Analyst: add 'social determinants' to Big Data to improve health risk predictions

Jeff Rowe
Analyst: add 'social determinants' to Big Data to improve  health risk predictions

There’s been no shortage of buzz around the potential of “Big Data” to transform the way many sectors, including healthcare, conduct their business.  Despite that potential, however, healthcare organizations still “largely depend on the analysis of traditional data sources to understand and predict patient and population health risk.”

That’s according to one risk analyst, who says the time has come for healthcare providers and policymakers to begin to take advantage of newly available types of data in order to better predict the health of both individual patients and entire populations.

Writing for Executive Insight, Kathy Mosbaugh, vice president of clinical solutions for LexisNexis Risk Solutions, argues that traditional data sources, including billing, laboratory, pharmacy, medical claims and patient health risk assessments, “are no longer enough to provide accurate risk and outcome predictions.”

Rather, she says, “nonclinical life events and socioeconomic information,” identified by the Institutes of Medicine as "social determinants of health,” are known to exacerbate a range of health conditions, “ranging from asthma, diabetes and high blood pressure to depression, metabolic syndrome and chronic obstructive pulmonary disease.” These “determinants” can  include street crime, illiteracy, income levels and “even lack of access to fresh, healthy food.”

As an example, she discusses the theoretical case of a single mother who has just experienced a divorce and had to move to a new neighborhood that has high crime rates. While the woman has always been in good health, having to find a job after she has been caring for her child at home and dealing with emotional effects of divorce could make her susceptible to serious stress that can set off a series of adverse health challenges. “If this type of change in an individual's social environment could be captured,” Mosbaugh argues, “it could serve as an early warning system to the possibility of increased clinical risks. Such an understanding would allow either her health plan or provider to be proactive in getting her the care she and her child need.”

This type of insight is equally critical for identifying and treating low or nonusers of the healthcare system who have unknown and rising risks. To be sure, a person's medical history will always provide the greatest insight into their future health risk, but the use of socioeconomic data  can increase the accuracy of traditional age/gender models for identifying and predicting future costs.