Transfers from nursing homes (NHs) or community living centers (CLCs) to hospitals place residents at risk for iatrogenic events, disruptions in care, disorientation, stress and increase costs. Over half a million hospital discharges a year emanate from community NHs and estimates of potentially avoidable hospitalizations have ranged from 23% to 67% of NH to hospital transfers. Avoidable hospitalizations are a significant issue that VA Geriatrics and Extended Care leadership has identified as a high priority area for study and intervention. The Interventions to Reduce Acute Care Transfers (INTERACT), a comprehensive, multi-level nursing staff training and quality improvement program, was designed by Dr. Ouslander and his colleagues and has been tested in demonstrations sponsored by the Centers for Medicare &Medicaid Services (CMS) and The Commonwealth Fund, and found to reduce hospitalizations by 17%, 24% among engaged facilities. The goal of this study is to implement and evaluate INTERACT in VHA CLCs. We are guided by the evidence based hypothesis that engagement in the INTERACT program will identify residents'clinical problems earlier, help evaluate and safely initiate management for them, improve communication and use of advance care planning and result in fewer hospital transfers. The INTERACT program, training curriculum and tools will be refined to be appropriate for implementation in a VA setting both in relation to standard CLC operating procedures and VHA language and policies. Based upon merged VIREC and Medicare inpatient and outpatient data and CLC nurse staffing and other organizational data, we will match CLCs in terms of hospitalization rate, proximity to their VAMC hospital and the mix of short and long stay CLC Veterans they serve. A total of 10 pairs of facilities will be identifie and from each pair one will be randomly selected to be approached about implementing the INTERACT program. We expect at least 8 of the facilities approached will accept, adopt and implement the program over the 6 month training period and the additional 12 month active intervention and monitoring period. Both members of the CLC matched pairs will be dropped from the study if the CLC that is randomly selected to be offered the INTERACT program declines to participate. During the 6 month training period and the 12 month active intervention and monitoring period, extensive process measures, ranging from staff time spent in on-line training as well as CLC staff time on the support conference calls, will be gathered for use in both the implementation analyses as well as the Budget Impact Analyses (BIS). Additionally, quality improvement tools applied by staff to examine the characteristics and "avoidability" of hospitalizations during the 6 month training and 12 month intervention monitoring period will be reviewed. There will be no contact with CLCs matched and randomly selected as "controls" since outcome data on their hospitalization rates will come exclusively from existing record data. Changes in the case-mix and seasonally adjusted hospitalization rates of intervention and control CLC pairs will be examined, comparing changes over the two years prior to introducing INTERACT and the two years thereafter. In addition, changes in rates of inappropriate hospitalizations will be examined. We hypothesize that INTERACT will reduce the hospitalization rate by at least 20% without increases in adverse events among Veterans in CLC. If effective, GEC leadership stands ready to integrate INTERACT into ongoing efforts to transform the culture of CLCs to make them more Veteran centered and focused by reducing re-hospitalization thereby increasing the likelihood of returning home.

Public Health Relevance

The decision to hospitalize a Veteran from CLC is complicated and common and evidence suggests up to 40% are avoidable. We will introduce and test the effect of a staff training and quality improvement intervention designed to reduce hospital admissions in 8 randomly selected pairs of VAMC Community Living Centers. We expect to reduce hospitalizations by 20% without adverse consequences for Veterans.

Agency
National Institute of Health (NIH)
Institute
Veterans Affairs (VA)
Type
Non-HHS Research Projects (I01)
Project #
1I01HX000978-01
Application #
8399284
Study Section
Special Emphasis Panel (HCR6)
Project Start
2013-07-01
Project End
2018-06-30
Budget Start
2013-07-01
Budget End
2014-06-30
Support Year
1
Fiscal Year
2013
Total Cost
Indirect Cost
Name
Providence VA Medical Center
Department
Type
DUNS #
182465745
City
Providence
State
RI
Country
United States
Zip Code
02908