Talk to the Veterans Crisis Line now
U.S. flag
An official website of the United States government

VA Health Systems Research

Go to the VA ORD website
Go to the QUERI website
2015 Conference Logo



2015 HSR&D/QUERI National Conference Abstract


3192 — The Impact of Integrative Health Coaching on Health Care Costs and Utilization among Homeless Veterans

Jordan N, Center of Innovation for Complex Chronic Healthcare, Hines VA Hospital; Northwestern University Feinberg School of Medicine; Ward CE, Northwestern University Feinberg School of Medicine; Etingen B, Center of Innovation for Complex Chronic Healthcare, Hines VA Hospital; Hill J, Center of Innovation for Complex Chronic Healthcare, Hines VA Hospital; Miscevics S, Center of Innovation for Complex Chronic Healthcare, Hines VA Hospital; LaVela S, Center of Innovation for Complex Chronic Healthcare, Hines VA Hospital; Northwestern University Feinberg School of Medicine;

Objectives:
Integrative health coaching (HC) is a strategy used to engage homeless patients in their care and connect them to preventive services. Our study objective was to examine the impact of HC on health care utilization and associated costs by comparing homeless Veterans receiving care at a VA facility in which HC was implemented (n = 1794) to homeless Veterans receiving care at a matched-control site (n = 3858).

Methods:
We conducted a retrospective cohort study using administrative and chart review data from 09/01/2011 - 03/31/2014. The control site was matched based on VA complexity rating and geographic proximity. We assessed utilization via inpatient, emergency department, and outpatient visits; average hospital length of stay; prescription medication fills. We assessed costs associated using data from the VA's Managerial Cost Accounting System (MCA). Bivariate and multivariable methods were used to compare mean health services utilization and costs per Veteran.

Results:
Veterans at the HC site had a shorter average hospital length of stay (2.7 vs. 4.2, p < .0001), fewer outpatient visits (36.7 vs. 53.8, p < .0001), and fewer mental health (19.2 vs. 32.9, p < .0001) but more primary care (5.2 vs. 4.5, p = 0.0004) visits. While the HC group had more no-shows (4.8 vs. 4.3, p = 0.0003), they had fewer ED visits than the non-HC group (2.4 vs. 2.7, p = 0.02). The HC group had fewer general prescription medication fills (26.9 vs. 29.0, p = 0.03), but a greater number of psychotropic medication fills (5.4 vs. 4.3, p < .0001). Average inpatient costs for Veterans in the HC group were lower ($4445 vs. $15,516, p < .0001), as were costs for overall outpatient services ($12,441 vs. $13,997, p < .0001) and outpatient mental health services ($4210 vs. $5523, p < .0001). Average costs of overall medications were also lower for the HC group ($489 vs. $666, p < .0001).

Implications:
Although Veterans in the HC group had more no-shows and fewer mental health visits, differences in primary care and medication use indicate that HC may be useful in engaging homeless Veteran patients in their care and fostering a more appropriate and efficient use of VA health care services.

Impacts:
HC is an intervention that should be evaluated in other practice settings.