If there were any doubts about the value of data analytics for healthcare organizations to turn data into actionable insights, a new book from the American Health Information Management Association attempts to put those doubts to rest and provide step-by-step instructions for analyzing data and using statistical techniques.
The co-authors of the book—David Marc and Ryan Sandefer, faculty members at the College of St. Scholastica in Duluth, Minn.—argue that proficiency in data analytics is increasingly important for all health information managers and health informaticians as the industry embraces quality improvement initiatives.
Their book, Data Analytics in Healthcare Research: Tools and Strategies, includes case studies on how to conduct healthcare data analytics, providing insights on databases and statistical software for data extraction, normalization, transformation, visualization, and statistical analyses.
Health Data Management spoke with Marc, an assistant professor at the College of St. Scholastica and the graduate program director for health informatics, who previously worked for a biotech company where he developed predictive data models for the diagnosis of neurological and immunological diseases.
HDM: How will your book help healthcare organizations as they venture into the area of data analytics?
Marc: The book offers them the opportunity to really learn how to do data analysis. And, the way that we formatted this book was to give a very hands-on delivery method to maximize their understanding of how to do analytics—a tremendously growing area in healthcare today. We have more and more data that’s being collected, particularly data in an electronic format because of the growth of electronic health records. As a result, there’s greater need for having trained professionals that know how to work with data. Those healthcare organizations that know how to leverage the use of data and to make decisions based on that data have a leg up on the competition.
HDM: You have argued that proficiency in data analytics is increasingly important for all health information managers and health informaticians. How so, and what about clinicians?
Marc: It really is important. We see it as a core competency within the health information management field and health informatics. No matter what your role is—whether you are a health information management director or an interface analyst or coder—you need to know how to work with data and to analyze it because it’s going to help you in performing your job. You will be using some form of data analysis throughout your career working in healthcare. And, I would say it’s even broader than that. It’s not just health information management or health informatics professionals. Data analytics is important for most people working in healthcare today. If you are a provider, you need to know how to identify trends—the hallmark of being a practitioner today—as we get to more personalized medicine that is tailored specifically to the individual needs of patients to make more informed decisions that result in better health outcomes. Organizations need to utilize that patient data collectively for decision making as well. So, data analytics really does impact a diverse group of professionals in a variety of different roles.
HDM: Do these professionals have the necessary skills and experience in terms of data analytics to perform at that level?
Marc: Currently, I think there is a skills gap. At healthcare organizations, in hospitals in particular, there’s more and more reporting requirements because of changes in the way that medical services are being paid for by payers and they need to have proof that they’re obtaining better outcomes. We’re at a point where many individuals don’t feel like they’re prepared to handle those reporting requirements or have a good understanding of how to work with this data. There’s generally a skills gap in the profession right now. That’s why there’s a main focus in building curriculum in colleges and universities so that we can ensure students graduating from these programs have the necessary skills to take on these tasks.
HDM: Putting human resource issues aside for a minute, how about computing resources? Are healthcare organizations making the investments they need to make in hardware and software for this kind of data analysis?
Marc: There’s been major growth in health information technology, particularly in the area of data analytics. Many IT vendors are creating products for that specific need. So, there’s growth in the vendor space. And, there’s beginning to be more growth on the clinical side as well. More hospitals and clinics are looking at adopting platforms that can help facilitate the analysis of this data. Now, many of these are costly systems that require quite a financial investment. But, there are other alternatives as well that healthcare organizations can look at using, such as open-source analytics tools. In fact, that’s what we cover in our book—how to use tools that are freely available and are open source. For instance, we use MySQL Workbench which is a free database software tool and it can help manage large databases, providing a nice visual for interacting with the data. And, then, there’s R—another tool. It’s free data analytics software and is one of the fastest growing platforms in the world. It’s beginning to be used in healthcare more and more.
HDM: You also talk in your book about the open-source data available from a variety of federal agencies including the Centers for Medicare and Medicaid Services, Health Resources and Services Administration, Office of the National Coordinator for Health IT, and the U.S. Census Bureau.
Marc: Yes, this is another important area. There is so much data available that can be leveraged by consumers for making more informed decisions about where they want to go to receive care, but also for healthcare organizations so they track their performance over time and can make decisions regarding performance relative to others using these public data sources. We have used this public data to build a database. Those that purchase our book actually get a relational database with it that essentially combines many of these public data sources so you can gain information not just by looking at a single public database in isolation.
HDM: Relational databases, such as EHR systems, organize data into columns, rows and tables, forcing information into predetermined categories. However, the problem is that while these databases are ideal for data that is easily structured, they cannot handle unstructured data as easily. What about non-relational databases as a solution for this kind of data?
Marc: NoSQL doesn’t follow the traditional tabular relations used in relational databases. However, in healthcare, we are still at a point where we’re trying to find structured means of storing the data. I do anticipate in the future we will see greater growth in healthcare utilizing the NoSQL non-relational database structure, particularly as the industry transitions to collecting different forms of data such as images or long narratives of text such as clinician notes and being able to store that in less of a schematic way. For now, the need is for people who understand data that is stored in a structured way using SQL and relational databases. While we will see growth in NoSQL, the necessary competencies and skills lie in having individuals who know how to work the relational database so they can help facilitate the storage, retrieval, and analysis of that data.
Register or login for access to this item and much more
All Health Data Management content is archived after seven days.
Community members receive:
- All recent and archived articles
- Conference offers and updates
- A full menu of enewsletter options
- Web seminars, white papers, ebooks
Already have an account? Log In
Don't have an account? Register for Free Unlimited Access