In the healthcare industry, big data can reveal influencing factors of health, target appropriate care for individuals and communities, enable new discoveries, shape outcomes, and reduce costs.
Nearly 80% of healthcare data is unstructured and does not fit well into traditional relational databases of the past. These data types include DICOM and PACS imagery, doctors’ free form SOAP notes, and structured EMR data. These issues add to the ongoing work still to be completed regarding improvement of data standardization and interoperability, promotion of information sharing, patient engagement, management of data governance, refinement in data science, education of the workforce in new technologies and generation of value of big data to individuals and clinicians at the point of care.
In addition, there are new data sources including medical device instrumentation, patient-provider engagement via smartphones and patient satisfaction information located on social media. These new sources could increase the volume of data captured by a factor of thirty over the next five years. Unless payers and providers are prepared to meet these new challenges, data will be embedded deep within various healthcare systems that will not be mined and analyzed for knowledge, insights and predictions to improve care quality across the healthcare spectrum.
Impacting and Broadening Care Delivery
New patient monitoring medical devices are now able to measure vital signs and send alerts to clinicians at the care facility for appropriate interventions. Blood sugar, pulse and heart beat monitoring can be performed by a variety of devices and sensors, some even attached to smart phones. This extends the care environment right to the patient’s residence including their home, their point of travel and assisted care facilities and ensures that doctors are able to monitor patient progress and outcomes even eternal to the realm of care facilities.
A Big Data Use Case for Weight Monitoring
In one innovative program at UC Irvine Health, congestive heart failure patients are given a wireless scale to take home with them. They are then instructed to weigh themselves at regular intervals as sudden weight gain can be a sign of fluid retention. Algorithms running in Hadoop determine unsafe weight gain thresholds and notify a physician to see the patient before emergency care is needed. This preventative approach is made possible by Big Data.