthcare worker [23, 24].

The current study

In a prior report, we described the development and implementation of BeHealth, a system-level digital mental health program designed to improve access to high-quality mental health services for employees and their dependents in a large, self-insured healthcare system [27]. Core program components included a dedicated website where employees could access help for mental health concerns at any time, a care navigation service that matched patients to treatments based on preference and clinical need, and an online evidence-based psychotherapy service.

The current study examined clinical outcomes of the online psychotherapy component as part of this broader systems level model. Specifically, we evaluated treatment engagement and changes in anxiety and depressive symptoms among employees and their dependents who initiated psychotherapy and reported clinically significant symptoms at baseline. We also examined outcomes among participants with moderate-to-severe baseline symptoms and explored the relationship between treatment engagement and clinical improvement. We hypothesized that participants would demonstrate significant reductions in anxiety and depressive symptoms during treatment, and that greater engagement would be associated with greater symptom improvement.

Methods

Study design and setting

This single-arm prospective cohort study evaluated outcomes of an employer-sponsored online psychotherapy program implemented at Northwell Health, a large integrated healthcare system in metropolitan New York. Northwell Health is a self-insured entity and employs over 85,000 individuals including 12,000 physicians and 19,000 nurses. The study analyzed data from program inception in March 2022 through July 2024. Enrollment and clinical assessments were completed through a secure web-based portal. Participants were followed prospectively with standardized outcome measures administered at baseline and monthly during treatment. The Northwell Health’s Institutional Review Board reviewed the project and determined it met criteria for quality improvement and was not considered human subjects research; therefore, informed consent was not required. All procedures were conducted in accordance with the ethical standards of the institutional research committee and with the Declaration of Helsinki.

Program description

The psychotherapy component of the BeHealth program was designed to deliver evidence-based treatment at scale within routine health system operations. Online psychotherapy was offered to eligible employees and their adult dependents at Northwell Health. Patients enrolled in online psychotherapy via a centralized digital platform that enabled self-referral, clinical assessment, and facilitated timely connection to care. Patients completed validated self-report clinical assessment measures upon sign-up and were then interviewed by a licensed mental health provider to assess whether they met eligibility criteria to receive weekly online psychotherapy. Patients were deemed ineligible to receive online psychotherapy if they met any of the following criteria: (1) any serious mental health condition resulting in severe functional impairment; (2) any alcohol withdrawal symptoms of greater than mild intensity; (3) active/imminent suicide risk; (4) regular drug use, defined as greater than weekly in frequency; (5) high severity of drinking, defined as 45 drinks per week or higher; (6) inpatient hospitalization for a substance use disorder or other psychiatric condition within the past 3 months [25, 26]. Patients assessed as ineligible or those who preferred a treatment modality other than online time-limited psychotherapy were referred to BeHealth’s care navigation services. Further details on program marketing, assessment procedures, and eligibility criteria are described in an earlier publication [27].

Patients were treated with the Unified Protocol (UP) for Transdiagnostic Treatment of Emotional Disorders—a cognitive-behavioral therapy (CBT) chosen for its flexibility in addressing a range of anxiety, depressive, and related disorders [28]. UP consists of 8 treatment modules targeting core emotion regulation processes and was delivered by trained clinicians using the published treatment manual as a guide. Treatment was structured as a course of up to 24 weekly sessions to facilitate completion of all 8 treatment modules in the context of routine clinical care. Measurement-based care was also implemented. Patients completed validated symptom measures at intake and monthly during treatment, and ratings were shared with clinicians to monitor progress and guide clinical decision-making. Sessions were conducted via a HIPAA-compliant video platform and delivered by licensed, Master’s- and Doctorate-level mental health professionals with at least 3 years of experience delivering CBT. Clinicians completed a training workshop from the Unified Protocol Institute and received weekly group supervision. Supervision focused on helping clinicians deliver manual guided UP in a clinical practice setting.

A digital platform was developed that supported the delivery of measurement-based care and manual guided UP. Digital treatment worksheets were developed using the UP manual that presented the didactic information and in-session practice exercises for each UP module. Patients completed intake and monthly clinical assessments in the platform and could view and complete treatment worksheets in between sessions. Therapists accessed the platform to view patient responses to clinical assessments. In addition, therapists could open treatment worksheets during sessions, share and review them with patients and then send patients the worksheets to complete as practice exercises. An internal audit of the treatment platform during a representative six-month period found that therapists accessed the worksheets during 94% of treatment sessions.

Participants

Participants were 496 adults who enrolled in the online psychotherapy program between March 2022 and July 2024 and reported at least mild anxiety or depressive symptoms, defined as a score ≥ 5 on either the Patient Health Questionnaire-9 (PHQ-9) ²⁹ or the Generalized Anxiety Disorder-7 (GAD-7) ³⁰ at baseline.

Outcomes and measures

Clinical outcomes were assessed with the PHQ-9 for depressive symptoms and the GAD-7 for anxiety symptoms. Both were administered at baseline and monthly during treatment. Continuous outcomes were defined as change from baseline to the last available follow-up score. Remission was defined as a follow-up score < 5 among participants with a baseline score ≥ 10 on that measure [31]. Engagement outcomes included number of sessions attended and treatment completion, defined as completing all 8 UP modules. For secondary analyses, a clinical severity subgroup was defined as participants with a baseline score ≥ 10 on either the PHQ-9 or GAD-7, consistent with established clinical cutoffs [28, 29]; this subgroup is referred to below as the “clinical sample”.

Data collection procedures

All clinical and engagement data were captured in the digital platform which was not part of health system’s electronic health record and accessible only to a small group of BeHealth staff to protect patient privacy. Participants were prompted via automated text message monthly during treatment to complete the PHQ-9 and/or GAD-7 if they scored ≥ 5 on that measure at baseline.

Statistical analyses

Participants completing at least 1 follow-up assessment were classified as “followed,” and baseline characteristics and engagement rates were compared between followed and not-followed groups for the full sample. Primary outcomes were change scores on the PHQ-9 and GAD-7, analyzed first with paired-sample t-tests. Linear regression models were then used to examine predictors of change in PHQ-9 and GAD-7 at last follow up. Change scores were modeled, rather than endpoint scores, to evaluate improvement from baseline and account for between-participant variability [32]. Baseline severity was included as a covariate.

Analyses were conducted for all followed participants and for the clinical subsample. Predictors included baseline symptom severity, demographics (age, sex), and treatment engagement (number of sessions attended, treatment completion). Predictors were first examined individually (controlling for baseline severity). Variables significant at p < .05 were entered into multivariable models, and non-significant predictors were removed stepwise to yield the most parsimonious models. Finally, binary logistic regression was used to test predictors of remission in the clinical subsample. Separate models evaluated remission of depression (PHQ-9 < 5) and anxiety (GAD-7 < 5), with predictors tested independently while controlling for baseline severity.

Results

Follow-up rates and baseline comparisons

Of the 496 participants who initiated psychotherapy, 348 (70.2%) completed at least 1 follow-up assessment and were classified as followed. The median time to last follow-up was 5.0 months (range = 1.0–15.0). Comparisons of baseline characteristics showed no significant differences between followed and not-followed participants. However, those who were followed attended significantly more therapy sessions (M = 14.9, SD = 5.2) compared to those who were not followed (M = 7.3, SD = 5.7), t(251.24) = -13.98, p < .001, Cohen’s d = -1.43). In addition, a greater proportion of participants who were followed completed treatment (81.7%) compared to those who were not followed (31.3%): χ²(1, N = 496) = 118.21, p < .001).

Participant descriptives

Participants were on average around 35 years old, and a majority were female (Table 1). About 60.1% (n = 301) of participants scored ≥ 10 on the PHQ-9 or GAD-7. Differences between all participants’ average baseline GAD-7 and PHQ-9 scores and those with clinically significant scores (≥ 10) was only about 2 points.

Table 1.

Demographic characteristics, treatment engagement, symptom change, and remission

Demographics	Full Sample N = 496	All Followed Participants N = 348	Clinical Subsample N = 211
Age, mean (SD)	35.3 (9.8)	34.7 (9.3)	33.6 (8.8)
No. (%) sex, female	80.0	80.4	79.1
No. (%) race
Asian	14.3	12.9	10.9
Black/African American	12.5	11.8	12.3
Caucasian	56.9	58.6	59.2
Other	16.3	16.7	17.6
No.(%) ethnicity, Hispanic	16.8	16.4	16.6
No. % education, bachelor’s degree or higher	78.4	78.2	74.4
Baseline Symptom Severity
GAD, mean (SD)	9.9 (4.9)	10.4 (4.7)	12.8 (4.2)
PHQ, mean (SD)	8.2 (5.9)	8.3 (5.9)	10.7 (5.8)
Treatment Engagement
Length of stay, mean (SD), months	4.6 (2.6)	5.2 (2.5)	5.3 (2.4)
Number of sessions, mean (SD)	12.7 (6.4)	15.0 (5.2)	14.9 (4.9)
No. (%) completed treatment	66.7	81.9	84.4
Clinical Outcomes
GAD last follow up, mean (SD)	-	4.5 (3.4)	5.0 (3.8)
PHQ last follow up, mean (SD)	-	4.1 (3.4)	4.7 (3.7)
GAD change score, mean (SD)	-	6.3 (4.9)*	8.1 (4.9)*
PHQ change score, mean (SD)	-	6.4 (5.0)*	7.7 (5.3)*
No. (%) GAD remission	-		48.8
No. (%) PHQ remission	-		36.5

^*Paired-samples t tests indicated significant pre-post improvements in both GAD symptoms (all followed participants: mean decrease = 6.3 points, 95% CI [5.8, 6.9], t(310) = 22.77, p < .001, d = 1.29; clinical subsample: mean decrease = 8.1 points, 95% CI [7.4, 8.8], t(198) = 23.54, p < .001, d = 1.67) and PHQ-9 symptoms (all followed participants: mean decrease = 6.4 points, 95% CI [5.7, 7.1], t(211) = 18.63, p < .001, d = 1.28; clinical subsample: mean decrease = 7.7 points, 95% CI [6.8, 8.6], t(139) = 17.95, p < .001, d = 1.52)

BeHealth descriptives

Table 1 also shows descriptives of the interventions. The average length of treatment was around 5 months, with an average of 12.7 sessions received. Over 66% completed treatment.

Change in GAD

Across all participants, GAD scores decreased on average by 6.3 points (SD = 4.9; 95% CI [5.8, 6.9]) from baseline to last follow-up: t(310) = 22.77, p < .001, Cohen’s d = 1.29. Among both the full and clinical subsamples, final models revealed that baseline GAD-7 score and completed treatment were significant predictors of change in GAD-7 score (see Table 2). Parameter estimates were similar in strength and direction across the two samples, with findings from all participants slightly weaker. In the clinical sample, those who completed treatment had on average 1.81 points greater change in GAD scores than those who did not complete treatment. Number of sessions attended showed a trend-level association with change in GAD scores (b=0.06, SE=0.04, p < .07).

Table 2.

Predictors of change in GAD from baseline to last follow up, final model

	All Participants with follow up data N=311a			Baseline GAD 10 + and/or PHQ 10+ N=201b
	b	SE	p	b	SE	p
Baseline GAD	0.81	0.04	< 0.001	0.83	0.07	< 0.001
Completed Treatment	1.26	0.48	0.09	1.81	0.70	0.01

^aPatients scoring ≥ 5 on the GAD-7 at baseline

^bPatients scoring ≥ 5 on the GAD-7 at baseline who were ≥ 10 on the GAD-7 and/or PHQ-9 at baseline

Change in PHQ

Across all participants, PHQ scores decreased on average by 6.4 points (SD = 5.0; 95% CI [5.7, 7.1]) from baseline to last follow-up: t(211) = 18.63, p < .001, Cohen’s d = 1.28. Among both the full and clinical subsamples, final models revealed that baseline PHQ-9 score and completed treatment were significant predictors of change in PHQ-9 score (Table 3). Parameter estimates were similar in strength and direction across the two samples, with findings from all participants slightly weaker. In the clinical sample, those who completed treatment had on average 3.3 points greater change in PHQ score than those who did not complete treatment. Number of sessions attended showed a trend-level association with change in PHQ scores (b=0.08, SE=0.05, p < .07).

Table 3.

Predictors of change in PHQ from baseline to last follow up, final model

	All Participants (N = 310)a			Baseline GAD 10+ and/or PHQ 10+ (N = 201)
	b	SE	p	b	SE	p
Baseline PHQ	0.81	0.05	< 0.001	0.90	0.07	< 0.001
Completed treatment	1.79	0.61	0.004	3.34	0.88	< 0.001

^aPatients scoring ≥ 5 on the PHQ-9 at baseline

^bPatients scoring ≥ 5 on the PHQ-9 at baseline who were ≥ 10 on the GAD-7 and/or PHQ-9 at baseline

Remission of depression and/or anxiety

Just under half of the clinical sample remitted from anxiety, while just over a third remitted from depression (Table 1). Table 4 shows the final models for remitting to < 5 on either the GAD or the PHQ among the clinical sample. For each unit increase in baseline GAD, odds of remission of anxiety decreased by 9%. Those who completed treatment had almost twice the odds of remission in GAD. For each unit increase in baseline PHQ, odds of remission of depression decreased by 5%. Finally, those who completed treatment had about 7.5 times the odds of remission from depression compared to those who did not complete treatment. Number of sessions attended was not significantly associated with remission from anxiety but was associated with remission from depression (b=0.13, SE=0.06, p < .05).

Table 4.

Symptom remission at last follow up among those with scores ≥ 10 on gad or phq (n = 201), final models

	GAD Remission						PHQ Remission
	b	SE	p	OR	LL	UL	b	SE	p	OR	LL	UL
Baseline GAD	− 0.09	0.04	0.02	0.92	0.55	0.99	--	--	--	--	--	--
Baseline PHQ	--	--	--	--	--	--	− 0.06	0.04	0.18	0.95	0.86	1.02
Completed Treatment	0.68	0.40	0.009	1.97	0.91	4.41	2.01	0.66	0.002	7.45	2.31	33.4

Post hoc analyses

We compared the mean number of sessions among followed participants who completed versus those who did not complete treatment. Completers had a significantly higher mean number of sessions (M = 16.5, SD = 4.1) compared to who did not complete treatment (M = 8.0, SD = 4.0), t(346) = -15.1, p < .001. The association between session attendance and symptom change was weaker than that observed for treatment completion. To further probe these results, we examined the distribution of session attendance and found it to be positively skewed, with less than 4% of the sample attending fewer than 6 sessions.

In addition, to assess the potential impact of loss to follow-up on our findings, we conducted a conservative sensitivity analysis using a baseline carried forward (BLCF) approach, whereby participants without follow-up data were assumed to have experienced no change from baseline. For GAD, the mean difference was weakened but still statistically significant from baseline (M = 9.8, SD = 4.9) compared to the last observation (M = 5.7, SD = 4.5), t(474) = 18.1, p < .001, demonstrating a 4.1 (95% CI [3.7, 4.6]) lower mean difference. For PHQ, the mean difference was also weakened yet still statistically significant from baseline (M = 8.2, 5.7, SD = 4.1) compared to the last observation (M = 5.3, SD = 5.2), t(474) = -13.6, p < .001, demonstrating a 2.9 (95% CI [2.5, 3.3]) lower mean difference. Remission rates among the clinical sample using BLCF for those without follow up for GAD and PHQ were also lower at 39.5% (n = 220) and 29.1% (n = 203), respectively.

Discussion

Healthcare workers experience high rates of burnout, depression, and anxiety [1–3]. A large body of research has examined individual-level, non-clinical interventions such as mindfulness, peer support, and resilience training to address these problems, but far less is known about the delivery and effectiveness of traditional mental health treatment for this population [13, 15]. This cohort study examined outcomes of an evidence-based online psychotherapy program delivered to a large cohort of HCWs and their dependents. Online psychotherapy was one component of a broader employer-sponsored initative designed to improve access to mental health treatment within a large health system [27]. On average, participants were in their mid-thirties, predominantly female, and ethnically diverse. Participants who were followed reported substantial reductions in both anxiety and depressive symptoms, with mean improvements exceeding 6 points on the GAD-7 and PHQ-9. Approximately half of those with clinically significant baseline anxiety scores and more than one-third with clinically significant depression scores achieved remission. Treatment engagement was high, with participants attending an average of 15 sessions and more than 80% completing a full course of UP.

Treatment completion consistently predicted outcomes across analyses, suggesting that exposure to the intervention’s core therapeutic components and full treatment sequence was associated with greater clinical benefit. By contrast, session attendance showed a weaker association with outcomes, which may reflect restricted variability in session counts as well as the nonlinear nature of psychotherapy response—characterized by larger symptom gains early in treatment followed by diminishing returns with additional sessions [33, 34].

In an earlier study, Doukas and colleagues reported significant symptom reductions among healthcare workers who completed online CBT during the pandemic [35]. The current study examined outcomes in a substantially larger sample, included separate analyses of participants with clinically significant baseline symptoms of anxiety or depression, and found somewhat larger pre–post reductions in anxiety and depressive symptoms. Overall, these findings extend the work of Doukas and colleagues and provide stronger support that evidence-based psychotherapy can be effectively delivered to healthcare workers at scale within a real-world health system context. The magnitude of symptom improvement observed in this study was similar to that reported in meta-analyses of the UP, which have shown large within-group effect sizes for anxiety and depression [36, 37]. In addition, UP has been studied extensively in randomized controlled trials, but few studies have examined its implementation in real-world settings [38, 39]. To our knowledge, this is among the first studies to report outcomes of UP delivered online and at scale in an outpatient mental health setting.

As described in our earlier study, individuals seeking care through BeHealth were matched to treatment based on clinical need and preference [27]. Patients receiving online psychotherapy typically presented with mild to moderate anxiety or depressive symptoms. Individuals with more serious mental health problems or needing substance use disorder treatment were referred to care navigation services which matched patients to an appropriate community provider. In addition, some patients requested other forms of treatment such as in-person or long-term psychotherapy. As noted in our earlier study, about 45% of BeHealth patients received online psychotherapy while the remainder engaged in care navigation services. This approach reflects a system-level, stepped care model designed to match patients to appropriate levels of care while expanding access to evidence-based treatment.

The high levels of engagement and favorable clinical outcomes observed in this study may reflect key features of the model, including centralized digital access to care, rapid connection to clinicians, online psychotherapy, use of a dedicated service embedded within the health system, delivery by trained clinicians using structured evidence-based protocols, and alignment of treatment with patient preferences and clinical need.

Limitations

This study has several limitations. First, the observational design without a comparison group precludes causal inferences about treatment effects and leaves open alternative explanations for the observed improvements including regression to the mean, spontaneous remission and placebo response. Regression to the mean is a well-recognized concern in single-arm studies because participants often enter treatment during periods of elevated symptom severity [40]. Likewise, previous studies have demonstrated that placebo response [41] and spontaneous remission [42] can contribute to symptom reduction. These factors should therefore be considered when interpreting the present findings. Second, follow-up assessments were incomplete for a subset of participants; while those not followed did not differ significantly from the followed sample on baseline characteristics, they had markedly lower rates of treatment engagement (mean sessions = 7.3 vs. 14.9; treatment completion = 31.3% vs. 81.7%). These large differences in treatment engagement raise questions about whether symptom reduction findings in the followed sample generalize to those not followed. Third, outcomes were based on a limited battery of self-reported symptom measures, which may be subject to reporting bias. Fourth, although multiple strategies were used to enhance adherence to UP, validated fidelity measures typically used in randomized controlled trials were not collected.

Conclusions

Taken together, these findings address a critical gap identified in the healthcare worker mental health literature by demonstrating that evidence-based psychotherapy can be feasibly implemented and delivered to healthcare workers within a large health system. The program achieved high rates of engagement and treatment completion, and among those followed, outcomes comparable to those reported in randomized controlled trials of UP. This model provides a scalable approach to improving access to high-quality mental health care for healthcare workers and may inform the design of similar programs in other health systems.

Abbreviations

CBT

Cognitive-Behavioral Therapy

GAD-7

Generalized Anxiety Disorder–7

HCWs

Healthcare Workers

HIPAA

Health Insurance Portability and Accountability Act

PHQ-9

Patient Health Questionnaire–9

UP

Unifed Protocol for Transdiagnotic Treatment of Emotional Disorders

Author contributions

J.M.: Conceptualization; Methodology; Investigation; Writing – Original Draft; Writing – Review & Editing; Supervision; Funding Acquisition; Project Administration. S.L.: Conceptualization; Methodology; Investigation; Writing – Review & Editing; Supervision; Funding Acquisition; Project Administration. S.M.: Methodology; Data Curation; Formal Analysis; Investigation; Writing – Review & Editing. A.K.: Data Curation; Formal Analysis; Writing – Review & Editing. F.M.: Conceptualization; Methodology; Investigation; Writing – Review & Editing.L.K.: Investigation; Supervision; Writing – Review & Editing.C.B.: Investigation; Supervision. M.G.: Project Administration; Funding Acquisition; Supervision. T.E.: Methodology; Data Curation; Formal Analysis; Investigation; Writing – Review & Editing. A.B.: Writing – Review & Editing. M.K.: Investigation; Writing – Review & Editing. M.S.: Conceptualization; Investigation; Writing – Review & Editing; Supervision; Funding Acquisition; Project Administration.

Funding

This study was supported by internal funds from Northwell Health. The funder had no role in the design and conduct of the study, collection, management, analysis, or interpretation of the data.

Data availability

The datasets used and analyzed during the current study are available from the corresponding author, Dr. Jon Morgenstern (jmorgenste@northwell.edu), upon reasonable request.

Declarations

Ethics approval and consent to participate

This study was reviewed and approved by the Institutional Review Board at the Feinstein Institutes for Medical Research at Northwell Health (HSRD22-0289). The project met criteria for quality improvement and was not considered human subjects research; therefore, informed consent was not required.

Consent for publication

There are no personally identifiable data or information in the manuscript.

Competing interests

The authors declare no competing interests.