Harnessing the Electronic Health Record and Computerized Provider Order Entry Data for Resource Management During the COVID-19 Pandemic: Development of a Decision Tree

Background

The COVID-19 pandemic has resulted in shortages of diagnostic tests, personal protective equipment, hospital beds, and other critical resources.

Objective

We sought to improve the management of scarce resources by leveraging electronic health record (EHR) functionality, computerized provider order entry, clinical decision support (CDS), and data analytics.

Methods

Due to the complex eligibility criteria for COVID-19 tests and the EHR implementation–related challenges of ordering these tests, care providers have faced obstacles in selecting the appropriate test modality. As test choice is dependent upon specific patient criteria, we built a decision tree within the EHR to automate the test selection process by using a branching series of questions that linked clinical criteria to the appropriate SARS-CoV-2 test and triggered an EHR flag for patients who met our institutional persons under investigation criteria.

Results

The percentage of tests that had to be canceled and reordered due to errors in selecting the correct testing modality was 3.8% (23/608) before CDS implementation and 1% (262/26,643) after CDS implementation (P<.001). Patients for whom multiple tests were ordered during a 24-hour period accounted for 0.8% (5/608) and 0.3% (76/26,643) of pre- and post-CDS implementation orders, respectively (P=.03). Nasopharyngeal molecular assay results were positive in 3.4% (826/24,170) of patients who were classified as asymptomatic and 10.9% (1421/13,074) of symptomatic patients (P<.001). Positive tests were more frequent among asymptomatic patients with a history of exposure to COVID-19 (36/283, 12.7%) than among asymptomatic patients without such a history (790/23,887, 3.3%; P<.001).

Conclusions

The leveraging of EHRs and our CDS algorithm resulted in a decreased incidence of order entry errors and the appropriate flagging of persons under investigation. These interventions optimized reagent and personal protective equipment usage. Data regarding symptoms and COVID-19 exposure status that were collected by using the decision tree correlated with the likelihood of positive test results, suggesting that clinicians appropriately used the questions in the decision tree algorithm.

Care pathways and healthcare management in a COVID-19 triage psychiatric inpatient ward at south london and maudsley nhs foundation trust

Introduction

The COVID-19 pandemic has enforced the restructuring of inpatient psychiatric services. In the UK, the South London and Maudsley NHS Foundation Trust has introduced a triage ward system to ensure all patients have a COVID test prior to admission to the general ward with the aim to reduce COVID transmission amongst psychiatric inpatients.

Objectives

To characterise the flow of patients through a COVID-19 psychiatric triage ward and the protocol of assessment and management used.

Methods

Descriptive analysis of patients admitted to a COVID-19 triage ward since its creation.

Results

The caseload of patients admitted to the COVID-19 triage ward since its inception will be presented. This will include the profile of patients admitted, their status (formal/informal) and their acceptance of COVID-19 tests. The protocol followed at this COVID-19 triage ward will be presented, and the challenges faced and suggestions to overcome them will be discussed.

Conclusions

This presentation aims to share the workflow and protocols adopted at a COVID-19 triage ward in the UK, discussing challenges experienced as well as good practices.

Conflict of interest

No significant relationships.

Delayed Diagnosis of Injury in Pediatric Trauma Patients at a Level I Trauma Center

BACKGROUND

Trauma care per Advanced Trauma Life Support addresses immediate threats to life. Occasionally, delays in injury diagnosis occur. Delayed diagnosis of injury (DDI) is a common quality indicator in trauma care, and pediatric DDI data are sparse.

OBJECTIVE

Our aim was to describe the DDI rate in a severely injured pediatric trauma population and identify any factors associated with DDI in the pediatric population.

METHODS

A prospective cohort of trauma activations in 0- to 16-year-old patients admitted to a pediatric level I trauma center over 12 months with injuries prospectively recorded were followed during admission to identify DDI.

RESULTS

A total of 170 trauma activations were enrolled. Twelve patients had type I DDI (7.1%), 15 patients had type II DDI (8.8%), and 5 patients had both type I and type II DDI (2.9%). DDI patients had twice as many injuries and higher Injury Severity Scores (ISS) as non-DDI patients. DDI patients were more likely to require intensive care unit (ICU) admission, longer hospital stay, and ventilator support. Controlling for age and ISS in multivariate analysis, the number of injuries found and requiring a ventilator were significantly associated with DDI.

CONCLUSIONS

This prospective study found a type I DDI rate of 7.1% and a type II DDI rate of 8.8% in the pediatric population. DDI patients had a greater number of injuries, higher ISS, higher rate of ICU admission, and were more likely to require mechanical ventilation. This study adds prospective data to the pediatric DDI literature, increases provider awareness of pediatric DDI, and lays the foundation for future study and quality improvement.

Neural Networks to Predict Radiographic Brain Injury in Pediatric Patients Treated with Extracorporeal Membrane Oxygenation

Brain injury is a significant source of morbidity and mortality for pediatric patients treated with Extracorporeal Membrane Oxygenation (ECMO). Our objective was to utilize neural networks to predict radiographic evidence of brain injury in pediatric ECMO-supported patients and identify specific variables that can be explored for future research. Data from 174 ECMO-supported patients were collected up to 24 h prior to, and for the duration of, the ECMO course. Thirty-five variables were collected, including physiological data, markers of end-organ perfusion, acid-base homeostasis, vasoactive infusions, markers of coagulation, and ECMO-machine factors. The primary outcome was the presence of radiologic evidence of moderate to severe brain injury as established by brain CT or MRI. This information was analyzed by a neural network, and results were compared to a logistic regression model as well as clinician judgement. The neural network model was able to predict brain injury with an Area Under the Curve (AUC) of 0.76, 73% sensitivity, and 80% specificity. Logistic regression had 62% sensitivity and 61% specificity. Clinician judgment had 39% sensitivity and 69% specificity. Sequential feature group masking demonstrated a relatively greater contribution of physiological data and minor contribution of coagulation factors to the model's performance. These findings lay the foundation for further areas of research directions.

Adaptive immune cell activation in acute pediatric traumatic brain injury

Traumatic brain injury (TBI) is mediated by mechanical injury to the brain resulting in tissue shearing, and characterized by neuronal injury and inflammation. We hypothesize that TBI activates brain-targeting immune responses which can induce secondary neuronal injury. Blood and cerebral spinal fluid (external ventricular drain, EVD) were collected from pediatric TBI patients (pTBI, 1–18 yrs.) at days 1, 3 and 5 following TBI and EVD placement. The blood from sick controls (0–18 yrs, no CNS injury) were collected at day 1 post injury. pTBI patients exhibited a higher quantity of leukocytes (CD45+) relative to controls (p=0.01) on day 1 post-TBI, indicating an immediate and robust immune modulation due to CNS injury, that decreased at days 3 and 5, relative to day 1. Major subsets of adaptive immune cells (CD3, CD4, CD8, and CD19) did not exhibit altered cellularity relative to controls but activated adaptive T cells (CD4+CD161+and CD8+CD161+) were substantially higher at day 1 versus controls. Neuronal autoreactivity was evaluated by labeling peripheral blood mononuclear cells with carboxyfluorescein succinimidyl ester (CFSE) and culturing with neuronal and control antigens. Response to antigen was determined by dilution of CFSE (proliferation) and expression of CD25 (activation). Autoreactivity analysis of pTBI patients revealed a decrease in the CD4 T-cell responses to CNS antigens relative to sick control patients. Interestingly, the CD8 T-cell analysis revealed CNS-specific autoimmune responses absent in pTBI patients which were present in sick controls. Based on these initial observations, we hypothesize that the adaptive immune system is activated upon TBI and can potentially target CNS resident cells in young patients.

Massive transfusion in pediatric trauma: An ATOMAC perspective

BACKGROUND/PURPOSE

Massive transfusion protocols (MTPs) are considered valuable in pediatric trauma. Important questions regarding the survival benefit and optimal blood component ratio remain unknown.

METHODS

The study time frame was January 2007 through December 2013 five Level I Pediatric Trauma Centers reviewed all trauma activations involving children ≤18 years of age. Included were patients who either had the institutional MTP or received >20 mL/kg or > 2 units packed red blood cells (PRBCs).

RESULTS

110/202 qualified for inclusion. Median age was 5.9 years (3.0-11.4). 73% survived to discharge; median hospitalization was 10 (3.1-22.8) days. Survival did not vary by arrival hemoglobin (Hgb), gender or age. Partial prothrombin time (PTT), INR, GCS and injury severity score (ISS) significantly differed for nonsurvivors (all p < 0.05). Logistic regression found increased mortality (OR 3.08 (1.10-8.57), 95% CI; p = 0.031) per unit increase over a 1:1 ratio of pRBC:FFP.

CONCLUSION

In pediatric trauma pRBC:FFP ratio of 1:1 was associated with the highest survival of severely injured children receiving massive transfusion. Ratios 2:1 or ≥3:1 were associated with significantly increased risk of death. These data support a higher proportion of plasma products for pediatric trauma patients requiring massive transfusion.

LEVEL OF EVIDENCE

Level IV.

Factors Associated with Pediatric Mortality from Motor Vehicle Crashes in the United States: A State-Based Analysis

OBJECTIVE

To examine geographic variation in motor vehicle crash (MVC)-related pediatric mortality and identify state-level predictors of mortality.

STUDY DESIGN

Using the 2010-2014 Fatality Analysis Reporting System, we identified passengers <15 years of age involved in fatal MVCs, defined as crashes on US public roads with ≥1 death (adult or pediatric) within 30 days. We assessed passenger, driver, vehicle, crash, and state policy characteristics as factors potentially associated with MVC-related pediatric mortality. Our outcomes were age-adjusted, MVC-related mortality rate per 100 000 children and percentage of children who died of those in fatal MVCs. Unit of analysis was US state. We used multivariable linear regression to define state characteristics associated with higher levels of each outcome.

RESULTS

Of 18 116 children in fatal MVCs, 15.9% died. The age-adjusted, MVC-related mortality rate per 100 000 children varied from 0.25 in Massachusetts to 3.23 in Mississippi (mean national rate of 0.94). Predictors of greater age-adjusted, MVC-related mortality rate per 100 000 children included greater percentage of children who were unrestrained or inappropriately restrained (P < .001) and greater percentage of crashes on rural roads (P = .016). Additionally, greater percentages of children died in states without red light camera legislation (P < .001). For 10% absolute improvement in appropriate child restraint use nationally, our risk-adjusted model predicted >1100 pediatric deaths averted over 5 years.

CONCLUSIONS

MVC-related pediatric mortality varied by state and was associated with restraint nonuse or misuse, rural roads, vehicle type, and red light camera policy. Revising state regulations and improving enforcement around these factors may prevent substantial pediatric mortality.

Validation of the HZETRN code for laboratory exposures with 1A GeV iron ions in several targets

A new version of the HZETRN code capable of validation with HZE ions in either the laboratory or the space environment is under development. The computational model consists of the lowest order asymptotic approximation followed by a Neumann series expansion with non-perturbative corrections. The physical description includes energy loss with straggling, nuclear attenuation, nuclear fragmentation with energy dispersion and downshift. Measurements to test the model were performed at the Alternating Gradient Synchrotron and the NASA Space Radiation Laboratory at Brookhaven National Laboratory with iron ions. Surviving beam particles and produced fragments were measured with solid-state detectors. Beam analysis software has been written to relate the computational results to the measured energy loss spectra of the incident ions for rapid validation of modeled target transmission functions.

Computational methods for the HZETRN code

Asymptotic expansion has been used to simplify the transport of high charge and energy ions for broad beam applications in the laboratory and space. The solution of the lowest order asymptotic term is then related to a Green's function for energy loss and straggling coupled to nuclear attenuation providing the lowest order term in a rapidly converging Neumann series for which higher order collisions terms are related to the fragmentation events including energy dispersion and downshift. The first and second Neumann corrections were evaluated numerically as a standard for further analytic approximation. The first Neumann correction is accurately evaluated over the saddle point whose width is determined by the energy dispersion and located at the downshifted ion collision energy. Introduction of the first Neumann correction leads to significant simplification of the second correction term allowing application of the mean value theorem and a second saddle point approximation. The regular dependence of the second correction spectral dependence lends hope to simple approximation to higher corrections. At sufficiently high energy nuclear cross-section variations are small allowing non-perturbative methods to all orders and renormalization of the second corrections allow accurate evaluation of the full Neumann series.

A space radiation transport method development

Improved spacecraft shield design requires early entry of radiation constraints into the design process to maximize performance and minimize costs. As a result, we have been investigating high-speed computational procedures to allow shield analysis from the preliminary design concepts to the final design. In particular, we will discuss the progress towards a full three-dimensional and computationally efficient deterministic code for which the current HZETRN evaluates the lowest-order asymptotic term. HZETRN is the first deterministic solution to the Boltzmann equation allowing field mapping within the International Space Station (ISS) in tens of minutes using standard finite element method (FEM) geometry common to engineering design practice enabling development of integrated multidisciplinary design optimization methods. A single ray trace in ISS FEM geometry requires 14 ms and severely limits application of Monte Carlo methods to such engineering models. A potential means of improving the Monte Carlo efficiency in coupling to spacecraft geometry is given in terms of re-configurable computing and could be utilized in the final design as verification of the deterministic method optimized design.

An improved Green's function for ion beam transport

Ion beam transport theory allows testing of material transmission properties in the laboratory environment generated by particle accelerators. This is a necessary step in materials development and evaluation for space use. The approximations used in solving the Boltzmann transport equation for the space setting are often not sufficient for laboratory work and those issues are the main emphasis of the present work. In consequence, an analytic solution of the linear Boltzmann equation is pursued in the form of a Green's function allowing flexibility in application to a broad range of boundary value problems. It has been established that simple solutions can be found for high charge and energy (HZE) ions by ignoring nuclear energy downshifts and dispersion. Such solutions were found to be supported by experimental evidence with HZE ion beams when multiple scattering was added. Lacking from the prior solutions were range and energy straggling and energy downshift with dispersion associated with nuclear events. Recently, we have found global solutions including these effects providing a broader class of HZE ion solutions.

A simple model for straggling evaluation

Some straggling models had largely been abandoned in favor of Monte Carlo simulations of straggling which are accurate but time consuming, limiting their application in practice. The difficulty of simple analytic models is the failure to give accurate values past 85% of the particle range. A simple model is derived herein based on a second order approximation upon which rapid analysis tools are developed for improved understanding of material charged particle transmission properties.

Zotepine in the prevention of recurrence: a randomised, double-blind, placebo-controlled study for chronic schizophrenia

RATIONALE

Zotepine is an antipsychotic drug with proven efficacy for treatment of acute episodes of schizophrenia. Antipsychotic drugs also require to be effective in prevention of recurrence.

OBJECTIVE

This trial was designed to compare the effects of zotepine and placebo in the prevention of recurrence of acute episodes in a population of patients with chronic schizophrenia.

METHODS

The study was a double-blind, parallel group, 26-week comparison of zotepine (300 mg daily, with fall back to 150 mg if necessary) versus placebo in 121 patients with chronic schizophrenia and a history of recurrence in the previous 18 months. The primary outcome measure was the time to recurrence. Other evaluations included the brief psychiatric rating scale (BPRS), the scale for the assessment of negative symptoms (SANS), the clinical global impression (CGI) severity and improvement scales, and the Simpson and Angus scale for extrapyramidal symptoms (EPS).

RESULTS

Fewer zotepine patients experienced recurrence over 26 weeks than placebo patients (4 versus 21, respectively). The estimated 26-week risk of recurrence was six times lower for zotepine than placebo (8.7% versus 52.8%; hazard ratio 0.16, 95% CI 0.053, 0.484; P<0.001). Scores on the BPRS and CGI supported the efficacy of zotepine. There was no difference between the treatments with respect to EPS.

CONCLUSIONS

Zotepine is effective in preventing recurrence in patients with chronic schizophrenia. The level of EPS was not different between zotepine and placebo.

A placebo-controlled comparison of zotepine versus chlorpromazine in patients with acute exacerbation of schizophrenia

OBJECTIVE

The aim of this study was to evaluate the efficacy of zotepine in the treatment of acute episodes of schizophrenia.

METHOD

Patients with acute exacerbation of schizophrenia (DSM-III-R criteria; n = 158) were allocated on a random, double-blind basis to receive zotepine (150 or 300 mg/day), chlorpromazine (300 or 600 mg/day) or placebo for 8 weeks. Symptoms were assessed on the BPRS, SANS and CGI scales at baseline and weeks 1, 2, 4, 6 and 8 and patients were assessed at these times for adverse effects. Analysis was by analysis of variance on the intent-to-treat population, with last observation carried forward.

RESULTS

Mean BPRS scores improved statistically significantly more with zotepine than chlorpromazine (point estimate of difference -12.4, 95% CI -18.3 to -6.5) or placebo (point estimate of difference -12.7, 95% CI -18.6 to -6.8). Zotepine produced significantly fewer extrapyramidal symptoms (EPS) than chlorpromazine.

CONCLUSION

Zotepine is an effective antipsychotic with low propensity for EPS.

An investigation into sound attenuation by earmould tubing

A major problem in high-gain post-aural hearing aids is acoustic feedback. One of the sources of feedback is the plastic tubing that guides the sound from the hearing aid hook to the ear canal. The objectives of this study were to determine the effects of increases in tubing wall thickness on the amount of acoustic feedback in the hearing aid system and to characterize earmould tubing. Five different samples of Tygon tubing were tested, of which four had different wall thicknesses; the study also compared Tygon tubing with medical grade PVC tubing. Measurement of the wall thickness, taken from scanning electron micrographs, showed that the wall thickness differed from the National Association of Earmould Laboratories standard by less than 6% and was well within the manufacturer's tolerances. Measurement of sound attenuation showed that an increase in the tubing wall thickness of 0.330 mm (Tygon 13 Standard to Tygon 13 Double Wall) caused an increase in sound attenuation of 2 dB, across the frequency range measured (0.5-4 kHz). It was also found that the medical grade PVC had a 5 dB greater sound attenuation than the equivalent Tygon tubing. While these differences are quite small, they could contribute to a reduction in acoustic feedback when used in conjunction with other hearing aid or earmould improvements.

A new exercise test for the assessment of heart failure: use of a self powered treadmill

Time limited exercise on a self powered treadmill was evaluated as a method of assessing functional capacity in patients with mild heart failure. The characteristics of the treadmill were established by exercising 11 controls at three speeds and two inclinations and comparing oxygen consumption with that on a motorised treadmill under the same conditions. Oxygen consumption on the self powered treadmill at an equivalent speed and inclination was significantly higher because of the work needed to overcome the friction of the belt. Unlike a conventional treadmill, increasing the gradient on the self powered treadmill did not increase oxygen consumption. The distance walked in 12 minutes on the self powered treadmill was measured in eight patients with mild heart failure and ten controls. Maximal oxygen consumption was measured in the same group on a conventional treadmill by a mass spectrometer. There was a significant correlation between the distance walked and maximum oxygen consumption. In patients with mild heart failure the distance travelled in 12 minutes on a self powered treadmill provides a practical, inexpensive, and sensitive method of assessing functional capacity.