Big data reveals trends and provides predictive analytics for retail, manufacturing, and has proven especially valuable for healthcare. Whether the Center for Disease Control is using big data to predict the latest flu epidemic or a hospital is trying to reduce post-operative infection, big data analytics improve patient care and practice operations. How big data isolates problems for healthcare is the same process used to isolate any other big data problem, there’s just more data to choose from.
Pharmaceutical companies have been aggregating and organizing data into medical databases for decades. Payers and healthcare providers have adopted digital records; in fact HIPAA regulations require digitization. Government agencies have been storing vast amounts of healthcare statistical data. Assimilating and analyzing that healthcare information is how big data isolates problems; by aggregating information and looking for patterns.
McKinsey and Company reports that in 2005 only about 30 percent of physicians’ offices used electronic medical records (EMRs). By the end of 2011 more than 50 percent of physicians’ offices and more than 75 percent of hospitals used EMRs. And about 45 percent of hospitals belong to regional health information exchanges (HIE), which means they have access to an even broader range of patient and healthcare information. And IDC indicates that the amount of healthcare data being generated is increasing by 40 percent annually.
With this much raw data at their fingertips, healthcare organizations can see how big data isolates problems that affect both patients and operations.
Big Data Solves Big Healthcare Problems
Medical professionals are anxious to apply big data to address problems such as diabetes, hypertension, melanoma, and other preventable ailments. There is plenty of data available for analysis; it’s just a matter of applying the right analytics.
At the UCSF Medical Center, for example, big data has played a key role in preventing heart disease. The Health eHeart Study has gathered data from 1 million people worldwide. The objective of the study is to look beyond the obvious causes of heart disease, such as high cholesterol and smoking, to identify other indicators that reveal prevalence for heart disease.
In addition to questionnaires, the Health eHeart Study team was able to use new technology, such as smartphone apps that could generate EKGs, monitor heart rate, and monitor exercise. They also use Bluetooth sensors to measure blood pressure, apps to monitor diet, and even track wellness activities through social media.
What makes this kind of research different is scale. Rather than conducting limited tests on a few hundred test subjects, analytic data is gathered from a million people around the world. The hope is that the Health eHeart Study will reveal traits that show a likelihood of developing heart disease, and other indicated to prevent heart attacks.
Real-Time Analytics Prevent Hospital Infection
IBM and OhioHealth have been collaborating to control hospital infections using a combination of wireless sensors and real-time big data analytics. By monitoring hand-washing practices, hospital administrators were able to dramatically reduce the number of health-care associated infections, which normally affect every one in 20 patients.
IBM installed sensors at hand washing stations throughout the hospital and interconnected them using a wireless mesh network. Using the sensors the system was able to capture time-stamped use of each station, and match data to when hospital personnel would enter and leave a patient’s room. Using cloud technology the data was stored and analyzed in real time to generate reports and compliance studies.
As a result of this project, OhioHealth’s pilot program in Columbus achieved a 90 percent compliance rate to hand-washing standards, well above the national compliance level of 50 percent.
Big Data Can Optimize Medical Practice
Of course, in addition to treating patients, doctors’ offices are also run as a business, and doctors are learning more about how big data can identify problems with their operations.
Big data could be used to manage billing, for example. Using big data, doctors can submit an insurance claim and determine if the codes match the appropriate treatments for billing. Or big data analytics can provide a better understand of what payers are willing to reimburse in specific instances or regions.
Or consider the medical office that started tracking patient visits against scheduled inoculations. They determined that patients who missed their shots were prone to more office visits. Using big data strategies the office was able to identify at risk patients and reach out to them directly to improve inoculation rates. The result was improved prescription management within 12 months, with 76 percent of patients receiving scheduled inoculations, up from 56 percent.
These are just a few examples of how big data isolates problems in healthcare. There are dozens of more examples. Where do you see yourself applying your expertise to help medical professionals benefit from big data?