Large Healthcare System Research Articles

Using a Comprehensive On-Site Assessment Process to Reduce Central Line-Associated Bloodstream Infection Rates.

Central line-associated bloodstream infection (CLABSI) rates increased substantially in the United States following the emergence of COVID-19 and subsequent surges. The pandemic resulted in hospital capacities being exceeded and crisis standards of care being implemented for sustained periods. As COVID-19 rates in the United States began to stabilize, some facilities did not return to previous CLABSI rates, indicating a change in practices that had a longer-term impact on CLABSI prevention. The authors' large health care system observed similar increases in CLABSI following the emergence of COVID-19, prompting investigation and intervention in the form of a quality improvement project. To identify changes related to ongoing increases in CLABSI, an assessment team conducted standardized on-site assessments at 11 facilities. Site assessments were considered an intervention, as they involved rigorous preassessment investigations and interviews, case reviews, practice observations, on-site staff interviews, and postassessment support for additional interventions. Nine facilities had enough postassessment data to analyze the impact of intervention. The overall CLABSI rate (infections per 1000 line days) at the 9 facilities in the 6 months prior to intervention was 1.42, and the postassessment rate in the 6 months following intervention was 0.44. This indicates the effectiveness of facility-specific investigation followed by targeted performance improvements to reduce the rate of CLABSI.

Persistent symptoms and conditions among children and adolescents hospitalised with COVID-19 illness: a qualitative study

ObjectivesThere is limited in-depth research exploring persistent symptoms and conditions among children and adolescents who contracted COVID-19 illness that required hospitalisation. The main objective of this study was to conduct...

Group vs Individual Prenatal Care and Gestational Diabetes Outcomes

The impact of group-based prenatal care (GPNC) model in the US on the risk of gestational diabetes (GD) and related adverse obstetric outcomes is unknown. To determine the effects of the GPNC model on risk of GD, its progression, and related adverse obstetric outcomes. This is a single-site, parallel-group, randomized clinical trial conducted between February 2016 and March 2020 at a large health care system in Greenville, South Carolina. Participants were individuals aged 14 to 45 years with pregnancies earlier than 21 weeks' gestational age; follow-up continued to 8 weeks post partum. This study used an intention-to-treat analysis, and data were analyzed from March 2021 to July 2022. Eligible participants were randomized to receive either CenteringPregnancy, a widely used GPNC model, with 10 group-based sessions or traditional individual prenatal care (IPNC). The primary outcome was the incidence of GD diagnosed between 24 and 30 weeks of gestation. The secondary outcomes included progression to A2 GD (ie, GD treated with medications) and GD-related adverse obstetric outcomes (ie, preeclampsia, cesarean delivery, and large for gestational age). Log binomial models were performed to estimate risk differences (RDs), 95% CIs, and P values between GPNC and IPNC groups, adjusting for all baseline covariates. Of all 2348 participants (mean [SD] age, 25.1 [5.4] years; 952 Black participants [40.5%]; 502 Hispanic participants [21.4%]; 863 White participants [36.8%]), 1176 participants were randomized to the GPNC group and 1174 were randomized to the IPNC group. Among all participants, 2144 (91.3%) completed a GD screening (1072 participants [91.3%] in GPNC vs 1071 [91.2%] in IPNC). Overall, 157 participants (6.7%) developed GD, and there was no difference in GD incidence between the GPNC (83 participants [7.1%]) and IPNC (74 participants [6.3%]) groups, with an adjusted RD of 0.7% (95% CI, -1.2% to 2.7%). Among participants with GD, GPNC did not reduce the risk of progression to A2 GD (adjusted RD, -6.1%; 95% CI, -21.3% to 9.1%), preeclampsia (adjusted RD, -7.9%; 95% CI, -17.8% to 1.9%), cesarean delivery (adjusted RD, -8.2%; 95% CI, -12.2% to 13.9%), and large for gestational age (adjusted RD, -1.2%; 95% CI, -6.1% to 3.8%) compared with IPNC. In this secondary analysis of a randomized clinical trial among medically low-risk pregnant individuals, the risk of GD was similar between participants who received GPNC intervention and traditional IPNC, indicating that GPNC may be a feasible treatment option for some patients. ClinicalTrials.gov Identifier: NCT02640638.

Prevalence and correlates of uncontrolled hypertension, persistently uncontrolled hypertension, and hypertensive crisis at a healthcare system.

Uncontrolled hypertension significantly increases risk of cardiovascular disease and death. This study examined the prevalence of uncontrolled hypertension, persistently uncontrolled hypertension, and hypertensive crisis and factors associated with these outcomes in a real-world patient cohort. Electronic medical records from a large healthcare system in North Carolina were used to identify adults with uncontrolled hypertension (last ambulatory blood pressure [BP] measurement ≥140/90; persistently uncontrolled hypertension (≥ 2 ambulatory BP measurements with all readings ≥140/90); and hypertensive crisis (any BP reading ≥ 180/120) in 2019. Generalized linear mixed models tested the association between patient and provider characteristics and each outcome. The study cohort included 213,836 patients (mean age 63.1 (±14.0) years, 55.5% female, 70.8% white). Of these, 29.7% and 13.1% had uncontrolled hypertension and hypertensive crisis, respectively. Among those experiencing hypertensive crisis, > 50% did not have uncontrolled hypertension. Of the 171,061 patients with ≥ 2 BP measurements, 5.9% had persistently uncontrolled hypertension. The likelihood of uncontrolled hypertension, persistently uncontrolled hypertension, and hypertensive crisis was higher in patients with black race (vs. white), self-pay (vs. private), prior emergency room visit, and no attributed primary care provider. Readings taken in the evening (vs. morning) and at specialty (vs. primary care) practices were more likely to meet thresholds for uncontrolled hypertension and hypertensive crisis. Hypertension control remains a significant challenge in healthcare. Health systems may benefit from segmenting their patient population based on factors such as race, prior healthcare use and timing of BP measurement to prioritize outreach and intervention.

Barriers to Human Papillomavirus Vaccination Initiation and Completion among Adults Aged 18-26 Years in a Large Healthcare System.

Human papillomavirus (HPV) is a common sexually transmitted infection, with over 40% prevalence in the US. Oropharyngeal cancers (OPCs) driven by high-risk HPV are increasing (up to 90%), with HPV vaccination being the only prevention available. The aim of this study was to investigate HPV vaccination among patients aged between 18 and 26 years old with at least one encounter at a large healthcare system and identify sociodemographic factors associated with vaccine initiation and completion. A cross-sectional retrospective study was conducted between 2018 and 2021, including 265,554 patients identified from the Clinical Data Warehouse. HPV vaccination status by age, sex, race/ethnicity, insurance type, primary care (PCP) visits in the past year, alcohol, tobacco, illicit drug use, and age at vaccination was examined. Overall, 33.6% of females and 25.4% of males have completed the HPV vaccine. Black Americans were 35% more likely to initiate the vaccine than White Americans but were less likely to complete the entire course. Overall, HPV vaccination prevalence was far below the Health People 2030 goal of 80%, especially in young males. This low rate is troubling, since many patients had a PCP visit and remained unvaccinated, which serves as a missed opportunity for vaccination.

Identifying Patients With Peripheral Artery Disease Using the Electronic Health Record: A Pragmatic Approach

BackgroundPeripheral artery disease (PAD) is underdiagnosed due to poor patient and clinician awareness. Despite this, no widely accepted PAD screening is recommended. ObjectivesThe authors used machine learning to develop an automated risk stratification tool for identifying patients with a high likelihood of PAD. MethodsUsing data from the electronic health record (EHR), ankle-brachial indices (ABIs) were extracted for 3,298 patients. In addition to ABI, we extracted 60 other patient characteristics and used a random forest model to rank the features by association with ABI. The model identified several features independently correlated with PAD. We then built a logistic regression model to predict PAD status on a validation set of patients (n = 1,089), an external cohort of patients (n = 2,922), and a national database (n = 2,488). The model was compared to an age-based and random forest model. ResultsThe model had an area under the curve (AUC) of 0.68 in the validation set. When evaluated on an external population using EHR data, it performed similarly with an AUC of 0.68. When evaluated on a national database, it had an AUC of 0.72. The model outperformed an age-based model (AUC: 0.62; P < 0.001). A random forest model with inclusion of all 60 features did not perform significantly better (AUC: 0.71; P = 0.31). ConclusionsStatistical techniques can be used to build models which identify individuals at high risk for PAD using information accessible from the EHR. Models such as this may allow large health care systems to efficiently identify patients that would benefit from aggressive preventive strategies or targeted-ABI screening.

Endovascular Aneurysm Repair Devices as a Use Case for Postmarketing Surveillance of Medical Devices

The US Food and Drug Administration (FDA) is building a national postmarketing surveillance system for medical devices, moving to a "total product life cycle" approach whereby more limited premarketing data are balanced with postmarketing surveillance to capture rare adverse events and long-term safety issues. To assess the methodological requirements and feasibility of postmarketing device surveillance using endovascular aneurysm repair devices (EVARs), which have been the subject of safety concerns, using clinical data from a large health care system. This retrospective cohort study included patients with electronic health record (EHR) data in the Veterans Affairs Corporate Data Warehouse. Implantation of an AFX Endovascular AAA System (AFX) device (any of 3 iterations) or a non-AFX comparator EVAR device from January 1, 2011, to December 21, 2021. The primary outcomes were rates of type III endoleaks and all-cause mortality; and rates of these outcomes associated with AFX devices compared with non-AFX devices, assessed using Cox proportional hazards regression models and doubly robust causal modeling. Information on type III endoleaks was available only as free-text mentions in clinical notes, while all-cause mortality data could be extracted using structured data. Device-specific information required by the FDA is ascertained using unique device identifiers (UDIs), which include factors such as model numbers, catalog numbers, and manufacturer-specific product codes. The availability of UDIs in EHRs was assessed. In total, 13 941 patients (mean [SD] age, 71.8 [7.4] years) received 1 of the devices of interest (AFX with Strata [AFX-S]: 718 patients [5.2%]; AFX with Duraply [AFX-D]: 404 patients [2.9%]; or AFX2: 682 patients [4.9%]), and 12 137 (87.1%) received non-AFX devices. The UDIs were not recorded in the EHR for any patient with an AFX device, and partial UDIs were available for 19 patients (0.1%) with a non-AFX device. This necessitated the development of advanced natural language processing tools to define the cohort of patients for analysis. The study identified a significantly higher risk of type III endoleaks at 5 years among patients receiving any of the AFX device iterations, including the most recent version, AFX2 (11.6%; 95% CI, 8.1%-15.1%) compared with that among patients with non-AFX devices (5.7%; 95% CI, 2.2%-9.2%; absolute risk difference, 5.9%; 95% CI, 2.3%-9.4%). However, there was no significantly higher all-cause mortality for any of the AFX device iterations, including for AFX2 (19.0%; 95% CI, 16.0%-22.0%) compared with non-AFX devices (18.0%; 95% CI, 15.0%-21.0%; absolute risk difference, 1.0%; 95% CI, -2.1% to 4.1%). The findings of this cohort study suggest that clinical data can be used for the postmarketing device surveillance required by the FDA. The study also highlights ongoing challenges to performing larger-scale surveillance, including lack of consistent use of UDIs and insufficient relevant structured data to efficiently capture certain outcomes of interest.

P135. Drivers of cost variability in single-level posterior and transforaminal lumbar interbody fusions across a large healthcare system in the US: results from 12,932 fusion procedures

P135. Drivers of cost variability in single-level posterior and transforaminal lumbar interbody fusions across a large healthcare system in the US: results from 12,932 fusion procedures

Quality of life after pancreatoduodenectomy: Is the outcome predetermined by the diagnosis?

Non-physiological factors tied to the disease process may drive the diminished quality of life (QoL) after pancreatoduodenectomy (PD). We compared postoperative QoL among patients undergoing PD for either benign or malignant pathology. From 2012 to 2021, 228 patients underwent PD in a large healthcare system. Eighty-two patients (36.0%) were interviewed using the EORTC QLQ-C30 questionnaire. A minimum of 6 months after surgery was required for the survey. QoL outcomes were compared based on diagnosis (benign vs. malignant). Patient mean age was 65 years (21-82), and forty-seven (57%) were men. Most patients underwent surgery for cancer, 76% (n = 62). Grade B postoperative pancreatic fistula incidence was higher in benign cases (30% vs. 6.5%, p = 0.024). Weight loss was more common in malignancy (79% vs. 50%, p = 0.016). Carcinoma patients felt less useful, hopeful, reported less control of their life and certainty of the future, and were less satisfied with their appearance. Carcinoma patients also reported diminished memory, fear of relapse, and greater financial burden. Long-term QoL is inferior in PD patients with carcinoma and is driven by the psychological and socioeconomic implications of malignancy. Supportive resources for pancreas cancer patients should be evaluated and optimized.

JAAD Game Changer: “Estimating the cost of skin cancer detection by dermatology providers in a large health care system”

JAAD Game Changer: “Estimating the cost of skin cancer detection by dermatology providers in a large health care system”

Telehealth-delivered Psychotherapy for the Treatment of Alcohol Use Disorder: Patient Perspectives in the Age of COVID-19.

The COVID-19 pandemic precipitated increases in alcohol use and ushered in virtually delivered health care, creating an opportunity to examine the impacts of telehealth on alcohol use disorder (AUD) treatment. To understand these impacts, we explored perspectives on telehealth-delivered psychotherapy among individuals with AUD. This was a qualitative study using semi-structured interviews. Participants (N = 31) were patients with AUD who had received telehealth-delivered AUD psychotherapy in the last 2 years (n = 11) or had never experienced AUD psychotherapy (n = 20), recruited from two large academically-affiliated health care systems in Michigan between July and August 2020. Participants were asked about perceived barriers and facilitators to AUD psychotherapy, benefits and drawbacks of telehealth-delivered AUD psychotherapy, and changes needed to improve psychotherapy delivery. Interviews were transcribed, coded, and analyzed iteratively using thematic analysis. Participants identified factors relating to perceptions of and experience with telehealth-delivered AUD psychotherapy. Findings reflected four major themes: treatment accessibility, treatment flexibility, treatment engagement, and stigma. Perceptions about telehealth's impact on treatment accessibility varied widely and included benefits (e.g., eliminating transportation challenges) and barriers (e.g., technology costs). Treatment flexibility and treatment engagement factors included the ability to use phone and video and perceptions of receiving care via telehealth, respectively. Telehealth impacts on treatment stigma were also a key theme. Overall, perceptions of telehealth treatment for AUD varied. Participants expressed the importance of options, flexibility, and collaborating on decisions with providers to determine treatment modality. Future research should explore who benefits most from telehealth and avenues to enhance implementation.

Associations between clinical and social factors and anticoagulant prescription among patients with atrial fibrillation: A retrospective cohort study from a large healthcare system.

Patient clinical factors and social determinants of health (SDOH) are associated with an increased risk of stroke for patients with atrial fibrillation (AF); however, the association between these factors and the management of AF is not well characterized, particularly among those factors commonly collected in electronic health records (EHRs). This study used EHR data to evaluate the associations between patient clinical factors and SDOH and prescribing of an oral anticoagulant (OAC) for stroke prevention in AF. This analysis included adult patients with newly diagnosed AF who had ≥2 encounters in the Advocate Aurora Health system in Wisconsin between May 2016 and May 2021. Patient-level demographics, comorbidities, medications, and SDOH were retrospectively extracted from EHRs. Area deprivation index (ADI) was linked to patient records as a measure of socioeconomic status. Of 16,656 patients with AF, 10,898 (65.4%) were prescribed an OAC within the first year of diagnosis. Patients were less likely to be prescribed an OAC (relative risk [95% CI]) if they were widowed (0.98 [0.96-0.99] vs single) or had a history of alcoholism (0.86 [0.79-0.95] vs no history). Most patients (53.3%) received prescriptions from a primary care provider. A linear relationship was found between worsening ADI and increased prescriptions for warfarin vs those for direct-acting OACs. Although guideline-concordant anticoagulant use remained suboptimal, clinical characteristics were strongly associated for whether a patient with AF would be prescribed an OAC. Disparities in patient care regarding the prescribing of OACs due to SDOH and associated behaviors were small but present, particularly for national ADI.

Engaging Emergency Nurses in Strategies to Address the Social Determinants of Health

IntroductionFor patients with social needs, emergency departments can be an essential bridge between the health care system and the community. Emergency nurses’ knowledge of and engagement in this work need to be examined to ensure that efforts for social determinants of health screening and the resulting community connections are effective. However, there is limited research in this area of nursing practice. The purpose of this study is to describe emergency nurses’ knowledge about social needs in their community, assess their knowledge of existing community resources, and examine their perceived confidence to respond to the social needs of their patients. MethodsA cross-sectional survey was conducted with 243 nurses employed in a large regional health care system. Data were collected using an adapted 81-item social determinants of health survey instrument to measure knowledge of social determinants of health, confidence, and frequency of discussing social determinants of health with emergency department patients and awareness of social resources available in the community. Survey participants were asked about barriers to incorporating social determinants of health into their emergency department workflow and to provide general demographic information. Descriptive statistics were used to analyze study results. ResultsMost of the 243 emergency nurse participants believed that addressing social determinants of health was important and that emergency nurses should be involved in issues around social determinants of health. However, most nurses reported limited knowledge about social determinants of health and had very limited knowledge about the resources available in their community to help patients with needs related to food, housing, medical care, and transportation. Nurses reported that although they know that their patients are unlikely to ask for help with social needs during an emergency department visit, they are still unlikely to ask their patients about social needs owing to low confidence about having social needs conversations, limited time, and competing care priorities. Participants advocated for greater case manager presence. Feeling connected to the community was significantly correlated to increased knowledge, confidence, and likelihood to ask about social needs (P < .05). DiscussionThe emergency department is a logical place for screening for social determinants of health and connecting patients with social needs to community resources. Emergency nurses included in this study acknowledged that they want to address the social needs of their patients but reported that they lack knowledge about both social determinants of health and more importantly about the resources available to help patients with social needs. In general, they did not feel confident discussing social determinants of health with their patients in the emergency department and are unlikely to ask about social needs. Barriers to screening include time, competing care priorities, and lack of knowledge and support needed from case managers. Findings from this study have implications for supporting emergency nurses employed in institutions that seek to address social influences of health for the patients in their communities.

One-Year Outcomes of Patients Requiring Tracheostomy Placement Due to Severe Acute Respiratory Syndrome Coronavirus 2 Infection.

What are the characteristics of patients who develop the need for tracheostomy due to SARS-CoV-2 with acute respiratory distress syndrome (ARDS)? What is their 90-day and 1-year survival and are there any identifiable risk factors for mortality and ventilator dependency? Retrospective, follow-up cohort study of adult patients with COVID-19 infection and ARDS who required tracheostomy placement in a large healthcare system. One hundred sixty-four consecutive patients with SARS-CoV-2 admitted to ICUs for ARDS who required tracheostomy placement between March 2020 and March 2021 were identified. One hundred nine (66.5%) were male. Average age was 63.5 years. The most common comorbidities were obesity, hypertension, diabetes mellitus, congestive heart failure, chronic kidney disease, chronic obstructive pulmonary disease (COPD), atrial fibrillation, and asthma. The most common complications during hospitalization were delirium, secondary infections, acute kidney injury, pneumothorax, and venous thromboembolism. Ninety-day and 1-year mortality were 29.9% and 44.5%, respectively. Ninety-six patients (58.5%) were liberated from the ventilator, and 84 (51.2%) had the tracheostomy tube decannulated. Asthma, COPD, atrial fibrillation, and renal replacement therapy requirement in the ICU correlated with increased risk of ventilator dependency. Among survivors at 1 year, 71 patients (43.3%) were residing at home and 20 patients (12.2%) remained in a skilled nursing facility. COVID-19 has resulted in a significant burden of acute critical illness and acute respiratory failure with the need for tracheostomy. A significant percentage of patients with SARS-CoV-2 requiring tracheostomy were alive and at home 1 year after tracheostomy placement. Long-term care support, including tracheostomy, beyond 90 days appears to be beneficial in this patient population and warrants further investigation.

Competitive Algorithms for the Online Minimum Peak Job Scheduling

Algorithms to schedule medical appointments This paper was inspired by a field collaboration effort to develop and disseminate a real-time appointment scheduling decision support tool for an outpatient cancer infusion center in a large healthcare system. Two challenging aspects of scheduling daily medical appointments are that each patient is scheduled upon arrival without knowledge on future patients and that the appointments typically consume scarce physical resources (e.g., chairs, nurses, and doctors). A desirable schedule should have relatively smooth utilization over the course of a day to minimize the peak demand for the scarce resources. This paper develops new real-time (online) algorithms to schedule appointments in medical and other settings. It establishes theoretical properties of these algorithms, showing that they perform close to algorithms that could exploit full retrospective information on all the appointments. Additionally, it provides important insights to guide efficient real-time appointment scheduling policies in practice.

A Multiyear Survey Evaluating Clinician Electronic Health Record Satisfaction.

We assessed how clinician satisfaction with a vendor electronic health record (EHR) changed over time in the 4 years following the transition from a homegrown EHR system to identify areas for improvement. We conducted a multiyear survey of clinicians across a large health care system after transitioning to a vendor EHR. Eligible clinicians from the first institution to transition received a survey invitation by email in fall 2016 and then eligible clinicians systemwide received surveys in spring 2018 and spring 2019. The survey included items assessing ease/difficulty of completing tasks and items assessing perceptions of the EHR's value, usability, and impact. One item assessing overall satisfaction and one open-ended question were included. Frequencies and means were calculated, and comparison of means was performed between 2018 and 2019 on all clinicians. A multivariable generalized linear model was performed to predict the outcome of overall satisfaction. Response rates for the surveys ranged from 14 to 19%. The mean response from 3 years of surveys for one institution, Brigham and Women's Hospital, increased for overall satisfaction between 2016 (2.85), 2018 (3.01), and 2019 (3.21, p < 0.001). We found no significant differences in mean response for overall satisfaction between all responders of the 2018 survey (3.14) and those of the 2019 survey (3.19). Systemwide, tasks rated the most difficult included "Monitoring patient medication adherence," "Identifying when a referral has not been completed," and "Making a list of patients based on clinical information (e.g., problem, medication)." Clinicians disagreed the most with "The EHR helps me focus on patient care rather than the computer" and "The EHR allows me to complete tasks efficiently." Survey results indicate room for improvement in clinician satisfaction with the EHR. Usability of EHRs should continue to be an area of focus to ease clinician burden and improve clinician experience.

Unindicated Cervical Cancer Screening in Adolescent Females Within a Large Healthcare System in the United States

ABSTRACT The increasingly complicated nature of screening guidelines for cervical cancer includes multiple component tests (cytology, genotyping, and high-risk human papilloma virus [HPV] testing) and evolving recommendations. The aim of this study was assessing single-large-healthcare-system trends over time associated with adolescent females and unindicated cervical cancer screening tests. This repeated cross-sectional study analyzed electronic medical records of provider- and patient-specific variables from the Yale New Haven Health System. Included in the study were all 13- to 20-year-old Yale New Haven Health System patients who underwent screening for cervical cancer via either HPV testing or cytology between January 1, 2012 and December 31, 2018. Age, medical histories, surgical histories, self-reported race and ethnicity, and prescribed medications were all patient-specific variables. Provider-specific variables included clinical practice affiliation, clinical degree, and adolescent patient volume (number of medical visits completed by a patient less than 21 every 6 months). In terms of this study, definition of practice setting was either academic (low-income teaching hospitals) or community-based (private practice setting with primarily insured women). Cytologies were categorized as either unindicated or indicated, based on cervical cancer screening guidelines from 2012 (recorded as unindicated for index specimens collected before the 21st birthday in patients lacking immunocompromised conditions). Should an HPV test be performed, any results other than “atypical squamous cells of undetermined significance” received an unindicated status. The final sample size in the study included 794 women from 118 providers (53% community setting and 47% academic setting). Results from analysis of 900 cervical cancer screenings demonstrated that unindicated tests made up the majority of tests (90%), with 87% of these being unindicated cytology testing only and 14% with unindicated HPV testing as well. Reasons for obtaining unindicated tests as recorded in the original clinical history were lacking, as 62% had no justification listed for testing, and those with reasons failed current guideline compliance. The incidence rate (IR) of adolescent unindicated tests decreased by 33% between 2012 and 2018. Although a significantly lower IR of unindicated screening existed in the academic setting, the rate of decline in the IR of unindicated screening during this period was significantly greater in the community than in the academic setting by 11%. A strength of the study included its ability to calculate IRs in all encounters, including sexually transmitted infection or contraception screenings. Limitations included utilization of hospital system cytology reports for obtaining data, thereby hindering any access to data necessary for validation of screening or surveillance indications from previously performed antecedent cytology. Furthermore, the study failed to collect information regarding another guideline-adherence marker: number of indicated index cervical cancer screenings for patients beginning at 21 years. According to an examination by Becerra-Culqui of women having an encounter at the age of 21 years or beyond, 65% missed opportunities for screening. Finally, the authors were unable to comment on both immunocompromised women and eligible young women who missed indicated screening due to collection of data via pathology reports rather than well-woman examinations. Although unindicated screening for women ages 21 and younger experienced significant reductions, areas for improvement remained. As recent guidelines updates may present greater provider and patient challenges, evidence-based strategies are more important than ever in terms of guideline adherence and dissemination.

Large-scale identification of undiagnosed hepatic steatosis using natural language processing

Nonalcoholic fatty liver disease (NAFLD) is a major cause of liver-related morbidity in people with and without diabetes, but it is underdiagnosed, posing challenges for research and clinical management. Here, we determine if natural language processing (NLP) of data in the electronic health record (EHR) could identify undiagnosed patients with hepatic steatosis based on pathology and radiology reports. A rule-based NLP algorithm was built using a Linguamatics literature text mining tool to search 2.15 million pathology report and 2.7 million imaging reports in the Penn Medicine EHR from November 2014, through December 2020, for evidence of hepatic steatosis. For quality control, two independent physicians manually reviewed randomly chosen biopsy and imaging reports (n=353, PPV 99.7%). After exclusion of individuals with other causes of hepatic steatosis, 3007 patients with biopsy-proven NAFLD and 42,083 patients with imaging-proven NAFLD were identified. Interestingly, elevated ALT was not a sensitive predictor of the presence of steatosis, and only half of the biopsied patients with steatosis ever received an ICD diagnosis code for the presence of NAFLD/NASH. There was a robust association for PNPLA3 and TM6SF2 risk alleles and steatosis identified by NLP. We identified 234 disorders that were significantly over- or underrepresented in all subjects with steatosis and identified changes in serum markers (e.g., GGT) associated with presence of steatosis. This study demonstrates clear feasibility of NLP-based approaches to identify patients whose steatosis was indicated in imaging and pathology reports within a large healthcare system and uncovers undercoding of NAFLD in the general population. Identification of patients at risk could link them to improved care and outcomes. The study was funded by US and German funding sources that did provide financial support only and had no influence or control over the research process.

Impact of an Inpatient Allergy Guideline on β-Lactam and Alternative Antibiotic Use.

A guideline identifying when inpatients with penicillin or cephalosporin antibiotic allergy labels (PCAAL) can receive β-lactam antibiotics increased β-lactam receipt at a large northeastern US health care system. To report outcomes of implementing a similar guideline and electronic order set (OS) at an independent academic health care system. Penicillin/cephalosporin receipt (percentage of inpatients receiving full doses) and alternative antibiotic use (days of therapy per 1000 patient-days [DOT/1000PD]) were compared over 3 periods before (February 1, 2017, to January 31, 2018) and after guideline implementation (February 1, 2018, to January 31, 2019), and after OS implementation (February 1, 2019, to January 31, 2020) among inpatients with PCAAL admitted on medical services with access to guideline/OS and education (Medical-PCAAL, n= 8721), surgical services with access to guideline/OS without education (Surgical-PCAAL, n= 5069), and obstetrics/gynecology services without interventions (Ob/Gyn-PCAAL, n= 798) and inpatients without PCAAL admitted on the same services (Medical-No-PCAAL, n= 50,840; Surgical-No-PCAAL, n= 29,845; Ob/Gyn-No-PCAAL, n= 6109). χ2 tests were used to compare categorical variables, and analysis of variance was used to compare continuous and interrupted time series analyses (ITSA) to investigate the guideline/OS implementation effect on penicillin/cephalosporin receipt. In the Medical-PCAAL group, penicillin/cephalosporin receipt increased (58%-68%, P < .001), specifically for cefazolin (8%-11%, P= .02) and third- to fifth-generation cephalosporins (43%-48%, P= .04), and aztreonam use decreased (12 DOT/1000PD, P= .03). In the Medical-No-PCAAL group, penicillin/cephalosporin receipt increased (88%-90%, P= .004), specifically for penicillin (40%-44%, P < .001), without changes in aztreonam use. Significant changes were not observed in these outcomes on surgical or obstetrics/gynecology services. Per ITSA, guideline/OS implementation was associated with increased penicillin/cephalosporin receipt in the Medical-PCAAL group only. Guideline and OS implementation was associated with improved antibiotic stewardship on inpatient services that also received allergy education.

Factors Associated with the Digital Patient Experience of Virtual Care Across Specialties.

We aimed to characterize patient experience with virtual care across medical specialties using validated survey data. Primary objective: to determine whether experience varied by visit modality (virtual vs. in-person) and whether relationships persisted after adjusting for patient and provider characteristics. Secondarily, among physicians with sufficient data, we compared virtual versus in-person patient experience scores at the physician level and identified characteristics associated with better experience scores for virtual care. This was a retrospective analysis of administrative databases from a large New England health care system, including all ambulatory visits from October 1, 2020 to September 30, 2021 with patient experience scores recorded. We compared experience between virtual and in-person at the visit level (score: 0-10) and the physician level for likelihood of recommending the physician to friends or family. We used a series of cross-classified hierarchical models with visits grouped by patient and by physician to decompose sources of variation. Among physicians with sufficient data, we compared physicians with higher virtual versus higher in-person net promoter score (NPS). Of 378,472 visits performed by 3368 physicians, 86,878 (23%) were conducted virtually. Most scored ≥9 for either modality, with a small preference for virtual versus in-person care (9.6 vs. 9.5, p < 0.001). We found that more variation in scores was explained by patient than by physician (22.9% vs. 3%). Visit modality was of minimal explanatory value. Most physicians' virtual and in-person NPS were similar, and virtual visit volume was not associated. We found robust evidence for the parity of patient experience between virtual and in-person modalities across specialties.