Search Results (285)

Background/Objectives: With the global rise in chronic diseases among older adults, rehabilitation services have become essential, particularly for those with cerebrovascular and central nervous system (CNS) disorders, which lead to significant long-term disabilities. To determine the impact of designated rehabilitation medical institutions on the readmission rates of older patients with CNS disorders who receive surgical interventions. Methods: This was a population-based cohort study. Data was obtained from the National Health Insurance Service database (2002–2019). Fifteen designated institutions participated in the pilot project for convalescent rehabilitation. We analyzed the data of 1019 patients before and after the implementation of the designated rehabilitation institution. The study sample included (1) patients admitted to 15 designated institutions participating in the pilot project for convalescent rehabilitation and (2) patients diagnosed with conditions classified under the rehabilitation patient group, Rehabilitation Impairment Category 1 to 7. The intervention was the pilot project for designated rehabilitation institutions, launched in October 2017. The primary outcome of interest was the readmission rate of older patients with CNS disorders who received surgical interventions. Interrupted time series analysis with segmented regression was used to assess changes in the 30-day readmission rates. Results: Post-intervention, an 8% reduction in 30-day readmission rates (estimate, 0.9225; 95% confidence interval: 0.9129–0.9322, p < 0.0001) was observed. Subgroup analysis showed a significant decline in readmission rates across various patient groups, including those with disabilities, high Charlson Comorbidity Index scores, and extended hospital stays. The regions outside Seoul (capital city), particularly Gyeonggi/Incheon (areas around Seoul) and other areas (i.e., rural), also showed a significant decrease in readmission trends after the intervention. Conclusions: Designated rehabilitation medical institutions led to a significant reduction in readmission rates of older patients with CNS disorders, suggesting that these institutions effectively support recovery and reduce the burden of readmission for patients with severe conditions and those residing in non-capital cities. Full article

Background/Objectives: Growing concerns about antibiotic-associated adverse events, including Clostridioides difficile infection, prompted implementation of an antibiotic stewardship program (ASP) in South Korea in November 2024. One year post-implementation, we evaluated changes in antibiotic consumption and C. difficile infection incidence. Methods: This study was conducted at Chung-Ang University Gwangmyeong Hospital, South Korea. Segmented regression and interrupted time series analyses were performed using weekly data on antibiotic use (days of therapy [DOT] per 1000 patient-days) and C. difficile infection or colonization (cases per 1000 patient-days) over 157 weeks (November 2022–October 2025). Weeks 1–105 defined the pre-ASP period, and weeks 106–157 the post-ASP period. A 4-week lag between antibiotic use and subsequent C. difficile infection was hypothesized. Results: Before ASP, weekly total antibiotic use increased (β₁ = 1.14, 95% CI, 0.76 to 1.51, p < 0.001). After ASP, the slope decreased significantly (β₃ = −1.50, 95% CI −2.62 to −0.39, p = 0.009), consistent across anti-pseudomonal penicillins and cephalosporins and fluoroquinolones. Pre-ASP C. difficile incidence increased (α₁ = 0.01, 95% CI, 0.01 to 0.02, p < 0.001); the upward trend attenuated post-ASP, though slope change was not significant (α₃ = −0.01, 95% CI, −0.03 to 0.004, p = 0.13). An increase of 1 DOT per 1000 patient-days was associated with a 0.005-case increase in C. difficile infection incidence after 4 weeks. Conclusions: The observed effects of proactive ASP strategies underscore the importance of maintaining stewardship in clinical practice. Further studies are warranted to assess the sustainability of these findings and evaluate additional factors influencing C. difficile infection incidence. Full article

Background and Objectives: Payment parity laws require commercial health plans to pay for telehealth on the same basis as in-person care. We systematically reviewed open-access empirical studies to identify and synthesize empirical U.S. studies that explicitly evaluated state telehealth payment parity (distinct from coverage-only parity) and to summarize reported effects on telehealth utilization, modality mix, quality/adherence, equity/access, and expenditures. Methods: Following PRISMA 2020, we searched PubMed/MEDLINE, Scopus, and Web of Science for U.S. studies that explicitly modeled state payment parity or stratified results by payment parity vs. coverage-only vs. no parity. We included original quantitative or qualitative studies with a time or geographic comparator and free full-text availability. The primary outcome was telehealth utilization (share or odds of telehealth use); secondary outcomes were modality mix, quality and adherence, equity and access, and spending. Because designs were heterogeneous (interrupted time series [ITS], difference-in-differences [DiD], regression, qualitative), we used structured narrative synthesis. Results: Nine studies met inclusion criteria. In community health centers (CHCs), payment parity was associated with higher telehealth use (42% of visits in parity states vs. 29% without; Δ = +13.0 percentage points; adjusted odds ratio 1.74, 95% CI 1.49–2.03). Among patients with newly diagnosed cancer, adjusted telehealth rates were 23.3% in coverage + payment parity states vs. 19.1% in states without parity, while cross-state practice limits reduced telehealth use (14.9% vs. 17.8%). At the health-system level, parity mandates were linked to a +2.5-percentage-point telemedicine share in 2023, with mental-health (29%) and substance use disorder (SUD) care (21%) showing the highest telemedicine shares. A Medicaid coverage policy bundle increased live-video use by 6.0 points and the proportion “always able to access needed care” by 11.1 points. For hypertension, payment parity improved medication adherence, whereas early emergency department and hospital adoption studies found null associations. Direct spending evidence from open-access sources remained sparse. Conclusions: Across ambulatory settings—especially behavioral health and chronic disease management—state payment parity laws are consistently associated with modest but meaningful increases in telehealth use and some improvements in adherence and perceived access. Effects vary by specialty and are attenuated where cross-state practice limits persist, and the impact of payment parity on overall spending remains understudied. Full article

In southern Brazil, dengue transmission in the state of Paraná has shown a significant increase in the number of cases since the first recorded occurrence in 1995, with more frequent outbreaks in the west, northwest, and north of the state. We evaluated the impact of a campaign of dengue vaccination administered to a fraction of the population in 30 municipalities in the state by conducting a 15-year interrupted time-series ecological study using data obtained from an official Brazilian data register. We modeled dengue incidence using Poisson regression adjusted by covariates (demographic, climate, and epidemiological factors), allowing for specific temporal variation for each site. A reduction of 18.7% in dengue incidence rate was estimated for a vaccination coverage of 100%. Although there was an increase in the crude dengue incidence rate, considering the three-dose coverage achieved in the municipalities, we estimated an 8.2% relative reduction in the incidence rate. This reduction would increase to 17% with a hypothetical coverage of 90%. The campaign was more effective in small municipalities since they had higher vaccination coverage. These findings underscore the significant impact of the vaccination campaign on reducing dengue incidence trends across the targeted municipalities. Full article

Hazardous gas leaks are a major trigger of chemical incidents. If not handled in time, they can easily lead to secondary disasters such as fires and explosions. In recent years, with the construction of hazardous chemical monitoring and early-warning systems in China, large volumes of field operating data from flammable and toxic gas sensors have been accumulated, providing a data foundation for leak-pattern studies grounded in real-world scenarios. In this study, 56 leak samples verified by site feedback were selected. Time-aware interpolation and Z-score normalization were used for preprocessing, and time-series features—including standard deviation of first differences, autocorrelation coefficients, and frequency-domain energy—were extracted. Leak patterns were then identified using two unsupervised approaches: K-Means clustering and a 1D-CNN autoencoder. Results show that K-Means effectively distinguishes macro-patterns such as sustained leaks, instantaneous leaks, fluctuating leaks, and interrupted leaks, while the autoencoder demonstrates stronger capability in extracting temporal features, revealing leak evolution and transition characteristics. The two methods are complementary and together provide a viable route to developing an end-to-end model for leak scenario identification and risk discrimination. This work not only verifies the feasibility of conducting leak-pattern recognition using real GDS data but also offers technical guidance for the intelligent upgrading of hazardous chemical monitoring and early-warning systems. Full article

Background/Objectives: We investigated multimodal strategies to reduce neonatal ventilator-associated pneumonia (VAP) and antimicrobial use across three periods: period 1 (2014–2017), environmental cleaning with sodium hypochlorite, installation of heat and moisture exchangers, elective high frequency oscillatory ventilation (HFOV) as the primary invasive mode, and nasal HFOV after extubation; period 2 (2018–2020), oral care with maternal milk; and period 3 (2021–2024), nasal synchronized intermittent positive pressure ventilation after extubation. Methods: We conducted a quasi-experimental study of all neonates admitted to a neonatal intensive care unit in Thailand. We compared the trends in VAP and antimicrobial use rates using interrupted time-series analysis with segmented regression. Results: During the 11-year study period, 45.6% of neonates were intubated (2470/5414), and the ventilator utilization ratio was 0.19 (17,820 ventilator days/95,151 patient days). The overall VAP incidence was 4.55 per 1000 ventilator days. The yearly VAP incidence density ratio was significantly lower than in 2014. The baseline trend of VAP incidence and colistin use decreased significantly during period 1; nonetheless, the level and slope did not differ significantly between periods 1, 2, and 3. Conclusions: Tailored implementations, namely environmental decontamination, ventilator circuit care, elective HFOV, and nasal HFOV, reduced VAP and colistin use during period 1. Moreover, additive interventions, including oral care in period 2 and nasal synchronized intermittent positive pressure ventilation in period 3, achieved sustained VAP reduction and limited colistin prescriptions in period 1. Full article

Background/Objectives: This study evaluated the impact of the COVID-19 pandemic on trends of acute coronary syndrome hospitalizations, all-cause deaths, and ischemic heart disease (IHD) deaths in Allegheny County, Pennsylvania. Methods: Inpatient hospital records from two hospital systems within Allegheny County, Pennsylvania, were aggregated from January 2017 to November 2020. The primary diagnoses were acute myocardial infarction (AMI) and unstable angina. The Pennsylvania Department of Health provided all-cause and IHD death counts for the same period. We compared absolute percentage changes in admissions by year (March–November) and trends by age-specific groups (<45, 45–64, 65–74, ≥75) from the pre-pandemic (January 2017–February 2020) to pandemic (March 2020–November 2020) period using an interrupted time-series analysis. Results: There were 11,913 AMI hospitalizations pre-pandemic and 2170 AMI hospitalizations during the pandemic period. AMI hospitalizations decreased by 14.8% and unstable angina hospitalizations decreased by 30.7% during the pandemic compared to 2019, with the largest decreases occurring in those aged ≥75. Total mortality increased by 9.2%, and IHD mortality increased by 2.4%. About 80% of the increase in deaths was due to COVID-19, and approximately 75% of deaths occurred in those aged ≥75 and in long-term care facility residents. Conclusions: The COVID-19 pandemic did not markedly alter the longitudinal declining trend of AMI hospitalizations and IHD deaths in Allegheny County. Full article

Background: Pediatric urinary tract infections (UTIs) are increasingly complicated to treat due to antimicrobial resistance (AMR). The coronavirus disease 2019 (COVID-19) pandemic has led to substantially reduced pediatric antibiotic prescribing, but its impact on resistance trends remains unclear. This study aimed to investigate the AMR trends in urinary pathogens from children under 24 months of age hospitalized with febrile UTI during the pre-, during-, and post-COVID-19 pandemic periods. Methods: We conducted a retrospective study of children aged <24 months who were hospitalized at a tertiary center in Korea between 2008 and 2023 for first febrile UTI. The patients were stratified by age (<100 days vs. 100 days to 24 months) and pandemic period (pre-, during-, and post-COVID-19). Interrupted time-series (ITS) analysis and multivariable logistic regression were used to assess the temporal trends and predictors of ciprofloxacin nonsusceptibility. Results: Ciprofloxacin susceptibility decreased significantly during the pandemic, especially among infants < 100 days. ITS analysis demonstrated an immediate 12.1% increase in ciprofloxacin nonsusceptibility at pandemic onset, which persisted thereafter. Extended-spectrum β-lactamase production was the strongest predictor of ciprofloxacin resistance. Conclusions: These findings suggest that adult antibiotic use and clonal dissemination may contribute to the persistence and spread of AMR, highlighting the need for integrated stewardship and genomic surveillance. Full article

Violence is a major public health concern, but long-term hospital-based analyses in Latin America remain scarce. This study examined trends, structural breaks, and seasonality of violence-related emergency visits at the José Casimiro Ulloa Emergency Hospital in Lima, Peru, between 2015 and 2024. A retrospective analysis of 14,570 visits was performed, classifying cases according to the World Health Organization typology into self-inflicted, interpersonal community, and interpersonal family violence. Descriptive statistics were stratified by sex, life stage, migratory status, and pandemic period: pre-pandemic (2015–2019), pandemic (2020–2021), and post-pandemic (2022–2024). Time-series analyses included segmented regression and seasonal-trend decomposition. Males (78.3%), youth, and adults were the most affected groups. Interpersonal community violence accounted for 94.2% of cases and closely mirrored the overall series, peaking at 327 visits in July 2019 and falling to 28 in April 2020. Segmented regression showed significant immediate decreases at the onset of the pandemic (−71.6 visits, p < 0.001) and partial rebounds after 2022. Family violence remained consistently low and stable, while self-inflicted violence displayed a pre-pandemic upward trend, a sharp decline in 2020, and higher post-pandemic levels. Seasonality was evident before 2020, disappeared during the pandemic, and re-emerged with reduced amplitude thereafter. The COVID-19 pandemic caused abrupt but heterogeneous disruptions in violence-related emergency visits in Lima. Full article

The coronavirus disease 2019 (COVID-19) pandemic in Tocantins, Brazil, exhibited distinct phases between 2020 and 2025, with high mortality concentrated in 2020–2021 and subsequent stabilization at residual levels. Using epidemiological data, statistical modeling, and genomic surveillance, we show that the crisis peaked in 2021, coinciding with the circulation of Gamma and Delta, when health system capacity was severely strained. From 2022 onwards, the spread of Omicron led to record incidence but proportionally low mortality, reflecting accumulated immunity, vaccination, and improved clinical management. Vaccination represented the turning point, reducing hospitalizations and deaths by over 90% and driving a clear decoupling between incidence and severity. Interrupted time-series and generalized additive model (GAM) analyses confirmed sustained reductions in transmission and severity associated with mass immunization. Genomic sequencing of 3941 severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) genomes identified 166 lineages and successive variant replacements, culminating in the predominance of LP.8.1.4 in 2025. To our knowledge, this is one of the few integrated, long-term analyses (2020–2025) combining epidemiological and genomic data, capturing the full succession of variants up to LP.8.1.4 and highlighting Tocantins as a strategic “variant corridor” linking Brazil’s North and Central-West regions. These findings underscore the dual role of vaccination and genomic surveillance in shaping the epidemic trajectory and the importance of sustaining both strategies to mitigate future health crises. Full article

Background/Objectives: Safety information in hospital laboratories is often fragmented or difficult to retrieve under time pressure. We developed an Essential Safety Sheet (ESS) to present critical, task-level safety information immediately and evaluated its effectiveness on safety performance and incidents. Methods: We conducted a mixed-methods evaluation across eight university hospital laboratories from March 2023 to August 2024, including a 13-month interrupted time series with a concurrent difference-in-differences comparison between ESS and control laboratories (pre-implementation 6 months, implementation month, post-implementation 6 months). Primary outcomes were (1) emergency escalation accuracy, (2) information search time for task-critical items and (3) laboratory incident rates. Segmented regression models with robust Standard errors estimated level and slope changes; parallel trends were assessed pre-intervention. Multiple comparisons across the three primary outcomes were controlled for using the Bonferroni correction. Qualitative usability feedback was analyzed to contextualize the quantitative effects. Results: ESS implementation was associated with significant improvements in information search time and reductions in incident rates that were sustained over the post-implementation period in the ESS laboratories relative to the controls. Escalation accuracy improved in direction but did not reach statistical significance after multiple comparison correction (Bonferroni-adjusted p = 0.150). Findings were robust to the sensitivity analyses of model specification and pre-trend assumptions. Conclusions: A concise, task-level safety sheet can enhance the speed of safety-critical information retrieval and contribute to lower incident rates in hospital laboratories. While escalation accuracy showed only a favorable trend after correction, overall results support ESS as low-cost, scalable interventions to strengthen laboratory safety performance. Future studies should test generalizability across more sites and tasks to assess longer-term sustainability. Full article

Background/Objectives: Cervical intraepithelial neoplasia grade 3 and worse (CIN3+) is a robust surrogate for cervical cancer risk. In Norway, organized cervical screening starts at 25 years of age (25–69 years). Norway introduced school-based HPV vaccination with the quadrivalent vaccine for 12-year-old girls in 2009 (birth cohorts ≥ 1997) with high 3-dose completion, and a catch-up program with the bivalent vaccine for women born 1991–1996 in 2016–2019 with lower uptake. We assessed whether increasing birth-cohort vaccination coverage (defined as ≥1 dose) was followed by reductions in CIN3+ at the age of entry to organized screening (25–29 years). Methods: We conducted a retrospective, population-based time-series of women aged 25–29 years in Troms and Finnmark screened in 2010–2024. CIN3+ was counted per unique woman and expressed per 1000 screened women per year. Cohort-level vaccination exposure was proxied by birth-year eligibility and national coverage (≥1 dose) by calendar year. Temporal trends were assessed using segmented linear regression (2010–2017; 2017–2024). Results: Among 42,253 screening tests, 865 women had CIN3+. CIN3+ rates were stable in 2010–2016 (≈15–24 per 1000), peaked in 2017–2018 (≈26–28 per 1000), and declined to 6.6 per 1000 in 2024 (~75% reduction from the peak). The 2010–2017 trend was not significant (p = 0.244), whereas 2017–2024 showed a significant annual decline (slope −3.04 per 1000 per year; p = 7.4 × 10⁻⁵). The decline coincided with an increase in the vaccinated share of the age group from an estimated 12% in 2017 to 78% in 2024. Cervical cancer was rare throughout and absent in 2024, and the 2023 transition to primary HPV testing did not interrupt the downward trend. Conclusions: As vaccinated birth cohorts—especially those vaccinated before sexual debut—entered organized screening at age 25, CIN3+ in women aged 25–29 years fell markedly. Estimates are based on coverage defined as ≥1 dose; future linkage to individual dose data and HPV type–specific CIN3+ is warranted. Full article

Real-time control of web traffic is a critical issue for network operators and service providers. It helps ensure robust service and avoid service interruptions, which has an important financial impact. However, due to the high speed and volume of actual internet traffic, standard multivariate time series models are inadequate for ensuring efficient real-time traffic management. In this paper we introduce a new model for functional time series analysis, developed by combining a local linear smoothing approach with an $L^1$-robust estimator of the quantile’s derivative. It constitutes an alternative, robust estimator for functional modal regression that is adequate to handle the stochastic volatility of high-frequency of web traffic data. The mathematical support of the new model is established under functional dependent case. The asymptotic analysis emphasizes the functional structure of the data, the functional feature of the model, and the stochastic characteristics of the underlying time-varying process. We evaluate the effectiveness of our proposed model using comprehensive simulations and real-data application. The computational results illustrate the superiority of the nonparametric functional model over the existing conventional methods in web traffic modeling. Full article

Data is an important resource for gaining knowledge about the behavior and condition monitoring of machines, enabling the estimation of parameters and the prediction of failures. However, in industrial environments, sensor interruptions often create gaps in the time series, which affects the reliability of the data. To overcome this challenge, this paper proposes an imputation strategy based on recurrent neural networks, in particular long short-term memory (LSTM) models, within a multivariate encoder–decoder architecture. This approach utilizes correlations between variables to reconstruct missing values, resulting in more complete and robust datasets. Experimental results with multivariate time series show that the proposed method achieves accurate imputation, with errors as low as RMSE = 2.33 and $R^2$ = 0.90 for some variables. Comparisons with alternative architectures, including GRU and Dense networks, show that LSTM excels in specific cases (e.g., VL3, $R^2$ = 0.45), while the Dense architecture provides more stable performance across most variables. In particular, the Dense model achieved the best overall balance between accuracy and robustness, reaching RMSE = 2.33 and $R^2$ = 0.90 for the best-performing variables, while the LSTM achieved the lowest error values in targeted scenarios, confirming its suitability for capturing complex temporal dependencies. Overall, this study highlights the feasibility of using recurrent neural networks to exploit temporal correlations for reliable data recovery, even under conditions of signal interruption in factory environments. Full article

Background: Delays in emergency cesarean section (CS) remain a major contributor to maternal and neonatal morbidity in low-resource settings. This study evaluated the combined effect of a mobile-based obstetric emergency system (MORES) and a midwife-led triage program on the decision-to-incision interval (DDI) and related outcomes in Liberia. Methods: A quasi-experimental study with an interrupted time series design was conducted in Bong County across two district hospitals receiving referrals from 20 rural health facilities. Seventy-two women referred for emergency CS were observed at baseline, midline, and endline. MORES used WhatsApp-based communication to improve referral coordination, while the triage program trained midwives to rapidly assess cases using a color-coded system. Data were analyzed using descriptive statistics, Wilcoxon rank-sum, chi-squared tests, and logistic regression. Results: By endline, the median DDI decreased by 117.5 min compared to baseline (95% CI: −205.1 to −29.9). Women were significantly more likely to receive a CS within 75 min (AOR: 11.7; 95% CI: 1.32 to 104.5). No maternal deaths occurred. Neonatal mortality was observed but not significantly associated with DDI. Conclusions: MORES and midwife-led triage substantially improved the timeliness of emergency CS in a resource-constrained setting. These low-cost, feasible strategies warrant further evaluation for sustainability and impact on neonatal outcomes. Full article