Search

This article revisits the Mincer earnings function and presents comparable estimates of the average monetary returns associated with an additional year of education across different regions worldwide. In contrast to the traditional Ordinary Least Squares (OLS) method commonly employed in the literature, this study applied a cutting-edge approach known as Support Vector Regression (SVR), which belongs to the family of machine learning (ML) algorithms. SVR is specifically chosen to address the bias arising from underfitting inherent in OLS. The analysis focuses on recent data spanning from 2010 to 2018, ensuring temporal homogeneity across the examined regions. The findings reveal that each additional year of education, on average, yields a private rate of returns of 10.4%. Notably, Sub-Saharan Africa exhibits the highest returns to education at 17.8%, while Europe demonstrates the lowest returns at 7.2%. Moreover, higher education is associated with the highest returns across the regions, with a rate of 12%, whereas primary education yields returns of 10%. Interestingly, women generally experience higher returns than men, with rates of 10.6 and 10.1%, respectively. Over time, the returns to education exhibit a modest decline, decreasing at a rate of approximately 0.1% per year, while the average duration of education demonstrates an increase of 0.16 years per year (1% per year). The application of the state-of-the-art ML technique, SVR, not only improves the accuracy of estimates but also enhances predictive performance measures such as the coefficient of determination (R[superscript 2]) and Root Mean Square Error (RMSE) when compared to the OLS method. The implications drawn from these findings emphasize the need for expanding university education, as well as investments in primary education, along with significant attention toward promoting girls' education. These findings hold considerable importance for policymakers who are tasked with making informed decisions regarding education expenditure and the implementation of education financing programs.

Through co-occurrence analysis of 1139 documents (1964-2018) we identified discussions about the implementation of student teaching evaluation (SET). We found that: (1) Attention to SET originated in the US in the 1970s, spreading to German-speaking countries in the mid-1990s and continuing in China and Latin America in the early 2000s. (2) SET is commonly viewed as a control tool deserving methodological improvement, while bias is debated in the US. We also found local trajectories: (3) Whereas in the US and Latin America SET is primarily seen as a management tool, German-speaking and Chinese authors reflect more on improving teaching. Chinese scholars consider SET a valid instrument for state control associated with artificial intelligence. Also, (4) SET is commonly used in medical education in the US and the German-speaking region and in physical education in China. We conclude that SET is discussed cross-nationally but affected by regional path dependencies.

For the past 70 years, researchers and experimental musicians have been working with computersynthesized music, forming a collaborative relationship with generative artificial intelligences known as human--AI co-creation. The last several years have shown that musical artists are quickly adopting AI tools to produce music for AI music competitions and for commercial production of songs and albums. The United States Copyright Office, in response to this trend, has released its latest policy revisions to clearly define what is eligible for copyright registration. Soon after, the Program for Cooperative Cataloging (PCC) also released new guidelines, providing recommendations for how library catalogers should address AI-generated materials. In both cases, they reject the notion of considering AI as a contributor. The language in each of these policies, however, is self-contradicting, showing that they are ill equipped to address generative AI. This study leverages critical textual analysis and qualitative content analysis and uses case examples to probe the manner in which these policies regard generative AI. Recommendations are made for addressing shortcomings in the PCC's policies, and moral philosophical frameworks such as virtue ethics and consequentialism support arguments for supplementing catalog item records with information from authoritative external sources, deviating from this policy for the sake of truth-seeking. [ABSTRACT FROM AUTHOR]

Carefully standardized facial images of 591 participants were taken in the laboratory while controlling for self-presentation, facial expression, head orientation, and image properties. They were presented to human raters and a facial recognition algorithm: both humans (r =.21) and the algorithm (r =.22) could predict participants' scores on a political orientation scale (Cronbach's α =.94) decorrelated with age, gender, and ethnicity. These effects are on par with how well job interviews predict job success, or alcohol drives aggressiveness. The algorithm's predictive accuracy was even higher (r =.31) when it leveraged information on participants' age, gender, and ethnicity. Moreover, the associations between facial appearance and political orientation seem to generalize beyond our sample: The predictive model derived from standardized images (while controlling for age, gender, and ethnicity) could predict political orientation (r ≈.13) from naturalistic images of 3,401 politicians from the United States, the United Kingdom, and Canada. The analysis of facial features associated with political orientation revealed that conservatives tended to have larger lower faces. The predictability of political orientation from standardized images has critical implications for privacy, the regulation of facial recognition technology, and understanding the origins and consequences of political orientation. Public Significance Statement: We demonstrate that political orientation can be predicted from neutral facial images by both humans and algorithms, even when factors like age, gender, and ethnicity are accounted for. This indicates a connection between political leanings and inherent facial characteristics, which are largely beyond an individual's control. Our findings underscore the urgency for scholars, the public, and policymakers to recognize and address the potential risks of facial recognition technology to personal privacy. [ABSTRACT FROM AUTHOR]

The objective of the study is to explore the associations of screen time, cyberbullying victimization, and suicidality during the COVID‐19 pandemic in female and male high school schools, especially the sex differences in the mediation effect of cyberbullying on the relation between screen time and suicidality. This study analysed the direct paths and mediation effects between variables among the 13,982 participants (female: 49%; male: 51%; age 15–17: 74%) in the Youth Risk Behaviour Survey 2021 (YRBS) using the Structural Equation Model and the Monte Carlo methods in Stata. Multiple‐group analysis was conducted to compare sex differences in the mediation effects. Elevated screen time increased the likelihood of suicide ideation (male: OR: 1.50, 95% CI: 1.26–1.79; female: OR: 1.47, 95% CI: 1.28–1.70), suicide plan (male: OR: 1.56, 95% CI: 1.29–1.89; female: OR: 1.45, 95% CI: 1.24–1.69), suicide attempts (female: OR: 1.23, 95% CI: 1.02–1.48). Cyberbully victims had higher odds of suicide ideation (female: OR: 3.69, 95% CI: 3.25–4.17; male: OR: 4.50; 95% CI: 3.80–5.34), suicide plan (female: OR: 3.74; 95% CI: 3.28–4.25; male: OR: 5.03; 95% CI: 4.22–6.01), and suicide attempt (female: OR: 4.24; 95% CI: 3.66–4.92; male: OR: 4.70; 95% CI: 3.81–5.79). Sex differences were revealed in the mediation effects on suicide ideations (χ2(2) = 8.72, p <.05), suicide attempts (χ2(2) = 8.80, p <.05), and suicide overall (χ2(2) = 6.42, p <.05), where the effects were stronger in female students than in males. Screen time and cyberbullying victimization were directly associated with suicide‐related behaviours in adolescents. Cyberbullying victimization in females had stronger mediation effects than in males. These findings emphasized the importance of understanding the new hybrid psychosocial dynamics and creating a healthy hybrid psychosocial environment, especially for female adolescents. [ABSTRACT FROM AUTHOR]

Recent studies have documented the type of content that is most likely to spread widely, or go "viral," on social media, yet little is known about people's perceptions of what goes viral or what should go viral. This is critical to understand because there is widespread debate about how to improve or regulate social media algorithms. We recruited a sample of participants that is nationally representative of the U.S. population (according to age, gender, and race/ethnicity) and surveyed them about their perceptions of social media virality (n = 511). In line with prior research, people believe that divisive content, moral outrage, negative content, high-arousal content, and misinformation are all likely to go viral online. However, they reported that this type of content should not go viral on social media. Instead, people reported that many forms of positive content—such as accurate content, nuanced content, and educational content—are not likely to go viral even though they think this content should go viral. These perceptions were shared among most participants and were only weakly related to political orientation, social media usage, and demographic variables. In sum, there is broad consensus around the type of content people think social media platforms should and should not amplify, which can help inform solutions for improving social media. [ABSTRACT FROM AUTHOR]

Objectives IgG4-related disease (IgG4-RD) can affect nearly any organ and is often treated with glucocorticoids, which contribute to organ damage and toxicity. Comorbidities and healthcare utilization in IgG4-RD are poorly understood. Methods We conducted a cohort study using claims data from a US managed care organization. Incident IgG4-RD cases were identified using a validated algorithm; general population comparators were matched by age, sex, race/ethnicity and index date. The frequency of 21 expert-defined clinical outcomes associated with IgG4-RD or its treatment and healthcare-associated visits and costs were assessed 12 months before and 36 months after the index date (date of earliest IgG4-RD-related claim). Results There were 524 cases and 5240 comparators. Most cases received glucocorticoids prior to (64.0%) and after (85.1%) the index date. Nearly all outcomes, many being common glucocorticoid toxicities, occurred more frequently in cases vs comparators. During follow-up, the largest differences between cases and comparators were seen for gastroesophageal reflux disease (prevalence difference: +31.2%, P  < 0.001), infections (+17.3%, P  < 0.001), hypertension (+15.5%, P  < 0.01) and diabetes mellitus (+15.0%, P  < 0.001). The difference in malignancy increased during follow-up from +8.8% to +12.5% (P  < 0.001). Some 17.4% of cases used pancreatic enzyme replacement therapy during follow-up. Over follow-up, cases were more often hospitalized (57.3% vs 17.2%, P  < 0.01) and/or had an emergency room visit (72.0% vs 36.7%, P  < 0.01); all costs were greater in cases than comparators. Conclusions Patients with IgG4-RD are disproportionately affected by adverse outcomes, some of which may be preventable or modifiable with vigilant clinician monitoring. Glucocorticoid-sparing treatments may improve these outcomes. [ABSTRACT FROM AUTHOR]

Task Force on 'Clinical Algorithms for Fracture Risk' commissioned by the American Society for Bone and Mineral Research (ASBMR) Professional Practice Committee has recommended that FRAX® models in the US do not include adjustment for race and ethnicity. This position paper finds that an agnostic model would unfairly discriminate against the Black, Asian and Hispanic communities and recommends the retention of ethnic and race-specific FRAX models for the US, preferably with updated data on fracture and death hazards. In contrast, the use of intervention thresholds based on a fixed bone mineral density unfairly discriminates against the Black, Asian and Hispanic communities in the US. This position of the Working Group on Epidemiology and Quality of Life of the International Osteoporosis Foundation (IOF) is endorsed both by the IOF and the European Society for Clinical and Economic Aspects of Osteoporosis, Osteoarthritis and Musculoskeletal Diseases (ESCEO). [ABSTRACT FROM AUTHOR]

Objective: Utilize machine learning models to identify factors associated with seeking medical care for migraine. Background: Migraine is a leading cause of disability worldwide, yet many people with migraine do not seek medical care. Methods: The web‐based survey, ObserVational survey of the Epidemiology, tReatment and Care Of MigrainE (US), annually recruited demographically representative samples of the US adult population (2018–2020). Respondents with active migraine were identified via a validated diagnostic questionnaire and/or a self‐reported medical diagnosis of migraine, and were then asked if they had consulted a healthcare professional for their headaches in the previous 12 months (i.e., "seeking care"). This included in‐person/telephone/or e‐visit at Primary Care, Specialty Care, or Emergency/Urgent Care locations. Supervised machine learning (Random Forest) and Least Absolute Shrinkage and Selection Operator (LASSO) algorithms identified 13/54 sociodemographic and clinical factors most associated with seeking medical care for migraine. Random Forest models complex relationships (including interactions) between predictor variables and a response. LASSO is also an efficient feature selection algorithm. Linear models were used to determine the multivariable association of those factors with seeking care. Results: Among 61,826 persons with migraine, the mean age was 41.7 years (±14.8) and 31,529/61,826 (51.0%) sought medical care for migraine in the previous 12 months. Of those seeking care for migraine, 23,106/31,529 (73.3%) were female, 21,320/31,529 (67.6%) were White, and 28,030/31,529 (88.9%) had health insurance. Severe interictal burden (assessed via the Migraine Interictal Burden Scale‐4, MIBS‐4) occurred in 52.8% (16,657/31,529) of those seeking care and in 23.1% (6991/30,297) of those not seeking care; similar patterns were observed for severe migraine‐related disability (assessed via the Migraine Disability Assessment Scale, MIDAS) (36.7% [11,561/31,529] vs. 14.6% [4434/30,297]) and severe ictal cutaneous allodynia (assessed via the Allodynia Symptom Checklist, ASC‐12) (21.0% [6614/31,529] vs. 7.4% [2230/30,297]). Severe interictal burden (vs. none, OR 2.64, 95% CI [2.5, 2.8]); severe migraine‐related disability (vs. little/none, OR 2.2, 95% CI [2.0, 2.3]); and severe ictal allodynia (vs. none, OR 1.7, 95% CI [1.6, 1.8]) were strongly associated with seeking care for migraine. Conclusions: Seeking medical care for migraine is associated with higher interictal burden, disability, and allodynia. These findings could support interventions to promote care‐seeking among people with migraine, encourage assessment of these factors during consultation, and prioritize these domains in selecting treatments and measuring their benefits. Plain Language Summary: In this study, we looked at factors that related to a person's decision to seek medical care for migraine in a representative sample of over 60,000 adults with migraine in the United States. We found that only about half (51.0%) of the participants reported seeking care for migraine. We also found that people were more likely to seek medical care when migraine got in the way of their lives during and between migraine attacks, and when they had allodynia (pain from stimuli that do not normally cause pain). [ABSTRACT FROM AUTHOR]

Simple Summary: As the treatment modalities for spinal metastatic tumors continue to evolve, the goal of improving the quality of life of patients with spinal metastases becomes more easily attainable. The best patient care is a combination of multiple treatment modalities, surgical and nonsurgical, that maximizes the advantages of each modality and should be individualized according to their unique clinical presentation, pathology, and life expectancy. In this review article, we highlight various new treatment options that have shown promising improvements in patient outcomes. Notably, we discuss the potential clinical applications of AI and NGS in the treatment of spinal metastases. Spinal metastases occur in up to 40% of patients with cancer. Of these cases, 10% become symptomatic. The reported incidence of spinal metastases has increased in recent years due to innovations in imaging modalities and oncological treatments. As the incidence of spinal metastases rises, so does the demand for improved treatments and treatment algorithms, which now emphasize greater multidisciplinary collaboration and are increasingly customized per patient. Uniquely, we discuss the potential clinical applications of AI and NGS in the treatment of spinal metastases. Material and Methods: A PubMed search for articles published from 2000 to 2023 regarding spinal metastases and artificial intelligence in healthcare was completed. After screening for relevance, the key findings from each study were summarized in this update. Results: This review summarizes the evidence from studies reporting on treatment modalities for spinal metastases, including minimally invasive surgery (MIS), external beam radiation therapy (EBRT), stereotactic radiosurgery (SRS), CFR-PEEK instrumentation, radiofrequency ablation (RFA), next-generation sequencing (NGS), artificial intelligence, and predictive models. [ABSTRACT FROM AUTHOR]

Fertility procedures recorded in health-care databases can be used to estimate the start of pregnancy, which can serve as a reference standard to validate gestational age estimates based on International Classification of Diseases codes. In a cohort of 17 398 US MarketScan pregnancies (2011-2020) in which conception was achieved via fertility procedures, we estimated gestational age at the end of pregnancy using algorithms based on (1) time (days) since the fertility procedure (the reference standard); (2) International Classification of Diseases, Ninth Revision (ICD-9)/ International Classification of Diseases, Tenth Revision (ICD-10) (before/after October 2015) codes indicating gestational length recorded at the end of pregnancy (method A); and (3) ICD-10 end-of-pregnancy codes enhanced with Z3A codes denoting specific gestation weeks recorded at prenatal visits (method B). We calculated the proportion of pregnancies with an estimated gestational age falling within 14 days (⁠|$\pm$| 14 days) of the reference standard. Method A accuracy was similar for ICD-9 and ICD-10 codes. After 2015, method B was more accurate than method A: For term births, within–14-day agreement was 90.8% for method A and 98.7% for method B. Corresponding estimates were 70.1% and 95.6% for preterm births; 35.3% and 92.6% for stillbirths; 54.3% and 64.2% for spontaneous abortions; and 16.7% and 84.6% for elective terminations. ICD-10–based algorithms that incorporate Z3A codes improve the accuracy of gestational age estimation in health-care databases, especially for preterm births and non–live births. [ABSTRACT FROM AUTHOR]

Background: The Bayesian group sequential design has been applied widely in clinical studies, especially in Phase II and III studies. It allows early termination based on accumulating interim data. However, to date, there lacks development in its application to stepped-wedge cluster randomized trials, which are gaining popularity in pragmatic trials conducted by clinical and health care delivery researchers. Methods: We propose a Bayesian adaptive design approach for stepped-wedge cluster randomized trials, which makes adaptive decisions based on the predictive probability of declaring the intervention effective at the end of study given interim data. The Bayesian models and the algorithms for posterior inference and trial conduct are presented. Results: We present how to determine design parameters through extensive simulations to achieve desired operational characteristics. We further evaluate how various design factors, such as the number of steps, cluster size, random variability in cluster size, and correlation structures, impact trial properties, including power, type I error, and the probability of early stopping. An application example is presented. Conclusion: This study presents the incorporation of Bayesian adaptive strategies into stepped-wedge cluster randomized trials design. The proposed approach provides the flexibility to stop the trial early if substantial evidence of efficacy or futility is observed, improving the flexibility and efficiency of stepped-wedge cluster randomized trials. [ABSTRACT FROM AUTHOR]

Artificial intelligence (AI) based predictive models for early detection of cardiovascular disease (CVD) risk are increasingly being utilised. However, AI based risk prediction models that account for right-censored data have been overlooked. This systematic review (PROSPERO protocol CRD42023492655) includes 33 studies that utilised machine learning (ML) and deep learning (DL) models for survival outcome in CVD prediction. We provided details on the employed ML and DL models, eXplainable AI (XAI) techniques, and type of included variables, with a focus on social determinants of health (SDoH) and gender-stratification. Approximately half of the studies were published in 2023 with the majority from the United States. Random Survival Forest (RSF), Survival Gradient Boosting models, and Penalised Cox models were the most frequently employed ML models. DeepSurv was the most frequently employed DL model. DL models were better at predicting CVD outcomes than ML models. Permutation-based feature importance and Shapley values were the most utilised XAI methods for explaining AI models. Moreover, only one in five studies performed gender-stratification analysis and very few incorporate the wide range of SDoH factors in their prediction model. In conclusion, the evidence indicates that RSF and DeepSurv models are currently the optimal models for predicting CVD outcomes. This study also highlights the better predictive ability of DL survival models, compared to ML models. Future research should ensure the appropriate interpretation of AI models, accounting for SDoH, and gender stratification, as gender plays a significant role in CVD occurrence. [ABSTRACT FROM AUTHOR]

Simple Summary: Breast cancer is a significant cause of female cancer-related deaths in the US. Checking how severe the cancer is helps in planning treatment. Modern AI methods are good at grading cancer, but they are not used much in hospitals yet. We developed and utilized ensemble deep learning algorithms for addressing the tasks of classifying (1) breast cancer subtype and (2) breast cancer invasiveness from whole slide image (WSI) histopathology slides. The ensemble models used were based on convolutional neural networks (CNNs) known for extracting distinctive features crucial for accurate classification. In this paper, we provide a comprehensive analysis of these models and the used methodology for breast cancer diagnosis tasks. Cancer diagnosis and classification are pivotal for effective patient management and treatment planning. In this study, a comprehensive approach is presented utilizing ensemble deep learning techniques to analyze breast cancer histopathology images. Our datasets were based on two widely employed datasets from different centers for two different tasks: BACH and BreakHis. Within the BACH dataset, a proposed ensemble strategy was employed, incorporating VGG16 and ResNet50 architectures to achieve precise classification of breast cancer histopathology images. Introducing a novel image patching technique to preprocess a high-resolution image facilitated a focused analysis of localized regions of interest. The annotated BACH dataset encompassed 400 WSIs across four distinct classes: Normal, Benign, In Situ Carcinoma, and Invasive Carcinoma. In addition, the proposed ensemble was used on the BreakHis dataset, utilizing VGG16, ResNet34, and ResNet50 models to classify microscopic images into eight distinct categories (four benign and four malignant). For both datasets, a five-fold cross-validation approach was employed for rigorous training and testing. Preliminary experimental results indicated a patch classification accuracy of 95.31% (for the BACH dataset) and WSI image classification accuracy of 98.43% (BreakHis). This research significantly contributes to ongoing endeavors in harnessing artificial intelligence to advance breast cancer diagnosis, potentially fostering improved patient outcomes and alleviating healthcare burdens. [ABSTRACT FROM AUTHOR]

Background: Balance disorders can give rise to sensations of instability, lightheadedness, vertigo, disequilibrium, or syncope, ultimately leading to grave medical, physical, emotional, and societal ramifications. These conditions are highly prevalent among individuals aged 40 and above. Screen time encompasses activities associated with television viewing, video game playing, and non-work-related computer usage. Prolonged screen exposure may engender a spectrum of health issues and even elevate overall mortality rates. However, the available evidence on the potential link between excessive screen time and balance dysfunction remains limited. Aims: The primary aim of this study was to explore the possible association between prolonged screen exposure and impaired balance function. Methods: This cross-sectional study utilized data from participants who completed a comprehensive questionnaire in the NHANES database between 1999 and 2002, all of whom were aged over 40 and under 85 years. Participants' screen time was categorized into two groups (< 4 h/d and ≥4 h/d) for subsequent data analysis. Logistic regression, combined with propensity score matching (PSM), was employed to investigate the correlation between screen time and balance disorders. Results: A total of 5176 participants were enrolled in this study, comprising 2,586 men and 2,590 women, with a prevalence rate of balance disorders at 25.7% (1331/5176). The incidence of balance disorders was found to be significantly higher among individuals who spent 4 hours or more per day on screen time compared to those with less screen time (P<0.001). Multivariate logistic analysis conducted on the unmatched cohort revealed a significant association between screen time and balance disorders, with an odds ratio (OR) 1.8 (95%CI 1.57 ∼ 2.05). These findings remained consistent even after adjusting for confounding factors, yielding an OR 1.43 (95%CI 1.24 ∼ 1.66). Moreover, the association persisted when employing various multivariate analyses such as propensity score matching adjusted model, standardized mortality ratio weighting model and pairwise algorithmic model; all resulting in ORs ranging from 1.38 to 1.43 and p-values < 0.001. Conclusions: After controlling for all covariates, screen time (watching TV, playing video games, and using computers outside of work) was associated with balance dysfunction among middle-aged and older adults. This finding may offer a possible idea for the prevention of dizziness and balance disorders. Nevertheless, additional research is imperative to further validate these results. [ABSTRACT FROM AUTHOR]

Artificial intelligence models represented in machine learning algorithms are promising tools for risk assessment used to guide clinical and other health care decisions. Machine learning algorithms, however, may house biases that propagate stereotypes, inequities, and discrimination that contribute to socioeconomic health care disparities. The biases include those related to some sociodemographic characteristics such as race, ethnicity, gender, age, insurance, and socioeconomic status from the use of erroneous electronic health record data. Additionally, there is concern that training data and algorithmic biases in large language models pose potential drawbacks. These biases affect the lives and livelihoods of a significant percentage of the population in the United States and globally. The social and economic consequences of the associated backlash cannot be underestimated. Here, we outline some of the sociodemographic, training data, and algorithmic biases that undermine sound health care risk assessment and medical decision-making that should be addressed in the health care system. We present a perspective and overview of these biases by gender, race, ethnicity, age, historically marginalized communities, algorithmic bias, biased evaluations, implicit bias, selection/sampling bias, socioeconomic status biases, biased data distributions, cultural biases and insurance status bias, conformation bias, information bias and anchoring biases and make recommendations to improve large language model training data, including de-biasing techniques such as counterfactual role-reversed sentences during knowledge distillation, fine-tuning, prefix attachment at training time, the use of toxicity classifiers, retrieval augmented generation and algorithmic modification to mitigate the biases moving forward. [ABSTRACT FROM AUTHOR]

Background Respiratory syncytial virus (RSV) is a major cause of morbidity and mortality among US infants. A child's calendar birth month determines their age at first exposure(s) to RSV. We estimated birth month-specific risk of medically attended (MA) RSV lower respiratory tract infection (LRTI) among infants during their first RSV season and first year of life (FYOL). Methods We analyzed infants born in the USA between July 2016 and February 2020 using three insurance claims databases (two commercial, one Medicaid). We classified infants' first MA RSV LRTI episode by the highest level of care incurred (outpatient, emergency department, or inpatient), employing specific and sensitive diagnostic coding algorithms to define index RSV diagnoses. In our main analysis, we focused on infants' first RSV season. In our secondary analysis, we compared the risk of MA RSV LRTI during infants' first RSV season to that of their FYOL. Results Infants born from May through September generally had the highest risk of first-season MA RSV LRTI—approximately 6–10% under the specific RSV index diagnosis definition and 16–26% under the sensitive. Infants born between October and December had the highest risk of RSV-related hospitalization during their first season. The proportion of MA RSV LRTI events classified as inpatient ranged from 9% to 54% (specific) and 5% to 33% (sensitive) across birth month and comorbidity group. Through the FYOL, the overall risk of MA RSV LRTI is comparable across birth months within each claims database (6–11% under the specific definition, 17–30% under the sensitive), with additional cases progressing to care at outpatient or ED settings. Conclusions Our data support recent national recommendations for the use of nirsevimab in the USA. For infants born at the tail end of an RSV season who do not receive nirsevimab, a dose administered prior to the onset of their second RSV season could reduce the incidence of outpatient- and ED-related events. [ABSTRACT FROM AUTHOR]

6–11-year-old children provide a critical window for physical activity (PA) interventions. The Virtual Fitness Buddy ecosystem is a precision health PA intervention for children integrating mixed reality technology to connect people and devices. A cluster randomized, controlled trial was conducted across 19 afterschool sites over two 6-month cohorts to test its efficacy in increasing PA and decreasing sedentary behavior. In the treatment group, a custom virtual dog via a mixed reality kiosk helped children set PA goals while sharing progress with parents to receive feedback and support. Children in the control group set PA goals using a computer without support from the virtual dog or parents. 303 children had 8+ hours of PA data on at least one day of each of the 3 intervention time intervals. Conversion of sedentary time was primarily to light-intensity PA and was strongest for children with low baseline moderate-to-vigorous PA than children above 45 min of baseline moderate-to-vigorous PA. Findings suggest that the VFB ecosystem can promote sustainable PA in children and may be rapidly diffused for widespread public health impact. [ABSTRACT FROM AUTHOR]

Background: Technological burden and medical complexity are significant drivers of clinician burnout. Electronic health record(EHR)-based population health management tools can be used to identify high-risk patient populations and implement prophylactic health practices. Their impact on clinician burnout, however, is not well understood. Our objective was to assess the relationship between ratings of EHR-based population health management tools and clinician burnout. Methods: We conducted cross-sectional analyses of 2018 national Veterans Health Administration(VA) primary care personnel survey, administered as an online survey to all VA primary care personnel (n = 4257, response rate = 17.7%), using bivariate and multivariate logistic regressions. Our analytical sample included providers (medical doctors, nurse practitioners, physicians' assistants) and nurses (registered nurses, licensed practical nurses). The outcomes included two items measuring high burnout. Primary predictors included importance ratings of 10 population health management tools (eg. VA risk prediction algorithm, recent hospitalizations and emergency department visits, etc.). Results: High ratings of 9 tools were associated with lower odds of high burnout, independent of covariates including VA tenure, team role, gender, ethnicity, staffing, and training. For example, clinicians who rated the risk prediction algorithm as important were less likely to report high burnout levels than those who did not use or did not know about the tool (OR 0.73; CI 0.61-0.87), and they were less likely to report frequent burnout (once per week or more) (OR 0.71; CI 0.60-0.84). Conclusions: Burned-out clinicians may not consider the EHR-based tools important and may not be using them to perform care management. Tools that create additional technological burden may need adaptation to become more accessible, more intuitive, and less burdensome to use. Finding ways to improve the use of tools that streamline the work of population health management and/or result in less workload due to patients with poorly managed chronic conditions may alleviate burnout. More research is needed to understand the causal directional of the association between burnout and ratings of population health management tools. [ABSTRACT FROM AUTHOR]

Positive airway pressure (PAP) therapy is first-line therapy for OSA, but consistent use is required for it to be effective. Previous studies have used Medicare fee-for-service claims data (eg, device, equipment charges) as a proxy for PAP adherence to assess its effects. However, this approach has not been validated in a US commercially insured population, where coverage rules are not standardized. In a commercially insured population in the United States, how well do claims-based algorithms for defining PAP adherence correspond with objective PAP device usage? Deidentified administrative claims data of commercially insured patients (aged 18-64 years) with OSA were linked to objective PAP therapy usage data from cloud-connected devices. Adherence was defined based on device use (using an extension of Centers for Medicare & Medicaid Services 90-day compliance criteria) and from claims-based algorithms to compare usage metrics and identify potential misclassifications. The final sample included 213,341 patients. Based on device usage, 48% were adherent in the first year. Based on claims, between 10% and 84% of patients were identified as adherent (accuracy, sensitivity, and specificity ranges: 53%-68%, 12%-95%, and 26%-92%, respectively). Relative to patients who were claims-adherent, patients who were device-adherent had consistently higher usage across all metrics (mean, 339.9 vs 260.0-290.0 days of use; 6.6 vs 5.1-5.6 d/wk; 6.4 vs 4.6-5.2 h/d). Consistent PAP users were frequently identified by claims-based algorithms as nonadherent, whereas many inconsistent users were classified by claims-based algorithms as adherent. In aggregate US commercial data with nonstandardized PAP coverage rules, concordance between existing claims-based definitions and objective PAP use was low. Caution is warranted when applying existing claims-based algorithms to commercial populations. [Display omitted] [ABSTRACT FROM AUTHOR]

Precisely and efficiently identifying subgroups with heterogeneous treatment effects (HTEs) in real-world evidence studies remains a challenge. Based on the causal forest (CF) method, we developed an iterative CF (iCF) algorithm to identify HTEs in subgroups defined by important variables. Our method iteratively grows different depths of the CF with important effect modifiers, performs plurality votes to obtain decision trees (subgroup decisions) for a family of CFs with different depths, and then finds the cross-validated subgroup decision that best predicts the treatment effect as a final subgroup decision. We simulated 12 different scenarios and showed that the iCF outperformed other machine learning methods for interaction/subgroup identification in the majority of scenarios assessed. Using a 20% random sample of fee-for-service Medicare beneficiaries initiating sodium-glucose cotransporter-2 inhibitors or glucagon-like peptide-1 receptor agonists, we implemented the iCF to identify subgroups with HTEs for hospitalized heart failure. Consistent with previous studies suggesting patients with heart failure benefit more from sodium-glucose cotransporter-2 inhibitors, iCF successfully identified such a subpopulation with HTEs and additive interactions. The iCF is a promising method for identifying subgroups with HTEs in real-world data where the potential for unmeasured confounding can be limited by study design. [ABSTRACT FROM AUTHOR]

Objective This study retrospectively compared false-positive result frequencies of 3 syphilis serology screening tests and assessed whether false positivity was associated with pregnancy and age. Methods Results for 3 screening tests were retrieved from the laboratory database, including rapid plasma reagin (RPR) assay between October 2016 and September 2019, BioPlex 2200 Syphilis Total immunoassay between May 2020 and January 2022, and Alinity i Syphilis TP assay between February 2022 and April 2023. The false-positive result frequencies were calculated based on testing algorithm criteria. Results False-positive result frequency for BioPlex was 0.61% (90/14,707), significantly higher than 0.29% (50/17,447) for RPR and 0.38% (55/14,631) for Alinity (both P <.01). Patients with false-positive results were significantly older than patients with nonreactive results for RPR (median age: 36 vs 28, P <.001), but not for BioPlex or Alinity. For all 3 tests, the positive predictive values in pregnant women were lower than those in nonpregnant women or men. However, pregnant women did not exhibit a higher false-positive result frequency. Conclusion Although false-positive result frequencies were low overall for all 3 syphilis serology tests, there is a significant difference between different tests. Pregnancy was not associated with more false-positive results for all 3 tests. [ABSTRACT FROM AUTHOR]

Artificial intelligence has demonstrated utility in orthopedic research. Algorithmic models derived from machine learning have demonstrated adaptive learning with predictive application towards outcomes, leading to increased traction in the literature. This study aims to identify machine learning arthroplasty research trends and anticipate emerging key terms. Published literature focused on machine learning in arthroplasty from 1992 to 2023 was selected through the Web of Science Core Collection of Clarivate Analytics. Following that, bibliometric indicators were attained and brought in to perform an additional examination using Bibliometrix and VOSviewer to identify historical and present patterns within the literature. A total of 235 documents were obtained through bibliometric sourcing based on machine learning applications within the arthroplasty literature. Thirty-four countries published articles on the topic, and the United States was demonstrated to be the largest global contributor. Four hundred-five institutions internationally contributed articles, with Harvard Medical School and the University of California system as the most relevant institutes, with 75 and 44 articles produced, respectively. Kwon YM was the most productive author, while Haeberle HS and Ramkumar PN were the most impactful based on h-index. The Thematic map and Co-occurrence visualization helped identify both major and niche themes present in the scientific databases. Machine learning in arthroplasty research continues to gain traction with a growing annual production rate and contributions from international authors and institutions. Institutions and authors based in the United States are the leading contributors to machine learning applications within arthroplasty research. This research discerns trends that have occurred, are presently ongoing, and are emerging within this field, aiming to inform future hotspot development. [ABSTRACT FROM AUTHOR]

Hidradenitis suppurativa (HS) is a chronic inflammatory follicular skin condition that is associated with significant psychosocial and economic burden and a diminished quality of life and work productivity. Accurate diagnosis of HS is challenging due to its unknown etiology, which can lead to underdiagnosis or misdiagnosis that results in increased patient and healthcare system burden. We applied machine learning (ML) to a medical and pharmacy claims database using data from 2000 through 2018 to develop a novel model to better understand HS underdiagnosis on a healthcare system level. The primary results demonstrated that high-performing models for predicting HS diagnosis can be constructed using claims data, with an area under the curve (AUC) of 81%-82% observed among the top-performing models. The results of the models developed in this study could be input into the development of an impact of inaction model that determines the cost implications of HS diagnosis and treatment delay to the healthcare system. [ABSTRACT FROM AUTHOR]

Introduction: The utility of ablation index (AI) to guide ventricular tachycardia (VT) ablation in patients with structural heart disease is unknown. The aim of this study was to assess procedural characteristics and clinical outcomes achieved using AI‐guided strategy (target value 550) or conventional non‐AI‐guided parameters in patients undergoing scar‐related VT ablation. Methods: Consecutive patients (n = 103) undergoing initial VT ablation at a single center from 2017 to 2022 were evaluated. Patient groups were 1:1 propensity‐matched for baseline characteristics. Single lesion characteristics for all 4707 lesions in the matched cohort (n = 74) were analyzed. The impact of ablation characteristics was assessed by linear regression and clinical outcomes were evaluated by Cox proportional hazard model. Results: After propensity‐matching, baseline characteristics were well‐balanced between AI (n = 37) and non‐AI (n = 37) groups. Lesion sets were similar (scar homogenization [41% vs. 27%; p =.34], scar dechanneling [19% vs. 8%; p =.18], core isolation [5% vs. 11%; p =.4], linear and elimination late potentials/local abnormal ventricular activities [35% vs. 44%; p =.48], epicardial mapping/ablation [11% vs. 14%; p =.73]). AI‐guided strategy had 21% lower procedure duration (−47.27 min, 95% confidence interval [CI] [−81.613, −12.928]; p =.008), 49% lower radiofrequency time per lesion (−13.707 s, 95% CI [−17.86, −9.555]; p <.001), 21% lower volume of fluid administered (1664 cc [1127, 2209] vs. 2126 cc [1750, 2593]; p =.005). Total radiofrequency duration (−339 s [−24%], 95%CI [−776, 62]; p =.09) and steam pops (−155.6%, 95% CI [19.8%, −330.9%]; p =.08) were nonsignificantly lower in the AI group. Acute procedural success (95% vs. 89%; p =.7) and VT recurrence (0.97, 95% CI [0.42–2.2]; p =.93) were similar for both groups. Lesion analysis (n = 4707) demonstrated a plateau in the magnitude of impedance drops once reaching an AI of 550–600. Conclusion: In this pilot study, an AI‐guided ablation strategy for scar‐related VT resulted in shorter procedure time and average radiofrequency time per lesion with similar acute procedural and intermediate‐term clinical outcomes to a non‐AI‐guided approach utilizing traditional ablation parameters. [ABSTRACT FROM AUTHOR]

Simple Summary: Endometrial cancer is the most common cancer of the female reproductive system in the United States and the second most common cancer of the female reproductive system worldwide. Treatment typically consists of upfront surgery. Postoperative therapy may include adjuvant radiation therapy and/or systemic therapy based on unique patient and pathologic characteristics. In this review, we explore the ways in which patient and tumor characteristics, current and emerging radiation technologies, and other cancer-directed treatments may be considered in developing personalized radiotherapy regimens for patients with endometrial cancer. Endometrial cancer is the most common gynecologic cancer in the United States and it contributes to the second most gynecologic cancer-related deaths. With upfront surgery, the specific characteristics of both the patient and tumor allow for risk-tailored treatment algorithms including adjuvant radiotherapy and systemic therapy. In this narrative review, we discuss the current radiation treatment paradigm for endometrial cancer with an emphasis on various radiotherapy modalities, techniques, and dosing regimens. We then elaborate on how to tailor radiotherapy treatment courses in combination with other cancer-directed treatments, including chemotherapy and immunotherapy. In conclusion, this review summarizes ongoing research that aims to further individualize radiotherapy regimens for individuals in an attempt to improve patient outcomes. [ABSTRACT FROM AUTHOR]

We estimate the impact of the Affordable Care Act (ACA) on emergency department (ED) visits and the composition of insurance coverage for White, Black, and Hispanic older adults. Our estimation strategy uses changes in the discontinuity of health insurance coverage at age 65 and the variation in state decisions about Medicaid expansion under the ACA. We find that uninsured ED visits decreased for older adults in all three racial and ethnic groups in Medicaid expansion and non-expansion states. The magnitude of the decreases varied from four visits per 1,000 people among White older adults in non-expansion states to 23 visits per 1,000 people among Black and Hispanic older adults in expansion states. The insurance coverage gains came primarily from Medicaid in expansion states and private insurance in non-expansion states, regardless of race or ethnicity. We find suggestive evidence of increased ED visits for Black and Hispanic populations that had low insurance coverage rates before 2014. [ABSTRACT FROM AUTHOR]

We evaluate the performance of a widely used technique for imputing the legal immigration status of U.S. immigrants in survey data—the logical imputation method. We validate this technique by implementing it in a nationally representative survey of U.S. farmworkers that includes a well-regarded measure of legal status. When using this measure as a benchmark, the imputation algorithm correctly identifies the legal status of 78% of farmworkers. Of all the variables included in the algorithm, we find that Medicaid participation poses the greatest challenge for accuracy. Using the American Community Survey, we show that increased Medicaid enrollments stemming from the implementation of the Affordable Care Act in 2014 led to sizable changes in the share of immigrants imputed as legal over time and across space. We explore the implications of these changes for two previous studies and conclude that including Medicaid criteria in the imputation algorithm can significantly impact research findings. We also provide tools to gauge the sensitivity of results. [ABSTRACT FROM AUTHOR]

As algorithmic decision making is increasingly deployed in every walk of life, many researchers have raised concerns about fairness-related bias from such algorithms. But there is little research on harnessing psychometric methods to uncover potential discriminatory bias inside decision-making algorithms. The main goal of this article is to propose a new framework for algorithmic fairness based on differential item functioning (DIF), which has been commonly used to measure item fairness in psychometrics. Our fairness notion, which we call differential algorithmic functioning (DAF), is defined based on three pieces of information: a decision variable, a "fair" variable, and a protected variable such as race or gender. Under the DAF framework, an algorithm can exhibit uniform DAF, nonuniform DAF, or neither (i.e., non-DAF). For detecting DAF, we provide modifications of well-established DIF methods: Mantel–Haenszel test, logistic regression, and residual-based DIF. We demonstrate our framework through a real dataset concerning decision-making algorithms for grade retention in K–12 education in the United States. [ABSTRACT FROM AUTHOR]

Background. Serum uric acid (SUA) and glycosylated hemoglobin (HbA1c) were closely related to the body's metabolism. This study aimed to investigate the relationship between HbA1c and SUA in adults. Methods. This study selected 7293 participants aged ≥20 from 2011 to 2020 in the National Health and Nutrition Examination Survey (NHANES). The multivariate linear regression model was used to test the association between HbA1c and SUA. Subgroup analysis was performed according to age, gender, race, and body mass index (BMI). This study solved the relationship between HbA1c and SUA by fitting a smooth curve. Finally, the inflection point in the nonlinear relationship was calculated by the recursive algorithm, and the relationship between HbA1c and SUA on both sides of the inflection point was analyzed by the two-segment piecewise linear regression model. Results. All 7293 participants found a negative correlation between HbA1c and SUA by completely adjusting the model (β = −7.93 and 95% CI: −9.49–−6.37). In addition, when this study was stratified by gender, age, race, and BMI status, this negative correlation was still statistically significant. In the subgroup analysis, we found that the relationship between the two had different results due to gender differences. In men, HbA1c had a significant negative correlation with SUA. However, in women, the HbA1c value was positively correlated with SUA before 6.8%, and the HbA1c value was negatively correlated with SUA after 6.8%, which indicates that the relationship between HbA1c and SUA in women has changed in prediabetes and diabetes. Conclusion. This study shows that HbA1c is positively correlated with SUA in American adults before 7%. There is a negative correlation after the HbA1c value of 7%. [ABSTRACT FROM AUTHOR]

Major Depressive Disorder (MDD) and Generalized Anxiety Disorder (GAD) pose significant burdens on individuals and society, necessitating accurate prediction methods. Machine learning (ML) algorithms utilizing electronic health records and survey data offer promising tools for forecasting these conditions. However, potential bias and inaccuracies inherent in subjective survey responses can undermine the precision of such predictions. This research investigates the reliability of five prominent ML algorithms—a Convolutional Neural Network (CNN), Random Forest, XGBoost, Logistic Regression, and Naive Bayes—in predicting MDD and GAD. A dataset rich in biomedical, demographic, and self-reported survey information is used to assess the algorithms' performance under different levels of subjective response inaccuracies. These inaccuracies simulate scenarios with potential memory recall bias and subjective interpretations. While all algorithms demonstrate commendable accuracy with high-quality survey data, their performance diverges significantly when encountering erroneous or biased responses. Notably, the CNN exhibits superior resilience in this context, maintaining performance and even achieving enhanced accuracy, Cohen's kappa score, and positive precision for both MDD and GAD. This highlights the CNN's superior ability to handle data unreliability, making it a potentially advantageous choice for predicting mental health conditions based on self-reported data. These findings underscore the critical importance of algorithmic resilience in mental health prediction, particularly when relying on subjective data. They emphasize the need for careful algorithm selection in such contexts, with the CNN emerging as a promising candidate due to its robustness and improved performance under data uncertainties. [ABSTRACT FROM AUTHOR]

Objective The purpose of this study was to apply network analysis methodology to better understand the relationships between pain-related measures among people with chronic pain. Methods We analyzed data from a cross-sectional sample of 4614 active duty service members with chronic pain referred to 1 military interdisciplinary pain management center between 2014 and 2021. Using a combination of Patient-Reported Outcomes Measurement Information System measures and other pain-related measures, we applied the "EBICglasso" algorithm to create regularized partial correlation networks that would identify the most influential measures. Results Pain interference, depression, and anxiety had the highest strength in these networks. Pain catastrophizing played an important role in the association between pain and other pain-related health measures. Bootstrap analyses showed that the networks were very stable and the edge weights accurately estimated in 2 analyses (with and without pain catastrophizing). Conclusions Our findings offer new insights into the relationships between symptoms using network analysis. Important findings highlight the strength of association between pain interference, depression and anxiety, which suggests that if pain is to be treated depression and anxiety must also be addressed. What was of specific importance was the role that pain catastrophizing had in the relationship between pain and other symptoms suggesting that pain catastrophizing is a key symptom on which to focus for treatment of chronic pain. [ABSTRACT FROM AUTHOR]

The American Academy of Pediatrics recommends utilizing hospitalizations as an opportunity to provide sexual health screenings for adolescents. This study aimed to describe the current practice of sexual history documentation (SHxD) and sexually transmitted infection (STI) testing among adolescents admitted to a pediatric hospital medicine service. Retrospective cross-sectional study of adolescents (14-19 years old) admitted to the PHM service from 2017-2019 was performed at an academic children's health system. Patient (demographics, history of complex chronic condition, and insurance), hospitalization (length of stay, diagnosis, STI tests ordered/results), and physician (level of training and gender) characteristics were extracted for each encounter. A natural language processing algorithm identified the presence of SHxD. Univariate analysis and multivariable analysis were performed to detect factors associated with SHxD and STI screening. The prevalence of STIs was calculated for those who were tested. Out of 2242 encounters, SHxD and STI testing rates were 40.9% and 17.2%, respectively. Patient gender, race, lack of complex chronic condition, and resident involvement were predictive of SHxD and STI testing. SHxD increased the odds of STI testing significantly (OR 5.06, CI 3.90-6.58). Among those who were tested, the prevalence of STIs was highest for chlamydia (37/329, 11.2%). Overall, sexual health screening rates remain low in the hospital setting and future improvement initiatives are needed. [ABSTRACT FROM AUTHOR]

Policy PointsPromoting healthy public policies is a national priority, but state policy adoption is driven by a complex set of internal and external factors.This study employs new social network methods to identify underlying connections among states and to predict the likelihood of new firearm‐related policy adoption given changes to this interstate network.This approach could be used to assess the likelihood that a given state will adopt a specific new firearm‐related law and to identify points of influence that could either inhibit or promote wider diffusion of specific laws. Context: US states are largely responsible for the regulation of firearms within their borders. Each state has developed a different legal environment with regard to firearms based on different values and beliefs of citizens, legislators, governors, and other stakeholders. Predicting the types of firearm laws that states may adopt is therefore challenging. Methods: We propose a parsimonious model for this complex process and provide credible predictions of state firearm laws by estimating the likelihood they will be passed in the future. We employ a temporal exponential‐family random graph model to capture the bipartite state law–state network data over time, allowing for complex interdependencies and their temporal evolution. Using data on all state firearm laws over the period 1979–2020, we estimate these models' parameters while controlling for factors associated with firearm law adoption, including internal and external state characteristics. Predictions of future firearm law passage are then calculated based on a number of scenarios to assess the effects of a given type of firearm law being passed in the future by a given state. Findings: Results show that a set of internal state factors are important predictors of firearm law adoption, but the actions of neighboring states may be just as important. Analysis of scenarios provide insights into the mechanics of how adoption of laws by specific states (or groups of states) may perturb the rest of the network structure and alter the likelihood that new laws would become more (or less) likely to continue to diffuse to other states. Conclusions: The methods used here outperform standard approaches for policy diffusion studies and afford predictions that are superior to those of an ensemble of machine learning tools. The proposed framework could have applications for the study of policy diffusion in other domains. [ABSTRACT FROM AUTHOR]

The healthcare sector is faced with challenges due to a shrinking healthcare workforce and a rise in chronic diseases that are worsening with demographic and epidemiological shifts. Digital health interventions that include artificial intelligence (AI) are being identified as some of the potential solutions to these challenges. The ultimate aim of these AI systems is to improve the patient's health outcomes and satisfaction, the overall population's health, and the well-being of healthcare professionals. The applications of AI in healthcare services are vast and are expected to assist, automate, and augment several healthcare services. Like any other emerging innovation, AI in healthcare also comes with its own risks and requires regulatory controls. A review of the literature was undertaken to study the existing regulatory landscape for AI in the healthcare services sector in developed nations. In the global regulatory landscape, most of the regulations for AI revolve around Software as a Medical Device (SaMD) and are regulated under digital health products. However, it is necessary to note that the current regulations may not suffice as AI-based technologies are capable of working autonomously, adapting their algorithms, and improving their performance over time based on the new real-world data that they have encountered. Hence, a global regulatory convergence for AI in healthcare, similar to the voluntary AI code of conduct that is being developed by the US-EU Trade and Technology Council, would be beneficial to all nations, be it developing or developed. [ABSTRACT FROM AUTHOR]

A trend reported by both academics and practitioners is that advertising on TV has become increasingly energetic. This study investigates the association between the energy level in ad content and consumers' ad-tuning tendency. Using a data set of over 27,000 TV commercials delivered to U.S. homes between 2015 and 2018, the authors first present a framework to algorithmically measure the energy level in ad content from the video of ads. This algorithm-based measure is then compared with human-perceived energy levels showing that the measure is related to the level of arousal stimulated by ad content. By relating the energy levels in ad content with the ad-tuning tendency using two empirical procedures, the authors document the following: overall, more energetic commercials are more likely to be tuned in or less likely to be avoided by viewers. The positive association between energy levels in ad content and ad tuning is statistically significant after controlling for placement and other aspects of commercials. However, the association varies across product categories and program genres. The main implication of this study is that advertisers should pay attention to components of ad content other than loudness, which has been regulated by law. [ABSTRACT FROM AUTHOR]

Over the past century, conscientiousness has become seen as the preeminent trait for predicting performance. This consensus is due in part to these employees' ability to work with traditional 20th-century technology. Such pairings balance the systematic nature of conscientious employees with the technology's need for user input and direction to perform tasks--resulting in a complementary match. However, the 21st century has seen the incorporation of intelligent machines (e.g., artificial intelligence, robots, and algorithms) into employees' jobs. Unlike traditional technology, these new machines are equipped with the capability to make decisions autonomously. Thus, their nature overlaps with the orderliness subdimension of conscientious employees--resulting in a non-complementary mismatch. This calls into question whether the consensus about conscientious employees' effectiveness with 20th-century technology applies to 21st-century jobs. Integrating complementarity and role theory, we refine this consensus. Across three studies using distinct samples (an experience sampling study, a field experiment, and an online experiment from working adults in Malaysia, Taiwan, and the United States), each focused on a different type of intelligent machine, we show not only that using intelligent machines has benefits and consequences, but, importantly, that conscientious (i.e., orderly) employees are less likely to benefit from working with them. [ABSTRACT FROM AUTHOR]

The Emergency Severity Index (ESI) is the most popular tool used to triage patients in the US and abroad. Evidence has shown that ESI has its limitations in correctly assigning acuity. To address this, AI can be incorporated into the triage process, decreasing the likelihood of assigning an incorrect ESI level. [ABSTRACT FROM AUTHOR]

Purpose: This study aimed to investigate the association between macronutrient intake and biological age. Methods: Data were collected from 26,381 adults who participated in the United States National Health and Nutrition Examination Survey (NHANES). Two biological ages were estimated using the Klemera-Doubal method (KDM) and PhenoAge algorithms. Biological age acceleration (AA) was computed as the difference between biological age and chronological age. The associations between macronutrient intakes and AA were investigated. Results: After fully adjusting for confounding factors, negative associations were observed between AA and fiber intake (KDM-AA: β – 0.53, 95% CI – 0.62, – 0.43, P < 0.05; PhenoAge acceleration: β – 0.30, 95% CI – 0.35, – 0.25, P < 0.05). High-quality carbohydrate intake was associated with decreased AA (KDM-AA: β – 0.57, 95% CI – 0.67, – 0.47, P < 0.05; PhenoAge acceleration: β – 0.32, 95% CI – 0.37, – 0.26, P < 0.05), while low-quality carbohydrate was associated with increased AA (KDM-AA: β 0.30, 95% CI 0.21, 0.38, P < 0.05; PhenoAge acceleration: β 0.16, 95% CI 0.11, 0.21, P < 0.05). Plant protein was associated with decreased AA (KDM-AA: β – 0.39, 95% CI – 0.51, – 0.27, P < 0.05; PhenoAge acceleration: β – 0.21, 95% CI – 0.26, – 0.15, P < 0.05). Long-chain SFA intake increased AA (KDM-AA: β 0.16, 95% CI 0.08, 0.24, P < 0.05; PhenoAge acceleration: β 0.11, 95% CI 0.07, 0.15, P < 0.05). ω-3 PUFA was associated with decreased KDM-AA (β – 0.18, 95% CI – 0.27, – 0.08, P < 0.05) and PhenoAge acceleration (β – 0.09, 95% CI – 0.13, – 0.04, P < 0.05). Conclusion: Our findings suggest that dietary fiber, high-quality carbohydrate, plant protein, and ω-3 PUFA intake may have a protective effect against AA, while low-quality carbohydrate and long-chain SFA intake may increase AA. Therefore, dietary interventions aimed at modifying macronutrient intakes may be useful in preventing or delaying age-related disease and improving overall health. [ABSTRACT FROM AUTHOR]

An unwelcoming policy climate can create barriers to health care access and produce a 'Chilling Effect' among immigrant communities. For undocumented immigrants, barriers may be unique and have a greater impact. We used administrative emergency department (ED) data from 2015 to 2019 for a Midwestern state provided under a data use agreement with the state hospital association. General linear modelling was used to estimate the impact of anti-immigrant rhetoric on ED visit intensity among non-elderly adults who were likely Hispanic/Latino with undocumented status. Compared to 2015, the average ED visit intensity among adults who were likely Hispanic/Latino with undocumented status was significantly higher during 2016–2019 when anti-immigrant rhetoric was heightened. The magnitude of this change increased over time (0.013, 0.014, 0.021, and 0.020, respectively). Additionally, this change over time was not observed in the comparison groups. Our findings suggest that anti-immigrant rhetoric may alter health care utilization for adults who are likely Hispanic/Latino with undocumented status. Limitations to our findings include the use of only those likely to be Hispanic/Latino, data from only one Midwestern state and the loss of data due to non-classification using the NYU ED algorithm. Further research should focus on validating these findings and investigating these identification methods and anti-immigrant rhetoric effects among other undocumented groups including children and adults of different race or ethnicity such as black, both those that identify as Hispanic/Latino and those that do not. Developing strategies to improve health care access for undocumented Hispanic/Latino adults also warrants future research. [ABSTRACT FROM AUTHOR]

Objective: Despite growing concerns that some digital algorithm-reliant fertility awareness-based methods of pregnancy prevention are marketed in an inaccurate, opaque, and potentially harmful manner online, there has been limited systematic examination of such marketing practices. This article therefore provides an empirical examination of how social media influencers have promoted the fertility tracking tool Daysy on Instagram. We investigate: (1) how the tool is framed in relation to pregnancy prevention using Health Belief Model (HBM) constructs, and (2) the promotional and disclosure practices adopted by influencers. Materials and Methods: We collected Instagram posts mentioning Daysy made between June 2018 and May 2022 using the tool CrowdTangle. Using a qualitative content analysis approach, we coded a random sample of 400 Daysy posts. This yielded 122 Instagram influencer posts promoting Daysy for pregnancy prevention that we coded for promotional content and HBM constructs. Results: Posts originated primarily from Europe (n = 62, 50.82%) and the United States (n = 37, 30.33%). Findings indicate that barriers to use (n = 18, 15.57%) and the severity of risks from unplanned pregnancy (n = 8, 6.56%) were rarely conveyed, whereas benefits of use (n = 122, 100%) and the severity of risks of hormonal contraception (n = 31, 25.41%) were covered more extensively. Only about one third of posts disclosed any formal relationship to the brand Daysy. Conclusions: With many posts emphasizing benefits and obscuring potential limitations, we argue that accurate and transparent information about the effectiveness and limitations of fertility tracking technologies is critical for supporting informed decision-making and, as such, should remain a public health priority. [ABSTRACT FROM AUTHOR]

Algorithms are increasingly used by human resource departments to evaluate employee performance. While the algorithms are perceived to be objective and neutral by removing human biases, they are often perceived to be less fair than human managers. This research proposes dignity as an important construct in explaining the discrepancy in perceived fairness and investigates remedial steps for improving dignity and fairness for algorithm-based employee evaluations. Three experiments' results show that those evaluated by algorithms perceive lower levels of dignity, leading them to believe the process is less fair. In addition, we find that providing justifications for algorithm usage in employee evaluations improves perceived dignity. However, human-algorithm collaboration does not enhance perceived dignity. [ABSTRACT FROM AUTHOR]

Objective Inpatient rehabilitation represents a critical setting for stroke treatment, providing intensive, targeted therapy and task-specific practice to minimize a patient's functional deficits and facilitate their reintegration into the community. However, impairment and recovery vary greatly after stroke, making it difficult to predict a patient's future outcomes or response to treatment. In this study, the authors examined the value of early-stage wearable sensor data to predict 3 functional outcomes (ambulation, independence, and risk of falling) at rehabilitation discharge. Methods Fifty-five individuals undergoing inpatient stroke rehabilitation participated in this study. Supervised machine learning classifiers were retrospectively trained to predict discharge outcomes using data collected at hospital admission, including patient information, functional assessment scores, and inertial sensor data from the lower limbs during gait and/or balance tasks. Model performance was compared across different data combinations and was benchmarked against a traditional model trained without sensor data. Results For patients who were ambulatory at admission, sensor data improved the predictions of ambulation and risk of falling (with weighted F1 scores increasing by 19.6% and 23.4%, respectively) and maintained similar performance for predictions of independence, compared to a benchmark model without sensor data. The best-performing sensor-based models predicted discharge ambulation (community vs household), independence (high vs low), and risk of falling (normal vs high) with accuracies of 84.4%, 68.8%, and 65.9%, respectively. Most misclassifications occurred with admission or discharge scores near the classification boundary. For patients who were nonambulatory at admission, sensor data recorded during simple balance tasks did not offer predictive value over the benchmark models. Conclusion These findings support the continued investigation of wearable sensors as an accessible, easy-to-use tool to predict the functional recovery after stroke. Impact Accurate, early prediction of poststroke rehabilitation outcomes from wearable sensors would improve our ability to deliver personalized, effective care and discharge planning in the inpatient setting and beyond. [ABSTRACT FROM AUTHOR]

Introduction: Pulmonary embolism (PE) is an imperative diagnosis to make given its associated morbidity. There is no current consensus in the initial workup of pregnant patients suspected of a PE. Prospective studies have been conducted in Europe using a pregnancy-adapted YEARS algorithm, which showed safe reductions in computed tomography pulmonary angiography (CTPA) imaging in pregnant patients suspected of PE. Our objective in this study was 1) to measure the potential avoidance of CTPA use in pregnant patients if the pregnancy-adapted YEARS algorithm had been applied and 2) to serve as an external validation study of the use of this algorithm in the United States. Methods: This study was a single-system retrospective chart analysis. Criteria for inclusion in the cohort consisted of keywords: pregnant; older than 18; chief complaints of shortness of breath, chest pain, tachycardia, hemoptysis, deep vein thromboembolism (DVT), and D-dimer--from January 1, 2019--May 31,2022. We then analyzed this cohort retrospectively using the pregnancy-adapted YEARS algorithm, which includes clinical signs of a DVT, hemoptysis, and PE as the most likely diagnosis with a D-dimer assay. Patients within the cohort were then subdivided into two categories: aligned with the YEARS algorithm, or not aligned with the YEARS algorithm. Patients who did not receive a CTPA were analyzed for a subsequent diagnosis of a PE or DVT within 30 days. Results: A total of 74 pregnant patients were included in this study. There was a PE prevalence of 2.7% (two patients). Of the 36 patients who did not require imaging by the algorithm, seven CTPA were performed. Of the patients who did not receive an initial CTPA, zero were diagnosed with PE or DVT within a 30-day follow-up. In total, 85.1% of all the patients in this study were treated in concordance with the pregnancy-adapted YEARS algorithm. Conclusion: The use of the pregnancy-adapted YEARS algorithm could have resulted in decreased utilization of CTPA in the workup of PE in pregnant patients, and the algorithm showed similar reductions compared to prospective studies done in Europe. The pregnancy-adapted YEARS algorithm was also shown to be similar to the clinical rationale used by clinicians in the evaluation of pregnant patients, which indicates its potential for widespread acceptance into clinical practice. [ABSTRACT FROM AUTHOR]

Introduction: Existing literature suggests that transgender women (TW) may be at high risk for adverse mental health due to stress attributed to combined experiences of stigma and complex social and structural vulnerabilities. Little research has examined how these co‐occurring experiences relate to mental health. We aimed to test a theoretically driven conceptual model of relationships between stigma, social and structural vulnerabilities, and mental health to inform future intervention tailoring. Design/Methods: Partial least square path modeling followed by response‐based unit segmentation was used to identify homogenous clusters in a diverse community sample of United States (US)‐based TW (N = 1418; 46.2% White non‐Hispanic). This approach examined associations between latent constructs of stigma (polyvictimization and discrimination), social and structural vulnerabilities (housing and food insecurity, unemployment, sex work, social support, and substance use), and mental health (post‐traumatic stress and psychological distress). Results: The final conceptual model defined the structural relationship between the variables of interest within stigma, vulnerability, and mental health. Six clusters were identified within this structural framework which suggests that racism, ethnicism, and geography may be related to mental health inequities among TW. Conclusion: Our findings around the impact of racism, ethnicism, and geography reflect the existing literature, which unfortunately shows us that little change has occurred in the last decade for TW of color in the Southern US; however, the strength of our evidence (related to sampling structure and sample size) and type of analyses (accounting for co‐occurring predictors of health, i.e., stigma and complex vulnerabilities, reflecting that of real‐world patients) is a novel and necessary addition to the literature. Findings suggest that health interventions designed to offset the negative effects of stigma must include anti‐racist approaches with components to reduce or eliminate barriers to resources that contribute to social and structural vulnerabilities among TW. Herein we provide detailed recommendations to guide primary, secondary, and tertiary prevention efforts. Clinical Relevance: This study demonstrated the importance of considering stigma and complex social and structural vulnerabilities during clinical care and design of mental health interventions for transgender women who are experiencing post‐traumatic stress disorder and psychological distress. Specifically, interventions should take an anti‐racist approach and would benefit from incorporating social support‐building activities. [ABSTRACT FROM AUTHOR]