The purpose of the project is to evaluate the impact of continuity of care on patient outcomes emerging from three hospital-home transition systems for cancer patients discharged with complex home care requirements. These systems represent varying levels of hospital and home care integration: an integrated system in which hospital staff, discharge planner and home care staff all work in one agency; a partially integrated system in which a home care agency has a liaison nurse assigned to work in specific member hospitals each with its own discharge planner; and a nonintegrated system in which a home care agency has no direct link with and is completely independent of the hospitals in its service area. Each hospital in a nonintegrated system has its own nurse or nonnurse discharge planner. A three year two phased project is planned. The descriptive phase is an investigation of the relationship between transition systems (and other independent variables), continuity of care (the mediating variables), and patient outcomes (the dependent variables, e.g. severity of complications, self-care ability, cost of home care, satisfaction with home care, and quality of life). This phase will be completed in 16 months with 420 patients (120 colorectal, 120 breast, 120 lung, and 60 head and neck cancer patients) distributed across transition systems. It is hypothesized that type of transition system impacts on continuity of care and that continuity of care, in turn, impacts on patient outcomes. In the experimental phase, the liaison nurse role is introduced to two Visiting Nurse Associations and a random sample of referring hospitals in previously nonintegrated systems (experimental conditions). Other hospitals in nonintegrated systems will not have the liaison nurse (control conditions). This phase will be completed in 18 months with a second set of 420 patients from the same diagnostic categories distributed as follows: 95 patients in each of four conditions (experimental and control conditions crossed with nurse or nonnurse discharge planners), 20 from integrated, and 20 from partially integrated systems. It is hypothesized that patients from integrated, partially integrated and experimental conditions will have better continuity of care and patient outcomes than those in control conditions; that within control conditions, systems with nurse discharge planners will have better continuity of care and outcomes than those with nonnurse discharge planners. The last 2 months are taken up with analysis and write-up.
