Hospital at home for elderly acute patients: a study protocol for a randomised controlled trial.

INTRODUCTION: The increasing elderly population has led to a growing demand for healthcare services. A hospital at home treatment model offers an alternative to standard hospital admission, with the potential to reduce readmission and healthcare consumption while improving patients' quality of life. However, there is little evidence regarding hospital at home treatment in a Danish setting. This article describes the protocol for a randomised controlled trial (RCT) comparing standard hospital admission to hospital at home treatment. The main aim of the intervention is to reduce 30-day acute readmission after discharge and improve the quality of life of elderly acute patients. METHODS AND ANALYSIS: A total of 849 elderly acute patients will be randomised in a 1:2 ratio to either the control or intervention group in the trial. The control group will receive standard hospital treatment in a hospital emergency department while the intervention group will receive treatment at home. The primary outcomes of the trial are the rate of 30-day acute readmission and quality of life, assessed using the European Quality of Life-5 Dimensions-5-Level instrument. Primary analyses are based on the intention-to-treat principle. Secondary outcomes are basic functional mobility, resource use in healthcare, primary and secondary healthcare cost, incremental cost-effectiveness ratio, and the mortality rate 3 months after discharge. ETHICS AND DISSEMINATION: The RCT was approved by the Ethical Committee, Central Denmark Region (no. 1-10-72-67-20). Results will be presented at relevant national and international meetings and conferences and will be published in international peer-reviewed journals. Furthermore, we plan to communicate the results to relevant stakeholders in the Danish healthcare system. TRIAL REGISTRATION NUMBER: NCT05360914.

Risk of death within 7 days of discharge from emergency departments with different organizational models.

OBJECTIVE: The aim of this study is to investigate the association between emergency department (ED) organizational models and the risk of death within 7 days of ED discharge. PATIENTS AND METHODS: We included Danish ED discharges between 1 January 2011 and 24 December 2014 that led to death within 7 days of discharge. The inclusion criterion was age older than 18 years. The exclusion criterion was further in-hospital admission. First model (Virtual): other departments employ interns who perform ED tasks. They are responsible for ED patient care and prioritize their task order between their own department and the ED. Second model (Hybrid): the ED/other departments perform tasks; interns/consultants are employed by the ED/other departments. The ED/other departments have patient care responsibility. Third model (Independent): the ED performs all tasks; employs interns/consultants; and have patient care responsibility. Sex, age, Charlson Comorbidity Index score, and primary diagnosis were used to describe patient characteristics. We calculated the risk of death within 7 days of discharge using multiple logistic regression analysis. RESULTS: In 805 out of 201 299 discharges included in the study, the patient died within 7 days. Compared with the Virtual model, the odds ratio for death within 7 days of discharge was 0.72 (95% confidence interval: 0.59-0.92) for the Independent model and 0.75 (95% confidence interval: 0.61-0.92) for the Hybrid+Virtual model. Increased risk was associated with male sex, older age, and a medium or a high Charlson Comorbidity Index score. CONCLUSION: Compared with discharges from a Virtual model, the risk of death within 7 days of discharge was lower if the ED had an Independent or a Hybrid+Virtual model.

The organization of Danish emergency departments.

INTRODUCTION: Twenty-one new Danish emergency departments (EDs) were established following a 2007 policy reform that included ED autonomy to self-organize. The aim of this study was to describe the organization of the 21 departments and their organizational challenges. PARTICIPANTS AND METHODS: We used a qualitative design based on COREQ guidelines. All 21 EDs participated, and 123 semi-structured interviews with hospital and ED leaders, physicians, nurses, and secretaries were performed between 2013 and 2015. We used the framework matrix method to investigate the ED goals, setting, structure, staff, task coordination, and incentive structure. RESULTS: We identified three generic models (virtual, hybrid, and independent). All had goals of high quality of care and high efficiency. The virtual model was staffed by junior physicians and tasks were coordinated by other departments. The hybrid model was staffed by junior physicians and senior physicians according to other departments and the ED. The ED coordinated all activities. The independent model was staffed by junior physicians and senior physicians, and activities were coordinated by the ED. Of the EDs, 19 utilized different organizational models at different times during a 24-h period and on weekdays and weekends. The main challenge of the virtual and hybrid models was high dependency on other departments. The main challenge of the independent model was establishing a high level of quality of emergency medicine. DISCUSSION AND CONCLUSION: We identified three organizational ED models (virtual, hybrid, and independent). Nineteen EDs used more than one organizational model depending on the time of day or day of the week.

Remote Real-Time Ultrasound Supervision via Commercially Available and Low-Cost Tele-Ultrasound: a Mixed Methods Study of the Practical Feasibility and Users' Acceptability in an Emergency Department.

Minor emergency departments (ED) struggle to access sufficient expertise to supervise learners of lung and cardiac point-of-care ultrasound (POCUS). Using tele-ultrasound (tele-US) for remote supervision may remedy this situation. We aimed to evaluate the feasibility of real-time supervision via tele-US when applied to an everyday ED clinic. We conducted a mixed methods study that assessed practical feasibility, determined performance, and explored users' acceptability of supervision via tele-US. Technical performance was assessed quantitatively by the ratio in mean gray value between images on site and as received by the supervisor, and by after-compression frame rate. Qualitatively, 12 exploratory semi-structured interviews were conducted with exposed junior doctors and supervisors. Remote supervision via tele-US was performed with 10 junior doctors scanning 45 included patients. During performance assessment, neither alternating internet connection nor software significantly changed the mean gray value ratio. The lowest median frame rate of 4.6 (interquartile range [IQR]: 3.1-5.0) was found by using a 4G internet connection; the highest of 28.5 (IQR: 28.5-29.0) was found with alternative computer and local area network internet connection. In interviews, supervisors stressed the importance of preserving frame rate, and junior doctors emphasized a need for shared ultrasound terminology. In the qualitative analysis, setup mobility, accessibility, and time consumption were emphasized as being of key importance for future clinical implementations. Remote supervision via a commercially available and low-cost tele-US setup is operational for both junior doctors and supervisors when applied to lung and cardiac POCUS scans of hospitalized patients.

Linking the severity of illness and the weekend effect: a cohort study examining emergency department visits.

BACKGROUND: Despite extensive research on the "weekend effect" i.e., the increased mortality associated with hospital admission during weekend, knowledge about disease severity in previous studies is limited. The aim of this study is to examine patient characteristics, including disease severity, 30-day mortality, and length of stay (LOS), according to time of admission to an emergency department. METHODS: Our study encompassed all patients admitted to a Danish emergency department in 2014-2015. Using data from electronic patient records, this study examines patient characteristics including age, gender, Charlson Comorbidity Index score, triage score, and primary diagnosis. Triage score and transfer to intensive care unit (ICU) were used as indicators of disease severity. LOS within the department and within the hospital was examined. Age- and sex-standardized 30-day mortality rates comparing patients with the same triage score admitted at daytime, evening, and nighttime on weekdays and on weekends were computed. To test differences, a Cox regression analysis was added. RESULTS: We included 35,459 patient visits, of which 10,435 (32%) started on a weekend. There were no large differences in baseline characteristics between patients admitted on weekdays and those admitted on weekends. The relative risk (RR) for being triaged orange or red was 1.16 (95% confidence interval (CI) 1.06-1.28, P = 0.0017) for weekend admissions as compared with weekday admissions. Weekend admissions were twice as likely as weekday admissions to be transferred to the ICU (RR, 1.96; 95% CI 1.53-2.52, P = 0.0000). No significant changes were found in LOS. The 30-day mortality rate increased with disease severity regardless of time of admission. When comparing the 30-day mortality rate for patients with the same triage score, the trend was toward a higher mortality when admission occurred during the weekend. Increasing mortality rate was significant for patients admitted at evening on weekends with a hazard ratio of 1.32 (95% CI 1.03-1.70, P = 0.027) when compared with patients admitted on daytime on weekdays. CONCLUSIONS: When comparing weekday and weekend admissions, the 30-day mortality rate increased for patients admitted at evening on weekends after adjusting for comorbidity and triage score, indicating that the weekend effect was independent of changes in illness severity.

Geographic information system data from ambulances applied in the emergency department: effects on patient reception.

BACKGROUND: Emergency departments (ED) recognize crowding and handover from prehospital to in-hospital settings to be major challenges. Prehospital Geographical Information Systems (GIS) may be a promising tool to address such issues. In this study, the use of prehospital GIS data was implemented in an ED in order to investigate its effect on 1) wait time and unprepared activations of Trauma Teams (TT) and Medical Emergency Teams (MET) and 2) nurses' perceptions regarding patient reception, workflow and resource utilization. INTERVENTION: From May 1st 2014 to October 31th 2014, GIS data was displayed in the ED. Data included real-time estimated time of arrival, distance to ED, dispatch criteria, patient data and ambulance contact information. Data was used by coordinating nurses for time activation of TT and MET involved in the initial treatment of severely-injured or critically-ill patients. In addition, it was used as a logistics tool for handling all other patients transported by ambulance to the ED. STUDY DESIGN: The study followed a mixed-methods design, consisting of a quantitative study (before and after intervention) and a qualitative study (survey and interviews). PARTICIPANTS: Participants included all patients received by TT or MET and coordinating nurses in the ED. RESULTS: 1.) Quantitative: 599 patients were included. The median wait time for TT and MET was 5 min both before and after the GIS intervention, showing no difference (p = 0.18). A significant reduction in the subgroup of waits >10 min was found (p < 0.05). No difference was found in unprepared TT and MET activations. 2.) Qualitative: Nurses perceived GIS data as a tool to optimize resource utilization and quality of all patients' reception, critically or non-critically ill. No substantial disadvantages were reported. DISCUSSION: The contradiction of measured median wait time and nurses perceived improved timing of team activation may result from having both RT- ETA and supplemental patient information not only for seriously-injured or critically-ill patients received by the TT and MET, but for all patients transported by ambulance. The reduction in waits > 10 minutes may have contributed to the overall perception of reduced wait time, as avoidance of long waits is clinically more important than reduction in the median wait time. CONCLUSION: A comparison of the use of prehospital GIS data in the ED with the control period showed no effect on median wait time for TT and MET, however, the number of waits of >10 min was reduced. On the other hand, nurses perceived implementation of GIS data as improving workflow, resource utilization and quality of all patients' reception, critically as well as non-critically ill. There were no substantial disadvantages to the GIS application. TRIAL REGISTRATION: ClinicalTrials.gov (NCT02188966).