Electronic Health Record (EHR) data are critical for advancing translational research and AI technologies. The ENACT network offers access to structured EHR data across 57 CTSA hubs. However, substantial information is contained in clinical narratives, requiring natural language processing (NLP) for research. The ENACT NLP Working Group was formed to make NLP-derived clinical information accessible and queryable across the network.

We established the ENACT NLP Working Group with 13 sites selected based on criteria including clinical notes access, IT infrastructure, NLP expertise, and institutional support. We divided sites into five focus groups targeting clinical tasks within disease contexts. Each focus group consisted of two development sites and two validation sites. We extended the ENACT ontology to standardize NLP-derived data and conducted multisite evaluations using the Open Health Natural Language Processing (OHNLP) Toolkit.

The working group achieved 100% site retention and deployed NLP infrastructure across all sites. We developed and validated NLP algorithms for rare disease phenotyping, social determinants of health, opioid use disorder, sleep phenotyping, and delirium phenotyping. Performance varied across sites (F1 scores 0.53–0.96), highlighting data heterogeneity impacts. We extended the ENACT common data model and ontology to incorporate NLP-derived data while maintaining Shared Health Research Informatics NEtwork (SHRINE) compatibility.

This demonstrates feasibility of deploying NLP infrastructure across large, federated networks. The focus group approach proved more practical than general-purpose approaches. Key lessons include the challenge of data heterogeneity and importance of collaborative governance. This work also provides a foundation that other networks can build on to implement NLP capabilities for translational research.

Urgent care centers (UCCs) have reported high rates of antibiotic prescribing for acute respiratory tract infections. Prior UCC studies have generally been limited to single networks. Broadly generalizable stewardship efforts targeting common diagnoses are needed. This study examines the effectiveness of an antibiotic stewardship intervention in reducing inappropriate prescribing for bronchitis and viral upper respiratory tract infections (URTIs) in UCCs.

A quality improvement study comparing inappropriate antibiotic prescribing rates in UCCs after the introduction of an antibiotic stewardship intervention.

Forty-nine UCCs in 27 different networks from 18 states, including 1 telemedicine site.

Urgent care clinicians from a national collaborative of UCCs, all members of the Urgent Care Association.

The intervention included signing a commitment statement and selecting from 5 different intervention options during 3 plan-do-study-act cycles. The primary outcome was the percentage of urgent care encounters for viral URTIs or bronchitis with inappropriate prescribing, stratified by clinician engagement and diagnosis. A 3-month baseline and 9-month intervention period were compared using a regression model using a generalized estimating equation.

Among 15,385 encounters, the intervention was associated with decreases in inappropriate antibiotic prescribing for bronchitis (48% relative decrease, aOR = 0.52; 95% CI, 0.33–0.83) and viral URTIs (33%, aOR = 0.67; 95% CI, 0.55–0.82) among actively engaged clinicians compared to baseline. The intervention did not result in significant changes for clinicians not actively engaged.

This intervention was associated with reductions in inappropriate prescribing among actively engaged clinicians. Implementing stewardship interventions in UCCs may reduce inappropriate antibiotic prescriptions for common diagnoses; however, active clinician engagement may be necessary.

Successfully educating urgent care patients on appropriate use and risks of antibiotics can be challenging. We assessed the conscious and subconscious impact various educational materials (informational handout, priming poster, and commitment poster) had on urgent care patients’ knowledge and expectations regarding antibiotics.

Stratified Block Randomized Control Trial.

Urgent care centers (UCCs) in Colorado, Florida, Georgia, and New Jersey.

Urgent care patients.

We randomized 29 UCCs across six study arms to display specific educational materials (informational handout, priming poster, and commitment poster). The primary intention-to-treat (ITT) analysis evaluated whether the materials impacted patient knowledge or expectations of antibiotic prescribing by assigned study arm. The secondary as-treated analysis evaluated the same outcome comparing patients who recalled seeing the assigned educational material and patients who either did not recall seeing an assigned material or were in the control arm.

Twenty-seven centers returned 2,919 questionnaires across six study arms. Only 27.2% of participants in the intervention arms recalled seeing any educational materials. In our primary ITT analysis, no difference in knowledge or expectations of antibiotic prescribing was noted between groups. However, in the as-treated analysis, the handout and commitment poster were associated with higher antibiotic knowledge scores.

Educational materials in UCCs are associated with increased antibiotic-related knowledge among patients when they are seen and recalled; however, most patients do not recall passively displayed materials. More emphasis should be placed on creating and drawing attention to memorable patient educational materials.

In response to the COVID-19 pandemic, we rapidly implemented a plasma coordination center, within two months, to support transfusion for two outpatient randomized controlled trials. The center design was based on an investigational drug services model and a Food and Drug Administration-compliant database to manage blood product inventory and trial safety.

A core investigational team adapted a cloud-based platform to randomize patient assignments and track inventory distribution of control plasma and high-titer COVID-19 convalescent plasma of different blood groups from 29 donor collection centers directly to blood banks serving 26 transfusion sites.

We performed 1,351 transfusions in 16 months. The transparency of the digital inventory at each site was critical to facilitate qualification, randomization, and overnight shipments of blood group-compatible plasma for transfusions into trial participants. While inventory challenges were heightened with COVID-19 convalescent plasma, the cloud-based system, and the flexible approach of the plasma coordination center staff across the blood bank network enabled decentralized procurement and distribution of investigational products to maintain inventory thresholds and overcome local supply chain restraints at the sites.

The rapid creation of a plasma coordination center for outpatient transfusions is infrequent in the academic setting. Distributing more than 3,100 plasma units to blood banks charged with managing investigational inventory across the U.S. in a decentralized manner posed operational and regulatory challenges while providing opportunities for the plasma coordination center to contribute to research of global importance. This program can serve as a template in subsequent public health emergencies.

To measure SARS-CoV-2 anti-nucleocapsid (anti-N) antibody seropositivity among healthcare personnel (HCP) without a history of COVID-19 and to identify HCP characteristics associated with seropositivity.

Prospective cohort study from September 22, 2020, to March 3, 2022.

A tertiary care academic medical center.

727 HCP without prior positive SARS-CoV-2 PCR testing were enrolled; 559 HCP successfully completed follow-up.

At enrollment and follow-up 1–6 months later, HCP underwent SARS-CoV-2 anti-N testing and were surveyed on demographics, employment information, vaccination status, and COVID-19 symptoms and exposures.

Of 727 HCP enrolled, 27 (3.7%) had a positive SARS-CoV-2 anti-N test at enrollment. Seropositive HCPs were more likely to have a household exposure to COVID-19 in the past 30 days (OR 7.92, 95% CI 2.44–25.73), to have had an illness thought to be COVID-19 (4.31, 1.94–9.57), or to work with COVID-19 patients more than half the time (2.09, 0.94–4.77). Among 559 HCP who followed-up, 52 (9.3%) had a positive SARS-CoV-2 anti-N antibody test result. Seropositivity at follow-up was associated with community/household exposures to COVID-19 within the past 30 days (9.50, 5.02–17.96; 2.90, 1.31–6.44), having an illness thought to be COVID-19 (8.24, 4.44–15.29), and working with COVID-19 patients more than half the time (1.50, 0.80–2.78).

Among HCP without prior positive SARS-CoV-2 testing, SARS-CoV-2 anti-N seropositivity was comparable to that of the general population and was associated with COVID-19 symptomatology and both occupational and non-occupational exposures to COVID-19.

To determine the prevalence of severe acute respiratory coronavirus virus 2 (SARS-CoV-2) IgG nucleocapsid (N) antibodies among healthcare personnel (HCP) with no prior history of COVID-19 and to identify factors associated with seropositivity.

Prospective cohort study.

An academic, tertiary-care hospital in St. Louis, Missouri.

The study included 400 HCP aged ≥18 years who potentially worked with coronavirus disease 2019 (COVID-19) patients and had no known history of COVID-19; 309 of these HCP also completed a follow-up visit 70–160 days after enrollment. Enrollment visits took place between September and December 2020. Follow-up visits took place between December 2020 and April 2021.

At each study visit, participants underwent SARS-CoV-2 IgG N-antibody testing using the Abbott SARS-CoV-2 IgG assay and completed a survey providing information about demographics, job characteristics, comorbidities, symptoms, and potential SARS-CoV-2 exposures.

Participants were predominately women (64%) and white (79%), with median age of 34.5 years (interquartile range [IQR], 30–45). Among the 400 HCP, 18 (4.5%) were seropositive for IgG N-antibodies at enrollment. Also, 34 (11.0%) of 309 were seropositive at follow-up. HCP who reported having a household contact with COVID-19 had greater likelihood of seropositivity at both enrollment and at follow-up.

In this cohort of HCP during the first wave of the COVID-19 pandemic, ∼1 in 20 had serological evidence of prior, undocumented SARS-CoV-2 infection at enrollment. Having a household contact with COVID-19 was associated with seropositivity.

Obstructive sleep apnea (OSA) is a sleep disorder that is highly comorbid with psychiatric disorders, including depression and anxiety. Excessive daytime sleepiness (EDS) is common in psychiatric disorders and OSA. In participants with OSA, EDS can persist despite use of positive airway pressure (PAP) therapy. This analysis of real-world data aimed to describe EDS and its relationship with PAP use in participants with and without depression.

US residents (≥18 years of age, self-reported physician diagnosis of OSA [from 1/1/2015 to 3/31/2020]) completed a survey in Evidation Health’s Achievement app assessing subjective levels of sleepiness (Epworth Sleepiness Scale [ESS]) and self-reported PAP usage, categorized as nonuse (no PAP use), nonadherent (<4 h/night or <5 d/wk), intermediate (4-6 h/night, ≥5 d/wk), or highly adherent (≥6 h/night, ≥5 d/wk). ESS score >10 defined EDS. A linear model assessed relationships between PAP use and ESS score. P-values are uncontrolled for multiplicity (nominal).

In total, 2289 participants (EDS, n=972; no EDS, n=1317) completed the survey (50.3% female; 82.5% White; mean±standard deviation [SD] age, 44.8 ± 11.1 years). Anxiety and depression were the most common comorbidities and were more common in participants with EDS (49% and 49%, respectively) than those without EDS (41% and 37%, respectively). Overall, EDS was more common among participants with comorbid depression (49%) than those without (38%), even among highly adherent PAP users (46% vs 30%, respectively). In a linear model (PAP users only), an additional 1 h/night of PAP use was associated with lower ESS scores in the subgroup of participants without depression (n=928; estimate [SE], −0.42 [0.09]; P<0.05), but not in the subgroup with depression (n=661; estimate [SE], −0.15 [0.10]; P>0.05). In a sensitivity analysis that excluded participants using medications that cause sleepiness, PAP use was associated with lower ESS scores regardless of depression status; however, EDS remained more common in participants with comorbid depression (46%) than in those without (36%).

In this real-world population of participants with OSA, those with EDS were more likely to have comorbid anxiety or depression. EDS was more common in participants with comorbid depression than those without, even with highly adherent PAP use. PAP use was associated with lower ESS scores in participants without comorbid depression, but not in those with comorbid depression; the use of medications that cause sleepiness may contribute to but does not fully explain this phenomenon.

Axsome Therapeutics and Jazz Pharmaceuticals

Obstructive sleep apnea (OSA) is a highly prevalent sleep disorder that is often associated with numerous medical and psychiatric comorbidities. Patients with OSA experience a variety of symptoms that can be burdensome and affect their quality of life and satisfaction with care. Excessive daytime sleepiness (EDS) is a common symptom of OSA, and can persist despite primary airway therapy (e.g., positive airway pressure [PAP]). This analysis aimed to characterize common comorbidities, as well as symptoms present at OSA diagnosis and their burden in a real-world population of participants with OSA.

US residents (≥18 years of age, self-reported clinician diagnosis of OSA [from 1/1/2015 to 3/31/2020]) completed a survey in Evidation Health’s Achievement app that assessed self-reported sleepiness (Epworth Sleepiness Scale [ESS]), self-reported PAP usage, self-reported physician-diagnosed comorbidities, and information on their symptoms at time of OSA diagnosis. Self-reported PAP use was categorized as nonuse (no PAP use), nonadherent (<4 h/night or <5 d/wk), intermediate (4–6 h/night, ≥5 d/wk), or highly adherent (≥6 h/night, ≥5 d/wk). EDS was defined as ESS score >10. All data were summarized descriptively.

In total, 2289 participants completed the survey (50.3% female; 82.5% White; mean ± standard deviation [SD] age, 44.8 ± 11.1 years; mean ± SD age at OSA diagnosis, 40.7 ± 11.4 years; mean ± SD body mass index, 35.4 ± 8.7 kg/m2); 42.5% had EDS. Among the total population, 30.6% were PAP non-users, 6.7% were nonadherent, 9.8% were intermediate adherent, and 52.9% were highly adherent. Across the study population, the most common self-reported physician-diagnosed comorbidities were anxiety (44%) and depression (42%) followed by hypertension (39%), dyslipidemia (26%), and asthma (21%). Among the symptoms participants reported having had at the time of OSA diagnosis, the most common were EDS (79%), fatigue (79%), snoring (75%), and awakening with a dry mouth or sore throat (63%). Concentration/Memory problems (48%) and mood changes (46%) were also common. In the overall population, the symptoms present at the time of OSA diagnosis that were most likely to be highly burdensome were fatigue (53%), EDS (46%), snoring (35%), difficulty concentrating/memory issues (31%), and mood changes (25%).

These real-world survey data identify anxiety and depression as the most frequently reported comorbidities in a population of participants with OSA, each affecting over 40% of participants. In addition to classic OSA symptoms (e.g., EDS, fatigue, snoring, and awakening with dry mouth/sore throat), concentration/memory problems and mood changes were also common at the time of OSA diagnosis and were among the presenting symptoms most frequently reported as highly burdensome, along with fatigue, EDS, and snoring.

Axsome Therapeutics and Jazz Pharmaceuticals

Dental healthcare personnel (DHCP) are at high risk of exposure to coronavirus disease 2019 (COVID-19). We sought to identify how DHCP changed their use of personal protective equipment (PPE) as a result of the COVID-19 pandemic, and to pilot an educational video designed to improve knowledge of proper PPE use.

The study comprised 2 sets of semistructured qualitative interviews.

The study was conducted in 8 dental clinics in a Midwestern metropolitan area.

In total, 70 DHCP participated in the first set of interviews; 63 DHCP participated in the second set of interviews.

In September–November 2020 and March–October 2021, we conducted 2 sets of semistructured interviews: (1) PPE use in the dental community during COVID-19, and (2) feedback on the utility of an educational donning and doffing video.

Overall, 86% of DHCP reported having prior training. DHCP increased the use of PPE during COVID-19, specifically N95 respirators and face shields. DHCP reported real-world challenges to applying infection control methods, often resulting in PPE modification and reuse. DHCP reported double masking and sterilization methods to extend N95 respirator use. Additional challenges to PPE included shortages, comfort or discomfort, and compatibility with specialty dental equipment. DHCP found the educational video helpful and relevant to clinical practice. Fewer than half of DHCP reported exposure to a similar video.

DHCP experienced significant challenges related to PPE access and routine use in dental clinics during the COVID-19 pandemic. An educational video improved awareness and uptake of appropriate PPE use among DHCP.

Attention deficit-hyperactivity disorder (ADHD) is related to depressive disorder, and adolescents with both present poor outcomes. However, evidence for the safety of concomitantly using a methylphenidate (MPH) and a selective serotonin reuptake inhibitor (SSRI) among adolescent ADHD patients is limited, a literature gap aimed to address through this investigation.

We conducted a new-user cohort study using a nationwide claims database in South Korea. We identified a study population as adolescents who were diagnosed both ADHD and depressive disorder. MPH-only users were compared with patients who prescribed both a SSRI and a MPH. Fluoxetine and escitalopram users were also compared to find a preferable treatment option. Thirteen outcomes including neuropsychiatric, gastrointestinal, and other events were assessed, taking respiratory tract infection as a negative control outcome. We matched the study groups using a propensity score and used the Cox proportional hazard model to calculate the hazard ratio. Subgroup and sensitivity analyses were conducted in various epidemiologic settings.

The risks of all the outcomes between the MPH-only and SSRI groups were not significantly different. Regarding SSRI ingredients, the risk of tic disorder was significantly lower in the fluoxetine group than the escitalopram group [HR 0.43 (0.25–0.71)]. However, there was no significant difference in other outcomes between the fluoxetine and escitalopram groups.

The concomitant use of MPHs and SSRIs showed generally safe profiles in adolescent ADHD patients with depression. Most of the differences between fluoxetine and escitalopram, except those concerning tic disorder, were not significant.

Subthreshold/attenuated syndromes are established precursors of full-threshold mood and psychotic disorders. Less is known about the individual symptoms that may precede the development of subthreshold syndromes and associated social/functional outcomes among emerging adults.

We modeled two dynamic Bayesian networks (DBN) to investigate associations among self-rated phenomenology and personal/lifestyle factors (role impairment, low social support, and alcohol and substance use) across the 19Up and 25Up waves of the Brisbane Longitudinal Twin Study. We examined whether symptoms and personal/lifestyle factors at 19Up were associated with (a) themselves or different items at 25Up, and (b) onset of a depression-like, hypo-manic-like, or psychotic-like subthreshold syndrome (STS) at 25Up.

The first DBN identified 11 items that when endorsed at 19Up were more likely to be reendorsed at 25Up (e.g., hypersomnia, impaired concentration, impaired sleep quality) and seven items that when endorsed at 19Up were associated with different items being endorsed at 25Up (e.g., earlier fatigue and later role impairment; earlier anergia and later somatic pain). In the second DBN, no arcs met our a priori threshold for inclusion. In an exploratory model with no threshold, >20 items at 19Up were associated with progression to an STS at 25Up (with lower statistical confidence); the top five arcs were: feeling threatened by others and a later psychotic-like STS; increased activity and a later hypo-manic-like STS; and anergia, impaired sleep quality, and/or hypersomnia and a later depression-like STS.

These probabilistic models identify symptoms and personal/lifestyle factors that might prove useful targets for indicated preventative strategies.