Wellness and Healthcare Informatics

It is now accepted that factors other than medical care play a high role in population health: genetics 20%; social,environmental, and behavioral factors 60%; medical care 20%. Recently, community-based interventions and high touch comprehensive care programs, have been introduced to address unmet social needs and physical-behavioral health challenges. However, to improve outcomes and reduce medical cost for patients with complex and costly care it requires an understanding of their unique needs and characteristics. Current linear predictive models have limitations, and superior

data models are necessary that improve:

• The ability to anticipate patients’ needs (clinical and non-clinical)

• Determine how impactable are the targeted patients and how interventions can impact them

• Predict the occurrence of expensive events: emergency room visits and hospitalization).

The authors have implemented predictive models based on activity-based intelligence (ABI) that rapidly integrates NLP and multiple data sources (clinical, claims, consumer, socioeconomics, media, environmental), through a platform evolved from use by the US Intelligence Community that, in real time, identifies, patients most likely to benefit from redesigned care delivery systems and recommend optimal interventions. Specifically, our methodology facilitates health enterprise zone initiatives to improve access to health care and health outcomes in underserved communities by designing detailed heat maps that recognize unmet social needs and neighborhood stress concentrations. Our predictive impactability models optimizes and guide the design of measurable interventions that are effective in achieving the desired objectives of improving social,personal and health conditions.