Hospitalization is a significant risk for the elderly (65 years or older) and is the largest component of U.S. health care spending accounting for more than 35% of health care dollars (Folland, Goodman, & Stano, 2001). It is estimated that by 2030, the number of elderly will be more than double to 70 million (Centers for Disease Control and Prevention [CDC], 2010). Because 20% of our future population will be elderly with many suffering from chronic diseases, health care and hospital costs for this group are expected to increase substantially (Lamont, Sampson, Matthias, & Kane, 1983). Compared with the general population and driving these predictions, the elderly have significantly higher rates of hospital admission and readmissions and are responsible for most hospitalization days and expenses (Parker, 2005; Victor, Healy, Thomas, & Seargeant, 2000). In the United States (Elixhauser, Yu, Steiner, & Bierman, 2000) and the United Kingdom (National Health Service Information Centre Hospital Episode Statistics, 2006), the elderly account for approximately 36% of hospital admissions; in Australia, they account for 52% (Karmel, Lloyd, & Hales, 2007). In addition to the frequency and expense of hospitalization during inpatient stays, many elderly experience deconditioning and functional decline that affects their future independence, autonomy, and quality of life (Covinsky et al., 2003; Creditor, 1993; Hoenig & Rubenstein, 1991). Published research shows that of 60 functionally independent individuals 75 years or older admitted to the hospital from their home for acute illness, 75% were no longer independent on discharge including 15% who were discharged to nursing homes (Creditor, 1993; Lamont et al., 1983). Reducing length of stay and aggressive discharge policies have become important features of the hospital experience, yet readmission following discharge has become a familiar and frustrating problem and may result from avoidable hospital, and/or community service failures (Colledge & Ford, 1994). Prior hospitalizations increase the probability of future hospitalizations and by identifying and managing patient-level risk factors for rehospitalization, patients' health care and quality of life would improve, whereas morbidity, mortality, costs, and the rate of future hospitalizations would decrease (Colledge & Ford, 1994; Creditor, 1993). Developing a predictive model for rehospitalization risk is one method to describe the factors associated with elderly rehospitalization, identify those at risk, and target individuals for appropriate preventive interventions (Blackman, Kamimoto, & Smith, 1999). Such a model may also predict hospital utilization costs more accurately allowing hospitals to better anticipate staffing and resource needs to efficiently serve elderly patients. However, research has shown that a risk factor must have a much stronger association with the outcome than we ordinarily see in etiologic and epidemiologic research if it is to provide a basis for prediction in individual patients. Despite the strong association between the risk factor and the outcome, it does not follow that the risk factor provides a basis for an effective prediction rule for individual patients (Ware, 2006). OBJECTIVES The purpose of this study is to identify demographic and clinical factors associated with nonelective, all-cause rehospitalization within 120 days of an index admission in the elderly. METHODOLOGY Our sampling frame included all patients 65 years or older admitted to Saint Vincent Catholic Medical Centers (SVCMC) of New York for nonelective admissions between June 21, 2003, and June 20, 2004. For all patients admitted more than once during the study period, we counted each admission as unique. To avoid the possibility of seasonal variation on our sample, we employed a random proportional stratified sampling strategy using the date of admission to stratify the sample by season. …