by Cher Zevala
Big data is a 21st century force to be reckoned with. From groceries on store shelves to online advertisements, the influence of big data is nearly ubiquitous and seems unlikely to diminish anytime soon.
The healthcare industry has jumped aboard the big data train and is working to find ways of analyzing and mining data to uncover insights that might help improve outcomes. The idea of using data and evidence-based research is not new. It has long been used to determine best practices for the care and treatment of patients. The sheer amount of data being collected now, however, is more than any researcher or provider has ever seen or used.
In the field of rehabilitation in particular, not only is there a significant influx of data, there isn’t yet a set of standards for most effectively gathering, manipulating, and analyzing data and putting it to use. As a result, most practitioners are taking a more experimental approach to working with data, and addressing the opportunities and challenges that this new paradigm presents.
Moving From Causation to Correlation
Consider the opportunities big data represents for a rehab practice. A noticeable shift occurring as a result of big data is a widespread change from looking at causes and subsequent responses to looking more at correlations between different variables. The typical methods of data analysis compare data to existing knowledge, and draws conclusions based on that comparison.
To use an example from the banking industry, data analysis is commonly used in fraud detection: When a customer’s debit card is used to make a purchase, the characteristics of that purchase are compared to vast stores of data related to fraudulent purchases, including the customer’s own purchasing patterns. If a certain number of markers are identified, the bank can then contact the customer to confirm the purchase, and decline the transaction if necessary. Within healthcare, the same principles apply: An individual’s condition is compared to data related to other patients with similar symptoms, and a treatment plan is devised.
By shifting to a data-mining perspective, though, and focusing on correlations rather than causations, practitioners can look for patterns within data sets that may bring forth new ideas and treatment protocols. With the amount of data being collected via electronic health records, electronic clinical outcome assessments in research and clinical trials, insurance claims, and more, patterns will become easier to identify—and new ways of working with the data will emerge. This will present additional opportunities for improving a practice, including:
• Improving communication with patients. One of the most common questions that patients have is “How long will treatment take?” Data mining can be used to compare similar patients to not only develop effective treatments, but also provide a better estimate of how long the therapy will take, how much pain or difficulty they may experience, and how to avoid setbacks. This improved communication will help create more satisfied patients and better outcomes.
• Creating measurable goals. Big data provides evidence of measurable outcomes in rehab—information that therapists can use to both set goals for their patients and find better ways for their patients to meet those goals more quickly. More efficient care can reduce cost and improve patient satisfaction.
• Improving provider skill sets. Comparing a facility’s performance against national and local performance, one can identify patterns, both in strengths and weaknesses, and put that information to work in creating employee development plans to improve the overall performance of a practice.
For all of the opportunities that big data provides, though, there are still challenges.
What to Do with This Data?
One of the major challenges to using big data to its greatest advantage in rehabilitation is that it is relatively new and there are not yet prescribed standards for the collection and use of the data. That is all changing, but other challenges remain for providers.
• Determining the best tools for data collection. The question is no longer if a practice should use an electronic medical record, but which medical record it should use to best match its practice. It’s also important to consider how well the system retrieves and uses data from all of the sources that are collecting data. Data needs to be put into usable, comparable formats if it’s going to be useful.
• Putting data to work. Perhaps the greatest challenge for rehabilitation providers is determining how to put the data they collect to work. Providing healthcare is no longer about seeing as many patients as possible to collect higher reimbursements. Rehab practitioners must now prove that they are improving outcomes, not just providing services. Data can help with this, but only when providers are able to use the data they collect to better inform treatment plans.
Big data is a disruptive trend in healthcare that still is in its early stages. Rehabilitation providers will be well served to address the challenges big data presents and find ways to put it to work if they want to remain competitive in the years to come. RM
Cher Zevala is a content coordinator specializing in topics associated with the healthcare industry and innovations in the healthcare field. She is also a contributing writer to Rehab Management. For more information, contact [email protected].
Are a lot of these data points kept in many cases of treatment? I can imagine when it comes to actual results like bloodwork and such, it would be much easier. But for something like a cold or something. How can it be used then?