Ebook: Connecting the System to Enhance the Practitioner and Consumer Experience in Healthcare
Health informatics plays a central role in the digital transformation of the healthcare sector. The integration and connection of health services, practitioners and consumers is critical to the realisation of the improvements promised by digital health, and the secondary use of health data has led to ground-breaking research discoveries. Increased reliance upon all types of digital media has also established health informatics as a viable specialisation in healthcare.
This book presents the proceedings of the 26th national Health Informatics Conference (HIC 2018), Australia’s premier health informatics event, held in Sydney, Australia, in July/August 2018. The conference provides an environment for clinicians, researchers, health IT professionals, industry and consumers to integrate, educate and share the knowledge which drives innovative thinking, and this year’s theme, ‘Today’s best practice, innovation today and preparing for tomorrow’, focuses on the important issues of connecting the system, being smart with data, and enhancing the practitioner and consumer experience in healthcare interactions. The papers presented here reflect this theme, highlighting cutting-edge research evidence, technology updates and innovations from the digital transformation of the healthcare sector.
Covering a wide spectrum of work, and encompassing major theoretical concepts, examples of key applications of new technologies and important new developments in the field of health informatics, the book will be of interest to all those working in the healthcare sector.
Integrating and connecting health services, practitioners and consumers is the critical step in realising the improvements promised by digital health. This is seen in the way the health informatics community is taking on a central role in the digital transformation of the healthcare sector, with health informatics roles being increasingly recognised as critical to that success. This is further emphasised in Safe, Seamless and Secure (Australia's National Digital Health Strategy), where a “workforce confidently using digital health technologies to deliver health and care” is acknowledged as one of the seven strategic priority outcomes to be achieved by 2022
The papers in this volume reflect this theme, highlighting the cutting-edge research evidence, technology updates and innovations that are seeing the digital transformation of the healthcare sector. The papers are indicative of the wide spectrum of work encompassing major theoretical concepts, examples of key applications of new technologies and important new developments in the field of health informatics.
This year's program maintains the high standard of papers for which the conference is well-known. All papers were blind-peer reviewed by three experts in the field of health informatics. These reviewers are widely considered to be prominent academics, digital health and industry specialists. The contribution of the Australasian College of Health Informatics, particularly the voluntary participation of Fellows, in supporting this review process is gratefully acknowledged. Similar contributions made by many senior and experienced members of the Health Informatics Society of Australia are also acknowledged. Forty papers underwent the initial review and feedback process. Resubmitted papers were then validated by the Scientific Program Committee to ensure that reviewers' recommendations were appropriately addressed or rebutted. In total 30 papers were selected for inclusion in this volume. We congratulate all the authors on their contribution and commend them to the readers.
Louise K. Schaper
To be considered fit for use in clinical care, health data should meet quality requirements, and data generated in remote patient monitoring settings are not exempt. Since patients take part of the responsibility in ensuring the quality of data during their flow from patient to the clinical setting, the quality management of these data is of great importance. This study aims to systematically review the literature to understand the current situation of quality management of patient generated data in remote patient monitoring particularly in use of wearable medical devices.
The rapid increase in the number of older adults in developed countries has raised concerns about their well-being and increasing need for healthcare. New technologies, including Internet of Things, are being used to monitor older adults' health and activities, thus enabling them to live safely and independently at home as they age. However, Internet of Things monitoring solutions create privacy challenges that need to be addressed. This review examines how privacy has been conceptualised in studies proposing new Internet of Things solutions for monitoring older adults. The literature reviewed mostly links privacy with information security and unauthorised accessibility threats. There is a limited consideration of other aspects of privacy such as confidentiality and secondary use of users' information. We argue that developers of Internet of Things solutions that aim to monitor and collect health data about older adults need to adopt an expanded view of privacy. This will ensure that safeguards are built in to Internet of Things devices to protect and maintain users' privacy while also enabling the appropriate sharing of data to support older adults' safety and wellbeing.
Internationally, shared digital health records are becoming an important addition to improving contemporary healthcare provision. In 2012, Australia launched its version of a shared digital health record, My Health Record, but enrolment is slow and there remain challenges in its practical implementation. Further, people living with complex and chronic conditions in rural and remote communities often experience challenges in obtaining equitable access to contemporary healthcare provision, including eHealth services. This paper reports on research that explored the experience of and engagement with My Health Record, in a rural Australian community setting. Based on the key research findings, recommendations are presented for improving national roll out of My Health Record. The findings highlight, to understand and engage vulnerable communities and support their adoption and use of shared digital health records, there is a need to move away from traditional models of healthcare delivery toward person-centred care delivered from a digital complex adaptive systems perspective.
Identification and prediction of patients who are at risk of hospital readmission is a critical step towards the reduction of the potential avoidable costs for healthcare organisations. This research was focused on the evaluation of LACE Index for Readmission – Length of stay (days), Acute (emergent) admission, Charlson Comorbidity Index and number of ED visits within six months (LACE) and Patients At Risk of Hospital Readmission (PARR) using New Zealand hospital admissions. This research estimates the risk for all readmissions rather than only those in a subset of referenced conditions. In total, 213,440 admissions between 1 Jan 2015 and 31 Dec 2016 were selected after appropriate ethics approvals and permissions from the three hospitals. The evaluation method used is the Receiver Operating Characteristics (ROC) analysis to evaluate the accuracy of both the LACE and PARR models. As a result, The LACE index achieved an Area Under the Curve (AUC) score of 0.658 in predicting 30-day readmissions. The optimal cut-off for the LACE index is a score of 7 or more with sensitivity of 0.752 and specificity of 0.564. Whereas, the PARR algorithm achieved an AUC score of 0.628 in predicting 30-day readmissions. The optimal cut-off for the PARR index is a score of 0.34 or more with sensitivity of 0.714 and specificity of 0.542.
Computerised dose prediction software assist clinicians in undertaking therapeutic drug monitoring by providing individualised dosing recommendations, typically communicated to prescribers in the form of a report. These software are highly sophisticated and accurate in predicting individualised dosage regimens, but if the information contained in the report is not understood by prescribers, the benefits of the software are not achieved. In this study, we set out to assess the perceived usability of a report generated from a dose prediction system. Fifteen prescribers were presented with a mock report and asked a number of questions to elicit their views of the report's content and design. Overall, we found that the mock report was effective in communicating the recommended dose of a drug, but this recommendation was presented alongside information that was not understood or was unlikely to be utilised by prescribers. In particular, the aspects of the report viewed negatively by end-users largely related to a lack of familiarity with the pharmacological terminology used in the report, which hindered understanding and caused confusion. Involving prescribers early on in the process of designing decision support systems is likely to result in systems and outputs that are more useful, usable and accessible to users.
Background. Few studies of sensor-based falls detection devices have monitored older people in their care settings, particularly in Australia. The present investigation addressed this gap by trialling the feasibility and acceptability of a non-contact smart sensor system (NCSSS) to monitor behaviour and detect falls in an Australian residential aged care facility (RACF).
Methods. This study used a mixed methods approach: a) Pilot study implementation at a RACF, b) Post-pilot interviews, c) Analysis and review of results.
Results and discussion. Data was collected for four RACF participants over four weeks of the NCSSS pilot. No falls were recorded during the uptime of the system. Numerous feasibility challenges were encountered, for example in the installation, configuration, and location of sensors for optimal detection, network and connectivity issues, and maintenance requirements. These factors may affect NCSSS implementation and adherence.
Apart from drug control, diabetes management should pay more attention to how lifestyle and daily routine affect blood sugar. This study aimed to explore the impact of m-health programme using a minimal psychological intervention on diabetes patients' knowledge, behavioural, and psychological health. A pretest-posttest single-group pre-experimental study was undertaken with 30 individuals to recruit patients with diabetes. A total of 22 participants completed the 10-week online programme. The pretest-posttest results demonstrated the differences in health behaviour, including foot care, diet control, and exercise. The adoption of m-health that changed and improved patient's self-care; however, the acceptance of using m-health remains a challenge for public dissemination.
Adherence to medication and treatment regimen continues to rank as one of the major clinical problems worldwide. Many of the currently available strategies lack the personalised interaction needed in order to effectively connect with the patients. Our solution is to develop an interactive avatar-based reminder application to simulate a human carer to provide customised conversations with the aim of improving adherence and satisfaction. The results of our initial trial showed that information and communication provided by the avatar were useful in reminding participants to take their health supplements. This paper discusses the importance of personalised communication and information in improving adherence and satisfaction, what we have learned from our initial trial with the avatar-based reminder application, and our proposed modifications for our next trial.
Automated conversational agents built with medical applications in mind, have the potential to reduce healthcare readmissions and improve accessibility to medical knowledge. In this work, we demonstrate the development and evaluation of an automated chatbot for triage and conditions assessment, based on user inputs in natural language. The implemented bot engages patients in conversation about symptoms experienced and provides a personalized pre-synopsis based on their symptoms and profile. Our chatbot system was able to predict user conditions correctly based on two sets of patient test cases with an average precision of 0.82. Our implementation demonstrates that a medical chatbot can help with automatic triage and pre-assessment of patients with simple symptom analysis and a conversational approach without the use of cumbersome form-based data entry.
Background: Large datasets of outcomes in cardiac rehabilitation (CR) are not often analysed using a combination of primary presentations and comorbidities, which has left important issues as yet unresolved. This study investigates the influence of history of infarction, measured pre-program resting blood pressure, and comorbidities including diagnosed hypertension, hypercholesterolemia, diabetes and obesity on CR outcomes in patients without a history of myocardial infarction (MI).
Methods: Retrospective exercise-based outpatient CR data were reviewed within the study (n = 4,202), collected between January 2000 and December 2011, consisting of observations made during pre- and post-program patient assessment processes. Resting heart rate (HR) and systolic (SBP) and diastolic (DBP) blood pressure observations in those without a history of MI (n = 1,703) were corrected for gender, assessed pre-program resting blood pressure, and comorbidities including diagnosed hypertension, hypercholesterolemia, diabetes and obesity (assessed BMI > 30).
Results: Gender differentials were observed in post-program outcomes. Additionally, assessed pre-program hypotensives (SBP < 100 mmHg) showed blood pressure increases while hypertensives (SBP > 130 mmHg) showed decreases post-program. These blood pressure outcomes were also influenced by the number of comorbidities, which further clarified post-program results and revealed potentially important trends.
Conclusion: Despite the limitations of administrative datasets, this study has demonstrated significant value in analysing such outpatient CR program data. The corrections applied in this retrospective review demonstrate the importance of considering gender, primary presentations and comorbidities when interpreting outcomes, as well as implying the need for a consideration of these factors in exercise prescription, towards pathology-specific CR and improving outcomes. More research is warranted.
There is a significant amount of anecdotal evidence of the benefits of individuals on the autism spectrum interacting with technology via natural language. Many of these individuals are non-verbal but still able to communicate via augmentative and alternative communication (AAC) aids. This paper presents the design of a AAC software program (app) with embedded artificial conversational agent, called Alex. Alex runs on a Android device and is able to engage with the user on a variety of topics using symbols and images. Alex may be programmed via speech and occupational therapists and other key stakeholders via speech and hence does not require any specialised computer skills. The importance of customisation, interoperability, personalisation and motor skill considerations is herein discussed.
Healthcare Social Question Answering (SQA) services provide an open and free way to share knowledge and experience about health related enquires. Examples of healthcare SQA services are MedHelp, BabyHub and Drugs.com. The quality and the source of the questions and answers vary widely. Hence, we work towards identifying the factors that affect the quality of answers shared on SQA services. This paper thus facilitates the reuse of archived answers in health care SQA services.
Clinical coding is done using ICD-10-AM (International Classification of Diseases, version 10, Australian Modification) and ACHI (Australian Classification of Health Interventions) in acute and sub-acute hospitals in Australia for funding, insurance claims processing and research. The task of assigning a code to an episode of care is a manual process. This has posed challenges due to increase set of codes, the complexity of care episodes, and large training and recruitment costs of clinical coders. Use of Natural Language Processing (NLP) and Machine Learning (ML) techniques is considered as a solution to this problem. This paper carries out a comparative analysis on a selected set of NLP and ML techniques to identify the most efficient algorithm for clinical coding based on a set of standard metrics: precision, recall, F-score, accuracy, Hamming loss and Jaccard similarity.
While it is widely accepted that whole of hospital solutions are necessary to reduce the ever-increasing burden on the public health system, little research has focussed on understanding the relationship between ambulance arrival related flow metrics and emergency department (ED) crowding. Queensland Ambulance Service (QAS) shares patient load across multiple hospitals, and receiving facilities strive to meet a Patient Off Stretcher Time (POST) target of 30 minutes. We examine ambulance arrival data from the QAS and ED patient arrival data from 15 major metropolitan hospitals across Queensland, to understand temporal variations in POST performance and examine the relationship between POST performance and ED crowding. The findings suggest a relationship between ED occupancy levels and both ambulances waiting at the ED door and average POST at larger hospitals. No relationship between POST and ED length of stay was found, perhaps due to competing ED National Emergency Access Targets (NEAT). Further modelling is recommended to formally test these observations.
Objective: To explore social network analysis (SNA) as an additional approach to elucidate quantifiable insight from qualitative health-related textual data.
Methods: Key concepts gained from thematic analyses of a set of qualitative health data obtained from an implementation study was analysed using the Excel Add-on module NodeXL.
Results: Our results show that SNA provided useful visualisation and quantifiable information of the relationship between key concepts obtained from the thematic analysis.
Discussion: SNA is a useful technique for exploring and analysing qualitative data, particularly when the research interest is in complex relationships that may exist among a large number of qualitative variables. In addition to providing a way to visualise the relationship between concepts, SNA provides metric measures that can be further analysed quantitatively.
Conclusion: The SNA approach allows researchers to explore deeper relationships that may exist among various variables and enable researchers to derive potentially a fuller and more complete appreciation and comprehension of health-related data.
Understanding the context and nuances of clinical performance at the front-line can provide valuable insights into why the effects of Health Information Technology implementation might differ across sites. It also helps to inform the design of systems that support performance flexibility as clinicians respond to the changing demands of the clinical environment. This study explored the use of an electronic result acknowledgement (eRA) system by physicians at two Emergency Departments (EDs) at Australian metropolitan teaching hospitals to understand how electronic result acknowledgement is managed differently in response to varying environmental demands. Semi-structured, in-depth interviews relating to physician electronic test result acknowledgement processes were conducted with 21 emergency physicians. Physician use of the eRA system differed in four aspects: i) responsibility (i.e. who acknowledges test results); ii) types of tests acknowledged; iii) coordination and synchronisation (i.e. when acknowledgement occurred); and iv) documentation of follow-up actions.
The implementation of Point-of-Care Testing (PoCT) services across rural and remote emergency departments (EDs) by NSW Health Pathology has the potential to significantly improve timely access to results for certain types of pathology laboratory tests and help to deliver timely patient care. The aim of this study was to examine the impact of the implementation of PoCT on the length of stay (LOS) of patients in rural and remote EDs. A total of 3808 patients with a circulatory system illness were treated and discharged at any one of 22 rural and remote EDs during the study period. Generalised Estimating Equation (GEE) modelling was applied to examine whether the implementation of PoCT impacted the ED LOS with adjustment for a range of clinical variables. More patients were treated and discharged from these rural and remote EDs within 4-hours after the PoCT implementation (post-PoCT 86.8% versus pre-PoCT 84.3%). Although average ED LOS was 11 minutes shorter in the post-PoCT period, the impact of PoCT on ED LOS was not conclusive after considering other important clinical factors (p=0.07). This study is the one of the few to examine changes in LOS following the introduction of PoCT in EDs in Australia. The study also identified areas where more robust methods could be applied in the future as the quality of PoCT data improves to further assess the potential effects of this technology on practice and outcomes.
Research on the impact of hospital technology on medication safety usually focuses on prescribing and administration. Less is known about the pharmacy-related processes of reviewing, ordering and dispensing medications and how technology supports this work. We carried out a qualitative exploratory study of a hospital in England (UK) with the aim of gaining insight into processes of digitalisation. We found that hospital pharmacy staff perform safety work with technology aiming to prevent harm, such as ‘scaffolding’ people's thinking processes, or linking-up unintegrated systems. Their work seems to ‘interweave’ safety in between others' medication activities, but they do so sometimes struggling with technological deficiencies.
During the last five years, research about mobile learning conducted with nurses, nurse supervisors and undergraduate students has provided insight into the complexity of this emerging issue, which has the potential to positively impact the workflow of nursing care and improve patient outcomes. Survey and focus group studies including confirmation of beliefs of nurses and nurse supervisors and interviews with representatives from nursing profession organisations were undertaken. Nursing student perspectives about mobile learning were also explored through an online survey. This paper draws on participant narratives from this research revealing ‘tales from the profession’, to demonstrate the complexity of installing mobile technology for learning at point of care for the benefit of healthcare professionals and their patients. This research demonstrates the urgency for introducing governance to provide guidance regarding safe and appropriate use of mobile learning at point of care. Teaching digital professionalism early in undergraduate nursing curricula and promotion of modelling digitally professional behaviour by nurses within healthcare environments is also imperative.
One in four pregnancies ends in miscarriage, a distressing event which can cause significant psychosocial impacts for many women, and yet often remains unseen and unspoken. Many would-be mothers turn to the internet for information and emotional support, and to share their experiences. In this paper, we present the results from 12 semi-structured interviews with women, investigating how and what online information they searched for at the time of miscarriage. We found that women are passive information seekers, searching for causes and preventive strategies to inform future pregnancies. Women want information presented in an easy to understand manner that is not overly clinical, and informed by credible sources. Women also seek psychological support and emotional relief through reading about others' experiences and sharing their stories online. The findings from this study provide a unique insight into the support and information needs of women, and will be used to guide the content, design and functionality of web-based technologies for women experiencing miscarriage.
This paper outlines the Australian experience in adopting the Classification Markup Language (ClaML) to represent two clinical classification systems: ICD-10-AM (International Statistical Classification of Diseases and Related Health Problems, 10th Revision, Australian Modification) and ACHI (Australian Classification of Health Interventions). The primary goal of this process is to share classification data efficiently with relevant parties in a consistent format. This paper outlines extensions implemented in ClaML to facilitate the representation of the above classifications, with validation of the resulting ClaML files verified through XML stylesheet transformation (XSLT) to render the data on a browser, in the same format of printed ICD-10-AM and ACHI books.
Benefits of telehealth have been demonstrated both internationally and through local assessments. Early diagnosis and treatment, reduced costs associated with patient travel, minimized time spent away from community and providing improved patient and staff satisfaction are key drivers for using telehealth. Uptake of telehealth in the NT has been limited, for a variety of reasons including inadequate broadband access. Through collaboration between stakeholders, high-end satellites have been deployed in three very remote clinics, uncontended internet provided and telehealth successfully implemented. Face-to-face consultation via video-conferencing, direct supervision and observation of patient examinations, showing patients and families pictures and videos from the internet allowing the supervising GP to demonstrate clearly what the problem is and the treatment required and the use of remote diagnostic systems for patient assessment in acute care is a “game changer” in remote Indigenous health service delivery. Early identification and decision making of malignancies can facilitate earlier intervention with better prognosis for the patient. Through collaboration, this program has demonstrated the value of uncontended and unlimited internet access in implementing telehealth. The question was: Is high quality internet required to improve service delivery? The service recognises the value and now relies heavily on this service and is committed to improving connectivity and implementing telehealth in more of their communities.
Conversational interfaces and speech recognition capabilities are being increasingly used to create more natural and intuitive user interaction with digital technology. While voice-activated technologies have been used to support clinical documentation, their use for reporting patient safety incidents has not been previously investigated. The purpose of this paper is to assess the technical feasibility of an application for reporting incidents that combines a conversational interface with speech recognition software, and to undertake a pilot study of its usability. We built a prototype finite state-based application where incidents involving digital health technologies could be reported by answering five questions about the task being performed, the response of the software and the outcome of the clinical situation. Pilot tests showed that the conversational interface was usable. However, participants expressed concerns with speaking out loud about sensitive patient safety and quality improvement issues, including human error and system failures, in busy clinical environments. Further work is required to identify the clinical contexts in which conversational interfaces can be used to support incident reporting.
In the field of clinical classifications, such as ICD-10 (International Classification of Disease, version 10), there is a transition from paper-based books towards digital systems to have an open process, powered by collaboration. Most clinical classification systems contain massive volume of terms that are hierarchically organised. There are variety of approaches to visualise hierarchical data, in digital systems. In this paper, a selected set of technologies, such as tree views, tree view diagrams, force-directed graphs are investigated to evaluate the suitability of those to accommodate a large clinical classification data sets. Our findings suggest that tree view diagram is the most appropriate technique, due to its ability to accommodate the multi-parent structure of some clinical classifications, and represent large amounts of data in a visually appealing manner.