Table of Contents
This issue of Vital Signs, released on August 31, 2013, explores the needs of an ever-expanding healthcare IT and interactive domain in a patient-centric model and identifies the criteria for true intelligent systems that are defining new approaches to care, facilitating the gathering and array of health information, including population management analytics that ultimately integrate with healthcare finance protocols to support rational decision-making for cost containment. Additionally, a company spotlight is provided for GenomeDx Biosciences in Vancouver, BC. The company has developed the Decipher Prostate Cancer Classifier test to predict the metastatic risk for patients that have undergone a prostatectomy as treatment or the surgical removal of the prostate gland.
Intelligent Systems for Health Informatics and Population Management
There is a global movement for nations to improve health, sustain quality outcomes, and reduce costs for healthcare (Dentzer, 2013). This triumvirate of objectives, or “Triple Aim,” first published in 2008 by Dr. Don Berwick and two other colleagues, is becoming a goal of countries in Europe, the Middle East, Latin America, the Pacific Rim, as well as the United States. British Prime Minister David Cameron recently appointed Berwick to head a task force for system improvements relating to quality of medicine, as the British National Health Service struggles to offset accusations of “appalling care.” At the same time, China is working to build an information technology (IT) infrastructure to align more than 20,000 hospitals to effectively share healthcare data as large segments of the population move between urban and rural settings and east to west. Canada, France, and Germany all are engaged in detailed technology assessments to re-shape payment policies related to diagnostic-related groups and cost structures based on real-value assessments of supplies and purchased items. In the Middle East, Saudi Arabia is building palatial hospitals in a massive effort to improve the patient experience so medical dollars remain in their country. Latin America is moving rapidly to employ electronic health records as part of an attempt to link care throughout the region and improve quality outcomes.
But pursuit of the triple-aim goal is still very much exploratory. Whatever the eventual outcome as each nation learns, adapts, and debates the efficacy of transforming their national healthcare system, one underlying principle, or requirement, is the unequivocal need for analysis of big data. This movement to robust big-data analytics is a direct result of the availability of more and more clinical information now housed in electronic health records and with payers, who are evolving systems to bill based on clinical results, not encounters (Barr, 2013). Hospital & Health Networks suggests, “With clinical-results data at the core of such new care and reimbursement models based on accountable care organizations and value-based purchasing, more healthcare systems executives are looking to make sure they are using data to the fullest to improve their care and work efficiently” (Barr, 2013).
In the United States, the second stage of meaningful use has emerged as a significant and new technology driver for EHRs to bridge patient engagement, including HIPAA-compliant communications between doctors and their patients (Morrissey, 2013). Moreover, this effort must provide a “downstream” view of all health information by patients who, in turn, have capability to upload, via telemedicine and mHealth assets, pertinent health data from remote locations in digital form, so a continuum of providers can review, interpret, diagnose, comment, prescribe, and intervene if clinically indicated.
The purpose of this white paper is: (1) to explore the needs of an ever-expanding healthcare IT and interactive domain in a patient-centric model, and (2) identity the criteria for true intelligent systems that are defining new approaches to care, facilitating the gathering and array of health information, including population management analytics that ultimately integrate with healthcare finance protocols to support rational decision-making for cost containment.
The Need for Intelligent Health Informatics
As technology continues to advance in virtually all corridors of patient care, diagnostics, and treatment, both providers and patients alike are calling for a means to aggregate all individual health information. But consolidation and documentation of data is not enough, because multiple-source storage and formatting does not allow for true interoperability. Proliferation of EHRs, each built on proprietary platforms, did not result in uniformity, but rather in multiple systems lacking the ability to “talk to each other.” Hence, the ever-increasing worldwide demand for true intelligent systems, where data flows throughout and across a connected and fully collaborative infrastructure. With this adaptation of intelligent systems, insight can be gained to move health data forward into actionable algorithms that will impact individual care plans. Multiple providers in diverse locations also need a technology partner for secure collaboration of all healthcare information along the care continuum that provides
real-time flow of accurate data.
The ability to leverage information to raise the level of quality outcomes is largely a consequence of using the right technology for integration, connectivity, and the incorporation of IT for the purpose of creating intelligent system design (Kilo, 2005). Moreover, the real-time demand for digital health information, coupled with mounting emphasis for enterprise-level solutions, has revolutionized health informatics. In just the past few years, which is a generation for health IT technology development, knowledge-based solutions have changed the way we view the collection and storage of health information. This rapid advance has simultaneously provided impetus for the development of true healthcare intelligent data processing. As a direct result, “big data” and the need for analytics is now a common component for any evolving intelligent system solution. The backbone on which these analytics tools are built is the recording, exchange and integration of data on systems which, in many cases, are built on non-interoperable IT software platforms.
This movement toward intelligent systems has emerged in parallel with the convergence of healthcare entities and the organizational transformation perhaps most visible in the shift to accountable care organizations (ACOs) in the United States. Supporting this shift is the massive migration of health information. This real-time convergence of clinical and payer data also holds promise for truly solving electronic health record (EHR) shortcomings with interoperability by forcing universal connectivity and tethered interface through integrated servers. Once all data is properly arrayed and has the ability to truly flow, analytic applications will then have the capability to query and segment populations by patient demographics, family histories, social and economic strata, payer eligibility, and health status or virtually any patient variable residing securely on a single platform.
With this infrastructure comes unique opportunity to move health information beyond retrievable data, accessed upon request, to knowledge artificial intelligence (AI)-driven systems that take action without human submission for requests to gather information or data mining to interpret and deduce archived data. This intelligent system revelation holds the potential for healthcare transformation for any government, health system, or hospital that chooses to accept the modern-day challenge for creating new health information workflows, thereby introducing true collaborative care focused on improving patient care, and ultimately delivering this quality of care at a lowercost differential.
Establishing an Objective for System Intelligence
Yet, even with financial incentives, healthcare, as an industry, does not have an IT solution that has as its objective improving patient outcomes leveraging what chief information officers (CIOs) have learned about the feasibility for intelligent systems. EHR introduction showed great theoretical promise to this end. But, according to a Health Affairs article, “[Government] health IT policy has focused more on core issues such as promoting technical interoperability, privacy and security. There has been implicit assumption that interoperable standards would automatically result in the seamless flow of information across settings and that this information flow would quickly result in improved care decisions and outcomes” (Carolyn M. Clancy, 2009). Yet, these assumptions have been ill founded. Little progress has been made toward improving outcomes as a result of evolving system intelligence capability. Quality is one implied objective of course, but to date, success has not been recognized in such magnitude to transform traditional delivery models for care in the United States or abroad.
Defining an end product for system intelligence as achieving holistic health becomes critical with healthcare intelligent system design. It is a fundamental requirement in the establishment of needed concepts and implanted response rules. If response rules are dominated by billing criteria, then the objective for improving outcomes becomes compromised. Therefore, by declaring improving patient outcomes as its objective, the true intelligent system architecture for healthcare applications should be engineered to sense and capture relevant healthcare data such as individual patient primary diagnosis, family history, co-morbidity, immunizations, laboratory test results, etc. The system will then store these as digital sensory variables or elementary concepts, not merely data fields.
With all the patient information expressed as concepts, the system then builds a relevant situational protocol or individual health situation. It then scans its memory fields, much the same way search engines like Google search for key words and phrases online looking for applicable response rules based on concept algorithms associated with medical intelligence. It locates applicable response rules, such as identifying the concept for patient history, and finds information concerning a family history of colon cancer variable and takes action.
Get Industry Insights. Simply.
Talk to Ahmad
+1 718 618 4302
The population health management market is projected to reach USD 42.54 billion by 2021 from USD 13.85 billion in 2016, at a CAGR of 25.2% in the next five years (2016 to 2021). The healthcare industry ...
The global E-Medical Records market is estimated to be USD 13.64 billion in 2016 and is projected to reach USD 22.01 billion by 2021, growing at a CAGR of 10.03% during the forecast period from 2016 to ...
The Global Electronic Health Records is poised to grow at a CAGR of around 6.1% over the next decade to reach approximately $36.3 million by 2025. Some of the prominent trends that the market is witnessing ...