Although pharmacogenomics can effectively guide antihypertensive treatment, relevant evidence from low-resource primary care settings is lacking. The aim of this study is to evaluate a pharmacogenomic-guided antihypertensive strategy in real-world, low-resource primary care settings.

We randomly assigned 94 villages (in 1:1:1 ratio, with stratification by villages) to receive a multifaceted intervention that included follow‐up, health education, and medication adjustment guided by pharmacogenomics (treatment) or physician's experience (control), or routine care (observation) over 4 weeks. The primary outcome was the proportion of patients with controlled blood pressure after 4 weeks of treatment. Generalized linear, linear mixed‐effect regression, and sensitivity analysis were conducted to assess the difference among groups. The intracluster correlation coefficient was calculated to assess the heterogeneity with cluster.

This study included 1031 hypertensive patients from 90 villages of Dajie and Jiuzhi towns in Daming County from May 10 to December 31, 2022. Overall, 79 loci on 39 antihypertensive‐related genes were tested for patients in the treatment group. After 4 weeks of treatment, 371 of 377 (98.4%) patients in the treatment group, 256 of 345 (74.2%) patients in the control group, and 232 of 309 (75.1%) patients in the observation group had their blood pressure under control. The between‐group net differences were 24.2% (95% confidence interval [CI]: 19.4%–29.0%) and 23.3% (95% CI: 18.3%–28.3%) for the treatment versus control groups and treatment versus observation groups, respectively. After 4 weeks of treatment, the proportion of patients with overall incident adverse events was no different among the three groups (all p > 0.05), and no serious adverse events or deaths related to antihypertensive treatment were reported during the study period.

A pharmacogenomic‐guided antihypertensive strategy can significantly enhance blood pressure control while maintaining a similar safety level compared to an experience‐guided antihypertensive strategy in low‐resource, primary care settings.

Large language models (LLMs) have shown considerable promise in supporting clinical decision‐making. However, their adoption and evaluation in dermatology remains limited. This study aimed to explore the preferences of Chinese dermatologists regarding LLM‐generated responses in clinical psoriasis scenarios and to assess how they prioritize key quality dimensions, including accuracy, traceability, and logicality.

A cross‐sectional, web‐based survey was conducted between December 25, 2024, and January 22, 2025, following the Checklist for Reporting Results of Internet E‐Surveys guidelines. A total of 1247 valid responses were collected from practicing dermatologists across 33 of China's provincial‐level administrative divisions. Participants evaluated responses to five categories of clinical questions (etiology, clinical presentation, differential diagnosis, treatment, and case study) generated by five LLMs: ChatGPT‐4o, Kimi.ai, Doubao, ZuoYiGPT, and Lingyi‐agent. Statistical associations between participant characteristics and model preferences were examined using chi‐square tests.

ChatGPT‐4o (Model 1) emerged as the most preferred model across all clinical tasks, consistently receiving the highest number of votes in case study (n = 740), clinical presentation (n = 666), differential diagnosis (n = 707), etiology (n = 602), and treatment (n = 656). Significant variation in model preference by professional title was observed only for the differential diagnosis task (χ² = 21.13, df = 12, p = 0.0485), while no significant differences were found across hospital tiers (p > 0.05). In terms of evaluation dimensions, accuracy was most frequently rated as “very important” (n = 635). A significant association existed between hospital tier and the most valued dimension (χ² = 27.667, df = 9, p = 0.0011), with dermatologists in primary hospitals prioritizing traceability more than their peers in higher‐tier hospitals. No significant associations were found across professional titles (p = 0.127).

Chinese dermatologists suggest a strong preference for ChatGPT‐4o over domestic LLMs in psoriasis‐related clinical tasks. While accuracy remains the primary criterion, traceability and logicality are also critical, particularly for clinicians in lower‐tier hospitals. These findings suggest that future clinical LLMs should prioritize not only content accuracy but also source transparency and structural clarity to meet the diverse needs of different clinical settings.

To evaluate sex‐related differences in the risk factors associated with nonhealing or recurrence of hyperthyroidism (NHRH) in patients with Graves' disease (GD) treated with radioactive iodine.

In total, 285 patients were enrolled. Data on radioactive iodine (RAI) dosage, ultrasound indexes of the thyroid, and other clinical factors were collected. Patients were divided into NHRH and non‐NHRH (hypothyroidism or euthyroidism) groups based on treatment outcomes. Univariate and multivariate weighted logistic regression analyses were used to identify factors associated with NHRH. Sex‐specific analyses of these risk factors were also conducted.

There were no significant differences between the two groups in terms of sex, thyroid shear wave elastography velocity values, or pretreatment serum free thyroxine (FT4) levels. Thyroid volume and age were independently associated with NHRH, with the odds of NHRH gradually decreasing as age increased. In subgroup analyses, both age and thyroid volume were independent risk factors for NHRH in female patients (p < 0.05), while in male patients, only FT4 was independently associated with NHRH (p < 0.05).

In patients of different sexes, the influence of thyroid volume, age, and FT4 on treatment outcomes exhibits distinct patterns.