What if doctors had the sum of all patient information, or “big data,” at their fingertips? Further, what if software could analyze the data and from that analysis make predictions about patient health so that the best course of action could be determined, and patient care could be optimized?
This is the promise of big data analytics, which can be divided into three categories:
- Descriptive analytics – Data describing past events, for example, lab tests, patient surveys, and clinical documentation.
- Predictive analytics – This likelihood of predicting a future outcome based on descriptive data
- Prescriptive analytics – Describes the course of action producing the highest likelihood of maximum benefit
Here are some areas where big data analytics are being employed in healthcare.
Identifying Mutations That Can Cause Genetic Disorders
Genetic testing by companies such as Progenity, founded by Executive Chair Harry Stylli, uses big data analytics to uncover patterns through the analysis of large-scale data sets. Diverse genomic data is combined with electronic health records (EHRs) to identify genetic variants for individualized diagnosis and therapy.
Reacting Quickly to Patient Deterioration
Data analytics can help hospitals react as quickly as possible to sudden changes in patients’ vitals, even before symptoms become apparent to staff. The University of Pennsylvania used an analytics tool combined with machine learning and EHRs to target patients likely to develop severe sepsis or septic shock 12 hours before onset.
Preventing Patient Suicide
To identify patients with suicidal tendencies, researchers using a combination of EHRs and a standard depression questionnaire found that suicide attempts and deaths among patients in the top 1 percent of predicted risk were 200 times more likely to attempt suicide than those in the bottom half of predicted risk.
Boosting Patient Involvement
Consumer relationship management is another tool used by providers and by insurance companies to promote wellness and thus reduce long-term spending. Predicting patient behavior is a key aspect of this tool. Consumer profiles allow the payer to send specific messaging. The result can be improved consumer retention as well as tailored strategies designed for each individual. Data analytics promises to help engage patients in all aspects of their care.
Developing Precision Medicine and New Therapies
Analytics can supplement traditional clinical trials and drug discovery techniques. For example, “In silico” testing uses computer modeling and simulation instead of recruiting patients for clinical trials, which can be complicated, invasive, and expensive. At the same time, in silico testing speeds up new therapy evaluation. This technique is also used to incorporate to identify patients who might need dose adjustments.
Managing the Supply Chain
The supply chain historically has been one of the largest cost centers for most providers. Thus, predictive tools are being used by hospital executives who are looking to reduce costs and improve efficiency to gain insights into ordering patterns and supply usage. Data analytics have been ranked as the top priority by supply chain organizations.
Big data analytics play a major part in this beginning of a transformative era in scientific, medical, and healthcare technology.