Clinical research plays a critical role in testing the efficacy of new interventions as it brings drugs and medical devices out of the controlled environment of the laboratory and into the real world. Although some findings may be favorable, research can also yield inconclusive or negative results, and funding structures may shape whether these findings are published. A review studying forty-eight clinical trials found that “the number of studies with favorable results is approximately 24% higher among industry-sponsored studies compared with non–industry-sponsored studies.” (Bero & Dawson, 2010). There are numerous papers on ghost writing, or companies covertly writing academic papers, as a means to manage scientific research to aid the industry sponsor’s marketing of their product. Far fewer studies focus on publication bias as a means of ghost management and gaining commercial influence over the research process. Using a case study from ED research at a private, nonprofit hospital as an anchor, this paper argues that non-publication is not simply the result of industry suppressing unfavorable outcomes to protect itself, but rather a negotiated decision between industry sponsors and principal investigators that balances institutional interests and power dynamics. The analysis demonstrates that science in the modern era is inherently social, shaped by hierarchies and organizational priorities, rather than an isolated endeavor driven to dissipate knowledge for the greater good.

By 2050, the number of people aged 50+ living with at least one chronic condition is projected to double compared to 2020. Amongst older Asian American and Pacific Islander (AAPI) populations, these challenges are impacted by distinct cultural attitudes toward health and pain, making it critical to examine how chronic pain intersects with mental health. This study examines how chronic conditions amongst elderly AAPI populations contribute to increased loneliness and depressive symptoms. A questionnaire evaluating chronic conditions, physical pain, depression, loneliness, and attitudes towards mental health was designed and distributed to community health fair participants aged 50+. Survey data was organized into distinct variables and analyzed in the SAS software to draw correlations between measured variables and an individual’s physical and mental health. Depression scores positively correlated with loneliness scores (r=0.67490, p=<0.0001). The number of chronic conditions was also significantly associated with depression (p=0.0084), whereas pain level was not (p=0.2099). Neither chronic conditions nor pain level showed significant associations with loneliness (p=0.1057, p=0.1227). This study’s findings will provide a more holistic understanding between chronic conditions and mental wellbeing for researchers and the geriatric AAPI community. With encouraged reflection on their physical and mental health, participants may experience an increase in self-awareness and encourage them to seek medical or emotional support.

Background Stroke is the leading cause of adult disability in the United States characterized by lasting impairment on Activities of Daily Living (ADLs). The link between ADL deficits measured by GG scores and lesion location is unknown. Objective The goal of this study was to determine the associations between post-stroke ADL deficits and lesion characteristics. Methods Stroke patients admitted to inpatient rehabilitation were assigned GG scores at admission and discharge. Stroke lesions were outlined on patient neuroimaging scans and processed for analysis. Voxel Lesion Symptom Mapping (VLSM) was used to find associations between lesion location, and both admission and change in GG scores. Results Stroke patients with complete data (N = 78, 44% female, Mage = 68.73, SD = 14.97) were included in the final analysis. Median total admission GG scores was 52 (25.25). Median change in GG scores was 45 (17). There was a significant association between lower GG scores at admission and lesioned voxels in the sensorimotor cortex near the central sulcus, corpus callosum, corticospinal tract, internal capsule, and CPC tract (Z = -3.65, p < 0.05). VLSM revealed significant associations between higher change in GG scores and injury to the putamen, central opercular cortex (sensorimotor region), and insula (Z = 3.35, p < 0.05). Significance Our findings are supported by previous literature and indicate there are clusters within the brain associated with baseline and change in ADL function post stroke.

Stroke affects nearly 800,000 Americans each year, and despite advances in mechanical thrombectomy, over half of patients experience severe disability or death. First-pass effect (FPE), or successful single-attempt clot removal, is a strong predictor of recovery but is achieved in only 20-40% of cases. Clot composition is a major barrier to FPE: immune cell-rich clots are rigid and difficult to retrieve, whereas red blood cell-rich clots are softer and more responsive to aspiration. While hemodynamic forces are known to influence thrombus formation and leukocyte recruitment, how flow drives the formation of immune cell-rich clot phenotypes remains poorly understood. To address this gap, this project investigated how flow conditions influence clot formation and immune cell recruitment. Clots were generated in a Chandler Loop under different flow conditions, analyzed with Martius Scarlet Blue staining, and characterized using single-cell RNA sequencing. Our results demonstrated that flow alters clot composition and gene expression: clots formed under high-flow conditions showed greater proportions of fibrin and immune cells, particularly neutrophils, whereas clots formed under low-flow conditions showed higher red blood cell content. These findings clarify how flow environment shapes clot composition and suggest that flow-induced differences in clot makeup and gene expression may be crucial in identifying novel therapeutic targets for stroke in upstream processes of thrombus formation.

Background: Adverse drug effects are more common in geriatric patients with multiple chronic conditions because multimorbidity and polypharmacy increase the risk of drug-on-drug interactions and medication errors. Machine learning (ML) may improve medication safety by identifying high-risk patients and supporting more personalized management strategies. This review discusses data-driven tools used to detect patterns, predict complications, and guide safer prescribing decisions in clinical settings. Objectives: This review evaluates how ML tools have been used to predict adverse drug events in multimorbid older adults and to identify strengths and limitations in the literature. Methods: A scoping review following PRISMA guidelines was conducted using the PubMed database. The inclusion criteria included studies that utilized ML to reduce adverse drug events for patients with multimorbidity and polypharmacy. A total of 48 studies met the inclusion criteria. Results: ML-based tools were effective in identifying potentially harmful medications and adverse drug interactions. Across the studies, ML algorithms demonstrated a moderate accuracy of around 70-75%. Key factors such as Drug Burden Index, mobility limitations, history of falls, and lifestyle factors were used by ML algorithms to identify high-risk patients. Conclusions: ML shows strong potential to improve medication safety in older adults. However, further studies need to be conducted due to limited sample sizes and limited long-term outcome evidence in current studies.