Ebook: Health Informatics: Digital Health Service Delivery – The Future Is Now!

We are living through a time of immense change, where innovation, learning from experiences and sharing ideas are essential to delivering the promises of e-health. As part of this exciting community we celebrate twenty one years of productive scientific output – a substantial achievement in any field of ICT. We have come of age. The Health Informatics Society of Australia, with the active support and involvement of the Australasian College of Health Informatics, is justifiably proud to have sustained the annual Australian National Health Informatics Conference (HIC) series of meetings through to this stage of maturity. As the largest national event of this type in Australia, with a dedicated scientific stream on health informatics, HIC provides a valued platform for academic and research contributions and interchange. It also consolidates the Australasian profile of strong and current research contributions in health informatics, marking our place on the world stage.

The HIC 2013 theme of “Digital Health Service Delivery – the Future is Now!” emphasises the need to embrace and advance the field of health informatics now. Health services are implementing major changes using technology and diverse groups of health informatics professionals are increasingly required to support these changes. At a time when large scale health information flows are being initiated, through both the inauguration of the personally controlled electronic health record (PCEHR) and the National Broadband Network rollout, the skills and expertise of health informaticians need to be demanded within these projects and within all healthcare organisations. The quality of these systems is of significant interest to consumers, clinicians, vendors and system decision makers. A more trusted environment for managing and using health information will help to consolidate and accelerate the use of health informatics solutions as change mechanisms to drive the establishment and adoption of new models of care and new technology-enabled healthcare processes.

Papers in this volume provide useful information on directions, successes and other factors which need to be considered by all when developing or implementing our systems. Papers represent experiences in Australia and New Zealand but also from other parts of the world. As always, it is a strength of HIC that a wide diversity of work is presented, and that a set of papers has been collected here that ranges from deeply theoretical to intensely practical. The careful reader will be rewarded with exposure to much diversity, and many elements of contemporary health informatics research endeavours.

The double blind peer review process established for HIC 2011 in a previous volume has been continued and augmented. All papers were reviewed by 3 experts in the field of health informatics, selected as prominent academic and industry specialists. The assistance of the Australasian College of Health Informatics in supporting this process through the voluntary efforts of a number of their Fellows is gratefully acknowledged, as is the similar contribution made by many senior members of the Health Informatics Society of Australia. This phase of reviewing resulted in the provisional acceptance of 28 papers from a much expanded submission field of 61. The Scientific Program Committee then undertook a validation process for all such papers that were resubmitted in amended form, to ensure that reviewers' recommendations were appropriately addressed or rebutted. This resulted in 28 papers finally being included for publication in this book.

Heather Grain

Louise K. Schaper

The Australian Government launched a personally controlled electronic health record (PCEHR) system in July 2012 committing $466.7m. Currently Australia lacks a clearly articulated implementation and evaluation framework and there remains limited detail on how this system's success will be determined. These problems are especially visible in primary healthcare. The UK and US, have been advocated as models, however they have started to report points of failure arising from their approaches. Evidence suggests that alternatives need to be considered, if mistakes are not to be replicated. Insights from e-health record implementation and evaluation approaches in Denmark and the Netherlands provide Australia with other approaches. The PCEHR requires different and radical thinking around the delivery of health services. Drawing on a range of English language articles identified between 1996 and 2012, the paper generates a conceptual framework for implementation and evaluation of the PCEHR. The generation of a grounded implementation and evaluation framework in primary healthcare will reduce provider scepticism and facilitate complex changes associated with PCEHR uptake.

Perinatal data (PD) is collected for mothers and babies Australia wide as mandated at a federal level. The data is used to monitor patterns in midwifery, obstetric and neonatal practice and health outcomes and is also used for research purposes, funding allocation and the education of midwives and medical officers. Accuracy in PD is most often reported via quantitative validation studies of PD collections both internationally and within Australia [1]. These studies report varying levels of accuracy in PD collection and suggest researchers need to be more aware of the quality of data they use. This paper presents findings from doctoral research that regarding issues of concern identified by midwives relating to their perceptions of the accuracy of computer PD records. Research, such as that presented in this paper, may improve the robustness of the PD collection and allow for more accurate planning of health services.

The research discussed focuses on improving the delivery of cancer services, specifically lymphoma, at a regional Australian hospital. The work examines whether an emerging patient journey modeling technique known as Essomenic, which has been successfully applied in the domains of; midwifery, chronic kidney disease, mental health, ambulatory care, bariatric and osteoarthritis can deliver similar results to the highly time critical domain of cancer diagnosis and treatment commencement. The work also analysed if enhancements were required to the modelling syntax to accommodate domain specific requirements.

Supporting self management of chronic disease in collaboration with primary healthcare has been a national priority in order to mitigate the emerging disease burden on the already strained healthcare system. However, in practice, the uptake of self-management programs and compliance with clinical guidelines remain poor. Time constraints due to work commitments and lack of efficient monitoring tools have been the major barrier to the uptake and compliance. In this paper, we present a newly integrated mobile health system with a clinical chronic disease management network called cdmNet, which has already been validated to facilitate General Practitioners (GPs) to provide collaborative disease management services. The newly integrated solution takes advantage of the latest mobile web and wireless Bluetooth communication techniques to enable patients to record health data entries through ubiquitous mobile phones, and allows the data to be simultaneously shared by multidisciplinary care teams. This integration would enable patients to self-manage their chronic disease conditions in collaboration with GPs and hence, improve the uptake and compliance. Additionally, the proposed integration will provide a useful framework encouraging the translation of innovative mobile health technologies into highly regulated healthcare systems.

Meticulous planning and preparation do not always guarantee that e-health programs unfold as predicted. e-health entails interdependent social interactions which are difficult to predict without past experience or reference to lessons learned. Judicious insight into past case studies and eventualities, therefore, is essential towards building a successful e-health solution. Australia's e-health program is at a crucial stage where appropriate policy considerations and operational changes are in order. In this paper, we present an initial exploration of prominent e-health initiatives of other countries to identify similarities, differences and to seek lessons towards making Australia's e-health initiative a better journey.

The aim of this study was to examine the impact of the Electronic Medical Record (EMR) on the quality of laboratory test orders made by clinicians. The study assessed the type and frequency of pre-analytical laboratory test order errors that were associated with the EMR across three hospitals and one pathology service. This involved a retrospective audit of the laboratory error logs for the period 1 March 2010 to 9 October 2011. Test order problems associated with the EMR occurred at a total rate of 1.34 per 1000 test order episodes across the three hospitals. In the majority of cases these errors were caused by the inappropriate use of the EMR system by clinicians. The errors resulted in increased data entry time for laboratory staff in the Central Specimen Reception area and led to a median increase of 181 minutes in test turnaround times for those test orders. The study highlights the importance of monitoring and comparing the impact of EMR systems in different locations over time in order to identify (and act upon) factors that can adversely impact on the effectiveness of pathology laboratory processes.

Behavioural mapping (BM) is a long established method of structured observational study used to understand where patients are and what they are doing within a hospital setting. BM is prominent in stroke rehabilitation research, where that research indicates patients spend most of their time at bed rest. We evaluate the technical feasibility of using the Microsoft Kinect to automate patient physical activity classification at bed rest.

Introduction: An exploratory study was conducted to understand patient propensity to use a Personal Health Record (PHR) linked to their primary care clinicians' Electronic Health Record (EHR). Individuals with long-term conditions rely on information exchange for effective self-care. PHRs could be helpful. The research questions were: How do we know if certain people will use PHRs for self-care?

Methods: Semi-structured interviews and observation were conducted with ten patients regarding their use of a PHR available to them.

Findings: Reasons for using PHR functions fell into two patterns: interaction and review. Patients used their PHR if they could interact with their GP, e.g. order a repeat prescription, and if they could review information, e.g. laboratory results, medicines list. Perceived usefulness (PU) of the PHR was strong, especially if linked to a self-care action. Interestingly, behavioural intent to use was not adversely affected by difficulties in using some aspects of the tool, if the PU of a function was strong. Conclusion Patient experience with a PHR linked to their EHR kept by their GP shows positive perceptions regarding the PHR usefulness. Future research could develop guidelines for clinicians to help select patients for PHR use in self-care.

Patient safety is the buzz word in healthcare. Incident Information Management System (IIMS) is electronic software that stores clinical mishaps narratives in places where patients are treated. It is estimated that in one state alone over one million electronic text documents are available in IIMS. In this paper we investigate the data density available in the fields entered to notify an incident and the validity of the built in classification used by clinician to categories the incidents. Waikato Environment for Knowledge Analysis (WEKA) software was used to test the classes. Four statistical classifier based on J48, Naïve Bayes (NB), Naïve Bayes Multinominal (NBM) and Support Vector Machine using radial basis function (SVM_RBF) algorithms were used to validate the classes. The data pool was 10,000 clinical incidents drawn from 7 hospitals in one state in Australia. In first part of the study 1000 clinical incidents were selected to determine type and number of fields worth investigating and in the second part another 5448 clinical incidents were randomly selected to validate 13 clinical incident types. Result shows 74.6% of the cells were empty and only 23 fields had content over 70% of the time. The percentage correctly classified classes on four algorithms using categorical dataset ranged from 42 to 49%, using free-text datasets from 65% to 77% and using both datasets from 72% to 79%. Kappa statistic ranged from 0.36 to 0.4. for categorical data, from 0.61 to 0.74. for free-text and from 0.67 to 0.77 for both datasets. Similar increases in performance in the 3 experiments was noted on true positive rate, precision, F-measure and area under curve (AUC) of receiver operating characteristics (ROC) scores. The study demonstrates only 14 of 73 fields in IIMS have data that is usable for machine learning experiments. Irrespective of the type of algorithms used when all datasets are used performance was better. Classifier NBM showed best performance. We think the classifier can be improved further by reclassifying the most confused classes and there is scope to apply text mining tool on patient safety classifications.

Although mobile applications and social media have emerged as important facets of the Internet, their role in healthcare is still not well-understood. We present design artefacts, inspired by persuasive technology concepts, from a study of social media as part of a diabetes mHealth application. We used the design science approach for mobile application design, and real-life user testing and focus group meetings to test the application over a 12-week period with 7 participants. Based on the System Usability Score (SUS), the mobile application scored an average of 84.6 (SD=13.2), which represents a fairly high usability score compared to the literature. Regression analysis on the daily blood glucose levels showed significant decreases for some patients, and although the study is not powered, the HbA1c showed a promising trend, and self-efficacy marginally increased. Incorporating persuasive elements such as blood glucose tracking and visualisation, and social media access directly from the mobile application produced promising results that warrant a larger study of behaviour change for people with diabetes.

The complexity of hospital operations ensures that one-size-fits-all solutions seldom work. As hospitals turn to evidence based strategies to redesign flow, it is critical that they tailor the strategies to suit their individual service. This paper analyses the effect of hospital occupancy on inpatient and emergency department patient flow parameters at the Caboolture hospital in Queensland, Australia, and identifies critical levels, or choke points, that result in performance decline. The effect of weekdays and weekends on patient flow is also investigated. We compare these findings to a previous study that has analysed patient flow across Queensland hospitals grouped by size, and discover several differences in the interaction between rising occupancy and patient flow parameters including rates of patient flow, length of stay, and access block. We also identify significantly higher choke points for Caboolture hospital as compared to other similarly sized Queensland hospitals, which suggest that patient flow here can be redesigned to operate at higher levels of occupancy without degrading flow performance. The findings support arguments for hospitals to analyse patient flow at a service level to deliver optimum service improvement.

Background: The e-health environment is a rapidly changing one. To effectively engage with technology for healthcare delivery, health professionals must be able to adapt to this constantly evolving environment very quickly. Learning and adapting to new e-health technologies is a life-long learning process.

Objective: This study examined the effectiveness of incorporating self-directed and transformative learning approaches to introduce health sciences students to e-health concepts and skills.

Methods: Two surveys were conducted to measure students' e-health knowledge and their perceived self-efficacy in using commonly available software to complete tasks required for an assessment in the unit of study. These surveys were conducted at the beginning and the end of the semester. Paired t-tests with Bonferroni adjustment were used to examine the effect of the teaching approach on students' self-perceived efficacy.

Results: It was found that students showed significant improvement in their knowledge of, and perceived efficacy in using, commonly available software to carry out spreadsheet, database and data manipulation operations after intervention.

Conclusion: This study shows that the combined self-directed and transformative teaching and learning approach is effective in helping students to identify their learning needs and develop skills to seek out resources that enable them to learn new e-health skills and concepts in a self-directed manner.

Electronic ordering systems have the potential to enhance the effective and efficient utilisation of pathology services. This study examined the influence of electronic ordering of pathology tests in four emergency departments on data entry times, a contributing factor to test turnaround time using failure time analysis. Data relating to tests ordered for 50,854 emergency department patients at four hospitals were analysed. The introduction of electronic test ordering significantly reduced turnaround time associated with laboratory data entry at the four hospitals, but there was significant variation between hospitals. Investigation of the impact of electronic ordering on efficiency of pathology services must take into account specific aspects of turnaround time measurement.

New Zealand has a rapidly expanding health information technology (IT) development industry and wide-ranging use of informatics, especially in the primary health sector. The New Zealand government through the National Health IT Board (NHITB) has promised to provide shared care health records of core information for all New Zealanders by 2014. One of the major barriers to improvement in IT use in healthcare is the dearth of trained and interested clinicians, management and technical workforce. Health Informatics New Zealand (HINZ) and the academic community in New Zealand are attempting to remedy this by raising awareness of health informatics at the “grass roots” level of the existing workforce via free “primer” workshops and by developing a sustainable cross-institutional model of educational opportunities. Support from the NHITB has been forthcoming, and the workshops started in early 2013, reaching out to clinical and other staff in post around New Zealand.

A Natural Language processing (NLP) classifier has been developed for the Victorian and NSW Cancer Registries with the purpose of automatically identifying cancer reports from imaging services, transmitting them to the Registries and then extracting pertinent cancer information. Large scale trials conducted on over 40,000 reports show the sensitivity for identifying reportable cancer reports is above 98% with a specificity above 96%. Detection of tumour stream, report purpose, and a variety of extracted content is generally above 90% specificity. The differences between report layout and authoring strategies across imaging services appear to require different classifiers to retain this high level of accuracy. Linkage of the imaging data with existing registry records (hospital and pathology reports) to derive stage and recurrence of cancer has commenced and shown very promising results.

Introduction. Over the 15 years since the Clinical Information Access Portal (CIAP) was launched, usage has continued to increase. This demonstrates the value that NSW Health staff place on access to point of care evidence-based clinical information.

Methods. Web log usage files for all resources available on CIAP have been collected since 1999. Analysis of resource usage was performed to identify patterns of use. Individual account demographics were analysed for the last 3 years to identify key clinical user groups in NSW Health.

Results. CIAP usage has increased significantly over the last 15 years when annual usage was only 2% of that seen in 2012. The most highly accessed resources include MIMS, eTheraputic Guidelines, ProQuest Nursing and Allied Health, Australian Medicines Handbook, and BMJ Best Practice.

Conclusions. CIAP remains an exemplary example of integrating evidence-based practice at the point of care by providing the best available clinical information for all NSW Health.

Population ageing is the demographic process that characterises the first half of the twenty-first century. Australia's population is already ageing and the states and territories are ageing at different rates. Our understanding of the dementias remains limited and diagnosis in primary care settings is poor. Locating where older people with dementia are and how they are coping is an emerging need in health information management. In this paper we discuss how a spatially informed health information management system could support population ageing and the disconnected systems that address ageing. We illustrate this with examples from our work to show how spatial informatics can advance our understanding of and response to the implications of population ageing.

Telehealth can be viewed as an intervention in the established and stable clinician-patient system with outcomes that are not always predictable. The success of a new telehealth service depends far more on changing the clinicians and patient's expectations and patterns of behaviour than on technology. Success in telehealth projects is critically dependent on understanding how the new telehealth offering changes the distinct value proposition for each of the parties – clinicians, patients and administrators. As with any intervention, design is the key to ensuring more predictable outcomes. An approach to the design of telehealth services is described that takes a holistic view of telehealth by acknowledging the multiple and interdependent dimensions of telehealth services (clinical process change, clinical risk and governance, adoption and change management, clinician and patient experience, operational sustainability, support, and least of all, technology). The approach draws on current themes in design including ‘design thinking’ and service design.

Electronic test ordering, via the Electronic Medical Record (EMR), which incorporates computerised provider order entry (CPOE), is widely considered as a useful tool to support appropriate pathology test ordering. Diagnosis-related groups (DRGs) are clinically meaningful categories that allow comparisons in pathology utilisation by patient groups by controlling for many potentially confounding variables. This study used DRG data linked to pathology test data to examine changes in rates of test ordering across four years coinciding with the introduction of an EMR in six hospitals in New South Wales, Australia. This method generated a list of high pathology utilisation DRGs. We investigated patients with a Chest pain DRG to examine whether tests rates changed for specific test groups by hospital emergency department (ED) pre- and post-EMR. There was little change in testing rates between EDs or between time periods pre- and post-EMR. This is a valuable method for monitoring the impact of EMR and clinical decision support on test order rates.

The expectations of adult children about their elderly parents regarding their care provision was surveyed. We found that the needs and expectations regarding their elderly parents included better information on entitlements of their parents, how to access relevant aged care services, the challenges of remotely dealing with dementia and depression of their parents, accessing medical and non-medical services and access to respite care. The aim was to identify needs that ICTs could potential to assist with. While the majority of respondents (67.2%) stated that they were satisfied or very satisfied with the frequency of contact with their elderly parent(s), they also cited logistical/transport difficulties, lack of time and stress as potential barriers in being in regular contact with their parents. The responses also indicated a high level of interest in a service that could act as a case manager to assist the adult child in discharging their responsibilities, manage access to services and to monitor the well-being of the parent. There is a need for further research to explore how this might be accomplished, whether such a service was viable and what funding models could be applied.

A Discharge Summary contains vocabulary that is difficult to understand for health consumers. We used iterative refinements in developing a system, SemLink, which dynamically generate synonyms and hyperlinks to appropriate Internet resources for difficult terms in discharge summary text to make the text more comprehensible to consumers. This paper describes our iterative refinement protocol to enhance the semantic annotation and dynamic hyperlinking algorithms to link topic-specific web pages for difficult terms found occurring in Discharge Summary text.

The discipline of health informatics is concerned with developing the information systems that facilitate collection, manipulation and dissemination of personal health information. It promotes the benefits of using personal health information for secondary purposes, including policy development, service planning and research. At the same time the health informatics community is a strong advocate of privacy and the need to protect individuals from negative consequences arising from unauthorised use of their personal health information. This creates a dilemma for health informatics professionals since there will be occasions when the rights and interests of individuals conflict with the rights and interests of the public, or particular sections of the public. In such instances, the community as a whole, and individual members, may be required to take a stance on whether to prioritise privacy over public needs and interests. Such instances are likely to increase in the future as demands for access to personal health information increase. This paper considers the way the health informatics community approaches the dilemma. It reports on a study which analysed various perspectives on the issue as expressed in HISA conference proceedings. The study identified six discourses, each of which focuses on different uses of personal health information. The study found that the discourses expressed strong support for expanded use of personal health information where the public interest was convincingly argued, although the interpretation of what constituted public interest varied between the discourses. The study also found that while higher level discussions highlight the potential for negative consequences arising from expanded uses of personal health information, this was not often discussed in the conference texts. It is argued that such concerns should be considered, particularly in the light of discussions around the Commonwealth government's Individual Health Identifier and Personally Controlled Electronic Health Records initiatives.

Introduction: A medicines list iPhone app, based on an existing paper product, was developed by NPS MedicineWise for consumers. Since its launch in June 2011 the app has been downloaded over 19,000 times. With plans to expand its development and increase its use, it was critical to discover how the app is used and how useful it is.

Method: A total of 61 study participants used the app for five days and then participated in online focus groups.

Results: While most people have a system for managing their medicines, these systems vary in their effectiveness. The biggest problem with managing medicines is when medicine doses are forgotten or taken late. Usability problems with the app were revealed. While the app was easy to install, it was time consuming and difficult to input medicine information. The majority of people did not enter information in all fields. While the app was deemed useful and will continue to be used by the majority of participants, not all participants were aware of the existing functions or the capabilities of the functions. Use was primarily for self-management. The majority of participants were comfortable sharing the information in the app (via the phone) with a health professional. Roughly half wanted to upload their information to the PCEHR. Recommendations were made and incorporated into design decisions for an updated version of the app (which includes potential linking to the PCEHR), and a multi-platform release (web, iOS, Android, Windows Mobile).

Information that is persistently collected, unrelated to direct clinical care may be a Pandora's Box for health improvement and monitoring. The use and linkage of disparate data sources, particularly those relating to geo-location and social media present enormous opportunities and challenges for healthcare. To date these sources have been difficult to harness and make sense of, yet they are now beginning to be exploited. It is the innovation in analytic methodology that may hold the key if the issues that may confound the results can be addressed. These issues are encompassed in privacy, and with the scientific quality and completeness of the data. The future for alternative application of geo-location and social media data is here and now, and if they can be harnessed carefully, healthcare may be the biggest beneficiary.