An Incremental EM-based Learning Approach for On-Line Prediction of Hospital Resource Utilization

Objective: Inpatient length of stay (LOS) is an important measure of hospital activity, health care resource consumption, and patient acuity. This research work aims at developing an incremental expectation maximization (EM) based learning approach on mixture of experts (ME) system for on-line prediction of LOS. The use of a batchmode learning process in most existing artificial neural networks to predict LOS is unrealistic, as the data become available over time and their pattern change dynamically. In contrast, an on-line process is capable of providing an output whenever a new datum becomes available. This on-the-spot information is therefore more useful and practical for making decisions, especially when one deals with a tremendous amount of data. Methods and material: The proposed approach is illustrated using a real example of gastroenteritis LOS data. The data set was extracted from a retrospective cohort study on all infants born in 1995-1997 and their subsequent admissions for gastroenteritis. The total number of admissions in this data set was n = 692. Linked hospitalization records of the cohort were retrieved retrospectively to derive the outcome measure, patient demographics, and associated co-morbidities information. A comparative study of the incremental learning and the batch-mode learning algorithms is considered. The performances of the learning algorithms are compared based on the mean absolute difference (MAD) between the predictions and the actual LOS, and the proportion of predictions with MAD < 1 day (Prop(MAD < 1)). The significance of the comparison is assessed through a regression analysis. Results: The incremental learning algorithm provides better on-line prediction of LOS when the system has gained sufficient training from more examples (MAD = 1.77 days and Prop(MAD < 1) = 54.3%), compared to that using the batch-mode learning. The regression analysis indicates a significant decrease of MAD (p-value = 0.063) and a significant (p-value = 0.044) increase of Prop(MAD

Psychotropic medication utilization in a child and adolescent mental health service

Objectives: This study aimed to identify rates and correlates of psychotropic drug utilization in children and adolescents in inpatient and outpatient settings. Methods: A retrospective chart review examined 122 inpatient and 126 outpatient charts from a metropolitan child and youth mental health service in Brisbane, Australia. Results: Inpatients received more psychotropic medication than outpatients (71% vs. 25%; p < 0.01). Patients receiving medication were older, had longer hospital admissions, and more complex presentations, including history of abuse or suicide attempts and more diagnoses (all p < 0.01). Selective serotonin reuptake inhibitors (SSRIs) were the most frequently used drug class (44% inpatients; 14% outpatients), primarily indicated for mood disorders (31%). SSRIs and newer antidepressants (ADs) were used more frequently in patients with a high suicide risk (p < 0.01). Atypical antipsychotics (APs) were also used (inpatients 23%; outpatients 3%), primarily for behavioral disturbances. Half of those receiving medication (51%) received polypharmacy (> 1 concurrent drug), with up to four drugs used at one time. Rates of polypharmacy were highest among patients receiving antipsychotics. Conclusions: Use of psychotropic medication is frequent in this population. Future research should initially focus on inpatients and intensive treatment settings and examine both safety and efficacy of interventions for depression in young people, atypical antipsychotics for behavioral disturbances, and polypharmacy.

The judicious utilization of new technologies through authentic learning in higher education: A case study

Though technology holds significant promise for enhanced teaching and learning it is unlikely to meet this promise without a principled approach to course design. There is burgeoning discourse about the use of technological tools and models in higher education, but much of the discussion is fixed upon distance learning or technology based courses. This paper will develop and propose a balanced model for effective teaching and learning for “on campus” higher education, with particular emphasis on the opportunities for revitalisation available through the judicious utilisation of new technologies. It will explore the opportunities available for the creation of more authentic learning environments through the principled design. Finally it will demonstrate with a case study how these have come together enabling the creation of an effective and authentic learning environment for one pre-service teacher education course at the University of Queensland.