Effectiveness and Safety of an Emergency Department Code Sepsis Protocol: A Pragmatic Clinical Trial.

RATIONALE: Sepsis care delivery - including initiation of prompt, appropriate antimicrobials - remains suboptimal. OBJECTIVE: Determine direct and off-target effects of emergency department (ED) sepsis care reorganization. METHODS: This pragmatic pilot trial enrolled adult patients presenting November 2019 to February 2021 to an ED in Utah before and after implementation of a multimodal, team-based "Code Sepsis" protocol. Patients presenting to two other EDs where usual care was continued served as contemporaneous controls. The primary outcome was door-to-antimicrobial time among patients meeting Sepsis-3 criteria before ED departure. Secondary and safety outcomes included all-cause 30-day mortality, antimicrobial utilization and overtreatment, and antimicrobial-associated adverse events. Multivariable regression analyses employed difference-in-differences methods to account for trends in outcomes unrelated to the studied intervention. RESULTS: Code Sepsis protocol activation (N=307) exhibited 8.5% sensitivity and 66% positive predictive value for patients meeting sepsis criteria before ED departure. Among 10,151 patients meeting sepsis criteria during the study, adjusted difference-in-differences analysis demonstrated a 13-minute (95% CI 7-19-minute) decrease in door-to-antimicrobial time associated with Code Sepsis implementation (p<0.001). Mortality and clinical safety outcomes were unchanged, but Code Sepsis implementation was associated with increased false-positive presumptive infection diagnosis among patients meeting sepsis criteria in the ED and increased antimicrobial utilization. CONCLUSIONS: Implementation of a team-based protocol for rapid sepsis evaluation and treatment during the COVID-19 pandemic's first year was associated with decreased ED door-to-antimicrobial time but also increased antimicrobial utilization. Measurement of both patient-centered and off-target effects of sepsis care improvement interventions is essential to comprehensive assessment of their value. CLINICAL TRIAL REGISTRATION: ClinicalTrials.gov (NCT04148989) This article is open access and distributed under the terms of the Creative Commons Attribution Non-Commercial No Derivatives License 4.0 (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Incorporating social determinants of health into transmission modeling of COVID-19 vaccine in the US: a scoping review.

During COVID-19 in the US, social determinants of health (SDH) have driven health disparities. However, the use of SDH in COVID-19 vaccine modeling is unclear. This review aimed to summarize the current landscape of incorporating SDH into COVID-19 vaccine transmission modeling in the US. Medline and Embase were searched up to October 2022. We included studies that used transmission modeling to assess the effects of COVID-19 vaccine strategies in the US. Studies' characteristics, factors incorporated into models, and approaches to incorporate these factors were extracted. Ninety-two studies were included. Of these, 11 studies incorporated SDH factors (alone or combined with demographic factors). Various sets of SDH factors were integrated, with occupation being the most common (8 studies), followed by geographical location (5 studies). The results show that few studies incorporate SDHs into their models, highlighting the need for research on SDH impact and approaches to incorporating SDH into modeling. Funding: This research was funded by the Centers for Disease Control and Prevention (CDC).

Factors associated with testing positive for SARS-CoV-2 and evaluation of a recruitment protocol among healthcare personnel in a COVID-19 vaccine effectiveness study.

Objective: The objective of this study was to determine factors associated with testing positive for SARS-CoV-2 among healthcare personnel. Secondary objectives were to assess representativeness of recruited participants and the effectiveness of a multiple-contact protocol for recruiting healthcare personnel in this COVID-19 study. Design: Survey study, conducted as part of an observational test-negative study of COVID-19 vaccine effectiveness. Setting: University of Utah Health system, including both inpatient and outpatient facilities. Participants: Clinical and non-clinical healthcare personnel at University of Utah Health. 1456 were contacted and 503 (34.5%) completed the survey. Cases were all eligible employees testing positive for COVID-19, with 3:1 randomly selected, matched controls (test negative) selected weekly. Methods: Online survey. Results: Significant differences in the demographics of participants and the source population were observed; e.g., nursing staff comprised 31.6% of participants but only 23.3% of the source population. The multiple-contact recruitment protocol increased participation by ten percentage points and ensured equal representation of controls. Potential exposure to illness outside of work was strongly predictive of testing positive for SARS-CoV-2 (OR = 3.74; 95% CI: 2.29, 6.11) whereas potential exposure at work was protective against testing positive (OR: 0.51, 95% CI: 0.29, 0.88). Conclusions: Carefully designed recruitment protocols increase participation and representation of controls, but bias in participant demographics still exists. The negative association between potential workplace exposure and positive test suggests testing bias in the test-negative design. Healthcare personnel's potential exposures to COVID-19 outside of the workplace are important predictors of SARS-CoV-2 seropositivity.

A Linguistic Analysis Examining the Impact of COVID-19 on Pneumonia Diagnosis and Disease Models.

Written clinical language embodies and reflects the clinician's mental models of disease. Prior to the COVID-19 pandemic, pneumonia was shifting away from concern for healthcare-associated pneumonia and toward recognition of heterogeneity of pathogens and host response. How these models are reflected in clinical language or whether they were impacted by the pandemic has not been studied. We aimed to assess changes in the language used to describe pneumonia following the COVID-19 pandemic.

An economic evaluation of the expansion of electronic case reporting in an academic healthcare setting.

Objectives: Determine the economic cost or benefit of expanding electronic case reporting (eCR) for 29 reportable conditions beyond the initial eCR implementation for COVID-19 at an academic health center. Materials and methods: The return on investment (ROI) framework was used to quantify the economic impact of the expansion of eCR from the perspective of an academic health system over a 5-year time horizon. Sensitivity analyses were performed to assess key factors such as personnel cost, inflation, and number of expanded conditions. Results: The total implementation costs for the implementation year were estimated to be $5031.46. The 5-year ROI for the expansion of eCR for the 29 conditions is expected to be 142% (net present value of savings: $7166). Based on the annual ROI, estimates suggest that the savings from the expansion of eCR will cover implementation costs in approximately 4.8 years. All sensitivity analyses yielded a strong ROI for the expansion of eCR. Discussion and conclusion: Our findings suggest a strong ROI for the expansion of eCR at UHealth, with the most significant cost savings observed implementing eCR for all reportable conditions. An early effort to ensure data quality is recommended to expedite the transition from parallel reporting to production to improve the ROI for healthcare organizations. This study demonstrates a positive ROI for the expansion of eCR to additional reportable conditions beyond COVID-19 in an academic health setting, such as UHealth. While this evaluation focuses on the 5-year time horizon, the potential benefit could extend further.

Navigating the Landscape: Barriers and Facilitators in Electronic Case Reporting Implementation Across Public Health Agencies.

OBJECTIVE: The objectives were to identify barriers and facilitators for electronic case reporting (eCR) implementation associated with "organizational" and "people"-based knowledge/processes and to identify patterns across implementation stages to guide best practices for eCR implementation at public health agencies. DESIGN: This qualitative study uses semistructured interviews with key stakeholders across 6 public health agencies. This study leveraged 2 conceptual frameworks for the development of the interview guide and initial codebook and the organization of the findings of thematic analysis. SETTING: Interviews were conducted virtually with informants from public health agencies at varying stages of eCR implementation. PARTICIPANTS: Investigators aimed to enroll 3 participants from each participating public health agency, including an eCR lead, a technical lead, and a leadership informant. MAIN OUTCOME MEASURES: Patterns associated with barriers and facilitators across the eCR implementation stage. RESULTS: Twenty-eight themes were identified throughout interviews with 16 informants representing 6 public health agencies at varying stages of implementation. While there was variation across these levels, 3 distinct patterns were identified, including themes that were described (1) solely as a barrier or facilitator for eCR implementation regardless of implementation stages, (2) as a barrier for those in the early stages but evolved into a facilitator for those in later stages, and (3) as facilitators that were unique to the late-stage implementation. CONCLUSION: This study elucidated critical national, organizational, and person-centric best practices for public health agencies. These included the importance of engagement with the national eCR team, integrated development teams, cross-pollination, and developing solutions with the broader public health mission in mind. While the implementation of eCR was the focus of this study, the findings are generalizable to the broader data modernization efforts within public health agencies.

Differences in perceptions and acceptance of COVID-19 vaccination between vaccine hesitant and non-hesitant persons.

Acceptance of the COVID-19 vaccination becomes more critical as new variants continue to evolve and the United States (US) attempts to move from pandemic response to management and control. COVID-19 stands out in the unique way it has polarized patients and generated sustained vaccine hesitancy over time. We sought to understand differences in perceptions and acceptance of COVID-19 vaccination between vaccine hesitant and non-hesitant patients, with the goal of informing communication and implementation strategies to increase uptake of COVID-19 vaccines in Veteran and non-Veteran communities. This qualitative study used interview data from focus groups conducted by the Department of Veterans Affairs (VA) and the University of Utah; all focus groups were conducted using the same script March-July 2021. Groups included forty-six United States Veterans receiving care at 28 VA facilities across the country and 166 non-Veterans across Utah for a total of 36 one-hour focus groups. We identified perceptions and attitudes toward COVID-19 vaccination through qualitative analysis of focus group participant remarks, grouping connections with identified themes within domains developed based on the questions asked in the focus group guide. Responses suggest participant attitudes toward the COVID-19 vaccine were shaped primarily by vaccine attitude changes over time, impacted by perceived vaccine benefits, risks, differing sources of vaccine information and political ideology. Veterans appeared more polarized, being either largely non-hesitant, or hesitant, whereas non-Veterans had a wider range of hesitancy, with more participants identifying minor doubts and concerns about receiving the vaccine, or simply being altogether unsure about receiving it. Development of COVID-19 vaccine communication strategies in Veteran and non-Veteran communities should anticipate incongruous sources of information and explicitly target community differences in perceptions of risks and benefits associated with the vaccine to generate candid discussions and repair individuals' trust. We believe this could accelerate vaccine acceptance over time.

Disparities in COVID-19 related outcomes in the United States by race and ethnicity pre-vaccination era: an umbrella review of meta-analyses.

Background: Meta-analyses have investigated associations between race and ethnicity and COVID-19 outcomes. However, there is uncertainty about these associations' existence, magnitude, and level of evidence. We, therefore, aimed to synthesize, quantify, and grade the strength of evidence of race and ethnicity and COVID-19 outcomes in the US. Methods: In this umbrella review, we searched four databases (Pubmed, Embase, the Cochrane Database of Systematic Reviews, and Epistemonikos) from database inception to April 2022. The methodological quality of each meta-analysis was assessed using the Assessment of Multiple Systematic Reviews, version 2 (AMSTAR-2). The strength of evidence of the associations between race and ethnicity with outcomes was ranked according to established criteria as convincing, highly suggestive, suggestive, weak, or non-significant. The study protocol was registered with PROSPERO, CRD42022336805. Results: Of 880 records screened, we selected seven meta-analyses for evidence synthesis, with 42 associations examined. Overall, 10 of 42 associations were statistically significant (p ≤ 0.05). Two associations were highly suggestive, two were suggestive, and two were weak, whereas the remaining 32 associations were non-significant. The risk of COVID-19 infection was higher in Black individuals compared to White individuals (risk ratio, 2.08, 95% Confidence Interval (CI), 1.60-2.71), which was supported by highly suggestive evidence; with the conservative estimates from the sensitivity analyses, this association remained suggestive. Among those infected with COVID-19, Hispanic individuals had a higher risk of COVID-19 hospitalization than non-Hispanic White individuals (odds ratio, 2.08, 95% CI, 1.60-2.70) with highly suggestive evidence which remained after sensitivity analyses. Conclusion: Individuals of Black and Hispanic groups had a higher risk of COVID-19 infection and hospitalization compared to their White counterparts. These associations of race and ethnicity and COVID-19 outcomes existed more obviously in the pre-hospitalization stage. More consideration should be given in this stage for addressing health inequity.

Scoping review of modelling studies assessing the impact of disruptions to essential health services during COVID-19.

BACKGROUND: Studies assessing the indirect impact of COVID-19 using mathematical models have increased in recent years. This scoping review aims to identify modelling studies assessing the potential impact of disruptions to essential health services caused by COVID-19 and to summarise the characteristics of disruption and the models used to assess the disruptions. METHODS: Eligible studies were included if they used any models to assess the impact of COVID-19 disruptions on any health services. Articles published from January 2020 to December 2022 were identified from PubMed, Embase and CINAHL, using detailed searches with key concepts including COVID-19, modelling and healthcare disruptions. Two reviewers independently extracted the data in four domains. A descriptive analysis of the included studies was performed under the format of a narrative report. RESULTS: This scoping review has identified a total of 52 modelling studies that employed several models (n=116) to assess the potential impact of disruptions to essential health services. The majority of the models were simulation models (n=86; 74.1%). Studies covered a wide range of health conditions from infectious diseases to non-communicable diseases. COVID-19 has been reported to disrupt supply of health services, demand for health services and social change affecting factors that influence health. The most common outcomes reported in the studies were clinical outcomes such as mortality and morbidity. Twenty-five studies modelled various mitigation strategies; maintaining critical services by ensuring resources and access to services are found to be a priority for reducing the overall impact. CONCLUSION: A number of models were used to assess the potential impact of disruptions to essential health services on various outcomes. There is a need for collaboration among stakeholders to enhance the usefulness of any modelling. Future studies should consider disparity issues for more comprehensive findings that could ultimately facilitate policy decision-making to maximise benefits to all.

Choosing Veterans Affairs: Determinants of post-9/11 Veterans' enrollment in Veterans Affairs health care.

Following recent policy changes, younger Veterans have particularly increased options for where to receive their health care. Although existing research provides some understanding of non-modifiable individual (e.g., age) and external community (e.g., non-VA provider supply) factors that influence VA enrollment, this study focused on modifiable facility access and quality factors that could influence Veterans' decisions to enroll in VA. In this cohort study, we examined enrollment in and use of VA services in the year following military separation as the binary outcome using mixed-effects logistic regressions, stratified by Active and Reserve Components. This study included 260,777 Active and 101,572 Reserve Component post-9/11 Veterans separated from the military in fiscal years 2016 to 2017. Independent variables included 4 access measures for timeliness of VA care and 3 VA quality measures, which are included in VA Medical Centers' performance plans. Eligible Veterans were more likely to enroll in VA when the closest VA had higher quality scores. After accounting for timeliness of VA care and non-modifiable characteristics, rating of primary care (PC) providers was associated with higher VA enrollment for Active Component (odds ratio [OR] = 1.014, 95% confidence interval [CI]: 1.007-1.020). Higher mental health (MH) continuity (OR = 1.039, 95% CI: 1.000-1.078) and rating of PC providers (OR = 1.009, 95% CI: 1.000-1.017) were associated with higher VA enrollment for Reserve Component. Improving facility-specific quality of care may be a way to increase VA enrollment. In a changing policy environment, study results will help VA leadership target changes they can make to manage enrollment of Veterans in VA and deliver needed foundational services.

Sex difference in the discordance between Abbott Architect and EuroImmun serological assays.

Background: This study evaluated the discordance between Abbott Architect SARS-CoV-2 IgG and EUROIMMUN SARS-COV-2 ELISA in a seroprevalence study. Methods: From June 10 to August 15, 2020, 8,246 specimens were dually evaluated by the Abbott Architect SARS-CoV-2 IgG (Abbott) and the EUROIMMUN SARS-CoV-2 ELISA (EI) assays. Sex-stratified phi correlation coefficients were calculated to evaluate the concordance between Abbott and EI assay's quantitative results. Multivariable mixed-effect logistic models were implemented to evaluate the association between assay positivity and sex on a low prevalence sample while controlling for age, race, ethnicity, diabetes, cardiovascular disease, hypertension, immunosuppressive therapy, and autoimmune disease. Results: EI positivity among males was 2.1-fold that of females; however, no significant differences in Abbott positivity were observed between sexes. At the manufacturer-recommended threshold, the phi correlation coefficient for the Abbott and EI qualitative results among females (Φ = 0.47) was 34% greater than males (Φ = 0.35). The unadjusted and fully adjusted models yielded a strong association between sex and positive EI result for the low prevalence subgroup (unadjusted OR: 2.24, CI: 1.63, 3.11, adjusted OR: 3.40, CI: 2.15, 5.39). A similar analysis of Abbott positivity in the low prevalence subgroup did not find an association with any of the covariates examined. Significant quantitative and qualitative discordance was observed between Abbott and EI throughout the seroprevalence study. Our results suggest the presence of sex-associated specificity limitations with the EI assay. As these findings may extend to other anti-S assays utilized for SARS-CoV-2 seroprevalence investigations, further investigation is needed to evaluate the generalizability of these findings.

Implementation of an Antibiotic Stewardship Initiative in a Large Urgent Care Network.

Importance: Urgent Care (UC) encounters result in more inappropriate antibiotic prescriptions than other outpatient setting. Few stewardship interventions have focused on UC. Objective: To evaluate the effectiveness of an antibiotic stewardship initiative to reduce antibiotic prescribing for respiratory conditions in a UC network. Design, Setting, and Participants: This quality improvement study conducted in a UC network with 38 UC clinics and 1 telemedicine clinic included 493 724 total UC encounters. The study compared the antibiotic prescribing rates of all UC clinicians who encountered respiratory conditions for a 12-month baseline period (July 1, 2018, through June 30, 2019) with an intervention period (July 1, 2019, through June 30, 2020). A sustainability period (July 1, 2020, through June 30, 2021) was added post hoc. Interventions: Stewardship interventions included (1) education for clinicians and patients, (2) electronic health record (EHR) tools, (3) a transparent clinician benchmarking dashboard, and (4) media. Occurring independently but concurrent with the interventions, a stewardship measure was introduced by UC leadership into the quality measures, including a financial incentive. Main Outcomes and Measures: The primary outcome was the percentage of UC encounters with an antibiotic prescription for a respiratory condition. Secondary outcomes included antibiotic prescribing when antibiotics were not indicated (tier 3 encounters) and first-line antibiotics for acute otitis media, sinusitis, and pharyngitis. Interrupted time series with binomial generalized estimating equations were used to compare periods. Results: The baseline period included 207 047 UC encounters for respiratory conditions (56.8% female; mean [SD] age, 30.0 [21.4] years; 92.0% White race); the intervention period included 183 893 UC encounters (56.4% female; mean [SD] age, 30.7 [20.8] years; 91.2% White race). Antibiotic prescribing for respiratory conditions decreased from 47.8% (baseline) to 33.3% (intervention). During the initial intervention month, a 22% reduction in antibiotic prescribing occurred (odds ratio [OR], 0.78; 95% CI, 0.71-0.86). Antibiotic prescriptions decreased by 5% monthly during the intervention (OR, 0.95; 95% CI, 0.94-0.96). Antibiotic prescribing for tier 3 encounters decreased by 47% (OR, 0.53; 95% CI, 0.44-63), and first-line antibiotic prescriptions increased by 18% (OR, 1.18; 95% CI, 1.09-1.29) during the initial intervention month. Antibiotic prescriptions for tier 3 encounters decreased by an additional 4% each month (OR, 0.96; 95% CI, 0.94-0.98), whereas first-line antibiotic prescriptions did not change (OR, 1.00; 95% CI, 0.99-1.01). Antibiotic prescribing for respiratory conditions remained stable in the sustainability period. Conclusions and relevance: The findings of this quality improvement study indicated that a UC antibiotic stewardship initiative was associated with decreased antibiotic prescribing for respiratory conditions. This study provides a model for UC antibiotic stewardship.

A qualitative evaluation of frontline clinician perspectives toward antibiotic stewardship programs.

OBJECTIVE: To examine the perspectives of caregivers that are not part of the antibiotic stewardship program (ASP) leadership team (eg, physicians, nurses, and clinical pharmacists), but who interact with ASPs in their role as frontline healthcare workers. DESIGN: Qualitative semistructured interviews. SETTING: The study was conducted in 2 large national healthcare systems including 7 hospitals in the Veterans' Health Administration and 4 hospitals in Intermountain Healthcare. PARTICIPANTS: We interviewed 157 participants. The current analysis includes 123 nonsteward clinicians: 47 physicians, 26 pharmacists, 29 nurses, and 21 hospital leaders. METHODS: Interviewers utilized a semistructured interview guide based on the Consolidated Framework for Implementation Research (CFIR), which was tailored to the participant's role in the hospital as it related to ASPs. Qualitative analysis was conducted using a codebook based on the CFIR. RESULTS: We identified 4 primary perspectives regarding ASPs. (1) Non-ASP pharmacists considered antibiotic stewardship activities to be a high priority despite the added burden to work duties: (2) Nurses acknowledged limited understanding of ASP activities or involvement with these programs; (3) Physicians criticized ASPs for their restrictions on clinical autonomy and questioned the ability of antibiotic stewards to make recommendations without the full clinical picture; And (4) hospital leaders expressed support for ASPs and recognized the unique challenges faced by non-ASP clinical staff. CONCLUSION: Further understanding these differing perspectives of ASP implementation will inform possible ways to improve ASP implementation across clinical roles.

Assessing Pathogen Transmission Opportunities: Variation in Nursing Home Staff-Resident Interactions.

OBJECTIVES: The Centers for Disease Control and Prevention (CDC) recommends implementing Enhanced Barrier Precautions (EBP) for all nursing home (NH) residents known to be colonized with targeted multidrug-resistant organisms (MDROs), wounds, or medical devices. Differences in health care personnel (HCP) and resident interactions between units may affect risk of acquiring and transmitting MDROs, affecting EBP implementation. We studied HCP-resident interactions across a variety of NHs to characterize MDRO transmission opportunities. DESIGN: 2 cross-sectional visits. SETTING AND PARTICIPANTS: Four CDC Epicenter sites and CDC Emerging Infection Program sites in 7 states recruited NHs with a mix of unit care types (≥30 beds or ≥2 units). HCP were observed providing resident care. METHODS: Room-based observations and HCP interviews assessed HCP-resident interactions, care type provided, and equipment use. Observations and interviews were conducted for 7-8 hours in 3-6-month intervals per unit. Chart reviews collected deidentified resident demographics and MDRO risk factors (eg, indwelling devices, pressure injuries, and antibiotic use). RESULTS: We recruited 25 NHs (49 units) with no loss to follow-up, conducted 2540 room-based observations (total duration: 405 hours), and 924 HCP interviews. HCP averaged 2.5 interactions per resident per hour (long-term care units) to 3.4 per resident per hour (ventilator care units). Nurses provided care to more residents (n = 12) than certified nursing assistants (CNAs) and respiratory therapists (RTs) (CNA: 9.8 and RT: 9) but nurses performed significantly fewer task types per interaction compared to CNAs (incidence rate ratio (IRR): 0.61, P < .05). Short-stay (IRR: 0.89) and ventilator-capable (IRR: 0.94) units had less varied care compared with long-term care units (P < .05), although HCP visited residents in these units at similar rates. CONCLUSIONS AND IMPLICATIONS: Resident-HCP interaction rates are similar across NH unit types, differing primarily in types of care provided. Current and future interventions such as EBP, care bundling, or targeted infection prevention education should consider unit-specific HCP-resident interaction patterns.

Electronic surveillance criteria for non-ventilator-associated hospital-acquired pneumonia: Assessment of reliability and validity.

OBJECTIVE: Surveillance of non-ventilator-associated hospital-acquired pneumonia (NV-HAP) is complicated by subjectivity and variability in diagnosing pneumonia. We compared a fully automatable surveillance definition using routine electronic health record data to manual determinations of NV-HAP according to surveillance criteria and clinical diagnoses. METHODS: We retrospectively applied an electronic surveillance definition for NV-HAP to all adults admitted to Veterans' Affairs (VA) hospitals from January 1, 2015, to November 30, 2020. We randomly selected 250 hospitalizations meeting NV-HAP surveillance criteria for independent review by 2 clinicians and calculated the percent of hospitalizations with (1) clinical deterioration, (2) CDC National Healthcare Safety Network (CDC-NHSN) criteria, (3) NV-HAP according to a reviewer, (4) NV-HAP according to a treating clinician, (5) pneumonia diagnosis in discharge summary; and (6) discharge diagnosis codes for HAP. We assessed interrater reliability by calculating simple agreement and the Cohen κ (kappa). RESULTS: Among 3.1 million hospitalizations, 14,023 met NV-HAP electronic surveillance criteria. Among reviewed cases, 98% had a confirmed clinical deterioration; 67% met CDC-NHSN criteria; 71% had NV-HAP according to a reviewer; 60% had NV-HAP according to a treating clinician; 49% had a discharge summary diagnosis of pneumonia; and 82% had NV-HAP according to any definition according to at least 1 reviewer. Only 8% had diagnosis codes for HAP. Interrater agreement was 75% (κ = 0.50) for CDC-NHSN criteria and 78% (κ = 0.55) for reviewer diagnosis of NV-HAP. CONCLUSIONS: Electronic NV-HAP surveillance criteria correlated moderately with existing manual surveillance criteria. Reviewer variability for all manual assessments was high. Electronic surveillance using clinical data may therefore allow for more consistent and efficient surveillance with similar accuracy compared to manual assessments or diagnosis codes.

Hospital admission decisions for older Veterans with community-onset pneumonia: An analysis of 118 U.S. Veterans Affairs Medical Centers.

OBJECTIVES: Age is important for prognosis in community-onset pneumonia, but how it influences admission decisions in the emergency department (ED) is not well characterized. Using clinical data from the electronic health record in a national cohort, we examined pneumonia hospitalization patterns, variation, and relationships with mortality among older versus younger Veterans. METHODS: In a retrospective cohort of patients ≥ 18 years presenting to EDs with a diagnosis of pneumonia at 118 VA Medical Centers January 1, 2006, to December 31, 2016, we compared observed, predicted, and residual hospitalization risk for Veterans < 70, 70-79, and ≥ 80 years of age using generalized estimating equations and machine learning models with 71 patient factors. We examined facility variation in residual hospitalization across facilities and explored whether facility differences in hospitalization risk correlated with differences in 30-day mortality. RESULTS: Among 297,498 encounters, 165,003 (55%) were for Veterans < 70 years, 61,076 (21%) 70-80, and 71,419 (24%) ≥ 80. Hospitalization rates were 52%, 67%, and 76%, respectively. After other patient factors were adjusting for, age 70-79 had an odds ratio (OR) of 1.39 (95% confidence interval [CI] 1.34-1.44) and ≥ 80 had an OR of 2.1 (95% CI 2.0-2.2) compared to age < 70. There was substantial variation in hospitalization across facilities among Veterans < 70 (<35% hospitalization at the lowest decile of facilities vs. > 66% at the highest decile) that was similar but with higher risk for patients 70-79 years (54% vs. 82%) and ≥ 80 years (59% vs. 85%) and remained after accounting for patient factors, with no consistently positive or negative associations with facility-level 30-day mortality. CONCLUSIONS: Older Veterans with community-onset pneumonia experience high risk of hospitalization, with widespread facility variation that has no clear relationship to short-term mortality.

Are Staphylococcus aureus Carrier Types Evidence of Population Heterogeneity?

Asymptomatic colonization by Staphylococcus aureus is a precursor for infection, so identifying the mode and source of transmission which leads to colonization could help in targeting interventions. Longitudinal studies have shown that some people are persistently colonized for years, while others seem to carry S. aureus for weeks or less, and conventional wisdom attributes this disparity to an underlying risk factor in the persistently colonized. We analyze published data with mathematical models of acquisition and carriage to compare this hypothesis with alternatives. The null model assumed a homogeneous population and still produced highly variable colonization durations (mean = 101.7 weeks; 5th percentile, 5.2 weeks; 95th percentile, 304.7 weeks). Simulations showed that this inherent variability, combined with censoring in longitudinal cohort studies, is sufficient to produce the appearance of "persistent carriers," "intermittent carriers," and "noncarriers" in data. Our estimates for colonization duration exhibited sensitivity to the assumption that false-positive test results can occur despite being rare, but our model-based approach simultaneously estimates specificity and sensitivity along with epidemiologic parameters. Our results show it is plausible that S. aureus colonizes people indiscriminately, and improved understanding of the types of exposures which result in colonization is essential.

Standard Vocabularies to Improve Machine Learning Model Transferability With Electronic Health Record Data: Retrospective Cohort Study Using Health Care-Associated Infection.

BACKGROUND: With the widespread adoption of electronic healthcare records (EHRs) by US hospitals, there is an opportunity to leverage this data for the development of predictive algorithms to improve clinical care. A key barrier in model development and implementation includes the external validation of model discrimination, which is rare and often results in worse performance. One reason why machine learning models are not externally generalizable is data heterogeneity. A potential solution to address the substantial data heterogeneity between health care systems is to use standard vocabularies to map EHR data elements. The advantage of these vocabularies is a hierarchical relationship between elements, which allows the aggregation of specific clinical features to more general grouped concepts. OBJECTIVE: This study aimed to evaluate grouping EHR data using standard vocabularies to improve the transferability of machine learning models for the detection of postoperative health care-associated infections across institutions with different EHR systems. METHODS: Patients who underwent surgery from the University of Utah Health and Intermountain Healthcare from July 2014 to August 2017 with complete follow-up data were included. The primary outcome was a health care-associated infection within 30 days of the procedure. EHR data from 0-30 days after the operation were mapped to standard vocabularies and grouped using the hierarchical relationships of the vocabularies. Model performance was measured using the area under the receiver operating characteristic curve (AUC) and F1-score in internal and external validations. To evaluate model transferability, a difference-in-difference metric was defined as the difference in performance drop between internal and external validations for the baseline and grouped models. RESULTS: A total of 5775 patients from the University of Utah and 15,434 patients from Intermountain Healthcare were included. The prevalence of selected outcomes was from 4.9% (761/15,434) to 5% (291/5775) for surgical site infections, from 0.8% (44/5775) to 1.1% (171/15,434) for pneumonia, from 2.6% (400/15,434) to 3% (175/5775) for sepsis, and from 0.8% (125/15,434) to 0.9% (50/5775) for urinary tract infections. In all outcomes, the grouping of data using standard vocabularies resulted in a reduced drop in AUC and F1-score in external validation compared to baseline features (all P<.001, except urinary tract infection AUC: P=.002). The difference-in-difference metrics ranged from 0.005 to 0.248 for AUC and from 0.075 to 0.216 for F1-score. CONCLUSIONS: We demonstrated that grouping machine learning model features based on standard vocabularies improved model transferability between data sets across 2 institutions. Improving model transferability using standard vocabularies has the potential to improve the generalization of clinical prediction models across the health care system.

Examining the impact of ICU population interaction structure on modeled colonization dynamics of Staphylococcus aureus.

BACKGROUND: Complex transmission models of healthcare-associated infections provide insight for hospital epidemiology and infection control efforts, but they are difficult to implement and come at high computational costs. Structuring more simplified models to incorporate the heterogeneity of the intensive care unit (ICU) patient-provider interactions, we explore how methicillin-resistant Staphylococcus aureus (MRSA) dynamics and acquisitions may be better represented and approximated. METHODS: Using a stochastic compartmental model of an 18-bed ICU, we compared the rates of MRSA acquisition across three ICU population interaction structures: a model with nurses and physicians as a single staff type (SST), a model with separate staff types for nurses and physicians (Nurse-MD model), and a Metapopulation model where each nurse was assigned a group of patients. The proportion of time spent with the assigned patient group (γ) within the Metapopulation model was also varied. RESULTS: The SST, Nurse-MD, and Metapopulation models had a mean of 40.6, 32.2 and 19.6 annual MRSA acquisitions respectively. All models were sensitive to the same parameters in the same direction, although the Metapopulation model was less sensitive. The number of acquisitions varied non-linearly by values of γ, with values below 0.40 resembling the Nurse-MD model, while values above that converged toward the Metapopulation structure. DISCUSSION: Inclusion of complex population interactions within a modeled hospital ICU has considerable impact on model results, with the SST model having more than double the acquisition rate of the more structured metapopulation model. While the direction of parameter sensitivity remained the same, the magnitude of these differences varied, producing different colonization rates across relatively similar populations. The non-linearity of the model's response to differing values of a parameter gamma (γ) suggests simple model approximations are appropriate in only a narrow space of relatively dispersed nursing assignments. CONCLUSION: Simplifying assumptions around how a hospital population is modeled, especially assuming random mixing, may overestimate infection rates and the impact of interventions. In many, if not most, cases more complex models that represent population mixing with higher granularity are justified.

Clinician documentation of patient centered care in the electronic health record.

BACKGROUND: In this study we sought to explore the possibility of using patient centered care (PCC) documentation as a measure of the delivery of PCC in a health system. METHODS: We first selected 6 VA medical centers based on their scores for a measure of support for self-management subscale from a national patient satisfaction survey (the Survey for Healthcare Experience-Patients). We accessed clinical notes related to either smoking cessation or weight management consults. We then annotated this dataset of notes for documentation of PCC concepts including: patient goals, provider support for goal progress, social context, shared decision making, mention of caregivers, and use of the patient's voice. We examined the association of documentation of PCC with patients' perception of support for self-management with regression analyses. RESULTS: Two health centers had < 50 notes related to either tobacco cessation or weight management consults and were removed from further analysis. The resulting dataset includes 477 notes related to 311 patients total from 4 medical centers. For a majority of patients (201 out of 311; 64.8%) at least one PCC concept was present in their clinical notes. The most common PCC concepts documented were patient goals (patients n = 126; 63% clinical notes n = 302; 63%), patient voice (patients n = 165, 82%; clinical notes n = 323, 68%), social context (patients n = 105, 52%; clinical notes n = 181, 38%), and provider support for goal progress (patients n = 124, 62%; clinical notes n = 191, 40%). Documentation of goals for weight loss notes was greater at health centers with higher satisfaction scores compared to low. No such relationship was found for notes related to tobacco cessation. CONCLUSION: Providers document PCC concepts in their clinical notes. In this pilot study we explored the feasibility of using this data as a means to measure the degree to which care in a health center is patient centered. PRACTICE IMPLICATIONS: clinical EHR notes are a rich source of information about PCC that could potentially be used to assess PCC over time and across systems with scalable technologies such as natural language processing.