The race-based stress reduction intervention (RiSE) study on African American women in NYC and Chicago: Design and methods for complex genomic analysis

RiSE study aims to evaluate a race-based stress-reduction intervention as an effective strategy to improve coping and decrease stress-related symptoms, inflammatory burden, and modify DNA methylation of stress response-related genes in older AA women. This article will describe genomic analytic methods to be utilized in this longitudinal, randomized clinical trial of older adult AA women in Chicago and NYC that examines the effect of the RiSE intervention on DNAm pre- and post-intervention, and its overall influence on inflammatory burden. Salivary DNAm will be measured at baseline and 6 months following the intervention, using the Oragene-DNA kit. Measures of perceived stress, depressive symptoms, fatigue, sleep, inflammatory burden, and coping strategies will be assessed at 4 time points including at baseline, 4 weeks, 8 weeks, and 6 months. Genomic data analysis will include the use of pre-processed and quality-controlled methylation data expressed as beta (β) values. Association analyses will be performed to detect differentially methylated sites on the targeted candidate genes between the intervention and non-intervention groups using the Δβ (changes in methylation) with adjustment for age, health behaviors, early life adversity, hybridization batch, and top principal components of the probes as covariates. To account for multiple testing, we will use FDR adjustment with a corrected p-value of <0.05 regarded as statistically significant. To assess the relationship between inflammatory burden and Δβ among the study samples, we will repeat association analyses with the inclusion of individual inflammation protein measures. ANCOVA will be used because it is more statistically powerful to detect differences.

Diastolic dysfunction and clinical outcomes after transcatheter or surgical aortic valve replacement in patients with atypical aortic valve stenosis

BACKGROUND

Diastolic dysfunction is a predictor of poor outcomes in many cardiovascular conditions. At present, it is unclear whether diastolic dysfunction predicts adverse outcomes in patients with atypical aortic stenosis who undergo aortic valve replacement (AVR).

METHODS

Five hundred and twenty-three patients who underwent transcatheter AVR (TAVR) (n = 303) and surgical AVR (SAVR) (n = 220) at a single institution were included in our analysis. Baseline left and right heart invasive hemodynamics were assessed. Baseline transthoracic echocardiograms were reviewed to determine aortic stenosis subtype and parameters of diastolic dysfunction. Aortic stenosis subtype was categorized as typical (normal flow, high-gradient) aortic stenosis, classical, low-flow, low-gradient (cLFLG) aortic stenosis, and paradoxical, low-flow, low-gradient (pLFLG) aortic stenosis. Cox proportional hazard models were utilized to examine the relation between invasive hemodynamic or echocardiographic variables of diastolic dysfunction, aortic stenosis subtype, and all-cause mortality. Propensity-score analysis was performed to study the relation between aortic stenosis subtype and the composite outcome [death/cerebrovascular accident (CVA)].

RESULTS

The median STS risk was 5.3 and 2.5% for TAVR and SAVR patients, respectively. Relative to patients with typical aortic stenosis, patients with atypical (cLFLG and pLFLG) aortic stenosis displayed a significantly higher prevalence of diastolic dysfunction (LVEDP ≥ 20mmHg, PCWP ≥ 20mmHg, echo grade II or III diastolic dysfunction, and echo-PCWP ≥ 20mmHg) and, independently of AVR treatment modality, had a significantly increased risk of death. In propensity-score analysis, patients with atypical aortic stenosis had higher rates of death/CVA than typical aortic stenosis patients, independently of diastolic dysfunction and AVR treatment modality.

CONCLUSION

We demonstrate the novel observation that compared with patients with typical aortic stenosis, patients with atypical aortic stenosis have a higher burden of diastolic dysfunction. We corroborate the worse outcomes previously reported in atypical versus typical aortic stenosis and demonstrate, for the first time, that this observation is independent of AVR treatment modality. Furthermore, the presence of diastolic dysfunction does not independently predict outcome in atypical aortic stenosis regardless of treatment type, suggesting that other factors are responsible for adverse clinical outcomes in this higher risk cohort.

Is There a Plateau to the Learning Curve for Acoustic Neuroma Resection?-Experience and Outcomes from a Single Interdisciplinary Team Over Thirty Years

Objective  The evolution of acoustic neuroma (AN) care continues to shift focus on balancing optimized tumor resection and control with preservation of neurological function. Prior learning curve analyses of AN resection have demonstrated a plateau between 20 and 100 surgeries. In this study of 860 consecutive AN surgeries, we investigate the presence of an extended learning curve tail for AN resection. Methods  A retrospective cohort study of AN resections by a single interdisciplinary team between 1988 and 2018 was performed. Proportional odds models and restricted cubic splines were used to determine the association between the timing of surgery and odds of improved postoperative outcomes. Results  The likelihood of improved postoperative House-Brackmann (HB) scores increased in the first 400 procedures, with HB 1 at 36% in 1988 compared with 79% in 2004. While the probability of a better HB score increased over time, there was a temporary decrease in slope of the cubic spline between 2005 and 2009. The last 400 cases continued to see improvement in optimal HB outcomes: adjusted odds of HB 1 score were twofold higher in both 2005 to 2009 (adjusted odds ratio [aOR]: 2.11, 95% confidence interval [CI]: 1.38-3.22, p  < 0.001) and 2010 to 2018 (aOR: 2.18, 95% CI: 1.49-3.19, p  < 0.001). Conclusion  In contrast to prior studies, our study demonstrates the steepest growth for learning, as measured by rates of preservation of facial function outcomes (HB 1), occurs in the first 400 AN resections. Additionally, improvements in patient outcomes continued even 30 years into practice, underlining the importance of lifelong learning.

Direct Oral Anticoagulants in Patients With Cardiac Amyloidosis: A Systematic Review and Meta-Analysis

Background and Objectives

Atrial fibrillation is common in patients with cardiac amyloidosis. However, the optimal anticoagulation strategy to prevent thromboembolic events in patients with cardiac amyloidosis and atrial fibrillation is unknown. This systematic review and meta-analysis compares direct oral anticoagulants (DOACs) vs. vitamin K antagonists (VKAs) in patients with cardiac amyloidosis and atrial fibrillation.

Methods

We performed a systematic literature review to identify clinical studies of anticoagulation therapies for patients with cardiac amyloidosis and atrial fibrillation. The primary outcomes of major bleeding and thrombotic events were reported using random effects risk ratios (RRs) with 95% confidence interval (CI).

Results

Our search yielded 97 potential studies and evaluated 14 full-text articles based on title and abstract. We excluded 10 studies that were review articles or did not compare anticoagulation. We included 4 studies reporting on 1,579 patients. The pooled estimates are likely underpowered due to small sample sizes. There was no difference in bleeding events for patients with cardiac amyloidosis and atrial fibrillation treated with DOACs compared to VKAs with a RR of 0.64 (95% CI, 0.38-1.10; p=0.10). There were decreased thrombotic events for patients with cardiac amyloidosis and atrial fibrillation treated with DOACs compared to VKAs with a RR of 0.50 (95% CI, 0.32-0.79; p=0.003).

Conclusions

This systematic review and meta-analysis suggests that DOACs are as safe and effective as VKAs in patients with cardiac amyloidosis and atrial fibrillation. However, more data are needed to investigate clinical differences in anticoagulation therapy in this patient population.

CT Findings in Laryngeal Trauma and the Clinical Implications

PURPOSE

Acute traumatic osseous and cartilaginous injuries to the larynx are rare injuries presenting to the emergency department. Despite the low reported incidence, laryngeal trauma carries a high morbidity and mortality. The purpose of this study is to identify fracture and soft tissue injury patterns in laryngeal trauma and explore associations with patient demographics, mechanisms of injury, urgent airway and surgical intervention.

METHODS

A retrospective review of patients with laryngeal injury who underwent multidetector computed tomography (MDCT) imaging was performed. The CT findings of laryngeal and hyoid fracture location, fracture displacement, and soft tissue injuries were recorded. Clinical data including patient demographics, mechanisms of injury, frequency of airway and surgical intervention were also recorded. Correlation of imaging characteristics with patient demographics, mechanism of injury and interventions were assessed for statistical significance using χ2 and Fisher's exact tests.

RESULTS

The median patient age was 40 years old with a strong male predominance. The most common mechanisms of injury included motor vehicle collisions and penetrating gunshot wounds. Thyroid cartilage fractures were the most common fracture type. Findings of fracture displacement and airway hematoma had a higher correlation with requiring urgent airway management.

CONCLUSION

Radiologists' early recognition and prompt communication of laryngeal trauma to the clinical service is important to reduce associated morbidity and mortality. Displaced fractures and laryngeal hematomas should be promptly conveyed to the clinical service as they are associated with more complex injuries and higher rates of urgent airway management and surgical intervention.

Risk Factors for CMV and BK Infections in an Elderly Veteran Population Following Kidney Transplantation: Implications for Immunosuppression Induction and Management

Cytomegalovirus (CMV) and BK Polyomavirus (BKPyV) are the most common opportunistic pathogens following kidney transplantation. We evaluated 102 patients with a median age of 63 at Edward Hines VA Hospital from November 2020 to December 2022. Our primary interest was the incidence of CMV and BKPyV infections, as well as CMV and BKPyV coinfection. Secondary interests included time to infection, rejection, and graft and patient survival. There were no statistically significant differences in patient age, donor age, race, transplant type, incidence of delayed graft function, or induction in both cohorts (any infection (N = 46) vs. those without (N = 56)). There was a 36% (37/102) incidence of CMV, a 17.6% (18/102) of BKPyV and an 8.8% (9/102) incidence of coinfection. There was a decreased incidence of CMV infection in Basiliximab induction versus antithymocyte globulin (21% and 43%). CMV risk status had no effect on the incidence of CMV infection following transplant. African American recipients had a lower incidence of BKPyV infection (12% vs. 39%), yet a higher incidence was observed in those with high cPRA (50% vs. 14%). Most CMV and/or BKPyV infections occurred within the first six months post-transplant (54%). Immunosuppression management of the elderly should continually be evaluated to reduce opportunistic infections post-transplant.

Oxygen saturation during donor warm ischemia time and outcome of donation after circulatory death (DCD) liver transplantation with static cold storage: A review of 1114 cases

The donor operation and the hemodynamics during declaration resulting in donor warm ischemia time have been linked to the outcomes in donation after circulatory death (DCD) liver transplantation (LT). Scrutiny of the donor hemodynamics at the time of withdrawal of life support concluded that a functional donor warm ischemia time may be associated with LT graft failure. Unfortunately, the definition for functional donor warm ischemia time has not reached a consensus-but has almost always incorporated time spent in a hypoxic state. Herein, we reviewed 1114 DCD LT cases performed at the 20 highest volume centers during 2014 and 2018. Donor hypoxia began within 3 minutes of withdrawal of life support for 60% of cases and within 10 minutes for 95% of cases. Graft survival was 88.3% at 1 year and 80.3% at 3 years. Scrutinizing the time spent under hypoxic conditions (oxygen saturation ≤ 80%) during the withdrawal of life support, we found an increasing risk of graft failure as hypoxic time increased from 0 to 16 minutes. After 16 minutes and up to 50 minutes, we did not find any increased risk of graft failure. In conclusion, after 16 minutes of time in hypoxia, the risk of graft failure in DCD LT did not increase. The current evidence suggests that an over-reliance on hypoxia time may lead to an unnecessary increase in DCD liver discard and may not be as useful for predicting graft loss after LT.

Long-Term Outcomes of Cardiac Sarcoid: Prognostic Implications of Isolated Cardiac Involvement and Impact of Diagnostic Delays

Background Isolated cardiac sarcoid (iCS) is reported to have more severe clinical presentation and greater risk of adverse events compared with cardiac sarcoid (CS) with extracardiac involvement (nonisolated CS). Delays in diagnosing specific organ involvement may play a role in these described differences. Methods and Results A retrospective observational study of patients with CS over a 20-year period was conducted. Objective evidence of organ involvement and time of onset based on consensus criteria were identified. CS was confirmed by histology in all patients from myocardium only (iCS) or extracardiac tissue (nonisolated CS). The primary end point was a composite of mortality, orthotopic heart transplant, and durable left ventricular assist device implantation. CS was isolated in 9 of 50 patients (18%). Among baseline characteristics, iCS and nonisolated CS differed significantly only in the frequency of sustained ventricular tachycardia at presentation (78% versus 37%; P=0.03) and delay in CS diagnosis >6 months (67% versus 5%; P<0.01). A nonsignificant trend toward lower left ventricular ejection fraction and more frequent heart failure in iCS was observed. Over a median follow-up of 9.7 years (95% CI, 6.8-10.8), 18 patients reached the primary end point (13 deaths, 2 orthotopic heart transplants, and 3 durable left ventricular assist device implantations). The 1-, 5-, and 10-year event-free survival rates were 96% (95% CI, 85%-99%), 79% (95% CI, 64%-88%), and 58% (95% CI, 40%-73%), respectively, without differences between groups. There were no significant predictors of the primary end point, including delayed CS diagnosis. Conclusions Long-term outcomes were similar between iCS and nonisolated CS in patients with histologically documented sarcoid. Diagnostic delays may contribute to differences in the dominant clinical presentation, despite similar outcomes.

Applying the Global Leadership Initiative on Malnutrition criteria in patients admitted with SARS-CoV-2 infection using computed tomography imaging

BACKGROUND

Patients with low muscle mass and acute SARS-CoV-2 infection meet the Global Leadership Initiative on Malnutrition (GLIM) etiologic and phenotypic criteria to diagnose malnutrition, respectively. However, available cut-points to classify individuals with low muscle mass are not straightforward. Using computed tomography (CT) to determine low muscularity, we assessed the prevalence of malnutrition using the GLIM framework and associations with clinical outcomes.

METHODS

A retrospective cohort was conducted gathering patient data from various clinical resources. Patients admitted to the COVID-19 unit (March 2020 to June 2020) with appropriate/evaluable CT studies (chest or abdomen/pelvis) within the first 5 days of admission were considered eligible. Sex- and vertebral-specific skeletal muscle indices (SMI; cm2 /m2 ) from healthy controls were used to determine low muscle mass. Injury-adjusted SMI were derived, extrapolated from cancer cut-points and explored. Descriptive statistics and mediation analyses were completed.

RESULTS

Patients (n = 141) were 58.2 years of age and racially diverse. Obesity (46%), diabetes (40%), and cardiovascular disease (68%) were prevalent. Using healthy controls and injury-adjusted SMI, malnutrition prevalence was 26% (n = 36/141) and 50% (n = 71/141), respectively. Mediation analyses demonstrated a significant reduction in the effect of malnutrition on outcomes in the presence of Acute Physiology and Chronic Health Evaluation II, supporting the mediating effects of severity of illness intensive care unit (ICU) admission, ICU length of stay, mechanical ventilation, complex respiratory support, discharge status (all P values = 0.03), and 28-day mortality (P = 0.04).

CONCLUSIONS

Future studies involving the GLIM criteria should consider these collective findings in their design, analyses, and implementation.

An objective assessment of the impact of tendon retraction on sleep efficiency in patients with full-thickness rotator cuff tears: a prospective cohort study

BACKGROUND

Sleep quality, quantity, and efficiency have all been demonstrated to be adversely affected by rotator cuff pathology. Previous measures of assessing the impact of rotator cuff pathology on sleep have been largely subjective in nature. This study was undertaken to objectively analyze this relationship through the use of activity monitors.

METHODS

Patients with full-thickness rotator cuff tears at a single institution were prospectively enrolled between 2018 and 2020. Waistworn accelerometers were provided for the patients to use each night for 14 days. Sleep efficiency was calculated using the ratio of the time spent sleeping to the total amount of time that was spent in bed. Retraction of the rotator cuff tear was classified using the Patte staging system.

RESULTS

This study included 36 patients: 18 with Patte stage 1 disease, 14 with Patte stage 2 disease, and 4 patients with Patte stage 3 disease. During the study, 25 participants wore the monitor on multiple nights, and ultimately their data was used for the analysis. No difference in the median sleep efficiency was appreciated amongst these groups (P&gt;0.1), with each cohort of patients demonstrating a generally high sleep efficiency.

CONCLUSIONS

The severity of retraction of the rotator cuff tear did not appear to correlate with changes in sleep efficiency for patients (P&gt;0.1). These findings can better inform providers on how to counsel their patients who present with complaints of poor sleep in the setting of full-thickness rotator cuff tears. Level of evidence: Level II.

Reported HIV-related stigma according to race and ethnicity

People living with HIV/AIDS (PLWHA) have long experienced structural, community, and personal stigma. We explored differences in experienced HIV-related stigma according to race/ethnicity using quantitative and qualitative measures. Sixty-four patients were enrolled in this study (22 White and 42 people of color [POC]). POC scored higher than White PLWHA on all 12 survey statements, with statistically significant differences in disclosure concerns and with one of the statements on public attitudes towards PLWHA. Common themes in the qualitative interview were HIV disclosure concerns and fear of rejection. These data demonstrate that stigma continues to be a significant concern for PLWHA, particularly POC, meaningfully impacting their lives. By acknowledging and working to reduce negative perceptions about PLWHA, physicians may improve care for their patients by developing more trusting relationships.

Binge drinking alcohol and circadian misalignment in night shift nurses is associated with decreased resiliency to COVID-19 infection

BACKGROUND

Nurses and other first responders are at high risk of exposure to the SARS-CoV2 virus, and many have developed severe COVID-19 infection. A better understanding of the factors that increase the risk of infection after exposure to the virus could help to address this. Although several risk factors such as obesity, diabetes, and hypertension have been associated with an increased risk of infection, many first responders develop severe COVID-19 without established risk factors. As inflammation and cytokine storm are the primary mechanisms in severe COVID-19, other factors that promote an inflammatory state could increase the risk of COVID-19 in exposed individuals. Alcohol misuse and shift work with subsequent misaligned circadian rhythms are known to promote a pro-inflammatory state and thus could increase susceptibility to COVID-19. To test this hypothesis, we conducted a prospective, cross-sectional observational survey-based study in nurses using the American Nursing Association network.

METHOD

We used validated structured questionnaires to assess alcohol consumption (the Alcohol Use Disorders Identification Test) and circadian typology or chronotype (the Munich Chronotype Questionnaire Shift -MCTQ-Shift).

RESULTS

By latent class analysis (LCA), high-risk features of alcohol misuse were associated with a later chronotype, and binge drinking was greater in night shift workers. The night shift was associated with more than double the odds of COVID-19 infection of the standard shift (OR 2.67, 95% CI: 1.18 to 6.07). Binge drinkers had twice the odds of COVID-19 infection of those with low-risk features by LCA (OR: 2.08, 95% CI: 0.75 to 5.79).

CONCLUSION

Working night shifts or binge drinking may be risk factors for COVID-19 infection among nurses. Understanding the mechanisms underlying these risk factors could help to mitigate the impact of COVID-19 on our at-risk healthcare workforce.

Deployment of Real-time Natural Language Processing and Deep Learning Clinical Decision Support in the Electronic Health Record: Pipeline Implementation for an Opioid Misuse Screener in Hospitalized Adults

BACKGROUND

The clinical narrative in electronic health records (EHRs) carries valuable information for predictive analytics; however, its free-text form is difficult to mine and analyze for clinical decision support (CDS). Large-scale clinical natural language processing (NLP) pipelines have focused on data warehouse applications for retrospective research efforts. There remains a paucity of evidence for implementing NLP pipelines at the bedside for health care delivery.

OBJECTIVE

We aimed to detail a hospital-wide, operational pipeline to implement a real-time NLP-driven CDS tool and describe a protocol for an implementation framework with a user-centered design of the CDS tool.

METHODS

The pipeline integrated a previously trained open-source convolutional neural network model for screening opioid misuse that leveraged EHR notes mapped to standardized medical vocabularies in the Unified Medical Language System. A sample of 100 adult encounters were reviewed by a physician informaticist for silent testing of the deep learning algorithm before deployment. An end user interview survey was developed to examine the user acceptability of a best practice alert (BPA) to provide the screening results with recommendations. The planned implementation also included a human-centered design with user feedback on the BPA, an implementation framework with cost-effectiveness, and a noninferiority patient outcome analysis plan.

RESULTS

The pipeline was a reproducible workflow with a shared pseudocode for a cloud service to ingest, process, and store clinical notes as Health Level 7 messages from a major EHR vendor in an elastic cloud computing environment. Feature engineering of the notes used an open-source NLP engine, and the features were fed into the deep learning algorithm, with the results returned as a BPA in the EHR. On-site silent testing of the deep learning algorithm demonstrated a sensitivity of 93% (95% CI 66%-99%) and specificity of 92% (95% CI 84%-96%), similar to published validation studies. Before deployment, approvals were received across hospital committees for inpatient operations. Five interviews were conducted; they informed the development of an educational flyer and further modified the BPA to exclude certain patients and allow the refusal of recommendations. The longest delay in pipeline development was because of cybersecurity approvals, especially because of the exchange of protected health information between the Microsoft (Microsoft Corp) and Epic (Epic Systems Corp) cloud vendors. In silent testing, the resultant pipeline provided a BPA to the bedside within minutes of a provider entering a note in the EHR.

CONCLUSIONS

The components of the real-time NLP pipeline were detailed with open-source tools and pseudocode for other health systems to benchmark. The deployment of medical artificial intelligence systems in routine clinical care presents an important yet unfulfilled opportunity, and our protocol aimed to close the gap in the implementation of artificial intelligence-driven CDS.

TRIAL REGISTRATION

ClinicalTrials.gov NCT05745480; https://www.clinicaltrials.gov/ct2/show/NCT05745480.

Reducing Disparities in Postpartum Care Utilization: Development of a Clinical Risk Assessment Tool

INTRODUCTION

Patients receiving care at Federally Qualified Health Centers (FQHCs) have low postpartum care attendance. Perinatal morbidity and mortality disproportionately affect patients with low-income and are potentially preventable. The purpose of this study was to develop a clinical decision support tool to identify FQHC patients less likely to return for postpartum care. To accomplish this purpose, we evaluated established predictors and novel risk factors in our patient population.

METHODS

This is a retrospective, secondary data analysis of 50,022 patients who received prenatal care past 24 weeks' gestation in FQHCs between 2012 and 2017. The postpartum visit was defined using Healthcare Effectiveness Data and Information Set measures as early care (birth to 21 days) and later care (21-84 days). Anderson's Behavioral Model for Access to Healthcare guided inclusion of potentially predictive factors. We stratified data by postpartum care attendance, and a final predictive model was selected by model fit statistics and clinical relevance.

RESULTS

In our sample, 64% of birthing persons attended postpartum care at FQHCs. Of those who returned for care, 38% returned within 21 days postbirth and 62% returned between 21 and 84 days, with 28% returning for both early and later care. Predictors for postpartum care attendance included maternal age, parity, gestational age at first visit, and number of prenatal care visits. A clinical decision support tool for identifying patients less likely to return for care was created.

DISCUSSION

An easy to implement clinical decision support tool can help identify FQHC patients at risk for postpartum nonattendance. Future interventions to improve adequacy of prenatal care can encourage early entry into prenatal care and sufficient prenatal visits. These efforts may improve postpartum care attendance and maternal health.

It's a Hard Knock Life: How Kids With Mild Traumatic Brain Injuries Are Treated

OBJECTIVES

The diagnosis and treatment of mild traumatic brain injuries (mTBIs) by emergency medicine providers is greatly varied. Because of the frequency and long-term consequences associated with pediatric head injuries, it is crucial that adequate counseling is provided in acute care settings. The purpose of our study is to evaluate existing practices at Loyola University Medical Center emergency department to address inconsistencies in diagnostic or discharge practices and determine future quality improvement measures.

METHODS

A retrospective cohort study was conducted at an academic hospital emergency department of patient records from 2017 to 2020. Demographic, diagnostic, and discharge data were summarized, and Pearson χ2 tests and Fisher exact tests were performed to determine associations among patient characteristics and provider practices.

RESULTS

A total of 1160 patients met inclusion criteria for analysis. In terms of diagnostic procedure, 31.6% of providers did not uniformly use existing screening criteria, such as PECARN, to determine if CT scans were needed for mTBI evaluation. Most discharge instructions were based on a generalized template on epic (91.9%). Only a minority of providers prepared patient-specific instructions through written, verbal, or other supplemental materials. The most common formats included epic only (46.1%), epic and personalized written instructions (20.2%), and epic and verbal instructions (12.4%). Follow-up care instructions were provided to 93% of patients who received discharged instructions, mainly for primary care (96.7%), sports medicine (1.58%), neurology (0.65%), or other providers (1.11%).

CONCLUSIONS

There is a lack of consistency in the evaluation and education of mTBI in pediatric patients. There is a need for personalized discharge instructions to ensure adequate patient and parent understanding and compliance. Further studies looking at long-term outcomes in these patients would also be beneficial.

The utility of augmented teledermatology to improve dermatologist diagnosis of cellulitis: a cross-sectional study

Dermatology consultation for cases of presumed cellulitis improves diagnostic accuracy and management. However, access to in-person consultation remains limited, a gap that could be filled with teledermatology. Augmented teledermatology may improve outcomes. In this cross-sectional study, 20 dermatologists (60% of whom reported conducting inpatient consults > 1 month per year) reviewed 10 real-life cases representing either cellulitis or pseudocellulitis as diagnosed by in-person dermatology consultation. For each case, respondents recorded their diagnosis, confidence, and management decisions after viewing the history and standard teledermatology photos, the responses to a physician-reported cellulitis questionnaire, and finally thermal images. Overall mean diagnostic accuracy increased from 84 ± 4% with the history and physical to 89 ± 3% when adding a cellulitis questionnaire and thermal images (p = 0.23). Accuracy for cellulitis cases specifically significantly increased from 76 ± 6% to 88 ± 4% when adding a cellulitis questionnaire and thermal images (p = 0.049). Accuracy for pseudocellulitis was consistently ≥ 94%. Augmented teledermatology with a standardized questionnaire and thermal images improved diagnostic accuracy for cases of cellulitis and may increase physician confidence. Dermatologists were able to accurately diagnose regardless of experience with inpatient consults, increasing the pool of potential dermatologists who could diagnose cellulitis remotely.

Review of culprit drugs associated with patients admitted to the burn unit with the diagnosis of Stevens-Johnson Syndrome and Toxic Epidermal Necrolysis Syndrome

PURPOSE

Stevens-Johnson Syndrome (SJS) and Toxic Epidermal Necrolysis Syndrome (TENS) are severe and potentially lethal adverse drug reactions characterized by acute inflammation of the skin, mucous membranes, and ocular surface that typically occurs within weeks of a culprit drug ingestion. The purpose of this study is to report a retrospective trend analysis of SJS spectrum diagnoses and associated culprit drugs in patients admitted to the Loyola University Medical Center (LUMC) Burn Unit, the major referral center in the Chicagoland region for patients with SJS disease spectrum.

METHODS

The electronic medical records (EMR) of 163 patients with a diagnosis of SJS/TENS admitted to the LUMC Burn Unit from 2000 to 2019 were reviewed. Clinical data in addition to the well-established algorithm of drug causality for epidermal necrolysis (ALDEN) allowed us to identify the single most probable culprit drug in 131 cases.

RESULTS

From 2000 to 2019, the most common spectrum classification was TENS (48.1%), followed by SJS (33.6%) and SJS-TEN Overlap Syndrome (18.3%). Anticonvulsants were found to be the most probable culprit class in 30% of cases followed by Trimethoprim-Sulfamethoxazole in 19% of cases. Beta-lactams were the most probable culprit class in 11% of cases while NSAIDs and allopurinol were each the most probable culprit class/drug in 8.4% of cases.

CONCLUSIONS

This is one of the largest single center series of SJS/TENS cases in the United States. Further study into culprit drug distribution by region as well as continuous monitoring of trends is crucial in order to advise prescribing practices.

Retrospective Clinical Evaluation of a Decision-Support Software for Adaptive Radiotherapy of Head and Neck Cancer Patients

Purpose

This study aimed to evaluate the clinical need for an automated decision-support software platform for adaptive radiation therapy (ART) of head and neck cancer (HNC) patients.

Methods

We tested RTapp (SegAna), a new ART software platform for deciding when a treatment replan is needed, to investigate a set of 27 HNC patients’ data retrospectively. For each fraction, the software estimated key components of ART such as daily dose distribution and cumulative doses received by targets and organs at risk (OARs) from daily 3D imaging in real-time. RTapp also included a prediction algorithm that analyzed dosimetric parameter (DP) trends against user-specified thresholds to proactively trigger adaptive re-planning up to four fractions ahead. The DPs evaluated for ART were based on treatment planning dose constraints. Warning (V₉₅<95%) and adaptation (V₉₅<93%) thresholds were set for PTVs, while OAR adaptation dosimetric endpoints of +10% (DE₁₀) were set for all D_max and D_mean DPs. Any threshold violation at end of treatment (EOT) triggered a review of the DP trends to determine the threshold-crossing fraction Fx when the violations occurred. The prediction model accuracy was determined as the difference between calculated and predicted DP values with 95% confidence intervals (CI₉₅).

Results

RTapp was able to address the needs of treatment adaptation. Specifically, we identified 18/27 studies (67%) for violating PTV coverage or parotid D_mean at EOT. Twelve PTVs had V₉₅<95% (mean coverage decrease of −6.8 ± 2.9%) including six flagged for adaptation at median Fx= 6 (range, 1–16). Seventeen parotids were flagged for exceeding D_mean dose constraints with a median increase of +2.60 Gy (range, 0.99–6.31 Gy) at EOT, including nine with DP>DE₁₀. The differences between predicted and calculated PTV V₉₅ and parotid D_mean was up to 7.6% (mean ± CI₉₅, −2.7 ± 4.1%) and 5 Gy (mean ± CI₉₅, 0.3 ± 1.6 Gy), respectively. The most accurate predictions were obtained closest to the threshold-crossing fraction. For parotids, the results showed that Fx ranged between fractions 1 and 23, with a lack of specific trend demonstrating that the need for treatment adaptation may be verified for every fraction.

Conclusion

Integrated in an ART clinical workflow, RTapp aids in predicting whether specific treatment would require adaptation up to four fractions ahead of time.

The Urobiomes of Adult Women With Various Lower Urinary Tract Symptoms Status Differ: A Re-Analysis

The discovery of the urinary microbiome (urobiome) has created opportunities for urinary health researchers who study a wide variety of human health conditions. This manuscript describes an analysis of catheterized urine samples obtained from 1,004 urobiome study participants with the goal of identifying the most abundant and/or prevalent (common) taxa in five clinically relevant cohorts: unaffected adult women (n=346, 34.6%), urgency urinary incontinence (UUI) (n=255, 25.5%), stress urinary incontinence (SUI) (n=50, 5.0%), urinary tract infection (UTI) (n=304, 30.4%), and interstitial cystitis/painful bladder syndrome (IC/PBS) (n=49, 4.9%). Urine was collected via transurethral catheter and assessed for microbes with the Expanded Quantitative Urine Culture (EQUC) technique. For this combined analytic cohort, the mean age was 59 ± 16; most were Caucasian (n=704, 70.2%), Black (n=137, 13.7%), or Hispanic (n=130, 13.0%), and the mean BMI was 30.4 ± 7.7. Whereas many control or IC/PBS cohort members were EQUC-negative (42.4% and 39.8%, respectively), members of the other 3 cohorts were extremely likely to have detectable microbes. The detected urobiomes of the controls and IC/PBS did not differ by alpha diversity or genus level composition and differed by only a few species. The other 3 cohorts differed significantly from the controls. As expected, Escherichia was both prevalent and highly abundant in the UTI cohort, but other taxa also were prevalent at more moderate abundances, including members of the genera Lactobacillus, Streptococcus, Staphylococcus, Corynebacterium, Actinomyces, and Aerococcus. Members of these genera were also prevalent and highly abundant in members of the UUI cohort, especially Streptococcus anginosus. Intriguingly, these taxa were also detected in controls but at vastly lower levels of both prevalence and abundance, suggesting the possibility that UUI-associated symptoms could be the result of an overabundance of typical urobiome constituents. Finally, prevalence and abundance of microbes in the SUI cohort were intermediate to those of the UUI and control cohorts. These observations can inform the next decade of urobiome research, with the goal of clarifying the mechanisms of urobiome community composition and function. There is tremendous potential to improve diagnosis, evaluation and treatment for individuals affected with a wide variety of urinary tract disorders.

Acupuncture for female bladder pain syndrome: a randomized controlled trial

INTRODUCTION

Growing evidence supports acupuncture for several pain conditions including chronic prostatitis. This study aimed to determine the safety, tolerability, and effectiveness of acupuncture in reducing pain in women with interstitial cystitis/bladder pain syndrome (IC/BPS).

MATERIALS AND METHODS

This prospective randomized single-blinded study compared electro-acupuncture (EA) to minimal acupuncture (MA) after 6 weekly treatments and again after 6 weeks of no treatment. Pain was assessed using the Brief Pain Inventory-Short Form (worst pain, average pain, pain severity, pain interference) and the Pain Catastrophizing Scale (PCS). Physical exams evaluated pelvic floor muscle tenderness. Mixed-effects models were used to estimate adjusted means over follow up.

RESULTS

Patients were randomized to EA (n = 11) or MA (n = 10). There were no adverse events. Both groups' worst pain improved at 6 weeks, -2.91 ± 0.59 and -2.09 ± 0.68 for EA and MA respectively with no difference between groups (p = 0.37). Results were similar at 12 weeks. The EA group had greater improvement in pain interference at 6 weeks, -3.28 ± 0.51 versus -1.67 ± 0.58 (p = 0.049). The between group difference was not maintained at 12 weeks (p = 0.13). Average pain and pain severity showed no difference between groups (p > 0.05). The PCS improved overall at 6 weeks, -6.2 ± 2.5 (p = 0.03), with no difference between groups (p = 0.39). On physical exam, only the EA group showed a significant decrease in levator ani tenderness (p = 0.031) after treatment.

CONCLUSIONS

Both EA and MA showed improvement in worst pain scores, however EA showed greater improvement in pain interference and pelvic floor muscle tenderness in women with IC/BPS.

Development and multimodal validation of a substance misuse algorithm for referral to treatment using artificial intelligence (SMART-AI): a retrospective deep learning study

Background

Substance misuse is a heterogeneous and complex set of behavioural conditions that are highly prevalent in hospital settings and frequently co-occur. Few hospital-wide solutions exist to comprehensively and reliably identify these conditions to prioritise care and guide treatment. The aim of this study was to apply natural language processing (NLP) to clinical notes collected in the electronic health record (EHR) to accurately screen for substance misuse.

Methods

The model was trained and developed on a reference dataset derived from a hospital-wide programme at Rush University Medical Center (RUMC), Chicago, IL, USA, that used structured diagnostic interviews to manually screen admitted patients over 27 months (between Oct 1, 2017, and Dec 31, 2019; n=54 915). The Alcohol Use Disorder Identification Test and Drug Abuse Screening Tool served as reference standards. The first 24 h of notes in the EHR were mapped to standardised medical vocabulary and fed into single-label, multilabel, and multilabel with auxillary-task neural network models. Temporal validation of the model was done using data from the subsequent 12 months on a subset of RUMC patients (n=16 917). External validation was done using data from Loyola University Medical Center, Chicago, IL, USA between Jan 1, 2007, and Sept 30, 2017 (n=1991 adult patients). The primary outcome was discrimination for alcohol misuse, opioid misuse, or non-opioid drug misuse. Discrimination was assessed by the area under the receiver operating characteristic curve (AUROC). Calibration slope and intercept were measured with the unreliability index. Bias assessments were performed across demographic subgroups.

Findings

The model was trained on a cohort that had 3·5% misuse (n=1 921) with any type of substance. 220 (11%) of 1921 patients with substance misuse had more than one type of misuse. The multilabel convolutional neural network classifier had a mean AUROC of 0·97 (95% CI 0·96–0·98) during temporal validation for all types of substance misuse. The model was well calibrated and showed good face validity with model features containing explicit mentions of aberrant drug-taking behaviour. A false-negative rate of 0·18–0·19 and a false-positive rate of 0·03 between non-Hispanic Black and non-Hispanic White groups occurred. In external validation, the AUROCs for alcohol and opioid misuse were 0·88 (95% CI 0·86–0·90) and 0·94 (0·92–0·95), respectively.

Interpretation

We developed a novel and accurate approach to leveraging the first 24 h of EHR notes for screening multiple types of substance misuse.

Funding

National Institute On Drug Abuse, National Institutes of Health.

Revisiting CT Signs of Unilateral Vocal Fold Paralysis: A Single, Blinded Study

BACKGROUND AND PURPOSE

Several CT findings are thought to be indicative of vocal cord paralysis; however, these signs have never been validated in a blinded fashion. This study attempts to compare and validate these signs and determine their accuracy in predicting vocal cord paralysis.

MATERIALS AND METHODS

A retrospective chart review was performed, and CT scans from patients with known unilateral vocal cord paralysis and known normal vocal cord movement were reviewed by 3 radiologists who were blinded to the status of the patients' laryngeal function. The scans were reviewed and scored for 8 accepted signs of vocal cord paralysis as well as for predicting a final diagnostic conclusion. Statistical analysis using odds ratios for signs and the Fleiss κ for criterion agreement among the radiologists was performed for diagnostic accuracy.

RESULTS

The presence of medial displacement of the posterior ipsilateral vocal fold margin and ipsilateral laryngeal ventricular dilation yielded the greatest positive predictive value. Other signs demonstrated high specificity, but interrater discrepancy was greater than expected and diminished the reliability of these signs in predicting vocal cord paralysis. Overall, sensitivity and negative predictive values were low.

CONCLUSIONS

Predicting vocal cord paralysis on the basis of CT findings is not as accurate or straightforward in prospectively predicting vocal cord paralysis as implied in prior studies.

Bias Assessment and Correction in Machine Learning Algorithms: A Use-Case in a Natural Language Processing Algorithm to Identify Hospitalized Patients with Unhealthy Alcohol Use

Unhealthy alcohol use represents a major economic burden and cause of morbidity and mortality in the United States. Implementation of interventions for unhealthy alcohol use depends on the availability and accuracy of screening tools. Our group previously applied methods in natural language processing and machine learning to build a classifier for unhealthy alcohol use. In this study, we sought to evaluate and address bias through the use-case of our classifier. We demonstrated the presence of biased unhealthy alcohol use risk underestimation among Hispanic compared to Non-Hispanic White trauma inpatients, 18- to 44-year-old compared to 45 years and older medical/surgical inpatients, and Non-Hispanic Black compared to Non-Hispanic White medical/surgical inpatients. We further showed that intercept, slope, and concurrent intercept and slope recalibration resulted in minimal or no improvements in bias-indicating metrics within these subgroups. Our results exemplify the importance of integrating bias assessment early into the classifier development pipeline.

The Addition of United States Census-Tract Data Does Not Improve the Prediction of Substance Misuse

Predictors from the structured data in the electronic health record (EHR) have previously been used for case-identification in substance misuse. We aim to examine the added benefit from census-tract data, a proxy for socioeconomic status, to improve identification. A cohort of 186,611 hospitalizations was derived between 2007 and 2017. Reference labels included alcohol misuse only, opioid misuse only, and both alcohol and opioid misuse. Baseline models were created using 24 EHR variables, and enhanced models were created with the addition of 48 census-tract variables from the United States American Community Survey. The absolute net reclassification index (NRI) was applied to measure the benefit in adding census-tract variables to baseline models. The baseline models already had good calibration and discrimination. Adding census-tract variables provided negligible improvement to sensitivity and specificity and NRI was less than 1% across substance groups. Our results show the census-tract added minimal value to prediction models.

Antibody Response to SARS-CoV-2 Infection and Vaccination in COVID-19-naïve and Experienced Individuals

Understanding the magnitude of responses to vaccination during the ongoing SARS-CoV-2 pandemic is essential for ultimate mitigation of the disease. Here, we describe a cohort of 102 subjects (70 COVID-19-naïve, 32 COVID-19-experienced) who received two doses of one of the mRNA vaccines (BNT162b2 (Pfizer-BioNTech) and mRNA-1273 (Moderna)). We document that a single exposure to antigen via infection or vaccination induces a variable antibody response which is affected by age, gender, race, and co-morbidities. In response to a second antigen dose, both COVID-19-naïve and experienced subjects exhibited elevated levels of anti-spike and SARS-CoV-2 neutralizing activity; however, COVID-19-experienced individuals achieved higher antibody levels and neutralization activity as a group. The COVID-19-experienced subjects exhibited no significant increase in antibody or neutralization titer in response to the second vaccine dose (i.e., third antigen exposure). Finally, we found that COVID-19-naïve individuals who received the Moderna vaccine exhibited a more robust boost response to the second vaccine dose (p = 0.004) as compared to the response to Pfizer-BioNTech. Ongoing studies with this cohort will continue to contribute to our understanding of the range and durability of responses to SARS-CoV-2 mRNA vaccines.

Abstract TP83: Benefit Of Multimodal Intraoperative Neuromonitoring Techniques During Carotid Procedures

Background: Carotid intervention is done for reduction of stroke risk. Intraoperative neuromonitoring (IOM), including electroencephalography (EEG), somatosensory evoked potential (SSEP), and cerebral oximetry/near-infrared spectroscopy (NIRS) may be used to guide operative decision-making and minimize the risk of perioperative complications. The utility of combining IOM methods is unclear. Therefore, we sought to investigate the value of multimodal IOM for carotid interventions.

Methods: A retrospective review of all patients undergoing carotid intervention from January 1, 2012 to December 31, 2019 at a single academic medical center was done. Patients with acute stroke undergoing catheter directed thrombectomy were excluded. Data collected included: patient demographics, IOM parameters, surgical intervention with endpoints of 30-day post-operative complications. Surrogate end-points of IOM change and intraoperative hemodynamic interventions were also assessed. Efficacy and diagnostic value of each IOM modality alone, and in combination, were assessed using multivariable logistic regression.

Results: 409 carotid interventions in 311 patients meeting inclusion criteria were identified. Patients were 63.3% male, 36.7% female, 83.4% white, 7.1% Hispanic, and 5.9% other. 162(39.6%) interventions were performed for ipsilateral symptoms including 111(27.1%) ipsilateral stroke, 91(22.3%) TIA. Use of a single versus multiple IOM modalities was not associated with a reduction in post-operative neurologic events. Changes in SSEP demonstrated a 2.97 odds ratio for predicting neurologic adverse events (AE), though this was not statistically significant.

Conclusion: IOM changes were associated with an increase in AEs but IOM impact on intraoperative changes in surgical plans were infrequent. It is unclear whether IOM had a significant impact on procedural modifications though this may reflect an underpowered sample size. The low impact of IOM on outcome raises the question of cost-effectiveness of these techniques in reducing intraoperative complications from carotid procedures.

Local Infiltrative Analgesia is Equivalent to Fascia Iliaca Block for Perioperative Pain Management for Prophylactic Cephalomedullary Nail Fixation

BACKGROUND

Impending pathologic fractures of the femur due to metastatic bone disease are treated with prophylactic internal fixation to prevent fracture, maintain independence, and improve quality of life. There is limited data to support an optimal perioperative pain regimen.

METHODS

A proof of concept comparative cohort analysis was performed: 21 patients who received a preoperative fascia iliacus nerve block (FIB) were analyzed retrospectively while 9 patients treated with local infiltrative analgesia (LIA) were analyzed prospectively. Primary outcomes included: visual analog scale (VAS) pain scores, narcotic requirements and hospital length of stay. Patient cohorts were compared via two-sample t-tests and Fischer's exact tests. Differences in VAS pain scores, length of stay and morphine milligram equivalents (MME) were assessed with Wilcoxon rank sum.

RESULTS

The LIA group had more patients treated with preoperative narcotics (p=0.042). There were no significant differences between the FIB and LIA groups in MME utilized intraoperatively (30.0 vs 37.5, p=0.79), on POD 0 (38.0 vs 30.0, p=0.93), POD 1 (46.0 vs 55.5, p=0.95) or POD 2 (40.0 vs 60.0 p=0.73). There were no significant differences in analog pain scale at any time point or in hospital length of stay (78 vs 102 hours, p=0.86).

CONCLUSION

Despite an increased number of patients being on preoperative narcotics in the LIA group, use of LIA compared with FIB is not associated with an increase in VAS pain scores, morphine milligram equivalents (MME), or length of hospital stay in patients undergoing prophylactic internal fixation of impending pathologic femur fractures.Level of Evidence: III.

Assessment of Interpretation Modality and Patient Comprehension in Spanish-speaking Limited English Proficiency (LEP) Patients Presenting to the Emergency Department

BACKGROUND

Provider-patient language discrepancies with limited English proficiency (LEP) patients can lead to misunderstandings about diagnoses and follow-up care.

OBJECTIVES

To assess interpretation modalities used in the emergency department (ED) in terms of patient satisfaction and understanding of discharge diagnosis.

METHODS

Spanish-speaking LEP patients completed a survey assessing overall satisfaction and discharge diagnosis comprehension. Modalities included in-person (interpreter or physician/nurse), remote (phone or video), or combination. Differences in proportions with correctly identified diagnoses were compared by modality using Fisher's exact test.

RESULTS

Patients preferred a Spanish-speaking staff member (52%) or in-person interpreter (33%) over other modalities. Almost 74% of surveyed patients accurately described their discharge diagnosis. Diagnostic accuracy was increased among patients using remote modalities alone compared with in-person alone or combination (p=.02).

CONCLUSION

Taking into account patient preferences and diagnostic accuracy, this study suggests the utility of having different interpreter modalities available for Spanish-speaking LEP ED patients.

Ki-67/MIB-1 and Recurrence in Pituitary Adenoma

Objectives Ki-67/MIB-1 is a marker of cellular proliferation used as a pathological parameter in the clinical assessment of pituitary adenomas, where its expression has shown utility in predicting the invasiveness of these tumors. However, studies have shown variable results when using Ki-67/MIB-1 association with recurrence. The purpose of this study is to determine if a high Ki-67/MIB-1 labeling index (LI) is predictive of recurrence in pituitary adenomas.

Methods A retrospective chart review was performed for patients undergoing pituitary adenoma resection with at least 1 year of follow-up. Additionally, systematic data searches were performed and included studies that correlated recurrence rate to Ki-67/MIB-1 LI. Our institutional data were included in a synthesis with previously published data.

Results Our institutional review included 79 patients with a recurrence rate of 26.6%. We found that 8.8% of our patients had a high Ki-67/MIB-1 LI (>3%); however, high Ki-67/MIB-1 was not associated with recurrence. The systematic review identified 244 articles and 49 full-text articles that were assessed for eligibility. Quantitative analysis was performed on 30 articles including our institutional data and 18 studies reported recurrence by level of Ki-67/MIB-1 LI. Among studies that compared Ki-67/MIB-1 ≥3 vs. <3%, 10 studies reported odds ratios (OR) greater than 1 of which 6 were statistically significant. A high Ki-67/MIB-1 had higher odds of recurrence via the pooled odds ratio (OR = 4.15, 95% confidence interval [CI]: 2.31–7.42).

Conclusion This systematic review suggests that a high Ki-67/MIB-1 should prompt an increased duration of follow-up due to the higher odds of recurrence of pituitary adenoma.

Evaluation of Stigma Toward Individuals With Alopecia

Importance

Perceived stigma among patients with alopecia is associated with impaired quality of life; however, the magnitude of laypersons' stigma toward individuals with alopecia is unknown.

Objective

To determine the prevalence and magnitude of laypersons' stigma toward individuals with varying degrees of alopecia and whether stigma increases with increased severity of alopecia.

Design, Setting, and Participants

This was a cross-sectional study using an internet survey administered to a convenience sample of adult respondents in the US participating on the Amazon Mechanical Turk platform. Portrait images of 6 individuals without hair loss were created using artificial intelligence and stock images. Each portrait was edited to create 2 additional versions, 1 with scalp hair loss and 1 with complete hair loss, for a total of 18 images. On January 9 to 10, 2020, the survey presented each internet respondent with 1 randomly selected portrait to be used in answering a series of stigma-related questions from 3 domains: stereotypes, social distance, and disease-related myths; the third domain was presented only to respondents who believed that the individual pictured had a medical condition.

Main Outcomes and Measures

The main outcome was the prevalence and magnitude of stigma of laypersons toward individuals with alopecia and the percentage of laypersons who believed the individual pictured had a medical condition as recorded in survey responses.

Results

The survey was completed by 2015 respondents (99.9% completion rate) with a mean age of 37 (range, 18-78) years; 1014 (50.3%) were men; 1596 (79.2%) were White; and 1397 (69.3%) had a college or postcollege education. Endorsement of every stigma item increased as alopecia severity increased (2.4%-27.6%). Absolute change on the stereotype (0.5-0.6) and social distance scales (0.2-0.5) also increased, indicating more stigma. The percentage of respondents believing the individual pictured had a medical condition increased as alopecia severity increased (33.6%-75.7%; P < .001). Among the subgroup of respondents who were asked to rate their agreement with disease-related myths, the absolute change on the myth scale decreased as alopecia severity increased, indicating decreased stigma (-0.7 to -1.2).

Conclusions and Relevance

This cross-sectional survey study found that stigmatizing attitudes of laypersons toward patients with alopecia exist across a multitude of social and professional scenarios. Stigma prevalence and magnitude vary by alopecia severity and possibly by whether alopecia is believed to be a medical condition.

Insurance Acceptance, Appointment Wait Time, and Dermatologist Access Across Practice Types in the US

Importance

In the 15 years since dermatology access was last investigated on a national scale, the practice landscape has changed with the rise of private equity (PE) investment and increased use of nonphysician clinicians (NPCs).

Objective

To determine appointment success and wait times for patients with various insurance types at clinics with and without PE ownership.

Design, Setting, and Participants

In this study, PE-owned US clinics were randomly selected and matched with 2 geographically proximate clinics without PE ownership. Researchers called each clinic 3 times over a 5-day period to assess appointment/clinician availability for a fictitious patient with a new and changing mole. The 3 calls differed by insurance type specified, which were Blue Cross Blue Shield (BCBS) preferred provider organization, Medicare, or Medicaid.

Main Outcomes and Measures

Appointment success and wait times among insurance types and between PE-owned clinics and control clinics. Secondary outcomes were the provision of accurate referrals to other clinics when appointments were denied and clinician and next-day appointment availability.

Results

A total of 1833 calls were made to 204 PE-owned and 407 control clinics without PE ownership across 28 states. Overall appointment success rates for BCBS, Medicare, and Medicaid were 96%, 94%, and 17%, respectively. Acceptance of BCBS (98.5%; 95% CI, 96%-99%; P = .03) and Medicare (97.5%; 95% CI, 94%-99%; P = .02) were slightly higher at PE-owned clinics (compared with 94.6% [95% CI, 92%-96%] and 92.8% [95% CI, 90%-95%], respectively, at control clinics). Wait times (median days, interquartile range [IQR]) were similar for patients with BCBS (7 days; IQR, 2-22 days) and Medicare (7 days; IQR, 2-25 days; P > .99), whereas Medicaid patients waited significantly longer (13 days; IQR, 4-33 days; P = .002). Clinic ownership did not significantly affect wait times. Private equity-owned clinics were more likely than controls to offer a new patient appointment with an NPC (80% vs 63%; P = .001) and to not have an opening with a dermatologist (16% vs 6%; P < .001). Next-day appointment availability was greater at PE-owned clinics than controls (30% vs 21%; P = .001).

Conclusions and Relevance

Patients with Medicaid had significantly lower success in obtaining appointments and significantly longer wait times regardless of clinic ownership. Although the use of dermatologists and NPCs was similar regardless of clinic ownership, PE-owned clinics were more likely than controls to offer new patient appointments with NPCs.

Associations Between Cholecalciferol Supplementation and Self-Reported Symptoms Among Women With Metastatic Breast Cancer and Vitamin D Deficiency: A Pilot Study

OBJECTIVES

To assess the potential effect of cholecalciferol supplementation to reduce symptom burden for women with metastatic breast cancer (MBC).

SAMPLE & SETTING

11 clinically stable women with estrogen receptor-positive MBC were recruited from a single cancer center for this phase 1, nonrandomized study (NCT02186015).

METHODS & VARIABLES

Women with insufficient serum 25-hydroxyvitamin D (25[OH]D) levels qualified to receive high-dose repletion therapy. Clinical and questionnaire data on common symptoms and quality of life were obtained prior to and following supplementation.

RESULTS

Serum 25(OH)D increased significantly pre- versus postintervention. Trends for improvements in endocrine symptoms, bone pain, and fatigue were observed following the intervention.

IMPLICATIONS FOR NURSING

Women achieved normal serum 25(OH)D levels after eight weeks of supplementation and reported reduced symptom burden. Vitamin D may be a low-cost supportive care therapy; however, future studies should be considered.

Clinical and histologic features associated with lentigo maligna clearance after imiquimod treatment

BACKGROUND

Imiquimod cream may be used as a non-surgical treatment for lentigo maligna or as adjuvant therapy following excision to decrease the risk of recurrence.

OBJECTIVES

To evaluate histologic and clinical factors associated with clinical clearance of lentigo maligna treated with imiquimod.

METHODS

We performed a retrospective review of all patients diagnosed with lentigo maligna and treated with imiquimod between 1997 and 2019 at our academic institution.

RESULTS

We observed clinical clearance in 93% (66/71) of participants who received adjuvant imiquimod following surgery and 79% (19/24) in the primary non-surgical treatment group over a median of 38 months of follow-up. In the adjuvant therapy group, positive surgical margins were associated with a decreased rate of clinical clearance when compared to cases with close (<1 mm) margins or background melanocytic dysplasia (83.3 vs. 100%, p = .01). The presence of an inflammatory response during treatment was associated with increased clearance (94.1 vs. 66.7%, p = .02).

CONCLUSIONS

Adjuvant imiquimod treatment may decrease LM recurrence rates in cases with background melanocytic dysplasia or close margins. LM cases with positive surgical margins need close clinical follow-up given higher recurrence rates.

Effectiveness of an interprofessional ambulatory care model on diabetes: evaluating clinical markers in a low-income patient population

The purpose of this study was to determine whether a new ambulatory care model, interprofessional collaborative care-coordinated team model (interprofessional model), based on the Wagner Care Model improved clinical indicators in a low-income population. This study was a retrospective 12-month pre-post (n = 204) and propensity matched (n = 171) comparative study of the interprofessional model in a primary clinic for patients with type 2 diabetes. Secondary data were collected from June 2014to February 2017 in an academic medical centre in a large Midwestern city. Findings demonstrated statistically and clinically significant improvement in A1C in both the pre/post arm of the study (↓ 0.8%) and the intervention/propensity matched arm (↓ 0.53%). Within the intervention group, there was a significant decrease in weight in the pre/post arm with 55% of cases losing weight, whereas 45% did not lose weight (p = .02). Diastolic blood pressure less than 90 also significantly improved in the pre/post arm of the study (10.1% n = 18, versus 3.9%, n = 7, p = .04). The interprofessional model showed that an ambulatory healthcare redesign incorporating an interprofessional team approach to optimise the health of this type 2 diabetes patient population can be effective. This study demonstrates the importance of using interprofessional collaborative practice teams to guide healthcare and improve patient outcomes.

Communicating Spiritual Care in the Electronic Health Record

Chaplains must document their ministry of care in electronic health records that primarily focus on the physical dimension of care. Creating chaplain documentation that reflects the spiritual dimension of care requires chaplains to participate in the screen design. This article describes how chaplain documentation was designed and refined using psychometric methods. The resulting system successfully supported chaplain workflow, provided an ability to aggregate chaplain workload, and integrated the chaplain into the interprofessional team by structuring, linking, and sharing both the chaplain and nursing assessment of spiritual distress in the electronic health record. Documentation used 5-point Likert scales to measure different dimensions of patient spirituality. Reliability and validity were further evaluated as part of a workshop at an Association of Professional Chaplains annual meeting. Findings supported interrater reliability and the ability to predict and discriminate change pre and post encounter. Documentation screen content is presented.

Radiographic evaluation of reactive osteitis in traumatic injury of sacroiliac joints

OBJECTIVES

After traumatic Sacroiliac (SI) joint injury, follow up radiographic imaging can demonstrate subchondral bone resorption resembling inflammatory sacroiliitis. No studies have described the incidence of marginal SI post-traumatic osteitis, the probable temporal relationship to the initial traumatic injury, or the possible effect of unilateral hardware fixation on the contralateral SI joint.

METHODS

A Level 1 trauma center imaging database was queried to identify patients with pelvic bony trauma between 2005 and 2017 with CT baseline preserved SI cortication and unilateral/bilateral traumatic SI diastasis. Serial radiographs were retrospectively evaluated by 2 musculoskeletal-trained radiologists at initial, 6 weeks, 3 months and 6 months following trauma, with documentation of diastasis, subchondral resorption, and operative fixation.

RESULTS

206 SI joints in 106 total patients met inclusion criteria. There was a statistically significant association between injury and presence of resorption at 6 weeks post-trauma for the right SI joint only. There was no other statistically significant relationship between injury and presence of resorption at any other post-trauma evaluation. There was no statistical relationship between resorption and surgical fixation. There was a statistically significant increased incidence of resorption in the post-traumatic population when compared to an atraumatic population undergoing CT pelvis study for non-SI related indications as well as compared to the incidence of inflammatory sacroiliitis in a general population.

CONCLUSIONS

This study confirms an incidence of sub-acute subchondral bone resorption following traumatic joint injury above that expected for a general, non-traumatic population. Accurate interpretation of this traumatic finding minimizes inappropriate consultation and intervention for inflammatory sacroiliitis.

Uterine perforation during brachytherapy for cervical cancer: Complications, outcomes, and best practices for forward treatment planning and management

PURPOSE

The purpose of the study was to determine the incidence of uterine perforations, review the associated complications, and propose guidelines for management of perforations after brachytherapy.

METHODS AND MATERIALS

A retrospective chart review was conducted for all patients with cervical cancer who received single or multiple high-dose-rate brachytherapy implants between April 2006 and May 2017 at a single academic institution. CT and MRI images were retrospectively evaluated to record incidences of uterine perforation of tandem during brachytherapy. Acute and long-term complications during and after treatment were scored using the Common Terminology Criteria for Adverse Events, Version 4.0, of the National Cancer Institute.

RESULTS

A total of 123 patients were included in the study. Perforations were observed in 22 patients (17.9%) with 31 (6.4%) of the 482 total implants. Of the different categories of adverse events, only the rate of acute infectious complications among those with perforations (n = 3, 13.6%) versus those without perforations (n = 3, 3.0%) was significant (p = 0.040). Two of the three perforated patients with acute infections had mild urinary tract infections, and all resolved without complications or treatment delays. The remaining one patient had a frank perforation of the anterior uterine wall with a subsequent Grade 3 pyometra infection despite administration of prophylactic antibiotics and 1-week treatment delay. This case was eventually resolved with cervical dilation and evacuation of fluid. Long-term complications were not different between the two arms.

CONCLUSIONS

Patients with cervical cancer with uterine perforations may be able to safely proceed with brachytherapy treatment without delay or need for prophylactic antibiotics in the acute setting. Further validating data would be able to assist in establishing a new standard of care and help prevent unnecessary and harmful breaks during treatment.

Comparison and interpretability of machine learning models to predict severity of chest injury

Objective

Trauma quality improvement programs and registries improve care and outcomes for injured patients. Designated trauma centers calculate injury scores using dedicated trauma registrars; however, many injuries arrive at nontrauma centers, leaving a substantial amount of data uncaptured. We propose automated methods to identify severe chest injury using machine learning (ML) and natural language processing (NLP) methods from the electronic health record (EHR) for quality reporting.

Materials and Methods

A level I trauma center was queried for patients presenting after injury between 2014 and 2018. Prediction modeling was performed to classify severe chest injury using a reference dataset labeled by certified registrars. Clinical documents from trauma encounters were processed into concept unique identifiers for inputs to ML models: logistic regression with elastic net (EN) regularization, extreme gradient boosted (XGB) machines, and convolutional neural networks (CNN). The optimal model was identified by examining predictive and face validity metrics using global explanations.

Results

Of 8952 encounters, 542 (6.1%) had a severe chest injury. CNN and EN had the highest discrimination, with an area under the receiver operating characteristic curve of 0.93 and calibration slopes between 0.88 and 0.97. CNN had better performance across risk thresholds with fewer discordant cases. Examination of global explanations demonstrated the CNN model had better face validity, with top features including “contusion of lung” and “hemopneumothorax.”

Discussion

The CNN model featured optimal discrimination, calibration, and clinically relevant features selected.

Conclusion

NLP and ML methods to populate trauma registries for quality analyses are feasible.

Targeting EphA2 suppresses hepatocellular carcinoma initiation and progression by dual inhibition of JAK1/STAT3 and AKT signaling

Hepatocellular carcinoma (HCC) remains one of the deadliest malignancies worldwide. One major obstacle to treatment is a lack of effective molecular-targeted therapies. In this study, we find that EphA2 expression and signaling are enriched in human HCC and associated with poor prognosis. Loss of EphA2 suppresses the initiation and growth of HCC both in vitro and in vivo. Furthermore, CRISPR/CAS9-mediated EphA2 inhibition significantly delays tumor development in a genetically engineered murine model of HCC. Mechanistically, we discover that targeting EphA2 suppresses both AKT and JAK1/STAT3 signaling, two separate oncogenic pathways in HCC. We also identify a small molecule kinase inhibitor of EphA2 that suppresses tumor progression in a murine HCC model. Together, our results suggest EphA2 as a promising therapeutic target for HCC.

Creation and Validation of Classification Criteria for Discoid Lupus Erythematosus

Importance

Classification criteria are the standardized definitions that are used to enroll uniform cohorts for research studies. They emphasize high specificity and are distinct from diagnostic criteria. No universally recognized classification criteria currently exist for discoid lupus erythematosus (DLE), which has led to problematic heterogeneity in observational and interventional clinical studies across the field.

Objective

To create and validate classification criteria for DLE using 12 previously defined candidate criteria items.

Design, Setting, and Participants

For this diagnostic study, candidate criteria items were prospectively applied by dermatologists and dermatopathologists at clinical visits of patients with DLE or a condition that could be confused for DLE, termed a DLE mimicker, at academic dermatology practices across the United States, Poland, Japan, and South Korea. Data were collected from December 1, 2017, to February 1, 2019, and analyzed from March 1 to September 19, 2019.

Main Outcomes and Measures

Clinical features among these 2 groups were calculated and compared with χ2 or Fisher exact tests. Candidate models were identified using best subsets logistic regression analysis. Improvement tests, fit statistics, and discrimination were considered to choose a final model.

Results

Nine sites contributed 215 patients, 15 of whom had missing or incomplete data. The final model for DLE classification criteria includes only clinical variables: atrophic scarring (3 points), location in the conchal bowl (2 points), preference for the head and neck (2 points), dyspigmentation (1 point), follicular hyperkeratosis and/or plugging (1 point), and erythematous to violaceous in color (1 point), with an area under the receiving operating characteristic curve of 0.91 (95% CI, 0.87-0.95). A score of at least 5 points yields a sensitivity of 84.1% and a specificity of 75.9% in the classification of DLE, with increasing scores yielding higher specificity.

Conclusions and Relevance

These findings provide the initial validation of classification criteria for DLE for use in observational and clinical trials.

Prediction of severe chest injury using natural language processing from the electronic health record

INTRODUCTION

Trauma injury severity scores are currently calculated retrospectively from the electronic health record (EHR) using manual annotation by certified trauma coders. Natural language processing (NLP) of clinical documents in the EHR may enable automated injury scoring. We hypothesize that NLP with machine learning can discriminate between cases of severe and non-severe injury to the thorax after trauma.

METHODS

Clinical documents from a trauma center were examined between 2014 and 2018. Severe chest injury was defined as a thorax abbreviated injury score (AIS) >2 and served as the reference standard for supervised learning. Free text unigrams and concept unique identifiers (CUIs) from the Unified Medical Language Systems (UMLS) were extracted from clinical documents collected at one hour, four hours, and eight hours after patient arrival to the emergency department. Logistic regression models with elastic net regularization were tuned to maximize area under the receiver operating characteristic curve (AUROC) using 10-fold cross-validation on the training dataset (80%) and tested on a hold-out 20% dataset.

RESULTS

There were 6,891 traumas that met inclusion criteria. The complete data corpus consisted of 473,694 documents. Models trained using the first hour of data had a mean AUROC of 0.88 (95%CI [0.86, 0.89]); model discrimination and reclassification from the first hour significantly improved after eight hours with a mean AUROC of 0.94 (95%CI [0.93, 0.95]). Performance of models using CUIs were similar to unigrams (p>0.05). Models demonstrated excellent clinical face validity.

CONCLUSIONS

Both CUIs and unigrams demonstrated excellent discrimination in predicting severity of chest injury using the first eight hours of clinical documents. Our model demonstrates that automated anatomical injury scoring is feasible and may be used for aggregation of data for trauma research and quality programs.

Development and application of a high throughput natural language processing architecture to convert all clinical documents in a clinical data warehouse into standardized medical vocabularies

OBJECTIVE

Natural language processing (NLP) engines such as the clinical Text Analysis and Knowledge Extraction System are a solution for processing notes for research, but optimizing their performance for a clinical data warehouse remains a challenge. We aim to develop a high throughput NLP architecture using the clinical Text Analysis and Knowledge Extraction System and present a predictive model use case.

MATERIALS AND METHODS

The CDW was comprised of 1 103 038 patients across 10 years. The architecture was constructed using the Hadoop data repository for source data and 3 large-scale symmetric processing servers for NLP. Each named entity mention in a clinical document was mapped to the Unified Medical Language System concept unique identifier (CUI).

RESULTS

The NLP architecture processed 83 867 802 clinical documents in 13.33 days and produced 37 721 886 606 CUIs across 8 standardized medical vocabularies. Performance of the architecture exceeded 500 000 documents per hour across 30 parallel instances of the clinical Text Analysis and Knowledge Extraction System including 10 instances dedicated to documents greater than 20 000 bytes. In a use-case example for predicting 30-day hospital readmission, a CUI-based model had similar discrimination to n-grams with an area under the curve receiver operating characteristic of 0.75 (95% CI, 0.74-0.76).

DISCUSSION AND CONCLUSION

Our health system's high throughput NLP architecture may serve as a benchmark for large-scale clinical research using a CUI-based approach.

Surgical training tools for dermatology trainees: porcine vs. synthetic skin for excision and repair

Since dermatologists routinely perform surgery in an outpatient setting, ensuring that dermatology trainees are provided with opportunities to develop sufficient proficiency in excisional surgery and suture technique is paramount. The objectives of this study are to assess trainee preference for silicone-based synthetic skin compared with porcine skin as a surgical training medium and to assess the ability of trainees to successfully demonstrate basic surgical skills using the simulated skin model. Participants were a convenience sample of dermatology residents from the greater Chicago area, who were asked to perform an elliptical excision and bilayered repair on a silicone-based synthetic skin model. Residents were then surveyed regarding their satisfaction with the model. Four blinded dermatologist raters evaluated digital photographs obtained during the performance of the procedures and graded the execution of each maneuver using a surgical task checklist. Nineteen residents were enrolled. Residents were more likely to prefer pig skin to simulated skin for overall use (p = 0.040) and tissue repair (p = 0.018), but the nominal preference for tissue handling was nonsignificant (p = 0.086). There was no significant difference between satisfaction with pig skin versus synthetic skin with regard to excision experience (p = 0.82). The majority of residents (10/19) performed all surgical checklist tasks correctly. Of those residents who did not perform all steps correctly, many had difficulty obtaining adequate dermal eversion and wound approximation. Synthetic skin may be conveniently and safely utilized for hands-on surgical practice. Further refinement may be necessary to make synthetic skin comparable in feel and use to animal skin.

Low medication adherence is associated with decline in health-related quality of life: results of a longitudinal analysis among older women and men with hypertension

OBJECTIVE

The aim of this study was to determine the association of low antihypertensive medication adherence with decline in health-related quality of life (HRQOL) over 1 year.

METHODS

We used data from older men and women with hypertension (n = 1525) enrolled in the Cohort Study of Medication Adherence among Older Adults. Adherence was measured using the validated self-report four-item Krousel-Wood Medication Adherence Scale (K-Wood-MAS-4) (low adherence = score ≥1) and prescription refill-based proportion of days covered (PDC) (low adherence = PDC < 0.80). We defined decline in HRQOL as a decrease in Mental Component Summary (MCS) or Physical Component Summary (PCS) score (from the RAND 36-Item Health Survey 1.0 administered at two time points - at the time of adherence assessment and 1 year later) equivalent to the minimal important difference (MID) for each respective summary score, calculated as the average of MID estimates derived from distribution and anchor-based approaches.

RESULTS

The prevalence of low adherence was 38.6% using the K-Wood-MAS-4 and 23.9% using PDC. On the basis of mean MID estimates of 4.40 for MCS and 5.16 for PCS, 21.8 and 25.2% of participants experienced a decline in MCS and PCS, respectively, over 1 year. Low adherence was associated with a decline in MCS for K-Wood-MAS-4 [prevalence ratio = 1.32, 95% confidence interval (95% CI) 1.08-1.62, P = 0.008], but not PDC (prevalence ratio  = 1.17, 95% CI 0.94-1.47, P = 0.168). Low adherence was not associated with decline in PCS (K-Wood-MAS-4: prevalence ratio  = 0.95, 95% CI 0.79-1.16; PDC: prevalence ratio  = 1.10, 95% CI 0.90-1.35).

CONCLUSION

Low self-report medication adherence is associated with decline in mental HRQOL over 1 year in older adults with hypertension.