An intelligent decision support system for acute postoperative endophthalmitis: design, development and evaluation of a smartphone application.

BACKGROUND: Today, clinical decision support systems based on artificial intelligence can significantly help physicians in the correct diagnosis and quick rapid treatment of endophthalmitis as the most important cause of blindness in emergency diseases. This study aimed to design, develop, and evaluate an intelligent decision support system for acute postoperative endophthalmitis. METHODS: This study was conducted in 2020-2021 in three phases: analysis, design and development, and evaluation. The user needs and the features of the system were identified through interviews with end users. Data were analyzed using thematic analysis. The list of clinical signs of acute postoperative endophthalmitis was provided to ophthalmologists for prioritization. 4 algorithms support vector machine, decision tree classifier, k-nearest neighbors, and random forest were used in the design of the computing core of the system for disease diagnosis. The acute postoperative endophthalmitis diagnosis application was developed for using by physicians and patients. Based on the data of 60 acute postoperative endophthalmitis patients, 143 acute postoperative endophthalmitis records and 12 non-acute postoperative endophthalmitis records were identified. The learning process of the algorithm was performed on 70% of the data and 30% of the data was used for evaluation. RESULTS: The most important features of the application for physicians were selecting clinical signs and symptoms, predicting diagnosis based on artificial intelligence, physician-patient communication, selecting the appropriate treatment, and easy access to scientific resources. The results of the usability evaluation showed that the application was good with a mean (± SD) score of 7.73 ± 0.53 out of 10. CONCLUSION: A decision support system with accuracy, precision, sensitivity and specificity, negative predictive values, F-measure and area under precision-recall curve 100% was created thanks to widespread participation, the use of clinical specialists' experiences and their awareness of patients' needs, as well as the availability of a comprehensive acute postoperative endophthalmitis clinical dataset.

Design and usability evaluation of a mobile-based-self-management application for caregivers of children with severe burns.

Paediatric burns are a major public health issue because of long-term physical, psychological and social consequences and the high cost of treatment. The aim of this study was to design and evaluate a mobile-based self-management application for caregivers of children with severe burns. A participatory design technique was employed to develop the Burn application, which included three main phases: the determination of application requirements, the design and evaluation of the low-fidelity prototype, and the design and evaluation of the high-fidelity prototypes. In the first phase, application requirements were determined via validated paper questionnaires using the Delphi technique. In the second step, a low-fidelity prototype was prepared using conceptual models and evaluated through a focus group with specialists. Seven specialists reviewed the application and evaluated how this prototype meets functional requirements and objectives. The third phase was performed in three stages. First, the high-fidelity prototype was designed and developed by the JAVA programming language. Second, a cognitive walk-through was carried out to show how users can interact with the mobile application and how it works. Third, this program was installed on the mobile phones of 28 caregivers of burned children, eight IT experts, and two general surgeries, and the prototype's usability was evaluated. In the present study, most caregivers of children with burns stated that after discharge, they face problems regarding infection control and wound care (4.07) and how to perform physical activity (4.12). User registration, educational materials, caregiver-clinician communication, chat box, and appointment booking, safe login were the most important characteristic of the Burn application. Mean usability evaluation scores were in the range of 7.92 ± 0.238 to 8.10 ± 0.103, which is considered at a "good" level. From the Burn program design experience, it can be concluded that co-design with health care specialists can significantly support and meet the specialists' and patients' needs and ensure the program's usefulness. In addition, application evaluation by users involved and not involved in the application design process can help enhance usability.

The relationship between user interface problems of an admission, discharge and transfer module and usability features: a usability testing method.

BACKGROUND: The admission, discharge and transfer (ADT) module is used in the hospital information system (HIS) for the purposes of managing appointments, patient admission, daily control of hospital beds, planning surgery procedures, keeping up-to-date on patient discharges, and registering patient transfers within or outside the hospital. The present study aimed to evaluate the usability of ADT module of a HIS through usability testing and assess the relationship between the number of user interface problems and usability features (i.e. effectiveness, efficiency, and satisfaction). METHODS: This descriptive analytical study was conducted in Shahid Beheshti hospital in Kashan, Iran, in 2017. The participating users were eight students in their last semester of a Bachelor of Health Information Technology Sciences degree. First, the users were introduced to the module functions in a two-hour session; ten days later, the users were asked to perform scenarios designed based on seven tasks and take notes of the problems encountered in performing each task after it was over. Effectiveness was measured based on the rate of completing the tasks, efficiency based on the time taken to perform each task, and satisfaction based on the users' answers to a satisfaction questionnaire. The relationship between these three usability features and the number of problems noted was assessed using Spearman's test in SPSS version 16. RESULTS: Thirteen unique usability problems were identified from the perspective of the users. Effectiveness was rated as 58.9%, efficiency as 53.3%, and mean user satisfaction as 53.4 ± 10.6. The number of problems in each task had significant relationships to the effectiveness (P = 0.009) and efficiency (P = 0.016) scores. User satisfaction also had a significant relationship with the effectiveness (P = 0.043) but not with the efficiency (P = 0.230) scores. CONCLUSIONS: In the view of the potential users, a HIS, used in more than 200 hospitals in a developing country, has several usability problems in its ADT module and its effectiveness, efficiency, and user satisfaction were not acceptable. The number of usability problems in the HIS user interface affected the effectiveness, efficiency and user satisfaction of the system.

Identifying and prioritizing the key performance indicators for hospital management dashboard at a national level: Viewpoint of hospital managers.

Participation of main users in identifying key performance indicators (KPIs) for management dashboards contributes to their success. The aim of this study was to identify and prioritize the KPIs of hospital management dashboards from the viewpoint of hospital managers. This study was conducted on managers of public hospitals at a national level in Iran in 2020. Data were collected using a self-administrated questionnaire. The KPIs were classified into five categories, namely financial, operational, human resources, safety and quality of care, services provided to patients. A total of 234 hospital managers participated in this study. Totally, 25 KPIs were determined for the hospital management dashboard, including the patient falls rate, waiting time for patients in the emergency department, patient satisfaction, total hospital revenue, financial balance, bed occupancy rate, patients' discharge with own agreement, average length of stay, and personnel satisfaction. For designing hospital management dashboards, the domains of services provided to patients, safety and quality of care, financial resources, human resources, and operational are important from the hospital managers' viewpoint, respectively. The results of this study can be helpful for developers of business intelligence tools, such as hospital management dashboards, to visualize the most important indicators for managers.

Effects of Computer-Aided Decision Support Systems on Appropriate Antibiotic Prescribing by Medical Interns: A Quasi-Experimental Study.

Literature suggests that the adoption of guidelines for antibiotic prescribing has a significant impact on improving prescription practices of physicians; thus, this study aimed to assess the effectiveness of computer-aided decision support systems (CA-DSS) on antibiotic prescribing among medical interns. A prospective before-and-after interventional study was conducted on 40 medical interns. The interns were asked to use the CA-DSS during a one-month internship course at the infectious disease department. The main outcome measure was the knowledge of medical interns regarding the type, name, volume, usual dosages, and administration route of antibiotics prescribed. Paired t-test was applied to assess the change of medical interns' knowledge before and after the study. There was a statistically significant difference between the mean score of interns' medical knowledge before 5.4±2 and after 9.1±2.8 using the CA-DSS (p = 0.000). CA-DSS as an IT-based training intervention was effective for the knowledge of medical interns to prescribe the right antibiotics for acute respiratory infections.

Information technology interventions to improve antibiotic prescribing for patients with acute respiratory infection: a systematic review.

OBJECTIVES: Information technology (IT) interventions provide physicians with easy and quick access to information at the point of care and can play a major role in clinical decision-making for antibiotic prescribing. This study aimed to examine the effects and characteristics of IT interventions on improving antibiotic prescribing for patients with acute respiratory infection (ARI). METHODS: A comprehensive search was performed in Medline (through PubMed), ISI web of science, Embase, and Cochrane databases from inception to 31 August 2020. Randomized controlled trial (RCT) and cluster RCT (CRCT) studies examining the effectiveness of IT interventions in improving antibiotic prescribing for patients with ARI were included. Participants were patients with ARI. IT interventions were used for improving antibiotic prescribing. Two researchers independently extracted data from studies on methods, characteristics of interventions, and results. The characteristics of interventions were extracted based on three dimensions of IT design, data entry source, and implementation characteristics. RESULTS: Eighteen studies (15 CRCTs and three RCTs) were included. Most of included studies (n = 11) were conducted in the United States. In 12 studies (66.7%), IT interventions improved the level of antibiotic prescribing, and in eight of the 12 studies the effect was statistically significant. In two studies the intervention had a statistically significant negative effect, and in two studies the level of antibiotic prescribing was not changed. Seventeen studies (94.4%) used clinical decision support systems (CDSSs) for the intervention. In 12 studies (66.7%) CDSSs were integrated with electronic health records (EHRs). CONCLUSIONS: Information technology interventions have the potential to improve prescription of antibiotics for patients with acute respiratory infection and to change physicians' behaviours in this regard. Factors affecting the acceptance of IT-based interventions to improve prescription of antibiotics should be investigated in future studies.

Design, development, and evaluation of an interactive training simulator for teaching hospital information systems.

BACKGROUND: In their apprenticeship program, health information technology (HIT) students are deprived of the ample opportunity to work with the hospital information system (HIS). This study aimed to design an interactive simulator for the HIS training and evaluate its effects on the informatics skills of HIT students. MATERIALS AND METHODS: This study was conducted on 16 Bachelor of Science students of HIT at Kashan University of Medical Sciences in 2019. After the functionalities and features of the simulator were determined based on similar existing simulators, expert opinions were received to simulate eight important processes of admission, discharge, and transfer module in HIS. The scores of students' skills and time taken to perform the processes were recorded and analyzed before and after the educational intervention. After they were trained by the simulator, the students filled out a usability evaluation questionnaire. The data were then analyzed in SPSS version 21. RESULTS: The simulators of health information systems were characterized by interactivity, multimedia applications, practice exercises, tests, and feedback. After the students were trained by the developed simulator, their skills scores improved significantly in 75% (6/8) of the processes, and the timespans of all processes decreased significantly (P < 0.05). The usability evaluation indicated the usability of the simulator was at a "good" level. CONCLUSIONS: According to the study results, using the simulator improves the informatics skills of HIT students in working with HIS. It is recommended that this method also be used in other apprenticeship programs to teach health information systems.

Towards a Comprehensive Quality Evaluation Model for Hospital Websites.

BACKGROUND: Hospital websites are important sources for patients to access health information. AIM: The purpose of this study was to develop the quality evaluation model for hospital websites. METHODS: The quantitative study was conducted through the modified Delphi method in 2014-2015. The population of the study includes 10 experts were chosen by targeting non-randomized method. A questionnaire was prepared based on the prototype that was developed by research papers and related models. The validity of the questionnaires was confirmed by face validity and CVI and CVR estimation. Reliability was obtained by split-half method (α = 0.8). Experts' opinions were collected through interview. Then, their frequency and percentage were determined. Items with options completely agree and agree over 75% was approved, items below 50% were removed, and items 50%-75% were removed after three interviews repetitions. RESULTS: Most of the experts agreed about the pleasant and harmonious colors, the readable and consistent fonts (100%). The least frequency was allocated to correct grammar and words, support for multilingual and rapidly changing of displaying pages in the website with a frequency of 2 (20%). CONCLUSION: The minimum qualitative criteria for a website are usability, efficiency, user friendly, service, reliability and interaction.