|
BACKGROUND/RATIONALE:
More than half of the VA enrollees are also eligible for Medicare. Several studies have documented the VA-Medicare dual utilization by VA patients and have compared the characteristics and health outcomes of the dual users in specific diseases. However, little is known about VA stroke patients (1) who have received healthcare within VA only compared to (2) VA stroke patients who have used Medicare only and/or (3) who have used both VA and Medicare simultaneously at the national level. Further, little is known about the characteristics and outcomes between these three groups of users. In order to provide the best quality of care across the continuum to our veterans with stroke, we must understand the scope of services provided both inside and outside the VA system.
OBJECTIVE(S):
This study's focus is to understand the process and outcomes of post-stroke care by VA stroke patients who are VA-only users versus those stroke patients who are Medicare-only users or VA-Medicare dual users. The immediate objectives of this study are: (1) to refine the patients' eligibility and utilization status for each healthcare program (VA and Medicare) in 2001-2002 based upon the methods we developed in our on-going project, VA and non-VA rehabilitation utilization by VA stroke patients living in the state of Florida, and to categorize the patients into three user groups (VA-only, Medicare-only, and the dual users), (2) to describe and compare the demographic and clinical characteristics between the three user groups, (3) to profile and compare 12-month pre-index and 3-, 6-, 9-, and 12-month post-index stroke healthcare utilization between the 3 user groups as well as the patterns of health services dual use by the dual users, (4) to predict the out-of-VA healthcare use, and (5) to estimate the differences in 3-, 6-, 9-, and 12-month post-index stroke mortality, rehospitalization in general, readmission for recurrent strokes, fractures, and discharge to community at index hospitalization between the three user groups and geographic regions.
METHODS:
For this retrospective observational national cohort study, our sample includes all veterans who were VHA enrollees, or Medicare Fee-for-Service plan enrollees, and identified in FY 2001-2002 VA and Medicare inpatient databases using the high specificity stroke ICD-9 codes. Our estimated overall sample would be 26,714 patients; this includes 19% VA only users, 42% Medicare only users and 39% dual users. Secondary data will be obtained from databases of two major sources: VA Austin Automation Center (AAC) for the VA data and VA Information Resource Center (VIReC) for the Veteran Medicare data. Summary statistics will be provided for sociodemographic, clinical and utilization characteristics of the different user groups as well as the dual use pattern by the dual users. Objective 1, 2 and 3 are descriptive studies. For objective 3, multinomial logit model will be constructed to predict the potential out-of-system utilization. For objective 5, we will use general linear-mixed models, with logit link for dichotomous outcomes (mortality, readmission for recurrent stroke, diagnosis of fractures, and discharge to home) and log link for rehospitalizations (any cause). This approach will allow us to consider intraclass correlation for patients within the same medical center by including a random center effect. In addition, for the longitudinal outcomes we will also include time and time by user group interaction as independent variables.
FINDINGS/RESULTS:
No findings/results to report at this time.
IMPACT:
This study is intended to provide needed information about: VA and Medicare healthcare utilization including post-stroke rehabilitation use, pattern of dual use, factors affecting the utilization variation, and several related outcomes of veterans with acute stroke. The study will allow us to develop a model to examine and compare the processes and outcomes of multiple user groups and it will be an important resource for future studies in stroke research. To clinicians, the findings will improve their understanding of the characteristics of their patients, patient VA and non-VA health use behavior, the continuum of treatment, and related outcomes. To healthcare planners and policy makers, this study will provide the evidence-based information about the processes and outcomes of stroke care by veterans who use VA and non-VA sources of care. To researchers, the study results will provide a basis for future studies in VA and non-VA healthcare utilization.
PUBLICATIONS:
None at this time.
DRA:
Chronic Diseases, Health Services and Systems
DRE:
Epidemiology, Resource Use and Cost, Rehabilitation, Quality of Care
Keywords:
Access, Chronic disease (other & unspecified), Stroke, Utilization patterns, Outcomes
MeSH Terms:
Outcome and Process Assessment (Health Care)
|
