Background. Although nursing homes are providing substantial end-of-life care for an increasingly elderly population with chronic, progressive illnesses, such care is often limited, with few receiving the "gold standard" of hospice services. Care directives (do-not-resuscitate [DNR], do-not-hospitalize [DNH], feeding restriction [FR], and medication restriction [MR]) may represent alternative approaches to defining wishes for end-of-life care. Yet, their use varies and seems to be influenced by factors other than health status, such as ethnicity and health system characteristics.

Methods. A descriptive correlational study design was employed using nursing home admission assessments from the Maryland 2000 Minimum Data Set (MDS) 2.0. Associations between resident characteristics (e.g., functional status, diagnoses, demographics) and the use of care directives and hospice were examined. Nursing home characteristics were measured using the Centers for Medicare and Medicaid Services On-Line Survey Certification and Reporting (OSCAR). Analyses reflected 10,023 Unduplicated admission records from 77 nursing homes.

Findings. The most frequently used care directive on admission was DNR (28%), followed by FR (9.5%), DNH (3.4%), and MR (1.3%). A very small percentage of residents received hospice on admission (1.7%). Appropriately, health-related characteristics had the strongest influence on use of care directives, however, multivariate logistical modeling found that Non-White race and Medicare insurance decreased the likelihood of having a DNR, DNH, and FR. Controlling for health-related factors, only non-Medicare payer significantly influenced the use of hospice care. Of subjects with two or more care directives, only 8.7% received hospice care. Although DNR, DNH, and FR increased the likelihood of receiving hospice care, the relationship was small, suggesting that care directives are used independently of hospice.

Implications. Since policy and reimbursement barriers to hospice use are likely to persist, care directives should be used to focus communication with residents, families, and providers. Future longitudinal studies should identify the pattern of how care directives are used and their interrelationship with hospice care, with the ultimate goal of more widespread implementation of hospice care principles.