Search

Purpose Some predictive systems using machine learning models have been developed to predict sepsis; however, they were mostly built with a low percent of missing values, which does not correspond with the actual clinical situation. In this study, we developed a machine learning model with a high rate of missing and erroneous data to enable prediction under missing, noisy, and erroneous inputs, as in the actual clinical situation. Materials and methods The proposed artificial neural network model was implemented using the MATLAB ANN toolbox, based on stochastic gradient descent. The dataset was collected over the past decade with approval from the appropriate institutional review boards, and the sepsis status was identified and labeled using Sepsis-3 clinical criteria. The imputation method was built by last observation carried forward and mean value, aimed to simulate clinical situation. Results The mean area under the receiver operating characteristic (ROC) curve (AUC) of classifying sepsis and nonsepsis patients was 0.82 and 0.786 at 0 h and 40 h prior to onset, respectively. The highest model performance was found for one-hourly data, demonstrating that our ANN model can perform adequately with limited hourly data provided. Conclusions Our model has the moderate ability to predict sepsis up to 40 h in advance under simulated clinical situation with real-world data.

Summary Introduction Mortality of trauma patients is strongly associated with Injury Severity Score (ISS). However, the interaction between ISS and age has not been studied extensively. We evaluated ISS in different age groups in relation to mortality rate. Materials and methods This was a retrospective review of 955 trauma patients who visited Mackay Memorial Hospital during January–December 2009. Age, sex, emergency department (ED) stay time, hospital stay time, out-of-hospital cardiac arrest, in-hospital cardiac arrest, ED survival rate, emergency operation, and mortality were analyzed. Patients were separated into two groups by age below or above 65 years, and hospital courses and mortality rate were compared. Results Nine hundred and fifty-five trauma patients visited our ED in 2009: 212 in the aged group (99 male patients; 46.7%) and 743 in the young group (472 male patients; 63.5%). The mean age of the elderly group was 77 ± 8 years (mean ± standard deviation), and 38 ± 16 years in the young group. Mean ISS was 12 ± 12 in the elderly group and 10 ± 13 in the young group. The mean ED stay time was 2.9 ± 3.8 hours for the elderly group and 3.0 ± 4.6 hours for the young group. One hundred and sixty-three patients were admitted and the mean hospital stay time was 12.7 ± 13.6 days in the elderly group, whereas 559 patients were admitted and the mean hospital stay time was 9.9 ± 13.7 days in the young group (p = 0.018). The mortality rate among hospitalized patients was 3.1% in the elderly group and 2.9% in the young group. On receiver operating characteristic curve analysis, the optimal cut-off ISS for mortality rate among young patients (area under the curve: 0.899) and elderly patients (area under the curve: 0.782) was 17 (sensitivity: 81.3%; specificity: 88.7%) and 14 (sensitivity: 80%; specificity: 82.3%), respectively. Conclusion The elderly group seemed to have no significant difference from the younger group for mortality rate, ED stay time and hospital stay time. In the elderly group, lower ISS predicted a higher mortality than in the younger group.

Summary Objectives Emergency room visits by nursing home patients with bloodstream infections (BSIs) are not uncommon and pose challenges to clinicians when selecting appropriate antimicrobial agents. This research aims to discover distinct bacteriology in nursing home patients by focusing particularly on antimicrobial-resistant bacteria. Methods This retrospective study enrolled patients aged 65 years or older who had documented positive blood culture reports between May 2006 and June 2008. Patients were further categorized into subgroups according to nursing home history and the timely trend of hospital exposure. Results The adjusted risk of BSIs associated with nursing home exposure was significantly higher in Methicillin-resistant Staphylococcus aureus (odds ratio (OR) 2.0; 95% CI 1.5–2.8), Methicillin-resistant coagulase-negative Staphylococcus species (OR 1.7; 95% CI 1.2–2.5), and Enterobacteriaceae-extended spectrum β-lactamase-producing Enterobacteriaceae (OR 2.4; 95% CI 1.5–3.9), but not in Pseudomonas aeruginosa (OR 1.4; 95% CI 0.9–2.2) or Acinetobacter baumannii (OR 0.9; 95% CI 0.5–1.5). Conclusion Host environments—nursing home or hospital exposure—were shown to be more significant than medical comorbid conditions for acquiring antimicrobial-resistant pathogens. Each unique environment increased the risk for acquiring some pathogens, but not all. In addition, these environmental factors may also exert cumulative effects toward some specific pathogens.

Corrigendum to “Pneumoperitoneum Presenting as Minimal Left Subphrenic Air in an Elderly Patient” [Int J Gerontol 6 (4) (2012) 298e299] Ding-Kuo Chien 1, , Wen-Han Chang 1, 3, 4 , Shu-Tien Huang 1 1 Department of Emergency Medicine, Mackay Memorial Hospital, 2 Mackay Medicine, Nursing and Management College, 3 School of Medicine, College of Medicine, Taipei Medical University, 4 Institute of Mechatronic Engineering, National Taipei University of Technology, Taipei, Taiwan

Background: Significant advances have been made over the past 2 decades in the use of mobile devices. So-called smart devices (e.g., smartphones, tablet computers) allow people greater access to information on the Internet and at any time. Furthermore, most smart devices now incorporate sensors such as accelerometers, global positioning satellite components, Bluetooth technology, and cameras, which have greatly impacted their scope and utilization as a result of their multi-functional capacity. In the past, mobile medical platforms or applications were considered tools that care providers could use to improve patient care. However, health information technology should also provide convenient assistance to individuals for their own health and wellness management, enabling them to better communicate with doctors, learn and share information about their own health, and take actions that will improve their quality of life. The aims of this quantitative and qualitative study are as follows: to explore the functionality patients want, and to better understand individuals’ opinions, behaviors, and motivations, as well as the barriers to utilizing mobile mHealth applications. Methods: We conducted a quantitative and qualitative semi-structured in-depth interview study on October 13, 15 and 16, 2017, with 220 patients from a medical center in Taipei. Ten of the 220 patients were audiotaped, transcribed verbatim and analyzed using a directed content analysis. We report our findings following the Consolidated Criteria for Reporting Qualitative Research (COREQ) checklist. Results: The top three functions requested by patients were as follows: outpatient department real-time progress (58%); the adverse effects of medicine (49%); and personal prescription lists and records (48%). Together, 83.2% patients were willing to use a hospital application if it offered the functionality they required, and 75% indicated they would do so for their family. The five major themes related to the influence of eHealth technology, as indicated by respondents, are as follows: notify patients and minimize prescription wastage; record patients’ daily medication adherence; convenient access to medication information on the mobile application; big data analyses of the data collected; and interaction between the patient and healthcare professionals, as well as with family and other patients. Conclusions: While many mobile health applications are being developed, large hospitals are lagging in building application that resonate with patients. Patients complain of a poor user experience with hospitals, and functionality of proprietary mobile applications often fails to individuals’ needs. The technological ability of different age group varies. Thus, understanding consumers’ experiences and expectations can encourage better mobile health application design and improve patients’ motivation to use such applications.

Background: Pulmonary embolism (PE) has been associated with the risk of arterial cardiovascular events, with diagnosis being difficult in older adults. Multiple factors contribute to mortality and morbidity in older adults. Hence, we aimed to investigate the difference of subsequent cardiovascular events (SCVE) and demography between younger and older adults. Methods: We undertook a retrospective cohort study using data from a 1100-bed urban medical center in Taipei, Taiwan. All patients who presented to the emergency department with a documented diagnosis of PE (ICD-9 of 415.19), from January 2003 to December 2008, were enrolled and specified into two age groups. Clinical follow-up was conducted until December 2009, with a median time of 3.5 years. The SCVE included new onset of ischemic stroke or heart attack and recurrence of PE. The time between when a patient was presented to the ER and when the diagnosis took place was assessed. Results: A total of 104 patients were studied, among which 58 patients were placed in the elderly group (≥ 65 years old) and 46 were placed in the young group (< 65 years old). There were significantly more females in the elderly group (n=43; 74.1%; p

A 61-year-old man, without previous disease was admitted to our emergency department (ED) due to syncope at home after soaking his feet in a solution containing 50% of saponated cresol. On physical examination, the patient was lethargic with a strong medicinal smell and brownish discolouration with bullae formation of his bilateral feet (figure 1A), and with otherwise normal vital signs. According to the family, …

A 49-year-old woman with no surgical history had abdominal pain and fullness for 3 days. Initially, the pain was localised at the epigastric area, and on the third day the pain had became severe and generalised. On presentation, the patient had rigorous nausea and vomiting with normal vital signs. Physical examination revealed a …

We describe a case of Solanum erianthum poisoning that happened to a 75-year-old man. He ate the S erianthum that he gathered from the countryside, believing that it would be helpful in improving the numbness of his distal limbs. S erianthum is a solanaceous plant that contains a variable concentration of solanum alkaloids, causing gastrointestinal irritation, and tropane alkaloids that have anticholinergic properties producing typical and sometimes severe atropinelike symptoms. The diagnosis of S erianthum poisoning was made based on clinical symptoms and signs of anticholinergic syndrome associated with a history of vegetable meal from countryside and, moreover, on identification of the vegetable obtained from his family. Gastrointestinal decontamination and symptomatic treatment were useful in treatment of acute poisoning.