The purpose of the study is to examine the rate of hospitalization of nursing home patients. Behavioral models used in the study are based on contingency theory and resource dependence theory. Data come from a sample of residents in 30 nursing homes in five states linked to Medicare claims data, On-line Survey of Certification Automated Records (OSCAR) and Area Resource File (ARF) data. The goal is to separate the proportion of explained variation in hospitalization rates that is due to market and facility factors from the portion explained by resident characteristics. Part of the analysis will focus on two potentially avoidable causes for hospitalization: dehydration and diabetics who are admitted with a primary admission diagnoses of hyperglycemia or hypoglycemia.
