Tristan Ling

Applying MCRDR to a multidisciplinary domain

This paper details updated results concerning an implementation of a Multiple Classification Ripple Down Rules (MCRDR) system which can be used to provide quality Decision Support Services to pharmacists practicing medication reviews (MRs), particularly for high risk patients. The system was trained on 126 genuine cases by an expert in the field; over the course of 19 hours the system had learned 268 rules and was considered to encompass over 80% of the domain. Furthermore, the system was found able to improve the quality and consistency of the medication review reports produced, as it was shown that there was a high incidence of missed classifications under normal conditions, which were repaired by the system automatically. However, shortcomings were identified including an inability to handle absent data, and shortcomings concerning standardization in the domain, proposals to solve these shortcomings are included.

Expert-Driven Knowledge Discovery

Knowledge discovery techniques find new knowledge about a domain by analysing existing domain knowledge and examples of domain data. These techniques typically involve using a human expert and automated software analysis (data mining). Often the human expertise is used initially to choose which data is processed, and then finally to determine which results are relevant. However studies have noted that some domains contain data stores too extensive and detailed, and existing knowledge too complex, for effective data selection or efficient data mining. A different approach is suggested which involves the human expert more pervasively, taking advantage of their expertise at each step, while using data mining techniques to assist in discovering data trends and in verifying the experts findings. Preliminary results suggest that the approach can be successfully applied to discover new knowledge in a complex domain, and reveal many potential areas for research and development.

A Computer Simulation of Community Pharmacy Practice for Educational Use

Objective. To provide a computer-based learning method for pharmacy practice that is as effective as paper-based scenarios, but more engaging and less labor-intensive. Design. We developed a flexible and customizable computer simulation of community pharmacy. Using it, the students would be able to work through scenarios which encapsulate the entirety of a patient presentation. We compared the traditional paper-based teaching method to our computer-based approach using equivalent scenarios. The paper-based group had 2 tutors while the computer group had none. Both groups were given a prescenario and postscenario clinical knowledge quiz and survey. Assessment. Students in the computer-based group had generally greater improvements in their clinical knowledge score, and third-year students using the computer-based method also showed more improvements in history taking and counseling competencies. Third-year students also found the simulation fun and engaging. Conclusion. Our simulation of community pharmacy provided an educational experience as effective as the paper-based alternative, despite the lack of a human tutor.

An Internet-Based Method for Extracting Nursing Home Resident Sedative Medication Data From Pharmacy Packing Systems: Descriptive Evaluation

Background Inappropriate use of sedating medication has been reported in nursing homes for several decades. The Reducing Use of Sedatives (RedUSe) project was designed to address this issue through a combination of audit, feedback, staff education, and medication review. The project significantly reduced sedative use in a controlled trial of 25 Tasmanian nursing homes. To expand the project to 150 nursing homes across Australia, an improved and scalable method of data collection was required. This paper describes and evaluates a method for remotely extracting, transforming, and validating electronic resident and medication data from community pharmacies supplying medications to nursing homes. Objective The aim of this study was to develop and evaluate an electronic method for extracting and enriching data on psychotropic medication use in nursing homes, on a national scale. Methods An application uploaded resident details and medication data from computerized medication packing systems in the pharmacies supplying participating nursing homes. The server converted medication codes used by the packing systems to Australian Medicines Terminology coding and subsequently to Anatomical Therapeutic Chemical (ATC) codes for grouping. Medications of interest, in this case antipsychotics and benzodiazepines, were automatically identified and quantified during the upload. This data was then validated on the Web by project staff and a “champion nurse” at the participating home. Results Of participating nursing homes, 94.6% (142/150) had resident and medication records uploaded. Facilitating an upload for one pharmacy took an average of 15 min. A total of 17,722 resident profiles were extracted, representing 95.6% (17,722/18,537) of the homes’ residents. For these, 546,535 medication records were extracted, of which, 28,053 were identified as antipsychotics or benzodiazepines. Of these, 8.17% (2291/28,053) were modified during validation and verification stages, and 4.75% (1398/29,451) were added. The champion nurse required a mean of 33 min website interaction to verify data, compared with 60 min for manual data entry. Conclusions The results show that the electronic data collection process is accurate: 95.25% (28,053/29,451) of sedative medications being taken by residents were identified and, of those, 91.83% (25,762/28,053) were correct without any manual intervention. The process worked effectively for nearly all homes. Although the pharmacy packing systems contain some invalid patient records, and data is sometimes incorrectly recorded, validation steps can overcome these problems and provide sufficiently accurate data for the purposes of reporting medication use in individual nursing homes.

RedUSe: reducing antipsychotic and benzodiazepine prescribing in residential aged care facilities

Objective: To assess the impact of a multi‐strategic, interdisciplinary intervention on antipsychotic and benzodiazepine prescribing in residential aged care facilities (RACFs).

More action needed: Psychotropic prescribing in Australian residential aged care

Objective: For at least two decades, concerns have been raised about inappropriate psychotropic prescribing in Australian residential aged care facilities, due to their modest therapeutic benefit and increased risk of falls and mortality. To date, the majority of prevalence data has been collected in Sydney exclusively and it is not known if recent initiatives to promote appropriate psychotropic prescribing have impacted utilisation. Thus, we aimed to comprehensively analyse psychotropic use in a large national sample of residential aged care facility residents. Method: A cross-sectional, retrospective cohort study of residents from 150 residential aged care facilities distributed nationally during April 2014–October 2015. Antipsychotic, anxiolytic/hypnotic and antidepressant utilisation was assessed, along with anticonvulsant and anti-dementia drug use. Negative binomial regression analysis was used to examine variation in psychotropic use. Results: Full psychotropic prescribing data was available from 11,368 residents. Nearly two-thirds (61%) were taking psychotropic agents regularly, with over 41% prescribed antidepressants, 22% antipsychotics and 22% of residents taking benzodiazepines. Over 30% and 11% were charted for ‘prn’ (as required) benzodiazepines and antipsychotics, respectively. More than 16% of the residents were taking sedating antidepressants, predominantly mirtazapine. South Australian residents were more likely to be taking benzodiazepines (p < 0.05) and residents from New South Wales/Australian Capital Territory less likely to be taking them (p < 0.01), after adjustment for rurality and size of residential aged care facility. Residents located in New South Wales/Australian Capital Territory were also significantly less likely to take antidepressants (p < 0.01), as were residents from outer regional residential aged care facilities (p < 0.01). Antipsychotic use was not associated with State, rurality or residential aged care facility size. Conclusion: Regular antipsychotic use appears to have decreased in residential aged care facilities but benzodiazepine prevalence is higher, particularly in South Australian residential aged care facilities. Sedating antidepressant and ‘prn’ psychotropic prescribing is widespread. Effective interventions to reduce the continued reliance on psychotropic management, in conjunction with active promotion of non-pharmacological strategies, are urgently required.

Multiple Classification Ripple Round Rules: A Preliminary Study

This paper details a set of enhancements to the Multiple Classification Ripple Down Rules methodology which enable the expert to create rules based on the existing presence of a conclusion. A detailed description of the method and associated challenges are included as well as the results of a preliminary study which was undertaken with a dataset of pizza topping preferences. These results demonstrate that the method loses none of the appeal or capabilities of MCRDR and show that the enhancements can see practical and useful application even in this simple domain.

A Fast Heuristic Algorithm for Similarity Search in Large DNA Databases

Knowledge Discovery techniques seek to find new information about a domain. These techniques can either be manually performed by an expert, or automated using software algorithms (Machine Learning). However some domains (such as the field of lung function testing) contain volumes of data too vast for effective manual analysis, and require background knowledge too complex for Machine Learning algorithms. This study examines how the Multiple Classification Ripple-Down Rules (MCRDR) Knowledge Acquisition process can be adapted to develop a new Knowledge Discovery method, Exposed MCRDR. A prototype system was developed and tested in the domain of lung function. Preliminary results suggest that the EMCRDR method can be successfully applied to efficiently discover new knowledge in a complex domain. The study also reveals many potential areas of study and development for the MCRDR method, and Knowledge Acquisition and Knowledge Discovery methods in general.

Simulation and Feedback in Health Education: A Mixed Methods Study Comparing Three Simulation Modalities

Background. There are numerous approaches to simulating a patient encounter in pharmacy education. However, little direct comparison between these approaches has been undertaken. Our objective was to investigate student experiences, satisfaction, and feedback preferences between three scenario simulation modalities (paper-, actor-, and computer-based). Methods. We conducted a mixed methods study with randomized cross-over of simulation modalities on final-year Australian graduate-entry Master of Pharmacy students. Participants completed case-based scenarios within each of three simulation modalities, with feedback provided at the completion of each scenario in a format corresponding to each simulation modality. A post-simulation questionnaire collected qualitative and quantitative responses pertaining to participant satisfaction, experiences, and feedback preferences. Results. Participants reported similar levels satisfaction across all three modalities. However, each modality resulted in unique positive and negative experiences, such as student disengagement with paper-based scenarios. Conclusion. Importantly, the themes of guidance and opportunity for peer discussion underlie the best forms of feedback for students. The provision of feedback following simulation should be carefully considered and delivered, with all three simulation modalities producing both positive and negative experiences in regard to their feedback format.

Problems detected by a ripple-down rules based medication review decision support system: are they relevant?

A ripple-down rules based clinical decision support system to detect drug-related problems (DRPs) has been previously designed and discussed. A commercial implementation of this system (MRM) was evaluated to determine how many additional DRPs would be identified by the reviewing pharmacist when supported by MRM, and whether these additional DRPs were clinically relevant. The DRPs identified by pharmacists were compared against those found by MRM on a dataset of 570 medication review cases, MRM found 2854 DRPs, pharmacists found 1974 DRPs, yet only 389 of the problems that MRM found were also found by the pharmacist. A sample of 20 of these cases were assessed by an expert panel to determine if the DRPs found by each source were clinically relevant. It was determined that DRPs found by both sources were clinically relevant. It is estimated that a pharmacist supported by MRM will find 2.25 times as many DRPs.