Chronic conditions often co-occur, and one in five individuals with chronic illness also presents with activity limitations. In addition, because of shared pathophysiology, drug-drug interactions or drug-disease interactions, chronically ill older individuals may also present with geriatric syndromes, such as falls, cognitive impairment, or urinary incontinence. Our group has demonstrated that it is the combination of chronic conditions (CC) together with functional limitations (FL) and/or geriatric syndromes (GS), rather than CCs alone, that are associated with unfavorable health outcomes (fair/poor health status, two-year health decline, and mortality). However, little is known about the specific combinations of conditions - within and across the broad rubrics of CC, FL, and GS - that have the greatest implications for clinical and resource management. Relying on statistical learning techniques, and using the 1992-2010 linked Health and Retirement Study (HRS) and Medicare data, we aim to identify and rank specific combinations of CC, FL, and GS according to a) frequency;b) health outcomes;c) health services utilization;and d) institutional and non-institutional cost of care. Our metrics for health services utilization include acute (admission to the emergency department or to the hospital and the number of inpatient days), post-acute (skilled nursing facility, home health care, or hospice), and long-term care. Furthermore, we will examine variations in multiple chronic conditions (MCC) combinations and outcomes across subgroups of older adults by age, sex, race/ethnicity, and socioeconomic status, as well as between fee-for-service Medicare beneficiaries and managed care enrollees. This study is innovative, in that, rather than focusing on CCs alone, it accounts for the impact of specific CC(s) in conjunction with presence and severity of FL and GS, on health outcomes, health services use and costs in older adults. Our use of the HRS-Medicare data will offer a unique perspective by combining the rich primary data on such pertinent measures cognitive status, by severity, from the HRS with utilization data from Medicare claims files. In addition, we rely on state-of-the-art approach of statistical learning to understand "what the data says" with regard to identifying specific combinations of clinical conditions in relation to health care use and outcomes. The findings will have important implications in both research and clinical practice. Methodologically, our study will highlight the importance of a more nuanced approach to defining MCC when examining health outcomes and resource use. From a clinical perspective, the findings will show the importance of a more comprehensive and personalized approach when characterizing patients with MCC. In addition, as we identify subgroups of patients with MCC with high risk for poor outcomes and high resource use, it will be possible to design targeted interventions to better meet these patients'health care needs and to curtail health care utilization.
In previous studies, when studying outcomes in older adults with multiple chronic conditions (MCC), we documented the incremental effects of functional limitations and geriatric syndromes above and beyond that of chronic conditions relative health outcomes. In the proposed study, we will identify and rank specific combinations of conditions, within and across the broader rubrics of chronic conditions, functional limitations, and geriatric syndromes according to a) frequency;b) health outcomes;c) health services utilization;and d) institutional and non-institutional cost of care. We will rely on statistical learning techniques t handle the very large number of possible combinations of clinical conditions. The subgroups of patients with MCC with high risk for poor outcomes and high resource can then be targeted for interventions to better meet these patients'health care needs and to curtail their health care utilization.