Bangladesh should engage the private sector for malaria elimination by 2030.

Bangladesh reduced malaria incidence by 93% from 2008 to 2020 through the action of governmental and non-governmental organizations. The Bangladesh context is unique to South Asia because its successful public sector malaria control programs have historically not engaged corporate partners (as undertaken in Sri Lanka and proposed in India). However, ∼18 million people continue to live at risk of infection in Bangladesh and for-profit private healthcare providers, catalytic for malaria elimination in many countries, are expected to benefit the national program. We distilled (from a large and complex literature) nine distinct strategies important in other developing settings and weighed them in the context of Bangladesh's flourishing private health care sector, driven by patient demand, self-interest and aspirations for public good, as well as heterogeneity in providers and malaria-prevalence. We propose a new model dependent on five strategies and its immediate deployment considerations in high endemic areas, to empower Bangladesh's phased agenda of eliminating indigenous malaria transmission by 2030.

Onyx embolization for salvage hemostasis after intraoperative rupture.

Video 1 shows three cases of intraoperative rupture during aneurysm coiling. Management of intraoperative aneurysm rupture is reviewed in brief, including reversal of anticoagulation/antiplatelets, intracranial pressure control, and rapid balloon deployment for control across the aneurysm neck. However, in all three cases, contrast extravasation continues despite aneurysm coiling, reversal of anticoagulation, and maximizing medical management. This is presumed to occur when the neck of the aneurysm is the site of rupture. We review the use of a salvage technique that can be considered as a last-ditch maneuver in these scenarios, which is deployment of Onyx liquid embolic (Medtronic). Onyx is the preferred liquid embolic for this use as it is cohesive instead of adhesive. The solvent used with Onyx, dimethyl sulfoxide, is also compatible with standard balloons.1 This is relevant because a balloon is needed both for control of hemorrhage during salvage embolization and for protection of the parent vessel from the embolic material.neurintsurg;jnis-2023-021402v1/V1F1V1Video 1Demonstrating the technique of Onyx embolization for salvage hemostasis after intraoperative aneurysm rupture during coiling. Three example cases are shown.

Hepatitis B virus (HBV) viremia despite tenofovir disoproxil fumarate-containing antiretroviral therapy in persons with HBV/HIV coinfection.

BACKGROUND: The goal of treatment of hepatitis B virus (HBV) and human immunodeficiency virus (HIV) coinfection is suppression of both viruses; yet incomplete HBV suppression on tenofovir (TFV) disoproxil fumarate (TDF)-based antiretroviral therapy (ART) is common. This study investigated TFV resistance-associated mutations (RAMs) in individuals with HBV/HIV coinfection with viremia on TDF/lamivudine (3TC)-containing ART. METHODS: Samples from individuals with HBV DNA levels ≥20 IU/mL in a cross-sectional study of 138 persons with HBV/HIV coinfection in Ghana were analyzed in the present study. HBV was sequenced for RAM analysis. TFV-diphosphate (TFV-DP) concentration in peripheral blood mononuclear cells (PBMCs) was used to assess ART adherence level. RESULTS: Nine of 138 participants (6.5 %) had detectable HBV DNA levels ≥20 IU/mL while on ART. Seven of the nine participants had TFV-DP concentrations commensurate with 7 doses per week, and six had suppressed HIV RNA. Phylogenetic analysis revealed that eight sequences were HBV genotype E, with one genotype E/A recombinant. Ten previously-reported TFV RAMs were present in the study samples; eight were wild-type for HBV genotype E. The non-genotype-E-wild-type point mutations M267L and K333Q were found in two and one patients, respectively. No 3TC RAMs were found. CONCLUSION: HBV viremia despite high adherence to TDF/3TC-based ART may be associated with the presence of TFV RAMs. These findings highlight the need for enhanced resistance monitoring and further research to examine the clinical significance of reported TFV RAMs. Individuals with HBV/HIV coinfection and TFV resistance on TDF-based ART may need alternative treatment strategies.

[Clinical Pathway for Suicide Risk Screening in Adult Primary Care Settings:Special Recommendations]. / Klinikai útvonal az öngyilkossági kockázat szurésére a felnott alapellátásban: Speciális ajánlások.

Suicide is a serious public health concern. On average, 80% of suicide decedents had contact with primary care within one year of their suicide. This and other research underscore the importance of screening for suicide risk within primary care settings, and implementation of suicide risk screening is already underway in many practices. However, while primary care practices may be familiar with screening for other mental health concerns (e.g., depression), many feel uncomfortable or unprepared for suicide risk screening. To meet the increasing demand for evidence-based suicide-risk screening guidance, we provide a clinical pathway for adult primary care practices (to include family medicine, internal medicine, women's health). The pathway was developed by experts with research, clinical expertise and experience in suicide risk screening and primary care. We also provide detailed guidance to aid primary care practices in their decisions about how to implement the clinical pathway.

'Like walking with someone as opposed to trying to catch up to them'-Dynamics at play when clinicians and young people formulate together.

OBJECTIVE: The aim of the present study was to explore the social process of formulation in talk therapy between young people and clinicians. DESIGN: Qualitative semi-structured interview study. METHOD: Ten young people (male = 6, female = 4, age range = 16-23 years) and nine clinicians from various disciplines within a youth mental health service were interviewed. Constructivist grounded theory was used for the analysis. RESULTS: Four themes were constructed from the data; a 'level playing field' between young person and clinician enables formulation, formulating is a constant process of getting it right and getting it wrong, emotional expression and attunement get us closer to each other and to understanding, and 'formulation versus diagnosis' can create tension in the therapy room. The constructivist grounded theory devised demonstrated how the dynamics of power, collaboration, openness, and the therapeutic relationship are constantly in flux during the process of formulation. CONCLUSION: The paper presents a constructivist grounded theory which incorporates dynamics relating to power, collaboration, and openness. The importance of the therapeutic relationship is also emphasised. The theory encourages continuous and recursive personal reflection by the therapist as to how they can be optimally attuned to the dynamics of power, collaboration, and openness with young people.

Comparing suicide risk screening strategies in Spanish-speaking pediatric patients.

BACKGROUND: Suicide and suicidal behaviors pose significant global public health challenges, especially among young individuals. Effective screening strategies are crucial for addressing this crisis, with depression screening and suicide-specific tools being common approaches. This study compares their effectiveness by evaluating the Ask Suicide-Screening Questions (ASQ) against item 9 of the Patient Health Questionnaire-A (PHQ-A). METHODS: This study is a secondary analysis of the Argentinean-Spanish version of the ASQ validation study, an observational, cross-sectional, and multicenter study conducted in medical settings in Buenos Aires, Argentina. A convenience sample of pediatric outpatients/inpatients aged 10 to 18 years completed the ASQ, PHQ-A, and Suicide Ideation Questionnaire (SIQ) along with clinical and sociodemographic questions. RESULTS: A sample of 267 children and adolescents were included in this secondary analysis. Results show that the ASQ exhibited higher sensitivity (95.1%; 95% CI: 83% - 99%) compared to PHQ-A item 9 (73.1%; 95% CI: 57% - 85%), and superior performance in identifying suicide risk in youth. LIMITATIONS: The study included a convenience sampling and was geographically restricted to Buenos Aires, Argentina. The study also lacked longitudinal follow-up to assess the predictive validity of these screening tools for suicide risk. CONCLUSION: The study highlights the ASQ's effectiveness in identifying suicide risk among youth, emphasizing the importance of specialized screening tools over depression screening tools alone for accurate risk assessment in this population.

The Power of Preference Signaling: A Monumental Shift in the Orthopaedic Surgery Application Process.

INTRODUCTION: Orthopaedic surgery has consistently been one of the most competitive specialties in the US residency selection process. This is due in part to the steady upward trend in average applications received per program and average applications submitted per applicant, which is of growing concern. With the implementation of the Preference Signaling Program, the total number of applications has now dropped for the first time in many years, indicating signaling may improve the application process. The hypothesis is that signaling has led to a decrease in applications sent by applicants and a decrease in applications received by programs. METHODS: A 7-question survey regarding their interview and match statistics was sent to orthopaedic surgery residency programs that participated in the Electronic Residency Application Service during the 2023-2024 application cycle. A response from the program director/administrator was then recorded. RESULTS: Our program search yielded 159 programs with 106 respondents (66.7%). 82 programs (78.8%) solely interviewed applicants who signaled their program. 92.7% of current interns signaled the program where they matched, and 88 programs (84.6%) matched only applicants who signaled. 95 programs (89.6%) revealed that implementing signaling has improved the application process. CONCLUSION: Most of the programs only interviewed applicants who also signaled, and nearly all matched orthopaedic surgery applicants from the 2022-2023 cycle signaled their matching program. Orthopaedic surgery applicants should consider only applying to 30 programs and using all 30 available signals. Applicants should also be more confident knowing that beyond the 30 signals they use, there is limited support to say that they will receive an interview outside of these 30 applications. Orthopaedic surgery programs will also now have the ability to allocate more time to applicants most interested in their program, given the reduction of applications.

Comparative Effectiveness of Second-Line Antihyperglycemic Agents for Cardiovascular Outcomes: A Multinational, Federated Analysis of LEGEND-T2DM.

BACKGROUND: Sodium-glucose cotransporter 2 inhibitors (SGLT2is) and glucagon-like peptide-1 receptor agonists (GLP-1 RAs) reduce the risk of major adverse cardiovascular events (MACE) in patients with type 2 diabetes mellitus (T2DM). However, their effectiveness relative to each other and other second-line antihyperglycemic agents is unknown, without any major ongoing head-to-head clinical trials. OBJECTIVES: The aim of this study was to compare the cardiovascular effectiveness of SGLT2is, GLP-1 RAs, dipeptidyl peptidase-4 inhibitors (DPP4is), and clinical sulfonylureas (SUs) as second-line antihyperglycemic agents in T2DM. METHODS: Across the LEGEND-T2DM (Large-Scale Evidence Generation and Evaluation Across a Network of Databases for Type 2 Diabetes Mellitus) network, 10 federated international data sources were included, spanning 1992 to 2021. In total, 1,492,855 patients with T2DM and cardiovascular disease (CVD) on metformin monotherapy were identified who initiated 1 of 4 second-line agents (SGLT2is, GLP-1 RAs, DPP4is, or SUs). Large-scale propensity score models were used to conduct an active-comparator target trial emulation for pairwise comparisons. After evaluating empirical equipoise and population generalizability, on-treatment Cox proportional hazards models were fit for 3-point MACE (myocardial infarction, stroke, and death) and 4-point MACE (3-point MACE plus heart failure hospitalization) risk and HR estimates were combined using random-effects meta-analysis. RESULTS: Over 5.2 million patient-years of follow-up and 489 million patient-days of time at risk, patients experienced 25,982 3-point MACE and 41,447 4-point MACE. SGLT2is and GLP-1 RAs were associated with lower 3-point MACE risk than DPP4is (HR: 0.89 [95% CI: 0.79-1.00] and 0.83 [95% CI: 0.70-0.98]) and SUs (HR: 0.76 [95% CI: 0.65-0.89] and 0.72 [95% CI: 0.58-0.88]). DPP4is were associated with lower 3-point MACE risk than SUs (HR: 0.87; 95% CI: 0.79-0.95). The pattern for 3-point MACE was also observed for the 4-point MACE outcome. There were no significant differences between SGLT2is and GLP-1 RAs for 3-point or 4-point MACE (HR: 1.06 [95% CI: 0.96-1.17] and 1.05 [95% CI: 0.97-1.13]). CONCLUSIONS: In patients with T2DM and CVD, comparable cardiovascular risk reduction was found with SGLT2is and GLP-1 RAs, with both agents more effective than DPP4is, which in turn were more effective than SUs. These findings suggest that the use of SGLT2is and GLP-1 RAs should be prioritized as second-line agents in those with established CVD.

Robust identification of environmental exposures and community characteristics predictive of rapid lung disease progression.

Environmental exposures and community characteristics have been linked to accelerated lung function decline in people with cystic fibrosis (CF), but geomarkers, the measurements of these exposures, have not been comprehensively evaluated in a single study. To determine which geomarkers have the greatest predictive potential for lung function decline and pulmonary exacerbation (PEx), a retrospective longitudinal cohort study was performed using novel Bayesian joint covariate selection methods, which were compared with respect to PEx predictive accuracy. Non-stationary Gaussian linear mixed effects models were fitted to data from 151 CF patients aged 6-20 receiving care at a CF Center in the midwestern US (2007-2017). The outcome was forced expiratory volume in 1 s of percent predicted (FEV1pp). Target functions were used to predict PEx from established criteria. Covariates included 11 routinely collected clinical/demographic characteristics and 45 geomarkers comprising 8 categories. Unique covariate selections via four Bayesian penalized regression models (elastic-net, adaptive lasso, ridge, and lasso) were evaluated at both 95 % and 90 % credible intervals (CIs). Resultant models included one to 6 geomarkers (air temperature, percentage of tertiary roads outside urban areas, percentage of impervious nonroad outside urban areas, fine atmospheric particulate matter, fraction achieving high school graduation, and motor vehicle theft) representing weather, impervious descriptor, air pollution, socioeconomic status, and crime categories. Adaptive lasso had the lowest information criteria. For PEx predictive accuracy, covariate selection from the 95 % CI elastic-net had the highest area under the receiver-operating characteristic curve (mean ± standard deviation; 0.780 ± 0.026) along with the 95 % CI ridge and lasso methods (0.780 ± 0.027). The 95 % CI elastic-net had the highest sensitivity (0.773 ± 0.083) while the 95 % CI adaptive lasso had the highest specificity (0.691 ± 0.087), suggesting the need for different geomarker sets depending on monitoring goals. Surveillance of certain geomarkers embedded in prediction algorithms can be used in real-time warning systems for PEx onset.

High baseline perivascular space volume in basal ganglia is associated with attention and executive function decline in Parkinson's disease.

BACKGROUND: Pathologic perivascular spaces (PVS), the fluid-filled compartments surrounding brain vasculature, may underlie cognitive decline in Parkinson's disease (PD). However, whether this impacts specific cognitive domains has not been investigated. OBJECTIVES: This study examined the relationship of PVS volume at baseline with domain-specific and global cognitive change over 2 years in PD individuals. METHODS: A total of 39 individuals with PD underwent 3T T1w magnetic resonance imaging to determine PVS volume fraction (PVS volume normalized to total regional volume) within (i) centrum semiovale, (ii) prefrontal white matter (medial orbitofrontal, rostral middle frontal, and superior frontal), and (iii) basal ganglia. A neuropsychological battery included assessment of cognitive domains and global cognitive function at baseline and after 2 years. RESULTS: Higher basal ganglia PVS at baseline was associated with greater decline in attention, executive function, and global cognition scores. CONCLUSIONS: While previous reports have associated elevated PVS volume in the basal ganglia with decline in global cognition in PD, our findings show such decline may affect the attention and executive function domains.

Development and validation of a patient-level model to predict dementia across a network of observational databases.

BACKGROUND: A prediction model can be a useful tool to quantify the risk of a patient developing dementia in the next years and take risk-factor-targeted intervention. Numerous dementia prediction models have been developed, but few have been externally validated, likely limiting their clinical uptake. In our previous work, we had limited success in externally validating some of these existing models due to inadequate reporting. As a result, we are compelled to develop and externally validate novel models to predict dementia in the general population across a network of observational databases. We assess regularization methods to obtain parsimonious models that are of lower complexity and easier to implement. METHODS: Logistic regression models were developed across a network of five observational databases with electronic health records (EHRs) and claims data to predict 5-year dementia risk in persons aged 55-84. The regularization methods L1 and Broken Adaptive Ridge (BAR) as well as three candidate predictor sets to optimize prediction performance were assessed. The predictor sets include a baseline set using only age and sex, a full set including all available candidate predictors, and a phenotype set which includes a limited number of clinically relevant predictors. RESULTS: BAR can be used for variable selection, outperforming L1 when a parsimonious model is desired. Adding candidate predictors for disease diagnosis and drug exposure generally improves the performance of baseline models using only age and sex. While a model trained on German EHR data saw an increase in AUROC from 0.74 to 0.83 with additional predictors, a model trained on US EHR data showed only minimal improvement from 0.79 to 0.81 AUROC. Nevertheless, the latter model developed using BAR regularization on the clinically relevant predictor set was ultimately chosen as best performing model as it demonstrated more consistent external validation performance and improved calibration. CONCLUSIONS: We developed and externally validated patient-level models to predict dementia. Our results show that although dementia prediction is highly driven by demographic age, adding predictors based on condition diagnoses and drug exposures further improves prediction performance. BAR regularization outperforms L1 regularization to yield the most parsimonious yet still well-performing prediction model for dementia.

Epigenome-wide association study of fluoride exposure during early adolescence and DNA methylation among U.S. children.

Exposure to fluoride in early childhood has been associated with altered cognition, intelligence, attention, and neurobehavior. Fluoride-related neurodevelopment effects have been shown to vary by sex and very little is known about the mechanistic processes involved. There is limited research on how fluoride exposure impacts the epigenome, potentially leading to changes in DNA methylation of specific genes regulating key developmental processes. In the Cincinnati Childhood Allergy and Air Pollution Study (CCAAPS), urine samples were analyzed using a microdiffusion method to determine childhood urinary fluoride adjusted for specific gravity (CUFsg) concentrations. Whole blood DNA methylation was assessed using the Infinium MethylationEPIC BeadChip 850 k Array. In a cross-sectional analysis, we interrogated epigenome-wide DNA methylation at 775,141 CpG loci across the methylome in relation to CUFsg concentrations in 272 early adolescents at age 12 years. Among all participants, higher concentrations of CUF were associated with differential methylation of one CpG (p < 6 × 10-8) located in the gene body of GBF1 (cg25435255). Among females, higher concentrations of CUFsg were associated with differential methylation of 7 CpGs; only three CpGs were differentially methylated among males with no overlap of significant CpGs observed among females. Secondary analyses revealed several differentially methylated regions (DMRs) and CpG loci mapping to genes with key roles in psychiatric outcomes, social interaction, and cognition, as well as immunologic and metabolic phenotypes. While fluoride exposure may impact the epigenome during early adolescence, the functional consequences of these changes are unclear warranting further investigation.

Fear of Reinjury, Psychological Factors, and Sport Played Have Negative Impact on Return to Sport Following Medial Patellofemoral Ligament Reconstruction for Patellar Instability.

PURPOSE: To analyze factors that affect return to sport after medial patellofemoral ligament reconstruction (MPFLR), such as psychological factors, sport played, and a positive apprehension test following surgery, and to determine the average return to sport rates and time to return to sport. METHODS: A literature search was conducted according to Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. Included studies met the following criteria: patients underwent MPFLR for patellar instability, return to sport was recorded, and a factor that affected return to sport was mentioned. Search terms included medial patellofemoral ligament, tibial tubercle osteotomy, tibial tubercle transfer, return to play, and return to sport. RESULTS: Eighteen of 632 identified studies met inclusion criteria, and 1,072 patients who underwent MFPLR were recorded. Return-to-sport rates and mean/median time ranged from 60.0% to 100% and 3 to 10.4 months, respectively. Of the patients, 55.6% to 84.0% returned to sport without decreasing the level of competition. Six of 12 studies (50.0%) reported fear of reinjury as the top reason for patients not returning or returning at a lower level of sport. Volleyball/handball had the lowest return to the same level following surgery (18.2%-50.0%). CONCLUSIONS: Athletes who underwent MPFLR following recurrent patellar instability returned to sport at a range of 60.0% to 100%. Return to sport at the same level or higher was found to have a lower maximum rate at 55.6% to 84.0%. Fear of reinjury and sport played were found to have a substantial impact on ability to return to sport. Surgeons can use this information to advise patients on expectations following surgery. LEVEL OF EVIDENCE: Level IV, systematic review of Level III and IV studies.

Variability in personal exposure to ultrafine and fine particles by microenvironment among adolescents in Cincinnati.

Personal exposure to air pollution is influenced by an individual's time-activity patterns, but data regarding personal exposure to air pollution among children populations is lacking. The objective of this study was to characterize personal exposure to both PM2.5 and ultrafine particles (UFPs) using two portable real-time monitors, combined with GPS logging, and describe the relationship between these exposures across time and microenvironments among adolescents with asthma. Participants completed personal exposure monitoring for seven consecutive days and PM2.5 and UFP concentrations experienced in five microenvironments were determined using GPS location and mobility data. Average UFP and PM2.5 exposure varied across microenvironments with the highest average UFP exposure concentrations observed in transit (10,910 ± 27,297 p/cc), though correlations between UFP and PM2.5 concentrations in transit were low (0.24) and did not reach statistical significance (p > 0.05). We calculated exposure time ratios for each participant. Across participants, UFP exposures within the transit environment demonstrated the highest ratio (average exposure-time ratio = 1.91) though only 3 % of overall sampling time among all participants was monitored in transit (74/2840 h). We did not observe similar trends among PM2.5 exposures. The correlations between UFP and PM2.5 exposures varied throughout the day, with an overall correlation ranging from moderate to high among participants. Identifying microenvironments and activities where high exposure to PM occurs may offer potential targets for interventions to reduce overall exposures among sensitive groups.

The Legacy of Redlining: Increasing Childhood Asthma Disparities Through Neighborhood Poverty.

RATIONALE: Identifying the root causes of racial disparities in childhood asthma is critical for health equity. OBJECTIVES: To determine if the 1930's racist policy of redlining led to present-day disparities in childhood asthma by increasing community-level poverty and decreasing neighborhood socioeconomic position (SEP). METHODS: We categorized census tracts at birth of participants from the Children's Respiratory and Environmental Workgroup birth cohort consortium into A, B, C, or D categories as defined by the Home Owners Loan Corporation (HOLC), with D being the highest perceived risk. Surrogates of present-day neighborhood-level SEP were determined for each tract including the percentage of low-income households, the CDC's social vulnerability index (SVI), and other tract-level variables. We performed causal mediation analysis, which, under the assumption of no unmeasured confounding, estimates the direct and mediated pathways by which redlining may cause asthma disparities through census tract-level mediators adjusting for individual-level covariates. MEASUREMENTS AND MAIN RESULTS: Of 4,849 children, the cumulative incidence of asthma through age 11 was 26.6% and 13.2% resided in census tracts with a HOLC grade of D. In mediation analyses, residing in grade D tracts (aOR = 1.03 [95%CI 1.01,1.05]) was significantly associated with childhood asthma, with 79% of this increased risk mediated by percentage of low-income households; results were similar for SVI and other tract-level variables. CONCLUSIONS: The historical structural racist policy of redlining led to present-day asthma disparities in part through decreased neighborhood SEP. Policies aimed at reversing the effects of structural racism should be considered to create more just, equitable, and healthy communities.

Particulate Matter Exposure and Default Mode Network Equilibrium During Early Adolescence.

Background: Air pollution exposure has been associated with adverse cognitive and mental health outcomes in children, adolescents, and adults, although youth may be particularly susceptible given ongoing brain development. However, the neurodevelopmental mechanisms underlying the associations among air pollution, cognition, and mental health remain unclear. We examined the impact of particulate matter (PM2.5) on resting-state functional connectivity (rsFC) of the default mode network (DMN) and three key attention networks: dorsal attention, ventral attention, and cingulo-opercular. Methods: Longitudinal changes in rsFC within/between networks were assessed from baseline (9-10 years) to the 2-year follow-up (11-12 years) in 10,072 youth (M ± SD = 9.93 + 0.63 years; 49% female) from the Adolescent Brain Cognitive Development (ABCD®) study. Annual ambient PM2.5 concentrations from the 2016 calendar year were estimated using hybrid ensemble spatiotemporal models. RsFC was estimated using functional neuroimaging. Linear mixed models were used to test associations between PM2.5 and change in rsFC over time while adjusting for relevant covariates (e.g., age, sex, race/ethnicity, parental education, and family income) and other air pollutants (O3, NO2). Results: A PM2.5 × time interaction was significant for within-network rsFC of the DMN such that higher PM2.5 concentrations were associated with a smaller increase in rsFC over time. Further, significant PM2.5 × time interactions were observed for between-network rsFC of the DMN and all three attention networks, with varied directionality. Conclusion: PM2.5 exposure was associated with alterations in the development and equilibrium of the DMN-a network implicated in self-referential processing-and anticorrelated attention networks, which may impact trajectories of cognitive and mental health symptoms across adolescence.

Knowledge-guided generative artificial intelligence for automated taxonomy learning from drug labels.

OBJECTIVES: To automatically construct a drug indication taxonomy from drug labels using generative Artificial Intelligence (AI) represented by the Large Language Model (LLM) GPT-4 and real-world evidence (RWE). MATERIALS AND METHODS: We extracted indication terms from 46 421 free-text drug labels using GPT-4, iteratively and recursively generated indication concepts and inferred indication concept-to-concept and concept-to-term subsumption relations by integrating GPT-4 with RWE, and created a drug indication taxonomy. Quantitative and qualitative evaluations involving domain experts were performed for cardiovascular (CVD), Endocrine, and Genitourinary system diseases. RESULTS: 2909 drug indication terms were extracted and assigned into 24 high-level indication categories (ie, initially generated concepts), each of which was expanded into a sub-taxonomy. For example, the CVD sub-taxonomy contains 242 concepts, spanning a depth of 11, with 170 being leaf nodes. It collectively covers a total of 234 indication terms associated with 189 distinct drugs. The accuracies of GPT-4 on determining the drug indication hierarchy exceeded 0.7 with "good to very good" inter-rater reliability. However, the accuracies of the concept-to-term subsumption relation checking varied greatly, with "fair to moderate" reliability. DISCUSSION AND CONCLUSION: We successfully used generative AI and RWE to create a taxonomy, with drug indications adequately consistent with domain expert expectations. We show that LLMs are good at deriving their own concept hierarchies but still fall short in determining the subsumption relations between concepts and terms in unregulated language from free-text drug labels, which is the same hard task for human experts.

Assessing Covariate Balance with Small Sample Sizes.

Propensity score adjustment addresses confounding by balancing covariates in subject treatment groups through matching, stratification, inverse probability weighting, etc. Diagnostics ensure that the adjustment has been effective. A common technique is to check whether the standardized mean difference for each relevant covariate is less than a threshold like 0.1. For small sample sizes, the probability of falsely rejecting the validity of a study because of chance imbalance when no underlying balance exists approaches 1. We propose an alternative diagnostic that checks whether the standardized mean difference statistically significantly exceeds the threshold. Through simulation and real-world data, we find that this diagnostic achieves a better trade-off of type 1 error rate and power than standard nominal threshold tests and not testing for sample sizes from 250 to 4000 and for 20 to 100,000 covariates. In network studies, meta-analysis of effect estimates must be accompanied by meta-analysis of the diagnostics or else systematic confounding may overwhelm the estimated effect. Our procedure for statistically testing balance at both the database level and the meta-analysis level achieves the best balance of type-1 error rate and power. Our procedure supports the review of large numbers of covariates, enabling more rigorous diagnostics.