A complex adaptive systems perspective of health information technology implementation.

Implementing health information technology (HIT) is a challenge because of the complexity and multiple interactions that define HIT implementation. Much of the research on HIT implementation is descriptive in nature and has focused on distinct processes such as order entry or decision support. These studies fail to take into account the underlying complexity of the processes, people and settings that are typical of HIT implementations. Complex adaptive systems (CAS) is a promising field that could elucidate the complexity and non-linear interacting issues that are typical in HIT implementation. Initially we sought new models that would enable us to better understand the complex nature of HIT implementation, to proactively identify problem issues that could be a precursor to unintended consequences and to develop new models and new approaches to successful HIT implementations. Our investigation demonstrates that CAS does not provide prediction, but forces us to rethink our HIT implementation paradigms and question what we think we know. CAS provides new ways to conceptualize HIT implementation and suggests new approaches to increasing HIT implementation successes.

The use of discrete-event simulation modeling to compare handwritten and electronic prescribing systems.

Electronic prescribing (e-prescribing) is expected to bring many benefits to Canadian healthcare, such as a reduction in errors and adverse drug reactions. As there currently is no functioning e-prescribing system in Canada that is completely electronic, we are unable to evaluate the performance of a live system. An alternative approach is to use simulation modeling for evaluation. We developed two discrete-event simulation models, one of the current handwritten prescribing system and one of a proposed e-prescribing system, to compare the performance of these two systems. We were able to compare the number of processes in each model, workflow efficiency, and the distribution of patients or prescriptions. Although we were able to compare these models to each other, using discrete-event simulation software was challenging. We were limited in the number of variables we could measure. We discovered non-linear processes and feedback loops in both models that could not be adequately represented using discrete-event simulation software. Finally, interactions between entities in both models could not be modeled using this type of software. We have come to the conclusion that a more appropriate approach to modeling both the handwritten and electronic prescribing systems would be to use a complex adaptive systems approach using agent-based modeling or systems-based modeling.

Isolation and characterization of highly pathogenic avian influenza virus subtype H5N1 from donkeys.

BACKGROUND: The highly pathogenic H5N1 is a major avian pathogen that crosses species barriers and seriously affects humans as well as some mammals. It mutates in an intensified manner and is considered a potential candidate for the possible next pandemic with all the catastrophic consequences. METHODS: Nasal swabs were collected from donkeys suffered from respiratory distress. The virus was isolated from the pooled nasal swabs in specific pathogen free embryonated chicken eggs (SPF-ECE). Reverse transcriptase polymerase chain reaction (RT-PCR) and sequencing of both haemagglutingin and neuraminidase were performed. H5 seroconversion was screened using haemagglutination inhibition (HI) assay on 105 donkey serum samples. RESULTS: We demonstrated that H5N1 jumped from poultry to another mammalian host; donkeys. Phylogenetic analysis showed that the virus clustered within the lineage of H5N1 from Egypt, closely related to 2009 isolates. It harboured few genetic changes compared to the closely related viruses from avian and humans. The neuraminidase lacks oseltamivir resistant mutations. Interestingly, HI screening for antibodies to H5 haemagglutinins in donkeys revealed high exposure rate. CONCLUSIONS: These findings extend the host range of the H5N1 influenza virus, possess implications for influenza virus epidemiology and highlight the need for the systematic surveillance of H5N1 in animals in the vicinity of backyard poultry units especially in endemic areas.

Evaluation of Heart Failure Apps to Promote Self-Care: Systematic App Search.

BACKGROUND: Heart failure (HF) is a chronic disease that affects over 1% of Canadians and at least 26 million people worldwide. With the continued rise in disease prevalence and an aging population, HF-related costs are expected to create a significant economic burden. Many mobile health (mHealth) apps have been developed to help support patients' self-care in the home setting, but it is unclear if they are suited to the needs or capabilities of older adults. OBJECTIVE: This study aimed to identify HF apps and evaluate whether they met the criteria for optimal HF self-care. METHODS: We conducted a systematic search of all apps available exclusively for HF self-care across Google Play and the App Store. We then evaluated the apps according to a list of 25 major functions pivotal to promoting HF self-care for older adults. RESULTS: A total of 74 apps for HF self-care were identified, but only 21 apps were listed as being both HF and self-care specific. None of the apps had all 25 of the listed features for an adequate HF self-care app, and only 41% (31/74) apps had the key weight management feature present. HF Storylines received the highest functionality score (18/25, 72%). CONCLUSIONS: Our findings suggest that currently available apps are not adequate for use by older adults with HF. This highlights the need for mHealth apps to refine their development process so that user needs and capabilities are identified during the design stage to ensure the usability of the app.

The Cost of Quality in Diabetes.

The adoption and use of Electronic Medical Records (EMRs) and Electronic Health Records (EHRs) is continuing to rise in North America. These systems contain data of varying degrees of quality, including poor quality or "dirty" data. Data entered into EMRs need to be clean or of high quality for them to be useful for a variety of reasons, including quality improvement, clinical decision support, population management, research and system management. There are two potential solutions to obtaining clean data from EMRs: data discipline and data cleansing. Data discipline focuses on ensuring that entry of data into EMRs is of high quality, while data cleansing focuses on cleaning data in the database. Clean data are necessary for healthcare providers to effectively manage chronic diseases and should lead to a reduction in the costs associated with those diseases. The objective of this paper is to compare the costs involved in implementing the two different data cleaning approaches by performing a Budget Impact Analysis (BIA) using diabetes as the exemplar in Canada. The BIA revealed that the cost to implement data discipline is $65 million whereas the cost to implement the data cleansing approach would be $21 million. Even though the cost may seem high, the cost of dirty data is even higher. Data discipline, data cleansing, or a combination of both approaches should be considered going forward.

A Platform to Collect Structured Data from Multiple EMRs.

Adoption and use of Electronic Medical Records (EMRs) is continuing to rise across Canada, leading to more data being generated. These data, however, are not being captured in a standardized manner, they are not available for research, surveillance or health system management, and they are not having a real-time impact on healthcare providers at the point of care. Multiple stakeholders, including researchers and system evaluators, require easy access to high quality, structured data. As current EMRs are not able to effectively meet their needs, we engaged multiple stakeholders to assist in designing a solution. A total of 90 stakeholders from various backgrounds participated in an iterative joint design process. After incorporating the feedback of all stakeholders, we developed the design for a scalable platform for capturing structured, evidence-based data from all EMRs in Canada for research, health system management, clinical decision support and other purposes. We discuss the design specification for our proposed solution and explain how, using clinical forms, we can not only capture structured, high quality data from multiple EMRs, but also provide real-time guideline advice to providers at the point of care. The scalability of this proposed solution across multiple diseases and multiple EMRs is also explained. We further discuss the benefits and limitations of this proposed solution to several key stakeholder groups and address issues of privacy and security.

Integrating Clinical Decision Support into EMR and PHR: a Case Study Using Anticoagulation.

Clinical decision support (CDS) for atrial fibrillation is expected to ease the implementation of often-complex guidelines for atrial fibrillation and anticoagulation. Most clinical decision support systems (CDSS) for anticoagulation are stand-alone systems that do not integrate with electronic medical records (EMR). We have developed an architecture that consists of a computerized CDS that can integrate with multiple EMRs and multiple patient health records (PHRs). The design process revealed some significant issues that were resolved through systematic business/clinical analysis and creative clinical design in the diagnostic and treatment domains. Key issues identified and resolved include: 1) how to correctly allocate existing patients into various CDSS states (e.g., MAINTENANCE, HOLD, DISCONTINUE, etc), 2) identify when a patient becomes eligible for CDSS guidance over time, 3) how the CDSS maintains information about the patient's anticoagulation state and 4) how to transform vague human-readable concepts to explicit computable concepts. The management of anticoagulation for atrial fibrillation is no easy task and we believe our architecture will improve patient care at all levels and ultimately better balance the reduction of stroke risk while minimizing harms from major bleeding. In addition, the architecture presented is scalable to other treatment guidelines and is scalable to multiple EMRs and PHRs, making it suitable for use in a platform approach.