Study Protocol for the Artificial Intelligence-Driven Evaluation of Structural Heart Diseases Using Wearable Electrocardiogram (ID-SHD).

Introduction: Portable devices capable of electrocardiogram (ECG) acquisition have the potential to enhance structural heart disease (SHD) management by enabling early detection through artificial intelligence-ECG (AI-ECG) algorithms. However, the performance of these AI algorithms for identifying SHD in a real-world screening setting is unknown. To address this gap, we aim to evaluate the validity of our wearable-adapted AI algorithm, which has been previously developed and validated for detecting SHD from single-lead portable ECGs in patients undergoing routine echocardiograms in the Yale New Haven Hospital (YNHH). Research Methods and Analysis: This is the protocol for a cross-sectional study in the echocardiographic laboratories of YNHH. The study will enroll 585 patients referred for outpatient transthoracic echocardiogram (TTE) as part of their routine clinical care. Patients expressing interest in participating in the study will undergo a screening interview, followed by enrollment upon meeting eligibility criteria and providing informed consent. During their routine visit, patients will undergo a 1-lead ECG with two devices - one with an Apple Watch and the second with another portable 1-lead ECG device. With participant consent, these 1-lead ECG data will be linked to participant demographic and clinical data recorded in the YNHH electronic health records (EHR). The study will assess the performance of the AI-ECG algorithm in identifying SHD, including left ventricular systolic dysfunction (LVSD), valvular disease and severe left ventricular hypertrophy (LVH), by comparing the algorithm's results with data obtained from TTE, which is the established gold standard for diagnosing SHD. Ethics and Dissemination: All patient EHR data required for assessing eligibility and conducting the AI-ECG will be accessed through secure servers approved for protected health information. Data will be maintained on secure, encrypted servers for a minimum of five years after the publication of our findings in a peer-reviewed journal, and any unanticipated adverse events or risks will be reported by the principal investigator to the Yale Institutional Review Board, which has reviewed and approved this protocol (Protocol Number: 2000035532).

Automated Diagnostic Reports from Images of Electrocardiograms at the Point-of-Care.

Timely and accurate assessment of electrocardiograms (ECGs) is crucial for diagnosing, triaging, and clinically managing patients. Current workflows rely on a computerized ECG interpretation using rule-based tools built into the ECG signal acquisition systems with limited accuracy and flexibility. In low-resource settings, specialists must review every single ECG for such decisions, as these computerized interpretations are not available. Additionally, high-quality interpretations are even more essential in such low-resource settings as there is a higher burden of accuracy for automated reads when access to experts is limited. Artificial Intelligence (AI)-based systems have the prospect of greater accuracy yet are frequently limited to a narrow range of conditions and do not replicate the full diagnostic range. Moreover, these models often require raw signal data, which are unavailable to physicians and necessitate costly technical integrations that are currently limited. To overcome these challenges, we developed and validated a format-independent vision encoder-decoder model - ECG-GPT - that can generate free-text, expert-level diagnosis statements directly from ECG images. The model shows robust performance, validated on 2.6 million ECGs across 6 geographically distinct health settings: (1) 2 large and diverse US health systems- Yale-New Haven and Mount Sinai Health Systems, (2) a consecutive ECG dataset from a central ECG repository from Minas Gerais, Brazil, (3) the prospective cohort study, UK Biobank, (4) a Germany-based, publicly available repository, PTB-XL, and (5) a community hospital in Missouri. The model demonstrated consistently high performance (AUROC≥0.81) across a wide range of rhythm and conduction disorders. This can be easily accessed via a web-based application capable of receiving ECG images and represents a scalable and accessible strategy for generating accurate, expert-level reports from images of ECGs, enabling accurate triage of patients globally, especially in low-resource settings.

Biometric contrastive learning for data-efficient deep learning from electrocardiographic images.

OBJECTIVE: Artificial intelligence (AI) detects heart disease from images of electrocardiograms (ECGs). However, traditional supervised learning is limited by the need for large amounts of labeled data. We report the development of Biometric Contrastive Learning (BCL), a self-supervised pretraining approach for label-efficient deep learning on ECG images. MATERIALS AND METHODS: Using pairs of ECGs from 78 288 individuals from Yale (2000-2015), we trained a convolutional neural network to identify temporally separated ECG pairs that varied in layouts from the same patient. We fine-tuned BCL-pretrained models to detect atrial fibrillation (AF), gender, and LVEF < 40%, using ECGs from 2015 to 2021. We externally tested the models in cohorts from Germany and the United States. We compared BCL with ImageNet initialization and general-purpose self-supervised contrastive learning for images (simCLR). RESULTS: While with 100% labeled training data, BCL performed similarly to other approaches for detecting AF/Gender/LVEF < 40% with an AUROC of 0.98/0.90/0.90 in the held-out test sets, it consistently outperformed other methods with smaller proportions of labeled data, reaching equivalent performance at 50% of data. With 0.1% data, BCL achieved AUROC of 0.88/0.79/0.75, compared with 0.51/0.52/0.60 (ImageNet) and 0.61/0.53/0.49 (simCLR). In external validation, BCL outperformed other methods even at 100% labeled training data, with an AUROC of 0.88/0.88 for Gender and LVEF < 40% compared with 0.83/0.83 (ImageNet) and 0.84/0.83 (simCLR). DISCUSSION AND CONCLUSION: A pretraining strategy that leverages biometric signatures of different ECGs from the same patient enhances the efficiency of developing AI models for ECG images. This represents a major advance in detecting disorders from ECG images with limited labeled data.

Biometric Contrastive Learning for Data-Efficient Deep Learning from Electrocardiographic Images.

Objective: Artificial intelligence (AI) detects heart disease from images of electrocardiograms (ECGs), however traditional supervised learning is limited by the need for large amounts of labeled data. We report the development of Biometric Contrastive Learning (BCL), a self-supervised pretraining approach for label-efficient deep learning on ECG images. Materials and Methods: Using pairs of ECGs from 78,288 individuals from Yale (2000-2015), we trained a convolutional neural network to identify temporally-separated ECG pairs that varied in layouts from the same patient. We fine-tuned BCL-pretrained models to detect atrial fibrillation (AF), gender, and LVEF<40%, using ECGs from 2015-2021. We externally tested the models in cohorts from Germany and the US. We compared BCL with random initialization and general-purpose self-supervised contrastive learning for images (simCLR). Results: While with 100% labeled training data, BCL performed similarly to other approaches for detecting AF/Gender/LVEF<40% with AUROC of 0.98/0.90/0.90 in the held-out test sets, it consistently outperformed other methods with smaller proportions of labeled data, reaching equivalent performance at 50% of data. With 0.1% data, BCL achieved AUROC of 0.88/0.79/0.75, compared with 0.51/0.52/0.60 (random) and 0.61/0.53/0.49 (simCLR). In external validation, BCL outperformed other methods even at 100% labeled training data, with AUROC of 0.88/0.88 for Gender and LVEF<40% compared with 0.83/0.83 (random) and 0.84/0.83 (simCLR). Discussion and Conclusion: A pretraining strategy that leverages biometric signatures of different ECGs from the same patient enhances the efficiency of developing AI models for ECG images. This represents a major advance in detecting disorders from ECG images with limited labeled data.

Detection of Left Ventricular Systolic Dysfunction From Electrocardiographic Images.

BACKGROUND: Left ventricular (LV) systolic dysfunction is associated with a >8-fold increased risk of heart failure and a 2-fold risk of premature death. The use of ECG signals in screening for LV systolic dysfunction is limited by their availability to clinicians. We developed a novel deep learning-based approach that can use ECG images for the screening of LV systolic dysfunction. METHODS: Using 12-lead ECGs plotted in multiple different formats, and corresponding echocardiographic data recorded within 15 days from the Yale New Haven Hospital between 2015 and 2021, we developed a convolutional neural network algorithm to detect an LV ejection fraction <40%. The model was validated within clinical settings at Yale New Haven Hospital and externally on ECG images from Cedars Sinai Medical Center in Los Angeles, CA; Lake Regional Hospital in Osage Beach, MO; Memorial Hermann Southeast Hospital in Houston, TX; and Methodist Cardiology Clinic of San Antonio, TX. In addition, it was validated in the prospective Brazilian Longitudinal Study of Adult Health. Gradient-weighted class activation mapping was used to localize class-discriminating signals on ECG images. RESULTS: Overall, 385 601 ECGs with paired echocardiograms were used for model development. The model demonstrated high discrimination across various ECG image formats and calibrations in internal validation (area under receiving operation characteristics [AUROCs], 0.91; area under precision-recall curve [AUPRC], 0.55); and external sets of ECG images from Cedars Sinai (AUROC, 0.90 and AUPRC, 0.53), outpatient Yale New Haven Hospital clinics (AUROC, 0.94 and AUPRC, 0.77), Lake Regional Hospital (AUROC, 0.90 and AUPRC, 0.88), Memorial Hermann Southeast Hospital (AUROC, 0.91 and AUPRC 0.88), Methodist Cardiology Clinic (AUROC, 0.90 and AUPRC, 0.74), and Brazilian Longitudinal Study of Adult Health cohort (AUROC, 0.95 and AUPRC, 0.45). An ECG suggestive of LV systolic dysfunction portended >27-fold higher odds of LV systolic dysfunction on transthoracic echocardiogram (odds ratio, 27.5 [95% CI, 22.3-33.9] in the held-out set). Class-discriminative patterns localized to the anterior and anteroseptal leads (V2 and V3), corresponding to the left ventricle regardless of the ECG layout. A positive ECG screen in individuals with an LV ejection fraction ≥40% at the time of initial assessment was associated with a 3.9-fold increased risk of developing incident LV systolic dysfunction in the future (hazard ratio, 3.9 [95% CI, 3.3-4.7]; median follow-up, 3.2 years). CONCLUSIONS: We developed and externally validated a deep learning model that identifies LV systolic dysfunction from ECG images. This approach represents an automated and accessible screening strategy for LV systolic dysfunction, particularly in low-resource settings.

Detection of left ventricular systolic dysfunction from single-lead electrocardiography adapted for portable and wearable devices.

Artificial intelligence (AI) can detect left ventricular systolic dysfunction (LVSD) from electrocardiograms (ECGs). Wearable devices could allow for broad AI-based screening but frequently obtain noisy ECGs. We report a novel strategy that automates the detection of hidden cardiovascular diseases, such as LVSD, adapted for noisy single-lead ECGs obtained on wearable and portable devices. We use 385,601 ECGs for development of a standard and noise-adapted model. For the noise-adapted model, ECGs are augmented during training with random gaussian noise within four distinct frequency ranges, each emulating real-world noise sources. Both models perform comparably on standard ECGs with an AUROC of 0.90. The noise-adapted model performs significantly better on the same test set augmented with four distinct real-world noise recordings at multiple signal-to-noise ratios (SNRs), including noise isolated from a portable device ECG. The standard and noise-adapted models have an AUROC of 0.72 and 0.87, respectively, when evaluated on ECGs augmented with portable ECG device noise at an SNR of 0.5. This approach represents a novel strategy for the development of wearable-adapted tools from clinical ECG repositories.

Metrics for Monitoring Preoperative Optimization Programs for Total Joint Arthroplasty: A Survey of Orthopaedic Nurses.

Preoperative optimization programs are becoming more common for patients seeking total joint arthroplasty; yet, limited research has been conducted to monitor the long-term effects of these programs on patient outcomes. Our aim was to develop a set of metrics that programs can use to monitor the success of preoperative optimization programs. As part of a larger survey of orthopaedic nurses, we collected data regarding current monitoring techniques for preoperative optimization programs and the feasibility of collecting specific variables. Surgical factors such as length of stay and 30-day readmissions were most often used to monitor the success of preoperative optimization programs. Surgical factors were the most likely to be accessible using the electronic medical record. Surgical factors and patient characteristics are the most feasible components for programs to monitor in order to track the outcomes of patients participating in preoperative optimization programs.

Industry Payments to Radiologists During the Last 5 Years and Impact of COVID-19 Pandemic.

OBJECTIVE: The aim of this study is to assess the trends in industry payments to radiologists and the impact of the COVID-19 pandemic, including trends in different categories of payments. METHODS: The Open Payments Database from CMS was accessed and analyzed for the period from January 1, 2016, to December 31, 2021. Payments were grouped into six categories: consulting fees, education, gifts, research, speaker fees, and royalties or ownership. The total number, value, and types of industry payments to radiologists were subsequently determined and compared pre- and postpandemic from 2016 to 2021. RESULTS: The total number of industry payments and the number of radiologists receiving these payments dropped by 50% and 32%, respectively, between 2019 and 2020, with only partial recovery in 2021. However, the mean payment value and total payment value increased by 177% and 37%, respectively, between 2019 and 2020. Gifts and speaker fees experienced the largest decreases between 2019 and 2020 (54% and 63%, respectively). Research and education grants were also disrupted, with the number of payments decreasing by 37% and 36% and payment value decreasing by 37% and 25%, respectively. However, royalty or ownership increased during the first year of the pandemic (8% for number of payments and 345% for value of payments). CONCLUSIONS: There was significant decline in overall industry payments coinciding with the COVID-19 pandemic, with biggest declines in gifts and speaker fees. The impact on the different categories of payments and recovery in the last 2 years has been heterogeneous.

The Use of Risk Scores in Patient Preoperative Optimization for Total Joint Arthroplasty: A Survey of Orthopaedic Nurse Navigators.

Preoperative optimization of patients seeking total joint arthroplasty is becoming more common, and risk scores, which provide an estimate for the risk of complications following procedures, are often used to assist with the preoperative decision-making process. The aim of this study was to characterize the use of risk scores at institutions that utilize nurse navigators in the preoperative optimization process. The survey included 207 nurse navigators identified via the National Association of Orthopaedic Nurses to better understand the use of risk scores in preoperative optimization and the different factors that are included in these risk scores. The study found that 48% of responding nurse navigators utilized risk scores in the preoperative optimization process. These risk scores often included patient comorbidities such as diabetes (85%) and body mass index (87%). Risk scores are commonly used by nurse navigators in preoperative optimization and involve a variety of comorbidities and patient-specific factors.

Orthopaedic Nurse Navigator Involvement in Preoperative Optimization for Total Joint Arthroplasty.

Patients seeking total joint arthroplasty frequently undergo preoperative optimization with the assistance of nurse navigators to facilitate interactions between patients, consulting services, and the orthopaedic surgical team. Given the enormous impact nurse navigator programs have on reducing postoperative complications, our aim is to characterize the involvement of nurse navigators in preoperative optimization programs across the country. We conducted a survey of nurse navigators identified through the National Association of Orthopaedic Nurses to assess the involvement of nurse navigators in the preoperative optimization process. Sixty-seven percent of responding nurse navigators were involved in preoperative optimization, including components such as heart disease (53%) and poorly controlled diabetes (52%). Orthopaedic nurse navigators are commonly involved in preoperative optimization programs for total joint arthroplasty but most of these involve gated yes/no checklists with limited established referral care pathways. Only some of the programs include standardized referrals for specific medical comorbidities.

Total Joint Arthroplasty Direct-to-Consumer Advertising by Medical Device Companies Lacks Patient Diversity.

Background: Obese and African American populations suffer from higher incidence of hip and knee osteoarthritis, yet African Americans are less likely to undergo total hip and knee arthroplasty (TJA). Patient interest in TJA is a necessary first step for surgery. Medical device company direct-to-consumer advertising for TJA represents 1 potential factor driving disparities in utilization. Here we analyze demographics of models represented in medical device company direct-to-consumer TJA advertisements to understand whether advertisement content correlates with the population in need. Methods: We analyzed medical device company pamphlets, websites, and banner and video advertisements of the top 4 medical device companies in US arthroplasty sales, collected via ad-specific search engine and direct correspondence. Variables include model race, sex, age, and weight. Pearson likelihood ratio tests were used to compare categorical variables. Results: Of the 116 advertisements collected, the model featured in the advertisement was white in 69.8%. The proportion of white models differed across medical device companies (company C, 75%) (P < .001) and advertisement type (video, 81.8%) (P < .001). Only 2.6% of advertisements featured obese models. Neither company C nor D, nor pamphlet or website advertisements used obese models. Conclusions: Direct-to-consumer advertising from the top 4 orthopedic US medical device companies does not represent the population in need: While TJA remains underutilized by African American/Hispanic patients, models were overwhelmingly white. While obese patients are known to need TJA, patients in the advertisements were overwhelmingly not obese. We advocate for medical device companies to refocus their advertising strategies to target diverse patients in need of TJA. Level of evidence: III (retrospective cohort study).

Charges for Initial Visits for Uninsured Patients at Musculoskeletal Urgent Care Centers in the US.

Importance: In recent years, specialized musculoskeletal urgent care centers (MUCCs) have opened across the US. Uninsured patients may increasingly turn to these orthopedic-specific urgent care centers as a lower-cost alternative to emergency department or general urgent care center visits. Objective: To assess out-of-pocket costs and factors associated with these costs at MUCCs for uninsured and underinsured patients in the US. Design, Setting, and Participants: In this survey study, a national secret shopper survey was conducted in June 2019. Clinics identified as MUCCs in 50 states were contacted by telephone by investigators using a standardized script and posing as uninsured patients seeking information on the out-of-pocket charge for a new patient visit. Exposures: State Medicaid expansion status, clinic Medicaid acceptance status, state Medicaid reimbursement rate, median income per zip code, and clinic region. Main Outcomes and Measures: The primary outcome was each clinic's out-of-pocket charge for a level 3 visit, defined as a new patient office visit requiring medical decision-making of low complexity. Linear regression was used to examine correlations of price with clinic policy against accepting Medicaid, median income per zip code, and Medicaid reimbursement for a level 3 visit. Results: Of 565 MUCCs identified, 558 MUCCs were able to be contacted (98.8%); 536 of the 558 MUCCs (96.1%) disclosed a new patient visit out-of-pocket charge. Of those, 313 (58.4%) accepted Medicaid insurance and 326 (60.8%) were located in states with expanded Medicaid at the time of the survey. The mean (SD) price of a visit to an MUCC was $250 ($110). Clinic policy against accepting Medicaid (ß, 22.91; 95% CI, 12.57-33.25; P < .001), higher median income per zip code (ß, 0.00056; 95% CI, 0.00020-0.00092; P = .003), and increased Medicaid reimbursement for a level 3 visit (ß, 0.737; 95% CI, 0.158-1.316; P = .01) were positively correlated with visit price. The overall regression was statistically significance (R2 = 0.084; P < .001). Conclusions and Relevance: In this survey study, MUCCs charged a mean price of $250 for a new patient visit. Medicaid acceptance policy, median income per zip code, and Medicaid reimbursement for a level 3 visit were associated with differences in out-of-pocket charges. These findings suggest that accessibility to orthopedic urgent care at MUCCs may be limited for underinsured and uninsured patients.

Evaluating musculoskeletal urgent care center triage and transfer of emergency conditions for emergency surgical assessment and intervention.

Musculoskeletal urgent care centers (MUCCs) are an alternative to emergency departments (EDs) for patients to seek care for low acuity orthopedic injuries such as ankle sprains or joint pain, but are not equipped to manage orthopedic emergencies that require a higher level of care provided in the ED. This study aims to evaluate telephone and online triage practices as well as ED transfer procedures for MUCCs for patients presenting with an orthopedic condition requiring urgent surgical intervention. We called 595 MUCCs using a standardized script presenting as a critical patient with symptoms of lower extremity compartment syndrome. We compared direct ED referral frequency and triage frequency for MUCCs for patients insured by either Medicaid or by private insurance. We found that patients presenting with an apparent compartment syndrome were directly referred to the ED by < 1 in 5 MUCCs. Additionally, < 5% of patients were asked additional triage questions that would increase clinician suspicion for compartment syndrome and allow MUCCs to appropriately direct patients to the ED. MUCCs provide limited telephone and online triage for patients, which may result in delays of care for life or limb threatening injuries that require ED resources such as sedation, reductions, and emergency surgery. However, when MUCCs did conduct triage, it significantly increased the likelihood that patients were appropriately referred to the ED. Level of Evidence: Level II, prognostic study.

Musculoskeletal Urgent Care Centers Restrict Access for Patients with Medicaid Insurance Based on Policy and Location.

BACKGROUND: As the urgent care landscape evolves, specialized musculoskeletal urgent care centers (MUCCs) are becoming more prevalent. MUCCs have been offered as a convenient, cost-effective option for timely acute orthopaedic care. However, a recent "secret-shopper" study on patient access to MUCCs in Connecticut demonstrated that patients with Medicaid had limited access to these orthopaedic-specific urgent care centers. To investigate how generalizable these regional findings are to the United States, we conducted a nationwide secret-shopper study of MUCCs to identify determinants of patient access. QUESTIONS/PURPOSES: (1) What proportion of MUCCs in the United States provide access for patients with Medicaid insurance? (2) What factors are associated with MUCCs providing access for patients with Medicaid insurance? (3) What barriers exist for patients seeking care at MUCCs? METHODS: An online search of all MUCCs across the United States was conducted in this cross-sectional study. Three separate search modalities were used to gather a complete list. Of the 565 identified, 558 were contacted by phone with investigators posing over the telephone as simulated patients seeking treatment for a sprained ankle. Thirty-nine percent (216 of 558) of centers were located in the South, 13% (71 of 558) in the West, 25% (138 of 558) in the Midwest, and 24% (133 of 558) in New England. This study was given an exemption waiver by our institution's IRB. MUCCs were contacted using a standardized script to assess acceptance of Medicaid insurance and identify barriers to care. Question 1 was answered through determining the percentage of MUCCs that accepted Medicaid insurance. Question 2 considered whether there was an association between Medicaid acceptance and factors such as Medicaid physician reimbursements or MUCC center type. Question 3 sought to characterize the prevalence of any other means of limiting access for Medicaid patients, including requiring a referral for a visit and disallowing continuity of care at that MUCC. RESULTS: Of the MUCCs contacted, 58% (323 of 558) accepted Medicaid insurance. In 16 states, the proportion of MUCCs that accepted Medicaid was equal to or less than 50%. In 22 states, all MUCCs surveyed accepted Medicaid insurance. Academic-affiliated MUCCs accepted Medicaid patients at a higher proportion than centers owned by private practices (odds ratio 14 [95% CI 4.2 to 44]; p < 0.001). States with higher Medicaid physician reimbursements saw proportional increases in the percentage of MUCCs that accepted Medicaid insurance under multivariable analysis (OR 36 [95% CI 14 to 99]; p < 0.001). Barriers to care for Medicaid patients characterized included location restriction and primary care physician referral requirements. CONCLUSION: It is clear that musculoskeletal urgent care at these centers is inaccessible to a large segment of the Medicaid-insured population. This inaccessibility seems to be related to state Medicaid physician fee schedules and a center's affiliation with a private orthopaedic practice, indicating how underlying financial pressures influence private practice policies. Ultimately, the refusal of Medicaid by MUCCs may lead to disparities in which patients with private insurance are cared for at MUCCs, while those with Medicaid may experience delays in care. Going forward, there are three main options to tackle this issue: increasing Medicaid physician reimbursement to provide a financial incentive, establishing stricter standards for MUCCs to operate at the state level, or streamlining administration to reduce costs overall. Further research will be necessary to evaluate which policy intervention will be most effective. LEVEL OF EVIDENCE: Level II, prognostic study.

Urgent care center wait times increase for COVID-19 results in August 2020, with rapid testing availability limited.

BACKGROUND: In a response to the pandemic, urgent care centers (UCCs) have gained a critical role as a common location for COVID-19 testing. We sought to characterize the changes in testing accessibility at UCCs between March and August 2020 on the basis of testing availability (including rapid antigen testing), wait time for test results, cost of visits, and cost of tests. METHODS: Data were collected using a secret shopper methodology. Researchers contacted 250 UCCs in 10 states. Investigators used a standardized script to survey centers on their COVID-19 testing availability and policies. UCCs were initially contacted in March and re-called in August. T-tests and chi-square tests were conducted to identify differences between March and August data and differences by center classification. RESULTS: Our results indicate that both polymerase chain reaction (PCR) tests to detect COVID-19 genetic material and rapid antigen COVID-19 tests have increased in availability. However, wait times for PCR test results have significantly increased to an average of 5.79 days. Additionally, a high proportion of UCCs continue to charge for tests and visits and no significant decrease was found in the proportion of UCCs that charge for COVID-19 testing from March to August. Further, no state reported a majority of UCCs with rapid testing available, indicating an overall lack of rapid testing. CONCLUSIONS: From March to August, COVID-19 testing availability gradually improved. However, many barriers lie in access to COVID-19 testing, including testing costs, visit costs, and overall lack of availability of rapid testing in the majority of UCCs. Despite the passage of the CARES Act, these results suggest that there is room for additional policy to improve accessibility to testing, specifically rapid testing.

Urgent Care Centers Delay Emergent Surgical Care Based on Patient Insurance Status in The United States.

OBJECTIVE: Patients may call urgent care centers (UCCs) with urgent surgical conditions but may not be properly referred to a higher level of care. This study aims to characterize how UCCs manage Medicaid and privately insured patients who present with an emergent condition. METHODS: Using a standardized script, we called 1245 randomly selected UCCs in 50 states on 2 occasions. Investigators posed as either a Medicaid or a privately-insured patient with symptoms of an incarcerated inguinal hernia. Rates of direct emergency department (ED) referral were compared between insurance types. RESULTS: A total of 1223 (98.2%) UCCs accepted private insurance and 981 (78.8%) accepted Medicaid. At the 971 (78.0%) UCCs that accepted both insurance types, direct-to-ED referral rates for private and Medicaid patients were 27.9% and 33.8%, respectively. Medicaid patients were significantly more likely than private patients to be referred to the ED [odds ratio (OR) 1.32, 95% confidence interval (CI) 1.09-1.60]. Private patients who were triaged by a clinician compared to nonclinician staff were over 6 times more likely to be referred to the ED (OR 6.46, 95% CI 4.63-9.01). Medicaid patients were nearly 9 times more likely to have an ED referral when triaged by a clinician (OR 8.72, 95% CI 6.19-12.29). CONCLUSIONS: Only one-third of UCCs across the United States referred an apparent emergent surgical case to the ED, potentially delaying care. Medicaid patients were more likely to be referred directly to the ED versus privately insured patients. All patients triaged by clinicians were significantly more likely to be referred to the ED; however, the disparity between private and Medicaid patients remained.

COVID-19 testing capabilities at urgent care centers in states with greatest disease burden.

While rapid and accessible diagnosis is paramount to monitoring and reducing the spread of disease, COVID-19 testing capabilities across the U.S. remain constrained. For many individuals, urgent care centers (UCCs) may offer the most accessible avenue to be tested. Through a phone survey, we describe the COVID-19 testing capabilities at UCCs and provide a snapshot highlighting the limited COVID-19 testing capabilities at UCCs in states with the greatest disease burden.