Bridging the Gap Between Big Data and Social Services

The goal of health and human service agencies is to benefit the general public as well as protect at-risk populations from worsening social concerns. While there has been a growing focus on prevention, predictive models can be hard to translate into solutions that can be effectively implemented. The recent proliferation of big data sources has created an unprecedented opportunity to leverage data in order to focus work with vulnerable populations and provide predictive-based intervention prior to the worsening of an individual’s situation. For example, publically available court records indicating an imminent eviction can be used in order to identify a population at a greater risk of becoming homeless. Prevention services can be provided to these identified individuals prior to their becoming homeless. This intervention, which precedes actual homelessness, not only helps an individual or family, but is also cost effective for the city. Such an approach requires integrating solutions across multiple levels: data integrity, predictive analytics, and implementing an effective intervention process. There are not many organizations that have the necessary tools, ability and knowledge to follow through on all these levels in order to deliver an effective outcome. In this perspective we would like to introduce a predictive-based social intervention approach and examine the associated challenges that must be addressed.