Disparities in use of physical restraints at an urban, minority-serving hospital emergency department.

BACKGROUND: Recent reports have identified associations between patient race and ethnicity and use of physical restraint while receiving care in the emergency department (ED). However, no study has assessed this relationship in hospitals primarily treating patients of color and underserved populations. The primary objective of this study was to evaluate the association between race/ethnicity and the use of restraints in an ED population at a minority-serving, safety-net institution. METHODS: For this cross-sectional study, chart review identified all adult patients presenting to the Boston Medical Center ED between January 2018 and April 2021. Generalized estimating equation logistic regression modeling was conducted to evaluate associations between race and use of restraints. RESULTS: Of 348,384 ED visits (22.9% White, 46.7% Black, 23.1% Hispanic), 1852 (0.5%) had an associated physical restraint order. Multivariable models showed significant interactions (p = 0.02) between race/ethnicity, behavioral health diagnosis, and sex on the primary outcome of physical restraint. Stratified analysis revealed that among patients with no behavioral health diagnoses, Black (odds ratio [OR] 0.50, 95% confidence interval [CI] 0.34-0.72, p = 0.0003) and Hispanic (OR 0.35, 95% CI 0.20-0.63, p = 0.0004) patients had lower odds of restraint than White patients. Among female patients with a mental health and/or substance use disorder diagnosis, Black (OR 1.95, 95% CI 1.49-2.54, p < 0.0001) and Hispanic (OR 2.13, 95% CI 1.49-3.03, p < 0.0001) patients had higher odds of restraint than White patients. Similar trends were observed for Black male patients (OR 1.60, 95% CI 1.34-1.91, p < 0.0001) but not for Hispanic male patients (OR 0.96, 95% CI 0.73-1.26, p = 0.77) with behavioral health diagnoses who had similar odds of restraint to White patients. Additional factors associated with physical restraint include younger age, public or lack of insurance, and ED visits during the pandemic. CONCLUSIONS: Racial disparities exist in restraint utilization at this minority-serving safety-net hospital; however, these disparities are modified by sex and by behavioral health diagnoses. The reasons for these disparities may be multifactorial and warrant further investigation.

Performance of three screening tools to predict COVID-19 positivity in emergency department patients.

BACKGROUND: COVID-19 symptoms vary widely. This retrospective study assessed which of three clinical screening tools-a nursing triage screen (NTS), an ED review of systems (ROS) performed by physicians and physician assistants and a standardised ED attending (ie, consultant) physician COVID-19 probability assessment (PA)-best identified patients with COVID-19 on a subsequent reverse transcription PCR (RT-PCR) confirmation. METHODS: All patients admitted to Boston Medical Center from the ED between 27 April 2020 and 17 May 2020 were included. Sensitivity, specificity and positive predictive value (PPV) and negative predictive value (NPV) were calculated for each method. Logistic regression assessed each tool's performance. RESULTS: The attending physician PA had higher sensitivity (0.62, 95% CI 0.53 to 0.71) than the NTS (0.46, 95% CI 0.37 to 0.56) and higher specificity (0.76, 95% CI 0.72 to 0.80) than the NTS (0.71, 95% CI 0.66 to 0.75) and ED ROS (0.62, 95% CI 0.58 to 0.67). Categorisation as moderate or high probability on the ED physician PA was associated with the highest odds of having COVID-19 in regression analyses (adjusted OR=4.61, 95% CI 3.01 to 7.06). All methods had a low PPV (ranging from 0.26 for the ED ROS to 0.40 for the attending physician PA) and a similar NPV (0.84 for both the NTS and the ED ROS, and 0.89 for the attending physician PA). CONCLUSION: The ED attending PA had higher sensitivity and specificity than the other two methods, but none was accurate enough to replace a COVID-19 RT-PCR test in a clinical setting where transmission control is crucial. Therefore, we recommend universal COVID-19 testing prior to all admissions.

Adaptation of a Simulation Model and Checklist to Assess Pediatric Emergency Care Performance by Prehospital Teams.

INTRODUCTION: Simulation tools to assess prehospital team performance and identify patient safety events are lacking. We adapted a simulation model and checklist tool of individual paramedic performance to assess prehospital team performance and tested interrater reliability. METHODS: We used a modified Delphi process to adapt 3 simulation cases (cardiopulmonary arrest, seizure, asthma) and checklist to add remote physician direction, target infants, and evaluate teams of 2 paramedics and 1 physician. Team performance was assessed with a checklist of steps scored as complete/incomplete by raters using direct observation or video review. The composite performance score was the percentage of completed steps. Interrater percent agreement was compared with the original tool. The tool was modified, and raters trained in iterative rounds until composite performance scoring agreement was 0.80 or greater (scale <0.20 = poor; 0.21-0.39 = fair, 0.40-0.59 = moderate; 0.60-0.79 = good; 0.80-1.00 = very good). RESULTS: We achieved very good interrater agreement for scoring composite performance in 2 rounds using 6 prehospital teams and 4 raters. The original 175 step tool was modified to 171 steps. Interrater percent agreement for the final modified tool approximated the original tool for the composite checklist (0.80 vs. 0.85), cardiopulmonary arrest (0.82 vs. 0.86), and asthma cases (0.80 vs. 0.77) but was lower for the seizure case (0.76 vs. 0.91). Most checklist items (137/171, 80%) had good-very good agreement. Among 34 items with fair-moderate agreement, 15 (44%) related to patient assessment, 9 (26%) equipment use, 6 (18%) medication delivery, and 4 (12%) cardiopulmonary resuscitation quality. CONCLUSIONS: The modified checklist has very good agreement for assessing composite prehospital team performance and can be used to test effects of patient safety interventions.

Patient navigation to address sociolegal barriers for patients with cancer: A comparative-effectiveness study.

BACKGROUND: Sociolegal barriers to cancer care are defined as health-related social needs like affordable healthy housing, stable utility service, and food security that may be remedied by public policy, law, regulation, or programming. Legal support has not been studied in cancer care. METHODS: The authors conducted a randomized controlled trial of patients who had newly diagnosed cancer at a safety-net medical center in Boston from 2014 through 2017, comparing standard patient navigation versus enhanced navigation partnered with legal advocates to identify and address sociolegal barriers. English-speaking, Spanish-speaking, or Haitian Creole-speaking patients with breast and lung cancer were eligible within 30 days of diagnosis. The primary outcome was timely treatment within 90 days of diagnosis. Secondary outcomes included patient-reported outcomes (distress, cancer-related needs, and satisfaction with navigation) at baseline and at 6 months. RESULTS: In total, 201 patients with breast cancer and 19 with lung cancer enrolled (response rate, 78%). The mean patient age was 55 years, 51% of patients were Black and 22% were Hispanic, 20% spoke Spanish and 8% spoke Haitian Creole, 73% had public health insurance, 77% reported 1 or more perceived sociolegal barrier, and the most common were barriers to housing and employment. Ninety-six percent of participants with breast cancer and 73% of those with lung cancer initiated treatment within 90 days. No significant effect of enhanced navigation was observed on the receipt of timely treatment among participants with breast cancer (odds ratio, 0.88; 95% CI, 0.17-4.52) or among those with lung cancer (odds ratio, 4.00; 95% CI, 0.35-45.4). No differences in patient-reported outcomes were observed between treatment groups. CONCLUSIONS: Navigation enhanced by access to legal consultation and support had no impact on timely treatment, patient distress, or patient needs. Although most patients reported sociolegal barriers, few required intensive legal services that could not be addressed by navigators. LAY SUMMARY: In patients with cancer, the experience of sociolegal barriers to care, such as unstable housing, utility services, or food insecurity, is discussed. Addressing these barriers through legal information and assistance may improve care. This study compares standard patient navigation versus enhanced navigation partnered with legal advocates for patients with breast and lung cancers. Almost all patients in both navigation groups received timely care and also reported the same levels of distress, needs, and satisfaction with navigation. Although 75% of patients in the study had at least 1 sociolegal barrier identified, few required legal advocacy beyond what a navigator who received legal information and coaching could provide.

Calculating power for the Finkelstein and Schoenfeld test statistic for a composite endpoint with two components.

The Finkelstein and Schoenfeld (FS) test is a popular generalized pairwise comparison approach to analyze prioritized composite endpoints (eg, components are assessed in order of clinical importance). Power and sample size estimation for the FS test, however, are generally done via simulation studies. This simulation approach can be extremely computationally burdensome, compounded by increasing number of composite endpoints and with increasing sample size. Here we propose an analytical solution to calculate power and sample size for commonly encountered two-component hierarchical composite endpoints. The power formulas are derived assuming underlying distributions in each of the component outcomes on the population level, which provide a computationally efficient and practical alternative to the standard simulation approach. Monte Carlo simulation results demonstrate that performance of the proposed power formulas are consistent with that of the simulation approach, and have generally desirable objective properties including robustness to mis-specified distributional assumptions. We demonstrate the application of the proposed formulas by calculating power and sample size for the Transthyretin Amyloidosis Cardiomyopathy Clinical Trial.

Racial equity in linkage to inpatient opioid use disorder treatment in patients that received emergency care.

OBJECTIVES: Opioid use disorder (OUD) is a national epidemic, and Black and Hispanic patients are less likely to receive treatment when compared to white patients. In this study, race was used as a proxy to assess potential effects of racism on the referral process for OUD treatment. Our primary aim was to examine whether Black or Hispanic patients experienced increased barriers to inpatient OUD detoxification (detox) placement at a community-integrated, substance use disorder support program based in an emergency department (ED). Our secondary aim was to determine if Black and Hispanic patients were more likely to have >3 referrals. METHODS: This retrospective cohort study was conducted at a large urban safety-net hospital and included patients seen in the ED from July 2018 to September 2019 with ICD-10 codes for an opioid-related visit and who sought placement to inpatient detox. A generalized linear mixed model controlling for multiple visits, age, sex, insurance, time, day of week, and time of year was used to assess the association between race/ethnicity and hypothesized barriers to placement. The proportion of patients with >3 visits for referral to inpatient detox was compared between Black and Hispanic patients and white patients using a chi-squared test. RESULTS: We identified 1733 encounters from 782 unique patients seeking connection to inpatient detox for OUD. Of the 1733 encounters, 45% were among Black and Hispanic patients. Hispanic and Black men had significantly lower odds of having a barrier to inpatient OUD detox than white men (OR = 0.734, 95% CI 0.542-0.995). No significant difference was found for Hispanic and Black women (OR = 1.212, 95% CI 0.705-2.082). More Black and Hispanic patients experienced >3 referrals to inpatient detox compared to white patients (19.2% vs 12.9%, p = 0.016). CONCLUSIONS: This study suggests in the context of near-universal health insurance coverage, an ED-based OUD support program staffed by diverse community members can mitigate inequities in access to inpatient detox. However, the increased number of ED visits for OUD detox placement by Black and Hispanic patients suggests racial inequities in OUD treatment exist after linkage to care. Additional research should explore the causes, specifically structural and interpersonal racism, and determine solutions to address racial inequities in detox placement as well as maintenance in treatment programs.

Persistent inter-observer variability of breast density assessment using BI-RADS® 5th edition guidelines.

OBJECTIVES: Due to most states' legislation, mammographic density categorization has potentially far-reaching implications, but remains subjective based on BIRADS® guidelines. We aimed to determine 1) effect of BI-RADS® 5th edition (5th-ed) vs 4th-edition (4th-ed) guidelines on reader agreement regarding density assessment; 2) 5th-ed vs 4th-ed density distribution, and visual vs quantitative assessment agreement; 3) agreement between experienced vs less experienced readers. METHODS: In a retrospective review, six breast imaging radiologists (BIR) (23-30 years' experience) visually assessed density of 200 screening mammograms performed September 2012-January 2013 using 5th-ed guidelines. Results were compared to 2016 data of the same readers evaluating the same mammograms using 4th-ed guidelines after a training module. 5th-ed density categorization by seven junior BIR (1-5 years' experience) was compared to eight experienced BIR. Nelson et al.'s kappas (κm, κw), Fleiss' κF, and Cohen's κ were calculated. Quantitative density using Volpara was compared with reader assessments. RESULTS: Inter-reader weighted agreement using 5th-ed is moderately strong, 0.73 (κw, s.e. = 0.01), similar to 4th-ed, 0.71 (κw, s.e. = 0.03). Intra-reader Cohen's κ is 0.23-0.34, similar to 4th-ed. Binary not-dense vs dense categorization, using 5th-ed results in higher dense categorization vs 4th-ed (p < 0.001). 5th-ed density distribution results in higher numbers in categories B/C vs 4th-ed (p < 0.001). Distribution for 5th-ed does not differ based on reader experience (p = 0.09). Reader vs quantitative weighted agreement is similar (5th-ed, Cohen's κ = 0.76-0.85; 4th-ed, Cohen's κ = 0.68-0.83). CONCLUSION: There is persistent subjectivity of visually assessed mammographic density using 5th-ed guidelines; experience does not correlate with better inter-reader agreement.

Methods of assessing categorical agreement between correlated screening tests in clinical studies.

Advances in breast imaging and other screening tests have prompted studies to evaluate and compare the consistency between experts' ratings of existing with new screening tests. In clinical settings, medical experts make subjective assessments of screening test results such as mammograms. Consistency between experts' ratings is evaluated by measures of inter-rater agreement or association. However, conventional measures, such as Cohen's and Fleiss' kappas, are unable to be applied or may perform poorly when studies consist of many experts, unbalanced data, or dependencies between experts' ratings exist. Here we assess the performance of existing approaches including recently developed summary measures for assessing the agreement between experts' binary and ordinal ratings when patients undergo two screening procedures. Methods to assess consistency between repeated measurements by the same experts are also described. We present applications to three large-scale clinical screening studies. Properties of these agreement measures are illustrated via simulation studies. Generally, a model-based approach provides several advantages over alternative methods including the ability to flexibly incorporate various measurement scales (i.e. binary or ordinal), large numbers of experts and patients, sparse data, and robustness to prevalence of underlying disease.

False negative point-of-care urine pregnancy tests in an urban academic emergency department: a retrospective cohort study.

STUDY OBJECTIVE: To determine the prevalence of false negative point-of-care (POC) urine pregnancy tests among emergency department (ED) patients and among those with abdominal pain or vaginal bleeding. METHODS: We identified all female patients, ages 14-50 years without prior hysterectomy who had a negative POC urine pregnancy test (beta subunit of human chorionic gonadotropin [ß-hCG]) performed by trained clinical staff in the ED between September 1, 2017 and December 31, 2018, as well as a subgroup we defined a priori as "high risk" for early pregnancy complications based on a triage chief complaint (text) of abdominal pain or vaginal bleeding. We identified those with a positive urine ß-hCG, serum ß-hCG >5 mIU/mL, or a diagnosis of pregnancy within 3 months of the initial ED visit (index visit). We used structured chart review with American College of Obstetrics and Gynecology guidelines to determine pregnancy diagnosis and outcomes (ectopic, intrauterine, abnormal including spontaneous abortion, and unknown), the date of conception, and whether the pregnancy was present at the index visit. RESULTS: Of 10,924 visits with a negative urine pregnancy test result that were screened for a pregnancy outcome, 171 (1.6%, 95% confidence interval [CI] = 1.4, 1.8) had a pregnancy present at the index visit. Diagnoses were ectopic (n = 12, 7.0%), intrauterine (n = 71, 41.5%), abnormal (n = 77, 45.0%), and unknown (n = 11, 6.4%). Of the 2732 patients with high-risk complaints, 97 (3.6%, 95% CI = 2.9, 4.3) had a pregnancy present at the index visit (relative risk of a pregnancy diagnosis 3.9, 95% CI = 2.9,5.3), including 10/12 ectopic (83%), 58/77 abnormal (75%), and 25/71 intrauterine pregnancies (35%). Serum ß-hCG ranged from 2 mIU/mL to above assay (median = 119.5, interquartile range = 957.5). CONCLUSION: Although false negative urine pregnancy tests were uncommon, multiple pregnancy diagnoses were missed, including ectopic pregnancies. False negatives were more common among patients with abdominal pain or vaginal bleeding. Concurrent serum ß-hCG levels demonstrated a broad distribution.

Measuring rater bias in diagnostic tests with ordinal ratings.

Diagnostic tests are frequently reliant upon the interpretation of images by skilled raters. In many clinical settings, however, the variability observed between experts' ratings plays a detrimental role in the degree of confidence in these interpretations, leading to uncertainty in the diagnostic process. For example, in breast cancer testing, radiologists interpret mammographic images, while breast biopsy results are examined by pathologists. Each of these procedures involves elements of subjectivity. We propose here a flexible two-stage Bayesian latent variable model to investigate how the skills of individual raters impact the diagnostic accuracy of image-related testing in large-scale medical testing studies. A strength of the proposed model is that the true disease status of a patient within a reasonable time frame may or may not be known. In these studies, many raters each contribute classifications on a large sample of patients using a defined ordinal grading scale, leading to a complex correlation structure between ratings. Our modeling approach considers the different sources of variability contributed by experts and patients while accounting for correlations present between ratings and patients, in contrast to currently available methods. We propose a novel measure of a rater's ability (magnifier) that, in contrast to conventional measures of sensitivity and specificity, is robust to the underlying prevalence of disease in the population, providing an alternative measure of diagnostic accuracy across patient populations. Extensive simulation studies demonstrate lower bias in estimation of parameters and measures of accuracy, and illustrate outperformance of the proposed model when compared with existing models. Receiver operator characteristic curves are derived to assess the diagnostic accuracy of individual experts and their overall performance. Our proposed modeling approach is applied to a large breast imaging study for known disease status and a uterine cancer dataset for unknown disease status.

A Comparison of Etomidate, Ketamine, and Methohexital in Emergency Department Rapid Sequence Intubation.

BACKGROUND: Rapid sequence intubation (RSI) is routinely used for emergent airway management in the emergency department (ED). It involves the use of induction, and paralytic agents help facilitate endotracheal tube placement. OBJECTIVE: In response to a previous national drug shortage resulting in the use of alternative induction agents for RSI, we describe the effectiveness and safety of ED RSI with ketamine or methohexital compared with etomidate. METHODS: We conducted a retrospective, single-center observational study from March 1-August 31, 2012 describing RSI with etomidate, ketamine, and methohexital. All adult patients undergoing RSI in the ED who received etomidate prior to its shortage and methohexital or ketamine during the shortage were included. RESULTS: The study included 47, 9, and 26 patients in the etomidate, ketamine, and methohexital groups, respectively. Successful intubation on the first attempt occurred in 74.5%, 55.6%, and 73.1% of the etomidate, ketamine, and methohexital groups, respectively. The mean number of intubation attempts and time to intubation seemed to be similar in all groups. At least three intubation attempts were required in 22.2% and 7.7% of the ketamine and methohexital groups, respectively, compared with none in the etomidate group. Two aspirations were observed in the etomidate group. CONCLUSION: Methohexital and etomidate had similar rates of successful intubation on the first attempt and seem to be more effective than ketamine. Etomidate may reduce the need for three or more intubation attempts. Larger, prospective studies are needed to determine if ketamine or methohexital are more effective than etomidate for RSI.

Point-of-care Lung Ultrasound Is More Sensitive than Chest Radiograph for Evaluation of COVID-19.

INTRODUCTION: Current recommendations for diagnostic imaging for moderately to severely ill patients with suspected coronavirus disease 2019 (COVID-19) include chest radiograph (CXR). Our primary objective was to determine whether lung ultrasound (LUS) B-lines, when excluding patients with alternative etiologies for B-lines, are more sensitive for the associated diagnosis of COVID-19 than CXR. METHODS: This was a retrospective cohort study of all patients who presented to a single, academic emergency department in the United States between March 20 and April 6, 2020, and received LUS, CXR, and viral testing for COVID-19 as part of their diagnostic evaluation. The primary objective was to estimate the test characteristics of both LUS B-lines and CXR for the associated diagnosis of COVID-19. Our secondary objective was to evaluate the proportion of patients with COVID-19 that have secondary LUS findings of pleural abnormalities and subpleural consolidations. RESULTS: We identified 43 patients who underwent both LUS and CXR and were tested for COVID-19. Of these, 27/43 (63%) tested positive. LUS was more sensitive (88.9%, 95% confidence interval (CI), 71.1-97.0) for the associated diagnosis of COVID-19 than CXR (51.9%, 95% CI, 34.0-69.3; p = 0.013). LUS and CXR specificity were 56.3% (95% CI, 33.2-76.9) and 75.0% (95% CI, 50.0-90.3), respectively (p = 0.453). Secondary LUS findings of patients with COVID-19 demonstrated 21/27 (77.8%) had pleural abnormalities and 10/27 (37%) had subpleural consolidations. CONCLUSION: Among patients who underwent LUS and CXR, LUS was found to have a higher sensitivity than CXR for the evaluation of COVID-19. This data could have important implications as an aid in the diagnostic evaluation of COVID-19, particularly where viral testing is not available or restricted. If generalizable, future directions would include defining how to incorporate LUS into clinical management and its role in screening lower-risk populations.

Measuring intrarater association between correlated ordinal ratings.

Variability between raters' ordinal scores is commonly observed in imaging tests, leading to uncertainty in the diagnostic process. In breast cancer screening, a radiologist visually interprets mammograms and MRIs, while skin diseases, Alzheimer's disease, and psychiatric conditions are graded based on clinical judgment. Consequently, studies are often conducted in clinical settings to investigate whether a new training tool can improve the interpretive performance of raters. In such studies, a large group of experts each classify a set of patients' test results on two separate occasions, before and after some form of training with the goal of assessing the impact of training on experts' paired ratings. However, due to the correlated nature of the ordinal ratings, few statistical approaches are available to measure association between raters' paired scores. Existing measures are restricted to assessing association at just one time point for a single screening test. We propose here a novel paired kappa to provide a summary measure of association between many raters' paired ordinal assessments of patients' test results before versus after rater training. Intrarater association also provides valuable insight into the consistency of ratings when raters view a patient's test results on two occasions with no intervention undertaken between viewings. In contrast to existing correlated measures, the proposed kappa is a measure that provides an overall evaluation of the association among multiple raters' scores from two time points and is robust to the underlying disease prevalence. We implement our proposed approach in two recent breast-imaging studies and conduct extensive simulation studies to evaluate properties and performance of our summary measure of association.

Loop Drainage Is Noninferior to Traditional Incision and Drainage of Cutaneous Abscesses in the Emergency Department.

BACKGROUND: There is limited research on loop drainage (LD) compared to incision and drainage (I&D) for treatment of cutaneous abscesses. We investigated whether LD was noninferior to I&D for abscess resolution and whether there was any difference in repeat ED visits or complication rates between these techniques. METHODS: We performed a prospective randomized controlled trial, using a convenience sample at an urban academic emergency department (ED). Subjects over 18 years who presented for first-time management of an abscess were eligible. Patients requiring specialist drainage or hospital admission or had previous treatment for the abscess were excluded. Enrolled subjects were seen 2 weeks after treatment for blinded reevaluation of abscess resolution, and the electronic medical record was reviewed for return ED visits/abscess complications. RESULTS: Of 2,889 patients screened, 238 subjects consented and were randomized to LD or I&D. Abscess resolution was achieved in 53/65 (81.5%) of patients in the I&D arm, compared to 66/75 (88%) in the LD arm. Fewer patients in the LD group compared to the I&D group returned to the ED for abscess-related management during the following 14 days (37.3% vs 67.1%, p = 0.002). Among returning subjects, there was a significant difference in mean visits per subject between LD and I&D groups (0.5 vs. 1.2, p = 0.001). There were fewer complications among LD than I&D subjects (9.3% vs. 24.6%, p = 0.01). CONCLUSION: Our study provides evidence that LD is noninferior to I&D in achieving complete abscess resolution at 14 days and is associated with fewer return ED visits and fewer complications. This makes it an attractive alternative treatment option for abscesses.

Measuring Agreement Among Prehospital Providers and Physicians in Patient Capacity Determination.

OBJECTIVES: If a patient wishes to refuse treatment in the prehospital setting, prehospital providers and consulting emergency physicians must establish that the patient possesses the capacity to do so. The objective of this study is to assess agreement among prehospital providers and emergency physicians in performing patient capacity assessments. METHODS: This study involved 139 prehospital providers and 28 emergency medicine physicians. Study participants listened to 30 medical control calls pertaining to patient capacity and were asked to interpret whether the patients in the scenarios had the capacity to refuse treatment. Participants also reported their comfort level using modified Likert scales. Inter-rater reliability was calculated utilizing Fleiss' and Model B kappa statistics. Fisher's exact tests were used to calculate p-values comparing the proportion in each cohort that responded "no capacity." Primary outcomes included inter-rater reliability in the physician and prehospital provider cohorts. RESULTS: The inter-rater agreement between the physicians was low (Fleiss' kappa = 0.31, standard error [SE] =0.06; model-based kappa = 0.18, SE = 0.04). Agreement was similarly low for the 135 prehospital providers (Fleiss' kappa = 0.30, SE = 0.06; model-based kappa = 0.28, SE = 0.04). The difference between the proportion of physicians and prehospital providers who responded "no capacity" was statistically significant in five of 30 scenarios. Median prehospital provider and physician confidence, on a 1 to 4 scale, was 2.00 (Q1-Q3 = 1.00-3.00 for prehospital providers and Q1-Q3 =1.0-2.0 for physicians). CONCLUSIONS: There was poor inter-rater reliability in capacity determination between and among the prehospital provider and physician cohorts. This suggests that there is need for additional study and standardization of this task.

Bayesian hierarchical latent class models for estimating diagnostic accuracy.

The diagnostic accuracy of a test or rater has a crucial impact on clinical decision making. The assessment of diagnostic accuracy for multiple tests or raters also merits much attention. A Bayesian hierarchical conditional independence latent class model for estimating sensitivities and specificities for a large group of tests or raters is proposed, which is applicable to both with-gold-standard and without-gold-standard situations. Through the hierarchical structure, not only are the sensitivities and specificities of individual tests estimated, but also the diagnostic performance of the whole group of tests. For a small group of tests or raters, the proposed model is further extended by introducing pairwise covariances between tests to improve the fitting and to allow for more modeling flexibility. Correlation residual analysis is applied to detect any significant covariance between multiple tests. Just Another Gibbs Sampler (JAGS) implementation is efficiently adopted for both models. Three real data sets from literature are analyzed to explicitly illustrate the proposed methods.

A paired kappa to compare binary ratings across two medical tests.

Agreement between experts' ratings is an important prerequisite for an effective screening procedure. In clinical settings, large-scale studies are often conducted to compare the agreement of experts' ratings between new and existing medical tests, for example, digital versus film mammography. Challenges arise in these studies where many experts rate the same sample of patients undergoing two medical tests, leading to a complex correlation structure between experts' ratings. Here, we propose a novel paired kappa measure to compare the agreement between the binary ratings of many experts across two medical tests. Existing approaches can accommodate only a small number of experts, rely heavily on Cohen's kappa and Scott's pi measures of agreement, and thus are prone to their drawbacks. The proposed kappa appropriately accounts for correlations between ratings due to patient characteristics, corrects for agreement due to chance, and is robust to disease prevalence and other flaws inherent in the use of Cohen's kappa. It can be easily calculated in the software package R. In contrast to existing approaches, the proposed measure can flexibly incorporate large numbers of experts and patients by utilizing the generalized linear mixed models framework. It is intended to be used in population-based studies, increasing efficiency without increasing modeling complexity. Extensive simulation studies demonstrate low bias and excellent coverage probability of the proposed kappa under a broad range of conditions. Methods are applied to a recent nationwide breast cancer screening study comparing film mammography to digital mammography.

Marginal analysis of ordinal clustered longitudinal data with informative cluster size.

The issue of informative cluster size (ICS) often arises in the analysis of dental data. ICS describes a situation where the outcome of interest is related to cluster size. Much of the work on modeling marginal inference in longitudinal studies with potential ICS has focused on continuous outcomes. However, periodontal disease outcomes, including clinical attachment loss, are often assessed using ordinal scoring systems. In addition, participants may lose teeth over the course of the study due to advancing disease status. Here we develop longitudinal cluster-weighted generalized estimating equations (CWGEE) to model the association of ordinal clustered longitudinal outcomes with participant-level health-related covariates, including metabolic syndrome and smoking status, and potentially decreasing cluster size due to tooth-loss, by fitting a proportional odds logistic regression model. The within-teeth correlation coefficient over time is estimated using the two-stage quasi-least squares method. The motivation for our work stems from the Department of Veterans Affairs Dental Longitudinal Study in which participants regularly received general and oral health examinations. In an extensive simulation study, we compare results obtained from CWGEE with various working correlation structures to those obtained from conventional GEE which does not account for ICS. Our proposed method yields results with very low bias and excellent coverage probability in contrast to a conventional generalized estimating equations approach.

Point-of-care influenza testing does not significantly shorten time to disposition among patients with an influenza-like illness.

INTRODUCTION: Availability of anti-viral agents and need to isolate infected patients increases the need to confirm the diagnosis of influenza before determining patient disposition. OBJECTIVES: We sought to determine if time-to-disposition (TTD) was shorter among patients tested for influenza using an Emergency Department (ED) Point-of-care (POC) test compared to core laboratory (lab) test and to determine difference in antibiotic use between groups. METHODS: We prospectively enrolled a convenience sample of ED patients for whom influenza testing was ordered during influenza season 2017. Participants were randomized to POC or lab. Data collected included demographics, chief complaint, influenza test results, turnaround time (TAT), whether antibiotics were given, and TTD. Descriptive statistics were calculated and group comparisons conducted using chi squared and Wilcoxon Rank Sum tests. RESULTS: Study population included 100 in the POC group and 97 in the lab group. Demographics were similar between POC and lab participants. More flu positive results were reported in the POC group compared to the lab group (51.0% vs. 33.0% p = 0.01). The median TTD was 146.5 min (IQR 98.5) for POC group and 165.5 min (IQR 127) for lab group (p = 0.26). The median TAT was 30.5 min (IQR 7.5) for POC group and 106.0 min (IQR 55) for core lab group (p = 0.001). Antibiotics were given to 14.0% of POC participants and 14.4% of lab participants (p = 0.93). CONCLUSIONS: Although use of a POC influenza test provided more rapid TAT than use of a core lab test, there was no significant difference in TTD or antibiotic use between groups.

Central venous catheter placement after ultrasound guided peripheral IV placement for difficult vascular access patients.

OBJECTIVES: Ultrasound guided peripheral intravenous catheters (USPIV) are frequently utilized in the Emergency Department (ED) and lead to reduced central venous catheter (CVC) placements. USPIVs, however, are reported to have high failure rates. Our primary objective was to determine the proportion of patients that required CVC after USPIV. Our secondary objective was to determine if classic risk factors for difficult vascular access were predictive of future CVC placement. METHODS: We performed a retrospective review for patients treated at a large academic hospital. Patients were identified by electronic health record and were restricted to age older than 21 years, had received USPIV, and admittance. Exclusion criteria included an existing CVC. Descriptive statistics, t-tests, chi-square proportions, and logistic regression were performed to test associations. RESULTS: Of 363 eligible patients, 20 were excluded allowing for 343 for analysis. Of 343, 45 (13.1% 95% CI 9.9-17.1%) required CVC after USPIV. For secondary outcomes, no expected characteristics (diabetes, end-stage renal disease, IV drug abuse, peripheral vascular disease, or sickle cell disease) were predictive of CVC placement. The only predictive variables were admission to ICU/stepdown and length of stay. Each additional day of hospitalization had an OR 1.11 (95% CI 1.06-1.16%) of having a CVC placed. CONCLUSION: Of those admitted after USPIV placement, approximately 7 out of every 8 patients did not require a subsequent CVC. Of the nearly 1 in 8 patients that required a CVC, factors associated with CVC placement were admission to a higher level of care and length of stay.