Tag: statistics

JMIR Med Inform. 2026 Apr 29;14:e84095. doi: 10.2196/84095.

ABSTRACT

BACKGROUND: Brain tumor is one of the most malignant diseases of the central nervous system, and early accurate detection is of great significance for improving patient survival rate. However, the heterogeneity of brain tumors in terms of morphology, size, and location on magnetic resonance imaging (MRI) image, as well as their similarity to surrounding normal brain tissue, poses significant challenges for tumor detection.

OBJECTIVE: This study aims to develop a high-performance brain tumor detection framework that integrates feature enhancement, channel attention, and progressive pruning, achieving an optimal balance between detection accuracy, model efficiency, and interpretability for slice-level MRI tumor localization tasks.

METHODS: This paper proposes a convolution Prewitt-and-pooling-based preprocessing (CSPP) approach, based on the “you only look once” version 11 (YOLOv11) framework, which highlights important structural detail more effectively than traditional statistics. A dynamic convolution-based C3k2 (DCC) module was integrated to more efficiently capture both local and global features. A channel prior convolutional attention (CPCA) module was introduced before the detection head, enabling the network to specifically focus on information-rich channels and key spatial regions. Through a progressive hybrid pruning strategy (PHPS), the model was optimized for efficient inference. Furthermore, Eigen-class activation mapping (Eigen-CAM) was used to interpret the prediction result, making them more transparent.

RESULTS: Extensive experiments on 3 brain tumor MRI datasets demonstrated the superior performance of CDCP-YOLO (CSPP-DCC-CPCA-PHPS-YOLO). On Br35H, the mean average precision (mAP) at an intersection-over-union (IoU) threshold of 0.5 (mAP0.5) increased by 2.6%, average mAP over several IoU thresholds (0.50-0.95; mAP0.5:0.95) increased by 5.9%, and number of floating-point operations (×10⁹; GFLOPs) decreased by 47.7%. On Roboflow, mAP0.5 increased by 19.5%, mAP0.5:0.95 increased by 7.7%, and GFLOPs decreased by 47.7%. On Capstone, mAP0.5 increased by 6.9%, mAP0.5:0.95 increased by 5.8%, and GFLOPs decreased by 47.7%.

CONCLUSIONS: The proposed CDCP-YOLO framework achieves an optimal balance between accuracy, efficiency, and interpretability, providing a lightweight and reliable solution for slice-level brain tumor detection in MRI images.

PMID:42054652 | DOI:10.2196/84095

JMIR Hum Factors. 2026 Apr 29;13:e89126. doi: 10.2196/89126.

ABSTRACT

BACKGROUND: Motivational interviewing (MI) is a patient-centered communication approach that supports health behavior change; yet, its integration into undergraduate medical curricula remains inconsistent. Combined learning models that comprise face-to-face instruction with structured web-based components may strengthen MI training, but evidence supporting their effectiveness among medical students, particularly in Asian contexts, is limited.

OBJECTIVE: This study evaluated the impact of a combined MI educational model on final-year medical students’ MI knowledge, confidence, and application in real patient encounters during clinical rotations.

METHODS: This study used a sequential explanatory mixed methods design. The quantitative component used a before-and-after study to evaluate changes in MI knowledge and confidence among final-year medical students enrolled in an Ambulatory Care course in 2024. All 130 students participated in a 2-hour interactive MI workshop, and 120 completed pre- and postintervention questionnaires assessing MI knowledge and self-reported confidence. Students were also provided access to a 3-hour web-based MI learning module, and learning-management system analytics were used to track engagement. The qualitative component consisted of semistructured interviews with 12 purposively selected students, conducted to explore their experiences applying MI during clinical encounters. Quantitative data were analyzed using paired-samples t tests, and qualitative data were analyzed using inductive conventional content analysis. Findings from both components were integrated during interpretation to provide a comprehensive understanding of the educational intervention.

RESULTS: Students demonstrated a significant improvement in MI knowledge following the educational intervention (pretest mean 8.87, SD 2.69; posttest mean 15.04, SD 2.99; t₁₁₉=-18.45; P<.001; η²=0.74). After the workshop, 96.9% (126/130) of students reported applying MI with patients, and 92.3% (n=120) agreed that the combined learning approach was adequate for supporting clinical use. Learning analytics data showed that 76.9% (n=100) of students enrolled in the web-based MI module, and 51% (n=51) completed all lessons. Students most frequently applied MI when counseling patients with diabetes, hypertension, and dyslipidemia, especially related to diet, physical activity, and medication adherence. Interview findings indicated that students mainly used brief MI, were most comfortable with engaging and focusing, and developed greater empathy, confidence, and patient-centered communication skills. Challenges included limited time during consultations, clinical workload, and difficulty applying all MI processes to complex cases.

CONCLUSIONS: A combined MI learning approach integrating a short workshop with a web-based course was associated with higher MI knowledge scores and greater self-reported confidence among students, as well as reported use of MI-informed communication strategies during clinical encounters. Students perceived MI as a practical and ethically grounded communication approach that can enhance patient engagement, particularly in the management of chronic diseases. Introducing MI training longitudinally through a spiral curriculum, with opportunities for repeated practice and reinforcement, may help strengthen behavior-change communication competencies in undergraduate medical education.

PMID:42054651 | DOI:10.2196/89126

JMIR Public Health Surveill. 2026 Apr 29;12:e88992. doi: 10.2196/88992.

ABSTRACT

BACKGROUND: It is important to understand the motives behind why people want to reduce their alcohol consumption to inform messaging for public health campaigns.

OBJECTIVE: This study aimed to identify the motives for reducing alcohol consumption among users of the Drink Less app in the United Kingdom.

METHODS: Content analysis of Drink Less app users’ responses to the prompt “I want to drink less because…” Users were aged 18 years or older, lived in the United Kingdom, and had downloaded the app between May 20, 2016, and June 24, 2024 (n=2520). Inductive content analysis was conducted to analyze users’ motives to drink less, and the frequencies were stratified by age, sex, and Alcohol Use Disorders Identification Test (AUDIT) score categories.

RESULTS: The most common motives to drink less were wanting to improve their physical health (1329/2520, 52.7%), feel better in their body (823/2520, 32.7%), improve their mental well-being (567/2520, 22.5%), regain agency (483/2520, 19.2%), live a different life (321/2520, 12.7%), and have better relationships (309/2520, 12.3%). The motives for drinking less,”improve physical health” and “feel better in their body,” had a lower prevalence among users in higher AUDIT risk zones compared with low-risk, while wanting to “improve their mental well-being,” “regain agency,” “live a different life,” and “have better relationships” had a higher prevalence among users in higher AUDIT risk zones.

CONCLUSIONS: Users of an alcohol reduction app in the United Kingdom most commonly reported wanting to improve their physical health, feel better in their bodies, and improve their mental well-being as their motives for drinking less alcohol. The pattern of prevalence of different motives varied by AUDIT risk zones, highlighting the importance of tailoring interventions based on what is most likely to be motivating for individuals.

PMID:42054642 | DOI:10.2196/88992

JMIR Public Health Surveill. 2026 Apr 29;12:e82792. doi: 10.2196/82792.

ABSTRACT

BACKGROUND: Although recovery is a central tenet of the US substance use disorder service delivery system, empirical research on youth recovery remains limited and underdeveloped. Notably, no population-based representative surveys, either in the United States or internationally, currently assess recovery status among secondary school-aged youth (aged 14-18 years). Consequently, little is known about how many youth identify as being in recovery or about their characteristics and needs.

OBJECTIVE: This study presents the first statewide representative estimate of adolescent self-reported recovery (ASR), derived from a large Midwestern state in the United States.

METHODS: We used data from the 2024 Illinois Youth Survey, a weighted, statewide representative survey of students from 8th, 10th, and 12th grades across Illinois. We examined the prevalence of ASR with a widely used single-item question, “Do you consider yourself to be in recovery?” The question was presented after an instruction directing students to consider only substance use when responding. We estimated the prevalence of ASR and conducted descriptive analyses to characterize this group.

RESULTS: Among the 6871 participating students from the 10th and 12th grades, the prevalence of ASR was 3.3% (95% CI 2.6%-4.1%). Among participants with ASR, 51.1% (118/231) were female, 39% (90/231) identified as Latino or Latina, 38.1% (88/231) identified as White, and 13% (30/231) identified as Black or African American. The average age of participants with ASR was 16.5 (SD 1.14) years. Participants with ASR were demographically diverse, and a little over half received free or reduced-price lunch.

CONCLUSIONS: Findings suggest that financial recovery capital may be particularly important for participants with ASR. This study provides the first population-based estimate of the prevalence of ASR and underscores the importance of including recovery status in large-scale surveys to inform and strengthen recovery support systems.

PMID:42054636 | DOI:10.2196/82792

J Med Internet Res. 2026 Apr 29;28:e81485. doi: 10.2196/81485.

ABSTRACT

BACKGROUND: Podcasts provide a platform for delivering patient information. They have the potential to enhance patient engagement in shared decision-making (SDM), reduce anxiety in relation to hospital visits, and improve patient satisfaction. However, their impact on these outcomes in the context of hospital visits remains underexplored.

OBJECTIVE: This study aimed to examine whether podcasts influence patients’ (1) engagement in SDM, (2) anxiety after the hospital visit, and (3) satisfaction with the hospital visit.

METHODS: A quasi-experimental design with a nonequivalent comparison group was used. The study was conducted in 3 specialized outpatient clinics at a Danish hospital. Patients were allocated to one of 2 groups: the intervention group, which received access to informational podcasts in addition to standard written information before their hospital visit, and the comparison group, which received only the standard written information. All patients received validated questionnaires to assess SDM (9-item Shared Decision-Making Questionnaire [SDM-Q-9]), anxiety (State-Trait Anxiety Inventory-State), and satisfaction after the consultation.

RESULTS: A total of 240 patients participated. Compared with the control group, the intervention group showed a 15% higher level of SDM (SDM-Q-9) scores (adjusted relative difference=1.15, 95% CI 1.05-1.26; P=.18). Subgroup analyses indicated a statistically significant effect among patients with low health literacy (adjusted relative difference=1.81, 95% CI 1.42-2.32; P=.002). Anxiety scores were 9% lower (adjusted relative difference=0.91, 95% CI 0.84-0.99; P=.23), and satisfaction with previsit information increased by 14% (adjusted relative difference=1.14, 95% CI 1.07-1.21; P=.003).

CONCLUSIONS: Informational podcasts, provided as a supplement to traditional written information, may offer modest support for patient engagement in SDM, particularly among patients with low health literacy. Podcasts also appear to improve satisfaction with previsit information more broadly. Effects on previsit anxiety were inconclusive.

PMID:42054634 | DOI:10.2196/81485

JCO Oncol Pract. 2026 Apr 29:OP2500963. doi: 10.1200/OP-25-00963. Online ahead of print.

ABSTRACT

PURPOSE: It has been theorized that antibiotic therapy (ABT) affects response to immune checkpoint inhibition (ICI) by inducing dysbiosis of the gut microbiome (GM). To investigate the association between ABT and real-world overall survival (rwOS)/progression-free survival (rwPFS) in patients with metastatic renal cell carcinoma (mRCC) receiving ICI versus tyrosine kinase inhibitors (TKIs).

METHODS: In total, 5,237 patients with mRCC from a nationwide electronic health record-derived deidentified database who received ICI or TKI first-line after diagnosis were included. ABT exposure was stratified by exposure (yes or no), timing (before v after treatment initiation v none), excretion modes (hepatic v renal excretion v none), and administration routes (oral v intravenous v none). Three-month landmark Kaplan-Meier estimation and log-rank tests were used to compare rwOS/rwPFS among ABT groups. Multivariable Cox proportional hazards models with time-varying coefficients investigated the association between rwPFS, rwOS, ABT, and treatment modality.

RESULTS: ABT exposure was negatively associated with rwOS/rwPFS in both ICI (rwOS [23.9 v 33.6 months, P = .029]; rwPFS [8.8 v 11.6 months, P < .001]) and TKI (rwOS [17.4 v 26.2 months, P < .001]; rwPFS [8.0 v 9.7 months, P < .001]) recipients. For ICI patients only, a negative correlation between ABT after treatment initiation (rwOS, P = .003, rwPFS <0.001) and oral administration route (rwOS P = .004, rwPFS P = .001) was identified. In time-varying Cox proportional models, the effect of ABT on rwPFS beyond 12 months was only statistically significant in ICI patients (ICI, hazard ratio [HR], 1.67, P = .013; TKI, HR, 0.95; P = .7).

CONCLUSION: In our observational study, we identified a potential unique and complex association between ABT and rwOS/rwPFS in patients with mRCC receiving ICI. We found a negative correlation between ABT use after treatment initiation or via the oral route on oncologic outcomes in ICI patients. Moreover, there appears to be an ICI-specific negative association of ABT on rwPFS beyond 1 year. Our findings are associative, but they emphasize the importance of antibiotic stewardship in this space.

PMID:42054627 | DOI:10.1200/OP-25-00963

Eur J Public Health. 2026 Apr 11;36(3):ckag064. doi: 10.1093/eurpub/ckag064.

ABSTRACT

The COVID-19 pandemic has changed work organization in hospitals, notably with the expansion of telework. This study aimed to assess perceptions of telework among hospital workers throughout the pandemic and to identify factors associated with telework perceptions in the post-pandemic period. An observational cross-sectional study was conducted from October to December 2023. All hospital workers, regardless of their occupation or status, were invited to participate in an online survey. Perceptions of telework were assessed using visual analog scales, ranging from 0 (very negative) to 100 (very positive), across three periods: before the COVID-19 pandemic, during the first French lockdown, and after the pandemic. A total of 882 hospital workers were included in the analysis. Throughout the pandemic, 41.4% reported adopting telework. Overall, perceptions of telework became significantly more positive over time, rising from a mean score of 54.3 ± 25.3 before the pandemic, to 62.7 ± 27.4 during the first lockdown, and to 66.0 ± 27.5 after the pandemic (P < .001). Experiencing telework for the first time was associated with more favorable perceptions, with high levels sustained among those who continued teleworking post-pandemic. Perceptions of telework improved significantly among hospital workers throughout the pandemic with notable shifts in work practices, with nearly one third of participants teleworking in the post-pandemic period. Experiencing telework was associated with more positive perceptions.

PMID:42054081 | DOI:10.1093/eurpub/ckag064

Eur J Public Health. 2026 Apr 11;36(3):ckag032. doi: 10.1093/eurpub/ckag032.

ABSTRACT

China’s “one-child policy” limited many households in China to only one child. This policy had an impact on birth outcomes due to the birth order effects, as firstborn infants typically have lower birth weights. This study aimed to estimate the impact of the “universal two-child policy” on birth weight in China by analyzing individual-level data collected from a major tertiary obstetrics hospital located in Shanghai, the largest metropolitan area in China. Medical records for all births were obtained from a major metropolitan obstetrics hospital between 2013 and 2018. Using difference-in-differences (DID) and quantile DID (QDID) methods while controlling for maternal characteristics and socioeconomic factors, we examined the policy’s impact on birth weight. Analyses included stratification by maternal migrant status, age, and delivery mode. Insurance was found to mediate the treatment effect significantly. Analysis of 133 358 live births showed the policy increased birth weight by 21 g, corresponding to approximately 0.04 standard deviations of birth weight in our sample, with effects varying across maternal age groups and residency status. Insurance coverage mediated 41.3% of the total effect on birth weight. The “universal two-child policy” demonstrated beneficial impact on birth weight in China during the study period, particularly affecting older women, Shanghai residents, and those with natural births.

PMID:42054080 | DOI:10.1093/eurpub/ckag032

JAMA Netw Open. 2026 Apr 1;9(4):e269091. doi: 10.1001/jamanetworkopen.2026.9091.

ABSTRACT

IMPORTANCE: Patients with inflammatory bowel disease (IBD) are at increased risk of major adverse cardiovascular events (MACE), driven by chronic inflammation, endothelial dysfunction, and use of certain therapeutic regimens. Whether different IBD treatments mitigate this risk remains unclear.

OBJECTIVE: To examine the association of the use of immunomodulators or biologics vs 5-aminosalicylic acid (5-ASA) with risk of MACE among older patients with IBD.

DESIGN, SETTING, AND PARTICIPANTS: This comparative effectiveness study was conducted among a 15% Medicare claims sample of patients with IBD aged 65 years or older was identified, with entry defined by the first prescription of immunomodulators, biologics, or 5-ASA between January 1, 2012, and December 31, 2020. Follow-up extended to the outcome, switch or discontinuation of study drug, the study’s administrative end, or up to 3 years, whichever occurred first. Propensity score matching at a 1:3 ratio balanced demographics, comorbidities, and medication use between groups. Data analyses were conducted between January and September 2025.

EXPOSURES: Exposures were use of immunomodulators or biologics compared with 5-ASA, identified using National Drug Codes and Healthcare Common Procedure Coding System procedure codes.

MAIN OUTCOMES AND MEASURES: The primary outcome was time to the first emergency department or inpatient visit caused by a MACE, defined as myocardial infarction, stroke, or all-cause mortality. Hazard ratios (HRs) and 95% CIs for MACE risk with immunomodulators or biologics vs 5-ASA were estimated using Cox proportional hazards models.

RESULTS: A total of 16 387 patients (mean [SD] age, 74.73 [6.79] years; 9861 [60.18%] female) were included in the analysis. In the immunomodulators vs 5-ASA and biologics vs 5-ASA cohort, the mean (SD) ages were 74.05 (6.43) years and 73.68 (6.01) years, respectively, after matching. Both cohorts had more female participants (2580 [58.85%] and 1780 [58.09%], respectively). Baseline comorbidities were mostly balanced between groups. Compared with 5-ASA, there was no statistically significant difference in the risk of MACE for immunomodulators (HR, 0.84 [95% CI, 0.61-1.17]) or biologics (HR, 0.86 [95% CI, 0.59-1.24]), although point estimates were below 1.

CONCLUSIONS AND RELEVANCE: In this comparative effectiveness study of Medicare beneficiaries with IBD, there was no statistically significant difference in MACE risk between those who used immunomodulators or biologics vs 5-ASA.

PMID:42054028 | DOI:10.1001/jamanetworkopen.2026.9091

JAMA Netw Open. 2026 Apr 1;9(4):e269673. doi: 10.1001/jamanetworkopen.2026.9673.

ABSTRACT

IMPORTANCE: Hypertrophic cardiomyopathy (HCM) is associated with an elevated risk of sudden cardiac death, often preceded by an out-of-hospital cardiac arrest (OHCA). However, population-based estimates of OHCA risk in patients with HCM are limited.

OBJECTIVE: To estimate the risk of OHCA in patients with HCM and identify characteristics associated with OHCA.

DESIGN, SETTING, AND PARTICIPANTS: This cohort study used multiple Danish registers during an observation period ranging from June 1, 2001, to December 31, 2022, and included a nested case-control study. All Danish residents aged 18 to 85 years during the study period constituted the source population. Patients with HCM were identified using codes from the International Statistical Classification of Diseases, Tenth Revision. The cohort included patients with a first-time HCM diagnosis and exposure-matched controls. In the nested case-control study, patients with HCM who experienced OHCA were risk-set matched with controls with HCM and no OHCA at the index time. Analyses were performed between September 1 and November 30, 2025.

EXPOSURE: First-time diagnosis of HCM.

MAIN OUTCOMES AND MEASURES: Time to OHCA from exposure or the matching date was the primary outcome. Risk estimates were determined using the Aalen-Johansen estimator. Association between covariates and OHCA was determined by conditional logistic regression.

RESULTS: The cohort included a total of 29 240 individuals: 5901 patients with HCM (median age, 65 [IQR, 54-75] years; 3277 male [55.5%]) and 23 339 matched controls (median age, 65 [IQR, 55-75] years; 12 982 male [55.6%]). In the group aged 61 to 85 years, the 10-year risk of OHCA was 4.3% (95% CI, 3.4%-5.1%) in patients and 3.3% (95% CI, 3.0%-3.7%) in controls. In the group aged 18 to 60 years, the 10-year risk was 2.8% (95% CI, 1.9%-3.7%) in patients and 1.5% (95% CI, 1.2%-1.8%) in controls. The nested case-control study included 250 cases with HCM and OHCA (167 male [66.8%]; median age, 68 [IQR, 59-76] years) and 1000 controls with HCM and no OHCA (668 male [66.8%]; median age, 68 [IQR, 59-76] years). Heart failure, both recent and longer term, was associated with an increased rate of OHCA (hazard ratio, 3.63 [95% CI, 1.55-8.50] and 2.82 [95% CI, 1.88-4.22], respectively).

CONCLUSIONS AND RELEVANCE: The findings of this cohort study suggest that HCM was associated with an increased risk of OHCA in people aged 18 to 85 years. The rate of OHCA was associated with heart failure, underscoring the need for improved risk stratification to optimize primary prevention.

PMID:42054027 | DOI:10.1001/jamanetworkopen.2026.9673