"Commentary:" Analytics and Healthcare: Ten Trends for 2016
Analytics continues to bring dramatic change to the healthcare industry in the United States and other countries, offering advances and challenges for the year ahead. Following are 10 trends to chart in 2016.
1. Increasing patient demand for health services spurs greater use of analytics/operations research models. As the number of patients with insurance increases, so do the wait lists for health services. An increasing rate of insured patients in the United States creates greater need to optimize the allocation of limited healthcare resources. Just as lowering inventory in a factory exposes problems with product flow, this increasing demand for healthcare will expose challenges in managing patient flow through health systems. Greater demand for services will create the need to prioritize patients on the basis of need and urgency. The plus side: it will open the door for new and innovative analytics approaches to healthcare delivery.
2. EMRs (electronic medical records) become the foundation for useful information, not just data. More data than ever before are available in EMRs, presenting new opportunities to build and calibrate operations research models for improving healthcare delivery. What’s more, EMRs provide more than just raw data about patients. Information generated from quantitative models that measure risk and other relevant factors will aid physician and patient decision making. Putting operations research/analytics models in the hands of physicians at the point of care will increase the accuracy and consistency of medical decision making.
3. Use of mobile health apps balloons. There are more than 100,000 healthcare apps for helping patients improve and maintain a high quality of life. These apps frequently incorporate statistical risk models and other analytics-related methods to aid decision making. These apps present new opportunities for analytics researchers and practitioners to push cutting-edge approaches directly into the hands of physicians and patients.
4. Medical devices drive analytics. New medical device technologies create an unprecedented need for new operations research/analytics methods to evaluate healthcare choices. New technologies abound, and in some cases, the costs are extraordinarily high. There is an increasing need to evaluate the short- and long-term benefits of new technologies to decide which technologies to adopt as best practices and which to avoid because they will drain health systems of resources, with high copays and little or no benefit to patients.
5. Pay for performance incentives induce demand for new ways to individualize patient care. With most people over 65 having multiple chronic conditions, there is an increasing burden of treatment requiring careful thought about what most affects a patient and what they really need to maintain the highest possible quality of life. Prioritizing treatment options requires the ability to individualize predictions about treatment outcomes for common conditions and diseases.
6. Demand for shared decision making drives the need for new decision aids. Patients need help weighing the harms and benefits of the treatment options that they are presented. A large inventory of operations research methods exists to aid decision making, but many physicians and patients need help interpreting the kinds of complex information that come from these models. Analytics approaches for implementing and visualizing personal health information will be critical to achieving new successes in personalized decision making.
7. Changes to the “common rule” lower the barrier to initiate needed healthcare studies. In September 2015, the Department of Health and Human Services and 15 other government agencies in the United States proposed a revision to the regulations governing ethical conduct for human subject research. The goal of the proposal is to enhance safeguards for research participants while improving the efficiency of the oversight and approval process for research projects. If the proposal passes, it will become easier to get approval for analytics and operations research projects that use de-identified data and/or pose minimal risk to patients.
8. Degree programs in analytics are increasing. As recently as five years ago only a handful of graduate programs turned out experts with the broad combination of analytical skills that are needed to visualize, analyze, and act on the vast amount of data being generated in healthcare. Now, more than 150 programs offer students the opportunity to specialize in analytics. Sites like https://www.informs.org/Build-Your-Career/Education-Resources/Analytics-Resources-for-Faculty-and-Program-Directors are helping program directors and instructors develop these programs.
9) Analytics certification is a growing trend. Nonprofit programs like the INFORMS Certified Analytics Professional (https://www.certifiedanalytics.org/) and vendor certifications like SAS certification (http://support.sas.com/certify/) are giving healthcare providers a new way of evaluating the credentials of healthcare analytics job applicants. They’re also allowing seasoned healthcare professionals and recent graduates to build and demonstrate their qualifications.
10) More collaboration, more team science. The National Institutes of Health (NIH) and other funders worldwide are increasing their focus on multidisciplinary collaboration. In the United States, a new initiative in precision medicine recognizes the need to harness discoveries, such as genetic biomarkers, to optimize treatment protocols based on personalized information (https://www.whitehouse.gov/precision-medicine). The benefits: operations researchers, with their diverse training and expertise, can contribute their knowledge to large team projects in search of new ways to cure and alleviate the burden of chronic diseases like cancer, diabetes, and heart disease.
Brian Denton, Associate Professor, University of Michigan, is the 2016 President-Elect of the Analytics Section of INFORMS (Institute for Operations Research and the Management Sciences.