Machine Learning And Predictive Analytics

Chapter five from the text Health Policy the organization, financing, and delivery of Health Care in America third edition by author Donald Barr M.D., Ph.D, confirms healthcare’s complexity. The text describes many healthcare, competitive options. Each time a health dilemma has arisen instead of repairing the current health care model, the government develops a new plan. Healthcare plans are similar to software upgrades. Each time a technological glitch occurs the software manufacturer creates an update of the previous version. Most often the software update is another expense for the consumer; not a viable option. Furthermore, the upgrade does not guarantee the new version will work any better. Typically, there is yet another glitch to fix in the latest version. In essence, the health care system should problem solve the healthcare plans that have failed.

This week’s article by Liz Welch titled, Physician, Healthy Industry discusses why our health care industry should not be run as a business. The doctor interviewed by Welch is a Boston-based physician, Dr. Rushika Fernandopulle who runs a company called Iora. He addresses the primary care experience including the coding, billing, and copays in our current health care industry. His argument is that the system should not make a physician do what they do not want to do. He states, “I’d seen 40 patients, and had to stay two hours after work to finish up notes, because that’s what these stupid systems make you do” (Welch, 2015, para. 2). Dr. Fernandopulle frustrations are with the paper pushing process that our health care system requires of all physicians, but he decides to focus on the patient and their concerns, rather than the paperwork aspect. The purpose of Dr. Fernandopulle interview was to reduce the rising health care costs and start looking at the patient closely to prevent unwanted and unnecessary emergency room visits.

It’s a value care for U.S patient population. Back to the digital health care prediction, Dr. Brown indicated that the vast amount of untapped data could have a great impact on health that exists outside of medical systems. However, there were challenges when it came to collecting healthcare data. First, the unstructured big data which presents in medical literature. How do we know which one to read? Second, a data associated with a single patient in an electronic medical record (EMR). An electronic medical record came with a structured data and an unstructured data. There was a question about the HIPPA regulation, and Dr. Brown assured that IBM adhered to the regulation completely. All these pooled data were placed in the Watson Cloud that aggregated the data together to perform different analytics. It’s all automated system. Then, IBM acquired Truvan that tracked insurance and reimbursement data and enabled to see overall

Anthem with its state of the art Health Care Analytics engine and access to wide range of comprehensive information including complete claims history is able to provide doctors meaningful actionable information like high

Change is inevitable when it is a faltering system; such as health care and insurance, which means in the health care industry there is nothing constant in the contemporary managed care systems. When speaking of managed health care, some find it to be an institution set forth as a response to the consistently rising health care costs, that some viewed as detrimental to the United States economy as a whole. Others viewed managed care as a documented response to health care quality problems that have long been ignored. That the problems managed health care would specifically help rectify patterns of over and underuse, misuse, and any geographic variation in the medical services provided to citizens (Kongstvedt, 2001). Both of these ideals had a hand in the wide spread adoption of managed health care in the United States, they do not define the foundational and personal changes that needed to occur in order to accommodate such a large change. However, following the introduction of the health maintenance organizations or HMO’s there has been an over flow of managed care programs and products being created in the United States.

Patients should prepare to have a higher health insurance cost or a higher “out of pocket” costs so they don’t freak out when they get their bill. Patients need to know and understand their health insurance policy, for example brand and generic drug costs, co-pay for doctor visits and etc. They should be assertive in dealing with their own insurers, providers, and case managers. I agree that many of the changes that have led to a managed care system are rooted deeply with the economic realities. Furthermore, I agree that while hospital closings and mergers create many issues and concerns, both the declining number of beds and the declining number of admissions is related to a decline in the number of in-patient

In 2004, it was reported that healthcare had one hundred fifty billion dollars in losses due to health insurance fraud and abuse. Health insurance fraud and abuse can waste up to one-tenth of healthcare resources. Fraud and abuse also affects the consumer as it causes for the rise in premiums, less benefits, increase taxes and higher copayments. The occurring fraud and abuse also puts patients through unnecessary test and procedures for health insurance claims. Having an aggressive stance on preventing fraud by having a outsourced program established can help prevention in loss and harm to patients. Educating the organization on how much fraud can cost, and that patients are putting trust into the care they receive, expecting to not be deceived.

Referring to a study conducted by IBM, over $30 billion are spent in US each year on unnecessary hospital admissions. Clinical and behaviour data can be used to identify patients at high-risk and lead healthcare providers to develop strategies to care for patients before it is too late. The prediction of patient’s problems can reducing the number of unnecessary hospitalizations and the related costs.

Myriad approaches to address their analytics needs, including combining claims with clinical data healthcare groupings are now taking. To cut down the lay on the line of the patient populations, this claims based bequest will be both a beginning point as well as a distraction from analyzing much richer clinical data from clinically housed sources as providers increasingly seek. (Elsevier 2012).

My observations in my role at CHS leaves me saddened at the current state of the health care industry. While working on the clinical side of health care, I basically only saw things from the patient care perspective. I did not understand why insurance companies would refuse coverage to patients or denied payment to the providers. I have witnessed the sick versus the healthy leading to disparity in health insurance coverage. Insurance companies tend to ostracize or avoid the sick people that need the coverage while covering the healthy people instead (Morone, Litman, & Robins, 2009). I thought that the payers were only concerned with not paying for health care. However, once I came to the financial side, I learned of how much hospitals

Evolent’s core functions center around their technology solutions and clinical acumen. Evolent developed a proprietary clinical engine to provide a 360 solution to providers and healthcare professionals. The intent of the system is to capture all the data that is attributed to a patient and centralize it within a single platform. This gives a provider the ability to understand what services their patients are receiving, any recent episodes in patient care, and adherence to medications, to name a few. Additionally, from an insurer’s perspective, it highlights the patients that are driving costs, are likely to have a hospital admission or readmission, what interventions would have the best clinical outcomes and examines how effective their

Clinical decision-support systems (CDSS) apply best-known medical knowledge to patient data for the purpose of generating case-specific decision-support advice. CDSS forms the cornerstone of health informatics research and practice. It is an embedded concept in almost all major clinical information systems and plays an instrumental role in helping health care achieve its ultimate goal: providing high quality patient care while, at the same time, assuring patient safety and reducing costs. This computer based systems designed to impact clinician decision making about individual patients at the point in time that these decisions are made. If used properly, CDSS have the potential to change the way medicine has been taught and

Predictive modeling is another form of data analytics that is focused on forecasting the future medical costs. This model utilizes patients’ medical information in an effort to evaluate health risks and forecast the future of medical utilization (Ingenix, 2006). A large variety of predictive modeling procedures are available, which are created to assign a specific risk level or score to the patients (Asparouhov, 2012). Risk scores are controlled by the risk markers and are given to each patient in a specific population (Ingenix, 2006). In using any past diagnoses and other information gathered, the predictive models will calculate the individual patient costs that can then be used by healthcare providers and the insurers. Therefore, certain

Hosseinzadeh et al. In [17] this work analyzed predictability of hospital readmission without targeting any specific disease. In this analysis considered a dataset ,covers demographic, clinical procedure-related and diagnostic-related features along with drug information of patients predict the readmission within 30 days (short-term readmission prediction) above 65 years of age. Therefore, it’s contained comprehensive results of feature reduction methods and the performance of the prediction models is modest. Consequently, this studied did not consider cost analysis and grouped features importance mining and these studied not targets towards specific disease prefer like diabetes.

Big data will continue to grow even bigger with the advancement of technology and the implementation of more interconnected health systems. The focus will continue to shift from episodic care to preventative care for larger populations especially the chronically ill. Healthcare delivery will continue to shift away from episodic to continuum model. The above changes inspire me to be a part of positive changes within the healthcare field that H.I.M. is providing. The federal government will increase regulations to do so and may or may not provide ongoing monetary incentives for providers to keep up.