JAMA

Background: The observational literature on comparative effectiveness is expanding rapidly but remains difficult to synthesize. Discordant findings often stem from structural differences in cohort definitions, inclusion criteria, and follow up windows, leaving stakeholders without a cohesive evidence base. Furthermore, studies typically focus on a narrow subset of outcomes, neglecting the broader needs of diverse healthcare stakeholders 1,2,3,4. Methods We developed a high throughput evidence generation workflow using linked EHR and administrative claims data. The cornerstone is a prespecified measurement architecture applied uniformly across clinical scenarios: six post index windows (acute to two year follow.up); 28 Elixhauser comorbidities; 14 healthcare resource utilization (HCRU) categories; 29 laboratory measures with 52 binary thresholds; and 42 adverse event categories. We generated unadjusted treatment comparisons across ~1,038 outcomes per scenario, including effect-measure modification (EMM) assessments across 130 baseline features. Results Across 40 clinical domains, the workflow produced approximately 32,982,552 outcome evaluations. An evaluation included a treatment comparison outcome population effect estimate with uncertainty bounds and supporting diagnostics. Approximately 5,000 narrative summaries underwent structured clinical and statistical quality control before dissemination. Conclusions Standardized, high throughput workflows can shift evidence generation away from fragmented studies toward comprehensive evidence packages. This shared evidence base supports precision medicine by making treatment effect heterogeneity visible across clinically meaningful subpopulations, reducing the need for redundant, stakeholder-specific studies.

Purpose: We systematically reviewed the evidence for three questions on screening for lung cancer with computed tomography (CT): benefits (from randomized trials) and harms of screening versus no screening/minimal intervention or alternative screening approaches (e.g., selection criteria, screening intervals); relative importance that informed patients place on the potential benefits and harms of screening (patient preferences); and comparative effects from observational studies of different screening selection criteria (using risk prediction models) or nodule classification systems compared with those used in the screening trials. Methods: A working group of primary care and specialist clinicians (previously members of Canadian Task Force on Preventive Health Care) and topic experts provided input into the eligibility criteria and key potential effect moderators, rated outcomes for their importance to decision making, and developed decision thresholds for use when making conclusions and assessing certainty of the evidence. Critical outcomes of screening effects included all-cause mortality, lung-cancer mortality, and overdiagnosis (via excess cancer incidence from screening). For patient preferences, we sought direct preference data via (i) disutilities of relevant health states (measuring their impact on ones health-related quality of life on a scale of 0 [perfect health] to 1 [death], mainly using EQ-5D), and (ii) other preference-based data, such as outcome trade-offs, as well as indirect preference data via (iii) the relative importance of benefits versus harms inferred from attitudes, intentions, and behaviors towards screening among eligible patients informed with estimates of the outcomes. For screening benefits and harms and for patient preferences, we searched three databases (MEDLINE, Embase, and Central & MEDLINE, Scopus, and EconLit, respectively) to July 11, 2025. For screening studies published prior to 2015 we relied on searches for other reviews, and for patient preferences our search was limited to 2012-onwards. For comparative effects, we searched MEDLINE and Embase from 2019 to September 23, 2025, with reliance on other reviews for studies published 2012-2018. Reference lists were scanned and trial registries searched. For the main searches, two independent reviewers screened titles and abstracts and then full texts; for search updates we applied AI to assist with title and abstract screening. Data extraction and analysis were undertaken by single reviewers, with verification; risk of bias and GRADE certainty assessments were undertaken independently by at least two reviewers. Data were pooled where suitable using random effects methods appropriate to the outcome metric and prevalence. Subgroup analyses explored heterogeneity (e.g., sex, number of rounds, type of comparator, sensitivity of nodule management, type of utility measurement). When not pooled (e.g., patient preferences based on screening intentions) data were analyzed by grouping studies based on factors such as population, setting, exposure, and outcomes, with consideration of study size and risk of bias. Conclusions and certainty assessments for screening effects were based on estimates of absolute effects. Results: We included 85 studies (N=640,537; 13 trials) on screening benefits and harms, 59 on patient preferences (33 [N=42,219] on disutilities and 26 [N=10,829] other studies), and 16 for comparing trial (National Lung Screening Trial [NLST]) and LungRADs nodule management, either directly (2 studies, N=26,978) or indirectly (14 studies, N=1,102,285). Screening benefits and harms: Findings from nine trials (N=94,530) examining low-dose CT (LDCT) screening on all-cause (RR 0.97, 95% CI 0.93 to 1.01; 3.7 fewer [8.5 fewer to 1.2 more] per 1000) and lung-cancer mortality (RR 0.87, 95% CI 0.79 to 0.96; 4.0 fewer [1.2 to 6.4 fewer] per 1000) offered low and moderate certainty, respectively, that screening previous/current 20-30 pack-year smokers 50-74 years old 3-4 times will probably result in at least 1 (all-cause) and 2 (cause-specific) fewer deaths per 1000 screened after 10-12 years. The absolute effects may not apply to participants at the lowest baseline risk for lung-cancer incidence (e.g., <1.5% over 6 years) or death. Seven trials (N=35,161) contributed to meta-analysis for overdiagnosis (RR 1.19, 95% CI 1.03 to 1.37; 8.4 [1.3 to 16.3] per 1000), and our certainty was moderate that LDCT screening 3-4 times will probably result in at least 2.5 cases of overdiagnosis per 1000 screened over 10 years. For important outcomes, we had high certainty that screening 3-4 times results in at least 75 people per 1000 screened (and probably at least 225) having at least one benign biopsy/false positive, 150 having one or more incidental findings (likely at least 450), and 50 (probably at least 100) having a clinically significant/actionable incidental finding, but probably does not have an important impact on major complications or death from invasive testing among those without cancer. Though undergoing a LDCT scan probably causes little-to-no psychosocial harm, having a positive screening result likely causes at least a small degree of harm (i.e., 4-8% change from baseline), especially for the 10-15% having to undergo invasive procedures where some may experience moderate harm. Among those without cancer, these effects may last for several months while the diagnostic process is underway, though moderate certainty evidence found little-to-no effects remaining after 6 months from diagnostic resolution. Comparative effects: Findings from applying different baseline predicted risks for lung-cancer incidence or mortality to the trial populations (i.e., alternative selection criteria) were considered with the effects from screening benefits and harms. Using LungRADs instead of NLST nodule management (among NLST eligible people) probably reduces the false positive rate substantially (about half), though the number of false positives still exceeded the decision threshold of 75 per 1000 and the effects for benefits or other harm outcomes are not known. Patient preferences: Findings showed little-to-no disutility (i.e., <0.04) from a positive screening test (moderate certainty) or a false positive result (low certainty). Low-certainty evidence found there may be little-to-no disutility from a stage I-IIIA cancer diagnosis (before treatment) but small but important disutilities from a stage IIIB/IV diagnosis, during first-line treatment of any stage (though possibly moderate disutility of about 0.09 for stage IIIB/IV), and after treatment for stage IIIB/IV but not stage I-IIIA (without progression) where effects were inconsistent but indicated that any disutility may not be long-lasting. Findings for stage I-IIIA are likely most relevant for understanding the consequences of overdiagnosis. For stated preferences between outcomes, there was low certainty evidence that a small majority (51-75%) of people may accept 69-122 false positives and at least 1.3 cases of overdiagnosis per prevented lung-cancer death, and think that the reduction in lung-cancer mortality is more important than experiencing one of the relevant harms. After being informed about anticipated benefits and harms from screening (with the largest screening effects shown to those at higher baseline risk), progressively more people preferred screening (mainly via intentions) as the net benefit of screening improved from low [25-50% preferred] to moderate [51-75%] to high [>75%]. Conclusions: This review provides contemporary data on the benefits and harms of LDCT screening after at least a decade of follow-up and makes conclusions based on absolute effects while considering thresholds for decision making. Across the reviews, findings indicate that screening those aged 50-74 years with 3-4 rounds of LDCT will lead to benefits and harms for which a majority, but not all, eligible people probably find acceptable and worthwhile. While current nodule management using LungRADs likely reduces false positives, whether it impacts the benefits of screening is less certain and worth further research. Further, comparative prospective studies are lacking to determine the effects from screening for those not meeting the minimum age (50 years) and smoking history criteria in the trials, despite having an equivalent risk for lung cancer.

BackgroundMobile health programs (MHPs) provide essential preventive services to uninsured and underserved communities. Following the 2024 regulatory approval of human papillomavirus (HPV) self-collection for cervical cancer screening, MHPs represent an access point for healthcare-based self-collection. However, little is known about patient perceptions of this approach in MHP and other healthcare settings. MethodsFrom May - August 2025, we surveyed individuals aged 25-65 years with a cervix who attended MHPs in Southern Arizona. The survey assessed interest in HPV self-collection, preferred locations, instructional preferences, and facilitators to attend follow-up after a positive result. Descriptive statistics summarized demographic characteristics and survey responses. ResultsFifteen female participants completed the survey (mean age 36 years). Ten (67%) identified as Hispanic or Latino, nine (60%) preferred Spanish, and 14 (93%) were uninsured. Interest in HPV self-collection was high, with ten (67%) very or extremely interested. Among those interested, nine (69%) preferred home-based self-collection, and four (31%) preferred clinic or MHP-based self-collection. Most common concerns regarding self-collection on the MHP were ensuring privacy (n=7; 47%) and knowing how to perform the test correctly (n=5; 33%). Most participants (n=11; 73%) reported being very or extremely confident they would attend follow-up after a positive result; language-concordant support, reminder calls, and scheduling assistance were the most endorsed facilitators. ConclusionHPV self-collection was highly acceptable among MHP attendees, although home-based self-collection was most preferred. Addressing privacy concerns, providing multiple modes of instruction, and offering navigation support may improve implementation success and ensure timely follow-up care in MHP settings.

Objectives Tuberculosis (TB) in the United States disproportionately affects non-U.S.-born individuals. While testing this population for TB infection is recommended, little is known about individuals' willingness to take treatment for latent TB infection (LTBI). To address this gap, we conducted a pilot preference survey among individuals from countries with high TB incidence. Design Cross-sectional survey supported by language concordant community health workers. Setting Federally qualified health center, serving a primarily Asian immigrant community, in San Francisco. Participants Adults eligible for risk-based LTBI testing based on place of birth seeking primary care. Outcome measures Perspectives on TB disease, risk of reinfection, and willingness to accept treatment if diagnosed with LTBI conditional on different factors, such as side effects, costs, and other treatment burden. Results Among 60 participants, the median age was 48 years (interquartile range 35-63 years), 52% were women, and 100% spoke Chinese. Infecting others (n=35, 58%), risk of death (n=30, 50%), and potential isolation (n=25, 42%) were the most worrisome consequences of TB disease. Reinfection risk, risk of liver damage, cost, TB progression risk, clinic visits, and blood draws were most often considered moderately or very important when deciding whether to take LTBI treatment (n=53 to 57, 88-95%). While most participants (n=56, 93%) were willing to take treatment if diagnosed with LTBI even at a 10-year TB progression risk below 1% and willing to accept a risk of liver damage (n=41, 68%), less than half would accept LTBI treatment if there were any associated cost (n=28, 47%). Finally, many participants had concerns about their reinfection risk after completing LTBI treatment (n=34, 57%). Conclusions Amongst surveyed participants, TB disease and its consequences such as hospitalization, death and infecting others were worrisome, and participants had a high level of willingness to take treatment if diagnosed with LTBI. Future assessments of how people weigh tradeoffs regarding LTBI diagnosis and treatment could inform interventions to increase LTBI treatment acceptance and completion.

Background: Previous recommendations on screening for prostate cancer relied on ongoing trials of screening with prostate-specific antigen (PSA), which may have lacked sufficient follow-up duration to fully examine effects on mortality and overdiagnosis. Findings which consider absolute effects by age and screening intensity, along with newer guidance for assessing evidence certainty, may lead to different interpretations. Adding magnetic resonance imaging (MRI) to PSA-based screening has been raised as a way to reduce false positives (FPs) and overdiagnosis. Methods: We systematically searched MEDLINE, Embase, and Central from 2014 to January 28, 2026, for randomized controlled trials (RCTs) and prospective observational studies of: (i) screening versus no screening and (ii) sequential screening with MRI for those with a positive PSA test versus PSA alone among men not known to be at high risk for prostate cancer. Studies on screening with PSA or digital rectal examination (DRE) published pre-2014 were identified from existing systematic reviews and reference lists. Studies on FPs and complications from biopsies after PSA screening did not require a control group. Paired reviewers screened titles/abstracts (assisted with artificial intelligence) and full texts, assessed risk of bias, and extracted data, by age when available. We pooled data when suitable using random-effects models, investigated heterogeneity, and assessed the certainty of evidence using GRADE with conclusions of effects based on decision thresholds based on absolute effect sizes. Results: Across both questions, we included 15 RCTs (N=856,000; 8 sites of ERSPC considered separate trials) and 8 observational studies (N=56,122). At 20 years, among 1000 men who underwent repeated PSA-based screening every 2-4 years starting from age 55-69 (mean 62), there is likely a reduction in prostate-cancer mortality ([&ge;]2 fewer) and metastatic cancer incidence ([&ge;]6 fewer), at the expense of prostate-cancer overdiagnosis ([&ge;]24 cases) and FPs ([&ge;]150 cases) (all moderate certainty). If screening starts at age 50-54 or age 55, the benefits are probably smaller (e.g., 1 vs. 2 fewer prostate-cancer related deaths) with similar harms. Adding DRE or screening with PSA annually does not add benefit. One round of PSA screening or starting screening later at age 70-74 may not offer any important benefit or harm (low to moderate certainty), and any benefit from screening primarily with DRE was not shown. Compared with PSA alone, sequential screening with PSA followed by MRI reduces FPs ([&ge;]33 fewer) and overdiagnosis (via [&ge;]10 fewer diagnoses of clinically insignificant [e.g., Gleason 6] cancers without impacting detection of clinically significant cancers) (moderate to high certainty), though findings were limited to one round of screening without long-term follow-up or measurement of mortality. Interpretation: This review provides clinicians and other interest holders with anticipated absolute effects by age, and assessments of certainty across critical and important outcomes and with approximately two decades of follow-up. Findings apply to a general population and may differ for specific groups. Results for most critical outcomes, both benefits and harms, exceeded thresholds for clinically important effect sizes, thereby demonstrating the complexity of guideline developers' and patients' decision-making regarding screening trade-offs. Findings about adding MRI for those with a positive PSA test were limited and would require additional consideration of costs, infrastructure, expertise, and equity. Protocol registration: PROSPERO - CRD420250651056.

Introduction: Preeclampsia (PE) is a leading cause of maternal and neonatal morbidity and mortality, and low-dose aspirin (LDA) prophylaxis is the cornerstone of evidence-based prevention. Despite guideline recommendations, LDA adherence remains poor, with 10-25% of moderate-risk patients taking aspirin. Objective personalized risk stratification using biomarkers has been shown to motivate behavior change in other disease contexts. Survey data suggest that patients are more motivated to take aspirin if informed by an objective predictive test. Here, we report real-world LDA adherence among patients who received a high-risk result from a cell-free RNA (cfRNA) PE risk prediction test. Methods: This retrospective, observational survey study included asymptomatic patients of advanced maternal age (AMA; [&ge;] 35 years at delivery) with singleton pregnancies without USPSTF-defined preexisting high-risk conditions for PE who received the cfRNA PE risk prediction test. Patients who opted in to receive text message surveys were asked about LDA use following receipt of test results. High adherence was defined as reporting LDA use on at least 6 of 7 days per week at least 85% of the time surveyed. The primary analysis included patients with a high-risk test result and at least one LDA frequency survey response following receipt of test result. The observed proportion of adherent patients was compared to a baseline estimate of 25% using an exact binomial test. Results: Of 166 patients who received a cfRNA PE risk prediction test result, 48 (28.9%) received a high-risk result. Of these, 29 (60%) opted in and responded to at least one survey, constituting the primary analysis population. Twenty-seven of the 29 (93.1%; 95% CI: 78.0-98.1%) were classified as highly adherent, significantly higher than the 25% baseline adherence estimate for moderate-risk patients (p < 0.0001). Conclusion: Among surveyed patients who received a high-risk cfRNA PE test result, the proportion classified as highly adherent to LDA (93%) substantially exceeded published estimates of adherence in a similar patient population and met the clinically meaningful threshold of [&ge;] 80% associated with reduced risk of preterm preeclampsia. These findings indicate that objective and personalized biomarker risk testing may be a powerful driver of behavior change that current guidelines have failed to produce.

Neurofibromatosis type 1 (NF1) is a common autosomal dominant genetic tumor predisposition syndrome.1 NF1 patients display remarkable phenotypic variability, even within families carrying the same NF1 mutation.2 With few exceptions, the identification of specific genotype-phenotype correlations has remained elusive.3-6 We utilized RNA-seq data and direct DNA sequencing to determine HLA genotypes for individuals with NF1-associated high-grade glioma (HGG, n=25), low-grade glioma (LGG, n=79), and malignant peripheral nerve sheath tumors (MPNST, n=105). Odds ratios (OR), binomial p-values and false discovery values were calculated by comparing observed carrier frequencies against expected frequencies derived from ethnicity-matched population data. We find that specific HLA class I and II alleles are associated with different NF1 tumor types. For example, HLA-B*40:02 is significantly associated with NF1-MPNST (OR=3.71, p=0.001, Q=0.02), increasing the lifetime risks for MPNST from 10% to about 29%. The relative cancer risk for an individual in the general population carrying a risk allele can be high, however, that individuals absolute risk for cancer typically remains very low. In contrast, individuals that carry a risk allele and are also burdened with a tumor predisposition syndrome will have a substantially higher absolute risk to develop a tumor, simply because they start at a higher baseline susceptibility for tumors. The identification of HLA-risk alleles for NF1 tumor development is therefore important, as it will allow for a risk-adapted screening or more aggressive treatment of individuals with a specific HLA haplotype. If confirmed, this study will thus improve clinical care and potential outcomes of individuals with NF1.

Congenital heart defects (CHD) are a common congenital anomaly and a leading cause of neonatal mortality. Even in ostensibly isolated cases, genetic testing can reveal monogenic causes of isolated CHD or identify syndromic conditions before additional features become clinically apparent. A timely and accurate genetic diagnosis can inform medical management and surveillance, reduce the need for unnecessary investigations, and offer families valuable information about prognosis, recurrence risk, and anticipatory guidance. In September of 2023, Primary Childrens Hospital introduced a universal genetic testing protocol that implemented whole genome sequencing for all neonates admitted to the cardiac intensive care unit (CICU) undergoing cardiac surgery before 30 days of life, with the goal of increasing the number of patients who receive a timely genetic diagnosis and improving clinical care. This is a retrospective chart review of patients who underwent whole genome sequencing (WGS) under the new universal genetic testing protocol at Primary Childrens Hospital from its initiation in September 2023 to February 2026. Over the study period, 217 neonates with CHD participated in the universal WGS protocol. Of these patients, 23 (10.6%) received a genetic diagnosis that was causative of their CHD, of which 11 patients (48%) had no major extracardiac features at the time testing was ordered. Twenty patients were diagnosed with a syndromic condition, and three patients were diagnosed with a non-syndromic condition. All of these patients received additional referrals to specialists following their new diagnosis, and six families used results to inform decisions regarding continuation of care. An additional 19 patients (8.8%) received WGS results that were clinically relevant but non-diagnostic for their CHD, including partial diagnoses, secondary findings, and carrier status. In total, 19.4% of patients (n=42) had clinically relevant variants identified on their WGS.

BackgroundA vision of lifelong genomic medicine, in which stored genomic data can inform a lifetime of care has long animated the field of genomic medicine. Component pieces of this vision are being researched or are already in clinical practice, including dozens of projects around the world sequencing healthy newborns, along with reanalysis of stored genomic data. Whether lifelong genomic medicine is desirable, and, if so, whether it is feasible, has not been explored in the literature. Methods and FindingsWe conducted and thematically analyzed interviews with over 50 US-based healthcare professionals, including clinical geneticists, genetic counselors, primary care clinicians, laboratory personnel, and those who have implemented genomic screening in health systems. We found broad endorsement of the value of lifelong genomic medicine across groups. Perceived clinical value stemmed from the existence of genomic information relevant at multiple stages of life, the ability to query the genome if an individuals medical circumstances change, and the ability to inform patients about relevant evolving scientific advances. Participants also articulated an efficiency argument for reanalyzing stored genomic data rather than retesting. The clinical value was contested by a few participants, who argued for more targeted testing for the clinical situation and disputed the efficiency argument. Many participants viewed the model as inevitable, with operational precedent already established for many component activities. The feasibility of lifelong genomic medicine was limited not by scientific barriers but by governance gaps spanning delivery models, consent, data stewardship, recontact, and the pediatric-to-adult transition. These gaps have equity implications that are cumulative and mutually reinforcing. ConclusionsThe concept of lifelong genomic medicine was widely viewed as acceptable and desired. However, until the governance infrastructure is established, including accountability, funding, data stewardship, and recontact mechanisms, population-scale genomic sequencing risks proceeding faster than the frameworks needed to make it responsible.

ObjectiveWe report secondary histologic and high-risk HPV (hrHPV) outcomes from the Acceptability and Feasibility of Combination Treatment for Cervical Precancer Among South African Women Living with HIV (ACT 2) Trial. MethodsWe conducted a double-blind Phase 2b feasibility trial of loop electrosurgical excision procedure (LEEP) combined with adjuvant intravaginal 5-fluorouracil (5FU) cream. Women living with HIV (WLWH) and cervical intraepithelial neoplasia (CIN) 2/3 underwent LEEP and were randomly assigned (1:1) to receive 8 doses of 5FU or placebo cream. Our secondary outcomes were (a) regression of cervical disease and (b) clearance of hrHPV. ResultsFrom March 2023 to January 2025, 180 participants underwent LEEP and were randomized to 5FU or placebo cream. Median age was 41 years (IOR: 35-45), 29% had HPV16, 18% had HPV18/45; 99% of women were virologically suppressed (<200 copies/mL) and median CD4 count was 636 cells/uL (IOR: 376-873). 172 participants (95.6%) completed follow-up. At week 24, 96.3% (78/81) in the 5FU group and 82.0% (73/89) in the placebo group regressed to CIN1 or normal histology (PD 14.3%, CI 5.3%, 23.3%). Among participants with positive LEEP margins at week 0, 88.0% (22/25) in the 5FU versus 61.3% (19/31) in the placebo group regressed to CIN1 or normal (PD 26.7%, CI 5.4%, 48.1%). Genotype-specific hrHPV clearance was similar in both groups (5FU: 58.0%, 40/69; Placebo: 53.8%, 43/80; PD 4.2%, CI -11.7%, 20.2%). ConclusionClinical outcomes from our Phase 2b trial demonstrates that intravaginal 5FU post-LEEP may be a beneficial adjuvant treatment for CIN2/3. Clinical Trial RegistrationNCT05413811

Background As part of Singapore's effort towards precision medicine tailored to Asian diversity, we describe the implementation of a nationwide reproductive carrier screening program. Using a customised 112-gene panel, incorporating population-specific recessive genetic diseases, we outline the overall program design, and initial efforts of community and stakeholder engagement, to inform culturally appropriate implementation. Methods Participants receive culturally tailored online education regarding our reproductive screening program and are provided results with genetic counselling and reproductive options. Community and stakeholder perspectives were assessed through questionnaires and consultations with religious leaders. Results Recruitment is nation-wide, and since initiation of our pilot phase in September 2024, 1,619 couples have registered interest, with 60% uptake of those deemed eligible. Among the 456 couples that have received results to date, four couples (0.9%) were identified to be at increased risk. Community questionnaire responses (n=1002), involving couples who participated in the program as well as the general public, indicated interest is high (59%) across the cohort but awareness, intent to participate and implications for reproductive options differed by sociodemographic factors such as ancestry and religion. Healthcare professional respondents (n=113) acknowledged carrier screening will be routine in medical care, but report limited confidence and resources. Engagement with religious leaders indicated support for the program. Conclusion These early program outcomes and community engagement are guiding the implementation of expanding population-based carrier screening in Singapore, contingent on addressing practical challenges through equitable outreach and professional training.

BackgroundAntiphospholipid syndrome (APS) complicating pregnancy carries significant obstetric morbidity. Secondary APS, arising in the context of systemic autoimmune disease, may confer worse outcomes than primary APS due to additional inflammatory and immunological mechanisms. This study aimed to compare pregnancy outcomes between autoimmune rheumatic disease-associated secondary APS and primary APS managed at a quaternary care hospital in Chennai. MethodsA retrospective observational study analysed 82 pregnancies (secondary APS n=46; primary APS n=36) managed between January 2025 and March 2026. Outcomes including live birth rate, miscarriage, fetal death, preterm birth, pre-eclampsia, and intrauterine growth restriction (IUGR) were compared using chi-square test, Fisher exact test, and independent t-test. Multivariable logistic regression identified independent predictors of adverse outcomes. ResultsLive birth rate was significantly lower in secondary APS compared to primary APS (69.6% vs 86.1%; p=0.048). Triple antiphospholipid antibody positivity was more prevalent in secondary APS (47.8% vs 25.0%; p=0.032). On multivariable analysis, secondary APS (aOR 2.71; 95% CI 1.08-6.81; p=0.033), triple positivity (aOR 3.45; 95% CI 1.39-8.57; p=0.007), and lupus anticoagulant (aOR 2.62; 95% CI 1.01-6.76; p=0.047) independently predicted adverse outcomes. Hydroxychloroquine (aOR 0.39; p=0.038) and combination aspirin plus low-molecular-weight heparin (aOR 0.31; p=0.019) were independently protective. ConclusionSecondary APS is associated with significantly worse pregnancy outcomes than primary APS. Triple antiphospholipid positivity and lupus anticoagulant independently increase obstetric risk. Hydroxychloroquine and combination antithrombotic therapy significantly improve live birth rates. Early rheumatology referral and multidisciplinary obstetric management are essential.

Background: Polygenic risk scores (PRS) for coronary artery disease (CAD) are associated with cardiovascular events, but the relationship between inherited risk and routinely reported coronary computed tomography angiography (CTA) findings has not been studied. Objectives: To evaluate associations between a genome-wide PRS for angiographic coronary disease burden and coronary CTA-derived measures of atherosclerotic severity in a real-world clinical cohort. Methods: We studied Penn Medicine BioBank participants with available genotypes and clinically obtained coronary CTA reports. A previously published PRS for angiographic CAD burden was calculated using pgsc_calc. CAD-RADS scores and coronary artery calcium (CAC) values were extracted from radiology reports using the large language model Llama 3.1 8B. Associations between PRS and CAD-RADS severity were evaluated using Bayesian cumulative ordinal logit regression, while associations with log-transformed CAC burden were assessed using Bayesian linear regression. Results: Among 630 participants, median age was 59 years (IQR 49 - 68), 53% were female, 62% were genetically similar to a European reference population, and 34% to an African reference population. LLM-extracted CAD-RADS and CAC values demonstrated near-perfect agreement with manual abstraction. Higher PRS was associated with greater coronary atherosclerotic burden on CTA. Each 1-standard deviation (SD) increase in PRS was associated with a 20% higher odds of belonging to a more severe CAD-RADS category (cumulative OR 1.20, 95% credible interval 1.06-1.44). Higher PRS was also associated with greater CAC burden ({beta} 0.38, 95% credible interval 0.15 - 0.61). Conclusions: Polygenic risk for angiographic coronary disease burden is reflected in clinically reported coronary CTA severity measures, including CAD-RADS and CAC. These findings demonstrate that inherited susceptibility to CAD manifests as greater anatomic atherosclerotic burden at the time of clinical presentation and support further investigation of genetic risk integration into imaging-based cardiovascular risk assessment.

Introduction Cesarean delivery accounts for nearly one-third of U.S. births and is associated with substantial maternal morbidity and health care costs. Persistent racial disparities have been documented, yet the structural factors contributing to these differences remain incompletely understood. The extent to which insurance coverage shapes racial disparities in cesarean delivery remains unclear. Objective To evaluate the independent and interactive associations of race/ethnicity and insurance coverage with cesarean delivery in the United States. Methods Population-based retrospective cohort study using singleton live births recorded in the United States Vital Statistics Natality files from 2014 to 2024. Multivariable logistic regression was used to estimate the independent effects of race/ethnicity and insurance status on cesarean delivery, including interaction terms to test effect modification, using national birth certificate data. Models were adjusted for maternal demographics, clinical factors, and temporal covariates. Adjusted odds ratios, predicted probabilities, and absolute risk differences were derived from post-estimation marginal effects. The main outcome measure was cesarean delivery (yes vs no). Results Among 41,543,568 deliveries from 2014 to 2024, 13,312,221 (32.0%) were cesarean deliveries. After adjustment, both race and ethnicity and insurance status were independently associated with cesarean delivery. Compared with non-Hispanic White women, non-Hispanic Black women had higher odds of cesarean delivery (odds ratio [OR], 1.22; 95% CI, 1.22-1.23). Relative to uninsured women, those with private insurance had 59% higher odds of cesarean delivery (OR, 1.59; 95% CI, 1.58-1.60). Significant interaction effects were observed, indicating that insurance coverage modified racial and ethnic differences in cesarean delivery. Non-Hispanic Black women had the highest predicted probabilities across all insurance categories, with the largest absolute disparities observed among uninsured women. Conclusion Racial and ethnic differences in cesarean delivery persist in the United States and are modified by insurance coverage, suggesting that coverage-related differences may contribute to inequities in obstetric care.

Background Open health data repositories receive billions in public funding, yet no systematic framework exists to evaluate their downstream scholarly impact, the composition of the research communities they cultivate, or the breadth of disciplines they reach. We introduce a two-degree citation methodology to quantify knowledge diffusion from open data, normalized by funding, and apply it to four major health data repositories. Methods Using the OpenAlex bibliometric database (January-February 2026), we identified all first-degree citing publications (n = 30,049) and their second-degree citing publications (n = 485,396), defined as papers citing those first-degree publications, for MIMIC (versions I-IV; retrospective EHR data; $14.4M), UK Biobank (prospective cohort with genomics; $525.5M), OpenSAFELY (federated EHR platform; $53.7M), and All of Us (prospective national cohort with biobanking and community engagement; $2,160M). We extracted author demographics (gender via Genderize.io, institutional country income via World Bank 2024 classifications) and research topics. Chi-square tests with odds ratios assessed demographic differences across repositories. Results Funding-normalized first-degree papers per $1M ranged from 689 (MIMIC) to 1 (All of Us), though these figures reflect total program investment, which included community engagement and biobanking for prospective cohorts in addition to data-curation costs. The citation amplification ratio was consistent across these four repositories (9.3-11.5x). Author demographics differed significantly (p < 0.001): LMIC authorship ranged from 41.8% (MIMIC) to 4.3% (All of Us), while female authorship showed the opposite pattern, lowest for MIMIC (31.8%) and highest for All of Us (43.2%). Female authors were consistently underrepresented in senior (last-author) compared with first-author positions across all repositories. Differences in scope, design, and what funding covers limit direct comparisons. Conclusions Open health data generates a consistent ~10x indirect citation amplification beyond its direct users, a ratio that held across repositories spanning over two orders of magnitude in funding. The large differences in funding-normalized output partly reflect structural differences between retrospective databases and prospective cohorts. Low-cost access combined with intentional community building attracted globally diverse research communities with LMIC investigators in intellectual leadership positions, while a persistent gender gap in senior authorship across all repositories reflects disciplinary and structural inequities that data access policies alone cannot address. Future evaluations of open data investments should examine who is producing research, from where, in what positions, and whether their participation translates into locally relevant knowledge production.

BackgroundArtificial intelligence and machine learning (AI/ML) are among the fastest-growing domains in NIH research funding, but whether children have shared equitably in this expansion is unknown. We characterized pediatric representation in NIH AI/ML funding from fiscal years (FY) 2020 to 2024. MethodsNIH grant data were obtained from Research Portfolio Online Reporting Tools Expenditures and Results bulk files for FY2020 to FY2024. AI/ML grants were identified using the NIH Research, Condition, and Disease Categorization "Machine Learning and Artificial Intelligence" category, and pediatric grants using the "Pediatric" category. Subprojects were excluded. Grants were deduplicated within each fiscal year by core project number for trend analyses and across all years retaining the most recent fiscal year for cross-sectional totals. Disease areas were identified by keyword searches of titles and abstracts. ResultsAcross FY2020 to FY2024, 5,624 unique NIH AI/ML grants totaling $3,371 million were identified. Of these, 836 grants (14.9%) were classified as pediatric, representing $401 million (11.9%) of total NIH AI/ML funding. Although this share was consistent with the historically reported overall NIH pediatric funding baseline of approximately 10% to 12%, it remained substantially below the US pediatric population share of approximately 22%. The pediatric share of NIH AI/ML funding declined from 12.3% in FY2020 to 10.8% in FY2024, despite growth in absolute pediatric funding. Indexed to FY2020, pediatric AI/ML funding grew approximately 2.6-fold compared with 3.0-fold growth in the total portfolio. Across disease areas, unadjusted adult/general-to-pediatric funding ratios ranged from 2.0-fold in mental health to 9.8-fold in cancer. ConclusionsPediatric representation in NIH AI/ML funding remained low and declined over time as the overall portfolio expanded. These findings suggest that growth in NIH AI/ML investment has not been matched by proportional gains for pediatric research.

BackgroundDiabetic retinopathy (DR) is the leading cause of preventable blindness among working-age adults worldwide, yet screening coverage remains inadequate, particularly in low-and middle-income countries. Automated deep learning systems offer potential to address the global shortage of expert graders, but most existing models lack lesion-level interpretability and are not aligned with established clinical referral frameworks. We developed and validated DRAGS (Diabetic Retinopathy Automated Grading System), a hybrid deep learning model that grades DR according to the UK Diabetic Eye Screening Programme (DESP) classification and provides lesion-level explainability. MethodsWe trained and validated a DenseNet-201-based convolutional neural network on 20,281 anonymised fundus images from two tertiary eye care institutions in Bangladesh. Images were graded by fellowship-trained retinal specialists using the UK DESP framework, resulting in 10 clinically interpretable classes that combine retinopathy grade (R0-R3) and maculopathy status (M0/M1). A companion dataset of 2,936 pixel-level lesion masks spanning nine pathological categories was used to train a parallel multi-label lesion-detection head. The dataset was partitioned 70:15:15 (patient-stratified). Performance was evaluated using macro-averaged AUROC (DeLong estimator), sensitivity, specificity, F1 score, quadratically weighted Cohens {kappa}, and expected calibration error (ECE), with 95% CIs from 2000 bootstrap resamples. Grad-CAM spatial alignment with ground-truth lesion masks was assessed using Dice and IoU. This study follows the TRIPOD+AI reporting guidelines. FindingsOn the held-out test set (Component I: n = 3,044; Component II: n {approx} 440), DRAGS achieved class-wise precision, recall, and F1 scores ranging from 0{middle dot}88 to 0{middle dot}99 across all ten UK DESP grades, with advanced proliferative stages (R3-M0, R3-M1) consistently exceeding 0{middle dot}95. Overall accuracy was approximately 91{middle dot}1% and quadratically weighted Cohens {kappa} was approximately 0{middle dot}90. For referable versus non-referable DR, sensitivity was 90{middle dot}7% and specificity was 91{middle dot}9%. The companion lesion-detection head achieved macro-averaged sensitivity of 93{middle dot}9%, specificity of 99{middle dot}5%, and AUC of 0{middle dot}997 across nine lesion classes; seven of nine classes achieved AUC = 1{middle dot}00. Grad-CAM activations showed progressive spatial shift from diffuse (normal) to lesion-dense peripheral patterns (proliferative DR), with maximal agreement for microaneurysms and exudates. Mean inference time was 110-160 ms per image. InterpretationDRAGS demonstrates high diagnostic accuracy for nine-class UK DESP-aligned DR grading, with clinically interpretable lesion-level explainability on a large real-world LMIC dataset. External validation and prospective clinical evaluation are warranted before deployment. FundingThe present study received no funding.

Background: Artificial intelligence chatbots (AICs), as a form of generative artificial intelligence (AI), are increasingly being considered for use in scholarly peer review to assist with tasks such as identifying methodological issues, verifying references, and improving language clarity. Despite these potential benefits, concerns remain regarding their reliability, ethical implications, and transparency. Evidence on how medical journal peer reviewers perceive the role and impact of AICs is limited. This study explored reviewers' familiarity with AICs, perceived benefits and challenges, ethical concerns, and anticipated future roles in peer review. Methods: We conducted a cross-sectional online survey of medical journal peer reviewers. Corresponding author information was extracted from MEDLINE-indexed articles added to PubMed within a two-month period using an R-based approach. A total of 72,851 authors were invited via email to participate; those who self-identified as peer reviewers were eligible. The 29-item survey assessed familiarity with AICs and perceptions of their benefits and limitations in peer review. The survey was administered via SurveyMonkey from April 28 to June 16, 2025, with two reminder emails sent during the data collection period. Results: A total of 1,260 respondents completed the survey. Most participants were familiar with AICs (86.2%) and had used tools such as ChatGPT for general purposes (87.7%), but the majority had not used AICs for peer review (70.3%). Most respondents reported that their institutions do not provide training on AIC use in peer review (69.5%), although many expressed interest in such training (60.7%). Perceptions of AIC benefits were mixed, while concerns were widely shared, particularly regarding potential algorithmic bias (80.3%) and issues related to trust and user acceptance (73.3%). Conclusions: While familiarity with AICs is high among medical journal peer reviewers, their use in peer review remains limited. There is clear interest in training and guidance, however, concerns related to ethics, data privacy, and research integrity persist and should be addressed before broader implementation.

Background: Cardiac rehabilitation (CR) is a cost-effective, evidence-based intervention that improves outcomes for patients with heart failure (HF), yet access remains inequitable, particularly among Medicaid enrollees. This study evaluates the state-by-state variability in Medicaid coverage for CR services and examines the implications for health equity in vulnerable populations. Methods: We conducted a cross-sectional policy analysis of all 50 U.S. states to assess Medicaid coverage for outpatient CR services billed under CPT codes 93797 (without ECG monitoring) and 93798 (with ECG monitoring). Publicly available Medicaid documents were reviewed and supplemented with direct communication with state Medicaid agencies. States were categorized into full, partial/inconclusive, or no coverage. Geographic trends were visualized through heat maps and contextualized using state-level Medicaid enrollment data. Results: Marked disparities in CR coverage were identified. Only 41 states reimbursed for CPT 93797, and 43 for CPT 93798. Eight states lacked coverage for either code, predominantly in the South and Mountain West, including Arkansas, Georgia, Louisiana, Mississippi, Nevada, and Utah. States with the highest Medicaid enrollment (e.g., Louisiana, Arkansas) often provided no CR coverage, compounding access barriers for high-risk, low-income populations. Conclusions: The absence of standardized Medicaid coverage for CR contributes to systemic inequities in cardiovascular care, disproportionately impacting disadvantaged communities. Aligning Medicaid policies to ensure universal CR access--particularly through tele-rehabilitation and value-based care models--could reduce hospitalizations, improve survival, and promote health equity across the U.S.

Background: Statewide immunization data are essential for monitoring vaccination trends and evaluating immunization program impact. In the United States, Immunization Information Systems (IIS) were established in the early 1990s to collect these data; however, operational, legal, and procedural details vary across states and over time. This study summarized differences in IIS characteristics, such as legal requirements and reporting procedures, across U.S. states and jurisdictions over time. Methods: We analyzed survey data from previous work in 2000 and the Centers for Disease Control and Prevention (CDC) in 2012, 2018, and 2024. Our review focused on legislation and reporting requirements for immunization registries across 50 states and 14 jurisdictions, including U.S. territories and Freely Associated States. Results: Between 2000 and 2024, legal frameworks and reporting practices for immunization registries expanded across U.S. states and jurisdictions. The number of states with laws or administrative rules authorizing immunization registries increased from 24 states in 2000 to all 50 states, the District of Columbia, five metropolitan areas, five U.S. territories, and three Freely Associated States in 2024. Over the same period, reporting requirements also became more widespread. The number of states and jurisdictions mandating providers to report immunization records increased from 12 in 2000 to 54 in 2024. Consent policies also changed over time. By 2024, most states and jurisdictions had adopted implicit consent for reporting children's immunization records (41; 64%), while a smaller proportion required explicit parental consent (7; 11%) or implemented mandatory reporting without consent (14; 22%). Discussion: IIS infrastructure and reporting requirements have expanded across U.S. states and jurisdictions over the past two decades, while heterogeneity in consent policies and reporting practices persists. These temporal changes may need to be considered when interpreting IIS data, particularly in longitudinal and cross-jurisdictional analyses.