Leveraging big data in population health management. Issue 1 (December 2016)

• Record Type: Journal Article
• Authors: Wells, Timothy; Ozminkowski, Ronald; Hawkins, Kevin; Bhattarai, Gandhi; Armstrong, Douglas
• Is Part Of: Big data analytics. Volume 1:Issue 1(2016)
• Page Start: 1
• Page End: 14
• Publication Date: 2016-12
• Subjects: Medicare -- Health information systems -- Public health surveillance -- Quality assurance health care; Big data -- Periodicals; Biology -- Data processing -- Periodicals; 570.28557
• DOI: 10.1186/s41044-016-0001-5
• Languages: English
• ISSNs: 2058-6345

Abstract

Abstract Background Population health management takes into account many determinants of health, including medical care, social and physical environments and related services, genetics, and individual behavior. Many different types of data may be used to guide population health management programs and to estimate program value. In addition to the variety of data required for these programs, big population health program data are characterized by large volume, high velocity, and inconsistent data flows. This manuscript describes how big data analytics have been used to craft a population health program to help improve the lives of about four million older adults who have an AARP® Medicare Supplement Insurance plan (i.e., a Medigap plan). Plan enrollees have access to a wellness program, holistic care coordination programs, two telephone-based advice lines, concierge support for insurance and medical care needs, and a program designed to help reduce unnecessary emergency room (ER) visits. Results During 2009–2011, these program components led to several improvements in health care. For example, increased duration in care coordination was associated with fewer hospital readmissions, and participants were significantly more likely to have recurring physician office visits and recommended laboratory tests. Participants in ER decision support reduced their ER visits by 1299 visits per 1000 insureds, compared with a reduction of 1121 visits per 1000 insureds for individuals who did not participate in the program. Better depression management helped reduce depression symptoms in 59 % of participants engaged in that program. Big data analytics of member data suggested the need for a wellness program feature, which began in 2014. Analytics of disease management services offered in 2009–2011 helped to combine and refocus these program features to enhance effectiveness in later years. Conclusions Using big data to help manage and evaluate a population health program has led to several improvements in health care. Program management, reporting, and evaluation processes generated additional data which, when analyzed, continues to refine program implementation and quality. Future improvements to this program may include enhanced integration of social service programs that will generate their own data streams for analyses designed to further improve health and wellbeing.