ABSTRACT Background Intensive care unit (ICU) nurses are exposed to numerous alarms, many of which are unnecessary and lead to alarm fatigue. This phenomenon, caused by desensitization to alarm sounds, reduces nurses' responsiveness to critical alerts and threatens both their well‐being and patient safety. Alarm fatigue has been linked to elevated stress, burnout and errors in patient care, underscoring the need for effective interventions. Progressive muscle relaxation involves the gradual and systematic contraction and relaxation of different muscle groups throughout the body and nature sounds use for decrease stress and alarm fatigue. Aim This study aimed to evaluate the effectiveness of progressive muscle relaxation (PMR) and exposure to nature sounds in reducing alarm fatigue among ICU nurses. Study Design A three‐arm, parallel, single‐blinded randomized controlled trial was conducted. Results Ninety‐six ICU nurses from three hospitals in southern Iran, were randomly assigned to one of three groups: PMR ( n = 32), nature sounds ( n = 32) or control ( n = 32). Alarm fatigue was assessed using a validated alarm fatigue scale at baseline, immediately after the intervention and 2 months post‐intervention. Both PMR and nature sound interventions significantly reduced alarm fatigue among ICU nurses ( p < 0.001). Between‐group comparisons showed significant differences between the PMR and control groups ( p < 0.001) and between the nature sound and control groups ( p < 0.001). No significant difference was observed between the PMR and nature sound groups ( p = 0.650). Conclusion PMR and exposure to nature sounds significantly reduced alarm fatigue among ICU nurses, as reflected in lower fatigue scores. These interventions improved nurses' capacity to manage stress and fatigue, thereby supporting better job performance and enhanced patient safety. Given their simplicity and accessibility, both approaches represent promising strategies for addressing alarm fatigue in critical care settings. Relevance to Clinical Practice Progressive muscle relaxation and nature sounds may offer practical, non‐pharmacological approaches to managing alarm fatigue among intensive care unit nurses. These interventions could be incorporated into nurses’ work routines, although further research is needed to confirm their effectiveness and long‐term benefits.
ABSTRACT Background Stroke‐related complications and frequent feeding interruptions often compromise nutritional intake, making enteral nutrition (EN) a cornerstone therapy to improve nutritional status and support recovery. Nevertheless, constipation is highly prevalent and clinically consequential among stroke patients receiving EN, particularly in critically ill populations, where enteral nutrition–associated constipation can impede rehabilitation, increase complications and prolong intensive care unit (ICU) stay. Aim This study aimed to explore potentially important determinants of enteral nutrition‐associated constipation (ENAC) in patients with stroke using artificial intelligence approaches, characterise relevant clinical subgroups, compare internally validated machine‐learning models and examine how the temporal availability of candidate predictors may influence model performance. Study Design Clinical data from patients admitted to a neurological intensive care unit between December 2019 and December 2023 were retrospectively analysed. K‐means clustering was used to identify clinical subgroups. Six algorithms—logistic regression, random forest, gradient boosting tree, XGBoost, LightGBM and Table PFN‐3—were compared under temporally restricted and expanded retrospective feature frameworks using five repeats of nested five‐fold cross‐validation. Performance was assessed using ROC‐AUC, average precision, Brier score, calibration and decision‐curve analysis. Treatment outcomes were summarised descriptively. Results Among 362 patients, 233 (64.4%) had documented constipation. Table PFN‐3 achieved the highest discrimination in both frameworks. The temporally restricted model yielded a ROC‐AUC of 0.629 (95% CI 0.567–0.689), whereas the expanded retrospective model achieved a ROC‐AUC of 0.771 (95% CI 0.715–0.820), average precision of 0.857, Brier score of 0.177 and calibration slope of 0.965. The ROC‐AUC increased by 0.142 (95% CI 0.090–0.201) after full‐hospitalisation information was included. K‐means clustering retained three subgroups ( K = 3; silhouette coefficient = 0.108), with constipation frequencies of 67.35%, 56.98% and 88.89% in Clusters 0, 1 and 2, respectively. Among evaluable patients, observed relief was 80.3% with glycerin enema, 60.0% with lactulose, 72.7% with combination treatment and 65.7% with neither treatment; these treatment comparisons were descriptive. Conclusions Table PFN‐3 achieved the strongest internally validated performance, with the expanded retrospective framework combining good discrimination, strong precision–recall performance and close calibration. The 0.142 increase in ROC‐AUC after completed hospital‐course variables were incorporated quantifies the substantial discriminative information that emerges as hospitalisation evolves. Together with the distinct clinical phenotypes and descriptive treatment findings, these results establish a clinically relevant framework for trajectory‐aware bowel monitoring and provide a clear rationale for developing dynamic, time‐updated prediction models. Relevance to Clinical Practice By integrating machine‐learning benchmarking, clinical phenotyping and observed treatment outcomes, this study identifies prolonged, treatment‐intensive hospital trajectories as a particularly high‐burden constipation phenotype and demonstrates that classification performance changes materially as hospital‐course information accumulates. These findings support structured reassessment of bowel risk during neurological intensive care rather than reliance on a single static evaluation and provide a practical rationale for future dynamic monitoring tools.
ABSTRACT Background Early warning scores ( EWS ) identify patient deterioration, but their impact depends on how they are integrated into workflows and escalation culture, particularly as hospitals recover from pandemic‐related bed pressures. Aims To evaluate changes in rapid response system ( RRS ) escalation processes and rare safety outcomes after embedding an EWS within proactive Critical Care Outreach Team ( CCOT ) rounds as hospital bed occupancy recovered. Study Design Observational longitudinal study at a single tertiary hospital in Japan comparing one pre‐intervention year (2021) with three post‐intervention years (2022–2024). In January 2022, a standardised EWS was embedded into proactive CCOT rounds and used as a prompt for shared clinical reasoning rather than a strict numeric trigger. Monthly aggregated ward‐level data were analysed using Poisson regression, adjusting for seasonality, patient characteristics and bed occupancy. Results Across 71 985 admissions and 692 942 general‐ward patient‐days, rates of nurse‐initiated escalations, RRS activations and resuscitation‐status updates per 1000 patient‐days increased in post‐intervention analyses, but these differences were attenuated after full adjustment. In fully adjusted models, post‐intervention incidence rate ratios ( IRRs ) were close to 1.0 for all process outcomes. For nurse‐initiated escalations, IRRs ranged from 0.94 to 1.14, and all 95% confidence intervals included 1.0. Rates of unexpected in‐hospital cardiac arrest and unplanned ICU admission remained low and stable. Bed occupancy increased over the study period, but within‐year associations with outcomes were weak. Conclusions After adjustment for patient characteristics and hospital bed occupancy, embedding an EWS within proactive CCOT rounds was associated with broadly stable escalation process indicators and no clear evidence of deterioration in rare safety outcomes. Relevance to Clinical Practice These findings suggest that approaches to the RRS afferent limb that embed EWS use within structured dialogue between ward nurses and outreach teams may provide one plausible mechanism for sustaining escalation behaviours under operational pressure.
ABSTRACT Background One Minute Wonders (OMWs) are workplace‐based microlearning resources that support continuing professional development in healthcare. Generative artificial intelligence (AI) is increasingly used to create educational materials, but little is known about its application in OMWs development. Aims To examine the use, experiences, opportunities and challenges related to AI‐assisted generation of OMWs as perceived among healthcare professionals participating in the German‐speaking OMW Network. Study Design A cross‐sectional online survey was conducted among members of the German‐speaking OMW Network in Germany, Austria and Switzerland. A study‐specific questionnaire assessed AI use, quality assurance practices, perceived benefits and risks and educational support needs. Quantitative data were analysed descriptively, and free‐text responses underwent qualitative content analysis. Results A total of 136 healthcare professionals were included, predominantly nurses ( n = 128, 94.1%). Overall, n = 63 (46.3%) reported using AI to create OMW. Among AI users, AI was most used for generating figures ( n = 41, 65.1%), graphics ( n = 35, 55.6%) and design elements ( n = 33, 52.4%), with ChatGPT being the most frequently used tool ( n = 60, 95.2%). Most AI users critically reviewed AI‐generated content ( n = 50, 79.4%). Inaccuracies requiring revision were reported by all AI users, most commonly ambiguous wording ( n = 32, 50.8%), inappropriate images ( n = 32, 50.8%) and incorrect content ( n = 29, 46.0%). The content analysis identified a main theme for opportunities: AI as a facilitator of educational resource development and knowledge translation and a main theme for risks: AI as a potential threat to the trustworthiness and professional integrity of educational resources. Conclusions The use of AI for developing educational materials such as OMWs seems to be established practice among participants of this study. AI offers benefits for efficiency and content creation but requires careful human oversight because inaccuracies and fabricated information are common. Relevance to Clinical Practice AI can support the rapid development of workplace‐based educational resources. However, critical review and quality assurance are essential to ensure accuracy and safety.
