Information Technology and Data in Healthcare

Using and Understanding Data

David Hartzband

• Editorial: Productivity Press
• Año: 2019
• ISBN:
  9780429590269
• Edición: 1
• Categoría: Medicina
• Categoría: Sanità, amministrazione e assistenza pubblica

Healthcare transformation requires us to continually look at new and better ways to manage insights – both within and outside the organization. Increasingly, the ability to glean and operationalize new insights efficiently as a byproduct of an organization's day-to-day operations is becoming vital for hospitals and health systems to survive and prosper. One of the long-standing challenges in healthcare informatics has been the ability to deal with the sheer variety and volume of disparate healthcare data and the increasing need to derive veracity and value out of it.

This book addresses several topics important to the understanding and use of data in healthcare. First, it provides a formal explanation based on epistemology (theory of knowledge) of what data actually is, what we can know about it, and how we can reason with it. The culture of data is also covered and where it fits into healthcare. Then, data quality is addressed, with a historical appreciation, as well as new concepts and insights derived from the author's 35 years of experience in technology.

The author provides a description of what healthcare data analysis is and how it is changing in the era of abundant data. Just as important is the topic of infrastructure and how it provides capability for data use. The book also describes how healthcare information infrastructure needs to change in order to meet current and future needs. The topics of artificial intelligence (AI) and machine learning in healthcare are also addressed. The author concludes with thoughts on the evolution of the role and use of data and information going into the future.

Chapter 1

Introduction – Data Is Essential

DOI: 10.4324/9780429061219-1
 

Making the statement “data is essential” seems premature given that we could argue that we don’t know specifically what data is^* yet. The next chapter will cover what data is in detail, so let’s agree to defer specific definition(s) until then and use data in the sense of factual information. So why is data essential in healthcare. Most of the work I’ve done in healthcare has focused on the use of factual information about an individual’s health status, or the use of factual information about a group or population’s health status in order to improve treatment and clinical outcomes for the individual or population. Healthcare policy currently has defined a set of goals embodied in the Triple (or Quadruple) Aim.^† These goals include enhancing patient experience, improving population health, reducing costs, and improving the work life of healthcare providers.

* While data is clearly the plural of the noun datum, it also has a standard use as an abstract, mass noun (like information) for which it takes a singular verb and modifiers (Merriam-Webster, https://www.merriam-webster.com/dictionary/data). This is the form that the word data is used in this book.

† T. Bodenheimer and C. Sinsky, 2014, From Triple to Quadruple Aim: Care of the Patient Requires Care of the Provider. Annals of Family Medicine, November, 12(6): 573–576. doi:10.1370/afm.1713

It is clear that data plays an important role in each of these goals:

• Enhancing patient experience
  • – There are two major aspects to enhancing the patient’s experience: first is improving the healthcare outcome for the patient and second is improving how the patient and their family, caregivers, etc., experience the interaction with the healthcare establishment (doctors, other providers, hospitals, health centers, other treatment facilities, etc.). Improving outcomes requires access to and analysis of the patient’s clinical data in order to do diagnosis, treatment planning, care coordination, and all the other things that constitute care.
     Improving the patient’s and their family’s and caregiver’s experiences requires the communication of data so that all parties have the information they need to interact with the healthcare professionals providing care and to make the necessary decisions. This communication facilitates the trust that’s needed in order for the entire healthcare experience to be improved for all parties.
• Improving population health
  • – Addressing population health requires:
    • Identifying specific populations
    • Stratifying the individuals in the population with respect to the prevalent healthcare risks
    • Planning and executing interventions to reduce those risks and improve individual outcomes
       This is not possible without access to comprehensive data about the population(s) and the analysis of this data in order to identify and stratify risks.
• Reducing costs
  • – Clinical and financial data need to be analyzed together in order to determine the cost of care (per capita and along other dimensions such as per provider, per location, etc.). These results can then be used to plan cost reduction measures as well as compare costs over time in order to define the trend in costs.
• Improving the work life of providers
  • – Data is also important in determining work and life conditions for providers and then in planning changes and/or interventions to improve the quality of their work life.

Today most healthcare organizations have 1–5 terabytes (1 TB = 10¹² or, more formally, 2⁴⁰ bytes) of data. This does not include image or device data. How much data is this? By comparison, the printed material in the Library of Congress is about 8–10 TBs. Of course, small organizations, such as community health centers and rural hospitals, have less data, in the 2–3 TB range, while large organizations, such as large urban hospitals (Mass General, Boston, Massachusetts, or Cedars Sinai, Los Angeles, California) may have much more – up to 1–5 petabytes (1 PB = 10¹⁵ or, more formally, 2⁵⁰ bytes, a lot of data). Some very large healthcare organizations, such as Kaiser Permanente, have 10s of petabytes of data. This data includes individual patient clinical (electronic health record, EHR); demographic and logistic (scheduling, etc.) information; patient financial claims and payments; social determinants; cost accounting; inventory and ordering; registry; and operational types of information. Increasingly, it will also include clinical data from heterogeneous sources such as partner organizations, various types of independent provider associations, health information exchanges, and accountable care organizations. Additional sources will include population and public health data from public and private sources, macro- and microeconomic data from state and federal sources, and other data that we don’t anticipate yet. The consensus is that this data will increase at 40%–60% a year – that is, it will double every 2 years! This means that most organizations will have 20 TBs of data in 6 years and the largest organizations may have in the range of 15–20 PBs in this timeframe. The technical implications of this growth in the amount of data to be managed and used will be addressed in Chapter 5.

It’s fall, 2018. According to the Centers for Disease Control (CDC), 86.9% of office-based physicians were using an EHR system as of (1st quarter) Q1 2017.^* This means that the vast majority of providers^† are generating electronic data as part of their clinical practice. How did we get here? Providers, for the most part, did not decide to convert their practices to electronic records spontaneously, although some healthcare practices did just this as early as the 1960s. The Mayo Clinic adopted an EHR system sometime around 1963 and Eclipsis (now part of Allscripts) was developed in 1971 for the El Camino Hospital (Mountain View, California) by the Lockheed Corporation.^‡ Most providers and healthcare organizations, however, adopted EHR systems (and other network-based record systems) because of federal regulations. These include the Health Information Technology for Economic and Clinical Health Act (HITECH), which is Title XIII of the American Recovery and Reinvestment Act (2/2009, Pub.L. 111-5, ARRA) and the Patient Protection and Affordable Care Act (3/2010, Pub.L. 111-148, ACA), as well as changes to the Center for Medicare and Medicaid Services (CMS) Physicians Fee Schedule (PFS 2012–2016) and the Medicare Access and CHIP Reauthorization Act of 2015 (MACRA, 4/2014, Pub.L. 114-10). In particular, the HITECH Act established guidelines for increased Medicare and Medicaid reimbursement if specific criteria were met for the use of electronic health records. These criteria were called Meaningful Use (MU) requirements and there were three stages, starting with Stage 1 in 2009, Stage 2 in 2011, and Stage 3 in 2017.^* In 2015, MACRA re-established the requirements for enhanced reimbursement to include a Value-Based Payment Modifier (VBPM), Physician Quality Improvement Program (PQRS), and Merit-Based Incentive Payment System (MIPS), and it continued the Meaningful Use (MU) program at Stage 3. One thing worth noting about the MIPS program is that it requires healthcare organizations to begin working to improve the health of populations as well as individuals. This requires both additional data and new types of analysis in order to identify populations that are at risk for various health outcomes and to be able to plan clinical and other types of interventions to address population-level issues.

* https://www.cdc.gov/nchs/fastats/electronic-medical-records.htm

† Healthcare provider is defined as: a doctor of medicine or osteopathy, podiatrist, dentist, chiropractor, clinical psychologist, optometrist, nurse practitioner, nurse-midwife, or clinical social worker, who are authorized to practice by the state and performing within the scope of their practice.

‡ https://www.beckershospitalreview.com/healthcare-information-technology/a-history-of-ehrs-10-things-to-know.html

* https://www.cdc.gov/ehrmeaningfuluse/timeline.html

This set of legislation and regulations ensures that healthcare organizations and individual providers continue to use EHR systems that are certified by the Office of the National Coordinator, Department of Health and Human Services (HHS).^† This, in turn, ensures that the clinical, demographic, and other data captured by EHRs continues to grow at ever increasing rates as described above, and that processes for the capture, management, and use of this data become more and more important over time. It is worth noting that this entire regulatory body of material is currently under revision and that the HHS appears to be going through a phase of minimizing regulation. It is not clear what the regulatory landscape will look like in 3–5 years or what requirements for data capture and use healthcare organizations will need to meet.

† https://www.cms.gov/Regulations-and-Guidance/Legislation/EHRIncentivePrograms/Certification.html

One thing is clear regardless of this future direction – the requirement to use electronic health records has changed the way we think about and use data in healthcare.

Chapter 2

What Is Data?

DOI: 10.4324/9780429061219-2

This is a book about data, but in order to understand and use data, or even to write about it, we need to have a common understanding of what it is. The following are several definitions of “data”:

• Factual information (such as measurements or statistics) used as a basis for reasoning, discussion, or calculation
• Information in digital form that can be transmitted or processed
• Information output by a sensing device or organ that includes both useful and irrelevant or red...

Índice

• Cover
• Half-Title
• Title
• Copyright
• Dedication
• Publisher’s Note
• Contents
• Preface
• About the Author
• 1 Introduction – Data Is Essential
• 2 What Is Data?
• 3 Data and Culture
• 4 Data Quality
• 5 What Is Data Analysis?
• 6 The Evolution of Infrastructure and Applications Required for Current and Near-Future HIT
• 7 Machine Intelligence in Healthcare
• 8 Evolution of Data and Analysis in Healthcare
• 9 Summary
• Index