Prediction of aminoglycoside response against methicillin-resistant Staphylococcus aureus infection in burn patients by artificial neural network modeling.

OBJECTIVE: To predict the response of aminoglycoside antibiotics (arbekacin: ABK) against methicillin-resistant Staphylococcus aureus (MRSA) infection in burn patients after considering the severity of the burn injury by using artificial neural network (ANN). Predictive performance was compared with logistic regression modeling. METHODOLOGY: The physiologic data and some indicators of the severity of the burn injury were collected from 25 burn patients who received ABK against MRSA infection. A three-layered ANN architecture with six neurons in the hidden layer was used to predict the ABK response. The response was monitored using three clinical criteria: number of bacteria, white blood cell count, and C-reactive protein level. Robustness of models was investigated by the leave-one-out cross-validation. RESULTS: The peak plasma level, serum creatinine level, duration of ABK administration, and serum blood sugar level were selected as the linear input parameters to predict the ABK response. The area of the burn after skin grafting was the best parameter for assessing the severity of the burn injury in patients to predict the ABK response in the ANN model. The ANN model with the severity of the burn injury was superior to the logistic regression model in terms of predicting the performance of the ABK response. CONCLUSION: Based on the patients' physiologic data, ANN modeling would be useful for the prediction of the ABK response in burn patients with MRSA infection. Severity of the burn injury was a parameter that was necessary for better prediction.

Contributions of pharmacists through the promotion of proper drug use.

The contents of pharmacist interventions, which were carried out by the ward pharmacists in their routine pharmacy service activities, were sorted and analyzed to evaluate the contributions of pharmacists. In the ward where pharmacists were stationed, there were a total of 196 cases of pharmacist intervention. The prescription was changed in 170 cases, giving a rate of prescription change of 86.7%. The breakdown of the pharmacist intervention was as follows: "efficacy/safety", 106 cases, followed by "dosage regimen" (48 cases) and "compliance" (10 cases). Cost savings achieved during the investigation period were calculated to be 440,639 yen, and cost avoidance was valued at 1,941,847-3,883,695 yen using the Diagnosis Procedure Combination (DPC). The results of the present investigation showed that pharmacists contribute to through not only their pharmacy services, but also through the promotion of proper drug use and risk management, thereby contributing to hospital management through cost savings and avoidance.

Structural equation modeling of qualification of pharmacists to improve subjected quality of life in cancer patients.

PURPOSE: To establish structural equation model (SEM) of subjected quality of life (QOL) in cancer patients taking into account qualification of pharmacists. METHOD: The SEM model was constructed from correlation matrix of the scores of answers of questions to both patients and pharmacists. Data were collected from 15 cancer patients who hospitalized and took opioid analgesics for pain control. The patients were asked 18 questions and pharmacists were asked seven questions. From the correlation matrix among scores of answers, a reasonable model was explored by SEM. RESULTS: Health-related QOL (HRQOL) in cancer patients can be modeled by latent variables consist of contributions from physical, emotional and functional domains. The fitting between data and the model was acceptable by statistical goodness-of-fit (GOF) index. The modeled HRQOL by SEM was weakly correlated with subjected QOL in patients, indicating that subjected QOL in patients would be affected not only by above latent variables but other variables. The model taking into account qualification of pharmacists to improve subjected QOL in patients was also made by SEM. The model was reasonably explained and fitting between data and the model was acceptable from some statistical index. The final model suggests that pharmacist can raise subjected QOL in patients through restraining unpleasant side effects. CONCLUSION: The qualification of pharmacists to improve subjected QOL in patients can be modeled by SEM. The final model suggests that pharmacists with qualification to assess patients' pain status contribute to raise subjected quality of life in cancer patients.

Artificial neural network modeling to predict the plasma concentration of aminoglycosides in burn patients.

The goal was to use an artificial neural network model to predict the plasma concentration of aminoglycosides in burn patients and identify patients whose plasma antibiotic concentration would be sub-therapeutic based on the patients' physiological data and taking into account burn severity. Physiological data and some indicators of burn severity were collected from 30 burn patients who received arbekacin. A three-layer artificial neural network with five neurons in the hidden layer was used to predict the plasma concentration of arbekacin. Linear modeling for prediction of plasma concentration and logistic regression modeling for the classification of patients were also used and the predictive performance was compared to results from the artificial neural network model. Dose, body mass index, serum creatinine concentration and amount of parenteral fluid were selected as covariates for the plasma concentration of arbekacin. Area of burn after skin graft was a good covariate for indicating burn severity. Predictive performance of the artificial neural network model including burn severity was much better than linear modeling and logistic regression analysis. An artificial neural network model should be helpful for the prediction of plasma concentration using patients' physiological data, and burn severity should be included for improved prediction in burn patients. Because the relationship between burn severity and plasma concentration of aminoglycosides is thought to be nonlinear, it is not surprising that the artificial neural network model showed better predictive performance compared to the linear or logistic regression models.