Impact of an Electronic Clinical Decision Support Tool for Emergency Department Patients With Pneumonia

Clinical Decision Support as a Prevention Tool for Medication Errors in the Operating Room: A Retrospective Cross-Sectional Study

BACKGROUND

Medication errors in the operating room have high potential for patient harm. While electronic clinical decision support (CDS) software has been effective in preventing medication errors in many nonoperating room patient care areas, it is not yet widely used in operating rooms. The purpose of this study was to determine the percentage of self-reported intraoperative medication errors that could be prevented by CDS algorithms.

METHODS

In this retrospective cross-sectional study, we obtained safety reports involving medication errors documented by anesthesia clinicians between August 2020 and August 2022 at a 1046-bed tertiary care academic medical center. Reviewers classified each medication error by its stage in the medication use process, error type, presence of an adverse medication event, and its associated severity and preventability by CDS. Informational gaps were corroborated by retrospective chart review and disagreements between reviewers were resolved by consensus. The primary outcome was the percentage of errors that were preventable by CDS. Secondary outcomes were preventability by CDS stratified by medication error type and severity.

RESULTS

We received 127 safety reports involving 80 medication errors, and 76/80 (95%) of the errors were classified as preventable by CDS. Certain error types were more likely to be preventable by CDS than others ( P < .001). The most likely error types to be preventable by CDS were wrong medication (N = 36, 100% rated as preventable), wrong dose (N = 30, 100% rated as preventable), and documentation errors (N = 3, 100% rated as preventable). The least likely error type to be preventable by CDS was inadvertent bolus (N = 3, none rated as preventable).

CONCLUSIONS

Ninety-five percent of self-reported medication errors in the operating room were classified as preventable by CDS. Future research should include a randomized controlled trial to assess medication error rates and types with and without the use of CDS.

Prognostic clinical decision support for pneumonia in the emergency department: A randomized trial

BACKGROUND

Hospitalization rates for childhood pneumonia vary widely. Risk-based clinical decision support (CDS) interventions may reduce unwarranted variation.

METHODS

We conducted a pragmatic randomized trial in two US pediatric emergency departments (EDs) comparing electronic health record (EHR)-integrated prognostic CDS versus usual care for promoting appropriate ED disposition in children (<18 years) with pneumonia. Encounters were randomized 1:1 to usual care versus custom CDS featuring a validated pneumonia severity score predicting risk for severe in-hospital outcomes. Clinicians retained full decision-making authority. The primary outcome was inappropriate ED disposition, defined as early transition to lower- or higher-level care. Safety and implementation outcomes were also evaluated.

RESULTS

The study enrolled 536 encounters (269 usual care and 267 CDS). Baseline characteristics were similar across arms. Inappropriate disposition occurred in 3% of usual care encounters and 2% of CDS encounters (adjusted odds ratio: 0.99, 95% confidence interval: [0.32, 2.95]). Length of stay was also similar and adverse safety outcomes were uncommon in both arms. The tool's custom user interface and content were viewed as strengths by surveyed clinicians (>70% satisfied). Implementation barriers include intrinsic (e.g., reaching the right person at the right time) and extrinsic factors (i.e., global pandemic).

CONCLUSIONS

EHR-based prognostic CDS did not improve ED disposition decisions for children with pneumonia. Although the intervention's content was favorably received, low subject accrual and workflow integration problems likely limited effectiveness. Clinical Trials Registration: NCT06033079.

The role of information systems in emergency department decision-making-a literature review

OBJECTIVES

Healthcare providers employ heuristic and analytical decision-making to navigate the high-stakes environment of the emergency department (ED). Despite the increasing integration of information systems (ISs), research on their efficacy is conflicting. Drawing on related fields, we investigate how timing and mode of delivery influence IS effectiveness. Our objective is to reconcile previous contradictory findings, shedding light on optimal IS design in the ED.

MATERIALS AND METHODS

We conducted a systematic review following PRISMA across PubMed, Scopus, and Web of Science. We coded the ISs' timing as heuristic or analytical, their mode of delivery as active for automatic alerts and passive when requiring user-initiated information retrieval, and their effect on process, economic, and clinical outcomes.

RESULTS

Our analysis included 83 studies. During early heuristic decision-making, most active interventions were ineffective, while passive interventions generally improved outcomes. In the analytical phase, the effects were reversed. Passive interventions that facilitate information extraction consistently improved outcomes.

DISCUSSION

Our findings suggest that the effectiveness of active interventions negatively correlates with the amount of information received during delivery. During early heuristic decision-making, when information overload is high, physicians are unresponsive to alerts and proactively consult passive resources. In the later analytical phases, physicians show increased receptivity to alerts due to decreased diagnostic uncertainty and information quantity. Interventions that limit information lead to positive outcomes, supporting our interpretation.

CONCLUSION

We synthesize our findings into an integrated model that reveals the underlying reasons for conflicting findings from previous reviews and can guide practitioners in designing ISs in the ED.

Optimizing Antimicrobial Stewardship in the Emergency Department

Antibiotic stewardship is a core component of emergency department (ED) practice and impacts patient safety, clinical outcomes, and public health. The unique characteristics of ED practice, including crowding, time pressure, and diagnostic uncertainty, need to be considered when implementing antibiotic stewardship interventions in this setting. Rapid advances in pathogen detection and host response biomarkers promise to revolutionize the diagnosis of infectious diseases in the ED, but such tests are not yet considered standard of care. Presently, clinical decision support embedded in the electronic health record and pharmacist-led interventions are the most effective ways to improve antibiotic prescribing in the ED.

ICU Utilization After Implementation of Minor Severe Pneumonia Criteria in Real-Time Electronic Clinical Decision Support

OBJECTIVES

To determine if the implementation of automated clinical decision support (CDS) with embedded minor severe community-acquired pneumonia (sCAP) criteria was associated with improved ICU utilization among emergency department (ED) patients with pneumonia who did not require vasopressors or positive pressure ventilation at admission.

DESIGN

Planned secondary analysis of a stepped-wedge, cluster-controlled CDS implementation trial.

SETTING

Sixteen hospitals in six geographic clusters from Intermountain Health; a large, integrated, nonprofit health system in Utah and Idaho.

PATIENTS

Adults admitted to the hospital from the ED with pneumonia identified by: 1) discharge International Classification of Diseases , 10th Revision codes for pneumonia or sepsis/respiratory failure and 2) ED chest imaging consistent with pneumonia, who did not require vasopressors or positive pressure ventilation at admission.

INTERVENTIONS

After implementation, patients were exposed to automated, open-loop, comprehensive CDS that aided disposition decision (ward vs. ICU), based on objective severity scores (sCAP).

MEASUREMENTS AND MAIN RESULTS

The analysis included 2747 patients, 1814 before and 933 after implementation. The median age was 71, median Elixhauser index was 17, 48% were female, and 95% were Caucasian. A mixed-effects regression model with cluster as the random effect estimated that implementation of CDS utilizing sCAP increased 30-day ICU-free days by 1.04 days (95% CI, 0.48-1.59; p < 0.001). Among secondary outcomes, the odds of being admitted to the ward, transferring to the ICU within 72 hours, and receiving a critical therapy decreased by 57% (odds ratio [OR], 0.43; 95% CI, 0.26-0.68; p < 0.001) post-implementation; mortality within 72 hours of admission was unchanged (OR, 1.08; 95% CI, 0.56-2.01; p = 0.82) while 30-day all-cause mortality was lower post-implementation (OR, 0.71; 95% CI, 0.52-0.96; p = 0.03).

CONCLUSIONS

Implementation of electronic CDS using minor sCAP criteria to guide disposition of patients with pneumonia from the ED was associated with safe reduction in ICU utilization.

Behavioral Strategies in Diagnostic Stewardship

Diagnostic stewardship refers to the responsible and judicious use of diagnostic tests to reduce low value care and improve patient outcomes. This article provides an overview of behavioral strategies, their relevance to diagnostic stewardship and highlights behavioral determinants that drive diagnostic testing behavior, drawing on theoretic frameworks. Additionally, we provide concrete examples of evidence-based behavioral strategies for promoting appropriate diagnostic testing while acknowledging associated challenges. Finally, we highlight the significance of evaluating these strategies and provide an overview of evaluation frameworks and methods.

A data-driven framework for clinical decision support applied to pneumonia management

Despite their long history, it can still be difficult to embed clinical decision support into existing health information systems, particularly if they utilise machine learning and artificial intelligence models. Moreover, when such tools are made available to healthcare workers, it is important that the users can understand and visualise the reasons for the decision support predictions. Plausibility can be hard to achieve for complex pathways and models and perceived "black-box" functionality often leads to a lack of trust. Here, we describe and evaluate a data-driven framework which moderates some of these issues and demonstrate its applicability to the in-hospital management of community acquired pneumonia, an acute respiratory disease which is a leading cause of in-hospital mortality world-wide. We use the framework to develop and test a clinical decision support tool based on local guideline aligned management of the disease and show how it could be used to effectively prioritise patients using retrospective analysis. Furthermore, we show how this tool can be embedded into a prototype clinical system for disease management by integrating metrics and visualisations. This will assist decision makers to examine complex patient journeys, risk scores and predictions from embedded machine learning and artificial intelligence models. Our results show the potential of this approach for developing, testing and evaluating workflow based clinical decision support tools which include complex models and embedding them into clinical systems.

The impact of antimicrobial de-escalation therapy in culture-negative pneumonia: a systematic review and meta-analysis

BACKGROUND/AIMS

Antimicrobial de-escalation (ADE) remains a challenging strategy in the treatment of pneumonia. We investigated the outcomes of ADE as measured by mortality and duration of the use of antibiotics in patients with culture- negative pneumonia.

METHODS

We performed a systematic review and meta-analysis in accordance with PRISMA guidelines. The primary outcome was inpatient mortality.

RESULTS

We examined six studies comprising 11,933 subjects, of whom 1,152 received ADE. Overall, the ADE strategy was associated with a statistically lower risk of in-hospital mortality compared with non-ADE (risk ratio [RR] = 0.60, 95% confidence interval [CI] = 0.38 to 0.93). Although substantial heterogeneity was found among the included studies (I2 = 66%), a meta-regression analysis could not reveal plausible sources of heterogeneity. And ADE was associated with a shorter duration of total and initial antibiotic therapies and total length of hospital stay compared with non-ADE.

CONCLUSION

Our findings suggest that ADE seems to be significantly associated with better clinical outcomes compared with non-ADE. Caution is demanded when interpreting data of this study because of substantial between-study heterogeneity.

Validation of Childhood Pneumonia Prognostic Models for Use in Emergency Care Settings

BACKGROUND

Unwarranted variation in disposition decisions exist among children with pneumonia. We validated three prognostic models for predicting pneumonia severity among children in the emergency department (ED) and hospital.

METHODS

We performed a two-center, prospective study of children 6 months to <18 years presenting to the ED with pneumonia from January 2014 to May 2019. We evaluated three previously developed disease-specific prognostic models which use demographic, clinical, and diagnostic predictor variables, with each model estimating risk for Very Severe (mechanical ventilation or shock), Severe (ICU without very severe features), and Moderate/Mild (Hospitalization without severe features or ED discharge) pneumonia. Predictive accuracy was measured using discrimination (concordance or c-statistic) and re-calibration.

RESULTS

There were 1088 children included in one or more of the three models. Median age was 3.6 years and the majority of children were male (53.7%) and identified as non-Hispanic White (63.7%). The distribution for the ordinal severity outcome was mild or moderate (79.1%), severe (15.9%), and very severe (4.9%). The three models each demonstrated excellent discrimination (C-statistic range across models [0.786-0.803]) with no appreciable degradation in predictive accuracy from the derivation cohort.

CONCLUSIONS

All three prognostic models accurately identified risk for three clinically meaningful levels of pneumonia severity and demonstrated very good predictive performance. Physiologic variables contributed the most to model prediction. Application of these objective tools may help standardize and improve disposition and other management decisions for children with pneumonia.

Antibiotic clinical decision support for pneumonia in the ED: A randomized trial

BACKGROUND

Electronic health record-based clinical decision support (CDS) is a promising antibiotic stewardship strategy. Few studies have evaluated the effectiveness of antibiotic CDS in the pediatric emergency department (ED).

OBJECTIVE

To compare the effectiveness of antibiotic CDS vs. usual care for promoting guideline-concordant antibiotic prescribing for pneumonia in the pediatric ED.

DESIGN

Pragmatic randomized clinical trial.

SETTING AND PARTICIPANTS

Encounters for children (6 months-18 years) with pneumonia presenting to two tertiary care children s hospital EDs in the United States.

INTERVENTION

CDS or usual care was randomly assigned during 4-week periods within each site. The CDS intervention provided antibiotic recommendations tailored to each encounter and in accordance with national guidelines.

MAIN OUTCOME AND MEASURES

The primary outcome was exclusive guideline-concordant antibiotic prescribing within the first 24 h of care. Safety outcomes included time to first antibiotic order, encounter length of stay, delayed intensive care, and 3- and 7-day revisits.

RESULTS

1027 encounters were included, encompassing 478 randomized to usual care and 549 to CDS. Exclusive guideline-concordant prescribing did not differ at 24 h (CDS, 51.7% vs. usual care, 53.3%; odds ratio [OR] 0.94 [95% confidence interval [CI]: 0.73, 1.20]). In pre-specified stratified analyses, CDS was associated with guideline-concordant prescribing among encounters discharged from the ED (74.9% vs. 66.0%; OR 1.53 [95% CI: 1.01, 2.33]), but not among hospitalized encounters. Mean time to first antibiotic was shorter in the CDS group (3.0 vs 3.4 h; p = .024). There were no differences in safety outcomes.

CONCLUSIONS

Effectiveness of ED-based antibiotic CDS was greatest among those discharged from the ED. Longitudinal interventions designed to target both ED and inpatient clinicians and to address common implementation challenges may enhance the effectiveness of CDS as a stewardship tool.

Parapneumonic effusions related to Streptococcus pneumoniae: serotype and disease severity trends from 2006 to 2018 in Bristol, UK

RATIONALE

Streptococcus pneumoniae epidemiology is changing in response to vaccination and some data suggest that empyema incidence is increasing. However, differences exist between the UK and US studies. We describe trends in the clinical phenotype of adult pneumococcal pleural infection, including simple parapneumonic effusions (SPE) in the pneumococcal conjugate vaccination (PCV) era.

OBJECTIVES

To determine whether there were differences in pneumococcal disease presentation and severity associated with pleural infection.

METHODS

A retrospective cohort study, all adults ≥16 years admitted to three large UK hospitals, 2006-2018 with pneumococcal disease. 2477 invasive pneumococcal cases were identified: 459 SPE and 100 pleural infection cases. Medical records were reviewed for each clinical episode. Serotype data were obtained from the UK Health Security Agency national reference laboratory.

RESULTS

Incidence increased over time, including non-PCV-serotype disease. PCV7-serotype disease declined following paediatric PCV7 introduction, but the effect of PCV13 was less apparent as disease caused by the additional six serotypes plateaued with serotypes 1 and 3 causing such parapneumonic effusions from 2011 onwards.Patients with pleural infection had a median survival 468 days (95% CI 340 to 590) vs 286 days (95% CI 274 to 335) in those with SPE. Pleural infection associated with frank pus had lower 90-day mortality than pleural infection without pus (0% vs 29%, p<0.0001). 90-day mortality could be predicted by baseline increased RAPID (Renal, Age, Purulence, Infection source, and Dietary factors) score (HR 15.01, 95% CI 1.24 to 40.06, p=0.049).

CONCLUSIONS

Pneumococcal infection continues to cause severe disease despite the introduction of PCVs. The predominance of serotype 1 and 3 in this adult UK cohort is in keeping with previous studies in paediatric and non-UK studies. Rising non-PCV serotype disease and limited impact of PCV13 on cases caused by serotypes 1 and 3 offset the reductions in adult pneumococcal parapneumonic effusion disease burden observed following the introduction of the childhood PCV7 programme.

AV, Mahale A. Multimodal medical tensor fusion network-based DL framework for abnormality prediction from the radiology CXRs and clinical text reports

Pulmonary disease is a commonly occurring abnormality throughout this world. The pulmonary diseases include Tuberculosis, Pneumothorax, Cardiomegaly, Pulmonary atelectasis, Pneumonia, etc. A timely prognosis of pulmonary disease is essential. Increasing progress in Deep Learning (DL) techniques has significantly impacted and contributed to the medical domain, specifically in leveraging medical imaging for analysis, prognosis, and therapeutic decisions for clinicians. Many contemporary DL strategies for radiology focus on a single modality of data utilizing imaging features without considering the clinical context that provides more valuable complementary information for clinically consistent prognostic decisions. Also, the selection of the best data fusion strategy is crucial when performing Machine Learning (ML) or DL operation on multimodal heterogeneous data. We investigated multimodal medical fusion strategies leveraging DL techniques to predict pulmonary abnormality from the heterogeneous radiology Chest X-Rays (CXRs) and clinical text reports. In this research, we have proposed two effective unimodal and multimodal subnetworks to predict pulmonary abnormality from the CXR and clinical reports. We have conducted a comprehensive analysis and compared the performance of unimodal and multimodal models. The proposed models were applied to standard augmented data and the synthetic data generated to check the model's ability to predict from the new and unseen data. The proposed models were thoroughly assessed and examined against the publicly available Indiana university dataset and the data collected from the private medical hospital. The proposed multimodal models have given superior results compared to the unimodal models.

Clinical decision support improves blood culture collection before intravenous antibiotic administration in the emergency department

OBJECTIVE

Surviving Sepsis guidelines recommend blood cultures before administration of intravenous (IV) antibiotics for patients with sepsis or moderate to high risk of bacteremia. Clinical decision support (CDS) that reminds emergency department (ED) providers to obtain blood cultures when ordering IV antibiotics may lead to improvements in this process measure.

METHODS

This was a multicenter causal impact analysis comparing timely blood culture collections prior to IV antibiotics for adult ED patients 1 year before and after a CDS intervention implementation in the electronic health record. A Bayesian structured time-series model compared daily timely blood cultures collected compared to a forecasted synthetic control. Mixed effects models evaluated the impact of the intervention controlling for confounders.

RESULTS

The analysis included 54 538 patients over 2 years. In the baseline phase, 46.1% had blood cultures prior to IV antibiotics, compared to 58.8% after the intervention. Causal impact analysis determined an absolute increase of 13.1% (95% CI 10.4-15.7%) of timely blood culture collections overall, although the difference in patients with a sepsis diagnosis or who met CDC Adult Sepsis Event criteria was not significant, absolute difference 8.0% (95% CI -0.2 to 15.8). Blood culture positivity increased in the intervention phase, and contamination rates were similar in both study phases.

DISCUSSION

CDS improved blood culture collection before IV antibiotics in the ED, without increasing overutilization.

CONCLUSION

A simple CDS alert increased timely blood culture collections in ED patients for whom concern for infection was high enough to warrant IV antibiotics.

A Pragmatic, Stepped-Wedge, Cluster-controlled Clinical Trial of Real-Time Pneumonia Clinical Decision Support

Rationale: Care of emergency department (ED) patients with pneumonia can be challenging. Clinical decision support may decrease unnecessary variation and improve care. Objectives: To report patient outcomes and processes of care after deployment of electronic pneumonia clinical decision support (ePNa): a comprehensive, open loop, real-time clinical decision support embedded within the electronic health record. Methods: We conducted a pragmatic, stepped-wedge, cluster-controlled trial with deployment at 2-month intervals in 16 community hospitals. ePNa extracts real-time and historical data to guide diagnosis, risk stratification, microbiological studies, site of care, and antibiotic therapy. We included all adult ED patients with pneumonia over the course of 3 years identified by International Classification of Diseases, 10th Revision discharge coding confirmed by chest imaging. Measurements and Main Results: The median age of the 6,848 patients was 67 years (interquartile range, 50-79), and 48% were female; 64.8% were hospital admitted. Unadjusted mortality was 8.6% before and 4.8% after deployment. A mixed effects logistic regression model adjusting for severity of illness with hospital cluster as the random effect showed an adjusted odds ratio of 0.62 (0.49-0.79; P < 0.001) for 30-day all-cause mortality after deployment. Lower mortality was consistent across hospital clusters. ePNa-concordant antibiotic prescribing increased from 83.5% to 90.2% (P < 0.001). The mean time from ED admission to first antibiotic was 159.4 (156.9-161.9) minutes at baseline and 150.9 (144.1-157.8) minutes after deployment (P < 0.001). Outpatient disposition from the ED increased from 29.2% to 46.9%, whereas 7-day secondary hospital admission was unchanged (5.2% vs. 6.1%). ePNa was used by ED clinicians in 67% of eligible patients. Conclusions: ePNa deployment was associated with improved processes of care and lower mortality. Clinical trial registered with www.clinicaltrials.gov (NCT03358342).

CheXED: Comparison of a Deep Learning Model to a Clinical Decision Support System for Pneumonia in the Emergency Department

PURPOSE

Patients with pneumonia often present to the emergency department (ED) and require prompt diagnosis and treatment. Clinical decision support systems for the diagnosis and management of pneumonia are commonly utilized in EDs to improve patient care. The purpose of this study is to investigate whether a deep learning model for detecting radiographic pneumonia and pleural effusions can improve functionality of a clinical decision support system (CDSS) for pneumonia management (ePNa) operating in 20 EDs.

MATERIALS AND METHODS

In this retrospective cohort study, a dataset of 7434 prior chest radiographic studies from 6551 ED patients was used to develop and validate a deep learning model to identify radiographic pneumonia, pleural effusions, and evidence of multilobar pneumonia. Model performance was evaluated against 3 radiologists' adjudicated interpretation and compared with performance of the natural language processing of radiology reports used by ePNa.

RESULTS

The deep learning model achieved an area under the receiver operating characteristic curve of 0.833 (95% confidence interval [CI]: 0.795, 0.868) for detecting radiographic pneumonia, 0.939 (95% CI: 0.911, 0.962) for detecting pleural effusions and 0.847 (95% CI: 0.800, 0.890) for identifying multilobar pneumonia. On all 3 tasks, the model achieved higher agreement with the adjudicated radiologist interpretation compared with ePNa.

CONCLUSIONS

A deep learning model demonstrated higher agreement with radiologists than the ePNa CDSS in detecting radiographic pneumonia and related findings. Incorporating deep learning models into pneumonia CDSS could enhance diagnostic performance and improve pneumonia management.

Association Between Procalcitonin and Antibiotics in Children With Community-Acquired Pneumonia

OBJECTIVE

To determine whether empirical antibiotic initiation and selection for children with pneumonia was associated with procalcitonin (PCT) levels when results were blinded to clinicians.

METHODS

We enrolled children <18 years with radiographically confirmed pneumonia at 2 children's hospitals from 2014 to 2019. Blood for PCT was collected at enrollment (blinded to clinicians). We modeled associations between PCT and (1) antibiotic initiation and (2) antibiotic selection (narrow versus broad-spectrum) using multivariable logistic regression models. To quantify potential stewardship opportunities, we calculated proportions of noncritically ill children receiving antibiotics who also had a low likelihood of bacterial etiology (PCT <0.25 ng/mL) and those receiving broad-spectrum therapy, regardless of PCT level.

RESULTS

We enrolled 488 children (median PCT, 0.37 ng/mL; interquartile range [IQR], 0.11-2.38); 85 (17%) received no antibiotics (median PCT, 0.32; IQR, 0.09-1.33). Among the 403 children receiving antibiotics, 95 (24%) received narrow-spectrum therapy (median PCT, 0.24; IQR, 0.08-2.52) and 308 (76%) received broad-spectrum (median PCT, 0.46; IQR, 0.12-2.83). In adjusted analyses, PCT values were not associated with antibiotic initiation (odds ratio [OR], 1.02, 95% confidence interval [CI], 0.97%-1.06%) or empirical antibiotic selection (OR 1.07; 95% CI, 0.97%-1.17%). Of those with noncritical illness, 246 (69%) were identified as potential targets for antibiotic stewardship interventions.

CONCLUSION

Neither antibiotic initiation nor empirical antibiotic selection were associated with PCT values. Whereas other factors may inform antibiotic treatment decisions, the observed discordance between objective likelihood of bacterial etiology and antibiotic use suggests important opportunities for stewardship.

Prevalence, Characteristics, and Outcomes of Emergency Department Discharge Among Patients With Sepsis

IMPORTANCE

Sepsis guidelines and research have focused on patients with sepsis who are admitted to the hospital, but the scope and implications of sepsis that is managed in an outpatient setting are largely unknown.

OBJECTIVE

To identify the prevalence, risk factors, practice variation, and outcomes for discharge to outpatient management of sepsis among patients presenting to the emergency department (ED).

DESIGN, SETTING, AND PARTICIPANTS

This cohort study was conducted at the EDs of 4 Utah hospitals, and data extraction and analysis were performed from 2017 to 2021. Participants were adult ED patients who presented to a participating ED from July 1, 2013, to December 31, 2016, and met sepsis criteria before departing the ED alive and not receiving hospice care.

EXPOSURES

Patient demographic and clinical characteristics, health system parameters, and ED attending physician.

MAIN OUTCOMES AND MEASURES

Information on ED disposition was obtained from electronic medical records, and 30-day mortality data were acquired from Utah state death records and the US Social Security Death Index. Factors associated with ED discharge rather than hospital admission were identified using penalized logistic regression. Variation in ED discharge rates between physicians was estimated after adjustment for potential confounders using generalized linear mixed models. Inverse probability of treatment weighting was used in the primary analysis to assess the noninferiority of outpatient management for 30-day mortality (noninferiority margin of 1.5%) while adjusting for multiple potential confounders.

RESULTS

Among 12 333 ED patients with sepsis (median [IQR] age, 62 [47-76] years; 7017 women [56.9%]) who were analyzed in the study, 1985 (16.1%) were discharged from the ED. After penalized regression, factors associated with ED discharge included age (adjusted odds ratio [aOR], 0.90 per 10-y increase; 95% CI, 0.87-0.93), arrival to ED by ambulance (aOR, 0.61; 95% CI, 0.52-0.71), organ failure severity (aOR, 0.58 per 1-point increase in the Sequential Organ Failure Assessment score; 95% CI, 0.54-0.60), and urinary tract (aOR, 4.56 [95% CI, 3.91-5.31] vs pneumonia), intra-abdominal (aOR, 0.51 [95% CI, 0.39-0.65] vs pneumonia), skin (aOR, 1.40 [95% CI, 1.14-1.72] vs pneumonia) or other source of infection (aOR, 1.67 [95% CI, 1.40-1.97] vs pneumonia). Among 89 ED attending physicians, adjusted ED discharge probability varied significantly (likelihood ratio test, P < .001), ranging from 8% to 40% for an average patient. The unadjusted 30-day mortality was lower in discharged patients than admitted patients (0.9% vs 8.3%; P < .001), and their adjusted 30-day mortality was noninferior (propensity-adjusted odds ratio, 0.21 [95% CI, 0.09-0.48]; adjusted risk difference, 5.8% [95% CI, 5.1%-6.5%]; P < .001). Alternative confounder adjustment strategies yielded odds ratios that ranged from 0.21 to 0.42.

CONCLUSIONS AND RELEVANCE

In this cohort study, discharge to outpatient treatment of patients who met sepsis criteria in the ED was more common than previously recognized and varied substantially between ED physicians, but it was not associated with higher mortality compared with hospital admission. Systematic, evidence-based strategies to optimize the triage of ED patients with sepsis are needed.

Effect of COVID 19 pneumonia on hyperglycemia: Is it different from non COVID pneumonia?

BACKGROUND AND AIMS

Glycemic control in critical illness has been linked to outcomes. We sought to investigate if COVID pneumonia was causing disrupted glycemic control compared to historically similar diseases.

METHODS

At Intermountain Healthcare, a 23-hospital healthcare system in the intermountain west, we performed a multicenter, retrospective cohort observational study. We compared 13,268 hospitalized patients with COVID pneumonia to 6673 patients with non -COVID-pneumonia.

RESULTS

Patients with COVID-19 were younger had fewer comorbidities, had lower mortality and greater length of hospital stay. Our regression models demonstrated that daily insulin dose, indexed for weight, was associated with COVID-19, age, diabetic status, HgbA1c, admission SOFA, ICU length of stay and receipt of corticosteroids. There was significant interaction between a diagnosis of diabetes and having COVID-19. Time in range for our IV insulin protocol was not correlated with having COVID after adjustment. It was correlated with ICU length of stay, diabetic control (HgbA1C) and prior history of diabetes. Among patients with subcutaneous (SQ) insulin only percent of glucose checks in range was correlated with diabetic status, having Covid-19, HgbA1c, total steroids given and Elixhauser comorbidity score even when controlled for other factors.

CONCLUSIONS

Hospitalized patients with COVID-19 pneumonia who receive insulin for glycemic control require both more SQ and IV insulin than the non-COVID-19 pneumonia counterparts. Patients with COVID-19 who received SQ insulin only had a lower percent of glucose checks in range.

Deploying an Electronic Clinical Decision Support Tool for Diagnosis and Treatment of Pneumonia Into Rural and Critical Access Hospitals: Utilization, Effect on Processes of Care, and Clinician Satisfaction

PURPOSE

Electronic clinical decision support (CDS) for treatment of community-acquired pneumonia (ePNa) is associated with improved guideline adherence and decreased mortality. How rural providers respond to CDS developed for urban hospitals could shed light on extending CDS to resource-limited settings.

METHODS

ePNa was deployed into 10 rural and critical access hospital emergency departments (EDs) in Utah and Idaho in 2018. We reviewed pneumonia cases identified through ICD-10 codes after local deployment to measure ePNa utilization and guideline adherence. ED providers were surveyed to assess quantitative and qualitative aspects of satisfaction.

FINDINGS

ePNa was used in 109/301 patients with pneumonia (36%, range 0%-67% across hospitals) and was associated with appropriate antibiotic selection (93% vs 65%, P < .001). Fifty percent of survey recipients responded, 87% were physicians, 87% were men, and the median ED experience was 10 years. Mean satisfaction with ePNa was 3.3 (range 1.7-4.8) on a 5-point Likert scale. Providers with a favorable opinion of ePNa were more likely to favor implementation of additional CDS (P = .005). Satisfaction was not associated with provider type, age, years of experience or experience with ePNa. Ninety percent of respondents provided qualitative feedback. The most common theme in high and low utilization hospitals was concern about usability. Compared to high utilization hospitals, low utilization hospitals more frequently identified concerns about adaptation for local needs.

CONCLUSIONS

ePNa deployment to rural and critical access EDs was moderately successful and associated with improved antibiotic use. Concerns about usability and adapting ePNa for local use predominated the qualitative feedback.

Understanding the Host in the Management of Pneumonia. An Official American Thoracic Society Workshop Report

Pneumonia causes a significant burden of disease worldwide. Although all populations are at risk of pneumonia, those at extremes of age and those with immunosuppressive disorders, underlying respiratory disease, and critical illness are particularly vulnerable. Although clinical practice guidelines addressing the management and treatment of pneumonia exist, few of the supporting studies focus on the crucial contributions of the host in pneumonia pathogenesis and recovery. Such essential considerations include the host risk factors that lead to susceptibility to lung infections; biomarkers reflecting the host response and the means to pursue host-directed pneumonia therapy; systemic effects of pneumonia on the host; and long-term health outcomes after pneumonia. To address these gaps, the Pneumonia Working Group of the Assembly on Pulmonary Infection and Tuberculosis led a workshop held at the American Thoracic Society meeting in May 2018 with overarching objectives to foster attention, stimulate research, and promote funding for short-term and long-term investigations into the host contributions to pneumonia. The workshop involved participants from various disciplines with expertise in lung infection, pneumonia, sepsis, immunocompromised patients, translational biology, data science, genomics, systems biology, and clinical trials. This workshop report summarizes the presentations and discussions and important recommendations for future clinical pneumonia studies. These recommendations include establishing consensus disease and outcome definitions, improved phenotyping, development of clinical study networks, standardized data and biospecimen collection and protocols, and development of innovative trial designs.

Community-acquired pneumonia

Community-acquired pneumonia is not usually considered a high-priority problem by the public, although it is responsible for substantial mortality, with a third of patients dying within 1 year after being discharged from hospital for pneumoniae. Although up to 18% of patients with community-acquired pneumonia who were hospitalised (admitted to hospital and treated there) have at least one risk factor for immunosuppression worldwide, strong evidence on community-acquired pneumonia management in this population is scarce. Several features of clinical management for community-acquired pneumonia should be addressed to reduce mortality, morbidity, and complications related to community-acquired pneumonia in patients who are immunocompetent and patients who are immunocompromised. These features include rapid diagnosis, microbiological investigation, prevention and management of complications (eg, respiratory failure, sepsis, and multiorgan failure), empirical antibiotic therapy in accordance with patient's risk factors and local microbiological epidemiology, individualised antibiotic therapy according to microbiological data, appropriate outcomes for therapeutic switch from parenteral to oral antibiotics, discharge planning, and long-term follow-up. This Seminar offers an updated view on community-acquired pneumonia in adults, with suggestions for clinical and translational research.

Appropriateness of antibiotic selection for pneumonia in the emergency department: pre- and post-order set changes

OBJECTIVES

Emergency department (ED) providers face pressure to meet sepsis mandates such as prompt administration of antibiotic therapy, which can lead to the overuse of broad-spectrum antibiotics. In recent years, there has also been a push to adhere to institutional antibiotic stewardship goals including decreasing inappropriate antibiotic therapy and limiting duration of therapy. Previous literature has demonstrated that the incorporation of clinical decision support (CDS) tools in electronic medical records can aid in guiding appropriate antibiotic prescribing. Therefore, the objective of this study was to determine whether the implementation of a CDS tool could improve antibiotic selection for pneumonia management in the ED.

METHODS

This was a retrospective single-centre observational study conducted in patients that presented to the ED with pneumonia. In November 2018, a CDS tool was incorporated into the ED sepsis order set to guide practitioners in selecting appropriate antibiotics for pneumonia. Antibiotic prescribing patterns were assessed pre-CDS (January-February 2018) and post-CDS (January-February 2019) implementation. Patients were included if they were 18 years of age or older, had an ED visit with ICD10 code reflective of pneumonia, and had at least one antibiotic ordered from the ED sepsis order set. The primary endpoint was the percentage of patients prescribed appropriate antibiotic therapy for pneumonia based on patient risk factors pre- and post-CDS implementation.

KEY FINDINGS

There were 161 patients in the pre-CDS group and 119 patients in the post-CDS group included in the study. There was a significant improvement in the selection of appropriate antibiotics in the post-CDS group (31.9% versus 65.3%, P < 0.0001) with no significant differences in duration of antibiotics, intubation rates, vasopressor initiation, length of stay, mortality or 30-day readmission.

CONCLUSION

The implementation of a CDS tool for empiric management of pneumonia in the ED significantly improved the selection of appropriate antibiotics.

Can evidence-based decision support tools transform antibiotic management? A systematic review and meta-analyses

OBJECTIVES

To assess the effectiveness of clinical decision support systems (CDSSs) at reducing unnecessary and suboptimal antibiotic prescribing within different healthcare settings.

METHODS

A systematic review of published studies was undertaken with seven databases from database inception to November 2018. A protocol was developed using the PRISMA-P checklist and study selection criteria were determined prior to performing the search. Critical appraisal of studies was undertaken using relevant tools. Meta-analyses were performed using a random-effects model to determine whether CDSS use affected optimal antibiotic management.

RESULTS

Fifty-seven studies were identified that reported on CDSS effectiveness. Most were non-randomized studies with low methodological quality. However, randomized controlled trials of moderate methodological quality were available and assessed separately. The meta-analyses indicated that appropriate antibiotic therapy was twice as likely to occur following the implementation of CDSSs (OR 2.28, 95% CI 1.82-2.86, k = 20). The use of CDSSs was also associated with a relative decrease (18%) in mortality (OR 0.82, 95% CI 0.73-0.91, k = 18). CDSS implementation also decreased the overall volume of antibiotic use, length of hospital stay, duration and cost of therapy. The magnitude of the effect did vary by study design, but the direction of the effect was consistent in favouring CDSSs.

CONCLUSIONS

Decision support tools can be effective to improve antibiotic prescribing, although there is limited evidence available on use in primary care. Our findings suggest that a focus on system requirements and implementation processes would improve CDSS uptake and provide more definitive benefits for antibiotic stewardship.

A Roadmap for Automating Lineage Tracing to Aid Automatically Explaining Machine Learning Predictions for Clinical Decision Support

Using machine learning predictive models for clinical decision support has great potential in improving patient outcomes and reducing health care costs. However, most machine learning models are black boxes that do not explain their predictions, thereby forming a barrier to clinical adoption. To overcome this barrier, an automated method was recently developed to provide rule-style explanations of any machine learning model’s predictions on tabular data and to suggest customized interventions. Each explanation delineates the association between a feature value pattern and an outcome value. Although the association and intervention information is useful, the user of the automated explaining function often requires more detailed information to better understand the patient’s situation and to aid in decision making. More specifically, consider a feature value in the explanation that is computed by an aggregation function on the raw data, such as the number of emergency department visits related to asthma that the patient had in the prior 12 months. The user often wants to rapidly drill through to see certain parts of the related raw data that produce the feature value. This task is frequently difficult and time-consuming because the few pieces of related raw data are submerged by many pieces of raw data of the patient that are unrelated to the feature value. To address this issue, this paper outlines an automated lineage tracing approach, which adds automated drill-through capability to the automated explaining function, and provides a roadmap for future research.

A Randomized Controlled Trial of an Electronic Clinical Decision Support Tool for Inpatient Antimicrobial Stewardship

BACKGROUND

The weighted incidence syndromic combination antibiogram (WISCA) is an antimicrobial stewardship tool that utilizes electronic medical record data to provide real-time clinical decision support regarding empiric antibiotic prescription in the hospital setting. The aim of this study was to determine the impact of WISCA utilization for empiric antibiotic prescription on hospital length of stay (LOS).

METHODS

We performed a crossover randomized controlled trial of the WISCA tool at 4 hospitals. Study participants included adult inpatients receiving empiric antibiotics for urinary tract infection (UTI), abdominal-biliary infection (ABI), pneumonia, or nonpurulent cellulitis. Antimicrobial stewardship (ASP) physicians utilized WISCA and clinical guidelines to provide empiric antibiotic recommendations. The primary outcome was LOS. Secondary outcomes included 30-day mortality, 30-day readmission, Clostridioides difficile infection, acquisition of multidrug-resistant gram-negative organism (MDRO), and antibiotics costs.

RESULTS

In total, 6849 participants enrolled in the study. There were no overall differences in outcomes among the intervention versus control groups. Participants with cellulitis in the intervention group had significantly shorter mean LOS compared to participants with cellulitis in the control group (coefficient estimate = 0.53 [-0.97, -0.09], P = .0186). For patients with community acquired pneumonia (CAP), the intervention group had significantly lower odds of 30-day mortality compared to the control group (adjusted odds ratio [aOR] .58, 95% confidence interval [CI], .396, .854, P = .02).

CONCLUSIONS

Use of WISCA was not associated with improved outcomes for UTI and ABI. Guidelines-based interventions were associated with decreased LOS for cellulitis and decreased mortality for CAP.

Antibiotic Stewardship in the Intensive Care Unit. An Official American Thoracic Society Workshop Report in Collaboration with the AACN, CHEST, CDC, and SCCM

Intensive care units (ICUs) are an appropriate focus of antibiotic stewardship program efforts because a large proportion of any hospital’s use of parenteral antibiotics, especially broad-spectrum, occurs in the ICU. Given the importance of antibiotic stewardship for critically ill patients and the importance of critical care practitioners as the front line for antibiotic stewardship, a workshop was convened to specifically address barriers to antibiotic stewardship in the ICU and discuss tactics to overcome these. The working definition of antibiotic stewardship is “the right drug at the right time and the right dose for the right bug for the right duration.” A major emphasis was that antibiotic stewardship should be a core competency of critical care clinicians. Fear of pathogens that are not covered by empirical antibiotics is a major driver of excessively broad-spectrum therapy in critically ill patients. Better diagnostics and outcome data can address this fear and expand efforts to narrow or shorten therapy. Greater awareness of the substantial adverse effects of antibiotics should be emphasized and is an important counterargument to broad-spectrum therapy in individual low-risk patients. Optimal antibiotic stewardship should not focus solely on reducing antibiotic use or ensuring compliance with guidelines. Instead, it should enhance care both for individual patients (by improving and individualizing their choice of antibiotic) and for the ICU population as a whole. Opportunities for antibiotic stewardship in common ICU infections, including community- and hospital-acquired pneumonia and sepsis, are discussed. Intensivists can partner with antibiotic stewardship programs to address barriers and improve patient care.

Clinical Policy: Critical Issues in the Management of Adult Patients Presenting to the Emergency Department With Community-Acquired Pneumonia

This clinical policy from the American College of Emergency Physicians is a revision of the 2009 "Clinical Policy: Critical Issues in the Management of Adult Patients Presenting to the Emergency Department With Community-Acquired Pneumonia." A writing subcommittee conducted a systematic review of the literature to derive evidence-based recommendations to answer the following clinical questions: (1) In the adult emergency department patient diagnosed with community-acquired pneumonia, what clinical decision aids can inform the determination of patient disposition? (2) In the adult emergency department patient with community-acquired pneumonia, what biomarkers can be used to direct initial antimicrobial therapy? (3) In the adult emergency department patient diagnosed with community-acquired pneumonia, does a single dose of parenteral antibiotics in the emergency department followed by oral treatment versus oral treatment alone improve outcomes? Evidence was graded and recommendations were made based on the strength of the available data.

Can emergency department provider notes help to achieve more dynamic clinical decision support?

OBJECTIVE

Assess whether clinical data were present in emergency department (ED) provider notes at time of order entry for cervical spine (c-spine) imaging that could be used to augment or pre-populate clinical decision support (CDS) attributes.

METHODS

This Institutional Review Board-approved retrospective study, performed in a quaternary hospital, included all encounters for adult ED patients seen April 1, 2013-September 30, 2014 for a chief complaint of trauma who received c-spine computed tomography (CT) or x-ray. We assessed proportion of ED encounters with at least 1 c-spine-specific CDS rule attribute in clinical notes available at the time of imaging order and agreement between attributes in clinical notes and data entered into CDS.

RESULTS

A portion of the clinical note was submitted before imaging order in 42% (184/438) of encounters reviewed; 59.2% (109/184) of encounters with note portions submitted before imaging order had at least 1 positive CDS attribute identified supporting imaging study appropriateness; 34.8% (64/184) identified exclusion criteria where CDS appropriateness recommendations would not be applicable. 65.8% (121/184) of encounters had either a positive CDS attribute or an exclusion criterion. Concordance of c-spine CDS attributes when present in both notes and CDS was 68.4% (κ = 0.35 95% CI: 0.15-0.56; McNemar P = 0.23).

CONCLUSIONS

Clinical notes are an underutilized source of clinical attributes needed for CDS, available in a substantial percentage of encounters at the time of imaging order. Automated pre-population of imaging order requisitions with relevant clinical information extracted from electronic health record provider notes may: (1) improve ordering efficiency by reducing redundant data entry, (2) help improve clinical relevance of CDS alerts, and (3) potentially reduce provider burnout from extraneous alerts.

Advances in community-acquired pneumonia

Community-acquired pneumonia is one of the commonest and deadliest of the infectious diseases, yet our understanding of it remains relatively poor. The recently published American Thoracic Society and Infectious Diseases Society of America Community-acquired pneumonia guidelines acknowledged that most of what we accept as standard of care is supported only by low quality evidence, highlighting persistent uncertainty and deficiencies in our knowledge. However, progress in diagnostics, translational research, and epidemiology has changed our concept of pneumonia, contributing to a gradual improvement in prevention, diagnosis, treatment, and outcomes for our patients. The emergence of considerable evidence about adverse long-term health outcomes in pneumonia survivors has also challenged our concept of pneumonia as an acute disease and what treatment end points are important. This review focuses on advances in the research and care of community-acquired pneumonia in the past two decades. We summarize the evidence around our understanding of pathogenesis and diagnosis, discuss key contentious management issues including the role of procalcitonin and the use or non-use of corticosteroids, and explore the relationships between pneumonia and long-term outcomes including cardiovascular and cognitive health.

Fusion of medical imaging and electronic health records using deep learning: a systematic review and implementation guidelines

Advancements in deep learning techniques carry the potential to make significant contributions to healthcare, particularly in fields that utilize medical imaging for diagnosis, prognosis, and treatment decisions. The current state-of-the-art deep learning models for radiology applications consider only pixel-value information without data informing clinical context. Yet in practice, pertinent and accurate non-imaging data based on the clinical history and laboratory data enable physicians to interpret imaging findings in the appropriate clinical context, leading to a higher diagnostic accuracy, informative clinical decision making, and improved patient outcomes. To achieve a similar goal using deep learning, medical imaging pixel-based models must also achieve the capability to process contextual data from electronic health records (EHR) in addition to pixel data. In this paper, we describe different data fusion techniques that can be applied to combine medical imaging with EHR, and systematically review medical data fusion literature published between 2012 and 2020. We conducted a systematic search on PubMed and Scopus for original research articles leveraging deep learning for fusion of multimodality data. In total, we screened 985 studies and extracted data from 17 papers. By means of this systematic review, we present current knowledge, summarize important results and provide implementation guidelines to serve as a reference for researchers interested in the application of multimodal fusion in medical imaging.

Computerised clinical decision support systems and absolute improvements in care: meta-analysis of controlled clinical trials

OBJECTIVE

To report the improvements achieved with clinical decision support systems and examine the heterogeneity from pooling effects across diverse clinical settings and intervention targets.

DESIGN

Systematic review and meta-analysis.

DATA SOURCES

Medline up to August 2019.

ELIGIBILITY CRITERIA FOR SELECTING STUDIES AND METHODS

Randomised or quasi-randomised controlled trials reporting absolute improvements in the percentage of patients receiving care recommended by clinical decision support systems. Multilevel meta-analysis accounted for within study clustering. Meta-regression was used to assess the degree to which the features of clinical decision support systems and study characteristics reduced heterogeneity in effect sizes. Where reported, clinical endpoints were also captured.

RESULTS

In 108 studies (94 randomised, 14 quasi-randomised), reporting 122 trials that provided analysable data from 1 203 053 patients and 10 790 providers, clinical decision support systems increased the proportion of patients receiving desired care by 5.8% (95% confidence interval 4.0% to 7.6%). This pooled effect exhibited substantial heterogeneity (I2=76%), with the top quartile of reported improvements ranging from 10% to 62%. In 30 trials reporting clinical endpoints, clinical decision support systems increased the proportion of patients achieving guideline based targets (eg, blood pressure or lipid control) by a median of 0.3% (interquartile range -0.7% to 1.9%). Two study characteristics (low baseline adherence and paediatric settings) were associated with significantly larger effects. Inclusion of these covariates in the multivariable meta-regression, however, did not reduce heterogeneity.

CONCLUSIONS

Most interventions with clinical decision support systems appear to achieve small to moderate improvements in targeted processes of care, a finding confirmed by the small changes in clinical endpoints found in studies that reported them. A minority of studies achieved substantial increases in the delivery of recommended care, but predictors of these more meaningful improvements remain undefined.

Performance of pneumococcal urinary antigen test in patients with community-onset pneumonia: a propensity score-matching study

BACKGROUND/AIMS

Although pneumococcal urinary antigen tests (PUATs) have universally been used for the diagnosis of pneumococcal pneumonia, data on the efficacy of these exams are limited. The objective of our study was to investigate the clinical impact of the PUAT in patients with community-onset pneumonia (CO-pneumonia).

METHODS

We conducted a retrospective cohort study of patients diagnosed with CO-pneumonia. Patients were classified according to their PUAT results and were matched using the propensity score-matching method. The primary outcome was 30-day mortality.

RESULTS

A total of 1,257 patients were identified and 163 (13.0%) demonstrated positive PUAT results. The sensitivity and specificity values of PUAT for overall pneumococcal pneumonia were 56.5% and 91.4%, respectively. In the full cohort, there were no significant differences in 30-day mortality between the two groups (6.1% in the positive PUAT group vs. 8.2% in the negative PUAT group, p = 0.357). However, in the propensity-matched cohort, the 30-day mortality rates were lower in the positive PUAT group (5.6% vs. 17.4%, p = 0.001). With respect to secondary outcomes, the proportion of patients with potentially drug-resistant pathogens, changes in antibiotics, and failure rates of initial antibiotic therapy were significantly lower in the positive PUAT group than in the negative PUAT group of the propensity-matched cohort.

CONCLUSION

We found that the sensitivity of the index test was low and specificity was high in this clinical setting. And our findings suggest that positive PUAT results may be associated with favorable clinical outcomes in patients with CO-pneumonia.

Antibiotic Prescribing Variability in a Large Urgent Care Network: A New Target for Outpatient Stewardship

Improving antibiotic prescribing in outpatient settings is a public health priority. In the United States, urgent care (UC) encounters are increasing and have high rates of inappropriate antibiotic prescribing. Our objective was to characterize antibiotic prescribing practices during UC encounters, with a focus on respiratory tract conditions. This was a retrospective cohort study of UC encounters in the Intermountain Healthcare network. Among 1.16 million UC encounters, antibiotics were prescribed during 34% of UC encounters and respiratory conditions accounted for 61% of all antibiotics prescribed. Of respiratory encounters, 50% resulted in antibiotic prescriptions, yet the variability at the level of the provider ranged from 3% to 94%. Similar variability between providers was observed for respiratory conditions where antibiotics were not indicated and in first-line antibiotic selection for sinusitis, otitis media, and pharyngitis. These findings support the importance of developing antibiotic stewardship interventions specifically targeting UC settings.

Application of human factors to improve usability of clinical decision support for diagnostic decision-making: a scenario-based simulation study

OBJECTIVE

In this study, we used human factors (HF) methods and principles to design a clinical decision support (CDS) that provides cognitive support to the pulmonary embolism (PE) diagnostic decision-making process in the emergency department. We hypothesised that the application of HF methods and principles will produce a more usable CDS that improves PE diagnostic decision-making, in particular decision about appropriate clinical pathway.

MATERIALS AND METHODS

We conducted a scenario-based simulation study to compare a HF-based CDS (the so-called CDS for PE diagnosis (PE-Dx CDS)) with a web-based CDS (MDCalc); 32 emergency physicians performed various tasks using both CDS. PE-Dx integrated HF design principles such as automating information acquisition and analysis, and minimising workload. We assessed all three dimensions of usability using both objective and subjective measures: effectiveness (eg, appropriate decision regarding the PE diagnostic pathway), efficiency (eg, time spent, perceived workload) and satisfaction (perceived usability of CDS).

RESULTS

Emergency physicians made more appropriate diagnostic decisions (94% with PE-Dx; 84% with web-based CDS; p<0.01) and performed experimental tasks faster with the PE-Dx CDS (on average 96 s per scenario with PE-Dx; 117 s with web-based CDS; p<0.001). They also reported lower workload (p<0.001) and higher satisfaction (p<0.001) with PE-Dx.

CONCLUSIONS

This simulation study shows that HF methods and principles can improve usability of CDS and diagnostic decision-making. Aspects of the HF-based CDS that provided cognitive support to emergency physicians and improved diagnostic performance included automation of information acquisition (eg, auto-populating risk scoring algorithms), minimisation of workload and support of decision selection (eg, recommending a clinical pathway). These HF design principles can be applied to the design of other CDS technologies to improve diagnostic safety.

Diagnosis and Treatment of Adults with Community-acquired Pneumonia. An Official Clinical Practice Guideline of the American Thoracic Society and Infectious Diseases Society of America

Background: This document provides evidence-based clinical practice guidelines on the management of adult patients with community-acquired pneumonia.

Methods: A multidisciplinary panel conducted pragmatic systematic reviews of the relevant research and applied Grading of Recommendations, Assessment, Development, and Evaluation methodology for clinical recommendations.

Results: The panel addressed 16 specific areas for recommendations spanning questions of diagnostic testing, determination of site of care, selection of initial empiric antibiotic therapy, and subsequent management decisions. Although some recommendations remain unchanged from the 2007 guideline, the availability of results from new therapeutic trials and epidemiological investigations led to revised recommendations for empiric treatment strategies and additional management decisions.

Conclusions: The panel formulated and provided the rationale for recommendations on selected diagnostic and treatment strategies for adult patients with community-acquired pneumonia.

Effect of Electronic Prescribing Strategies on Medication Error and Harm in Hospital: a Systematic Review and Meta-analysis

BACKGROUND

Computerized physician order entry and clinical decision support systems are electronic prescribing strategies that are increasingly used to improve patient safety. Previous reviews show limited effect on patient outcomes. Our objective was to assess the impact of electronic prescribing strategies on medication errors and patient harm in hospitalized patients.

METHODS

MEDLINE, EMBASE, CENTRAL, and CINAHL were searched from January 2007 to January 2018. We included prospective studies that compared hospital-based electronic prescribing strategies with control, and reported on medication error or patient harm. Data were abstracted by two reviewers and pooled using random effects model. Study quality was assessed using the Effective Practice and Organisation of Care and evidence quality was assessed using Grading of Recommendations Assessment, Development, and Evaluation.

RESULTS

Thirty-eight studies were included; comprised of 11 randomized control trials and 27 non-randomized interventional studies. Electronic prescribing strategies reduced medication errors (RR 0.24 (95% CI 0.13, 0.46), I² 98%, n = 11) and dosing errors (RR 0.17 (95% CI 0.08, 0.38), I² 96%, n = 9), with both risk ratios significantly affected by advancing year of publication. There was a significant effect of electronic prescribing strategies on adverse drug events (ADEs) (RR 0.52 (95% CI 0.40, 0.68), I² 0%, n = 2), but not on preventable ADEs (RR 0.55 (95% CI 0.30, 1.01), I² 78%, n = 3), hypoglycemia (RR 1.03 (95% CI 0.62-1.70), I² 28%, n = 7), length of stay (MD - 0.18 (95% - 1.42, 1.05), I² 94%, n = 7), or mortality (RR 0.97 (95% CI 0.79, 1.19), I² 74%, n = 9). The quality of evidence was rated very low.

DISCUSSION

Electronic prescribing strategies decrease medication errors and adverse drug events, but had no effect on other patient outcomes. Conservative interpretations of these findings are supported by significant heterogeneity and the preponderance of low-quality studies.

Provincial dissemination of HEARTSMAP, an emergency department psychosocial assessment and disposition decision tool for children and youth

Background

This article describes the provincial dissemination of HEARTSMAP, an evidence-based emergency department (ED) psychosocial assessment and disposition decision tool for clinician use with children and youth.

Methods

HEARTSMAP was disseminated in partnership with local, child and youth mental health teams, as part of a quality improvement initiative implemented in British Columbia EDs. The target audience of education sessions were clinicians working in ED settings responsible for paediatric psychosocial assessments. We used the RE-AIM framework to evaluate the reach, effectiveness, adoption, implementation, and maintenance of HEARTSMAP dissemination, analyzing data from session evaluation forms and online tool data.

Results

Education sessions reached 475 attendees, in 52 of 95 British Columbia EDs. HEARTSMAP training was well received by clinicians with 96% describing effective content including increased comfort in conducting paediatric psychosocial assessments and confidence in disposition planning after training. Clinicians identified unclear processes and lack of local resources as the main barriers to implementation. One-third of the attendees expressed willingness to use the tool, and 27% of registered clinicians have used the tool postimplementation.

Conclusions

Our approach reached and effectively trained clinicians from over half of the province's EDs to use HEARTSMAP for emergency paediatric psychosocial assessments. For some, this provided greater comfort and confidence for these assessments and the following disposition decisions. This evaluation provides valuable insights on training clinicians to use a paediatric mental health tool within diverse ED settings and emphasized the need for ongoing support and institutional engagement to facilitate local, infrastructural, and operational processes for adoption and maintenance, postdissemination.

Clinical Decision Support Systems and Their Role in Antibiotic Stewardship: a Systematic Review

PURPOSE OF REVIEW

The purpose of this article is to perform a systematic review over the past 5 years on the role and effectiveness of clinical decision support systems (CDSSs) on antibiotic stewardship.

RECENT FINDINGS

CDDS interventions found a significant impact on multiple outcomes relevant to antibiotic stewardship. There are various types of CDSS implementations, both active and passive (provider initiated). Passive interventions were associated with more significant outcomes; however, both interventions appeared effective. In the reviewed literature, CDSSs were consistently associated with decreasing antibiotic consumption and narrowing the spectrum of antibiotic usage. Generally, guideline adherence was improved with CDSS, although this was not universal. The effect on other outcomes, such as mortality, Clostridiodes difficile infections, length of stay, and cost, inconsistently showed a significant difference. Overall, CDDS implementation has effectively decreased antibiotic consumption and improved guideline adherence across the various types of CDSS. Other positive outcomes were noted in certain settings, but were not universal. When creating a new intervention, it is important to identify the optimal structure and deployment of a CDSS for a specific setting.

Development and validation of a pragmatic natural language processing approach to identifying falls in older adults in the emergency department

BACKGROUND

Falls among older adults are both a common reason for presentation to the emergency department, and a major source of morbidity and mortality. It is critical to identify fall patients quickly and reliably during, and immediately after, emergency department encounters in order to deliver appropriate care and referrals. Unfortunately, falls are difficult to identify without manual chart review, a time intensive process infeasible for many applications including surveillance and quality reporting. Here we describe a pragmatic NLP approach to automating fall identification.

METHODS

In this single center retrospective review, 500 emergency department provider notes from older adult patients (age 65 and older) were randomly selected for analysis. A simple, rules-based NLP algorithm for fall identification was developed and evaluated on a development set of 1084 notes, then compared with identification by consensus of trained abstractors blinded to NLP results.

RESULTS

The NLP pipeline demonstrated a recall (sensitivity) of 95.8%, specificity of 97.4%, precision of 92.0%, and F1 score of 0.939 for identifying fall events within emergency physician visit notes, as compared to gold standard manual abstraction by human coders.

CONCLUSIONS

Our pragmatic NLP algorithm was able to identify falls in ED notes with excellent precision and recall, comparable to that of more labor-intensive manual abstraction. This finding offers promise not just for improving research methods, but as a potential for identifying patients for targeted interventions, quality measure development and epidemiologic surveillance.

Broad-spectrum antibiotic use and poor outcomes in community-onset pneumonia: a cohort study

QUESTION

Is broad-spectrum antibiotic use associated with poor outcomes in community-onset pneumonia after adjusting for confounders?

METHODS

We performed a retrospective, observational cohort study of 1995 adults with pneumonia admitted from four US hospital emergency departments. We used multivariable regressions to investigate the effect of broad-spectrum antibiotics on 30-day mortality, length of stay, cost and Clostridioides difficile infection (CDI). To address indication bias, we developed a propensity score using multilevel (individual provider) generalised linear mixed models to perform inverse-probability of treatment weighting (IPTW) to estimate the average treatment effect in the treated. We also manually reviewed a sample of mortality cases for antibiotic-associated adverse events.

RESULTS

39.7% of patients received broad-spectrum antibiotics, but drug-resistant pathogens were recovered in only 3%. Broad-spectrum antibiotics were associated with increased mortality in both the unweighted multivariable model (OR 3.8, 95% CI 2.5-5.9; p<0.001) and IPTW analysis (OR 4.6, 95% CI 2.9-7.5; p<0.001). Broad-spectrum antibiotic use by either analysis was also associated with longer hospital stay, greater cost and increased CDI. Healthcare-associated pneumonia was not associated with mortality independent of broad-spectrum antibiotic use. In manual review we identified antibiotic-associated events in 17.5% of mortality cases.

CONCLUSION

Broad-spectrum antibiotics appear to be associated with increased mortality and other poor outcomes in community-onset pneumonia.

Short-Term Air Pollution and Incident Pneumonia. A Case-Crossover Study

RATIONALE

The relationship between air pollution and pneumonia is poorly understood.

OBJECTIVES

To examine relationships between short-term air pollution exposure and number and severity of pneumonia cases along the Wasatch Front in Utah, a region with periodic high levels of outdoor air pollution.

METHODS

We applied time-stratified case-crossover analyses with distributed lag to patients presenting to seven emergency departments with pneumonia over a 2-year period. We compared levels of particulate matter less than or equal to 2.5 μm in aerodynamic diameter, nitrogen dioxide, and ozone at patient residences with emergency department cases, hospitalizations, objectively defined severe pneumonia, and mortality. We calculated direct cost impacts of particulate matter less than or equal to 2.5 μm in aerodynamic diameter reduction.

RESULTS

We evaluated 4,336 pneumonia cases in seven hospitals. Among adults aged 65 years and older, we found consistently positive associations between particulate matter less than or equal to 2.5 μm in aerodynamic diameter within 6 days of presentation and instances of pneumonia (Lag Day 1 adjusted odds ratio, 1.35 per 10 μg/m³ over 12 μg/m³; 95% confidence interval, 1.16-1.57), severe pneumonia (Lag Day 1 adjusted odds ratio, 1.38; 95% confidence interval, 1.06-1.80), and inpatient mortality (Lag Day 5 adjusted odds ratio, 1.50; 95% confidence interval, 1.03-2.16). Smaller associations were found between nitrogen dioxide exposure and pneumonia occurrence, severity, and inpatient and 30-day mortality. Ozone exposure was modestly associated with increased instance and severity of pneumonia in younger adults. Particulate matter less than or equal to 2.5 μm in aerodynamic diameter and nitrogen dioxide effects were greatest in colder months, and ozone effects were greatest in warmer months. Reduction of particulate matter less than or equal to 2.5 μm in aerodynamic diameter levels to less than 12.0 mg/m³ could prevent 76-112 cases of pneumonia per year in these hospitals serving approximately half of the Wasatch Front's population, reducing direct medical facility costs by $807,000 annually.

CONCLUSIONS

Among older adults, short-term ambient particulate matter less than or equal to 2.5 μm in aerodynamic diameter exposure is associated with more emergency department visits and hospitalizations for pneumonia, severe pneumonia, increased mortality, and increased healthcare costs. Nitrogen dioxide and ozone modestly increase pneumonia risk and illness severity.

Antibiotic Use and Outcomes After Implementation of the Drug Resistance in Pneumonia Score in ED Patients With Community-Onset Pneumonia

BACKGROUND

To guide rational antibiotic selection in community-onset pneumonia, we previously derived and validated a novel prediction tool, the Drug-Resistance in Pneumonia (DRIP) score. In 2015, the DRIP score was integrated into an existing electronic pneumonia clinical decision support tool (ePNa).

METHODS

We conducted a quasi-experimental, pre-post implementation study of ePNa with DRIP score (2015) vs ePNa with health-care-associated pneumonia (HCAP) logic (2012) in ED patients admitted with community-onset pneumonia to four US hospitals. Using generalized linear models, we used the difference-in-differences method to estimate the average treatment effect on the treated with respect to ePNa with DRIP score on broad-spectrum antibiotic use, mortality, hospital stay, and cost, adjusting for available patient-level confounders.

RESULTS

We analyzed 2,169 adult admissions: 1,122 in 2012 and 1,047 in 2015. A drug-resistant pathogen was recovered in 3.2% of patients in 2012 and 2.8% in 2015; inadequate initial empirical antibiotics were prescribed in 1.1% and 0.5%, respectively (P = .12). A broad-spectrum antibiotic was administered in 40.1% of admissions in 2012 and 33.0% in 2015 (P < .001). Vancomycin days of therapy per 1,000 patient days in 2012 were 287.3 compared with 238.8 in 2015 (P < .001). In the primary analysis, the average treatment effect among patients using the DRIP score was a reduction in broad-spectrum antibiotic use (OR, 0.62; 95% CI, 0.39-0.98; P = .039). However, the average effects for ePNa with DRIP on mortality, length of stay, and cost were not statistically significant.

CONCLUSIONS

Electronic calculation of the DRIP score was more effective than HCAP criteria for guiding appropriate broad-spectrum antibiotic use in community-onset pneumonia.

A roadmap for semi-automatically extracting predictive and clinically meaningful temporal features from medical data for predictive modeling

Predictive modeling based on machine learning with medical data has great potential to improve healthcare and reduce costs. However, two hurdles, among others, impede its widespread adoption in hdealthcare. First, medical data are by nature longitudinal. Pre-processing them, particularly for feature engineering, is labor intensive and often takes 50-80% of the model building effort. Predictive temporal features are the basis of building accurate models, but are difficult to identify. This is problematic. Healthcare systems have limited resources for model building, while inaccurate models produce sub-optimal outcomes and are often useless. Second, most machine learning models provide no explanation of their prediction results. However, offering such explanations is essential for a model to be used in usual clinical practice. To address these two hurdles, this paper outlines: 1) a data-driven method for semi-automatically extracting predictive and clinically meaningful temporal features from medical data for predictive modeling; and 2) a method of using these features to automatically explain machine learning prediction results and suggest tailored interventions. This provides a roadmap for future research.

Computerized Clinical Decision Support Systems and Antibiotic Prescribing: A Systematic Review and Meta-analysis

PURPOSE

The aim of this study was to perform a systematic review and meta-analysis of studies performed in primary care centers and hospital facilities that evaluated the effectiveness of computerized clinical decision support systems (CDSSs) in decision making on the prescription of any given antibiotic.

METHODS

We conducted a search of the MEDLINE and EMBASE databases. A meta-analysis was then conducted of all variables with results reported in >2 studies.

FINDINGS

A total of 42 of the 46 studies included in the review identified a statistically significant advantage for CDSSs in ≥1 study variables. The effect of CDSSs on the percentage accuracy of the antibiotic spectrum prescribed empirically with respect to the microbial agent's susceptibility, which is one of the most frequently studied outcome variables, was examined in 7 studies, all undertaken in hospital settings. In all these studies but one, CDSSs resulted in a statistically significant increase in percentage accuracy. The other study variables present in >2 studies had more inconsistent results. Although the results of the meta-analysis of the variables percentage accuracy, antibiotic prescription rate in hospital, percentage adherence to antibiotic prescription guidelines in primary care or hospital, and percentage of inappropriate prescriptions for antibiotics in primary care were statistically significantly favorable to CDSSs; in the case of hospital length of stay and mortality, they were favorable although not statistically significantly.

IMPLICATIONS

CDSSs appear to be useful for variables such as the percentage of appropriate empirical treatment in the hospital setting or to induce changes in antibiotics prescription rate. Even so, more better quality studies are required to draw clearer conclusions in respect of morbidity and mortality outcome variables and other settings.

Scope and Influence of Electronic Health Record-Integrated Clinical Decision Support in the Emergency Department: A Systematic Review

STUDY OBJECTIVE

As electronic health records evolve, integration of computerized clinical decision support offers the promise of sorting, collecting, and presenting this information to improve patient care. We conducted a systematic review to examine the scope and influence of electronic health record-integrated clinical decision support technologies implemented in the emergency department (ED).

METHODS

A literature search was conducted in 4 databases from their inception through January 18, 2018: PubMed, Scopus, the Cumulative Index of Nursing and Allied Health, and Cochrane Central. Studies were included if they examined the effect of a decision support intervention that was implemented in a comprehensive electronic health record in the ED setting. Standardized data collection forms were developed and used to abstract study information and assess risk of bias.

RESULTS

A total of 2,558 potential studies were identified after removal of duplicates. Of these, 42 met inclusion criteria. Common targets for clinical decision support intervention included medication and radiology ordering practices, as well as more comprehensive systems supporting diagnosis and treatment for specific disease entities. The majority of studies (83%) reported positive effects on outcomes studied. Most studies (76%) used a pre-post experimental design, with only 3 (7%) randomized controlled trials.

CONCLUSION

Numerous studies suggest that clinical decision support interventions are effective in changing physician practice with respect to process outcomes such as guideline adherence; however, many studies are small and poorly controlled. Future studies should consider the inclusion of more specific information in regard to design choices, attempt to improve on uncontrolled before-after designs, and focus on clinically relevant outcomes wherever possible.

CDS in a Learning Health Care System: Identifying Physicians' Reasons for Rejection of Best-Practice Recommendations in Pneumonia through Computerized Clinical Decision Support

BACKGROUND

Local implementation of guidelines for pneumonia care is strongly recommended, but the context of care that affects implementation is poorly understood. In a learning health care system, computerized clinical decision support (CDS) provides an opportunity to both improve and track practice, providing insights into the implementation process.

OBJECTIVES

This article examines physician interactions with a CDS to identify reasons for rejection of guideline recommendations.

METHODS

We implemented a multicenter bedside CDS for the emergency department management of pneumonia that integrated patient data with guideline-based recommendations. We examined the frequency of adoption versus rejection of recommendations for site-of-care and antibiotic selection. We analyzed free-text responses provided by physicians explaining their clinical reasoning for rejection, using concept mapping and thematic analysis.

RESULTS

Among 1,722 patient episodes, physicians rejected recommendations to send a patient home in 24%, leaving text in 53%; reasons for rejection of the recommendations included additional or alternative diagnoses beyond pneumonia, and comorbidities or signs of physiologic derangement contributing to risk of outpatient failure that were not processed by the CDS. Physicians rejected broad-spectrum antibiotic recommendations in 10%, leaving text in 76%; differences in pathogen risk assessment, additional patient information, concern about antibiotic properties, and admitting physician preferences were given as reasons for rejection.

CONCLUSION

While adoption of CDS recommendations for pneumonia was high, physicians rejecting recommendations frequently provided feedback, reporting alternative diagnoses, additional individual patient characteristics, and provider preferences as major reasons for rejection. CDS that collects user feedback is feasible and can contribute to a learning health system.

Emergency Department Crowding Is Associated With Delayed Antibiotics for Sepsis

STUDY OBJECTIVE

Barriers to early antibiotic administration for sepsis remain poorly understood. We investigated the association between emergency department (ED) crowding and door-to-antibiotic time in ED sepsis.

METHODS

We conducted a retrospective cohort study of ED sepsis patients presenting to 2 community hospitals, a regional referral hospital, and a tertiary teaching hospital. The primary exposure was ED occupancy rate, defined as the ratio of registered ED patients to licensed ED beds. We defined ED overcrowding as an ED occupancy rate greater than or equal to 1. We used multivariable regression to measure the adjusted association between ED crowding and door-to-antibiotic time (elapsed time from ED arrival to first antibiotic initiation). Using Markov multistate models, we also investigated the association between ED crowding and pre-antibiotic care processes.

RESULTS

Among 3,572 eligible sepsis patients, 70% arrived when the ED occupancy rate was greater than or equal to 0.5 and 14% arrived to an overcrowded ED. Median door-to-antibiotic time was 158 minutes (interquartile range 109 to 216 minutes). When the ED was overcrowded, 46% of patients received antibiotics within 3 hours of ED arrival compared with 63% when it was not (difference 14.4%; 95% confidence interval 9.7% to 19.2%). After adjustment, each 10% increase in ED occupancy rate was associated with a 4.0-minute increase (95% confidence interval 2.8 to 5.2 minutes) in door-to-antibiotic time and a decrease in the odds of antibiotic initiation within 3 hours (odds ratio 0.90; 95% confidence interval 0.88 to 0.93). Increasing ED crowding was associated with slower initial patient assessment but not further delays after the initial assessment.

CONCLUSION

ED crowding was associated with increased sepsis antibiotic delay. Hospitals must devise strategies to optimize sepsis antibiotic administration during periods of ED crowding.

Antimicrobial Stewardship in the Emergency Department

The emergency department (ED) is the hub of the US health care system. Acute infectious diseases are frequently encountered in the ED setting, making this a critical setting for antimicrobial stewardship efforts. Systems level and behavioral stewardship interventions have demonstrated success in the ED setting but successful implementation depends on institutional support and the presence of a physician champion. Antimicrobial stewardship efforts in the ED should target high-impact areas: antibiotic prescribing for nonindicated respiratory tract conditions, such as bronchitis and sinusitis; overtreatment of asymptomatic bacteriuria; and using two antibiotics (double coverage) for uncomplicated cases of cellulitis or abscess.