Tag: nevin manimala

Fam Med. 2025 Apr;57(4):286-291. doi: 10.22454/FamMed.2025.925429.

ABSTRACT

BACKGROUND AND OBJECTIVES: The Council of Academic Family Medicine Educational Research Alliance (CERA) is a unique collaboration of academic family medicine organizations (Society of Teachers of Family Medicine [STFM], Association of Family Medicine Residency Directors, North American Primary Care Research Group, Association of Departments of Family Medicine) that facilitates and improves educational research in family medicine. CERA conducts approximately five surveys per year, including residency program directors, clerkship directors, department chairs, and general membership. Members of these organizations propose modules of 10 questions for these surveys. Proposals are peer-reviewed, and the top proposals are incorporated, along with standardized demographic questions, into an omnibus survey. We sought to determine the impact of self-reported gender of the primary submitter on survey module acceptance rates.

METHODS: We conducted a bibliometric analysis to explore author characteristics and quantify dissemination efforts. We conducted ꭓ2 analyses to determine gender differences in proposal acceptance. We used the exact binomial test to compare proportions of women authors to the benchmark proportion of women in STFM.

RESULTS: Overall, women submitted 66% (460/699) of CERA survey module proposals and authored 65% of accepted CERA modules (157/241) with the highest proportion concentrated among Clerkship Surveys (73%, 40/55). The acceptance rate did not differ significantly by gender (χ2=0.07, df=1, P=.80). A total of 73.4% (177/241) of module authors went on to present or publish their findings; we found no significant differences in scholarly output by gender (χ2=0.70, df=1, P=.41).

CONCLUSIONS: These findings indicate that the CERA module submission process has been successful in achieving comparable acceptance rates for men and women submitters. Other specialties should consider a similar model as a means to support early career educational researchers, including women.

PMID:40272837 | DOI:10.22454/FamMed.2025.925429

JAMA Ophthalmol. 2025 Apr 24. doi: 10.1001/jamaophthalmol.2025.0834. Online ahead of print.

ABSTRACT

IMPORTANCE: Recently, it was proved that the large language model Generative Pre-trained Transformer 4 (GPT-4; OpenAI) can fabricate synthetic medical datasets designed to support false scientific evidence.

OBJECTIVE: To uncover statistical patterns that may suggest fabrication in datasets produced by large language models and to improve these synthetic datasets by attempting to remove detectable marks of nonauthenticity, investigating the limits of generative artificial intelligence.

DESIGN, SETTING, AND PARTICIPANTS: In this quality improvement study, synthetic datasets were produced for 3 fictional clinical studies designed to compare the outcomes of 2 alternative treatments for specific ocular diseases. Synthetic datasets were produced using the default GPT-4o model and a custom GPT. Data fabrication was conducted in November 2024.

EXPOSURE: Prompts were submitted to GPT-4o to produce 12 “unrefined” datasets, which underwent forensic examination. Based on the outcomes of this analysis, the custom GPT Synthetic Data Creator was built with detailed instructions to generate 12 “refined” datasets designed to evade authenticity checks. Then, forensic analysis was repeated on these enhanced datasets.

MAIN OUTCOMES AND MEASURES: Forensic analysis was performed to identify statistical anomalies in demographic data, distribution uniformity, and repetitive patterns of last digits, as well as linear correlations, distribution shape, and outliers of study variables. Datasets were also qualitatively assessed for the presence of unrealistic clinical records.

RESULTS: Forensic analysis identified 103 fabrication marks among 304 tests (33.9%) in unrefined datasets. Notable flaws included mismatch between patient names and gender (n = 12), baseline visits occurring during weekends (n = 12), age calculation errors (n = 9), lack of uniformity (n = 4), and repetitive numerical patterns in last digits (n = 7). Very weak correlations (r < 0.1) were observed between study variables (n = 12). In addition, variables showed a suspicious distribution shape (n = 6). Compared with unrefined datasets, refined ones showed 29.3% (95% CI, 23.5%-35.1%) fewer signs of fabrication (14 of 304 statistical tests performed [4.6%]). Four refined datasets passed forensic analysis as authentic; however, suspicious distribution shape or other issues were found in others.

CONCLUSIONS AND RELEVANCE: Sufficiently sophisticated custom GPTs can perform complex statistical tasks and may be abused to fabricate synthetic datasets that can pass forensic analysis as authentic.

PMID:40272814 | DOI:10.1001/jamaophthalmol.2025.0834

JAMA. 2025 Apr 24. doi: 10.1001/jama.2025.4605. Online ahead of print.

ABSTRACT

IMPORTANCE: Children living in poverty are at increased risk of mental health and neurodevelopmental disorders. Little is known about the trends in diagnoses of these disorders among children enrolled in public insurance programs, such as Medicaid, which insure more than 1 in 3 US children.

OBJECTIVE: To provide comprehensive, multistate estimates of changes in the percentage of publicly insured children with mental health and/or neurodevelopmental disorder diagnoses.

DESIGN, SETTING, AND PARTICIPANTS: This serial, cross-sectional study used administrative claims data from 22 states to test trends from 2010 to 2019 in the percentage of publicly insured children aged 3 to 17 years with mental health or neurodevelopmental disorder diagnoses. Regression models included a dummy variable for each year, controlled for child demographics, county-level metropolitan status, median household income, and US Census region. Adjusted risk differences were estimated, with standard errors clustered at the state level.

EXPOSURE: Calendar year.

MAIN OUTCOMES: Any mental health or neurodevelopmental disorder diagnosis in the calendar year, and any diagnosis in 1 of 13 specific diagnostic categories.

RESULTS: A total of 129 306 637 child-year observations (29 925 633 unique publicly insured children) were included. The percentage of publicly insured children with any diagnosed mental health or neurodevelopmental disorder increased from 10.7% in 2010 to 16.5% in 2019; this change remained significant after adjustment for covariates (adjusted risk difference [aRD], 6.7 percentage points [95% CI, 5.0-8.4]). Statistically significant increases were also observed in 9 of the 13 diagnostic categories examined. The largest absolute increases were observed for attention-deficit/hyperactivity disorder (aRD, 2.3 percentage points [95% CI, 1.4-3.3]), trauma- and stressor-related disorders (aRD, 1.7 percentage points [95% CI, 0.9-2.5]), anxiety disorders (aRD, 1.6 percentage points [95% CI, 1.2-2.1]), autism spectrum disorders (aRD, 1.1 percentage points [95% CI, 0.9-1.4]), depressive disorders (aRD, 0.9 percentage points [95% CI, 0.6-1.3]), and other neurodevelopmental disorders (aRD, 2.6 percentage points [95% CI, 1.8-3.5]).

CONCLUSIONS AND RELEVANCE: The percentage of publicly insured children receiving any mental health or neurodevelopmental disorder diagnosis significantly increased between 2010 and 2019, with increases observed for most diagnostic categories examined. These findings highlight the need for access to appropriate services in safety net systems and other settings that serve this population.

PMID:40272810 | DOI:10.1001/jama.2025.4605

JAMA Netw Open. 2025 Apr 1;8(4):e256852. doi: 10.1001/jamanetworkopen.2025.6852.

ABSTRACT

IMPORTANCE: Guidelines recommend annual imaging surveillance after endovascular abdominal aortic aneurysm repair (EVAR). How these guidelines translate into practice among veterans remains poorly described.

OBJECTIVE: To characterize post-EVAR surveillance among veterans.

DESIGN, SETTING, AND PARTICIPANTS: This retrospective cohort study evaluated veterans who underwent EVAR between January 1, 2000, and December 31, 2023, at US Department of Veterans Affairs (VA) hospitals and received follow-up care at VA and non-VA hospitals and imaging centers with payment via Medicare or the VA. Veterans treated with EVAR in VA hospitals during the study period were included.

EXPOSURE: Years after EVAR.

MAIN OUTCOMES AND MEASURES: The primary outcome was annual surveillance adherence, measured as 1 or more imaging studies in the abdomen or pelvis each year after EVAR. Stepwise logistic regression modeling was used to determine factors associated with poor adherence. Secondary outcomes were imaging type (cross-sectional, ultrasonography, or ultrasonography followed by cross-sectional imaging) and factors associated with lower adherence.

RESULTS: The cohort included 27 792 veterans (27 624 male [99.4%]; 22 521 aged ≥65 years [81.0%]). Mean (SD) follow-up was 6.0 (4.0) years. The mean (SD) proportion of time that veterans were surveillance adherent was 71.1% (28.5%). Surveillance was initially high, with 25 026 of 27 792 veterans (90.0%) undergoing surveillance imaging in year 1 after EVAR. However, this proportion decreased further out from EVAR, with 12 401 of 21 384 veterans (58.0%) undergoing surveillance imaging by year 4 after EVAR. Veterans were most likely to undergo imaging with computed tomography scans (21 911 veterans [78.8%]). However, the proportion with surveillance via ultrasonography alone increased from 823 of 25 026 veterans (3.3%) in year 1 after EVAR to 2567 of 12 401 veterans (20.7%) in year 4 after EVAR. White race (odds ratio [OR] vs all other racial groups, 0.84; 95% CI, 0.72-0.98), married status (OR vs all other social status categories, 0.80; 95% CI, 0.71-0.89), having a service-connected disability (OR, 0.69; 95% CI, 0.62-0.77), and a higher Charlson Comorbidity Index score (OR per 1-unit increase, 0.93; 95% CI, 0.91-0.95) were associated with lower odds of poor surveillance adherence.

CONCLUSIONS AND RELEVANCE: In this study, post-EVAR imaging surveillance was high, although surveillance lapses were more likely further out from EVAR and for patients with certain characteristics. This information may inform future patient-centered efforts to improve post-EVAR imaging adherence.

PMID:40272801 | DOI:10.1001/jamanetworkopen.2025.6852

JAMA Netw Open. 2025 Apr 1;8(4):e257115. doi: 10.1001/jamanetworkopen.2025.7115.

ABSTRACT

IMPORTANCE: Mental and somatic health is often impaired among transgender persons. Studies regarding coexisting mental and somatic health outcomes among transgender persons are limited.

OBJECTIVE: To assess health diagnoses and medicine use among transgender persons compared with cisgender controls.

DESIGN, SETTING, AND PARTICIPANTS: This register-based national cohort study included data from January 1, 2000, to December 31, 2021. Transgender persons were included on the first date of receipt of a transgender identity contact code. Controls included 10 age-matched cisgender men (n = 5) and women (n = 5) for each transgender person. Statistical analyses were conducted from September to December 2024.

MAIN OUTCOMES AND MEASURES: The main outcomes were International Statistical Classification of Diseases and Related Health Problems, Tenth Revision diagnosis codes and medicine use in a 5-year period up to the first date of transgender contact code for most commonly occurring mental and physical illnesses. The main outcomes were determined after data collection.

RESULTS: The cohort included 3812 transgender persons (1993 transmasculine persons [52.3%] with a median age of 19 years [IQR, 15-24 years] and 1819 transfeminine persons [47.7%] with a median age of 23 years [IQR, 19-33 years]) and 38 120 cisgender controls. The odds for a mental health diagnosis was up to 12 times higher among transgender persons compared with cisgender controls. Among transmasculine and transfeminine persons, neurotic, stress-related disorders (transmasculine: adjusted odds ratio [AOR], 4.70 [95% CI, 4.02-5.50]; transfeminine: AOR, 5.27 [95% CI, 4.28-6.49]); developmental disorders, including autism (transmasculine: AOR, 11.67 [95% CI, 8.85-15.39]; transfeminine: AOR, 9.39 [95% CI, 7.05-12.50]); mood (affective) disorders (transmasculine: AOR, 5.41 [95% CI, 4.32-6.77]; transfeminine: AOR, 5.61 [95% CI, 4.16-7.57]); and behavioral disorders (transmasculine: AOR, 4.50 [95% CI, 3.61-5.62]; transfeminine: AOR, 4.15 [95% CI, 3.19-5.39]) were the most frequent mental health diagnoses compared with cisgender controls of the opposite sex at birth. Transmasculine persons had higher odds for somatic diagnosis codes of diabetes (AOR, 2.00 [95% CI, 1.12-3.56]), asthma (including chronic obstructive lung disease; AOR, 1.40 [95% CI, 1.06-1.85]), injury and poisoning (AOR, 1.28 [95% CI, 1.15-1.41]), and pain (AOR, 1.29 [95% CI, 1.12-1.49]) compared with control cisgender women. Among transfeminine persons, somatic diagnosis codes of infection (AOR, 1.68 [95% CI, 1.33-2.13]), anemia (AOR, 3.08 [95% CI, 1.36-6.97]), diabetes (AOR, 1.95 [95% CI, 1.25-3.05]), sleep apnea (AOR, 3.41 [95% CI, 1.84-6.31]), and pain (AOR, 1.31 [95% CI, 1.08-1.58]) were more frequent compared with control cisgender men. Transgender persons had higher use of psychopharmacologic medicine, antacids, and laxatives compared with cisgender controls (transmasculine persons vs control cisgender women, antipsychotics: AOR, 6.20 [95% CI, 5.07-7.59]; hypnotics-sedatives: AOR, 4.45 [95% CI, 3.78-5.23]; antacids: AOR, 1.25 [95% CI, 1.07-1.45]; and laxatives: AOR, 1.53 [95% CI, 1.17-1.99]; transfeminine persons vs control cisgender men, antipsychotics: AOR, 4.74 [95% CI, 3.92-5.74]; hypnotics-sedatives: AOR, 3.01 [95% CI, 2.53-3.57]; and antacids: AOR, 1.32 [95% CI, 1.12-1.56]). Mental health diagnoses and use of psychopharmacologic drugs were coexisting with somatic diagnoses and use of drugs for somatic diseases.

CONCLUSIONS AND RELEVANCE: This cohort study of Danish transgender persons and cisgender controls found significantly higher risks for mental and somatic health diagnoses among transgender persons. Coexistence of mental health outcomes and somatic health outcomes among transgender persons could be associated with stress encountered due to belonging to a gender identity or sexual orientation minority group; mental and physical morbidity should be considered an integrated part of transgender care.

PMID:40272800 | DOI:10.1001/jamanetworkopen.2025.7115

JAMA Netw Open. 2025 Apr 1;8(4):e257202. doi: 10.1001/jamanetworkopen.2025.7202.

ABSTRACT

IMPORTANCE: Bloodstream infections (BSIs) associated with peripheral intravenous catheters (PIVCs) are rare but preventable adverse events. The association of dwell time with the risk of BSIs with PIVCs remains controversial.

OBJECTIVE: To analyze the risk of BSIs during PIVC maintenance therapy.

DESIGN, SETTING, AND PARTICIPANTS: In this observational cohort study, all patients hospitalized at Geneva University Hospitals with at least 1 PIVC insertion on the upper extremity (N = 371 061) between January 1, 2016, and February 29, 2020, were evaluated. Statistical analysis was performed from January 2023 to January 2025.

EXPOSURE: At least 1 PIVC insertion on the upper extremity.

MAIN OUTCOMES AND MEASURES: The primary outcome was BSIs with PIVCs; data were collected by prospective BSI surveillance. The daily risk of BSIs with PIVCs was analyzed using the hazard rate function by kernel-based methods. Multivariable logistic models were performed to evaluate the risk of BSIs with PIVCs comparing different cutoff values of dwell times (>3 vs ≤3 days, >4 vs ≤4 days, >5 vs ≤5 days, and >6 vs ≤6 days).

RESULTS: A total of 371 061 PIVCs (median patient age, 63 years [IQR, 41-79 years]; 187 786 women [51%]) with documented catheter duration were included. A total of 140 178 PIVCs (38%) had a dwell time of 1 to 2 days, 119 252 (32%) had a dwell time of 3 to 4 days, and 111 631 (30%) had a dwell time of more than 4 days. The instantaneous risk of BSIs with PIVCs was low in the first 2 days of dwell time and increased rapidly thereafter. The risk of BSIs was significantly increased after 3 days of catheter maintenance (adjusted odds ratio [AOR], 13.55; 95% CI, 5.44-34.00). This risk was the highest after 3 days and remained increased thereafter (>4 days: AOR, 8.53; 95% CI, 4.47-16.28; >5 days: AOR, 5.38; 95% CI, 3.23-8.96; and >6 days: AOR, 7.63; 95% CI, 4.57-12.74).

CONCLUSIONS AND RELEVANCE: In this cohort study of 371 061 PIVCs, dwell time was associated with the development of BSIs with PIVCs. After day 3, PIVC indication should be reviewed and PIVC replacement considered.

PMID:40272799 | DOI:10.1001/jamanetworkopen.2025.7202

Facial Plast Surg Aesthet Med. 2025 Apr 24. doi: 10.1089/fpsam.2024.0250. Online ahead of print.

ABSTRACT

Background: Chemodenervation is a mainstay in the treatment of nonflaccid facial paralysis (NFFP). This study investigates how electromyography (EMG) guidance during chemodenervation may allow for more accurate needle placement. Study Objective: To compare the accuracy with needle EMG chemodenervation compared with standard non-EMG approaches in patients receiving treatment for NFFP with botulinum toxin (BT) as measured by physician and EMG verification. Methods: The use of needle EMG across three surgeons was recorded to determine if the use of EMG changed the location, angle, or depth of needle insertion into the targeted facial muscle. The determination of accurate needle placement was self-reported by the physician coupled with the physician assessment of muscle EMG activity. Accuracy was determined as the ratio of the number of successful EMG needle insertions (placed within muscle with adequate signal) over the total number of EMG needle insertions. Results: A total of 137 patients underwent treatment for NFFP, with an average age of 54 (interquartile range [IQR]: 44-65) and weight of 73 kg (IQR: 64-90). Of the 14 facial mimetic muscles, the posterior belly of digastric (accuracy = 19%, standard deviation [SD] 0.4), risorius (accuracy = 46%, SD 0.5), and zygomaticus (accuracy = 47%, SD 0.5) had the lowest accuracy in BT placement (Table 3). Of all the muscles included, only the buccinator (p = 0.015), platysma (p = 0.0093), and zygomaticus (p = 0.024) had statistically significant variation in the accuracy of BT placement between surgeons. Conclusions: EMG guidance for the treatment of NFFP may improve the precision of needle placement and BT delivery, particularly in the mid and lower facial muscles.

PMID:40272786 | DOI:10.1089/fpsam.2024.0250

J Racial Ethn Health Disparities. 2025 Apr 24. doi: 10.1007/s40615-025-02424-7. Online ahead of print.

ABSTRACT

Discrepancies in healthcare outcomes and quality of treatment between White people and people of color are known and statistically relevant in the USA. The possible root causes of this issue, however, are not discussed enough, further enhancing an unequal access to proper healthcare and not acknowledging the role of racism and White supremacy in the consolidation of the medical field in the country. This article used discourse and document analysis to understand the role in which racism and narratives regarding Black bodies during the consolidation of medicine and gynecology had in the USA during the nineteenth century to create discrepancies in access and quality of care for Black patients today. The findings suggest that the dehumanizing narratives of Black patients in the past still impact the quality of care, creating higher mortality rates, worse prognosis, and lack of proper pain management, reinforcing health disparities.

PMID:40272774 | DOI:10.1007/s40615-025-02424-7

J Cancer Educ. 2025 Apr 24. doi: 10.1007/s13187-025-02630-1. Online ahead of print.

ABSTRACT

Esophageal cancer is a prevalent cancer worldwide, with incidence rates rising significantly in recent years, particularly in high-risk regions where case numbers continue to escalate. As one of the most visited social media platforms globally, TikTok has emerged as a key source for health information. This study aims to evaluate Chinese TikTok videos on esophageal cancer, focusing on content integrity, informational quality, source credibility, usefulness, and reliability.: We assessed 246 esophageal cancer-related videos using the Content Integrity Score based on Goobie’s coding scheme for content evaluation and the DISCERN instrument for assessing information reliability and treatment guidance. Videos were categorized by duration, source, publication year, and user engagement metrics (likes, comments, favorites, shares). Statistical analysis was conducted using Shapiro-Wilk, Kruskal-Wallis and post hoc Bonferroni tests. Additionally, Spearman’s correlation analysis was applied to explore relationships between key quantitative variables. Of the videos analyzed, 212(86.2%) were uploaded by doctors, and 168(68.3%) lacked content related to risk factors. and 217(88.2%) had a quality score rated as poor or below. User engagement metrics, including likes, comments, saves, and shares, exhibited strong intercorrelations. However, only shares showed a weak correlation with DISCERN scores (R = 0.134, p = 0.036). Additionally, DISCERN scores were positively correlated with video duration (R = 0.352, p < 0.001). Chinese TikTok videos on esophageal cancer generally offer low-quality information that lacks accuracy and is insufficient for guiding patients in making informed health decisions. Given the low quality and reliability of such content, TikTok is not a suitable source for patient education.

PMID:40272765 | DOI:10.1007/s13187-025-02630-1

Cancer Causes Control. 2025 Apr 24. doi: 10.1007/s10552-025-01998-1. Online ahead of print.

ABSTRACT

PURPOSE: Acute leukemia is the most common childhood malignancy, with suspected contributions from environmental factors and immune responses to common pathogens. A recent meta-analysis indicated possible spatiotemporal clustering, though the findings were hindered by data quality limitations. We investigated spatial and spatiotemporal clustering of childhood leukemia using advanced methods and complete residential histories.

METHODS: We included patients aged 0-17 years diagnosed in 1990-2019, using data from the Finnish Cancer Registry. A 1:3 age- and sex-matched case-control design was employed and residential history data with exact coordinates was collected. Clustering was evaluated using the Cuzick-Edwards test, Knox test, Kulldorff’s scan statistic, and Jacquez’s Q statistic.

RESULTS: The dataset included 1,626 childhood leukemia cases (median age 5.0 years, 54% male). The Knox test revealed no evidence of spatiotemporal clustering. However, the Cuzick-Edwards test revealed spatial clustering at diagnosis addresses for children under 1 year (OR 1.35, 95% CI 1.14-1.57). Further analysis with Jacquez’s Q test using complete residential histories identified significant spatiotemporal clustering in young children (ages 1.5-5.99 years) with acute lymphoblastic leukemia (ALL, p = 0.037). We also tested for co-incidence between leukemia and type 1 diabetes but found no clustering.

CONCLUSION: Overall, we found limited evidence for clustering. In the subgroup analyses, significant spatiotemporal clustering in acute lymphoblastic leukemia cases among children aged 1.5-5.99 years was observed, coinciding with the peak incidence in early childhood. Previous research has shown that this age group has distinct genetic characteristics and may possess a unique etiology.

PMID:40272762 | DOI:10.1007/s10552-025-01998-1