6 Ways That Big Data Could Transform Health Care

Big Data
Kamruzzaman Ratan/E+/Getty Images.

Big data refers to large volumes of data from multiple sources which is rapidly generated, collected, and stored. These data sets are too large and complex to analyze with traditional processes. Big data analytics involves discovering and communicating patterns in the data, and making sense of the data. Retailers like Amazon use big data approaches to promote the right products to the right customers at the right time.

Dr David W. Bates, in an interview with the American Academy of Family Physicians, emphasized the potential role of big data in population health management: “Historically, we've focused more on patients who come to see us. Big data approaches can help in population management so that you are thinking about all your patients and can identify risky situations even in those who don't come in.”

Despite the buzz surrounding big data in health care, the actual uptake has been slow for a variety of reasons. The following article is a high-level summary of 6 ways that big data could transform health care, as described by Dr Bates and colleagues in a July 2014 article in Health Affairs.

1) Reduce health care costs

Spending on health care is not uniformly distributed across the US population. A small group of patients accounts for a relatively large portion of health care spending. Reducing hospitalizations, emergency department visits, expensive medications, and unnecessary testing in this subset of patients may be an effective way to reduce overall spending.

Predictive analytics can identify patients who are likely to incur high costs without further intervention. Characteristics associated with high costs include uncontrolled or severe chronic disease, mental health problems, poverty and other socioeconomic factors. Programs should focus on specific gaps in care that can be addressed in such a way to reduce the need for high cost encounters.

One example would be to identify and treat patients with depression. Or assist patients with issues related to transportation and medication nonadherence.

2) Prevent readmissions

Many readmissions to the hospital can be prevented by delivering the right kind of follow-up care at the right time. This requires hospitals to monitor high risk patients after discharge to detect early signs of decompensation, and to deliver tailored interventions to prevent further decline in health status. Analyzing data from remote monitoring (e.g. smart phones and other devices) may be effective in identifying patients who would benefit from a specific invention shortly after discharge.

3) Improve accuracy of triage

Traditional clinical assessment tools for triaging patients are sometimes too subjective or they depend on data that is only available after the patient has taken a turn for the worse. By evaluating a wider range of data in electronic health records (EHRs) in real time, big data analytics can identify patients who need a higher level of care in the hospital -- and equally important -- those who don’t.

4) Predict decompensation

Decompensation -- a rapid, severe decline in health status -- can happen in the home, nursing facility, emergency department, hospital ward, or even the intensive care unit. Predicting decompensation is relevant to first three categories discussed. One advantage of using big data analytics to identify patients who are at risk for decompensation is the increase in signal-to-noise ratio. By incorporating multiple streams of data, analytics tools may generate fewer false alarms.

5) Predict specific adverse events

Another promising application of big data analytics is to predict which patients are at risk for preventable adverse events, such as infection, renal failure, and adverse drug events. Data on medication exposures, vital signs, lab results, and even genetic profiles allow earlier detection and intervention in these events.

6) Tailor treatment for patients with multi-organ diseases

Chronic diseases which affect multiple organ systems are challenging to manage in an effective and cost-efficient manner. Examples of these diseases include lupus, scleroderma, and rheumatoid arthritis. By analyzing longitudinal data in the EHR, it may be possible to select treatments that are tailored to a patient’s expected disease course.

There are many other potential uses for big data in health care. The sources of big data for health care continue to expand. But real progress will also depend on improving analytical techniques and empowering health care providers and organizations to act on big data findings.


American Academy of Family Physicians. Given Time, 'Big Data' Promises to Transform Patient Care. Accessed on August 29, 2014.

Bates DW et al. Big data in health care: using analytics to identify and manage high-risk and high-cost patients. Health Aff (Millwood) 2014;33(7):1123-31. doi: 10.1377/hlthaff.2014.0041. Accessed on August 29, 2014.

Continue Reading