Thursday, October 3, 2019

Learning Health Systems in Australia Analysis

Learning Health Systems in Australia Analysis Submitted by : Jaison Prabhath Jaiprakash INTRODUCTION A Learning Health System (LHS) aims to deliver the best possible care to patients, each time, and to learn and improve itself with each care experience. Its vision guarantees to change healthcare services, by empowering the health professionals to change the entire health care system into a highly reliable industry. A learning health system combines quality patient care with the routine collection of data. This is aimed at improving patient outcome. A fully functional system like this would advance the overall quality of healthcare and improve patient and provider safety. The data collected through electronic health records are vast and expanding, which helps in creating new knowledge about the effectiveness of the given treatment and helps in predicting outcomes. An LSH emphasises on an approach that shares data and insights across boundaries to drive better, more efficient medical practice and patient care. The key to achieve their objectives are linked to the collection of data th at is commonly called Big Data from various types of clinical practices. The big data movement in computer science has brought dramatic changes in what counts as data, how that data is analysed, and what can be done with that data. Big data has only recently begun to influence clinical practice. (Iwashyna and Liu, 2014). Enormous amounts of health care data are collected from patients and populations and the interpretation of that data is very important in meeting the needs of the patients. Combining big data and next-generation analytics into population health research and clinical practice requires new data sources, new thinking, training, and tools. If properly used, these pools of data can be an infinite source of knowledge to power a learning health care system. Clinical trials help to manage and improve the health care system. It is all about conducting studies and investigations into various diseases and conditions and eventually hope to eradicate the illnesses. It helps to harness the information for improved clinical trial design, patient recruitment, site selection, monitoring insight and decision making. Data produced through clinical trials like randomized control trials (RCT) often include many treatments and patients from different groups, to improve the reliability of participants and to access the data, these records are digitized, this is where big data helps to store large amount of data sets. By mining the area of clinical practice, we can learn a lot about the patient care. METHODS Search Strategy The SCOPUS and PubMed databases were searched for articles related to the role of learning health systems and clinical practice. Most articles were taken from the year 2014. The search was limited to articles published in journals. Search terms A Boolean search was performed using the following terms: learning health system AND clinical practice, learning healthcare system AND clinical practice, learning health system AND clinic and learning healthcare system AND clinic. Selection / inclusion Criteria The literature review was conducted and articles chosen were from the existing learning health systems such as PEDSnet which are already being used for various clinical practices.   The search was later filtered into aspects that are essential to clinical practice as well as learning health systems, namely, big data. RESULT The role of the health care system is important to deliver the quality care and treatment to the patients. Learning health systems have shown remarkable developments in clinical practices, for example formation of Clinical Data Research Networks (CDRN) consist of many health care systems which conducts research as a network on topics like health care delivery, population health, assessing health disparities and so on. A few of these healthcare systems are listed below. PEDSnet: A National Pediatric Learning Health System   Ã‚   PEDSnet is a clinical data research network (CDRN) that provides the infrastructure to support a national paediatric learning health system. The PEDSnet clinical data research network is an association of eight childrens hospitals, two existing patient-centred disease-specific paediatric networks addressing inflammatory bowel disease and complex congenital heart disease, a newly formed paediatric obesity network, and two national data partners. Together they form the essential components of the National Paediatric Learning Health System (NPLHS). The NPLHS will establish the data sharing environment to enable a community of patients and clinicians, interacting at the point of care, to generate data that can be reused for research and quality improvement and to support continuous monitoring of outcomes that identify specific management practices as targets for comparative effectiveness research (CER). (Forrest et al., 2014) All the information about the patients are recorded using Patient Reported Data (PRD) for quality improvement, clinical practice, or research applications. Table 1: PEDSnet overview (Forrest et al., 2014) Point of Care Research (POC-R) Point of Care Research (POC-R) is a clinical study design that is used to compare two or more treatments that are considered equal. It takes advantage of Electronic health records to enable participant recruitment and data collection of the patients. The goal of POC-R is to embed research into clinical practice, contributing to a Learning Healthcare System (Weir et al., 2014). pSCANNER (part of the PCORnet) The patient-centred Scalable National Network for Effectiveness Research (pSCANNER), is a part of the recently formed PCORnet (Patient Centred Outcomes Research net), which is a national network composed of learning healthcare systems and patient-powered research networks funded by the Patient Centred Outcomes Research Institute (PCORI). Its mission is to provide health related data available to clinicians, researchers and other stakeholders to improve the health-related policies, decision-making and governance. It uses a distributed architecture to integrate data from three existing networks VA Informatics and Computing Infrastructure (VINCI), University of California Research exchange (UC-ReX) and SCANNER, a consortium of UCSD covering over 21 million patients in all 50 states of the USA providing ambulatory care and community-based outpatient clinics with claims and health information exchange data. (Ohno-Machado et al., 2014). pSCANNER shares the data but also protects the privacy of patients at the same time. Only summary statistics are shared between the researcher and clinician. Initial use cases will focus on three conditions: congestive heart failure, Kawasaki disease and obesity. Stakeholders, such as patients, clinicians, and health service researchers, will be engaged to prioritize research questions to be answered through the network. The distributed system will be based on a common data model that allows the construction and evaluation of distributed multivariate models for a variety of statistical analyses. (Ohno-Machado et al., 2014) Learn From Every Patient (LFEP) The merging of three major trends in medicine, namely conversion to electronic health records (EHRs), prioritization of translational research, and the need to control healthcare expenditures, has created unique interests and chances to develop systems that advance healthcare while reducing the overall cost. But making a learning health system operational requires regular changes that have not yet been widely demonstrated in clinical practice. The authors developed, implemented, and evaluated a model of EHR-supported care in a cohort of 131 children with cerebral palsy that integrated clinical care, quality improvement, and research, entitled Learn from Every Patient (LFEP). Children treated in the LFEP Program for a 12-month period experienced a 43% reduction in total inpatient days, a 27% reduction in inpatient admissions, a 30% reduction in emergency department visits, and a 29% reduction in urgent care visits. LFEP Program implementation also resulted in reductions in healthcare costs of 210% (US$7014/child) versus a Time control group, and reductions of 176% ($6596/child) versus a Program Activities control group. Importantly, clinical implementation of the LFEP Program has also driven the continuous accumulation of robust research-quality data for both publication and implementation of evidence-based improvements in clinical care. These results demonstrate that a learning health system can be developed and implemented in a cost-effective manner, and can integrate clinical care and research to systematically drive simultaneous clinical quality improvement and reduced healthcare costs. (Lowes et al., 2017) Figure 1: The Learn From Every Patient (LFEP) model PaTH PaTH provides an informatics supported infrastructure for cohort identification and data sharing within the network of three targeted conditions: idiopathic pulmonary fibrosis (IPF), atrial fibrillation (AF), and obesity. It helps in linking the electronic patients records and understand the survey methods used in research. It uses an open source tools (i2b2 and SHRINE) to aggregate, analyze the distributed data, and facilitate patient centered, comparative effective research. It also helps in improving the decision making capability of both patients and physicians through collaborative process that brings each partner closer to the ideals of a learning health system. (Waqas Amin, 2014). DISCUSSION Big Data is an important but diverse intellectual movement seeking to bring new technologies of data acquisition, data integration, and data analysis into clinical research, hospital operations, and clinical practice. These trends will only accelerate for the foreseeable future, as they build on decades of others doing exactly those same things. Big Data will not solve fundamental challenges of either logical inference or of human behaviour. (Weir et al., 2014). Big Data will continue to provide new knowledge and decision-making support for an array of real and pressing clinical problems (Iwashyna and Liu, 2014). PEDSnet will transform paediatric healthcare and childrens health by developing an extensive and efficient digital infrastructure that enables all participants to work together in the work of producing new knowledge and improving health and care delivery. PEDSnet benefits from robust pre-existing resources and a unique history of collaboration by childrens hospitals that has fundamentally reshaped outcomes for previously fatal diseases, such as cystic fibrosis and many childhood cancers. As the basic digital structure to a learning health system, PEDSnet enables the quick application of new evidence into clinical practice and will address fundamental questions of clinical effectiveness for children and their families, particularly for individuals affected by serious, and generally rare, illness that persists into adulthood. (Forrest et al., 2014) The Point of Care Research (POC-R) highlights several possible factors important to a nationwide implementation of a pragmatic trial program. Participants were significantly concerned with added burden, changes in the provider-patient relationship, ethical implications, validity of results, and integration with workflow. To encourage and support provider buy-in, programs might consider provider training, marketing, and electronic support for decision-making. Providing evidence of equipoise and the validity of data capture might be essential for buy-in. Work process analysis should be part of the proposal. (Weir et al., 2014) pSCANNER will encode a significant portion of policies in software, use a flexible strategy to harmonize data, and use privacy-preserving technology that enables highly diverse institutions to join the network and allow stakeholders to participate. Significant challenges in terms of providing sufficient incentives for patients, clinicians, and health systems to participate and ensuring the sustainability of the network, which were not the focus of this article, will also need to be addressed. The pSCANNER project offers a unique opportunity to make progress toward these objectives, and share results with a community of researchers and representatives from a broader group of stakeholders. (Ohno-Machado et al., 2014) The introduction of EHR-supported care that integrated clinical care, quality improvement, and research resulted in large reductions in healthcare utilization, with associated reductions in charges. Direct comparisons with two distinct comparison groups, to account for the effects of time and LFEP Program activities, confirmed that patients in the LFEP Program had greater reductions both in healthcare utilization and healthcare charges than either control group. Together, these early results confirm that it is both feasible and cost-effective to operationalize key components of an LHS in a large academic medical center. Furthermore, such a system is able to simultaneously improve clinical care and efficiency, and reduce healthcare expenditures, while creating a robust research-quality data set enabling healthcare systems to systematically Learn from Every Patient. (Lowes et al., 2017) The PaTH network will adhere to best practices by using as its backbone open source tools (i2b2 and SHRINE) to aggregate data using standard vocabularies and provide distributed, de-identified cohort queries. PaTH will test these systems in three targeted disease conditions. PaTH will provide a robust informatics supported platform to facilitate comparative effectiveness research, support the conduct of clinical trials, and improve the decision-making capability of both patients and physicians through a collaborative process that brings each partner closer to the ideals of a learning health system. (Waqas Amin, 2014) CONCLUSION The ongoing feedback of insights from data to patients, clinicians, managers and policymakers can be a powerful motivator for change as well as provide an evidence base for action. Many studies and systems have demonstrated that routine data can be a powerful tool when used appropriately to improve the quality of care. A learning healthcare system may address the challenges faced by our health systems, but for routinely collected data to be used optimally within such a system, simultaneous development is needed in several areas, including analytical methods, data linkage, information infrastructures and ways to understand how the data were generated. (Deeny and Steventon, 2015) These results demonstrate that a learning health system can be developed and implemented in a cost-effective manner, and can integrate clinical care and research to steadily drive simultaneous clinical quality improvement and reduce the overall cost of healthcare. (Lowes et al., 2017) REFERENCES BRODY, H. MILLER, F. G. 2013. The Research-Clinical Practice Distinction, Learning Health Systems, and Relationships. Hastings Center Report, 43, 41-47. DEENY, S. R. STEVENTON, A. 2015. Making sense of the shadows: Priorities for creating a learning healthcare system based on routinely collected data. BMJ Quality and Safety, 24, 505-515. FORREST, C. B., MARGOLIS, P. A., CHARLES BAILEY, L., MARSOLO, K., DEL BECCARO, M. A., FINKELSTEIN, J. A., MILOV, D. E., VIELAND, V. J., WOLF, B. A., YU, F. B. KAHN, M. G. 2014. PEDSnet: A national pediatric learning health system. Journal of the American Medical Informatics Association, 21, 602-606. GRANT, R. W., URATSU, C. S., ESTACIO, K. R., ALTSCHULER, A., KIM, E., FIREMAN, B., ADAMS, A. S., SCHMITTDIEL, J. A. HEISLER, M. 2016. Pre-Visit Prioritization for complex patients with diabetes: Randomized trial design and implementation within an integrated health care system. Contemporary Clinical Trials, 47, 196-201. IWASHYNA, T. J. LIU, V. 2014. Whats so different about big data?: A primer for clinicians trained to think epidemiologically. Annals of the American Thoracic Society, 11, 1130-1135. LOWES, L. P., NORITZ, G. H., NEWMEYER, A., EMBI, P. J., YIN, H., SMOYER, W. E., LEARN FROM EVERY PATIENT STUDY, G., TIDBALL, A., LOVE, L., SCHMIDT, J., GOLIAS, J. MILLER, M. 2017. Learn From Every Patient: implementation and early results of a learning health system. Developmental Medicine and Child Neurology, 59, 183-191. OHNO-MACHADO, L., AGHA, Z., BELL, D. S., DAHM, L., DAY, M. E., DOCTOR, J. N., GABRIEL, D., KAHLON, M. K., KIM, K. K., HOGARTH, M., MATHENY, M. E., MEEKER, D. NEBEKER, J. R. 2014. pSCANNER: Patient-centered scalable national network for effectiveness research. Journal of the American Medical Informatics Association, 21, 621-626. STEINER, J. F., SHAINLINE, M. R., BISHOP, M. C. XU, S. 2016. Reducing missed primary care appointments in a learning health system. Medical Care, 54, 689-696. WAQAS AMIN, F. R. T., CHARLES BORROMEO, CYNTHIA H CHUANG, 2014. PaTH: towards a learning health system in the Mid-Atlantic region. Journal of the American Medical Informatics Association, 21, 633-636. WEIR, C. R., BUTLER, J., THRAEN, I., WOODS, P. A., HERMOS, J., FERGUSON, R., GLEASON, T., BARRUS, R. FIORE, L. 2014. Veterans Healthcare Administration providers attitudes and perceptions regarding pragmatic trials embedded at the point of care. Clinical Trials, 11, 292-299.

No comments:

Post a Comment

Note: Only a member of this blog may post a comment.